Belay Zeleke Dilnesa - Eawag-Empa Library

Fe-containing hydrates and their fate during cement 

hydration: thermodynamic data and experimental 

study 

THÈSE N O 5262(2011) 

PRÉSENTÉE le 07 DECEMBRE 2011 

À LA FACULTE SCIENCES ET TECHNIQUES DE L'INGÉNIEUR 

LABORATOIRE DES MATÉRIAUX DE CONSTRUCTION 

PROGRAMME DOCTORAL EN STRUCTURES 

ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE 

POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES 

PAR 

Belay Zeleke Dilnesa 

acceptée sur proposition du jury: 

Prof. Nava Setter, président du jury 

Prof. Karen Scrivener, Dr. Barbara Lothenbach, directeur de thèse 

Dr. Guillaume Renaudin, rapporteur 

Dr. Thomas Matschei, rapporteur 

Dr. Paul Bowen, rapporteur 

Proposée en decembre, 2011

ABSTRACT 

Thermodynamic modeling is a versatile tool for predicting the chemical composition 

cement during the hydration of cement. The quality of the thermodynamic modeling 

depends directly on the quality and completeness of thermodynamic database used. One 

of the main limitations of modeling the hydration of cement is the lack of thermodynamic 

data for Fe containing hydrates. In addition, the formation of solid solutions between Fe- 

and Al-containing hydrates could stabilize mixed solids. However, it is unclear to what 

extent such solid solution formation occurs. Also experimentally it is very difficult to 

identify Fe-containing hydrates in hydrating cements by standard analytical techniques as 

the signals from Fe-containing phases significantly overlap with those from the 

corresponding Al-containing phases. 

Thus, in this study, potential Fe-containing hydrates like Fe-hemicarbonate (Fe-Hc), Fe- 

monocarbonate (Fe-Mc), Fe-monosulphate (Fe-Ms), Fe-Friedel’s salt (Fe-Fr), Fe- 

strätlingite (Fe-St), Fe-katoite (C3FH6) and Fe-siliceous hydrogarnet (Fe-Si-Hg) were 

synthesised at 20, 50 and 80 °C. The solid phases were characterized by X-ray powder 

diffraction (XRD), Thermogravimetric analysis (TGA), scanning electron microscopy 

(SEM), vibrational spectroscopy (Raman and Infrared spectroscopy) and Extended X-ray 

absorption fine structure spectroscopy (EXAFS). The compositions of the liquid phases 

were analyzed using inductively-coupled plasma optical emission spectrometry and mass 

spectrometry (ICP-OES and MS). At ambient temperature Fe-Mc, Fe-Ms, Fe-Fr and Fe- 

Si-Hg were stable, while Fe-Hc, Fe-katoite and Fe-St were metastable. Fe-Mc, Fe-Ms, 

Fe-Fr and Fe-Si-Hg were stable also at 50°, but the Fe-AFm phases were unstable at 80 

°C while Fe-Si-Hg were stable up to above 100 °C. 

i

The measured composition of the liquid phase was used to calculate the solubility 

products at 20 and 50 °C and to derive the data for standard conditions (25 °C, 1 atm). 

The solubility products of Fe-Fr was similar to the solubility product of Al-Fr, while the 

solubility products of Fe-Mc and Fe-Ms were about 3 log unit lower than that of Al-Mc 

and Al-Ms indicating that in Fe-Friedel’s salt is probably not stable in cements. The very 

low solubility product of Fe-Si-Hg (5 to 7 log units lower than that of Al-Si-Hg) implies 

that Fe-Si-Hg could be a stable phase in hydrated cements. 

Also the mixed Al- and Fe-containing hydrates were synthesized to study the extent of 

solid solution formation. Both XRD and thermodynamic modelling of the liquid 

compositions indicated that Al- and Fe-monosulphate and Al- and Fe-Friedel’s formed 

solid solutions with a miscibility gap, while Al- and Fe- monocarbonate existed as two 

separate hydrates due to their different crystal structure (Al-Mc: monoclinic, Fe-Mc: 

rhombohedral). The formation of solid solution between Al and Fe-siliceous hydrogarnet 

seemed probable. 

To understand to what extent the findings from the synthesised hydrates were relevant for 

real cements, the speciation of iron was determined in hydrating cement using EXAFS 

spectroscopy. Identification of Fe-containing hydrates and quantification of their 

contributions was achieved by combining principal component analysis with iterative 

target tests, and linear combination. The results show that several Fe species already 

contributed to the overall Fe K-edge spectra of cement pastes during the first day of 

hydration. While ferrite was the dominant Fe-containing phase in the unhydrated cement, 

Fe-hydroxide was detected shortly after starting the hydration process. With time the 

formation of stable Al/Fe-siliceous hydrogarnet was observed, while the amounts of Fe- 

hydroxide and ferrite clinker slowly decreased. The latter finding agrees with results from 

ii

thermodynamic modeling of the hydration process, which predicts formation of stable 

Al/Fe-siliceous hydrogarnet in cement system. 

The determination of the solubility products of these hydrates will help to extend the 

thermodynamic data base of cement minerals and establish whether and to which extent 

Fe-containing hydrates are stable in fresh and in leached cementitious systems. The 

results from this study on the Fe speciation in cementitious systems are important for a 

better understanding of cement-water interactions with a view to the durability of 

cementitious materials. 

Keywords: Fe-containing hydrates; solubility product; solid solution; crystal structure; 

thermodynamic modeling; thermodynamic data 

iii

ZUSAMMENFASSUNG 

Thermodynamische Modellierung ermöglicht die Mineral-Zusammensetzung von Zement 

während der Hydratisierung zu berechnen. Die Qualität der Modellierung hängt dabei 

stark von der Qualität und der Vollständigkeit der verwendeten thermodynamischen 

Datenbanken ab. Eine wesentliche Einschränkung bei der Modellierung der 

Hydratisierung von Zement ist das Fehlen von thermodynamischen Daten für die 

eisenhaltigen Zementhydrate. Zudem könnte die Bildung von festen Lösungen (solid 

solution) von Fe- und Al-haltigen Hydraten gemischte Festphasen stabilisieren. Zurzeit 

ist allerdings nicht bekannt, ob und in welchem Ausmass diese festen Lösungen 

entstehen. Die Identifikation der eisenhaltigen Hydrate in Zementstein mittels 

Standardtechniken ist sehr schwierig, weil die charakteristischen Signale der Fe-haltigen 

Phasen oft stark mit denjenigen der Al-haltigen Phasen überlappen. 

In dieser Studie wurden potentiell Fe-haltige Hydrate, wie Fe-Hemikarbonat (Fe-Hc), Fe- 

Monokarbonat (Fe-Mc), Fe-Monosulfat (Fe-Ms), Fe-Friedel’s Salz (Fe-Fr), Fe-Strätlingit 

(Fe-St), Fe-katoite (C3FH6) und Fe-Si-Hydrogranat (Fe-Si-Hg), bei 20, 50 und 80 C 

synthetisert. Die Festphasen wurden mittels Röntgenpulverdiffraktometrie (XRD), 

Thermo-gravimetrie (TGA), Raserelektronenmikroskopie (SEM), Raman und Infrarot- 

Spektroskopie, und synchrotron-basierter Röntgenabsorptionsspektroskopie (EXAFS) 

charakterisiert. Die Zusammensetzung der Flüssigphase wurde mittels induktiv 

gekoppelter Emissions-spektroskopie mit optischer oder massenspektrometrischer 

Detektion (ICP-OES oder MS) bestimmt. Die Untersuchungen zeigen, dass bei 

Raumtemperatur Fe-Mc, Fe-Ms, Fe-Fr und Fe-Si-Hg stabil sind während Fe-Hc, Fe- 

katoite und Fe-St metastabil sind. Fe-Mc, Fe-Ms, Fe-Fr und Fe-Si-Hg sind auch bei 50° C 

iv

stabil. Die Fe-AFm Phasen sind nicht stabil bei 80° C während Fe-Si-Hg bis 100° C 

stabil ist. 

Die gemessenen Lösungszusammensetzungen wurden verwendet um die Löslichkeits- 

produkte der Festphasen bei 20° C und 50° C und die thermodynamischen Parameter 

unter Standardbedingungn (25° C, 1 atm) zu berechnen. Das Löslichkeitsprodukt von Fe- 

Fr ist vergleichbar mit demjenigen der entsprechenden Al Phase (Al-Fr) während die 

Löslichkeits-produkte von Fe-Mc und Fe-Ms etwa 3 Grössenordnungen tiefer liegen als 

diejenigen von Al-Mc und Al-Ms. Dies deutet darauf hin, dass Fe-Friedel’s Salz in 

hydratisiertem Zement wahrscheinlich nicht stabil ist während Fe-haltige AFm Phasen 

sich bilden könnten. Das sehr tiefe Löslickeitsprodukt von Fe-Si-Hg (5-7 logarithmische 

Einheiten tiefer als dasjenige von Al-Si-Hg) impliziert, dass Fe-Si-Hg in hydratisiertem 

Zement eine stabile Phase ist. 

Im Weiteren wurden gemischte Al- und Fe-haltigen Hydrate synthetisiert um die 

Möglichkeit der Bildung von festen Lösungen (solid solution) zu untersuchen. Sowohl 

XRD Messungen an den Festphasen wie auch die thermodynamische Modellierung der 

Lösungszusam-mensetzung zeigen, dass Al-/Fe-Monsulfat wie auch Al-/Fe-Friedel’s Salz 

feste Lösungen mit und ohne Mischungslücke bilden während die Al-/Fe-Monokarbonate 

aufgrund der unterschiedlichen Kristallstrukturen (Al-Mc: monoklinisch, Fe-Mc: 

rhomboedrisch) als zwei separate Hydrate existieren. Möglicherweise findet auch die 

Bildung einer festen Lösung bei Al-/Fe-Si-Hydrogranat statt. 

Die Speziation von Fe wurde mittels EXAFS Spektroskopie in hydratisiertem Zement 

bestimmt um festzustellen, ob die Resultate aus den Untersuchungen mit Einzelphasen 

auch auf reale Zementsysteme übertragen werden können. Durch Faktoranalyse 

(principal component analysis, iterative target transformation) und Linearkombination 

v

konnten die Fe-haltigen Hydrate im Zementstein identifiziert und deren Anteile 

quantifiziert werden. Die Resultate zeigen, dass bereits nach einem Tag Hydratisierung 

des Zements mehrere Fe Spezies zum EXAFS Spektrum von Zementstein, das an der Fe 

K-Kante bestimmt wurde, beitragen. Während Ferrit die dominierende Fe Spezies im 

unhydratisierten Zement ist, erfolgte bereits kurz nach Beginn des 

Hydratisierungsprozesses die Bildung von Fe-Hydroxid. Mit der Zeit wurde die Bildung 

von stabilem Al/Fe-Si-Hydrogranat beobachtet, während die Anteile von Ferrit und Fe- 

Hydroxid langsam abnahmen. Diese Beobachtungen stehen im Einklang mit der 

thermodynamischen Modellierung, welche die Bildung von stabilem Al/Fe-Si- 

Hydrogranat in Zementstein voraussagt. 

Die Bestimmung der Löslichkeitsprodukte der einzelnen Hydratphasen ermöglicht es die 

bestehende, thermodynamische Datenbasis für Zementmineralien zu erweitern und eine 

quantitative Beurteilung, ob und in welchem Ausmass Fe-haltige Hydrate in frischen und 

gealterten Zementsystemen stabil sind, vorzunehmen. Die Resultate aus dieser Studie zur 

Speziation von Fe in Zement sind für ein besseres Verständnis der Wechselwirkung von 

Wasser und Zement wichtig und damit für die Beurteilung der Dauerhaftigkeit von 

zementartigen Materialien von grosser Bedeutung. 

Stichworte: Eisenhaltige Hydratphasen, Löslichkeitsprodukt, Feste Lösungen, 

Krystallstruktur; thermodynamische Modellierung; thermodynamische Daten 

vi

ACKNOWLEDGEMENT 

First I would like to thank Swiss National Foundation (SNF) for financial assistance 

during my study. I owe my deepest gratitude to my thesis supervisor and director Dr. 

Barbara Lothenbach, EMPA for excellent supervision. The thesis would not have been 

possible without the assistance and guidance of her. I will never find words to thank her 

for sharing her time. All you have done for me as a supervisor and as a friend are 

unforgettable. 

I would like to express my sincere acknowledgement to my thesis director Prof. Karen 

Scrivener for all the supervision, supports and valuable discussion throughout my thesis. 

It has been a great pleasure to be a member of her team and work together with her. 

It is with immense gratitude that I acknowledge my co-supervisor Dr. Erich Wieland, PSI 

Switzerland for teaching and guiding me during my study in particular on EXAFS 

techniques. He has made available his support all the time. 

It gives me great pleasure in acknowledging the support and help of Dr. Guillaume 

Renaudin and Dr. Adel Mesbah for synchrotron X-ray diffraction and Raman 

measurements and data analysis. I would like to thank many people who have helped me 

through the completion of this dissertation: Dr. Rainer Daehn for his assistance on XAS, 

Dr. Adrian Wichser for ICP measurement, Dr. Gwenn La Saout for Rietveld refinement, 

Dr. Mohsen Ben haha and Florian Deschner for assisting me on the SEM measurement, 

Dr. Frank Winnefeld for all the valuable discussions. Dr.Göril Möschner preparation for 

old samples and Dr. Konstantin Rozov for Fe-hydrotalcite sample. 

vii

I owe to thank all the laboratory technicians at EMPA particularly Luigi Brunetti, Boris 

Ingold and Angela Steffen for helping me in the laboratory. I would like to thank all my 

members of lab 135. who have given me love and respect thought my study. All the good 

times with friends (Wolfi, Walti, Lucy, Flo, Laura) are memorable. 

Many thanks to Trindler family for their care during my stay in Switzerland. My 

gratitude goes to my Ethiopian friends living in Switzerland who has given me their 

encouragement, care and affection during my study. 

Most especially I am grateful to my family particularly my mother Tiruwork and my 

father Zeleke for giving me all their cares and love throughout my life. Their support and 

inspiration as a parent from my childhood up to now is immense. This is for you. 

Last but not least this work is not possible without the support, love, care of my wife 

Fasika. Thanks to the almighty God for all you have done for me. 

viii

LIST OF TABLES 

Table 1 Oxide and the phase composition of the cements used. ...................................... 17 

Table 2 The compositions of the synthetic Fe-cement mixes with varying gypsum 

(CsH2) and calcite (Cc) content. ............................................................................... 18 

Table 3 Reference reactions used to estimate unknown heat capacities of cement 

minerals. .................................................................................................................... 25 

Table 4 Dissolution reaction used for thermodynamic calculation. ................................. 29 

Table 5 Thermodynamic data at standard conditions (298 K, 1 atm) used for the 

calculation of the liquid phase compositions and for computation of 

thermodynamic parameters for the synthesized solids. ............................................ 30 

Table 6 Multi pattern refinement (from two sample-to-detector distances: 1/ 150 mm, 

and 2/ 350 mm) and crystal data of Fe-Mc. .............................................................. 43 

Table 7 Fractional coordinate of non hydrogen atoms and isotropic displacement. ........ 45 

Table 8 Selected interatomic distances (Å) in Fe-Mc. ...................................................... 45 

Table 9 EXAFS structural parameters of Fe-Mc equilibrated for three years. ................. 48 

Table 10 IR vibrations of Ca4[(AlxFe1-x)2(OH)12] . CO3 . nH2O. .......................................... 53 

Table 11 Measured ion concentrations and calculated solubility products at different 

equilibration times. ................................................................................................... 57 

Table 12 Compositions of Al/Fe-monocarbonate after synthesis at 20 °C equilibrated 

for 3 years at supersaturated and undersaturated condition. ..................................... 58 

Table 13 Thermodynamic parameters of carbonate containing AFm phases at 

standard conditions (25°C, 1 atm). ........................................................................... 61 

Table 14 Quantitative phases analyses from Rietveld refinement. ................................... 71 

Table 15 Refined structural parameters of Fe-monosulfate (standard deviation in 

parentheses). .............................................................................................................. 71 

Table 16 Refined interatomic distances in Fe-monosulfate (standard deviation is 

given in parentheses). ................................................................................................ 74 

Table 17 Measured ion concentrations and calculated solubility products at different 


Table 18 Thermodynamic parameters of Fe-monosulfate at standard conditions 

(25°C, 1 atm). ............................................................................................................ 83 

i

Table 19 Compositions of Al/Fe-monosulfate after synthesis at 20°C equilibrated for 

680 days in supersaturated condition. ....................................................................... 85 

Table 20 Solubility products of all the solids formed during the synthesis of Al/Femonosulfate 

solid solution series at 20 °C equilibrated for 680 days under 

supersaturated condition. .......................................................................................... 87 

Table 21 Refined structural parameters of 3CaO.Fe2O3.CaCl2.10H2O (standard 

deviation is given in parentheses). ............................................................................ 97 

Table 22 Measured ion concentrations and calculated solubility products at 20 °C and 

sampled after different equilibration times synthesized from FeCl3.6H2O and 

CaO in 0.1M K OH. ................................................................................................ 102 

Table 23 Measured ion concentrations and calculated solubility products at 20 and 

50°C and sampled after different equilibration times synthesized from C2F, 

CaCl2.2H2O and CaO in distilled water and in 0.1 M KOH. .................................. 102 

Table 24 Thermodynamic parameters of Friedel’s salt at standard conditions (25 °C, 1 

atm). ........................................................................................................................ 104 

Table 25 Compositions of Al/Fe-Friedel’s salt synthesized at 20°C and equilibrated 

for 270 days under supersaturated condition. ......................................................... 105 

Table 26 Measured ion concentrations at different equilibration times in 0.1 M KOH . 114 

Table 27 Thermodynamic parameters at standard conditions determined in this study 

(25°C, 1 atm). .......................................................................................................... 116 

Table 28 Measured concentration of mixed C3AH6-C3FH6 systems equilibrated for 

three years ............................................................................................................... 121 

Table 29 Measured ion concentrations in the solution of solids synthesized at 110°C 

and re dissolved and equilibrated for 4 months at 20 °C and 50 °C. ...................... 126 

Table 30 Refined structure parameters of Fe siliceous hydrogarnet (standards 

deviation are indicated in parentheses). .................................................................. 133 

Table 31 Measured ion concentrations of solids synthesized at 110 °C (re dissolved 

and equilibrated for 4 months at 20 °C and 50 °C) and at 20 °C(equilibrated for 3 

years under oversaturated condition). ..................................................................... 134 

Table 32 Measured ion concentration of Ca3(AlxFe1-x)2(SiO4)(OH)8 equilibrated for 

four months from dissolution (undersaturation) experiment. ................................. 140 

Table 33 Summary of the results obtained in chapter 3 and comparison with their Alanalogues. 

................................................................................................................ 146 

Table 34 Relative weights of Fe-containing phases in hydrated OPC at 20 °C and 50 

°C obtained from LC fitting. ................................................................................... 165 

ii

Table 35 Relative weights of Fe-containing phases in HS hydrated at 20 °C and 50 °C 

obtained from LC fitting. ........................................................................................ 166 

iii

LIST OF FIGURES 

Fig. 1 Calculated volume changes during the hydration of OPC. ...................................... 5 

Fig. 2 Calculated (lines) and measured (dots) composition of the liquid phase of 

ordinary Portland cement during hydration [6]. ......................................................... 5 

Fig. 3 A sample of edges and the corresponding electronic transitions [25]. ..................... 8 

Fig. 4 Sample XAS Spectrum of FeO with XANES and EXAFS region [26]. .................. 9 

Fig. 5 Time-dependent XRD pattern of Fe-Hc (and Fe-Mc) synthesized at 20 °C; C2F: 

2CaOFe2O3, Fe-Mc: Fe-monocarbonate, Fe-Hc: Fe-hemicarbonate. ...................... 38 

Fig. 6 TGA and DTG curves of Fe-Hc formation at 20 °C for different equilibration 

times. CH: Portlandite, C: carbonates. ...................................................................... 38 

Fig. 7 Time-dependent XRD pattern of Fe-Mc formed at 20 °C. * unidentified ............. 40 

Fig. 8. TGA and DTG curves of Fe-Mc formation at 20 °C for different equilibration 

times. CH: Portlandite, C: carbonates. ...................................................................... 41 

Fig. 9 Comparison of XRD pattern of Fe-Mc equilibrated for one year at 20, 50 and 

80 °C. CH: portlandite, C: carbonate, Fe2O3: hematite. ............................................ 42 

Fig. 10 Rietveld plot from powder pattern recorded with a sample-to-detector distance 

of 150 mm (red crosses are experimental data, black line is calculated pattern, 

blue line is the difference pattern, green sticks are Bragg peaks positions for Fe- 

Mc and calcite). ......................................................................................................... 44 

Fig. 11a. Projection of the Fe-Mc structure along b axis (the interlayer part of the 

structure is ordered for clarity; i.e. the statistical distribution between one 

carbonate and two water molecule has been alternatively ordered). b. 3D 

cohesion in Fe-Mc structure (representation of the main hydrogen bonds). ............ 46 

Fig. 12. Fe K-edge EXAFS data of Fe-Mc: Experimental (solid line) and theoretical 

(dots) Fourier transform (modulus) obtained from k 3 -weighted, normalized, 

background-subtracted spectrum (inset). .................................................................. 47 

Fig. 13 a) Raman spectra on Fe-Mc in the frequencies range 200 cm -1 – 1800 cm -1 b) 

Raman spectra on Fe-Mc in the frequencies range 2800 cm -1 – 4000 cm -1 . ............. 49 

Fig. 14 SEM micrographs of Fe-Mc. ................................................................................ 50 

Fig. 15 Thermal analysis (DTG and TGA) of Ca3(AlxFe1-x)2O3.CaCO3.nH2O. ............... 51 

Fig. 16 IR spectra of Al-Mc and Fe-Mc. .......................................................................... 52 

iv

Fig. 17 XRD pattern of the Al/Fe-Mc after 3 years hydration time at 20 °C * peak due 

to additional water in Mc. ......................................................................................... 54 

Fig. 18 Layer thickness for Al-Mc and Fe-Mc after refinement by Le Bail fitting and 

Rietveld analysis. C4FcH12: Fe-Mc, C4AcH11: Al-Mc. ............................................. 55 

Fig. 19 Values of a-parameters for Al-Mc and Fe-Mc. .................................................... 55 

Fig. 20. Calculated solubility products of Fe-Mc and Fe-Hc from the solubility 

experiments. Squares: experimental solubility product of Fe-Hc, Triangles: 

experimental solubility product of Fe-Mc. ............................................................... 60 

Fig. 21 Measured (symbols) and calculated (lines) concentrations in the liquid phases 

of the synthesized monocarbonate at different Al/Al+Fe ratios. .............................. 62 

Fig. 22 Changes in the total volume of phases of a hydrated model mixture consisting 

of Al2O3, Fe2O3 and a fixed SO3/(Al,Fe)2O3 ratio of 1 as a function of the calcite 

content (CO2/(Al,Fe)2O3 ratio) at 20 °C at constant amount of solids: (Al2O3 + 

Fe2O3 + CaSO4 + CaO + CaCO3). ............................................................................ 63 

Fig. 23 XRD pattern of C4FsH12 formed at 20 °C after different equilibration times. ..... 68 

Fig. 24 TGA and DTG curves of C4FsH12 formation at 20°C after different 


Fig. 25 XRD pattern of C4FsH12 equilibrated for 360 days at 20, 50 and 80 °C. ............. 70 

Fig. 26. Rietveld plot for Fe-monosulfate samples (synthesized at 20 °C: top, and at 

50 °C: bottom) with = 1.5418Å. ............................................................................ 72 

Fig. 27 Details of the Rietveld plot from the sample Fe-Ms-50 °C. ................................. 73 

Fig. 28 Spectral range 100 cm -1 – 1500 cm -1 of Raman spectra from sample Fe-Ms-20 

°C (comparison with Al-monosulfate spectra [79]). ................................................. 75 

Fig. 29 Spectral range 2800 cm -1 – 4000 cm -1 of Raman spectra from sample Fe-Ms- 

50 °C (comparison with Al-monosulfate spectra [79]). ............................................ 76 

Fig. 30 Thermal analysis (TGA and DTG) of Al and Fe-monosulfate after 680 days. .... 77 

Fig. 31 XRD pattern of the C4AsH12-C4FsH12 series after 680 days equilibration at 

20 °C. Al-monosulfate (2θ = 19.89°), Fe-monosulfate (2θ = 19.99°) and *Alettringite. 

# Al-monocarbonate (2θ = 23.50°). .......................................................... 78 

Fig. 32 Layer thickness observed for the C4(A,F)sH12 solid solution in this study and 

reported in literature [12, 87, 88]. ............................................................................. 79 

Fig. 33 Values of a-parameters for the C4(A,F)sH12 solid solution series determined in 

this study and reported in literature [12, 87, 88]. ...................................................... 79 

v

Fig. 34 Calculated solubility products of Fe-monosulfate from the solubility 

experiments. symbols: experimental data. ................................................................ 84 

Fig. 35 Lippmann diagram illustrating the total solubility products of Al/Femonosulfate 

solid solution series: total experimentally determined solubility 

product (symbols), modeled total solubility products assuming ideal solid 

solution (dashed lines), modeled total solubility products assuming a non-ideal 

solid solution with a miscibility gap (a0= 1.26 and a1= 1.57) (solid lines) and 

solubility products assuming no solid solution (dotted lines). X-axis: Al/(Al + 

Fe) ratios in the solid and Al/(Al + Fe) ratios in the liquid. ...................................... 89 

Fig. 36 Measured (points) and calculated (lines) concentrations in the liquid phases of 

the synthesized monosulfate with different Al/(Al+Fe) mole ratio, assuming a 

continuous solid solution with a miscibility gap. ...................................................... 90 

Fig. 37 XRD pattern of 3CaO . Fe2O3 . CaCl2 . 10H2O (Fe-Fr) synthesized at 20 °C and 

sampled after different equilibration times from FeCl3.6H2O and CaO in 0.1 M 

KOH. ......................................................................................................................... 93 

Fig. 38 Comparison of the XRD patterns of Fe-Friedel’s salts equilibrated for three 

years at different pH values: synthesized a). FeCl3.6H2O and CaO in 0.1M KOH 

(pH = 11.94), b). C2F, CaCl2.2H2O and CaO in distilled water (pH = 12.39) and 

c). C2F, CaCl2.2H2O, and CaO in 0.1 M KOH (pH = 12.84), CH-portlandite. ........ 94 

Fig. 39 TGA-DTG curves of Fe-Friedel’s salt synthesized at 20 °C and sampled after 

different equilibration times from FeCl3.6H2O and CaO in 0.1M KOH. ................. 95 

Fig. 40 Rietveld plot for Fe-Friedel’s salt recorded at = 0.697751 Å and at a 

sample-to-detector distance of 150 mm. ................................................................... 96 

Fig. 41 Thermal analysis (TGA and DTG) of Al and Fe-Friedel’s salt synthesized 

from FeCl3.6H2O and CaO in 0.1 M KOH and equilibrated for 270 days. .............. 98 

Fig. 42 Raman spectra recorded for Fe-Friedel’s salt crystal. .......................................... 99 

Fig. 43 Values of a-parameters for the Al/Fe-Friedel’ salt solid solution determined in 

this study compared to the findings by Kuzel et al. [88], Goetz Neunhoeffer et al. 

[100], Rapin et al. [99] and Rousselot et al. [96]. ................................................... 100 

Fig. 44 Experimental determined solubility products of Fe-Friedel’s salt as a function 

of pH. ...................................................................................................................... 103 

Fig. 45 Calculated solubility products of Fe-Friedel’s salt from the solubility 

experiments compared with the solubility product Al-Friedel’s salt calculated 

from measured concentrations reported in literature [54, 94, 101-103]. ................ 104 

Fig. 46 Measured (points) and calculated (lines) concentrations in the liquid phases of 

the synthesized Friedel’s salt at different Al/(Al+ Fe) ratio, assuming ideal solid 

solution. ................................................................................................................... 106 

vi

Fig. 47 Nomenclature of minerals of the hydrogarnet group. ......................................... 110 

Fig. 48 Time-dependent XRD pattern of C3AH6 synthesized at 20 °C, * C4AcH11. ...... 112 

Fig. 49 Thermal analysis (TGA and DTG) of C3AH6 and C3FH6 synthesized at 20 °C 

and sampled after different equilibration times. ..................................................... 113 

Fig. 50 Solubility products of C3AH6 calculated from the solubility experiments 

carried out in this study and from different published data [7, 111, 112, 114- 

116]. ........................................................................................................................ 115 

Fig. 51 Time-dependent XRD pattern of C3FH6 synthesized at 20 °C and the sample 

synthesized at 110 °C and equilibrated for 5 days. ................................................. 118 

Fig. 52 XRD pattern of mixed Al and Fe hydrogarnets after 3 years equilibration. ...... 120 

Fig. 53 Calculated solids in the CaO-Al2O3-Fe2O3-H2O system in 0.1 M KOH using 

the solubility products as given in Table 28. .......................................................... 122 

Fig. 54 The XRD pattern of Al containing Si-hydrogarnet synthesized at 20 °C and 

110 °C. * Al-Si-hydrogarnet with two different compositions (see inlet); o 

C3AH6; - KNO3 present as impurity; +CaF2 added as an internal standard. ........... 123 

Fig. 55 Thermal analysis (TGA and DTG) of Al-and Fe-Si hydrogarnet synthesized at 

20 °C and 110 °C. The circle region indicates the water loss of hydrogarnets 

with different compositions. ................................................................................... 124 

Fig. 56 Estimation of the silica content for synthesized Al-containing hydrogarnet; 

PDF: Powder Diffraction File. ICSD: Inorganic Crystal Structure Database. The 

composition of the synthesized solid solution series was estimated from the unit 

cell size as indicated by the line. ............................................................................. 125 

Fig. 57 Comparison of published solubility products of Al-Si-hydrogarnet calculated 

in this study from the data reported in [7, 110, 112, 117], C3AH6 (dashed line), 

C3AS0.41H5.18 (solid line), C3AS0.84H4.32 (dotted line).............................................. 128 

Fig. 58 Solubility products as a function of Si content between C3AH6 and C3AS3 end 

members at 25 °C. ................................................................................................... 129 

Fig. 59 Time-dependent XRD pattern of C3FSH4 synthesized at 20 °C from C2F, * the 

solid synthesized at 110 °C. R: rutile. ..................................................................... 130 

Fig. 60 Rietveld plot for Fe-Si-Hydrogarnet sample with = 0.697751Å and a 

sample-to-detector distance of 150 mm (top) and 400 mm (bottom). .................... 132 

Fig. 61 Zoom of the Rietveld plot from pattern recorded for a sample-to-detector 

distance of 400 mm showing the two hydrogarnet phases (systematic shoulders, 

right side, for hydrogarnet diffraction peaks). ........................................................ 133 

vii

Fig. 62 Calculated solubility products of Fe-Si-hydrogarnets from the solubility 

experiments. (lines show calculated values, full symbols show the measured 

values from undersaturation and empty symbols from oversaturation). ................. 135 

Fig. 63 XRD pattern of the solid solution series of Ca3Fe2(SiO4)3-y(OH)4y, + CH. The 

dotted lines indicate the peak shifts. *Main reflections of the hydroandradite end 

members. Note that Xsi = y = 3 ............................................................................... 136 

Fig. 64 Solubility products as a function of Si content in between C3FH6 and C3FS3 

end members at 25 °C. The dotted line connects the solubility products of C3FH6 

and C3FS3. ............................................................................................................... 138 

Fig. 65 XRD pattern of the solid solution series of Ca3(AlxFe1-x)2(SiO4)(OH)8 

synthesized at 110 °C. ............................................................................................. 139 

Fig. 66 Lippmann diagram illustrating the total solubility products of Al/Fe-siliceous 

hydrogarnet solid solution series Ca3(AlxFe1-x)2(SiO4)0.9(OH)8.4 at a) 20 °C b) 50 

°C: .experimentally determined total solubility products (filled symbols), 

modeled total solubility products assuming ideal solid solution (dashed lines). In 

addition also the solubility product of C3AS0.84H4.32 and C3FS0.95H4.1 derived 

from the experimental data (empty symbols) and the solubility products 

assuming no solid solution (dotted lines) are given. X-axis: Al/(Al + Fe) ratio in 

the solid or liquid phases, respectively. .................................................................. 143 

Fig. 67 XRD patterns of OPC (+) and HS (*) cements hydrated at 20 °C. .................... 149 

Fig. 68 TGA-DTG curves of OPC (+) and HS (*) cements hydrated at 20 °C. ............. 150 

Fig. 69 XRD patterns of OPC (+) and HS (*) cement hydrated for 1 year at 50 °C: 

The XRD peak at 2θ ~ 11.30 is between the monosulfate and monocarbonate 

peaks. ...................................................................................................................... 151 

Fig. 70 TGA-DTG curves of OPC (+) and HS (*) cement hydrated for 1 year at 50 °C 152 

Fig. 71 XRD patterns of OPC (+) and HS (*) hydrated at 20 °C after selective 

dissolution using SAM. Note that the samples suffered from carbonation during 

SAM extraction. ...................................................................................................... 153 

Fig. 72 TGA-DTG curves of OPC (+) and HS (*) hydrated at 20 °C after selective 

dissolution with SAM. ............................................................................................ 154 

Fig. 73 XRD patterns of OPC (+) and HS (*) hydrated at 50 °C for 150 days after 

selective dissolution with SAM. ............................................................................. 154 

Fig. 74 SEM/EDX of hydrated OPC for 2 years at (a) 20 °C and (b) 50 °C after 


Fig. 75 Atomic ratio of hydrated OPC for 2 years at (a) 20 °C and (b) 50 °C after 


viii

Fig. 76 Fe K-edge XANES spectra of Fe-containing hydrates. The broken lines 

indicate the position of related spectral features. .................................................... 159 

Fig. 77 k 3 -weigthed experimental bulk-EXAFS spectra of Fe-containing phases used 

as reference compounds. The broken lines indicate the position of related 

spectral features. ..................................................................................................... 160 

Fig. 78 EXAFS spectra of hydrated OPC at 20 °C and at different ages (line: 

experimental data; dots: modelled data). The broken lines outline selected 


Fig. 79 EXAFS spectra of hydrated OPC at 50 °C and at different ages (line: 

experimental data; dots: modeled data). The broken lines outline selected 


Fig. 80 EXAFS spectra of hydrated HS at 20 °C and at different ages (line: 



Fig. 81 EXAFS spectra of hydrated HS at 50 °C and at different ages (line: 



Fig. 82 Volume changes of hydrated phases at different hydration ages during 

hydration of OPC at room temperature. .................................................................. 168 

Fig. 83 Heat flow of the hydration of C2F and synthetic Fe-cement in the presence of 

different amounts of gypsum. ................................................................................. 172 

Fig. 84 XRD (above) and TGA-DTG (below) analysis of C2F and synthetic Fecement 

after 3 days of hydration in the presence of different amounts of gypsum 

*Fe-OH-AFm +unidentified. .................................................................................. 173 

Fig. 85 XRD (above) and TGA-DTG (below) analysis of C2F and synthetic Fecement 

after 3 months of hydration in the presence of different amounts of 

gypsum *Fe-AFm hydroxyl. ................................................................................... 175 

Fig. 86 Calculated phase diagram of thermodynamic stable hydrate assemblages of 

synthetic Fe-cement with different amounts of gypsum. ........................................ 176 

Fig. 87 Conduction calorimeter curve of the hydration of synthetic Fe-cement with 

different amounts of gypsum and calcite. ............................................................... 177 

Fig. 88 XRD (above) and TGA-DTG (below) analysis of synthetic Fe-cement after 3 

months of hydration with different amounts of gypsum and calcite ...................... 179 

Fig. 89 Calculated phase diagram of thermodynamic stable hydrate assemblages of 

Fe-synthetic cement with different amounts of gypsum and calcite. ...................... 180 

ix

Table of Contents 

TABLE OF CONTENTS 

Abstract ................................................................................................................................ i 

Zusammenfassung .............................................................................................................. iv 

Acknowledgement ............................................................................................................. vii 

List of Tables ........................................................................................................................ i 

List of Figures ...................................................................................................................... iv 

1 INTRODUCTION ........................................................................................................... 1 

1.1 Ordinary Portland cement (OPC) ............................................................................ 1 

1.2 Cement hydration and thermodynamic modeling ................................................. 1 

1.3 The fate of iron oxides during the hydration of cements ....................................... 6 

1.4 Characterization of cementitious system ............................................................... 7 

1.4.1 Standard analytical techniques ....................................................................... 7 

1.4.2 X‐ray absorption spectroscopy (XAS) .............................................................. 7 

1.5 Objective of this study .......................................................................................... 10 

1.6 Outline of the thesis .............................................................................................. 11 

2 MATERIALS AND METHODS ...................................................................................... 12 

2.1. Synthesis of Fe‐containing phases .................................................................... 12 

2.1.1. Fe‐hemicarbonate and Fe/Al‐monocarbonate ............................................. 12 

2.1.2. Fe‐monosulfate ............................................................................................. 13 

2.1.3. Fe‐Friedel’s Salt ............................................................................................. 14 

i


2.1.4. Fe‐strätlingite ................................................................................................ 14 

2.1.5. Synthesis of hydrogarnets ............................................................................ 14 

2.1.5.1. Silica free hydrogarnets: Ca3(AlxFe1‐x)2(OH)12 ....................................... 14 

2.1.5.2. Siliceous hydrogarnets: Ca3(AlxFe1‐x)2(SiO4)(OH)8 ................................. 15 

2.1.6. Hydrated cement samples ............................................................................ 16 

2.1.7. Synthesis of synthetic Fe‐cement ................................................................. 17 

2.2. Analytical methods ........................................................................................... 18 

2.2.1. Powder X‐ray diffraction ............................................................................... 18 

2.2.2. Synchrotron powder diffraction ................................................................... 19 

2.2.3. Thermogravimetric analysis .......................................................................... 19 

2.2.4. Vibrational spectroscopy (Raman and Infrared spectroscopy) .................... 20 

2.2.5. Scanning electron microscopy (SEM) ........................................................... 20 

2.2.6. Liquid phase analysis .................................................................................... 20 

2.2.7. Selective dissolution ..................................................................................... 21 

2.2.8. Synchrotron‐based X‐ray absorption spectroscopy (XAS) ............................ 22 

2.2.8.1. Data collection and reduction............................................................... 22 

2.2.8.2. Data analysis and fitting ........................................................................ 23 

2.2.9. Calorimetry ................................................................................................... 24 

2.3. Thermodynamic modeling ................................................................................ 24 

2.3.1. Estimation of heat capacity of Fe‐containing phases ................................... 25 

2.3.2. Determination of solubility products ............................................................ 25 

2.3.3. Thermodynamics of solid solutions .............................................................. 31 

2.3.4. Thermodynamic modeling of cement hydration .......................................... 34 

ii


3. SYNTHETIC FE‐CONTAINING HYDRATES ................................................................... 35 

3.1. Iron containing carbonate AFm phases ............................................................ 35 

3.1.1. Introduction .................................................................................................. 35 

3.1.2. Fe‐hemicarbonate ......................................................................................... 37 

3.1.3. Fe‐monocarbonate ....................................................................................... 40 

3.1.3.1. Kinetics of formation ............................................................................ 40 

3.1.3.2. Effect of temperature ........................................................................... 42 

3.1.3.3. Structure of Fe‐Mc ................................................................................ 43 

3.1.3.4. Comparison of pure Fe‐ and Al‐Mc ....................................................... 50 

3.1.4. Mixed CaO.(AlxFe1‐x)2O3.CaCO3.nH2O systems .............................................. 53 

3.1.5. Solubility ........................................................................................................ 55 

3.1.5.1. Determination of solubility products at 20 °C and 50 °C ...................... 56 

3.1.5.2. Estimation of the solubility product under standard conditions ......... 59 

3.1.5.3. Modeling of mixed CaO(AlxFe1‐x)2O3CaCO3nH2O systems ................. 61 

3.1.6. Modeling of C3A‐C2F‐CaCO3‐CaSO4‐H2O system in cement hydration ......... 62 

3.1.7. Conclusions ................................................................................................... 64 

3.2. Fe‐containing monosulfate ............................................................................... 67 

3.2.1. Introduction .................................................................................................. 67 

3.2.2. Kinetics of formation .................................................................................... 67 

3.2.3. Effects of temperature .................................................................................. 69 

3.2.4. Structure of C4FsH12 ...................................................................................... 70 

3.2.5. Comparison of C4AsH12 with C4FsH12 ............................................................ 74 

3.2.6. Solid solution between Al and Fe‐monosulfate (C4(A,F)sH12) ...................... 77 

iii


3.2.7. Solubility of Al/Fe‐monosulfate .................................................................... 80 

3.2.7.1. Determination of solubility products at 20, 50 and 80 °C .................... 80 

3.2.7.2. Determination of solubility products under standard condition ......... 82 

3.2.7.3. Determination of solubility product of the solid solution and modeling 

of the liquid phase ................................................................................................ 84 

3.2.8. Conclusions ................................................................................................... 91 

3.3. Fe‐Friedel’s salt (3CaO . Fe2O3 . CaCl2 . 10H2O) ....................................................... 92 

3.3.1. Introductions ................................................................................................. 92 

3.3.2. Kinetics of formation .................................................................................... 92 

3.3.3. Structure of Fe‐Friedel’s salt ......................................................................... 95 

3.3.4. Comparison of Al‐Friedel’s salt and Fe‐Friedel’s .......................................... 97 

3.3.5. Solid solution between Al and Fe‐Friedel’s salt (3CaO(AlxFe1‐x)2CaCl2.10H2O .. 

....................................................................................................................... 99 

3.3.6. Solubility ...................................................................................................... 100 

3.3.6.1. Solubility of Fe‐Friedel’s salt ............................................................... 100 

3.3.6.2. Determination of the solubility products of the solid solution and 

modeling of the liquid phase .............................................................................. 105 

3.3.7. Conclusions ................................................................................................. 106 

3.4. Fe‐strätlingite .................................................................................................. 108 

3.5. Hydrogarnets .................................................................................................. 109 

3.5.1. Introduction ................................................................................................ 109 

3.5.2. Al‐Katoite, C3AH6 ......................................................................................... 111 

3.5.3. Fe‐Katoite, C3FH6 ......................................................................................... 117 

iv


3.5.4. Solid solution between aluminum and iron katoite, C3(A,F)H6 .................. 119 

3.5.5. Aluminum siliceous hydrogarnet, C3ASH4 ................................................... 122 

3.5.6. Iron siliceous hydrogarnet, C3FSH4 ............................................................. 129 

3.5.7. Solid solution between Ca3Fe2(OH)12 and Ca3Fe2O6(SiO2)3 (hydroandradite, 

Ca3Fe2(SiO4)3‐y(OH)4y) .............................................................................................. 136 

3.5.8. Solid solution between aluminum and iron siliceous hydrogarnet, C3(A,F)SH4 

..................................................................................................................... 138 

3.5.9. Conclusions ................................................................................................. 143 

3.6. Summary ......................................................................................................... 145 

4. FE‐CONTAINING HYDRATES IN HYDRATED CEMENT .............................................. 147 

4.1. Identification of Fe‐containing hydrates in hydrated cement ........................ 147 

4.1.1. Introduction ................................................................................................ 147 

4.1.2. Characterization of hydrated cement using standard analytical techniques ... 

..................................................................................................................... 149 

4.1.3. Spectroscopic investigation ........................................................................ 158 

4.1.3.1. XANES and EXAFS spectra of Fe‐containing reference compounds ... 158 

4.1.3.2. Identification of Fe‐containing hydrates ............................................ 160 

4.1.4. Thermodynamic modeling .......................................................................... 167 

4.1.5. Conclusions ................................................................................................. 169 

4.2. Synthetic Fe‐cement ....................................................................................... 171 

4.2.1. Introduction ................................................................................................ 171 

4.2.2. Effects of gypsum on the of hydration of synthetic Fe‐cement ................. 172 

4.2.3. Effects of calcite on the of hydration of synthetic Fe‐cement ................... 177 

v


4.2.4. Conclusions ................................................................................................. 180 

5. GENERAL CONCLUSION AND OUTLOOK ................................................................. 182 

5.1. General conclusion ......................................................................................... 182 

5.2. Outlook ........................................................................................................... 186 

ABBREVATIONS ............................................................................................................... 188 

APPENDIX ........................................................................................................................ 191 

Appendix A: Additional fitted structural parameters ................................................. 191 

Appendix B: Additional figures ................................................................................... 192 

REFERENCES .................................................................................................................... 197 

vi

1 INTRODUCTION 

CHAPTER 1 INTRODUCTION 

In this chapter a general overview about the hydration of Portland cements, different 

characterization techniques, thermodynamic modeling and the reaction of ferrite phases is 

given. Moreover, the objective of the study is briefly explained. 

1.1 Ordinary Portland cement (OPC) 

The raw materials for Portland cement production are a mixture of limestone and clay 

minerals containing calcium oxide, silicon oxide, aluminum oxide, ferric oxide, and 

magnesium oxide. The raw materials are ground together in a raw mill and then heated in 

a cement kiln at a temperature between 1400-1500 °C which produces nodules of clinker. 

The clinker is mixed with a few percent of gypsum and finely ground to make cement. 

The clinker contains four major phases, called alite (Ca3SiO5 or C3S), belite (C2S or 

Ca2SiO4), aluminate (C3A or Ca3Al2O3) and ferrite (C2(A,F)). The formulas given are 

idealized, as all clinker phases contain in addition a number of minor elements [1]. Ferrite 

designates a solid solution series with the formula Ca2(AlxFe1-x)2O5 with 0 ≤ x < 0.7. It 

crystallizes in the orthorhombic crystal system, the unit-cell dimensions vary with the 

Al2O3/Fe2O3 ratio. In ferrite as present in cement clinker, a part of Fe 3+ can be replaced 

by Mg 2+ in combination with Si 4+ or Ti 4+ resulting in the typical clinker ferrite 

composition of approximately Ca2AlFe0.6Mg0.2Si0.15Ti0.05O5. 

1.2 Cement hydration and thermodynamic modeling 

Alite and belite constitute over 80 wt.% of most Portland cements. Alite is the most 

important phase for strength development during the first month, while C2S reacts much 

slowly and contributes rather to the long-term strength of the cement. Both the silicate 

1


phases react with water as shown below to form calcium hydroxide and calcium-silicate 

hydrate (C-S-H) with Ca/Si ratio of 1.5 to 1.9: 

C3S + 5.3H2O → C1.7SH4 + 1.3CH 

C2S + 4.3H2O → C1.7SH4 + 0.3CH 

Tricalcium aluminate (Ca3Al2O3) constitutes 5-10% of most Portland cement clinkers. In 

the absence of any additives, C3A reacts with water to form two intermediate hexagonal 

phases, C2AH8 and C4AH13. The structure of C2AH8 is not precisely known, but C4AH13 

has a layered structure based on the calcium hydroxide structure. All of the aluminum in 

C4AH13 is octahedral. C2AH8 and C4AH13 are metastable phases that transform with time 

into the thermodynamically more stable cubic phase C3AH6. 

2C3A + 21H2O → C2AH8 + C4AH13 → 2C3AH6 + 9H2O 

In the presence of gypsum, anhydrite or bassanite, C3A reacts slowly and forms Al- 

ettringite, which can convert to Al-monosulfate after the depletion of calcium sulfates 

and further Al-ettringite. 

C3A + 3CsH2 + 26H2O → C6As3H32 

2C3A + C6As3H32 + 4H2O → 3C4AsH12 

In the presence of carbonate, C3A forms Al-hemicarbonate or Al-monocarbonate, 

depending of the availability of calcite [2]. 

C3A + 0.5Cc + 0.5CH + 11.5H2O → C4Ac0.5H12 

C3A + Cc + 11H2O → C4AcH11 

The reaction of ferrite (C2(A,F)) is similar to the reactions of C3A though the presence of 

Fe makes it more complicated. The Al from C2(A,F) can form as discussed above Al- 

containing OH-AFm phases (C2AH8 and C4AH13) or Al-katoite (C3AH6). In the presence 

of gypsum or carbonate, monosulfate, hemicarbonate, monocarbonate and ettringite can 

2


be formed. The fate of Fe is unclear. If we assume partial substitution of Al by Fe, a solid 

solution can be formed as in 

C2(A,F) + 2CH + 19H2O → C2(A,F)H8 + C4(A,F)H13 → 2C3(A,F)H6 + 9H2O 

In the presence of gypsum, 

C2(A,F) + 6CsH2 + 2CH + 50H2O → 2C6(A,F)s3H32 

C2(A,F) + C6(A,F)s3H32 + 2CH + 2H2O → 3C4(A,F)sH12 

In the presence of calcite, 

C2(A,F) + Cc + 11H2O → C4(A,F)cH11 

The hydration of cement is far more complex than the sum of the hydration reactions of 

the individual minerals. The major constituents of OPC are alite, belite, aluminate and 

ferrite and in addition a number of other minerals such as calcium sulfates (gypsum, 

hemihydrate and/or anhydrite), calcite, calcium oxide, magnesium oxide, Na- and K- 

sulfates are usually present. In contact with water, the easily soluble solids in the cement, 

such as gypsum, alkali sulfate and calcium oxide, react until equilibrium with the pore 

solution is reached or they are dissolved completely. The clinker phases hydrate slowly 

and release continuously Ca, Si, Al, Fe and OH - to solution, which then precipitate as 

calcium silicate hydrates (C-S-H), ettringite or as other hydrate phases. The balance 

between dissolution rates of the clinker phases and precipitation rates of the secondary 

phases determines the amount of Ca, Al, Fe, Si, and OH - released and the rate of 

formation of C-S-H, ettringite and the other hydrates. 

Thermodynamic modeling of the interactions between solid and liquid phase in cements 

using geochemical speciation codes helps to provide a basis for the interpretation of the 

hydration process [3, 4]. Furthermore, it allows the composition of the hydrate 

3


assemblages to be predicted under different conditions (e.g., initial clinker composition, 

water-to-cement (w/c) ratios, etc.) and for longer time scales. 

Hydration models have been developed in the past years to quantify the composition of 

the solid phases and liquid phases in cementitious systems during hydration [3-6]. For 

OPC systems, thermodynamic modeling in combination with calculated hydration rates 

correctly predicts the depletion of gypsum within the first day of hydration (Fig. 1). In 

this phase a strong decrease of the sulfate concentration in the pore solution is observed 

as ettringite continues to precipitate (Fig. 2). The depletion of aqueous sulfate during the 

first day is compensated by the release of OH - to fulfill conditions of electroneutrality in 

solution, which gives rise to a significant increase in pH. This increase in pH decreases 

the Ca concentration constrained by the portlandite solubility (Fig. 2). After the first day 

the precipitation of ettringite (6Ca(OH)2·(AlxFe1-x)2(SO4)3·26H2O(s)) ends as gypsum is 

exhausted. Subsequently, calcium monocarbonate (3CaO·(AlxFe1- 

x)2O3·CaCO3·11H2O(s)) and hydrotalcite (Mg4Al2(OH)14·3H2O) start forming. Calcite is 

slowly consumed due to the formation of monocarbonate. After hydration time of one 

month and longer, the solid paste is mainly composed of C-S-H, portlandite, ettringite 

and monocarbonate. With the exception of ettringite, the amount of hydration products 

continues to slowly increase with time. 

4

cm 3 /100 g cement 

85 

80 

75 

70 

65 

60 

55 

50 

45 

40 

35 

30 

25 

20 

15 

10 

gypsum 

C AF 4 C3A C S 2 

C S 3 


ettringite 

monocarbonate 

C-S-H 

portlandite 

5 

0 

0.01 0.1 1 10 100 1000 

hydration time [days] 

pore solution 

Fig. 1 Calculated volume changes during the hydration of OPC. 

[mM] 

600 

500 

400 

300 

200 

100 

0 

CaSO 4,Ca(OH) 2 

C-S-H, ettringite, 

brucite 

Ca 

SO 4 

K 

K OH- 

Na SO4 

Ca Si 

C-S-H, Ca(OH) 2, 

ettringite, 

monocarbonate, 

hydrotalcite 

calcite 

hydrotalcite 

0.01 0.1 1 10 100 1000 10000 

time [hours] 

Fig. 2 Calculated (lines) and measured (dots) composition of the liquid phase of ordinary Portland 

cement during hydration [6]. 

Application of thermodynamic models requires that the thermodynamic data of the 

hydrates formed in OPC are known. In the past years a set of thermodynamic data for 

selected hydrates have been critically reviewed [6]. Additionally, the solubility of 

OH - 

Na 

Si 

5


numerous hydrates particularly Al-containing phases have been investigated 

experimentally between 5 to 85 °C, which served as a basis to extend the cement database 

[3, 7]. However, there is a lack of thermodynamic data on Fe-hydrates that limits 

thermodynamic modeling to predict the fate of Fe-during cement hydration. 

1.3 The fate of iron oxides during the hydration of cements 

Ferrite C2(A,F) (Ca4(Fex-1Alx)4O10) is an important clinker phase in Portland cements 

(5-15%). The rate at which it reacts with water appears to be somewhat variable perhaps 

due to differences in composition or other characteristics, but it reacts fast initially and 

much more slowly at later ages [8, 9]. 

In pure system, i.e. in the presence of Ca, Al, Fe, and sulfate or carbonate only, Fe- 

containing ettringite, monosulfate and monocarbonate were found to precipitate and to 

form solid solutions with their Al-containing analogues [10-17]. Further, the formation of 

an amorphous iron hydroxide phase was reported [16-20]. In the complex cement matrix, 

however, the situation appears to be unclear due to the presence of silica. It was 

suggested that Fe-containing siliceous hydrogarnets might form in cementitious systems 

[21-23]. Harchand et al. [24] found that no Fe(OH)3 was present in hydrated cements 

based on Mössbauer spectroscopy but they could not gain any further information from 

the spectra concerning the kind of Fe-containing hydrates formed. Whether and to what 

extent Al/Fe-ettringite, Al/Fe-monosulfates, Al/Fe-monocarbonate, amorphous Fe(OH)3 

or Al/Fe (siliceous) hydrogarnets, respectively, might form in Portland cement is poorly 

understood. 

6


1.4 Characterization of cementitious system 

1.4.1 Standard analytical techniques 

Commonly used analytical techniques to characterize cementitious system include XRD, 

TGA, microscopic techniques (SEM and TEM) and vibrational spectroscopies (IR and 

Raman). X-ray diffraction (XRD) is a key technique for characterizing the crystalline 

phase composition of materials. Moreover, it allows phase identification and provides 

information about crystal structure. However, it does not give sufficient information 

about poorly crystalline and amorphous phases. Thermogravimetric analysis (TGA) helps 

to characterize and identify phases from complex cement matrix based on the weight loss 

over a specific temperature range. The limitation of TGA is due to the difficulty of 

distinguishing different phases within the complex cement matrix which have the weight 

loss at the same temperature. Scanning electron microscopy (SEM) is used to study the 

microstructure of cement and cementitious materials and in combination with EDX 

(energy dispersive X-ray spectroscopy) to characterize the chemical composition of the 

different phases and their spatial distribution. The above standard analytical techniques 

cannot provide a clear identification of Fe-containing hydrates in hydrated cement as 

signals from Fe-containing phases overlap with its Al-analogues. 

1.4.2 X-ray absorption spectroscopy (XAS) 

X-ray absorption spectroscopy (XAS) is a technique used to obtain structural information 

of a compound. It is element specific and accounts for the local geometric and electronic 

structures. A synchrotron light source is used as the X-ray photon source. The energy is 

tuned to an energy at which the incident photon can excite a core electron of the 

absorbing atom to a continuum state. The electron is now considered a photoelectron and 

7


propagated as a spherical wave. The energy of this photoelectron is equal to the energy of 

the absorbed photon minus the binding energy of the electron to the atom. The energy at 

which these photoelectrons are absorbed is related to the edges seen in XAS, K, L and M 

which correspond to the particular electronic transitions (Fig. 3). 

Fig. 3 A sample of edges and the corresponding electronic transitions [25]. 

The number of X-ray photons that are transmitted through a sample (It) is equal to the 

number of X-ray photons shone on the sample (I0) multiplied by a decreasing exponential 

factor that depends on the absorption coefficient (μ) of the type of atoms in the sample 

and the thickness of the sample x. 

It = I0e 

– μx 

There are two main regions of the XAS spectrum providing structural information: 

XANES and EXAFS (Fig. 4). 

8


Fig. 4 Sample XAS Spectrum of FeO with XANES and EXAFS region [26]. 

The X-ray absorption near edge structure, XANES, is the part of the spectrum that gives 

qualitative data based on modeling and simulation. XANES is used to give information 

about the average oxidation state and coordination environment. By taking unknown 

spectra and fitting a linear combination of known reference spectra, one can get an 

estimate of the contribution of each reference to the unknown spectra. 

Extended X-ray absorption fine structure, EXAFS, is the part of spectrum that gives 

quantitative data on the local structure around the absorber atom. From EXAFS mainly 

information on the type of neighboring atoms, their distance from absorber atom (bond 

length), the number of neighboring atoms (coordination numbers) and ordering effects 

(Debye-Waller factor) can be extracted. As described above, the photoelectron can be 

thought of as a wave centered at an atom. The wave vector of the photoelectron is related 

to the difference in binding energy of the electron, E0, and the energy of the photon, E, as 

shown below: 

9

, k=2π/λ 


When this wave interacts with other atoms, there is either a destructive or constructive 

interference. The phase and amplitude of interference that occurs is related to the type 

and location of the incident atom. Therefore, analysis of EXAFS data allows structural 

information about the type of atom and its coordination environment to be determined. 

XANES and EXAFS techniques allow dilute samples to be examined (concentration of 

the X-ray absorber down to a few tens of ppm). Most importantly, XAS can be used to 

study amorphous solids, surface adsorbed complexes, or species in solution in addition to 

crystalline materials. There is growing interest in the application of this technique for 

quantification of species in a complex mixture [27]. 

Synchrotron-based X-ray absorption spectroscopy (XAS) can be used as a 

complementary technique to gain molecular-level information from cementitious systems 

[28-30]. Furthermore, advanced high resolution synchrotron-based X-ray micro-probe 

allows to obtain spatially resolved information on the speciation of the X-ray absorber of 

interest in compact matrices, such as cementitious materials [28, 30]. 

1.5 Objective of this study 

As discussed above, the fate of iron during cement hydration is poorly known. Moreover, 

experimentally determined thermodynamic solubility products and other thermodynamic 

parameters are lacking for Fe-hydrates. The general objectives of this study are the 

following: 

Synthesis and characterization of Fe-hydrates and investigation of their solid 

solution formation with the Al-analogous. Experimental determination of 

solubility products and other thermodynamic parameters of Fe-hydrates. 

10


Identification of Fe-hydrates in hydrated Portland cements using XAS technique. 

Thermodynamic modelling of Portland cement hydration including the newly 

determined thermodynamic data for iron phases and compare them to the 

experimental data in Portland cements and in synthetic Al-free cements. 

1.6 Outline of the thesis 

The thesis contains five chapters: 

Chapter 1: contains the introduction and the objective of the thesis. 

Chapter 2: presents the materials and methods used to study Fe-containing hydrates 

possibly present in cementitious system. It explains the procedures followed to synthesize 

pure Fe-containing phases and their solid solutions with Al. In the course of this chapter 

the analytical techniques used to characterize both the solid and the liquid phases are 

presented. Furthermore, the application of thermodynamics in the framework of this 

study is explained. 

Chapter 3: briefly presents the results obtained on formation of Fe-containing phases, 

their crystal structure, and formation of solid solution with their Al-analogues and 

determination of thermodynamic data. 

Chapter 4: describes identification of hydrated phases in cements particularly Fe- 

containing hydrates using EXAFS. It also presents thermodynamic modeling of Portland 

cement and hydration study of Al-free Fe-synthetic cement. 

Chapter 5: presents the general conclusions of the study and the outlook for future 

investigations. 

11

CHAPTER 2 MATERIALS AND METHODS 

2 MATERIALS AND METHODS 

2.1. Synthesis of Fe-containing phases 

C3A and C2F clinkers were used as starting materials for the synthesis. C3A and C2F were 

prepared by mixing appropriate amounts of CaCO3 with Al2O3 and Fe2O3 powders and 

burning at 1400 °C and 1350 °C respectively for 24 hours. The powders were ground to 

63 µm. XRD analysis indicated that no other solids than C3A or C2F were present. CaO 

was synthesized by burning CaCO3 at 1000 °C. 

2.1.1. Fe-hemicarbonate and Fe/Al-monocarbonate 

Pure Fe-Mc and Fe-Hc were synthesized by the addition of appropriate amounts of C2F, 

CaCO3, and CaO to 0.1 M KOH solution (50 ml) at liquid/solid ratio ~ 20. The 

stoichiometry of the reaction is given by: 

2CaO . Fe2O3 + CaCO3 + CaO + 12H2O → 3CaO . Fe2O3 . CaCO3 . 12H2O 

2CaO . Fe2O3 + 0.5CaCO3 + 1.5CaO + 10H2O → 3CaO . Fe2O3 . Ca(CO3)0.5 . 10H2O 

0.1 M KOH solution was used to simulate the high pH present in the pore solution of 

Portland cement. Al/Fe-monocarbonates were synthesized by precipitation from 

supersaturated solutions. Appropriate amounts of C3A, C2F, CaCO3, and CaO were added 

to 0.1 M KOH solution (pH = 13.0). The mole fraction of Al varied from x = 0 to 1. The 

overall stoichometric reaction is given by: 

xC3A + (1-x)C2F + CaCO3 + (1-x)CaO + nH2O → 3CaO(AlxFe1-x)2O3CaCO3nH2O. 

The samples were stored in closed PE-bottles at different temperatures (20, 50 and 80 °C) 

and sampled up to three years. After equilibration the solid and liquid phases were 

separated by vacuum filtration through 0.45µm nylon filters. All sample preparation and 

12


handling were done in a glove box filled with N2-atmosphere to minimize CO2 

contamination. 

The mixes used in the undersaturation experiments correspond to those prepared for the 

oversaturation experiments. After an equilibration time of 3 years, an additional amount 

of 0.1 M KOH solution was added to duplicate the volume of the solution (resulting in 

undersaturation) and equilibrated for further 15 months. Note that, all the 3 years old 

samples prepared during the previous PhD project [31]. 

2.1.2. Fe-monosulfate 

Pure Fe-monosulfate was synthesized by the addition of appropriate amounts of C2F, 

CaSO42H2O and CaO to 50 ml of 0.4 M KOH solution (pH = 13.6) at liquid/solid ratio ~ 

20. The overall stoichometric reaction is given by: 

2CaOFe2O3 + CaSO42H2O + CaO + 10H2O → 3CaOFe2O3CaSO412H2O 

0.4 M KOH solution was used to mimic the high pH conditions in the pore solution of 

Portland cement in which monosulfate is formed. At lower KOH concentrations the 

formation of Fe-ettringite instead of Fe-monosulfate is favored [15]. Mixed Al/Fe- 

monosulfate was synthesized by precipitation from supersaturated solutions. Again 

appropriate amounts of C3A, C2F, CaSO4 2H2O, and CaO were added to 0.4 M KOH 

solution (pH = 13.6). The mole fraction of Al varied from x = 0 to 1. The overall 

stoichiometric reaction is given by: 

xC3A +(1-x)C2F +CaSO42H2O + (1-x)CaO +10H2O→3CaO(AlxFe1x)2O3CaSO412H2O 

Sample handling was as described in section 2.1.1. 

13

2.1.3. Fe-Friedel’s Salt 


Pure Fe-Friedel’s salt was synthesized in three different ways: 

a. By the addition of appropriate amounts of C2F, CaCl2 . 2H2O, and CaO to 

distilled water (50 ml) at liquid/solid ratio ~ 20 according to: 

2CaOFe2O3 + CaCl22H2O + CaO + 8H2O → 3CaOFe2O3 . CaCl210H2O 

b. By mixing appropriate amounts of C2F, CaCl22H2O, and CaO in 0.1 M KOH 

solution (50 ml) at liquid/solid ratio ~ 20 according to the above reaction. 

c. By the addition of appropriate amounts of AlCl36H2O, FeCl36H2O, and CaO 

in 50 ml of 0.1 M KOH at a liquid/solid ratio ~ 20 to obtain 

4CaO (AlxFe1-x)2O3Cl210 H2O. 


2.1.4. Fe-strätlingite 

Different methods were also used to obtain Fe-strätlingite (C2FSH8). The first method 

was mixing appropriate amounts of Fe(OH)3, Na2SiO35H2O and CaO in 0.1 M KOH. 

The second method was by mixing 2FeCl36H2O, Na2SiO35H2O, and 2Ca(NO3)24H2O 

in 0.1 M KOH. The samples were equilibrated at 7, 28 and 200 days at 20, 50 and 80 °C. 

2.1.5. Synthesis of hydrogarnets 

2.1.5.1. Silica free hydrogarnets: Ca3(AlxFe1-x)2(OH)12 

C3AH6 and C3FH6 were synthesized by mixing appropriate amounts of C3A or C2F and 

CaO in 50 ml 0.1 M KOH to obtain a liquid/solid ratio of ~ 20. The suspensions were 

equilibrated at 20, 50 and 80 °C up to three years. The stoichiometry of the reactions is 

given by: 

14

3CaOAl2O3 + 6H2O → 3CaOAl2O36H2O 


2CaOFe2O3 + CaO + 6H2O → 3CaOFe2O36H2O 

The samples were stored in closed PE-bottles and sampled after different reaction times. 


2.1.5.2. Siliceous hydrogarnets: Ca3(AlxFe1-x)2(SiO4)(OH)8 

a) Synthesis at ambient temperature (supersaturation experiments) 

In a first attempt, C3ASH4 and C3FSH4 were synthesized by mixing stoichiometric 

amounts of C3A or C2F with CaO and Na2SiO35H2O at 20 °C in 50 ml 0.1 M KOH at 

liquid/solid ratio of ~ 20. The samples were stored in closed PE-bottles at different 

temperatures (20, 50 and 80 °C) and sampled up to three years under supersaturated 

conditions. Sample handling was as described in section 2.1.1. 

b) Hydrothermal synthesis (undersaturation experiments) 

Due the slow reaction of the C2F clinkers and the poor crystallinity of the products 

formed at 20 °C, mixed Ca3(AlxFe1-x)2(SiO4)(OH)8 solids were also prepared 

hydrothermally at 110 °C. Stoichiometric amounts of AlCl36H2O, FeCl36H2O, 

Na2SiO35H2O and Ca(NO3)24H2O were mixed with 200 ml of 1 M KOH at a 

liquid/solid ratio of ~ 25 to obtain 3CaO(AlxFe1-x)2O3SiO24H2O. A pH of approximately 

13.5 (measured at 20 °C) was observed after mixing. The mixes were stored for 5 days in 

closed teflon vessels at approximately 110 °C. The same procedure was used to prepare 

Fe-hydrogarnets Ca3Fe2(SiO4)3-y(OH)4y containing different quantities of silica and 

hydroxide. 

15


After aging for 5 days at 110 °C, the solid and liquid phases were separated by vacuum 

filtration through 0.45µm nylon filters. The residues were dried in N2-filled desiccators 

over saturated CaCl2 solutions for 1 week. The dried solids were re-dissolved in 0.1 M 

KOH at liquid/solid-ratio of ~ 20 in HDPE bottles and equilibrated at 20 °C and 50 °C for 

4 months (undersaturation experiments). 

2.1.6. Hydrated cement samples 

To study the fate of iron in hydrated cements, hydration experiment were carried out 

using an ordinary Portland cement (OPC), CEM I 32.5 R and a sulfate resistant cement 

HS (CEM I 42.5 N). The chemical composition of the cements used for this study is 

listed in Table 1. The cement pastes were prepared at a water/cement (w/c) ratio of 0.425 

and hydrated at 20 and 50 °C. The latter temperature was chosen since the composition of 

the hydration assemblage is expected to change around 48 °C [3, 32]. The cements were 

hydrated for 4, 8, 16 hours, 1, 28, 150 days, 1 and 3 years at 20 and 50 °C. 

The hydration of the cements was stopped using isopropanol. The sample were dried in 

an oven at 40 °C for 1 hour. The sample was ground by hand for XRD, EXAFS and TGA 

analysis. 

16


Table 1 Oxide and the phase composition of the cements used. 

OPC HS OPC HS 

CEM I 32.5 R CEM I42.5 N CEM I 32.5 R CEM I42.5 N 

a Chemical analysis (g/100g) (g/100g) b 

Phase composition (g/100g) (g/100g) 

SiO2 20.34 17.55 C3S 53.5 60.0 

Al2O3 5.17 4.58 C2S 18.0 5.1 

Fe2O3 3.09 7.2 C3A 8.5 0.0 

CaO 63.38 60.34 C2(A,F) 9.4 21.9 

MgO 2.53 1.98 CaSO4 2.5 2.7 

K2O 0.91 1.02 

c 

K2SO4 1.5 1.5 

Na2O 0.2 0.33 

c 

Na2SO4 0.2 0.2 

SO3 2.41 2.67 

d 

K2O 0.1 0.2 

CO2 0.12 0.73 

d 

Na2O 0.1 0.2 

TiO2 0.32 0.53 CaO(free) 1.2 - 

Mn2O3 0.06 0.07 CaCO3 0.8 - 

P2O5 0.25 0.34 

d 

MgO 2.5 2.0 

Cl 0.03 0.073 

d 

SO3 0.2 0.3 

Loss of ignition 1.01 3.24 

a 

XRF data corrected for ignition loss. 

b 

Calculated from the chemical analysis. 

c 

Estimated based on the alkali content and on the alkali distribution given in Taylor (1987). 

d 

Present as solid solution in the major clinker phases 

2.1.7. Synthesis of synthetic Fe-cement 

In addition, the hydration products of Al-free synthetic cements were investigated. The 

clinker composition of the synthetic Fe-cement consisted of 78.5% C3S, 19.3% C2F, 

0.5% Na2SO4 and 1.7 % K2SO4. The alkalis were added to mimic real Portland cement. 

No Al-phases were present to allow the identification of Fe-containing hydrates. 

The synthetic cements were hydrated in the presence and the absence of calcite and 

gypsum with a liquid/solid ratio of 1. Different quantities of gypsum and calcite were 

added while the ratio of C3S to C2F was kept constant (Table 2). The phase assemblage 

was investigated after 3 days and 3 months. The results from the studies were compared 

to those from studies on the synthesized phases [10-14, 19] and the studies in OPC 

systems [21-23, 33-35]. 

17


Table 2 The compositions of the synthetic Fe-cement mixes with varying gypsum (CsH2) and 

calcite (Cc) content. 

Sample ID C3S C2F Na2SO4 K2SO4 CsH2 Cc 

g/100g g/100g g/100g g/100g g/100g g/100g 

Gyp-0% 78.5 19.3 0.5 1.7 0.0 

Gyp-6% 73.7 18.1 0.5 1.7 6.0 

Gyp-26% 57.6 14.2 0.5 1.7 26.0 

Cc1 70.7 19.3 0.5 1.7 7.8 

Cc 2 66.3 18.1 0.5 1.7 6 7.4 

Cc 3 56.2 15.4 0.5 1.7 20 6.2 

C2F-pure 97.8 0.5 1.7 

C2F-Gyp 65.2 0.5 1.7 32.6 

C2F-Gyp-Cc 58.6 0.5 1.7 32.6 6.6 

C2F-Cc 88 0.5 1.7 9.8 

2.2. Analytical methods 

2.2.1. Powder X-ray diffraction 

X-ray powder diffraction (XRD) measurements were carried out using CuKα radiation on 

a PANalytical X’Pert Pro MPD diffractometer in a -2 configuration with an angular 

scan 5°-75° 2θ and an X’Celerator detector. To study the effect of relative humidity, a 

climatic chamber (Anton Paar) specially designed for the X- ray diffractometer in a - 

configuration was used. The sample was placed in a sample tray of the climatic chamber 

of the X-ray diffractometer where both temperature and relative humidity can be 

controlled. The diffractograms of the synthesized pure phases and identification of new 

phases in cement paste were verified using the PDF database of the International Centre 

for Diffraction Data (ICDD). CaF2 was mixed to the powder samples as internal standard 

to determine the unit cell parameters for some Fe-containing phases. 

18


2.2.2. Synchrotron powder diffraction 

Synchrotron powder diffraction data were collected at the Swiss-Norwegian Beam Line 

(SNBL) at the European Synchrotron Radiation Facility (ESRF), Grenoble, France. The 

powder material was introduced into glass capillaries (0.5 mm diameter). Data collection 

was performed at 295 K at a wavelength of = 0.72085 Å using a MAR345 image plate 

detector with the highest resolution (3450 x 3450 pixels with a pixel size of 100 m). The 

calculated absorption coefficient mR (m = powder packing factor, = linear absorption 

coefficient, R = radius of the capillary) was estimated at 0.65. Three sample-to-detector 

distances were used (150, 250 and 350 mm) in order to combine the advantages of high 

resolution and extended 2 range. The detector parameters and the wavelength were 

calibrated with NIST LaB6. The exposure time was 60s with a rotation of the capillary by 

60°. The two-dimensional data were integrated with the Fit2D program which produced 

the correct intensity in relative scale [36]. This 2D detector was used in order to perfectly 

define the background, to observe very weak diffraction peaks, and to improve the 

accuracy of the integrated intensities by achieving a better powder average. Uncertainties 

of the integrated intensities were calculated at each 2-point applying Poisson statistics to 

the intensity data, considering the geometry of the detector. The instrument resolution 

function was determined from the LaB6 data. 

2.2.3. Thermogravimetric analysis 

Thermogravimetric analysis (TGA) was carried out to determine the weight loss and 

characterize the thermal behavior of the solids. The analysis was carried out in a N2 

atmosphere on about 8-12 mg of crushed material at a heating rate of 20°C/min over the 

temperature range from 30-980°C using a Metter Toledo TGA instrument. 

19


2.2.4. Vibrational spectroscopy (Raman and Infrared spectroscopy) 

Micro-Raman spectra were recorded at room temperature in the back scattering 

geometry, using a Jobin-Yvon T64000 device. The Raman detector was a charge coupled 

device (CCD) multichannel detector cooled by liquid nitrogen to 140 K. The laser beam 

was focused onto the sample through an Olympus confocal microscope with x100 

magnification. The laser spot was about 1 μm². The spectral resolution obtained with an 

excitation source at 514.5 nm (argon ion laser line, spectra physics 2017) was ~ 1 cm -1 . 

The measured power at the sample level was kept low (< 5mW) in order to avoid any 

damage of the material. The Raman scattered light was collected with microscope 

objectives at 360° angle from the excitation and filtered with an holographic Notch filter 

before being dispersed by a single grating (1800 grooves per mm). Infrared spectroscopy 

(FTS 6000 Spectrometer using KBr pellets technique) was used to characterize the solid 

phases. The IR spectra were collected in transmission mode in the region 4000 cm -1 to 

600 cm -1 . 

2.2.5. Scanning electron microscopy (SEM) 

The samples were coated with carbon wire and examined using a Philips SEM FEG XL 

30 scanning electron microscopy (SEM). Secondary electron images (SE) were taken to 

analyze the hydrates morphology. Energy dispersive X-ray spectroscopy (EDX) was 

applied to determine the elemental composition of the hydrates. 

2.2.6. Liquid phase analysis 

A pH electrode (Knick pH-meter 766 with a Knick SE 100 pH/Pt 1000 electrode) was 

used to measure the pH in an aliquot of the undiluted solutions immediately after 

20


filtration. The electrode was calibrated with KOH solutions of known concentrations 

prior to the measurements. Another aliquot of the filtered solution was diluted by a factor 

of 10 with HNO3 (6.5% supra pure) and analyzed for Ca, Al, Si and K by inductively- 

coupled plasma optical emission spectrometry (ICP/OES; Varian, VISTA Pro) and for Fe 

by inductively-coupled plasma mass spectrometry (ICP/MS; Finnigan MAT, 

ELEMENT2). The Cl concentrations were determined in diluted, non-acidified solutions 

using a Dionex ion chromatography system (ICS) 3000 and chloride standards from 

Fluka. 

2.2.7. Selective dissolution 

Selective dissolution allows major phases to be dissolved from the cement matrix, thus 

causing to an enrichment of minor phases in the residue. The minor phases can then be 

identified more clearly using standard methods [37]. A salicylic acid/methanol (SAM) 

extraction was used to dissolve alite, belite, portlandite, C-S-H, AFt and AFm phases 

leaving residues of ferrite, siliceous hydrogarnet and hydrotalcite. 5 g of hydrated cement 

were stirred in a flask containing 300 ml methanol and 20 g salicylic acid for 2 hours. 

The suspension was allowed to settle for about 15 minutes. The solid was then filtered by 

vacuum filtration using 0.45µm nylon filters, washed with methanol, dried at 90°C for 45 

minutes and then analyzed by XRD and TGA. 

21


2.2.8. Synchrotron-based X-ray absorption spectroscopy (XAS) 

2.2.8.1. Data collection and reduction 

Synchrotron-based XANES and EXAFS spectra were used to determine the Fe- 

containing phases in the complex cement matrix. The spectra were collected at the Fe K- 

edge (7112 eV) at beamline BM26A (Dubble) at the European Synchrotron Radiation 

Facility (ESRF) Grenoble, France, and beamline X10DA (SuperXAS) at the Swiss Light 

Source (SLS). Both beamlines are equipped with a Si (111) crystal monochromator. The 

monochromator angle was calibrated by assigning the energy of 7112 eV to the first 

inflection point of the K-adsorption edge of Fe metal foil. The XANES and EXAFS 

measurements were carried out at room temperature in transmission (ionization 

chambers, Oxford Instruments) and in fluorescence mode (BM26A: 9 channel monolithic 

Ge-solid-state detector; X10DA: 13 element Ge-solid-state detector). A minimum of 

three scans were collected up to k ~ 13 Å −1 and averaged for each sample. 

EXAFS data reduction was performed with the IFEFFIT (ATHENA) software package 

following standard procedures [38, 39]. Background subtraction and normalization were 

carried out by fitting a first-degree polynomial to the pre-edge and a third-degree 

polynomial to the post-edge range of the spectra. The energy was converted to 

photoelectron wave vector units (Å -1 ) by assigning the ionization energy of the Fe K-edge 

(7112 eV), E0, to the first inflection point of the absorption edge or half height of the 

edge step, respectively. The radial structural function (RSF) was obtained by Fourier 

transforming k 3 -weighted χ(k) functions between 2.0 and 11.5 Å −1 using a Kaiser-Bessel 

window function with a smoothing parameter of 4. A multi shell approach was employed 

for data fitting. Theoretical single scattering paths were calculated with FEFF8 using the 

Al-analogues structure as model compounds. The amplitude reduction factor (S0 2 ) was set 

22


at 0.75. For the hydrated cement spectra, pre-edge background subtraction and 

normalization were performed. The spectrum of at least one reference compound was 

used at each campaign to evaluate energy calibration at the beam lines and when needed 

to adjust the data on the same energy scale. 

2.2.8.2. Data analysis and fitting 

Normalized XANES spectra and the k 3 -weighted EXAFS spectra in the k-range between 

2.0 and 11.5 Å −1 were used for data analysis and fitting. Analysis of XANES and EXAFS 

spectra from the cement pastes was based on the assumption that relevant species of the 

absorber atom contributed to the overall signal. The latter analysis was based on principal 

component analysis (PCA) and target transformation (TT) according to procedures 

described in detail elsewhere and using the Labview software package of beamline 10.3.2 

[27, 30, 40, 41]. Principal component analysis was used to determine whether a set of 

spectra could be represented as linear combination of a smaller number of independent 

component spectra. The indicator value of Malinowsky (IND) is commonly used as 

criterion to determine the minimum number of independent components sufficient to 

reconstruct the set of experimental spectra [42]. Calculation of the IND factor was 

performed for the XANES range (30 eV below the absorption edge to 150 eV above the 

absorption edge) and the EXAFS range (3.0 Å −1 ≤ k ≤ 10.0 Å −1 ). Both approaches led to 

similar results. The use of TT allowed testing which spectra of the Fe-containing cement 

minerals were necessary to reconstruct the set of spectra. 

In this study PCA/TT was applied to gain information on the number and type of Fe 

species (Fe-containing minerals) in the cement matrix. Based on this information a least- 

square linear combination (LC) fitting as implemented in the IFEFFIT software package 

23


(ATHENA) was applied to determine the contribution of the XANES or EXAFS spectra 

of the individual reference compounds, respectively, to the experimental XANES or 

EXAFS spectra of the hydrated cement paste (HCP) samples. Thus, all HCP spectra can 

be represented as a sum of the spectra of Fe reference compounds with coefficients 

(weights) between 0 and 1. For example, a reference spectrum with a coefficient ~1 

indicates the contribution of a single component to the HCP spectra. This further implies 

that the coordination environments of Fe in the reference compound and that in the HCP 

samples are identical. 

2.2.9. Calorimetry 

A conduction calorimeter (Thermometric TAM Air) was used to determine the rate of 

hydration heat during the first 72 hours. 2.00 g synthetic Fe-cements were weighed into a 

flask and the corresponding amount of water was added (2.00 ml). The flask was tightly 

closed and placed into the calorimeter. Internal mixing was done by a small stirrer for 

1 min. 

2.3. Thermodynamic modeling 

Thermodynamic modeling was carried out using the geochemical code GEMS [43]. 

GEMS is a broad-purpose geochemical modeling code, which computes equilibrium 

phase assemblage and speciation in a complex chemical system from its total bulk 

elemental composition. Chemical interactions involving solids, solid solutions, gas 

mixture and aqueous electrolyte are considered simultaneously. The default database of 

24


the GEMS code was used, which is based on the PSI chemical thermodynamic database 

[44] merged with the slop98.dat database for temperature and pressure corrections [45]. 

2.3.1. Estimation of heat capacity of Fe-containing phases 

The heat capacity, Cp°, of a solid can be expressed as a function of temperature using the 

parameters (a0, a1, a2 and a3) according to: 

Cp o = a0 + a1T + a2T -2 + a3T -0.5 

In this study the heat capacities Cp° for Fe-containing phases were calculated based on 

the Cp° of the structurally similar Al-analogues [46] using reference reactions as shown 

in Table 3. If such reference reactions involve only solids and no “free” water, the change 

in heat capacity and the entropy is approximately zero [47, 48]. 

Table 3 Reference reactions used to estimate unknown heat capacities of cement minerals. 

Phases Reference reaction 

3CaOFe2O3CaCO312H2O → 3CaOAl2O3CaSO412H2O - Al2O3 - CaSO4 + Fe2O3 +CaCO3 

3CaOFe 2O 3Ca(CO 3) 0.510H 2O → 3CaOAl 2O 3CaSO 412H 2O - CaSO 42H 2O - Al 2O 3 + Fe 2O 3 + 0.5CaCO 3 + 0.5CaO 

3CaOFe2O3CaSO412H2O → 3CaOAl2O3CaSO412H2O - Al2O3 + Fe2O3 

3CaOFe2O3CaCl210H2O → 3CaOAl2O3CaSO412H2O - CaSO4 . 2H2O - Al2O3 + Fe2O3 + CaCl2 

3CaOFe2O3Ca(OH)212H2O → 3CaOAl2O3CaSO412H2O - CaSO4 - Al2O3 + Fe2O3 + Ca(OH)2 

3CaOFe 2O 36H 2O → 3CaOAl 2O 36H 2O - Al 2O 3 + Fe 2O 3 

3CaOFe2O30.95SiO24.1H2O →3CaOAl2O36H2O - Al2O3 +0.95SiO2 + 1.9CaO -1.9Ca(OH)2+ Fe2O3 

3CaO.Fe2O31.52 SiO2.2.96H2O →3CaOAl2O36H2O - Al2O3 +1.52SiO2 + 3.04CaO -3.04Ca(OH)2+ Fe2O3 

3CaOAl2O30.41 SiO25.18H2O →3CaOAl2O36H2O +0.41SiO2 + 0.82CaO + 0.82Ca(OH)2 

3CaOAl 2O 30.84 SiO 24.32H 2O →3CaOAl 2O 36H 2O +0.84SiO 2 + 1.68CaO + 1.68Ca(OH) 2 

2.3.2. Determination of solubility products 

The measured composition of the liquid phase was used to calculate solubility products 

for the different solids. The activity coefficients of the dissolved species were calculated 

25


from the measured concentration. The activity coefficients of aqueous species yi were 

computed with the expanded extended Debye-Hückel equation in Truesdell-Jones form 

with individual parameters ai (dependent on ion size) and common third parameter by: 

2 

Ay 

zi 

log i 

1 

B a 

I 

by 

I 

I 

y i 

where zi denotes the charge of species i, I the effective molal ionic strength, by is a semi- 

empirical parameter (0.064 was used to calculate the solubility of Fe- 

monocarbonate/hemicarbonate and 0.123 for KOH electrolytes for the other Fe- 

containing phases at 25 °C), and Ay and By are P, T-dependent coefficients. This activity 

correction is thought to be applicable up to 1-2 M ionic strength [43]. The aqueous ion 

activities and speciation were calculated using the GEMS database relevant to the 

particular calculations. The solubility products were calculated from the activities 

obtained according to the reactions given in Table 4. From the calculated total solubility 

products, the Gibbs free energy of the reaction, 

according to: 

 

0 

r, 

T G RT 

ln K S 0, 

T 

at a temperature T was computed 

0 

r, 

T G 

where R = 8.31451J/(mol K) is the universal gas constant and T is the temperature in 

Kelvin. 

The temperature dependence of the solubility product of Fe-containing phases was 

computed based on the solubility measured at 20, 50 and 80 °C with the help of GEMS, 

using the built-in three-term temperature extrapolation function and the relationships 

shown in the equations below [49]. The three-term temperature extrapolation assumes 

26


0 

that the heat capacity of the reaction, rC p , is constant in the considered temperature 

range. 

A2 

A0 

A lnT 

; 

T 

log KT 3 

0. 

4343 

R 

0 

0 

SCp 1lnT 

A0 r T0 

r T0 

0. 

4343 

 

R 

0 

0 

H Cp T 

A2 r T0 

r T0 

0. 

4343 

A 

R 

 

 

0 

3 r T0 

Cp 

 

 

 

T 

ln 

0 0 

0 

rST 

rST 

Cp 

0 r T0 

T0 

0 

0 

0 

rH 

T rH 

T Cp ( 0 r T0 

0 

G H T 

r 

0 

T 

r 

0 

T 

r 

S 

0 

T 

T T 

) 

0 

0 

where T0 is the reference temperature (298.15 K), S0 the entropy, H 0 the enthalpy and G 0 

the Gibbs free energy. A more detailed description of the temperature corrections used in 

GEMS is given elsewhere [50] and in the online documentation of GEMS. The value of 

0 

the heat capacity of the reaction, rC p , has little influence on the calculated log K values 

in the temperature range 0-100 °C. Generally, the difference between the 3-term 

extrapolation and the more complete description using 7 terms is usually negligible in the 

temperature range from 0 to 100 °C [51]. 

The entropy S° was adjusted to obtain the best fit between the measured solubility data at 

different temperatures and the calculated solubility products. As only solubility products 

at two or three different temperatures were available, only the entropy was fitted, while 

27


the heat capacities Cp° for Fe-containing phases was calculated based on a reference 

reaction in Table 3. In addition, as discussed above, the value of the heat capacity has 

only little influence on the calculated solubility products in the temperature range 0-100 

°C, which makes it insensitive to the fitting procedure. Table 4 and Table 5 summarize 

the thermodynamic data used and obtained in this study. 

28


Table 4 Dissolution reaction used for thermodynamic calculation. 

Phases Reactions log KS0 ref 

Al-katoite Ca3Al2 (OH)12 → 3Ca 2+ +2Al(OH)4 − + 4OH − -20.56 d 

Fe-katoite Ca3Fe2 (OH)12 → 3Ca 2+ +2Fe(OH)4 − + 4OH − -25.56 d 

Si-poor Al-siliceous hydrogarnet Ca3Al2(SiO4)0.41(OH)10.36→ 3Ca 2+ +2AlO2 − + 0.41HSiO3 2− + 3.59OH − + 3.18H2O -25.47 d 

Si-rich Al-siliceous hydrogarnet Ca3Al2(SiO4)0.84(OH)8.64 → 3Ca 2+ +2AlO2 − + 0.84HSiO3 2− + 3.16OH − + 2.32H2O -26.70 d 

Si poor Fe-siliceous hydrogarnet Ca3Fe2(SiO4)0.95(OH)8.2 → 3Ca 2+ +2FeO2 − + 0.95HSiO3 2− + 3.05OH − + 2.1H2O -32.75 d 

Si rich Fe-siliceous hydrogarnet Ca3Fe2(SiO4)1.52(OH)5.92 → 3Ca 2+ +2FeO2 − + 1.52HSiO3 2− + 2.48OH − + 0.96H2O -34.68 d 

Al-ettringite Ca6Al2(SO4)3(OH)12·26H2O → 6Ca 2+ +2Al(OH)4 − + 3SO4 2− +4OH − +26H2O -44.90 a, c 

Fe-ettringite Ca6Fe2(SO4)3(OH)12·26H2O → 6Ca 2+ +2Fe(OH)4 − + 3SO4 2− +4OH − +26H2O -44.00 a, b 

Al- hydroxy AFm Ca4Al2(OH)12·7H2O → 4Ca 2+ +2Al(OH)4 − + 6OH − +6H2O -25.40. a 

Fe- hydroxy AFm Ca4Fe2(OH)12·7H2O → 4Ca 2+ +2Fe(OH)4 − + 6OH − +6H2O -30.64 d 

Al-monosulfate Ca4Al2(SO4)(OH)12·6H2O → 4Ca 2+ +2Al(OH)4 − + SO4 2− +4OH − +6H2O -29.26 a, c 

Fe-monosulfate Ca4Fe2(SO4)(OH)12·6H2O → 4Ca 2+ +2Fe(OH)4 − + SO4 2− +4OH − +6H2O -31.57 d 

Al-monocarbonate Ca4Al2(CO3)(OH)12·5H2O → 4Ca 2+ +2Al(OH)4 − + CO3 2− +4OH − + 5H2O -31.47 a, c 

Fe-monocarbonate Ca4Fe2(CO3)(OH)12·6H2O → 4Ca 2+ +2Fe(OH)4 − + CO3 2− +4OH − + 6H2O -34.59 d 

Al-hemicarbonate Ca4Al2(CO3)0.5(OH)12·6H2O → 4Ca 2+ +2Al(OH)4 − + 0.5CO3 2− +5OH − + 5.5H2O -29.13 a, c 

Fe-hemicarbonate Ca4Fe2(CO3)0.5(OH)12·4H2O → 4Ca 2+ +2Fe(OH)4 − + 0.5CO3 2− +5OH − + 3.5H2O -30.83 d 

Al-Friedel’s salt Ca4Al2(Cl2)(OH)12·4H2O → 4Ca 2+ +2Al(OH)4 − + 2Cl − +4OH − + 4H2O -27.69 f 

Fe-Friedel’s salt Ca4Fe2(Cl2)(OH)12·4H2O → 4Ca 2+ +2Fe(OH)4 − + 2Cl − +4OH − + 4H2O -28.62 d 

Fe(OH)3 (am.) Fe(OH)3 + OH - → Fe(OH)4 - -2.60 e 

Fe(OH)3 (microcr.) Fe(OH)3+ OH - → Fe(OH)4 - -4.10 d 

Al(OH)3 (am.) Al(OH)3+ OH - → Al(OH)4 - 0.24 a 

Gibbsite Al(OH)3+ OH - → Al(OH)4 - -1.24 e 

Gypsum CaSO4 2H2O → Ca 2+ + SO4 2- + 2H2O -4.58 e 

Portlandite Ca(OH)2 → Ca 2+ + 2OH - -5.2 d 

Calcite CaCO3 →Ca 2+ + HCO3 − 1.85 e 

(a ) Lothenbach et al [3], (b) Möschner et al [15], (c) Matschei. et al [52], (d) this study 

(e) GEMS/PSI TDB [45, 53]. (f) Balonis et al [54] 

29


Table 5 Thermodynamic data at standard conditions (298 K, 1 atm) used for the calculation of the 

liquid phase compositions and for computation of thermodynamic parameters for the 

synthesized solids. 

fG° fH° S 0 C 0 p 1 a0 a1 a2 a3 V 0 Ref. 

Phase [kJ/mol] [kJ/mol] [J/K/mol] [J/mol/K] [J/(mol.K)] [J/(mol.K 2 )] [J K/mol] [J/(mol.K 0.5 )] [cm3/mol] 

Al-katoite -5008.5 -5535 432 459 292 0.5610 150 d 

Fe-katoite -4118.6 -4724 165 485 275 0.0627 0 155 d 

Si-poor Al-siliceous hydrogarnet -5193.5 -5717 342 441 198 0.5967 -9.98E+05 1312 151 d 

Si-rich Al-siliceous hydrogarnet -5365.2 -5867 310 422 100 0.6342 -2.05E+06 2688 142 d 

Si-poor Fe-siliceous hydrogarnet -4523.5 -4854 855 612 582 0.6094 2.02E+06 -3040 156 d 

Si-rich Fe-siliceous hydrogarnet -4752.8 -5044 847 688 766 0.5988 2.29E+06 -4864 161 d 

Al-ettringite -15205.9 -17535 1900 2174 1939 0.7890 0 0 707 a 

Fe-ettringite -14282 -16600 1937 2200 1922 0.8550 2.02E+06 0 717 a, b 

Al-hydroxy AFm -7326.6 -8300 708 930 711 1.0470 0 -1600 274 b 

Fe-hydroxy AFm -6438 -7431 640 956 694 1.1134 2.02E+06 -1600 286 d 

Al-monosulfate -7778.5 -8750 821 942 594 1.1680 0 0 309 a, c 

Fe-monosulfate -6873.2 -7663 1430 968 577 1.2340 2.02E+06 0 321 d 

Al-monocarbonate -7337.5 -8250 657 881 618 0.9820 -2.59E+06 0 262 a, c 

Fe-monocarbonate -6674 -7485 1230 950 612 1.1600 -5.73E+05 0 292 d 

Al-hemicarbonate -7336 -8270 713 906 664 1.0140 -1.30E+06 -800 285 a, c 

Fe-hemicarbonate -5952.9 -6581 1270 841 308 1.2014 -9.08E+05 3200 273 d 

Al-Friedel’s salt -6814.6 -7625 731 829 498 0.8900 -2.03E+06 1503 272 f 

Fe-Friedel’s salt -5900.1 -6525 1286 855 481 0.9611 -16130 1503 208 d 

Al(OH)3 (am.) -1143 -1281 70 93 36 0.1910 0 0 32 a 

Gibbsite -1151 -1289 70 93 36 0.1910 0 0 32 e 

Fe(OH)3 (am.) -700 -879 88 43 28 0.0520 0 0 34 e 

Fe(OH)3 (microcr.) -709 -841 88 43 28 0.0520 0 0 34 d 

Gypsum -1798 -2023 194 186 91 0.3180 0 0 75 e 

Portlandite -897 -985 83 88 187 -0.0220 0 -1600 33 e 

Calcite -1129 -1207 93 82 105 0.0220 -2.59E+05 0 37 e 

CaO -604 -635 40 43 49 0.0040 -6.53E+05 0 17 e 

Al2O3 -1568 -1662 51 79 115 0.0180 -3.51E+06 0 26 e 

Fe2O3 -8214 -8214 88 105 98 0.0780 -1.49E+06 0 30 e 

References: a ) Lothenbach et al. [3], b) Möschner et al. [15], (c) Matschei et al. [52], d) this study, e) Thoenen et al. and Hummel et 

al. [44, 45], f) Balonis et al. [54] 

1 Cp o = a0 + a1T + a2T -2 + a3T -0.5 , Si-poor and Si-rich Al-siliceous hydrogarnet do not form at room temperature 

30


2.3.3. Thermodynamics of solid solutions 

Solid solutions are frequently encountered in cementitious systems [7, 16, 54-56]. 

According to the definition given in Bruno et al. [57], a solid solution is a homogeneous 

crystalline structure in which one or more types of atoms or molecules are partly 

substituted without changing the structure, although the lattice parameters may vary. If 

the size and crystal lattice between host and substituting ion are similar, the formation of 

an ideal solid solution is probable. The larger the difference, the stronger is the tendency 

to non-ideality [57, 58] and thus the tendency for the presence of miscibility gaps. 

Thermodynamically, an ideal solid solution forms if the enthalpy of mixing is zero. Any 

other solid solution is called non-ideal. The presence of ideal or non-ideal solid solutions 

can stabilize the formation of these solids and may result in a significant lowering of 

dissolved ion concentrations. Ideal solid solutions are always more stable than the 

mechanical mixture of the end-members and thus stabilize the formation of the solid 

solution with respect to other solids. 

A very short summary of the thermodynamics of solid solution is given here, based on 

the comprehensive books of Bruno et al. [57] and Anderson and Crerar [47]. A pure 

phase has only one fixed, constant stoichiometry, i.e. it consists of only one mineral e.g. 

portlandite Ca(OH)2; its activity is equal to one. In contrast, a solid solution has a 

variable bulk composition and can be described by two (or more) end-members which are 

present in varying concentrations in the solid solution. Each end-member, however, has a 

fixed elemental stoichiometry. Generally one distinguishes between ideal and non-ideal 

solid solutions. 

The molar Gibbs free energy of a solid solution between different end-members can be 

calculated according to: 

31


0 

X i 

f Gi 

RT 

X i ln X i RT 

G 

G 

G 

G 

 

X ln 

ss 

mm 

is 

ex 

i i 

i 

The first term of the above equation, Gmm, is related to mechanical mixing of the end- 

members and is calculated using the mole fraction Xi = ni/ni (ni is the mole amount of 

the end member i; Xi = 1) and fGi 0 - the standard molar Gibbs energy of formation of 

end-member i. The second term is the Gibbs free energy related to the ideal solid solution 

Gis with activity coefficients equal to 1. The last term is the excess free-energy of 

mixing, Gex, due to non-ideality. The excess Gibbs energy of mixing is only needed to 

compute thermodynamic properties of non-ideal solid solutions and is calculated as 

i 

RT ln . In the case of an ideal solid solution, all activity coefficients, I, equal to 1 

X i 

i 

and thus the excess Gibbs free energy of mixing is zero. An ideal solid solution is always 

more stable than the mechanical mixture of the end-members as the Gibbs free energy of 

the solid solution is lower than the Gibbs free energy of mechanical mixing (the second 

term of the above equation is negative (Xi


The activity coefficients of the non-ideal binary solid solution are calculated as: 

lnγi = X 2 2 [0 –1 (3 X1 – X2)] 

lnγ2 = X 2 1[0 –1 (3 X2 – X1)] 

The software MBSSAS [59] was used to derive the Guggenheim parameters a0 and a1 

based on experimentally-observed compositional boundaries of the miscibility gap in 

the investigated binary solid solution series. A detailed description of MBSSAS is 

given elsewhere [59]. 

Lippmann developed an algorithm to describe the phase diagram of a binary solid 

solution and its relation to the composition of the aqueous phase. A total solubility 

product (ΣΠ) was introduced. If the system is in equilibrium, the total solubility product 

of a binary solid solution (B1-xCxA) is the sum of the partial solubility products of each 

end member. The total solubility product of a binary solid solution (B1-xCxA) can be 

calculated from the sum of the partial solubility product of each end members (see the 

equations below). 

ΣΠSolidus = KBA.XBA.γBA + KCA.XCA.γCA 

KBA.XBA.γBA = [B + ][A - ] 

KCA.XCA.γCA = [C + ][A - ] 

where KBA and KCA are the solubility products of the end members BA and CA; XBA and 

XCA are mole fractions of BA and CA in the solid; γBA and γCA are the activity 

coefficients as expressed first by the Guggenheim expansion series and then modified by 

Redlich and Kister [60]. The above equations are used to derive the solidus curve of the 

Lippmann phase diagram as a function of the solid phase. The solutus curve in the 

33


Lippmann phase diagram is described below as a function of the mole fraction of the 

liquid phase in the solid solution series. 

ΣΠsolutus= 1/( XB,liq/KB.γB+ XC,liq/KC.γC) 

The mole fraction of the liquid phase XB,liq and XC,liq are calculated as: 

XB,liq = [B + ]/[B + ] +[C + ] = KBA.XBA.γBA/ ΣΠSd 

XC,liq = [C + ]/[B + ] +[C + ] = KCA.XCA.γCA/ ΣΠSd 

The total solubility products of the solidus and solutus curve of the Lippmann phase 

diagram allow the properties of binary solid solution or miscibility gaps to be determined. 

The use of solid solution for cement minerals is also explained in details in the paper of 

Möschner et al. [16, 61, 62] and in the thesis of Matschei et al. [63]. 

2.3.4. Thermodynamic modeling of cement hydration 

Cement hydration was modeled based on the measured composition of the cements used 

in this study (Table 1). GEMS together with a set of equations that describe the 

dissolution as a function time was used to predict the solid phase formed during hydration 

process [4, 6, 64]. The calculations were carried out using the cement database 

Cemdata2007 [3, 7, 15]. The thermodynamic properties of Fe-containing hydrates were 

updated using the data derived in this thesis and Al-containing siliceous hydrogarnets 

were updated using the data derived in this thesis (see Table 5). As the Al-containing 

siliceous hydrogarnets did not form at room temperature, but only hydrothermally, their 

formation was suppressed in the calculations of cement hydration. A more detailed 

procedure of modeling of the hydration of cement is reported e.g. in [3, 6, 65]. 

34

CHAPTER 3 SYNTHETIC FE-CONTAINING HYDRATES 

3. SYNTHETIC FE-CONTAINING HYDRATES 

3.1. Iron containing carbonate AFm phases 1 

3.1.1. Introduction 

The main hydration products of Portland cements include C-S-H (calcium silicate 

hydrate), portlandite, ettringite and AFm (Al2O3-Fe2O3-mono) phases. AFm phases are 

formed from C3A (3CaOAl2O3) and C2(A;F) (Ca2(AlxFe1-x)O5) phases in the presence of 

carbonates, sulfates, chlorides and hydroxide during the hydration of Portland cement. 

The general formula is Ca2(Al,Fe)(OH)6XnH2O, where X denotes a single charged or 

half of a double charged anion which occupies the interlayer sites. Among possible 

anions are OH - , SO4 2- , CO3 2- and Cl - . AFm phases have a layered structure composed of 

two layers, a positively charged main layer [Ca2(Al,Fe) (OH)6] + and a negatively charged 

[XnH2O] - interlayer. The main layer consists of sheets of Ca(OH)6 octahedral ions, as in 

portlandite, in which every third Ca 2+ is substituted by Al 3+ and/or Fe 3+ . 

Cements are sensitive to carbonation which can lead to the formation of hemicarbonate 

3CaO(AlxFe1-x)2O3(CaCO3)0.5(Ca(OH)2)0.5nH2O and/or monocarbonate 3CaO(AlxFe1- 

x)2O3(CaCO3)mH2O, x=0 to 1. Al-monocarbonate (Al-Mc) has a triclinic 

pseudohexagonal symmetry [66]. The solubility products of Al-monocarbonate and 

hemicarbonate have been determined experimentally in the range of 5 to 85 °C [7]. The 

stability of these phases has an impact on the bulk chemistry of cements as the formation 

of hemi- and/or monocarbonate indirectly stabilizes ettringite. This results in a higher 

volume of hydrated phases which can contribute to the improvement in mechanical 

properties of cements [65, 67]. It has been shown that Al-hemicarbonate (Al-Hc) and 

1 This chapter has been published in Cement and Concrete Research, 41 (2011). 

35


hydroxy-AFm are unstable with respect to Al-monocarbonate in the presence of calcite 

[2, 65]. 

OPC contains around 3-7 % of Fe2O3. During hydration, Fe-containing AFm and/or Fe- 

AFt phases may form. The extent to which Fe is present in AFm and AFt-phases will 

influence strongly the amount of AFt and AFm phases present and thus the volume of the 

hydrates and the properties of the hydrated cement. Understanding the characteristic of 

the hydrates in complex cement matrices is important since the material properties of 

cement-based materials are related to the chemical environment and the thermodynamic 

properties of the hydrated phases. 

Until recently only rough estimates of the solubility products of Fe-containing 

monocarbonate and hemicarbonate have been available [3], where the solubility has been 

estimated based on the solubility of the Al-containing phases. The first experimental data 

on the solubility of Fe-monocarbonate were estimated from Fe-ettringite experiments 

where contamination with CO2 led to the formation of Fe-Mc [15]. 

The formation of solid solutions can play an important role in stabilizing these solids. 

Solid solutions between anions in the interlayer structure of Al-containing AFm phases 

are common [52, 68, 69]. The existence of a solid solution between the Al- and Fe- 

ettringite has been reported [16]. It is unclear, however, to what extent Fe and Al in the 

main layer of AFm phases form solid solutions. 

In this section Fe-containing monocarbonate and hemicarbonate and the solid solution 

series with their aluminum analogues were synthesized to study their structure and 

solubility. Different techniques were used to characterize the synthesized solids. 

Synchrotron X-ray powder diffraction and Raman spectroscopy were used to determine 

the crystal structure of the Fe-monocarbonate. The solubility products of Fe-Mc and Fe- 

36


Hc were determined experimentally and compared with their Al analogues. The solubility 

products were used together with thermodynamic data for the other cement minerals [3] 

to model the hydrate assemblages of hydrated Portland cement in the presence of CaCO3 

and CaSO4. 

3.1.2. Fe-hemicarbonate 

The formation of Fe-Hc was studied at different equilibration times in samples containing 

less calcite than the samples used to prepare Fe-Mc. The XRD pattern shows the 

formation of an AFm phase, labeled Fe-Hc in Fig. 5 with a peak around 7.48 Å. It is 

known from Al-containing AFm phases that generally Al-Hc is formed first and converts 

to Al-Mc with time if calcite is present [2, 65]. As the solutions contain only calcium, 

iron, hydroxide and carbonate, it was tentatively concluded that the observed phase 

corresponds to Fe-hemicarbonate. After 180 days and longer, the formation of Fe-Mc was 

observed. In addition, significant quantities of calcite and portlandite were observed at all 

times. A reddish color indicated the presence of X-ray amorphous iron hydroxide. 

37

Intensity (arb. untits) 


Fe-Mc 

Fe-Hc 

C 2F 

10 11 12 13 14 

2θCuKα 

1 year 

180 days 

28 days 

7 days 

Fig. 5 Time-dependent XRD pattern of Fe-Hc (and Fe-Mc) synthesized at 20 °C; C2F: 

differentiated relative weight Weight loss in % 

100 

-0.1 

-0.2 

-0.3 

2CaOFe2O3, Fe-Mc: Fe-monocarbonate, Fe-Hc: Fe-hemicarbonate. 

90 

80 

70 

60 

Fe-Hc/Fe-Mc 

CH 

200 400 600 800 

Temperature °C 

C 

7 days 

180 days 

1 year 

Fig. 6 TGA and DTG curves of Fe-Hc formation at 20 °C for different equilibration times. CH: 

Portlandite, C: carbonates. 

Ecker et al. [70] also observed a peak at 7.49 Å when the sample was dried at 35% 

relative humidity, which they attributed to the formation of a triclinic Fe-Mc. They also 

found a peak at around 8.05 Å that was attributed to Fe-Hc. However, the assignments 

were done by interpolation of the data from the study of the 3CaO.Al2O3.CaCO3.11H2O - 

38


3CaO.Fe2O3.CaCO3.12H2O systems and they could not synthesize the phase without the 

presence of Al. 

In the system studied here, i.e free of Al, the peak at 7.48 Å shows no change with 

variation from 90 to 20% relative humidity in the XRD in situ climate chamber and no 

peak at 8.05 Å has been observed. The difference with previous results [70] may be 

explained by the absence of Al in the preparation. 

The large difference in the layer thickness of Fe-Hc (d = 7.48 Å) compared to Al- 

hemicarbonate (3CaO.Al2O3.(CaCO3)0.5.(CaO)0.5.12H2O) (d = 8.24 Å) indicates the 

presence of less water (and/or carbonate) in Fe-Hc than in Al-Hc. In the CaO-Al2O3- 

CaCO3-H2O system, the Al-Hc appears first and then disappears with time to form Al-Mc 

[2, 65]. Depending on the drying condition, the Al-Hc layer thickness may vary from 6.6 

Å (3CaO.Al2O3.(CaCO3)0.5.(CaO)0.5.6.5H2O) to 8.2Å 

(3CaO.Al2O3.(CaCO3)0.5.(CaO)0.5.12H2O) [71]. Based on the comparison with the Al 

system and on the TGA data after 180 days (Fig. 6) an interlayer water content of 3 to 4 

H2O molecules could be roughly estimated for the Fe-Hc phase investigated. Thus we 

may suggest that the peak at 7.48 Å can be attributed to Fe-Hc with an amount of water 

close to 10: 3CaO.Fe2O3.(CaCO3)0.5.(CaO)0.5.10H2O. However, the instability of this 

phase and its conversion to Fe-Mc did not permit structural investigation and the 

carbonate and water content could not be precisely determined. The formation of C4FH13 

would also have been possible, but again the measured interlayer thickness does not fit 

either the 7.94 Å reported for C4AH13 or the 7.35 Å reported for C4AH11 [71]. In addition, 

the amount of calcite clearly reduced after 180 days. 

Fig. 6 shows the TGA curve caused by the loss of weight from Fe-Hc at different 

39


equilibration times. After 180 days equilibration the main loss of weight was between 

110 and 180 °C, caused by the loss of 3-4 interlayer water molecules from Fe-Hc. 

However, after 1 year equilibration, the loss of weight in the same temperature range was 

due to loss of water from both Fe-Hc and Fe-Mc. 

3.1.3. Fe-monocarbonate 

3.1.3.1. Kinetics of formation 

Fig. 7 shows the XRD patterns of Fe-Mc at 20 °C as a function of equilibration time. The 

reaction of pure ferrite was slow and its counter was significantly lowered solely after 

120 days. Both Fe-Mc and Fe-Hc were present after 120 days. However, after 3 years, 

only Fe-Mc was observed with some traces of calcite and portlandite. The latter sample 

was further used for the structural determination (solution and refinement) of Fe-Mc 

reported below. In the XRD patterns a shoulder at around 11.39° 2θ was observed, in 

which the intensity decreased with drying. The synthesized solids had a slightly reddish 

color, suggesting the presence of small amounts of Fe-hydroxide not detectable by XRD. 

Intensity [arb . units] 

Fe-Mc 

* 

Fe-Hc 

C 2F 

3 years 

1 year 

120 days 

28 days 

7 days 

10 11 12 13 14 

2CuK 2CuK 

Fig. 7 Time-dependent XRD pattern of Fe-Mc formed at 20 °C. * unidentified 

40


C4FH13 (4CaOFe2O313H2O) was not observed in any of the experiments. The presence 

of calcite destabilized the C4FH13 phase, which led to the formation of carbonate 

containing Fe-AFm as previously reported [12, 15]. This phenomenon was also observed 

for the Al analogues [52, 72]. 

The thermogravimetric curve of Fe-Mc (Fig. 8) shows several weight loss between 80° 

and 800°C. The first mass losses below 240 °C indicates the loss of the 6 waters 

molecules from the interlayer of Fe-Mc as previously observed for Al-Mc [73]. The 

weight loss up to 500 °C is due the removal of the remaining 6 waters molecules from the 

main layer and decomposition of traces of portlandite. The weight loss at about 700 °C is 

due to the loss of CO2 from Fe-Mc and from calcite. The peak areas of calcite and 

portlandite were found to decrease with hydration time while the Fe-Mc peaks increased. 

This finding further substantiates that the formation of synthetic Fe-Mc was completed 

only after long hydration times. 

weight loss in % 

differentiated relative weight 

100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

Fe-Mc 

Fe-Mc 

CH 

200 400 600 800 

Temperature (°C) 

C 

7days 

120 days 

3 years 

Fig. 8. TGA and DTG curves of Fe-Mc formation at 20 °C for different equilibration times. CH: 

Portlandite, C: carbonates. 

41


3.1.3.2. Effect of temperature 

X-ray diffraction and the TGA measurements revealed that the formation of Fe-Hc and 

Fe-Mc was faster at 50 °C (Table 11). First indications of the formation of Fe-Mc were 

observed after 7 days of equilibration. The TGA and XRD data further revealed that after 

28 days and longer the intensities of portlandite and calcite increased while the peaks of 

Fe-Mc decreased, thus suggesting instability of Fe-Mc with regard to calcite, portlandite 

and Fe-oxide/hydroxide at higher temperature and longer equilibration time. At 50 °C 

and 80 °C the diffraction pattern also indicates the presence of hematite (Fe2O3) (Fig. 9). 

At 80 °C, neither Fe-Hc nor Fe-Mc was observed. Hematite, calcite and portlandite were 

the only phases identified. The color of the solid formed was dark red, confirming the 

presence of Fe-oxide/hydroxide [15]. The findings show that Fe-Mc is unstable at 80 °C 

and decomposes to Fe2O3, calcite and portlandite. 

Intensity (arb. units) 

Fe-Mc 

Fe-Hc 

CH 

Fe-Mc 

Fe 2O 3 

Fe-Hc 

CH 

10 20 

2CuK 2CuK 

30 

C 

Fe 2 O 3 

CH 

80 °C 

50 °C 

20 °C 

Fig. 9 Comparison of XRD pattern of Fe-Mc equilibrated for one year at 20, 50 and 80 °C. CH: 

portlandite, C: carbonate, Fe2O3: hematite. 

42


3.1.3.3. Structure of Fe-Mc 

The virtually single phase (absence of Fe-Hc phase) Fe-Mc sample obtained after 3 years 

reaction time was subject to crystallographic structure determination using synchrotron 

powder XRD. High quality diffraction data allowed the structure of Fe-Mc to be solved 

and refined. The sample was composed of 89 wt. % (weight percent) of the studied Fe- 

Mc phase with some impurities of calcite (11 wt. %). The crystal data and multi-pattern 

refinement (using data from two samples to detector distances) parameters are 

summarized in Table 6 while a Rietveld plot (corresponding to data from the sample to 

detector distance of 150 mm) is shown in Fig. 10. 

Table 6 Multi pattern refinement (from two sample-to-detector distances: 1/ 150 mm, and 2/ 350 

mm) and crystal data of Fe-Mc. 

Compound Iron monocarbonate 

Formula 3CaO·Fe2O3·CaCO3·12.18(4)H2O 

Structural formula [Ca2Fe(OH)6] + [½CO3·3.08(2)H2O] - 

Calculated formula weight (g.mol -1 ) 646.91 

T(K) 293 K 

System Rhombohedral 

Space group c 

R3 

a (Å) 5.9196 (1) 

c (Å) 47.8796 (10) 

V (Å 3 ) 1453.01 (4) 

Z / Dx (g cm -3 ) 6 / 2.22 

Wavelength (Å)1 0.720852 

Angular range 2 (°) 1, 2 3.14- 49.16, 2.50- 26.35 

Nobs 1, 2 1283, 1111 

Excluded regions (°) 5.40-5.61 and 9.03-9.71 

Nref 1, 2 288, 52 

Rp 1, 2 (%) 3.27, 3.89 

Rwp 1, 2 (%) 4.40, 5.37 

RBragg 1, 2(%) 4.38, 3.43 

RF 1, 2 (%) 4.73, 3.31 

N of profile parameters 20 

N intensity dependent parameters 15 

Fe-Mc was found to crystallise in the rhombohedral R3 c space group, i.e. the highest 

symmetry observed for AFm phases, which further corresponds to the symmetry of the 

high temperature (HT)-polymorph of Friedel’s salt [74, 75]. The structure is composed of 

43


a positively charged main layer [Ca2Fe(OH)6] + and a negatively charged interlayer 

[½CO3·3.08(2)H2O] - . The main layer contains of trivalent Fe 3+ cations in hydroxide 

octahedral coordination (i.e. substituting Al 3+ cations usually encountered in AFm 

phases) and bivalent Ca 2+ cations, which are seven-fold coordinated (6 hydroxyls + 1 

water molecule from interlayer). The structure of Fe-Mc can be described by 7 non-H 

atomic positions: one atomic position for iron, for calcium, for hydroxyl ions, for water 

molecule bonded to Ca 2+ (labeled Ow1), for water molecule weakly bonded in the center 

of the interlayer (labeled Ow2 with a refined partial occupancy of 0.36), for carbon atom 

from carbonate group (with a fixed partial occupancy of ½ in agreement with the 

electroneutrality of the compound) and for oxygen atoms from carbonate (labeled Oc). 

Fig. 10 Rietveld plot from powder pattern recorded with a sample-to-detector distance of 150 mm 

(red crosses are experimental data, black line is calculated pattern, blue line is the 

difference pattern, green sticks are Bragg peaks positions for Fe-Mc and calcite). 

The interlayer can be described in terms of a statistical distribution between one 

carbonate group and two water molecules. The refined composition is very close to 

44


3CaO·Fe2O3·CaCO3·12H2O, which is the composition determined by TGA. Atomic 

coordinates of the 7 crystallographic sites are indicated in Table 7, whereas interatomic 

distances are given in Table 8. 

Table 7 Fractional coordinate of non hydrogen atoms and isotropic displacement. 

Atom site X y z Ueq x 10 3 (A 2 ) occ 

Fe 6b 0 0 0 9.6 (8) 1.000 

Ca 12c 1/3 2/3 0.01134 (4) 8.3 (8) 1.000 

OH 36f 0.3863 (6) 0.4006 (5) 0.1447 (1) 1.7 (1) 1.000 

Ow1 12c 1/3 2/3 0.0635 (1) 36 (2) 1.000 

C 6a 0 0 1/4 65 (5) 0.5(-) 

Oc 18e 0 -0.2153(9) 1/4 = Ueq (C) 0.5(-) 

Ow2 18e 0 = y (Oc) 1/4 = Ueq (C) 0.368(8) 

Table 8 Selected interatomic distances (Å) in Fe-Mc. 

Atom atom Distances (Å) 

Fe 6 x OH 2.043(5) 

Ca 3 x OH 2.385 (8) 

3 x OH 2.472 (8) 

Ow1 2.50 (1) 

C 3 x Oc 1.275 (1) 

Ow1 ½ (3 x Oc) 2.58 (2) 

½ (2 x Ow2) 2.58 (2) 

(Oc,Ow2) 2 x Ow1 2.58 (2) 

2 x OH 3.273 (1) 

2 x OH 3.50 (1) 

Ow2 Ow2 “2.208 (1)” 

The accuracy of the refined structural model is reflected by the refined values for the 

interatomic distances in the main layer and the interlayer. The only unrealistic Ow2-Ow2 

distance of 2.208 Å is attributed to partial occupancies in this region of the structure: one 

carbonate anion is statistically distributed with two water molecules in the location at the 

center of interlayer. 

An equivalent situation (statistical distribution between one carbonate group and three 

water molecules) was described earlier in the case of the disordered 4 11 AcH C D 

structure [76]. The structure of Fe-Mc is presented in Fig. 11, showing a general 

45


representation (Fig. 11a) and details of the network of hydrogen bonding assuming the 

cohesion between main and interlayer (Fig. 11b). 

a. 

b. 

Fig. 11a. Projection of the Fe-Mc structure along b axis (the interlayer part of the structure is 

ordered for clarity; i.e. the statistical distribution between one carbonate and two water 

molecule has been alternatively ordered). b. 3D cohesion in Fe-Mc structure 

(representation of the main hydrogen bonds). 

46


EXAFS spectroscopy was carried out to obtain information on the local arrangement of 

Fe in Fe-Mc. Synchrotron-based X-ray absorption spectroscopy is a local probing 

technique, which provides information on the coordination environment of the X-ray 

absorbing atom within a distance of up to ~ 5 Å. Experimental and theoretical Fourier 

transforms (modulus) obtained from the spectrum are shown in Fig. 12 while the 

structural parameters are summarized in Table 9. The central atom Fe has six neighboring 

O atoms at a distance of 2.02 Å and six neighboring Ca atoms at 3.47 Å. The former 

finding confirms that Fe is octahedrally coordinated in Fe-Mc. Furthermore, the Fe-O and 

Fe-Ca distances agree with the refined XRD data (2.04 and 3.46 Å). The absence of any 

Fe-Fe backscattering contributions, which would be at 3.01 Å [15, 77], suggesting that, if 

at all, Fe-hydroxide was present in the 3 years old sample below the detection limit of the 

method (~ 5 wt%). 

Fourier tansform magnitude 

K 3 

2 4 6 8 10 

k Å-1 0 2 4 

R +R( Å ) 

( 

) 

Experimental 

Modeled 

Fig. 12. Fe K-edge EXAFS data of Fe-Mc: Experimental (solid line) and theoretical (dots) 

Fourier transform (modulus) obtained from k 3 -weighted, normalized, background 

subtracted spectrum (inset). 

47


Table 9 EXAFS structural parameters of Fe-Mc equilibrated for three years. 

Atomic pair N R(Å) R(Å) from XRD σ 2 (Å 2 ) ΔE0 (eV) R-factor 

Fe-O 6.0 a 2.02 2.04 0.006 1.18 0.06 

Fe-Ca 6.0 a 3.47 3.46 0.008 1.18 

N: Coordination number of the neighboring atom (uncertainty ± 20%); a fixed parameter 

R: Distance to the neighboring atom (uncertainty ± 0.02Å) 

σ: Debye-Waller factor 

ΔE0: inner potential correction 

R-factor: deviation between experimental data and fit 

Raman spectra from Fe-Mc confirmed the refined structure of Fe-Mc, namely the 

interlayer description (Fig. 13). The symmetric stretching band of carbonate [CO3] 

groups was observed at 1085 cm -1 (Fig. 13a). In a recent study the 1085 cm -1 value for 

the carbonate 1 mode was attributed to carbonate weakly bonded at the centre of 

interlayer [78]. This mode of vibration is clearly shifted from 1068 cm -1 as observed in 

the case of carbonate bonded to the main layer of, for example, Al-Mc. The broad and 

unresolved band of vibration observed in the frequency range 2800 cm -1 – 4000 cm -1 

characterizes a disordered interlayer region (Fig. 13b). The hydrogen bond network is not 

well defined neither in space due to statistic disorder nor in time due to dynamical 

disorder (liberation of carbonate group around the trigonal axis and/or movement of 

weakly bonded Ow2 water molecules). The latter observations have previously been 

reported for AFm phases [79, 80]. 

48

a). 

b). 


Fig. 13 a) Raman spectra on Fe-Mc in the frequencies range 200 cm -1 – 1800 cm -1 b) Raman 

spectra on Fe-Mc in the frequencies range 2800 cm -1 – 4000 cm -1 . 

ESEM micrographs of the synthesized solids showed a platy crystal with hexagonal 

symmetry (Fig. 14). This indicates a preferred orientation of the crystals might formed as 

previously observed for Al-containing AFm phases [71, 81] 

49


Fig. 14 SEM micrographs of Fe-Mc. 

3.1.3.4. Comparison of pure Fe- and Al-Mc 

The Fe-Mc and analogous Al-Mc compounds exhibit different symmetries. The Fe-Mc 

structure is represented by the highly symmetric R3 c space group whereas the Al-Mc 

structure is described by the triclinic symmetry: one ordered structure described in the P1 

space group [66] and one disordered structure in the P 1 space group [76]. The two 

monocarbonate analogues have different layer spacings i.e about 7.98 Å for Fe-Mc and 

about 7.57 Å for Al-Mc. This difference is attributed to the location of carbonate anions. 

CO3 2- anions are bonded to the main layer in Al-Mc as one of the three oxygen atoms of 

the carbonate group occupies the seventh coordination position of a seven fold 

coordinated Ca 2+ cation from the main layer. In the case of Fe-Mc, however, carbonate 

anions are located in the center of the interlayer, weakly bonded via hydrogen bonds in a 

position parallel with the main layer (Fig. 11). Such pronounced differences in symmetry 

and carbonate locations are expected to be incompatible with the existence of a complete 

solid solution series expressed by Ca4[(AlxFe1-x)2(OH)12]CO3(6-x)H2O. The unit cell 

50


parameter a of Fe-Mc is at 5.92 Å somewhat larger than for Al-Mc (a = 5.78 Å) due to 

the larger ion radius of iron (0.64 Å) compared to aluminum (0.54 Å) [82]. 

The TGA/DTG results obtained from Al-Mc and Fe-Mc in Fig. 15 allow the following 

steps to be distinguished: 

Ca4[(AlxFe1-x)2(OH)12]CO35-6H2O → Ca4[(AlxFe1-x)2(OH)12]CO3 + 5-6H2O (interlayer 

water removal) 

Ca4[(AlxFe2-x)(OH)12]CO3 → Ca4[(AlxFe1-x)2O6]CO3 + 6H2O (dehydroxylation) 

Ca4[(AlxFe2-x)O6]CO3 → Ca4[(AlxFe1-x)2O7] + CO2 (decarbonation) 


100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

Al-Mc 

Fe-Mc 

inter layer water removal 

Al-Mc 

Al-Mc 

Fe-Mc 

dehydroxilation 

CH 

200 400 600 800 


C 

decarbonation 

Fe-Mc 

Al-Mc 

Fig. 15 Thermal analysis (DTG and TGA) of Ca3(AlxFe1-x)2O3.CaCO3.nH2O. 

From the TGA analysis the amount of the interlayer water of Fe-Mc was calculated. The 

number of interlayer water molecules per unit cell was found to be approximately 5.8 

resulting in total water content of 11.8. On heating Ca3[Fe2(OH)6]2CaCO36H2O, all the 

molecules of water molecules from the interlayer were released around 240 °C which 

agrees with the findings of Ecker et al. [12] and with the amount of interlayer water 

51


determined by XRD. The weight loss up to 500 °C indicates the dehydroxylation of the 

water molecules associated with the main layer and traces of portlandite. Finally, the 

weight loss at around 700 °C represents decarbonation of calcite and Fe-Mc. The 

interlayer water of Ca3[Al2(OH)6]2.CaCO3.5H2O was removed at somewhat lower 

temperatures than in the case of Fe-Mc, i.e. in the range between 80 and 270 °C in several 

steps. The number of interlayer water molecule of Al-Mc was found to be around 5.2. 

The IR results of Fe-Mc and Al-Mc are summarized in Fig. 16 and Table 10. The IR 

spectrum of Fe-Mc are correlated with the Al-analogues spectra to assign the type of 

bonds at different absorption bands and compared with the study from Ecker et al. [12]. 

The IR bands of Al-Mc have been assigned based on the study of Fischer et al. [71] and 

Trezza et al. [83]. The IR frequencies at 670 cm -1 , 817 cm -1 and 952 cm -1 are due to the 

vibrations of the AlO6 bond in the main layer. The strong IR frequencies at 1360 cm -1 and 

1415 cm -1 are bands attributed to the asymmetric stretching vibration of ٧3-CO3 2- while 

the sharp peak at 880 cm -1 is related to the asymmetric starching vibration of ٧2-CO3 2- . 

Fig. 16 IR spectra of Al-Mc and Fe-Mc. 

52


The absorption band in the range between 3000 cm -1 and 3600 cm -1 is due to OH 

stretching vibrations resulting from the interlayer water. In the case of Fe-Mc the weak 

bands at 661 cm -1 and 960 cm -1 are attributed to FeO6 vibrations. Like the Al-analogues, 

٧2-CO3 2- vibrations in Fe-Mc spectrum are found at about at 710 cm -1 and 875 cm -1 

respectively. The sharp band at 1382 cm -1 is related to the vibration of ٧3-CO3 2- . The 

broad band between 2700 cm -1 and 3320 cm -1 is due to the vibration of OH bonds in the 

interlayer water of Fe-Mc. The bands at frequencies higher than 3600 cm -1 can be related 

to the vibrations of OH in the main layer [Ca2.Al/Fe(OH)12] + . The peaks between 3000 

cm -1 and 3600 cm -1 are sharper for Al-Mc than Fe-Mc indicating more highly coordinated 

interlayer water in Al-Mc. 

Table 10 IR vibrations of Ca4[(AlxFe1-x)2(OH)12] . CO3 . nH2O. 

Al-Mc Fe-Mc 

Wavenumbers (cm -1 ) Vibrations Wavenumbers (cm -1 ) Vibrations 

670 AlO6 661 FeO6 

720 4-CO3 2- 710 4-CO3 2- 

817 AlO6 

880 2-CO3 2- 875 2-CO3 2- 

952 AlO6 960 FeO6 

1360 3-CO3 2- 1382 3-CO3 2- 

1415 3-CO3 2- 

1651 2-H2O 

3007 1-H2O 

3369 1-H2O 2700-3320 1-H2O 

3518 OH - 3505 OH - 

3616 OH -a 

3668 OH -a 3653 OH -a 

a 

associated to the main layer 

3.1.4. Mixed CaO.(AlxFe1-x)2O3.CaCO3.nH2O systems 

Variations of the Al/Fe ratio in the 3CaOAl2O3CaCO311H2O - 

3CaOFe2O3CaCO312H2O system resulted in the formation of two separate stable 

53


phases (no systematic peak shifts) while no intermediate phase formed (Fig. 17). The 

intensity of the small intermediate peak at 11.39° 2θ was found to depend on the water 

content as it decreased upon drying. To index the reflections and to determine the layer 

thickness and the unit cell parameter a, a Le Bail fitting was performed. The results are 

shown in Fig. 18 and Fig. 19, respectively. No significant variation of both parameters 

with varying Al mole fraction was observed compared to end member values. These 

results clearly confirm the absence of a solid solution between Al-Mc and Fe-Mc, in 

contrast to the findings reported elsewhere [70]. Thus, comparison of the XRD pattern 

further corroborate that 3CaOAl2O3CaCO311H2O and 3CaOFe2O3CaCO312H2O have 

different structural symmetries, carbonate location and different amounts of water 

molecules in the interlayer. 

Intensity Intensity [arb. [arb. units] 

Fe-Mc 

* 

Al-Mc 

1.0 

0.7 

0.5 

10 11 12 2CuK 2CuK 

13 0.3 

0.0 

14 

Al/(Al+Fe) ratio 

Fig. 17 XRD pattern of the Al/Fe-Mc after 3 years hydration time at 20 °C * peak due to 

additional water in Mc. 

54

o 

layer thickness (A) 

8.0 

7.8 

7.6 


Fe Mc (this study) 

Fe Mc (Ecker, 1997) 

Al Mc O(Renaudin, 1999) 

Al Mc D(Renaudin, 1999) 

Al/Fe Mc (this study) 

7.4 

0.0 0.2 0.4 0.6 0.8 1.0 

C FcH 4 12 


C AcH 4 11 

Fig. 18 Layer thickness for Al-Mc and Fe-Mc after refinement by Le Bail fitting and Rietveld 

unit cell parameter a (A) 

° 

analysis. C4FcH12: Fe-Mc, C4AcH11: Al-Mc. 

5.95 

5.90 

5.85 

5.80 

Fe Mc (this study) 

Fe Mc (Ecker, 1997) 

Al Mc (Renaudin, 1999) 

Al/Fe Mc (this study) 

5.75 

0.0 0.5 1.0 

C FcH 4 12 


C AcH 4 11 

Fig. 19 Values of a-parameters for Al-Mc and Fe-Mc. 

3.1.5. Solubility 

The composition of the solutions in equilibrium with pure Al-Mc, Fe-Mc, Fe-Hc and 

mixtures were determined at different equilibration times and at 20 °C and 5°C (Table 11 

and Table 12). The very low Fe concentrations detected in solution indicated the presence 

55


of an additional phase, presumably Fe-hydroxide. Note that the presence of Fe- 

oxide/hydroxide is consistent with the slightly red coloring of the samples. 

3.1.5.1. Determination of solubility products at 20 °C and 50 °C 

The measured concentrations of calcium, hydroxide, aluminum and iron were used to 

calculate the solubility products using GEMS: 

KS0 Al-monocarbonate = {Ca 2+ } 4. {Al(OH)4 - } 2 . {CO3 2- } . {OH - } 4 . {H2O} 5 

KS0,Fe-monocarbonate = {Ca 2+ } 4 . {Fe(OH)4 - } 2 {CO3 2- } . {OH - } 4. {H2O} 6 

KS0,Fe-hemicarbonate = {Ca 2+ } 4 . {Fe(OH)4 - } 2. {CO3 2- } 0.5 {OH - } 4. {H2O} 4 

where {} denotes the activity. 

As dissolved carbonate, CO3 2- , could not be determined precisely, the concentration of 

CO3 2- was calculated based on the assumption that all solutions were in equilibrium with 

calcite. The calculated solubility products are listed in Table 11 and Table 12. 

The total solubility products at 20 °C were determined to be log KS0 Al-Mc = -31.55, log 

KS0 Fe-Mc = -34.51 ± 0.50 and log KS0 Fe-Hc = -30.55 ± 0.67. The value determined for Al- 

Mc at 20 °C is nearly identical to the value of -31.47 reported by Matschei et al. [52] at 

25 °C. Möschner et al. [15] reported a tentative solubility product of -35.9 for Fe-Mc at 

20 °C, which is one log unit lower than the value determined in this study. 

At 50 °C a solubility product of log KS0 = - 35.27 ± 0.17 was determined for Fe-Mc and a 

log KS0 = -32.58 ± 0.61 for Fe-Hc. Thus, the solubility products at 50 °C are lower than 

those determined at 20 °C. Even though Fe-Mc was stabilized at 50 °C, the phase became 

unstable with time with respect to Fe-hydroxide, calcite and portlandite. At 80 °C both 

Fe-Mc and Fe-Hc were found to be unstable with respect to calcite and hematite (Fe2O3). 

56


Table 11 Measured ion concentrations and calculated solubility products at different equilibration times. 

Equilibration 

time in days Temperature Ca[mmol/l] Al[mmol/l] Fe[mmol/l] K[mmol/l] pH 1 logKs0 logKs0 

Fe-Hc 

Solid phases present Fe-Mc Fe-Hc 

7 20 5.18


Table 12 Compositions of Al/Fe-monocarbonate after synthesis at 20 °C equilibrated for 3 years at supersaturated and undersaturated condition. 

Mole fraction of Al in Al [mmol/l] Ca [mmol/l] Fe [mmol/l] K [mmol/l] pH 

the solids 

1 Solids present logKs0 logKs0, 

Supersaturation 

Al-Mc Fe-Mc 

1 2.08 0.12


Additionally, the solutions equilibrated at 50 °C and 80 °C were thermodynamically 

oversaturated with respect to Fe-hydroxide and hematite. The kinetics of formation of 

hematite, however, was very slow, so that the solid did not form at 20 °C or 50 °C within 

the experimental period. 

For both Al-Mc and Al-Hc, an increase in the solubility with increasing temperature was 

observed [52], while in contrast, for both Fe-Hc and Fe-Mc a slight decrease was 

observed suggesting a slight stabilization of Fe-Hc and Fe-Mc (Fig. 20). 

3.1.5.2. Estimation of the solubility product under standard conditions 

The solubility products at standard conditions were calculated with the help of GEMS- 

PSI using temperature extrapolation from the solubility products calculated at 20 °C and 

50 °C as previously demonstrated by Matschei et al. [7]. The calculated solubility 

products at 20 °C and 50 °C (Table 11 and Table 12) were used to develop the 

temperature-dependent ‘log K’ function which allowed the solubility products to be 

calculated at different temperatures (Fig. 20) as described in section 2.3. The entropy was 

adjusted until good agreement between measured and calculated solubility was reached. 

The thermodynamic properties of the solids at 25 °C are listed in Table 13. 

59

Calc.solubility product log K sp 

-29 

-30 

-31 

-32 

-33 

-34 

-35 

-36 


Fe-Mc 

Fe-Hc 

-37 

5.0 25.0 45.0 


65.0 85.0 

Fig. 20. Calculated solubility products of Fe-Mc and Fe-Hc from the solubility experiments. 

Squares: experimental solubility product of Fe-Hc, Triangles: experimental solubility 

product of Fe-Mc. 

The solubility products of Fe-Mc (-34.59) and Fe-Hc (-30.83) are about 2-3 log units 

lower than those of Al-Mc (-31.47) and Al-Hc (-29.12), respectively. Note that a similar 

difference in the solubility products between Al- and Fe-monosulfate was found (see next 

section), while the solubility product of Fe-ettringite (-44.0) was found to be slightly 

higher than that of Al-ettringite (-44.9). This indicates that the Fe-AFm phases are 

potentially stable under conditions where Al-AFm can be formed while the formation of 

Fe-ettringite is less probable under conditions where Al-ettringite is stable. 

60

Phase log KSo 


Table 13 Thermodynamic parameters of carbonate containing AFm phases at standard conditions 

(25°C, 1 atm). 

fG° 

[kJ/mol] 

fH° 

[kJ/mol] 

S 0 

[J/K/mol] 

C 0 p 

[J/mol/K] 

a0 

[J/(mol.K)] 

a1 

[J/(mol.K 2 )] 

a2 

[JK/mol] 

a3 

[J/(mol.K 0.5 )] 

V 0 

[cm 3 /mol] Ref. 

C4FcH12 -34.59±0.5 -6674.0 -7485 1230 950 612 1.160 -5.73e+05 0 292 a 

C4Fc0.5H10 -30.83±0.5 -5952.9 -6581 1270 841 308 1.200 -9.08e+05 3200 273 a 

C4AcH11 -31.47±0.5 -7337.5 -8250 657 881 618 0.982 -2.59e+06 0 262 b, c 

C4Ac0.5H12 -29.13±0.5 -7336.0 -8270 713 906 664 1.014 -1.30e+06 -800 285 b, c 

(a) This study, (b) Lothenbach et al. [3], (c) Matschei et al. [52]. 

3.1.5.3. Modeling of mixed CaO(AlxFe1-x)2O3CaCO3nH2O systems 

Table 12 shows the composition of the liquid phase with varying Al mole fraction. The 

concentrations of all species (Ca, Al, Fe, K) between 0.1 and 0.9 mole fraction of Al in 

the 3CaO(AlxFe1-x)2O3.CaCO3.nH2O system were found to be similar within the 

uncertainty range, indicating the presence of two separate solid phases (Table 12). Fig. 21 

shows that the calculated and measured concentrations are comparable on the assumption 

that two separate phases are present. This supports the findings from XRD, which 

showed no solid solution between the Al- and Fe-Mc end members. Note that the 

presence of a solid solution would be indicated by gradual changes in the elemental 

concentrations (see dotted line in Fig. 21). 

61

log c(Ca, Al, Fe, K, OH) in mmol/L 

3 

1 

-1 

-3 


-5 

0 0.2 0.4 0.6 0.8 1 

C4FcH11 Al/(Al+Fe) ratio 

C4AcH12 Fig. 21 Measured (symbols) and calculated (lines) concentrations in the liquid phases of the 

Ca 

Al 

Fe 

K 

OH- 

synthesized monocarbonate at different Al/Al+Fe ratios. 

3.1.6. Modeling of C3A-C2F-CaCO3-CaSO4-H2O system in cement hydration 

Thermodynamic modeling was used to calculate the changes of the hydrate assemblage in 

the system C3A-C2F-CaCO3-CaSO4-H2O (SO3/(Al,Fe)2O3 = 1) in the absence and 

presence of CaCO3 using the thermodynamic data given in Table 5 with the aim of 

assessing differences in the properties of Fe-and Al-analogues and the effect of calcite. In 

the model calculations a fixed SO3/(Al,Fe)2O3 ratio of 1 was used while the calcite 

content (CO2/(Al,Fe)2O3 ratio) varied. A constant amount of solids (Al2O3 + Fe2O3 + 

CaSO4 + CaO + CaCO3) was maintained. 

62

volume [cm 3 ] 

700 

600 

500 

400 

300 


200 

Al-Ms 

100 

0 

Fe-Ms 

Al-Hc 

Fe-Hc 

Fe-Mc 

0.0 0.2 0.4 0.6 0.8 1.0 1.2 

CO 2 /(Al,Fe) 2 O 3 

Al-Ett 

portlandite 

calcite 

Fig. 22 Changes in the total volume of phases of a hydrated model mixture consisting of Al2O3, 

Fe2O3 and a fixed SO3/(Al,Fe)2O3 ratio of 1 as a function of the calcite content 

(CO2/(Al,Fe)2O3 ratio) at 20 °C at constant amount of solids: (Al2O3 + Fe2O3 + CaSO4 + 

CaO + CaCO3). 

In the absence of calcite, Al- and Fe-monosulfate were calculated to be stable in presence 

of small amounts of portlandite (Fig. 22). Upon sequential addition of calcite, first Fe-Hc 

and Al-ettringite are expected to form at the expense of Al- and Fe-Ms. Appearance of 

Al-Hc, Fe-Mc and finally Al-Mc occurs with increasing supply of carbonate. The 

influence of calcite on the composition of the phase assemblage of a hydrated model 

Portland cement is similar to that on a pure Al-system [67]. In the presence of calcite, 

monosulfate (Al-Ms and Fe-Ms) is expected to be unstable, and ettringite (Al-Ett) and 

monocarbonate (Fe-Mc and Al-Mc) form instead, which leads to a higher degree of space 

filling (Fig. 22). In contrast to a pure Al-system, however, Fe-ettringite is not expected to 

be stable in a mixed system, which results in Fe-Hc and Fe-Mc. Note that at higher 

CaSO4 contents (SO3/(Al,Fe)2O3 > 2), Fe-ettringite can be stable in a Portland cement 

Al-Mc 

63


system. During the first hours of the hydration of Portland cements, an excess of CaSO4 

is present, which allows both Al- and Fe-ettringite to be formed. With progressing 

hydration, however, the CaSO4/(Al,Fe)2O3 is reduced, which results in a situation where 

only Al-ettringite will be stable and both Al- and Fe-monocarbonate (or monosulfate in 

the absence of calcite) can be formed. The predicted presence of Fe-containing ettringite 

during the early hydration is consistent with the observations of Neubauer et al. [84]. The 

authors observed a peak shift of ettringite to a larger d-value during the first hours of 

OPC hydration, which could indicate the presence of Fe- (or of CO3) in the ettringite 

structure. After a few hours, however, d-values decreased, thus suggesting the 

transformation to Al-ettringite. 

Note that the presence of silicates was not included in the current assessment because 

data on the stability of Fe-containing Si-hydrogarnets (C3FSH4) and Fe-strätlingite 

(C2FSH8) were not yet available. There are indications, however, that the formation of 

Fe-containing Si-hydrogarnets is kinetically possible at ambient temperatures. 

3.1.7. Conclusions 

A crystalline Fe-Mc was synthesized by mixing appropriate amounts of C2F, CaCO3 and 

CaO at 20, 50 and 80 °C. At ambient temperature, the kinetics of the reaction was slow; 

C2F transformation was completed only after one year and longer. After 3 months of 

equilibration, Fe-Hc was detected; after 1 year and longer Fe-Hc transformed to Fe-Mc. 

At 50 °C the kinetics were found to be faster. The presence of Fe-Hc and some Fe-Mc 

was already observed after 7 days. The amount of Fe-Mc increased while Fe-Hc 

disappeared over time. At 80 °C Fe-Mc was unstable with respect to Fe- 

hydroxide/hematite, portlandite and calcite. 

64


The structure of Fe-Mc was solved and refined using synchrotron powder diffraction 

data. Fe-Mc was described in the rhombohedral R3 c symmetry. It belongs to the AFm 

family with a positively charged main layer and a negatively charged interlayer. The 

structure of Fe-Mc was found to diverge from its Al analogous Al-Mc compound. The 

main difference consists in the carbonate location. Carbonate is bonded to the main layer 

in Al-Mc while it is weakly bonded in the interlayer of Fe-Mc. 

EXAFS spectroscopy data supported the formation of stable Fe-Mc in which iron is 

octahedrally surrounded by six oxygen and calcium atoms. The Fe-Ca backscattering 

contributions revealed that Fe is associated with the Fe-Mc structure while the absence of 

Fe-Fe backscattering contributions in the synthesized material confirmed the absence of 

significant amounts of amorphous Fe-hydroxide [77]. The coordination environment of 

Fe in Fe-Mc corresponds to that of Al in the Al-analogue [77]. 

TGA and IR measurements revealed characteristics of the interlayer water molecules of 

Fe-Mc and Al-Mc. The weight loss of the interlayer water molecules observed by TGA 

was found to occur in the temperature range around 200 °C for Fe-Mc and from 80 to 270 

°C for Al-Mc. The IR spectra of Fe-Mc further showed vibrations of bonds that are on 

same range of frequency with those of Al-Mc but with different shape of the peaks 

composed of the spectra of Al-Mc. 

In the mixed Ca4[(AlxFe1-x)2(OH)12].CO3.nH2O system XRD data and measurements of 

the elemental composition in solution were not consistent with the formation of a solid 

solution between Al- and Fe-Mc, presumably due to the structural differences between 

Al- and Fe-Mc. Al-Mc has triclinic structure and a layer thickness of 7.57 Å, while Fe- 

Mc has a rhombohedral structure and a layer thickness of 7.98 Å. 

65


The solubility products were determined experimentally at 20 °C for Fe-Mc (logKS0,20 °C 

= -34.51), for Fe-Hc (-30.55) and for Al-Mc (-31.55). At 50 °C, the logKS0,50 °C for Fe-Mc 

was -35.27 and for Fe-Hc -32.58. At standard conditions (25 °C, 1 atm) a logKS0 of - 

34.59 for Fe-Mc and of -30.83 for Fe-Hc was determined. Thus, the experimentally 

derived solubility products of Fe-Mc and Fe-Hc were approximately 3 log units lower 

than those reported for Al-Mc and Al-Hc. 

Thermodynamic modeling further indicates that in a system containing C3A-C2F-CaSO4- 

CaO-H2O (absence of calcite), Al-Ms and Fe-Ms are expected to dominate the phase 

assemblage. In the presence of CaCO3, however, Al-ettringite, Al-Mc and Fe-Mc are 

expected to be stable, while Fe-ettringite will not be present. Only at higher 

SO3/(Al,Fe)2O3 ratios (>2), Fe-ettringite was predicted to be stable. High SO3/(Al,Fe)2O3 

ratios are achieved during the first hours of OPC hydration when only small amounts of 

the aluminate and ferrite clinkers have reacted. Hence, Fe-ettringite could potentially 

form in this stage of the hydration process. In the later stages, however, when a lower 

SO3/(Al,Fe)2O3 ratio is achieved, Fe-Mc (or Fe-Ms in the absence of calcite) are expected 

to be stable, together with Al-ettringite and Al-Mc (or Al-Ms). 

In summary, no experimental data have been available to date which allowed the 

formation of Fe-containing phases might exist in hydrated cements. The data obtained in 

this study offer a possibility to predict the fate of iron oxides in Portland cement. Based 

on the available data, iron oxide can be expected to be present in hydrated cements rather 

as AFm phases than as ettringite. 

66


3.2. Fe-containing monosulfate 


The formation of monosulfate (Ca4(Al,Fe)2(OH)12SO46H2O) in cement paste has been 

extensively studied [32, 52, 85, 86]. Mainly the formation Al-monosulfate 

(Ca4Al(OH)12SO46H2O) from C3A and gypsum or Al-ettringite was reported. Al- 

monosulfate can also be formed from the reaction of C2(A,F). Moreover, the presence of 

iron might lead to the formation of Fe-monosulfate (Ca4Fe(OH)12SO46H2O) or mixed 

Al/Fe-monosulfate might formed. The formation of Fe-monosulfate in the absence of 

cements has been reported [12, 15]. 

Experimental thermodynamic data are available for Al-monosulfate [7, 61]. Only a rough 

estimation of the thermodynamic data has been reported for Fe-monosulfate from Fe- 

ettringite experiments where the formation of Fe-monosulfate was observed at high pH 

values [15]. Further, the crystal structure of Fe-monosulfate is poorly understood. 

In this section Fe-containing monosulfate and mixed monosulfate containing both Al and 

Fe were synthesized and characterized. Their crystal structure and their solubility were 

determined. 

3.2.2. Kinetics of formation 

Fe-monosulfate (C4FsH12) is stable at high pH values [15]. In this study, C4FsH12 was 

synthesized in 0.4 M KOH and studied after different equilibration times. An attempt to 

synthesis Fe-monosulfate in 0.1 M KOH (using stoichiometric amounts of C2F, CsH2 and 

C) resulted in Fe-ettringite formation only after an equilibration time up to 1100 days. 

Fig. 23 shows the XRD patterns of the phase synthesized at 20 °C as a function of 

equilibration time. The reaction of pure ferrite was slow and the formation of C4FsH12 

67


was observed after 120 days by a peak at 2θ = 9.96° giving rise to basal spacing d = 8.88 

Å. Portlandite co-precipitated with the target phase at all equilibration times, gypsum was 

present up to 28 days. The formation of X-ray-amorphous Fe-oxide/hydroxides was 

easily recognized from the reddish color of the samples. As stoichiometric amounts of 

iron, calcium and sulfate had been mixed, the precipitation of portlandite and Fe- 

oxide/hydroxide seems to have been kinetically faster than the formation of Fe- 

monosulfate. No gypsum or any other sulfate bearing phase was observed in any of the 

sample older than 120 days due to the high dissolved sulfate concentration present in the 

solutions (see section 3.2.7). 

Intensity [arb. units] 

C 4FsH 12 

C 2F 

CH 

10 20 30 

2CuK 

C 4FsH 12 

C 4FsH 12 

Fig. 23 XRD pattern of C4FsH12 formed at 20 °C after different equilibration times. 

C 2F 

CH 

680 days 

360 days 

120 days 

28 days 

7 days 

The TGA-DTG curves indicate the removal of the interlayer water in several steps (see 

Fig. 24). The trend is similar as for Al-monosulfate. The weight loss up to 300 °C is due 

68


to the loss of water molecule from the interlayer and the main layer of C4FsH12. A weight 

loss due to the presence of portlandite (CH) was observed at around 450 °C. The TGA- 

DTG curves also show the existence of traces of calcite in the sample after 7 days 

equilibration time, probably due to CO2 contamination during drying and the presence of 

unreacted gypsum during the first 28 days. 

differentiated relative weight weight loss in % 

100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

C 4 FsH 12 

CsH 2 

CH 

Carbonate 

200 400 600 800 


7 days 

28 days 

120 days 

340 days 

680 days 

Fig. 24 TGA and DTG curves of C4FsH12 formation at 20°C after different equilibration times. 

3.2.3. Effects of temperature 

Fe-monosulfate (C4FsH12) formation at 50 °C and 80 °C is much faster than at 20 °C. At 

50 °C, the Fe-monosulfate remained stable up to 360 days. At 80 °C, Fe-monosulfate 

decomposed to portlandite and Fe-hydroxide and finally to hematite after 28 days and 

longer. Fig. 25 shows the XRD pattern of the samples equilibrated for 360 days at 

different temperatures, revealing the instability of C4FsH12 with respect to portlandite and 

69


hematite at 80 °C. For the sample prepared at 50 °C, besides to C4FsH12 a small quantity 

of C4FsH10 was observed. 


C 4 FsH 12 

C 4 FsH 10 

CH 

C 4 FsH 12 

C 4 FsH 12 

CH 

10 20 30 

2CuK 

C 2 F 

Fe 2 O 3 

Fig. 25 XRD pattern of C4FsH12 equilibrated for 360 days at 20, 50 and 80 °C. 

3.2.4. Structure of C4FsH12 

CH 

80 °C 

50 °C 

20 °C 

The laboratory XRD patterns showed that well crystallized Fe-monosulfate (Fig. 23 and 

Fig. 25) had formed both at 20 °C and 50 °C. The solids synthesized at 20 °C 

(equilibrated for 680 days) and 50 °C (equilibrated for 360 days) were used for 

crystallographic investigation on the structure of Fe-monosulfate using synchrotron X-ray 

powder diffraction. The structure of Fe-monosulfate was described by multipattern 

Rietveld refinement as shown in Fig. 26. The refined data show the formation of multiple 

phases for both samples synthesized at 20 °C and 50 °C. The major phase at both 

temperatures was portlandite, while the amount of amorphous Fe-hydroxide could not be 

70


quantified. The solid synthesized at 20 °C is composed of two types of AFm phases (38 

wt.% and 6 wt.%) and Ca(OH)2 (53 wt.%). The solid at 50 °C is composed of the same 

two types of AFm phases (37 wt.% and 4 wt.%) and Ca(OH)2 (45 wt.%). Details of the 

quantitative analyses extracted from Rietveld refinement are given in Table 14. 

Table 14 Quantitative phases analyses from Rietveld refinement. 

Compound Fe-Ms-20°C (wt. %) Fe-Ms-50°C (wt. %) 

Fe-Ms 38 37 

AFm N°2 6 4 

Ca(OH)2 53 45 

Fe2O3 3 8 

CaCO3 – vaterite - 4 

CaCO3 – calcite - 2 

The main Fe-monosulfate peaks indicates an interlayer distance of 8.87 (1) Å 

corresponding to C4FsH12. The structural parameters of C4FsH12 were refined, using the 

C4AsH12 structure [87] as a model. The refined parameters of Fe-monosulfate reveal that 

the phase crystallizes in the rhombohedral R3 c symmetry with a = 5.8874 (3) Å and c = 

26.598 (3) Å at 20 °C and a = 5.8832 (2) Å and c = 26.6181 (1) Å at 50 °C. The structure 

is composed of a positively charged main layer [Ca2Fe(OH)6] + and a negatively charged 

interlayer [ 1 /2SO4 . 3H2O] - . The atomic coordinates of the eight crystallographic sites are 

given in Table 15. 

Table 15 Refined structural parameters of Fe-monosulfate (standard deviation in parentheses). 

Atom Wyckoff X Y Z Biso (Å 3 ) Occupancy 

Ca 6c 2/3 1/3 0.0220 (3) 0.7 (2) 1 

Fe 3ª 0 0 0 = Biso (Ca) 1 

O1 (OH) 18f 0.283 (4) -0.041 (4) 0.0337 (7) = Biso (Ca) 1 

O2 (H2O) 6c 2/3 1/3 0.1156 (9) 1.5 (4) 1 

S 3b 0 0 ½ = Biso (O2) 1/2 (-) 

O3 (SO4) 18f 0.245 (5) 0.090 (9) 0.474 (1) = Biso (O2) 0.25 (-) 

O3’ (H2O) 18f = x (O3) = y (O3) = z (O3) = Biso (O2) 0.20 (2) 

O4 (SO4) 6c 0 0 0.555 (1) = Biso (O2) 1/4 (-) 

71


Fig. 26. Rietveld plot for Fe-monosulfate samples (synthesized at 20 °C: top, and at 50 °C: 

bottom) with = 1.5418Å. 

72


The presence of a second AFm phase was indicated by an intensive diffraction peak at 2θ 

= 10.7° (Fig. 27). This could indicate the presence of a second Fe-monosulfate compound 

with less water. The second Fe-monosulfate has an interlayer distance of 8.36(1) Å which 

could correspond to C4FsH10. The decrease in the interlayer distance from C4FsH12 (8.87 

Å) to C4FsH10 (8.36 Å) is comparable to the difference in interlayer distance reported for 

C4AsH12 (8.94 Å) [88] and C4AsH10 (8.42 Å) [89]. The interlayer distance of the second 

AFm (C4FsH10) phase is between the value for C4FsH12 and the value of C4FcH12 (7.98 

Å). 

Fig. 27 Details of the Rietveld plot from the sample Fe-Ms-50 °C. 

The structure of Fe-monosulfate is isotypic to Al-monosulfate. It has the same symmetry 

and same atomic positions with two kinds of disorder in the interlayer region: statistical 

distribution between one sulfate group with 2.3 water molecules and orientation disorder 

of sulfate groups (up or down). The refined composition is 

73


[Ca2Fe(OH)6]·[(SO4)0.5·3.2(1)H2O]; i.e. C4AsH12.4(2). The crystallographic structure 

representations are identical to those previously determined for Al-monosulfate [87]. The 

interatomic distances determined for the refined Fe-monosulfate structure are listed in 

Table 16. 

Table 16 Refined interatomic distances in Fe-monosulfate (standard deviation is given in 

parentheses). 

Bond type Atoms Distances (Å) 

Ca-OH Ca-O1 3 x 2.25 (2) 

3 x 2.40 (2) 

Ca-H2O Ca-O2 2.49 (3) 

Fe-OH Fe-O1 6 x 2.01 (2) 

S-O S-O3 3 x 1.44 (4) 

S-O4 1.48 (3) 

SO4-OH O3-O1 2.95 (5) 

O4-O1 2.80 (3) 

SO4-H2O O2-O3 3 x 2.87 (5) 

3.2.5. Comparison of C4AsH12 with C4FsH12 

Both Al-monosulfate and Fe-monosulfate crystallize in the rhombohedral R3 c 

symmetry. This structural similarity could enable a substitution of Al by Fe in the main 

layer structure. Fe 3+ (0.64 Å) has a slightly larger ion radius than Al 3+ (0.54 Å) which 

resulted in a larger unit cell parameter (a = 5.88 Å) for Fe-monosulfate than for Al- 

monosulfate (a = 5.76 Å). 

Raman spectroscopy was applied to study the vibrational frequency of the molecules in 

the structure of Fe-monosulfate. The results were compared with Al-monosulfate [79]. 

Raman spectroscopy measurements were carried out for the solids synthesized at 20 °C 

and 50 °C (Fig. 28 and Fig. 29). 

74


Fig. 28 Spectral range 100 cm -1 – 1500 cm -1 of Raman spectra from sample Fe-Ms-20 °C 

(comparison with Al-monosulfate spectra [79]). 

The small peak at 1085 cm -1 was assigned to carbonate which was present due to carbon 

dioxide contamination. Sulfate vibrations are less intense in the Fe-monosulfate 

compound compared to Al-monosulfate compound. The vibration mode of sulfate was 

shifted from the more intense 1 mode: 992 cm -1 for Fe-Ms compared with 982 cm -1 for 

Al-monosulfate (or 981 cm -1 on Kuzel’s salt) [79, 90]. The 1 position of Fe-monosulfate 

synthesized at 20 °C corresponds to that of Al-monosulfate treated at 117 °C (i.e. 993 cm - 

1 ) [79]. The band position of Fe(OH)6 appeared at 508 cm -1 , which is significantly lower 

than for Al(OH)6 (532 cm -1 ). The shifts of the bands could indicate a possible substitution 

of Al by Fe as was previously observed between Al- and Fe- ettringite [79]. 

The hydrogen bond network (Fig. 29) is highly modified in Fe-monosulfate compared to 

that in Al-monosulfate. 

75

Intensity 

4500 

4000 

3500 

3000 

2500 

2000 

1500 

1000 

500 

0 


2800 3000 3200 3400 3600 3800 4000 

Raman shift (cm -1 ) 

Fe-Ms 50°C 

Al-Ms 

Fig. 29 Spectral range 2800 cm -1 – 4000 cm -1 of Raman spectra from sample Fe-Ms-50 °C 

(comparison with Al-monosulfate spectra [79]). 

The weight loss of monosulfate observed by TGA/DTG analysis (Fig. 30) was assigned 

as follows: 

Ca4[(Al, Fe)2(OH)12]SO4 6H2O → Ca4[(Al, Fe)2(OH)12]SO4 + 6H2O: interlayer water 

removal 

Ca4[(Al, Fe)(OH)12]SO4 → Ca4[(Al, Fe )2O6]SO4 + 6H2O : dehydroxylation 

The loss of water is controlled by the interlayer structure and occurs in several steps for 

both Al- and Fe-monosulfate. The 6 molecules of the interlayer water of Al-monosulfate 

escaped in the temperature range up to 260 °C comparable to other observations [61, 73]. 

The remaining water molecule bond in main layer escaped above 260 °C. The water loss 

observed for Fe-monosulfate had less well defined steps due to the water loss of Fe(OH)3 

in this temperature range [16]. Water loss up to 260 °C was 4.5 in the case of Fe- 

monosulfate taking account the presence of other solids. 

76


Fig. 30 Thermal analysis (TGA and DTG) of Al and Fe-monosulfate after 680 days. 

3.2.6. Solid solution between Al and Fe-monosulfate (C4(A,F)sH12) 

The similarity in the ionic radii of Al 3+ and Fe 3+ and the structure of Al- and Fe- 

monosulfate could allow Al-Fe substitutions in the structure and the formation of a solid 

solution. Solid solution formation can stabilize the solids by decreasing their solubility. 

Möschner et al. [16] observed the formation of a solid solution between Al- and Fe- 

ettringite with a miscibility gap. In contrast, no solid solution was found between Al- and 

Fe-monocarbonate due to the different symmetries of Al-monocarbonate (triclinic) and 

Fe-monocarbonate (rhombohedral) [19]. 

In this study, mixed 3CaO(AlxFe1-x)2O3(CaSO4)12H2O were synthesized at room 

temperature by varying the Al/(Al + Fe) ratio from 0 to 1. Fig. 31 shows the XRD pattern 

of the 3CaO(AlxFe1-x)2O3(CaSO4)12H2O series. 

77


A peak shift from 0.0 to 0.45 Al/(Al + Fe) ratio was clearly observed caused by the 

substitution of Al 3+ by Fe 3+ in the main layer structure of monosulfate for the 104 peak at 

2θ = 22°. A peak splitting was observed between 0.45 and 0.95 Al/(Al + Fe) ratio. This 

indicated a mixture of two monosulfate instead of one solid solution containing Al and 

Fe. This suggested a miscibility gap in the solid solution series exists in the range 0.45 


found to vary at 0.0 < Al/(Al+Fe) < 0.45. Two distinct unit cell parameters in the samples 

at 0.45


3.2.7. Solubility of Al/Fe-monosulfate 

The composition of the solutions in equilibrium with Fe-monosulfate was determined 

after different equilibration times and at different temperature (Table 17). The very low 

Fe-concentration indicated the presence of Fe-hydroxide as also visible from the red color 

of the sample. 

3.2.7.1. Determination of solubility products at 20, 50 and 80 °C 

The measured concentrations of calcium, iron, sulfur and hydroxides were used to 

calculate the solubility products using GEMS. The solubility products are given by: 

Ks0,Fe-monosulfate = {Ca 2+ } 4 . {FeO2 - } 2 {SO4 2- } . {OH - } 4. {H2O} 10 

Ks0,Fe-hydroxide = {FeO2 - } 2 {OH - } -1. {H2O} 1 

Ks0,portlandite = {Ca 2+ } {OH - } 2 

where {} denotes the activity. 

The calculated solubility products at different equilibration times and temperature are 

listed in Table 17. The ion activity product of portlandite calculated at 20 °C is 

comparable with the theoretical solubility product of -5.2 at 25 °C [45] indicating the 

presence of portlandite in agreement with the XRD and TGA data and the validity of the 

measured calcium concentrations and pH values. The calculated ion activity product of 

Fe-hydroxide in the kinetic experiments suggested the presence of Fe-hydroxide. This is 

consistent with the red color observed in all Fe-containing samples. 

80


Table 17 Measured ion concentrations and calculated solubility products at different equilibration times. 

Equlibration 

time 

Temperature Ca Fe K S +pH log Ks0 log Ks0 log Ks0 Solid phases 

[days] [°C] [mmol/l] [mmol/l] [mmol/l] [mmol/l] C4FsH12 Fe(OH)3 CH present 

7 20 2.94 0.00057 357 64.0 13.05 -31.63 -5.67 -4.93 CH, C2F, Fe(OH)3 

28 20 2.84 0.00047 373 64.2 13.18 -31.80 -5.78 -4.90 CH, C2F, Fe(OH)3 

120 20 1.97 0.00195 346 50.8 13.39 -31.21 -5.16 -5.05 C4FsH12, CH, C2F, Fe(OH)3 

360 20 3.22 0.00107 349 38.6 13.40 -30.82 -5.47 -4.74 C4FsH12, CH, C2F, Fe(OH)3 

680 20 2.04 0.00079 366 47.0 13.44 -31.86 -5.60 -4.96 C4FsH12, CH, C2F, Fe(OH)3 

Average -31.30±0.30 -5.53±0.01 -4.92±0.10 

7 50 1.57 0.00122 356 57.4 13.08 -33.04 -5.39 -5.40 C4FsH12, CH, C2F, Fe(OH)3 

28 50 1.35 0.00097 372 50.5 13.21 -33.40 -5.50 -5.39 C4FsH12, CH, C2F, Fe(OH)3 

120 50 1.12 0.00281 341 38.6 13.40 -33.02 -5.11 -5.48 C4FsH12, CH, C2F, Fe(OH)3 

360 50 2.87 0.00143 361 49.5 13.37 -33.02 -5.95 -5.08 C4FsH12, CH, Fe(OH)3 

Average -33.12±0.30 -5.49±0.01 -5.34±0.10 

7 80 0.8 0.00351 354 60.6 13.06 -34.92 -5.10 -6.02 C4FsH12, CH, Fe(OH)3 

28 80 1.07 0.00278 368 66.4 13.13 -5.36 -5.89 CH, Fe2O3 

120 80 1.17 0.00283 358 67.2 13.30 -5.41 -5.88 CH, Fe2O3 

360 80 3.44 0.00161 326 69.3 13.20 -6.26 -5.50 CH, Fe2O3 

Average -34.92±0.30 -5.53±0.01 -5.82±0.10 CH, Fe2O3 

Detection limits [mmol/l]: Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0,0002 measurement uncertainty ±10% , + pH measured at 20 °C, Due to the strong dependence of the H + 

activity on temperature, a pH value of 13.2 at 20 °C corresponds to12.3 at 50 and to11.6 at 80 °C. The solubility product for portlandite are -5.15(20 °C), -5.51(50 °C) and -5.95(80 °C) and for Fehydroxide 

-4.77 (20 °C), -3.81(50 °C) and -3.00(80 °C) [44, 45] 

81


The solutions equilibrated up to 680 days were saturated with respect to Fe-hydroxide. 

The mean solubility product of Fe-monosulfate between 120 and 680 days was used at 20 

°C. The value of -31.30 is approximately 2 log units higher than the value of -33.06 

recalculated from the concentrations measured between pH 13.2 and 13.4 in the Fe- 

ettringite experiments by Möschner et al. [15] from under- and oversaturation. Note that 

Möschner et al. [15] had calculated somewhat lower solubility products (-32.8 to -34.2) 

using a different by of 0.064 (for NaCl electrolyte) instead of the 0.123 (for KOH 

electrolyte) used here for the activity corrections in the Truesdell-Jones form as given in 

section 2.3.3. The solubility product at 50 °C and 80 °C were determined to be -33.12 and 

-34.92, respectively. At 80 °C Fe-monosulfate was unstable with respect to portlandite 

and hematite within 28 days. The solubility of Fe-monosulfate decreases at higher 

temperature (Fig. 34), indicating that Fe-monosulfate is more stable at higher 

temperature. Nevertheless, also the stability of Fe-hydroxide, portlandite and hematite 

and also their kinetic of formation is changing. 

3.2.7.2. Determination of solubility products under standard condition 

The solubility product at standard conditions was calculated with the help of GEMS-PSI 

using the three term temperature extrapolation from the solubility products obtained at 

20, 50 and 80 °C. The procedures to calculate temperature-dependent log K value was 

described in chapter 2.3. The heat capacity of Fe-monosulfate was estimated from the 

heat capacity of Al-monosulfate (Cp =942 J/(mol K)) measured Ederova and Satava [46] 

using the reference reaction given in Table 3. The measured solubility products at 20, 50 

and 80 °C were used to fit the entropy of reaction, as previously described elsewhere [7, 

51, 62]. The heat capacity, enthalpy and entropy obtained are compiled in Table 18. The 

82


calculated values as a function of temperature as shown in Fig. 34 are also used to derive 

the solubility product of Fe-monosulfate at 25 °C. The value is -31.57, roughly 2 log unit 

lower than the solubility product of Al-monosulfate [52]. This shows that Fe-monosulfate 

is more stable than Al-monosulfate, similar to the observations made for Fe- and Al- 

monocarbonate [19]. While the heat capacity, Cp of Fe-monosulfate (968 J/K.mol) is 

identical to the heat capacity reported in Möschner et al. [15], the entropy fitted to the 

measured data (1430 J/(K . mol), (Table 18) is considerably higher than the entropy of 858 

J/(K . mol) estimated by Möschner et al. [15] based on the reference reaction given in 

Table 3 and the value of 833.3 J/(K . mol) estimated by Blanc et al. [91] using a similar 

reference reaction. The calculated thermodynamic parameters are compiled in Table 18. 

Table 18 Thermodynamic parameters of Fe-monosulfate at standard conditions (25°C, 1 atm). 

Phases log Ks0 

ΔfG° 

[kJ/mol] 

Δf H° 

[kJ/mol] 

S 0 

[J/mol/K] 

C 0 p 

[J/mol/K] 

a0 

[J/(mol.K)] 

a1 

[J/(mol.K 2 )] 

a2 

[J K/mol] 

a3 

[J/(mol.K 0.5 )] 

V 0 


C4AsH12 -29.26 -7778.5 -8750.0 821 942 594 1.168 0 0 309 [7] 

C4FsH12 -31.57 -6873.2 -7662.6 1430 968 577 1.234 2.E+06 0 321 t.s 

t.s: this study 

83

log Ks0 

-30 

-31 

-32 

-33 

-34 

-35 

-36 

-37 


this study 

Möschner et al 2008 

0 20 40 60 80 100 


Fig. 34 Calculated solubility products of Fe-monosulfate from the solubility experiments. 

symbols: experimental data. 

3.2.7.3. Determination of solubility product of the solid solution and 

modeling of the liquid phase 

The compositions of the liquid phases in equilibrium with monosulfate at varying Al/(Al 

+Fe) ratios were determined (Table 19). In the presence of pure Fe-monosulfate, at 

relatively high sulfate concentrations, intermediate calcium concentrations and very low 

iron concentrations were measured. Iron concentrations were limited by the presence of 

iron hydroxide, calcium by the solubility of portlandite. In the presence of aluminum, the 

sulfate concentrations decreased gradually from 47 mmol/l to 1.8 mmol/l, resulting in a 

corresponding increase in hydroxide concentrations and a decrease in the calcium 

84


concentrations from 2 to 1 mmol/l. For the Fe-free Al-monosulfate, a lower calcium 

concentration, sulfate and aluminum concentrations were measured. Generally the 

aluminum concentrations were 100 to 1’000 times larger than the iron concentrations 

which were limited by the precipitation of Fe(OH)3. In the aluminum free sample 

(XAl/(Al+Fe) = 0.0), 0.63mmol/l of Al was detected. This indicates that the samples were 

contaminated by Al. Additional investigations indicated that aluminum contamination 

had occurred during dilution of the samples with HNO3; thus all measurements were 

corrected by 0.63 mmol/l Al “blank” value (see Table 19). However, even after this 

correction the calculated solubility products of Al-monosulfate and Al-ettringite were 

approximately 1 log unit higher than expected (see Table 20). 

Table 19 Compositions of Al/Fe-monosulfate after synthesis at 20°C equilibrated for 680 days in 

supersaturated condition. 

Al/Fe+Al Al Al* Ca Fe K S OH +pH Solid phases 

Ratio [mmol/l] [mmol/l] [mmol/l] [mmol/l] [mmol/l] [mmol/l] [mmol/l] present 

0 0.63


GEMS is set up for one ion substituting another ion in one crystallographic site per end 

member. To fulfill this condition the stoichiometry of both end members, the logKs0 

values and the G° as given in Table 18 were downscaled by a factor 2: 

KAl-Ms . XAl-Ms . γAl-Ms = 2×({Ca 2+ } 2 {AlO2 - } 1 {SO4 2- } 0.5 {OH - } 2. {H2O} 5 ) 

KFe-Ms . XFe-Ms . γFe-Ms = 2×({Ca 2+ } 2 {FeO2 - } 1 {SO4 2- } 0.5 {OH - } 2 {H2O} 5 ) 

ΣΠ= 2×(0.5KAl-Ms.XAl-Ms.γAl-Ms + 0.5KFe-Ms.XFe-Ms.γFe-Ms) = 2×({Ca 2+ } 2 [{AlO2 - }+{FeO2 - }] 

{SO4 2- } 0.5 {OH} 2 {H2O} 5 ) 

KAl-Ms and KFe-Ms are the solubility products of the end members of Al-monosulfate and 

Fe-monosulfate; XAl-Ms and XFe-Ms mole fractions of Al-Ms and Fe-Ms in the solid; γAl-Ms 

and γAl-Ms the activity coefficients. The activity coefficients are calculated according to: 

ln.γAl-Ms= X 2 Fe-Ms[a0 –a1(3 XAl-Ms – XFe-Ms)] 

ln.γFe-Ms= X 2 Al-Ms[a0 –a1(3 XFe-Ms – XAl-Ms)] 

The software MBSSAS [59] was used to derive the Guggenheim parameters a0 = 1.26 

and a1 = 1.57 based on the experimentally-observed miscibility gap between 0.45 < 

Al/(Al + Fe) < 0.95 in the monosulfate binary solid solution series. A detailed 

description of MBSSAS is given elsewhere [59]. 

86


Table 20 Solubility products of all the solids formed during the synthesis of Al/Fe-monosulfate 

solid solution series at 20 °C equilibrated for 680 days under supersaturated condition. 

Al/(Al+Fe) logKs0 logKs0 logΣΠ logKs0 logKs0 logKs0 

Ratio C4AsH12 C4FsH12 C4(A,F)sH12 Fe(OH)3 CH C6As3H32 

0.0 n.d -31.86 -31.86 -5.60 -4.96 

0.1 n.d -33.95 n.d -6.34 -4.98 

0.2 n.d -34.72 n.d -6.27 -4.99 

0.3 -28.83 -34.91 -28.83 -6.28 -4.97 -43.74 

0.4 -28.69 -33.92 -28.69 -5.82 -4.95 -43.62 

0.5 -28.41 -33.79 -28.41 -5.80 -4.92 -43.35 

0.6 -28.84 -33.10 -28.83 -5.32 -4.97 -43.95 

0.7 -28.46 -32.59 -28.45 -5.14 -4.94 -43.47 

0.8 -28.38 -34.75 -28.38 -6.27 -4.92 -43.32 

0.9 -28.59 -35.08 -28.59 -6.31 -4.99 -43.61 

1.0 -28.43 -28.43 -43.46 

Average -5.91 -4.96 

Theoretical Values* -29.40 -31.3 -4.77 -5.15 -44.7 

* The theoretical solubility products at 20 °C were calculated using GEMS and the implemented standard database [3, 7, 44, 45]. 

For Fe-monosulfate the values derived in this study from Table 17 were given. n.d: not determined 

A Lippmann diagram (for details see section 2.3.3) was drawn to describe the solid 

solution formation between Al and Fe-monosulfate. In the Lippmann diagram two 

different superimposed x-axis are used. The total solubility products are plotted as a 

function of Xsolid, the solid Al/(Al + Fe) ratio (solidus) and as a function of Xliquid, the 

Al/(Al + Fe) ratios in the solution (solutus). The solidus-solutus phase diagram helps to 

display all possible equilibrium states of the solid solution series during Al-Fe 

substitution in the main layer structure of monosulfate. The measured solubility products 

of Fe-monosulfate logKs0 = -31.86 and Al-monosulfate logKs0 = -28.43 as given in 

(Table 20) were used to model the solidus and solutus solubility curve. The solubility 

product of Al-monosulfate obtained in this study at 20 °C was -28.43, which is within the 

range of reported literature values -27.7 [92] to -29.8 [93]. This value is approximately 

one log unit higher than the value (-29.26) obtained by Matschei et al. [7] for 25 °C. 

87


The presence of a continuous solid solution with a miscibility gap between Al- and Fe- 

monosulfate has been observed by XRD (Fig. 32 and Fig. 33). In the Lippmann diagram 

both an ideal (dashed lines) and a solid solution with miscibility gap (solid line) were 

plotted (Fig. 35). In the case of ideality the parameters a0 and a1 are equal to zero and the 

activity coefficients of the end members equal to one (γAl-Ms = γFe-Ms=1). For the non-ideal 

solid solution the dimensionless parameters a0 = 1.26 and a1 = 1.57 were used as 

described above. 

If an ideal solid solution is assumed, the modeled total solubility products of the solidus 

underestimate the total experimentally determined solubility products as shown in Fig. 

35. The non-ideal solid solution with a miscibility gap at 0.45 < Al/(Al + Fe) ratio < 0.95 

(solid lines) shows somewhat better agreement with the experimental values. 

88


Fig. 35 Lippmann diagram illustrating the total solubility products of Al/Fe-monosulfate solid 

solution series: total experimentally determined solubility product (symbols), modeled 

total solubility products assuming ideal solid solution (dashed lines), modeled total 

solubility products assuming a non-ideal solid solution with a miscibility gap (a0= 1.26 

and a1= 1.57) (solid lines) and solubility products assuming no solid solution (dotted 

lines). X-axis: Al/(Al + Fe) ratios in the solid and Al/(Al + Fe) ratios in the liquid. 

The liquid phase composition in the presence of (CaO)3(AlxFe1-x)2O6CaSO412H2O 

(x=0.0, 0.1,….,0.9, 1.0) was calculated with the derived thermodynamic parameters as 

compiled in Table 20 using GEMS. As discussed above, the solubility products were 

89


normalized so that the number of exchangeable sites equaled to 1: CaO1.5(AlxFe1- 

x)O3 1 /2CaSO46H2O. 

The modeling was done assuming the non-ideal solid solution model with the miscibility 

gap using a0 = 1.26 and a1 = 1.57 for Al and Fe-monosulfate as shown in Fig. 35. The co- 

precipitation of portlandite, Fe-hydroxide and Al-ettringite was calculated in agreement 

with the experimental observations in the solid phase (see Table 19). Fig. 36 shows the 

calculated and measured compositions of the liquid phase. Both the calculated and the 

measured data show a reasonable agreement. 

Fig. 36 Measured (points) and calculated (lines) concentrations in the liquid phases of the 

synthesized monosulfate with different Al/(Al+Fe) mole ratio, assuming a continuous 

solid solution with a miscibility gap. 

90



The formation of Fe-monosulfate was slow as also noted for the other Fe-containing 

phases. Portlandite and Fe-oxide/hydroxide co-precipitated with Fe-monosulfate at all 

equilibration times. At 80 °C Fe-monosulfate was unstable with respect to portlandite and 

Fe-oxide/hydroxide after a few days. 

Fe-monosulfate crystalizes in rhombohedral R3 c symmetry. The structure of Fe- 

monosulfate is similar to that of Al-monosulfate so that substitution of Al by Fe in the 

main layer structures seems possible. In the dried samples generally two Fe-monosulfates 

with different water content were observed. The main phase found was C4FsH12 with an 

interlayer distance of 8.87 Å and in addition the interlayer distance at 8.36 Å tentatively 

assigned to C4FsH10. Raman data showed shifting of the band positions that could 

indicate a possible solid solution formation. Similarly, also the XRD data showed a 

systematic shift of XRD peaks with different Al/(Al +Fe) ratios that confirmed the 

formation of a solid solution between the Al- and Fe-monosulfate with miscibility gap 

between 0.45 < Al/(Al + Fe) < 0.95. 

The solubility products of Fe-monosulfate were determined to be -31.30, -33.12 and - 

34.92 at 20, 50 and 80 °C respectively. The solubility product at standard condition was 

calculated to be -31.57 which is 2 log units lower than the one of Al-monosulfate (- 

29.26). 

In conclusion, the solubility products as well as the solution compositions changed as a 

function of the Al/(Al+Fe) ratios which indicated a solid solution formation between Al 

and Fe-monosulfate. Stability of the solid increased as Fe substituted Al in monosulfate. 

In the presence of iron, portlandite and iron hydroxide were always present in addition to 

monosulfate. 

91


3.3. Fe-Friedel’s salt (3CaO . Fe2O3 . CaCl2 . 10H2O) 

3.3.1. Introductions 

A number of investigations have been carried out on the chloride binding capacity of 

cement forming Al-bearing Friedel’s salt [54, 68, 74, 75, 78, 94-96]. Moreover, chlorides 

can react with ferrite and form Fe-bearing Friedel’s salts. The formation of Fe-Friedel’s 

salt was observed in cement free systems [88, 95-97]. 

The aim of this study was to determine the thermodynamic properties Fe-Friedel’s salt. 

Moreover, the formation of this phase was investigated and characterized under different 

condition. 

3.3.2. Kinetics of formation 

In this study Fe-Friedel’s salt (3CaO.Fe2O3.CaCl2.10H2O) was synthesized using 3 

different methods (see section 2.1.3). Fig. 37 shows the XRD patterns of the solid 

synthesized from FeCl3.6H2O and CaO in 0. 1M KOH. The sharp peaks of the Fe- 

Friedel’s salt indicate the presence of a well crystalline solid. Fe-Friedel’s salt started to 

form after 7 days of equilibration and the peak position is at 2θ = 11.37° (basal spacing d 

= 7.76 Å) (see Fig. 37). 

92


Fe-Fr 


Fe-Fr 

Fe-Fr Fe-Fr 

10 15 20 

2CuK 

25 30 

Cc 

3 years 

270 days 

180 days 

28 days 

7 days 

Fig. 37 XRD pattern of 3CaO . Fe2O3 . CaCl2 . 10H2O (Fe-Fr) synthesized at 20 °C and sampled after 

different equilibration times from FeCl3.6H2O and CaO in 0.1 M KOH. 

Another synthesis of Fe-Friedel’s salt was carried out using C2F, CaO and CaCl2.2H2O to 

obtain a different hydroxide concentration and understand the possible uptake of 

additional hydroxide in Fe-Friedel’s salt. Fe-Friedel’s salt started to form from C2F after 

7 days equilibration time. Portlandite and Fe-hydroxide co-precipitated with Fe-Friedel’s 

as observed previously for other Fe-AFm phases [19]. Traces of portlandite, Fe- 

hydroxide and possibly carbonate containing AFm phases were detected after three years 

of equilibration times. Fig. 38 showed that the XRD patterns of Fe-Friedel’s salt 

synthesized at different pH values (pH=11.94, 12.39 and 12.84) were very similar. At 

higher pH values, however, a slight peak shift of the 003 peaks towards higher 2θ values 

was observed, which could indicate the uptake of additional hydroxide in the interlayer 

of Fe-Friedel’s salt and thus a shift towards the position of C4FHx (2θ = 11.65°). Note 

that extensive solid solution formation has been observed between the Al-Friedel’s salt 

93


and C4AH13 [54]. The XRD data indicate that Fe-Friedel’s salt is thermodynamically 

stable within the pH range of 11.9 to 12.9. Possibly, also carbonate could replace chloride 

in the interlayer as has been observed between Al-monocarbonate and Al-Friedel’s salt 

[54, 78]. 

Fig. 38 Comparison of the XRD patterns of Fe-Friedel’s salts equilibrated for three years at 

different pH values: synthesized a). FeCl3.6H2O and CaO in 0.1M KOH (pH = 11.94), b). 

C2F, CaCl2.2H2O and CaO in distilled water (pH = 12.39) and c). C2F, CaCl2.2H2O, and 

CaO in 0.1 M KOH (pH = 12.84), CH-portlandite. 

In addition, the TGA-DTG analysis indicated fast formation of Fe-Friedel’s salt. The 

weight loss from the interlayer structure occurred in two steps as observed for other AFm 

phases [19]. The 4 water molecules from the interlayer are lost at around 140 °C and the 

water of the main layer above 200 °C. The peak at 575 °C corresponds to the loss of CO2 

94


present due to CO2 contamination. The peak of carbonate decreased over time. The 

varying carbonate content indicates that carbonation occurred mainly during sample 

drying of the solids (Fig. 39). A substitution of Cl - by CO3 2- in C3FCaCl2H10 - C4FcH12 

system could be possible as reported for the Al containing Friedel’s salt [54]. 



100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

Fe-Fr 

Fe-Fr 

carbonate 

200 400 600 800 


Fe-Fr-7days 

Fe-Fr-28days 

Fe-Fr-180days 

Fe-Fr-270days 

Fe-Fr-3years 

Fig. 39 TGA-DTG curves of Fe-Friedel’s salt synthesized at 20 °C and sampled after different 

equilibration times from FeCl3.6H2O and CaO in 0.1M KOH. 

3.3.3. Structure of Fe-Friedel’s salt 

The solid synthesized by mixing appropriate amounts of C2F, CaCl2.2H2O and CaO in 

0.1 M KOH equilibrated for 500 days (pH = 12.39) was used for the crystallographic 

investigation of Fe-Friedel’s salt. Laboratory XRD and TGA results of this solid are 

shown in the appendix. 

Multipattern Rietveld refinement was performed as shown in Fig. 40. The sample 

consisted mainly of 3CaO.Fe2O3.CaCl2.10H2O (90 wt.%), some portlandite (3 wt.%) and 

calcite (7 wt.%). 

95


Fig. 40 Rietveld plot for Fe-Friedel’s salt recorded at = 0.697751 Å and at a sample-to-detector 

distance of 150 mm. 

The refined structural model was found to be identical to that of the data published by 

Rousselot et al. [96]. Fe-Friedel’s salt crystallized in rhombohedral structure with R3 c 

symmetry. The unit cell parameters were: a = 5.8567 (2) Å and c = 23.314 (1) Å (V = 

692.57 (5) Å 3 ). The Fe-Friedel’s salt is composed of a positively charged main layer 

[Ca2Fe(OH)6] + with a possible substitution of Fe 3+ by Al 3+ and negatively charged 

interlayer [Cl 2H2O] - with a possible substitution of Cl - by 1/2CO3 2- or OH - . 

Table 21 showed the refined structural parameters of Fe-Friedel’s salt. The structure 

corresponds to the high temperature phase of Al-Friedel’s salt (the rhombohedral HT- 

structure is observed above 35 °C for pure Friedel’s salt; i.e. calcium aluminate without 

carbonate contamination) [74, 75]. The large Biso values of the interlayer species 

(chloride and water molecules) can be explain either by dynamical disorder within the 

96


interlayer region or by carbonate contamination with CO3 2- , which substituted chloride in 

the 3b site. 

Table 21 Refined structural parameters of 3CaO.Fe2O3.CaCl2.10H2O (standard deviation is given 

in parentheses). 

Atom Wyckoff x y z Biso (Å 3 ) Occupancy 

Ca 6c 2/3 1/3 0.0255 (2) 1.2 (1) 1 

Fe 3a 0 0 0 1.2 (1) 1 

O (OH) 18f 0.270 (1) -0.053 (1) 0.0445 (3) 0.9 (2) 1 

Cl 3b 0 0 1/2 7.0 (4) 1 

O (H2O) 6c 2/3 1/3 0.1355 (5) 4.0 (4) 1 

3.3.4. Comparison of Al-Friedel’s salt and Fe-Friedel’s 

The structure of 3CaO.Al2O3.CaCl2.10H2O has been investigated by different researches 

[54, 75, 78, 94, 96]. The layered structure consists of a `brucite-like` main layer 

[Ca2(Al(OH)6] + separated by an interlayer [Cl H2O] - . The low temperature polymorph is 

monoclinic α-Friedel’s salt and at higher temperature (>35 °C) as rhombohedral β- 

Friedel’s salt [75, 98]. In this study, the structure of Fe-Friedel’s salt was found 

crystallize in rhombohedral symmetry. Structural transition could be possible in Fe- 

Friedel’s salt at lower temperatures. The rhombohedral structure of Al and Fe-Friedel’s 

salt indicates a possible substitution of aluminum by iron in the main layer and chloride 

by carbonate or hydroxide in the interlayer. 

Fig. 41 compares the thermal analysis of Al-Friedel’s salt and Fe-Friedel’s salt. Al- 

Friedel’s salt lost water from its structure in three steps. At a temperature up to 180 °C, 

the 4 interlayer water molecules were removed. The main layer water was removed in 

two steps at higher temperatures where each time 3 water molecules were lost around the 

two peaks at 265 and 380 °C. In the case of Fe-Friedel’s salt the water loss occurred in 

two distinct steps only. The 4 interlayer water molecules are at around 140 °C. For the 

main layer water, only one peak at around 260 °C was observed. However, water loss 

97


continued up to 400 °C. The theoretical total water loss of 3CaO . Al2O3 . CaCl2 . 10H2O 

corresponds with 32 wt.% well with the water loss observed in Fig. 41, while the water 

loss of 3CaO.Fe2O3.CaCl2.10H2O is due to the dilution by the presence of carbonate 

somewhat lower than the theoretical 29 wt.%. The decarbonation observed in the Fe- 

Friedels salt sample could be due to the presence of calcite or carbonate containing AFm 

phases. 

Fig. 41 Thermal analysis (TGA and DTG) of Al and Fe-Friedel’s salt synthesized from 

FeCl3.6H2O and CaO in 0.1 M KOH and equilibrated for 270 days. 

Raman spectrum was collected for Fe-Friedel’s salt as shown in Fig. 42. The carbonate 

contamination of the Fe-Friedel’s salt is clearly visible by the CO3 2- vibration at 1088 cm - 

1 

. Spectrum relative to hydrogen bonds network is observed in the spectral range from 

98


3000 cm -1 to 4000 cm -1 with sharp signals. In contrast, the O–H bands of vibrations in the 

structure of Al-Friedel’s salt are characterized by a broad signal [78]. 

Fig. 42 Raman spectra recorded for Fe-Friedel’s salt crystal. 

3.3.5. Solid solution between Al and Fe-Friedel’s salt (3CaO(AlxFe1- 

x)2CaCl2.10H2O 

Kuzel et al. [88] observed only limited solid solution formation between Al- and Fe- 

Friedel’s salt at a temperature below 100 °C. In contrast, Goetz-Neunhoeffer et al. [97] 

and Rapin et al. [99] observed the existence of a continuous solid solution in Al/Fe- 

Friedel’s salt at room temperature (see Fig. 43). In this study, only a few samples with 

different Al/(Al + Fe) ratios were investigated. The samples were prepared from CaO and 

AlCl3.6H2O or FeCl3.6H2O in 0.1 M KOH. The results of this mixed Al/Fe-Friedel’s salt 

did not agree well with previously reported findings as shown in Fig. 43. Carbonation and 

the small number of samples made the results inconclusive. However, based on the study 

of Goetz-Neunhoeffer et al. [97] and Rapin et al. [99] a solid solution formation seems 

probable. 

99

unit cell parameter a (Å) 

6.0 

5.9 

5.8 

5.7 


0.0 0.2 0.4 0.6 0.8 1.0 


Goetz-Neunhoeffer 1996 

Rapin 2001 

Kuzel 1968 

Rousselot 2003 

this study 

Fig. 43 Values of a-parameters for the Al/Fe-Friedel’ salt solid solution determined in this study 

compared to the findings by Kuzel et al. [88], Goetz Neunhoeffer et al. [100], Rapin et al. 

[99] and Rousselot et al. [96]. 

3.3.6. Solubility 

The concentrations of the solutions were measured after different equilibration times for 

pure Fe-Friedel’s salt (Table 22 and Table 23) and for mixed Al/Fe-Friedel’s salts (Table 

25). The measured ion concentrations were used for the calculation of the solubility 

products of Friedel’s salt. 

3.3.6.1. Solubility of Fe-Friedel’s salt 

The Fe-Friedel’s salt prepared from FeCl3.6H2O and CaO in 0.1 M KOH did not contain 

portlandite as also indicated by the low portlandite ion activity product (-6), thus 

indicating effectively undersaturation. The concentrations of calcium, chloride and iron in 

the aqueous phase were relatively high compared to the other samples; while the 

100


hydroxide concentrations were lower (pH 11.6 to 11.9). In contrast, Fe-Friedel’s salt 

prepared from C2F, CaCl2.2H2O and CaO co-precipitated with portlandite and Fe- 

hydroxide at pH values from 12.4 to 12.8. The aqueous phase had lower concentrations 

of calcium and chloride. The calculated ion activity products for Fe(OH)3 were (10 -4 to 

10 -3 ) unexpectedly high, which indicates that either the presence of chloride ions could 

result in a kinetic hindrance for Fe(OH)3 formation or alternatively mixed complexes 

between Fe 3+ , OH - and Cl - occur. Note that the latter complexes are not accounted for in 

the thermodynamic database (the known iron chloride complexes are rather weak and 

give no significant contribution at high pH values: FeCl 2+ , log K = 1.48, FeCl2 + , log K = 

2.13, and FeCl3 0 , log K = 1.13 in GEMS-PSI TDB. A literature search gave no evidence 

for the existence of FeCl4 - or mixed hydroxide-chloride-iron(III) complexes). 

The solubility product of the solid was calculated according to: 

Ks0, Fe-Friedel’s salt = {Ca 2+ } 4 . {FeO2 - } 2 {Cl - } 2 {OH - } 4. {H2O} 8 

The Ks0 values were calculated based on the measured concentrations of Ca, Fe, Cl and K 

and the activity coefficients obtained with GEMS. The pH values were adapted by 

varying the concentrations of KOH to the measured pH values. The solubility products of 

Fe-Friedel’s salt prepared at different pH values are presented in Table 22 and Table 23. 

101


Table 22 Measured ion concentrations and calculated solubility products at 20 °C and sampled 

after different equilibration times synthesized from FeCl3.6H2O and CaO in 0.1M K 

OH. 

Age Ca K Fe Cl #pH +pH log Ks0 log Ks0 log Ks0 

(days) [mmol/l] [mmol/l]l [mmol/l] [mmol/l] Fe-Fr Fe(OH)3 CH 

7 162 93 0.0027 422 11.62 11.61 -28.40 -3.14 -6.55 

28 161 98 0.0020 419 11.40 11.42 -29.45 -3.08 -6.94 

180 146 87 0.0023 427 11.61 11.53 -29.04 -3.13 -6.76 

270 144 88 0.0028 399 11.69 11.70 -28.23 -3.22 -6.41 

1100 149 88 0.0059 410 11.94 11.88 -26.83 -3.07 -6.06 

Average -28.39±0.50 -3.13±0.20 -6.55±0.30 

Detection limits [mmol/l]: Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0.0002, Cl=0.003 , measurement 

uncertainty ±10%, #pH measured at 20°C, + calculated pH by GEMS. 

Table 23 Measured ion concentrations and calculated solubility products at 20 and 50°C and 

sampled after different equilibration times synthesized from C2F, CaCl2.2H2O and CaO 

in distilled water and in 0.1 M KOH. 

Age Temp. Ca K Fe Cl #pH +pH log Ks0 log Ks0 log Ks0 

(days) [°C] [mmol/l] [mmol/l] [mmol/l] [mmol/l] Fe-Fr Fe(OH)3 CH 

180 20 35.0


available to describe this possible solid solution series and thus for further calculations 

only the solubility product of -28.39 determined in the pH range from 11.4 to 11.8 was 

used. The solubility product at 50 °C was -29.89 (Table 23). The above values were used 

to obtain the temperature-dependent ‘log K’ function, which allowed the solubility 

products to be calculated at different temperatures as described in chapter 2.3. The 

temperature dependent solubility product of Fe-Friedel’s salt was computed as shown in 

Fig. 45. The heat capacity of Fe-Friedel’s salt (855 J/(mol.K)) was estimated similar as 

that for the Al-Friedel’s salt (829 J/(mol.K)). The entropy was adjusted until good 

agreement between measured and calculated solubility product was reached. The 

solubility product at standard conditions was also calculated with the help of GEMS-PSI 

using temperature extrapolation from the solubility products calculated at 20 °C and 50 

°C. The thermodynamic properties of the solids at 25 °C are listed in Table 24. 

logKs0 

‐25 

‐27 

‐29 

‐31 

‐33 

11.5 12 

pH 

12.5 13 

Fig. 44 Experimental determined solubility products of Fe-Friedel’s salt as a function of pH. 

103

log Ks0 

‐27 

‐28 

‐29 

‐30 

‐31 

‐32 

‐33 

‐34 


calculated 

Fe‐Fr, this study 

Al‐Fr, Balonis et al 2010 

Al‐Fr, Abate et al 1995 

Al‐Fr, Bothe et al 2004 

Al‐Fr, Birnin‐Yauri et al 1998 

Al‐Fr, this study 

Al‐Fr, Hobbs et al 2001 

0 20 40 60 80 100 


Fig. 45 Calculated solubility products of Fe-Friedel’s salt from the solubility experiments 

compared with the solubility product Al-Friedel’s salt calculated from measured 

concentrations reported in literature [54, 94, 101-103]. 

Table 24 Thermodynamic parameters of Friedel’s salt at standard conditions (25 °C, 1 atm). 

Phases log KS0 

∆fG° 

[kJ/mol] 

ΔfH° 

[kJ/mol] 

S 

t.s: this study, Al-Fr: composition is C3ACl1.95H10.05 

0 

[J/(mol/K)] 

C 0 p 

[J/(mol/K)] 

a0 

[J/(mol.K)] 

a1 

[J/(mol.K 2 )] 

a2 

[JK /(mol)] 

a3 

[J/(mol.K 0.5 )] 

V 0 


Fe-Fr -28.62 -5900.1 -6525 1286 855 481 0.9611 -16130 1503 208 t.s 

Al-Fr -27.69 -6814.6 -7625 731 829 498 0.89 -2.03e+6 1503 272 [54] 

104


3.3.6.2. Determination of the solubility products of the solid solution and 

modeling of the liquid phase 

The total solubility products of the solid solution and the solubility products of Al and 

Fe-Friedel’s salt were calculated based on the solution compositions for Al/Fe-Friedel’s 

salt as presented in Table 25. The concentrations of calcium and chloride were lower for 

the Fe-containing end members and increased when Fe was substituted by Al. At the 

same time, the pH values decreased. 

Table 25 Compositions of Al/Fe-Friedel’s salt synthesized at 20°C and equilibrated for 270 days 

under supersaturated condition. 

Al/(Al+Fe) Al Ca K Fe Cl #pH +pH log Ks0 log Ks0 logΣΠ log Ks0 log Ks0 

Ratio [mmol/l] [mmol/l] [mmol/l] [mmol/l] [mmol/l] Fe-Fr Al-Fr Al/Fe-Fr Fe(OH)3 CH 

0


Fig. 46 Measured (points) and calculated (lines) concentrations in the liquid phases of the 

synthesized Friedel’s salt at different Al/(Al+ Fe) ratio, assuming ideal solid solution. 


The formation of Fe-Friedel’s salt was faster than the other AFm phases. For the 

aluminum based 3CaO.Al2O3.CaCl2.10H2O it was observed that, in the presence of 

carbonate and sulfate, the carbonate and sulfate containing AFm phases are more stable, 

if equimolar dissolved chloride, carbonate or sulfate concentrations are present [54]. At 

high chloride concentrations, however, all AFm phases convert to Friedel’s salt. Based on 

the comparison of the solubility product of 3CaO.Fe2O3.CaCl2.10H2O (-28.44) with Fe- 

monosulfate (-31.57) and Fe-monocarbonate (-34.59), a similar behavior can be expected 

for the Fe-system. As the solubility products of other Fe-containing AFm phases are 2 to 

3 log units lower than those of Al-AFm phases, this indicates a relatively low stability of 

106


Fe-Friedel’s salt in the presence of Al. Thermodynamic properties of Fe-Friedel’s salt 

have not been reported previously. 

The formation of solid solution in Friedel’salt is possible both in the main layer (Al and 

Fe substitution) and in the interlayer (chloride-hydroxy or carbonate substitution). For the 

Al-system, solid solution between carbonate and chloride AFm and the formation of an 

intermediate solid (Kuzel’s salt) is known. The extent of carbonate, hydroxy or sulfate 

substitution in the Fe-based Friedel’s salt is unclear and will need further investigations. 

Fe-Friedel’s salt was found to crystalize in a rhombohedral structure, such as also Fe- 

monocarbonate and Fe-monosulfate. 

107


3.4. Fe-strätlingite 

Different methods were used to synthesis Fe-strätlingite (C2FSH8). Appropriate amounts 

of Fe(OH)3, Na2SiO3 . 5H2O and CaO and alternatively of 2FeCl3.6H2O, Na2SiO3.5H2O, 

and 2Ca(NO3)2.4H2O were mixed in 0.1 M KOH and equilibrated for up to 200 days at 

20, 50 or 80 °C. The synthesis was not successful using either method. No C2FSH8 was 

observed but the formation of portlandite, C-S-H and iron hydroxide occurred, indicating 

the chemical instability of C2FSH8 with respect to these phases. 

108


3.5. Hydrogarnets 


In Portland cement hydrated at ambient temperatures only minor quantities of 

hydrogarnet have been observed [1, 4, 22, 65]. The formation of significant quantities of 

siliceous-hydrogarnet was reported for cements hydrated at higher temperatures [3, 37, 

104, 105] or in the presence of excess Fe(OH)3 [21, 104]. In general, the formation of 

aluminum-iron intermixed siliceous hydrogarnet (Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y) was 

reported [22, 23, 35, 106]. At 200 °C and higher, their composition is strongly influenced 

by the curing temperature as additional amount of silica are built in the structure at higher 

temperatures [107]. In contrast to Portland cement systems, C3AH6 is a major hydrate 

formed during the hydration of calcium aluminate cements [108]. 

Garnet minerals have a cubic structure with the general formula X3Y2(SiO4)3. The X site 

is usually occupied by divalent cations (Ca 2+ , Mg 2+ , Fe 2+ ) and the Y site by trivalent 

cations (Al 3+ , Fe 3+ , Cr 3+ ) in an octahedral/tetrahedral framework with [SiO4] 4− occupying 

the tetrahedral positions. The anhydrous end-members of the Ca3(Al,Fe)2(SiO4)3 series 

are grossular (Ca3Al2(SiO4)3) and andradite (Ca3Fe2(SiO4)3). Hydrogarnet 

(Ca3(Al,Fe)2(SiO4)3-y(OH)4y); y=0-3) includes a group of minerals where the [SiO4] 4− 

tetrahedral are partially or completely replaced by OH - . The Al-containing hydrogarnet 

includes hydrogrossular (Ca3Al2(SiO4)3-y(OH)4y); y=0-3) with the endmember katoite 

(Ca3Al2(OH)12 or C3AH6 in cement notation). The Fe-containing hydrogarnet is 

designated as hydroandradite (Ca3Fe2(SiO4)3-y(OH)4y; y=0-3) and Fe-katoite 

(Ca3Fe2(OH)12) or C3FH6. The nomenclature of minerals of the hydrogarnet group 

Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y as recommended by Passaglia et al. [109] is given in Fig. 

47. 

109

C 3FH 6 

y=3 

Hydroandradite 

(x=0) 

C 3FS 3-yH 2y 

y=0 

Andradite x=0 

C3FS3 CHAPTER 3 SYNTHETIC FE-CONTAINING HYDRATES 

Grandites 

(y=0) 

C 3A xF 1-xS 3 

Katoite 

3≥y>1.5) 

C 3AS 3-yH 2y 

Hibschite 

1.5≥y>0) 

C 3AS 3-yH 2y 

Fig. 47 Nomenclature of minerals of the hydrogarnet group. 

Katoite 

C 3AH 6 

Hydrogrossular 

(x=1) 

C 3AS 3-yH 2y 

x=1 Grossular 

C 3AS 3 

The replacement of the [SiO4] 4− tetrahedral by 4OH - can result in a range of intermediate 

compositions. The extent of the solid solution formation depends strongly on the 

temperature. For Al-containing hydrogrossulars synthesized between 200 and 350 °C, the 

formation of a continuous solid solution from y=0 (C3AS3) to y=2.2 (C3AS0.8H4.4) was 

observed [107]. For samples synthesized at 95 °C, however, a miscibility gap from 

C3AS0.76H4.48 to C3AS0.42H5.52 was reported [110]. While the solubility of C3AH6 is well 

known [7, 91, 111-116], only a few studies determined solubility data for solids with 

composition from C3AS1.5H3 to C3AS0.42H5.52 [7, 91, 112, 117]. For the C3FH6-C3FS3 

solid solutions only estimated thermodynamic data exist [3, 118] and no information on 

the extent of the solid solution or possible miscibility gaps is available. C3FH6 is reported 

to be metastable and to decompose at ambient temperature [119]. It is also unclear to 

what extent Fe substitutes for Al in hydrogarnet. 

110


In present study different hydrogarnet compositions (Ca3(Al,Fe)2(SiO4)3-y(OH)4y); y ≤ 1) 

containing various amounts of aluminum, iron, OH - and SiO4 4- have been investigated. 

The solids were characterized by X-ray powered diffraction, thermogravimetric analysis 

and scanning electron microscopy. Synchrotron diffraction was used to study the 

structure of Fe siliceous hydrogarnet. The solubility products of the different 

hydrogarnets was calculated based on the measured concentrations. 

3.5.2. Al-Katoite, C3AH6 

Several studies have been carried out to date to determine the stability of CaO-Al2O3- 

H2O at different temperatures [7, 111-116]. C3AH6 is thermodynamically stable in the 

CaO-Al2O3-H2O system over the temperature range from 5 to 250 °C [7, 115, 116]. In 

many cases OH-AFm (C4AHx) precipitated in the initial stage, which generally converted 

to C3AH6 with time. Note that C4AHx is metastable with respect to C3AH6 at 20 °C and 

above. Solubility data for C4AHx have been determined from 1 to 90 °C [7, 113-116]. In 

the present study the formation of C3AH6 at 20, 50 and 80 °C was determined in 0.1 M 

KOH solutions (pH values ~ 13 at 20 °C). 

The XRD pattern shows that C3AH6 together with a small amount of C4AHx was already 

observed after 1 week (Fig. 48). The strong reflection assigned to C3AH6 at 2θ = 17.27° 

corresponds to a basal spacing d = 5.13 Å. Minor peaks with basal spacing d = 7.69 Å (2θ 

= 11.49°) were observed, which could be assigned to C4AHx. After three years of 

equilibration, a weak reflection was found with basal spacing d = 8.11 Å (2θ = 10.90°) 

indicating the presence of C4Ac0.5H12 due to CO2 contamination of the sample with time. 

At 50 and 80 °C, C3AH6 formed rapidly. The C4AHx present initially at 50 °C was 

replaced by hemi/mono-carbonate due to CO2 contamination at later ages (see Table 26). 

111


Fig. 49 shows the TGA-DTG curve of C3AH6 where the main loss of water occurred at 

around 300 °C. The water loss up to 350 °C equal to approximately 5H2O. The total 

water loss up to 600 °C (28 wt.%) is consistent with the presence of 6 waters in C3AH6. 

A minor weight loss at around 126 °C indicates the presence of traces of AFm phases 

(C4AHx and hemicarbonate) in the sample after three years of equilibration. 

Fig. 48 Time-dependent XRD pattern of C3AH6 synthesized at 20 °C, * C4AcH11. 

112


Fig. 49 Thermal analysis (TGA and DTG) of C3AH6 and C3FH6 synthesized at 20 °C and sampled 

after different equilibration times. 

113


Table 26 Measured ion concentrations at different equilibration times in 0.1 M KOH 

Age Temp. Ca Al Fe K +pH logKs0 logKs0 logKs0 logKs0 Solid phases present 

(days) [°C] [mmol/l] [mmol/l] [mmol/l] [mmol/l] 

C3AH6/ 

C3FH6 

C4AH13/ 

C4FH13 CH FH3 

C3AH6 

7 20 2.1 1.3 < D.L. 98 13.0 -20.14 -25.57 C3AH6, C4AHx 

28 20 1.8 1.4 < D.L. 103 13.0 -20.28 -25.78 C3AH6, C4AHx 1100 20 0.8 3.4 < D.L. 89 13.1 -20.64 -26.54 C3AH6, C4AHx 

Average -20.35±0.20 -25.96±0.25 

7 50 2.0 2.0 < D.L. 105 13.0 -20.45 -26.11 C3AH6, C4AHx, C4AcH11 

28 50 1.8 1.6 < D.L. 101 13.0 -20.83 -26.60 C3AH6, C4AHx, C4AcH11 

Average -20.64±0.20 -26.36±0.25 

7 80 2.2 2.4 < D.L. 113 13.0 -20.99 C3AH6, C4AcH11 

28 80 2 2.7 < D.L. 106 13.0 -21.02 C3AH6, C4AcH11 

Average -21.01±0.20 

C3FH6 

7 20 5.5 < 0.1 0.0124 95 13.0 -4.97 -3.98 C4FHx,C2F, CH 

28 20 4.7 < 0.1 0.0004 86 13.1 -26.06 -31.12 -5.05 -5.46 C4FHx,C2F, CH 

365 20 5.6 < 0.1 0.0003 96 13.0 -26.07 -31.05 -4.71 -5.62 C3FH6, C4FHx, CH 

1100 20 5.1 < 0.1 0.0004 92 13.0 -25.95 -4.96 -5.59 C3FH6,CH, C4FcH12 

Average >-26.03±0.20 -31.08±0.25 -4.92±0.04 -5.16±0.05 

7 50 4.8 < 0.1 0.0099 98 13.0 -23.89 -29.15 C3FH6, C4FHx, CH, C2F 

28 50 4.7 < 0.1 0.0002 88 13.1 CH, Fe2O2 

365 50 2.1 < 0.1 0.0004 101 13.0 CH, Fe2O3 

Average >-23.89±0.20 -27.1±0.25 

7 80 4.4 < 0.1 0.0043 91 13.0 -5.57 -4.42 CH, Fe2O3 28 80 1.8 < 0.1 0.0009 101 13.1 -5.97 -5.08 CH, Fe2O3 365 80 1.8 < 0.1 0.0005 96 13.0 -5.97 -5.34 CH, Fe2O3 

Average -5.84±0.20 -4.95±0.20 

Detection limits (mmol/l): Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0.0002 measurement uncertainty ±10% , + pH measured at 20 °C. Due to the strong dependence of the H + 

activity on temperature, a pH of 13.0 at 20 °C corresponds to 12.1 at 50 °C and to 11.5 at 80 °C. 

114


Based on the measured compositions of the liquid phase (as given in Table 26), the 

solubility products at 20, 50 and 80 °C were calculated according to: 

KS0 (C3AH6) = {Ca 2+ } 3 . {AlO2 - } 2 . {OH - } 4 . {H2O} 4 

The values estimated for different equilibration times are given in Table 26 and shown in 

Fig. 50. The calculated solubility products are comparable with previously published data 

within the experimental error [7, 111, 112, 115]. The temperature dependent solubility 

product of C3AH6 was determined based on the solubility measured at 20, 50 and 80 °C 

in this study and from solubilities reported in the literature. The entropy was visually 

fitted until a good agreement between measured and calculated data was achieved (Fig. 

50). 

logKs0 

-20 

-20.5 

-21 

-21.5 

-22 

-22.5 

-23 

-23.5 

-24 

0 20 40 60 80 100 120 


Matschei et al 2007 

Atkins et al 1991 

Bennet et al 1992 

Peppler et al 1954 

this study 

Lothenbach et al 2011 

Roberts et al 1969 

Calculated 

Carlson et al. 1968 

Wells et al. 1943 

Fig. 50 Solubility products of C3AH6 calculated from the solubility experiments carried 

out in this study and from different published data [7, 111, 112, 114-116]. 

115


The data of Carlson et al. [114] and Wells at el. [116] obtained at 1 and 21 °C were not 

considered as they deviate strongly from the other published data. 

The co-precipitation of C4AH13 with the target phase in all equilibration experiment 

enabled us to estimate the ion activity products of the OH-AFm based on the following 

equation: 

KS0 (C4AH13) = {Ca 2+ } 4 . {AlO2 - } 2 . {OH - } 6 . {H2O} 10 

The calculated data are given in Table 26. The ion activity products were determined to 

be -26.02 and -26.45 at 20 °C and 50 °C, respectively. These values are slightly lower 

than the value of -25.53 (20 °C) and -25.06 (50 °C) derived elsewhere [3]. The solubility 

at 50 °C is difficult to measure as the C4AH13 transformed relatively fast to C3AH6 or 

monocarbonate and thus the system possibly was undersaturated at the time of sampling. 

The standard molar thermodynamic properties of C3AH6 determined in this study are 

summarized in Table 27. 

Table 27 Thermodynamic parameters at standard conditions determined in this study (25°C, 1 

atm). 

Phases log KS0 

fG° 

[kJ/mol] 

fH° 

[kJ/mol] 

S 0 

[J/mol/K] 

C 0 p 

[J/mol/K] 

a0 

[J/(mol.K)] 

a1 

[J(/mol.K 2 )] 

a2 

[JK /mol] 

a3 

[J/(mol.K 0.5 )] 

C3AH6 -20.56 -5008.8 -5535 432 459 292 0.5610 150 

116 

V 0 

[cm 3 /mol] 

C4FH13 -30.64 -6438.0 -7431 640 956 694 1.1134 2.02E+06 -1600 286 

C3FS0.95H4.1 -32.75 -4523.5 -4854 855 612 582 0.6094 2.19E+06 -3040 156 

C3FS1.52H2.96 -34.68 -4752.8 -5044 847 688 766 0.5988 2.29E+06 -4864 161 

C3AS0.41H5.18 -25.47 -5193.5 -5717 342 441 198 0.5967 -9.98E+05 1312 151 

C3AS0.84H4.32 -26.70 -5365.2 -5867 310 421 100 0.6342 -2.05E+06 2688 142


3.5.3. Fe-katoite, C3FH6 

C3FH6 was attempted to synthesize at 20, 50 and 80 °C by mixing C2F and CaO in 0.1 M 

KOH. Trace amounts of metastable C3FH6 were observed in the samples equilibrated up 

to 1 year at 20 °C (2θ = 19.71, Fig. 51). However, C3FH6 disappeared and decomposed to 

portlandite and iron hydroxide with time. At longer equilibration time C4FcH12 was 

formed due to CO2 contamination, which destabilized the CaO-Fe2O3-H2O system. This 

shows the metastability of C3FH6 with respect to Fe-hydroxide and portlandite and in the 

presence of CO2 to hemi/monocarbonate. This metastability of C3FH6 agrees with the 

findings of Ecker et al. [12] and Rogers et al. [119]. The CaO-Fe2O3-H2O system is very 

sensitive to carbonation at room temperature. The sample synthesized hydrothermally at 

110° C contained portlandite and Fe-oxide/hydroxides and traces of C3FH6. Another 

weak reflection was observed corresponding to a basal spacing d = 7.58 Å (2θ=11.65°) 

which persisted at longer age. It does fit neither to C4Fc0.5H10 nor to C4FcH12. The peak 

was tentatively assigned to C4FHx with an unknown number of water molecules. All the 

samples showed a reddish color of the solids indicating the formation of XRD amorphous 

Fe-oxide/hydroxide in the CaO-Fe2O3-H2O system. The TGA-DTG curve also revealed 

the formation of traces of metastable C3FH6 (see Fig. 49). 

117


Fig. 51 Time-dependent XRD pattern of C3FH6 synthesized at 20 °C and the sample 

synthesized at 110 °C and equilibrated for 5 days. 

At 50 °C small amount of C3FH6 and C4FHx were observed in the solid equilibrated for 7 

days. Portlandite and iron hydroxide were the main phases observed at all equilibration 

times. At longer age, hematite (Fe2O3) started to crystallize as clearly observed by XRD. 

At 80 °C portlandite and hematite were the main constituents in the solid (see Table 26). 

This indicates the decomposition of iron hydroxide to hematite at higher temperature. 

Carbonation was not encountered at higher temperatures. 

The ion activity product of the metastable C3FH6 was estimated according to: 

Ks0 (C3FH6) = {Ca 2+ } 3 {FeO2 - } 2 {OH - } 4 {H2O} 4 

Synthesis of the solid shows that OH-AFm persisted up to one year. Assuming C4FHx is 

stable in the absence of carbonate, the solubility products were estimated as follows: 

Ks0 (C4FH13) = {Ca 2+ } 4 {FeO2 - } 2 {OH - } 6 {H2O} 10 

118


The values are listed in Table 26. The calculated ion activity product of C3FH6 represents 

a lower limit and will be higher in reality. The presence of both, the C3FH6 and the 

C4FH13, makes it impossible to determine the solubility product of the metastable C3FH6. 

As C4FH13 converted to monocarbonate with time, this solubility product should be 

considered as a rough estimate. The solubility product of C4FH13 is approximately 5 log 

units lower than that of C4AH13, (see Table 26). The solubility product of C3FH13 at 

standard conditions was calculated based on the solubility measured at 20 °C and 50 °C 

(see Table 27) 

3.5.4. Solid solution between aluminum and iron katoite, C3(A,F)H6 

The C3AH6 and C3FH6 have similar cubic structures with the unit cell parameter a = 

12.58 and 12.72 Å, respectively. It is very likely that similarity of the crystal structure 

could lead to a solid solution between those two compounds. Flint et al. [120] indicated 

that a solid solution between C3AH6 and C3FH6 exists only in the presence of silica. 

Roger et al. [119] pointed out that in the presence of iron, C3(A,F)H6, is unstable and 

converts to form portlandite, Fe-hydroxide, hematite and C3AH6. 

119


Fig. 52 XRD pattern of mixed Al and Fe hydrogarnets after 3 years equilibration. 

As shown above in Fig. 51, C3FH6 is unstable and the same was observed for mixed 

C3(A,F)H6 systems. In agreement with previous observations [119, 120], no substitution 

between Al 3+ by Fe 3+ in C3(A,F)H6 samples occurred in the absence of silica. In the 

absence of iron, C3AH6, hemicarbonate and OH-AFm were the dominant phases. The 

presence of iron led to the formation of portlandite, Fe-hydroxide, C3AH6, C4FHx and Al 

and Fe carbonate containing AFm phases (due to CO2 contamination) (see Fig. 52 ). No 

peak shift was observed for C3AH6, indicating the absence of a solid solution between 

C3AH6 and C3FH6. 

The solubility products of all the phases in the solid were calculated and listed in Table 

28. The Fe-containing solutions are also saturated with respect to portlandite and Fe- 

hydroxide. The calculated ion activity products of portlandite and Fe-hydroxide also 

served as an independent quality check of the concentration measurements. The 

calculated solubility products for Al-katoite and the OH-AFm remained constant 

independent of the mixing proportions. This indicates no solid solution formation, thus 

120


confirming the XRD results. The calculated average solubility products from these mixed 

systems are comparable with the values determined in the systems which contained only 

Al or Fe (see Table 26). 

Table 28 Measured concentration of mixed C3AH6-C3FH6 systems equilibrated for three years 

at oversaturation. 

Al/(Al+Fe) Al Ca K Fe pH log Ksp log Ks0 log Ks0 log Ks0 log Ks0 log Ks0 

ratio [mmol/l] [mmol/l] [mmol/l] [mmol/l] C3FH6 C4FH13 C3AH6 C4AH13 CH FH3 1 3.37 0.8 89.0 13.1 -20.64 -26.54 

0.9 0.12 5.4 88.9 0.0003 13.0 -26.11 -31.10 -20.91 -25.90 -4.98 -5.59 

0.8 0.15 4.9 90.0 0.0005 13.0 -25.82 -30.86 -20.87 -25.91 -5.03 -5.37 

0.7 0.06 5.1 90.5 0.0004 13.0 -25.95 -30.97 -21.60 -26.61 -5.01 -5.47 

0.6 0.11 5.5 91.3 0.0004 13.0 -25.86 -30.84 -20.98 -25.97 -4.97 -5.47 

0.5 0.13 5.1 90.7 0.0055 13.0 n.d n.d -20.94 -25.96 -5.00 n.d 

0.4 0.15 5.1 88.3 0.0005 13.0 -25.77 -30.79 -20.82 -25.84 -5.01 -5.37 

0.3 0.15 4.8 90.1 0.0013 13.0 -25.01 -30.06 -20.89 -25.94 -5.04 -4.95 

0.2 0.16 5.1 89.7 0.0011 13.0 -25.08 -30.11 -20.76 -25.78 -5.01 -5.03 

0.1 0.15 5.1 90.3 0.0003 13.0 -26.20 -31.22 -20.81 -25.82 -5.01 -5.59 

0 0.02 5.1 91.7 0.0004 13.0 -25.95 -30.96 -5.01 -5.47 

Average -25.75±0.20 -30.77±0.25 -20.92±0.20 -26.03±0.25 -5.01±0.04 -5.35±0.05 

Detection limits (mmol/l): Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0.0002 measurement uncertainty ±10%, 

+ pH measured at 20°C, n.d: not determined due to unexpected higher iron concentrations indicating contamination of the sample. 

Based on the derived solubility products as given in Table 28 the influence of varying 

Al/(Al+Fe) ratio on the stable hydrates in the CaO-Al2O3-Fe2O3–H2O system was 

calculated. A small amount of CO2 was added to conform to the CO2 contamination 

observed in the experiments. The formation of C3AH6, portlandite, Fe-hydroxide and 

carbonate containing AFm phase was calculated but no C3FH6 formation (see Fig. 53), in 

agreement with the XRD data. 

121

Mass of the solids in grams 

2.5 

2.0 

1.5 

1.0 

0.5 


Fe-hydroxide 

Portlandite 

C 4Ac 0.5H 12 

0.0 

0.0 0.2 0.4 0.6 0.8 1.0 

(Al/Al+Fe) ratio 

C 3 AH 6 

Fig. 53 Calculated solids in the CaO-Al2O3-Fe2O3-H2O system in 0.1 M KOH using the solubility 

products as given in Table 28. 

3.5.5. Aluminum siliceous hydrogarnet, C3ASH4 

The C3ASH4 system is more complex than the silicon free C3AH6 system. The presence 

of two different hydrogarnets can be expected, a silica-poor and a silica-rich with a 

miscibility gap in between. For hydrogarnets prepared at 95 °C, a miscibility gap between 

C3AS0.42H5.16 and C3AS0.76H4.48 was reported [110]. In this study the synthesis of C3ASH4 

(target composition) was carried out at both 20 °C and 110 °C. Fig. 54 shows the XRD 

pattern of the Al-containing siliceous hydrogarnet synthesized at 110 °C (equilibrated for 

5 days) and at 20 °C (equilibrated for 3 years). The peaks of the sample synthesized at 

110 °C are relatively broad. A minor splitting of the peaks was observed at d-spacing ~ 

2.75 Å, indicating that the solid could be composed of two siliceous hydrogarnets with 

different compositions. The peaks of the solid synthesized at 20 °C indicated the presence 

122


of C3AH6 only even after a reaction time of 3 years. This shows that the synthesized solid 

contains no or very little silica. 

Fig. 54 The XRD pattern of Al containing Si-hydrogarnet synthesized at 20 °C and 110 °C. * Al- 

Si-hydrogarnet with two different compositions (see inlet); o C3AH6; - KNO3 present as 

impurity; +CaF2 added as an internal standard. 

The TGA-DTG curve shows a weight loss at 145 °C due to the co-precipitation of C-S-H 

(see Fig. 55). The water loss of Al-Si-hydrogarnet synthesized at 20 °C corresponds to 

that observed for C3AH6 (see Fig. 49) with a main peak at around 300 °C. This supports 

the XRD results and indicates that the Al-Si-hydrogarnet synthesized at 20 °C contains 

little or no silica. The main water loss of Al-Si-hydrogarnet synthesized at 110 °C 

occurred around 330 °C. The presence of C-S-H gel in the samples with the target 

composition C3ASH4 was persistent. 

123


Fig. 55 Thermal analysis (TGA and DTG) of Al-and Fe-Si hydrogarnet synthesized at 20 °C and 

110 °C. The circle region indicates the water loss of hydrogarnets with different 

compositions. 

The complete absence of silica containing hydrogarnet in the samples synthesized at 20 

°C even after 3 years aging indicates that the formation at room temperature is kinetically 

hindered and does not occur not even after prolonged hydration. 

Rietveld refinement was performed with the addition of CaF2 as an internal standard to 

determine the unit cell of the solid synthesized at 110 °C. The structure of silica poor (a ~ 

12.38Å, ICSD N o 49772) and silica rich (a ~ 12.27Å, ICSD N o 172076) siliceous- 

hydrogarnet were used as starting values for the refinement. The resulting unit cell 

parameters were a ~ 12.47Å and a ~ 12.37Å for a silica poor and rich siliceous 

124


hydrogarnet respectively. The silica content of the two hydrogarnets was estimated by 

assuming a linear relationship of unit cell parameters between C3AH6 (a ~ 12.58Å PDF 

24217) and grossular C3AS3 (a ~ 11.85Å PDF 1741087) [7, 110] (see Fig. 56). The 

compositions of the synthesized siliceous-hydrogarnet were estimated to be C3AS0.41H5.18 

and C3AS0.84H4.32. In agreement with the published data a miscibility gap occurred [107, 

110, 112]. The miscibility gap observed in the present study ranged from C3AS0.41H5.18 to 

C3AS0.84H4.32. According to the reflection intensity, the silica poor hydrogarnet was 

dominant (see inlet in Fig. 54). 

Garnet silica content 

3 

2 

1 

C 3AS 3 PDF1741087 

C3AS2.3H1.4 PDF1842016 

C3AS2H2 Kyritsis(2009) 

C 3AS 2H 2 PDF1731654 

C 3AS 1.5H 3 Kyritsis(2009) 

C3ASH3.5 PDF451447 

C3ASH4 Kyritsis(2009) 

C3AS1.09H3.82 ICSD 172076 

C3AS0.84H4.32 (this study) 

Miscibility gap 

C 3AS 0.64H 4.72 ICSD 49772 

C 3AS 0.8H 4.4 (Matschei 2007) 

C 3AS 0.41H 5.18 (this study) 

0 

11.8 12 12.2 

Unit cell size [Å] 

12.4 12.6 

C 3AH 6PDF24217 

Fig. 56 Estimation of the silica content for synthesized Al-containing hydrogarnet; PDF: Powder 

Diffraction File. ICSD: Inorganic Crystal Structure Database. The composition of the 

synthesized solid solution series was estimated from the unit cell size as indicated by the 

line. 

The solubility products of the silica poor hydrogarnet and silica rich hydrogarnet were 

125


determined based on the chemical formula obtained from XRD analysis of the solids in 

the dissolution experiment (Table 28). 

The solubility products of both hydrogarnets were calculated using the following 

equations: 

KS0 (Ca3Al2(SiO4)0.41(OH)10.36) = {Ca 2+ } 3 {AlO2 - } 2 {HSiO3 - } 0.41 {OH - } 3.59 {H2O} 3.18 

KS0 (Ca3Al2(SiO4)0.41(OH)8.64) = {Ca 2+ } 3 {AlO2 - } 2 {HSiO3 - } 0.84 {OH - } 3.16 {H2O} 2.32 

Table 29 Measured ion concentrations in the solution of solids synthesized at 110°C and re 

dissolved and equilibrated for 4 months at 20 °C and 50 °C. 

Phases 

Temp. 

[°C] 

Al 

[mmol/l] 

Ca 

[mmol/l] 

Si 

[mmol/l] 

K 

[mmol/l] 

Si 

[mmol/l] 

+ pH log Ks0 

Ca3Al2 (SiO4)0.41(OH)10.36 20 2.99 0.66 0.51 151 0.51 13.12 -25.58±0.20 

Ca3Al2 (SiO4)0.84(OH)8.64 20 2.99 0.66 0.51 151 0.51 13.12 -26.83±0.20 

Ca3Al2 (SiO4)0.41(OH)10.36 50 4.63 0.07 0.67 163 0.67 13.04 -25.19±0.30 

Ca3Al2 (SiO4)0.84(OH)8.64 50 4.63 0.07 0.67 163 0.67 13.04 -26.38±0.30 

Detection limits [mmol/l]: Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0.0002 measurement uncertainty ±10%, 

+ 

pH measured at 20°C. 

The resulting solubility products are log Ks0 = -25.58 and -26.83 at 20 °C and log Ks0 = - 

25.19 and -26.38 at 50 °C for C3AS0.41H5.18 and C3AS0.84H4.32, respectively. The 

solubility products of Al containing siliceous hydrogarnets from different studies are 

plotted in Fig. 57. The solubility products calculated from the concentrations measured 

by Matschei et al. [7] for C3AS0.8H4.4, which had been equilibrated for 4 weeks at 

undersaturation, were log Ks0 = -30.46, -30.03, -29.34, -29.50 and -29.74 at 5, 25, 55, 70 

and 85 °C, respectively. From the concentrations and compositions determined by Bennet 

et al. [112] for C3AS0.41H5.18 and C3ASH4 equilibrated for 4 weeks at 25 °C at 

undersaturation log Ks0 of -29.59 (C3AS0.41H5.18) and -30.17 (C3ASH4) were calculated. 

These values are very similar to the values determined by Matschei et al. [7] as shown in 

126


Fig. 57. From the data of Atkins et al. [117] a significantly higher solubility product (log 

Ks0 = -27.20 and -28.09) was obtained for C3ASH4 indicating a lower stability. The 

samples from the study of Atkins et al. [117] had been equilibrated for 10 weeks and 6 

months at 25 °C under oversaturation. Jappy et al. [110] determined the solubility at 95 

°C under oversaturation for C3AS0.09H5.82, C3AS0.76H4.48 and C3AS1.14H3.72. The 

calculated solubility products (log Ks0 = -25.02, -27.19 and -29.96) decrease strongly 

with increasing silica content as shown in Fig. 57. Thus, the solubility products calculated 

from different studies are significantly dependent on the amount of silica, temperature 

and the experimental conditions (oversaturation or undersaturation) employed. The 

values determined in this study at undersaturation experiments (equilibration time 4 

months) are similar to the values of Atkins et al. [117] obtained at oversaturation and 

clearly higher than the values obtained at undersaturation experiments where the solids 

had been equilibrated for 4 weeks [7, 112] as shown in Fig. 57. This could be related to 

the longer equilibration time and the amount of silica present in the solids. 

127

logKs0 

-20 

-22 

-24 

-26 

-28 

-30 

-32 

Si 0.8 


Si 0.84 

Si 0.41 

Si 0.8 

Si 1.0 

Si 1.0 

Si 0.41 

Si 1.0 

C 3AH 6 

Si 0.41 

Si 0.84 

Si 0.8 

0 10 20 30 40 50 60 70 80 90 100 

Temperature [°C] 

C 3AS 0.41H 5.18 

C 3AS 0.84H 4.32 

Si 0.8 

Si 0.8 

Si 0.09 

Si 0.76 

Si 1.14 

Atkins 1992 

Jappy 1991 

Bennet 1992 

this study 

Matschei 2007 

Fig. 57 Comparison of published solubility products of Al-Si-hydrogarnet calculated in this study 

from the data reported in [7, 110, 112, 117], C3AH6 (dashed line), C3AS0.41H5.18 (solid 

line), C3AS0.84H4.32 (dotted line). 

The solubility products calculated from the solution composition determined at 20 °C and 

25 °C were also plotted as a function of the silica content. Note that the solubility product 

of C3AS3 (logKs0= -41.78 and -38.6) was calculated from fG° = -6278.50 kJ/mol (Robie 

et al [121]) and fG° = -6260.55 kJ/mol (Sverjensky et al. [122] ) using the following 

equilibration reaction: 

Ca3Al2Si3O12 + 2H2O ↔ 3Ca 2+ + 2AlO2 - + 3HSiO3 - + OH - 

The thermodynamic data of water and the dissolved species used in the above reaction 

were taken from the GEMS thermodynamic database [121-125]. Fig. 58 shows that the 

solubility product decreases strongly with increasing silica content. This confirms that the 

128


presence of silica stabilizes (hydrothermally prepared) hydrogarnets. 

The relatively low solubility product obtained for C3AS0.41H5.18 compared to C3AH6 and 

C3AS0.84H4.32 could indicate (i) that this sample was not yet equilibrated or (ii) a strong 

stabilization of C3AS0.41H5.18. 

Fig. 58 Solubility products as a function of Si content between C3AH6 and C3AS3 end members at 

25 °C. 

3.5.6. Iron siliceous hydrogarnet, C3FSH4 

The formation of Fe-containing siliceous hydrogarnet was reported [120, 126, 127]. 

C3FH6 is stabilized by the presence of silica and Ca3Fe(SiO4)3-x(OH)4x is formed due to 

the substitution of (OH)4 4- by SiO4 4- . Fe siliceous hydrogarnet was synthesized from C2F 

and NaSiO3.5H2O at 20 °C and investigated at different equilibration times. Fig. 59 

129


shows the slow reaction of C2F resulting also in slow formation of Fe siliceous 

hydrogarnet at 20 °C. After 3 years of equilibration very poorly crystalline Fe siliceous 

hydrogarnet was observed. In contrast, the formation of Fe siliceous hydrogarnet was fast 

at 110 °C, producing a well crystalline phase. This material was used for structural 

studies and dissolution experiments to determine the thermodynamic properties of Fe- 

siliceous hydrogarnet. 

Fig. 59 Time-dependent XRD pattern of C3FSH4 synthesized at 20 °C from C2F, * the solid 

synthesized at 110 °C. R: rutile. 

The TGA-DTG curve shown in Fig. 55 illustrates the main water loss from the structure 

of Fe siliceous hydrogarnet at 275 °C. The curve also shows a weight loss at around 145 

°C and 610 °C due to the precipitation of C-S-H gel and traces of CaCO3, respectively. 

Evidently, it not possible to avoid co-precipitation of C-S-H in CaO-Al2O3-Fe2O3-SiO2 

systems. 

130


The sample prepared at 110 °C was used for the structural determination. Multipattern 

Rietveld refinement was performed (Fig. 60). Traces of impurity were identified as 

indicated by weak diffraction peaks observed at d = 3.77 Å and 3.47 Å (see Fig. 60, 

bottom for 2 = 10.6° and 11.5°). The sample is not single-phased, and two Fe siliceous 

hydrogarnet phases have been observed (Fig. 61). The measured sample is composed of 

Fe siliceous hydrogarnet phase N°1 (64 weight %), Fe siliceous hydrogarnet phase N°2 

(27 wt. %), calcite CaCO3 (8 wt. %) and C3FH6 (1 wt. %). 

The two Fe siliceous hydrogarnet phases correspond to the structure already described for 

hydroandradite, the Al-free iron-hydrogarnet compound [126]. The two Fe siliceous 

hydrogarnet phases with cubic symmetry have different lattice parameters (a = 12.5424 

(5) Å and 12.4297 (7) Å, respectively for Fe siliceous hydrogarnet N°1 and Fe siliceous 

hydrogarnet N°2) and different silica contents (refined composition are 

Ca3Fe2(SiO4)0.95(2)(OH)8.20(2) and Ca3Fe2(SiO4)1.52(4)(OH)5.92(4)). A miscibility gap seems 

to appear between these two hydroandradites. Table 30 compiles the refined parameters 

for both hydroandradite phases which crystallized in the cubic Ia3 d space group. 

131


Fig. 60 Rietveld plot for Fe-Si-Hydrogarnet sample with = 0.697751Å and a sample-to-detector 

distance of 150 mm (top) and 400 mm (bottom). 

132


Fig. 61 Zoom of the Rietveld plot from pattern recorded for a sample-to-detector distance of 400 

mm showing the two hydrogarnet phases (systematic shoulders, right side, for 

hydrogarnet diffraction peaks). 

Table 30 Refined structure parameters of Fe siliceous hydrogarnet (standards deviation are 

indicated in parentheses). 

Hydrogarnet N°1, a = 12.5424 (5) Å, Ca3Fe2(SiO4)0.95(2)(OH)8.20(2) 

Atom Wyckoff x Y z B iso (Å 3 ) Occupancy 

Ca 24 1/8 0 1/4 1.10 (5) 1 

Fe 16 0 0 0 = Biso (Ca) 1 

Si 24 3/8 0 1/4 = Biso (Ca) 0.316 (8) 

O 96 0.0336 (2) 0.0521 (2) 0.6492 (3) = Biso (Ca) 1 

Hydrogarnet N°2, a = 12.4297 (7) Å, Ca 3Fe 2(SiO 4) 1.52(4)(OH) 5.92(4) 

Atom Wyckoff x Y z B iso (Å 3 ) Occupancy 

Ca 24 1/8 0 1/4 = Biso (Ca) 1 

Fe 16 0 0 0 = Biso (Ca) 1 

Si 24 3/8 0 1/4 = Biso (Ca) 0.51 (1) 

O 96 0.0336 (2) 0.0521 (2) 0.6492 (3) = B iso (Ca) 1 

133


Using the compositions determined from XRD, the solubility products of both 

hydroandradite phases were calculated according to the following equations: 

KS0 (Ca3Fe2(SiO4)0.95(2)(OH)8.20(2)) = {Ca 2+ } 3 . {FeO2 - } 2 {HSiO3 - } 0.95 . {OH - } 3.05 . {H2O} 2.1 

KS0 (Ca3Fe2(SiO4)1.52(2)(OH)5.92(4)) = {Ca 2+ } 3 . {FeO2 - } 2 {HSiO3 - } 1.52 . {OH - } 2.48 . {H2O} 0.96 

The calculated values from the dissolution experiments are given in Table 31. The 

resulting solubility products are log Ks0= -32.34 and –34.50 at 20 °C and log Ks0 = -33.68 

and -35.76 at 50 °C for C3FS0.95H4.1 and C3FS1.52H2.26, respectively. The solubility 

products of Fe-Si-hydrogarnet synthesized at 20 °C and equilibrated under oversaturation 

are very similar (log Ks0 = -32.98 and –34.63 at 20 °C for C3FS0.95H4.1 and C3FS1.52H2.96). 

The values are 7 log units lower than those of the Al-containing siliceous hydrogarnet 

determined in this study. Hence, hydroandradites are significantly more stable than the Al 

containing hydrogrossulars. 

Table 31 Measured ion concentrations of solids synthesized at 110 °C (re dissolved and 

equilibrated for 4 months at 20 °C and 50 °C) and at 20 °C(equilibrated for 3 years 

under oversaturated condition). 

Phases 

Temp. 

[°C] 

Ca 

[mmol/l] 

K 

[mmol/l] 

Si 

[mmol/l] 

Fe 

[mmol/l] 

+ pH log Ks0 

Ca3Fe2 (SiO4)0.95(OH)8.2 20 0.10 118 0.05 0.0061 13.04 -32.34±0.20 

*Ca3Fe2 (SiO4) 0.95(OH) 8.2 20 0.20 85 0.26 0.0008 13.10 -32.98±0.20 

Ca3Fe2 (SiO4)1.52(OH)5.92 20 0.10 118 0.05 0.0061 13.04 -34.50±0.20 

*Ca3Fe2 (SiO4)1.52(OH)5.92 20 0.20 85 0.26 0.0008 13.10 -34.63±0.20 

Ca3Fe2 (SiO4)0.95(OH)8.2 50 0.06 125 0.06 0.0044 13.02 -33.68±0.20 

Ca3Fe2 (SiO4) 1.52(OH) 5.92 50 0.06 125 0.06 0.0044 13.02 -35.76±0.20 

Detection limits (mmol/l): Al = 0.001 Ca = 0.0004, Fe = 0.0000006, K = 0.0024, S=0.0014, Si=0.0002 measurement uncertainty ±10%, 

+ 

pH measured at 20°C, *solubility of Fe-siliceous hydrogarnet equilibrated for 3 years under oversaturated condition. 

134


Fig. 62 Calculated solubility products of Fe-Si-hydrogarnets from the solubility experiments. 

(lines show calculated values, full symbols show the measured values from 

undersaturation and empty symbols from oversaturation). 

The temperature dependent solubility product of hydroandradite was computed based on 

the solubility measured at 20 °C and 50 °C. The entropy was fitted using measured 

solubility products as described before (Fig. 62) for both hydroandradite. The solubility 

products decreases with increasing temperature indicating a higher stability at higher 

temperature. The thermodynamic parameters at standard conditions are in Table 27. 

135


3.5.7. Solid solution between Ca3Fe2(OH)12 and Ca3Fe2O6(SiO2)3 

(hydroandradite, Ca3Fe2(SiO4)3-y(OH)4y) 

It is known that in the tetrahedral sites of the cubic hydrogarnet (OH)4 4- can be replaced 

by SiO4 4- . The existence of a solid solution between C3AH6 and C3AS3 with a miscibility 

gap between C3AS0.41H5.18 and C3AS0.84H4.32 has been demonstrated. Hydroandradite, 

Ca3Fe2(SiO4)3-y(OH)4y, was synthesized at 110 °C with varying y = 0, 1.5, 1.75, 2, 2.25, 

2. 75 and 3 to study the substitution of (OH)4 by SiO4. Fig. 63 shows the XRD pattern of 

the solid solution series of Ca3Fe2(SiO4)3-y(OH)4y. If no SiO2 was present, the iron 

hydrogarnet was metastable with respect to portlandite and Fe-oxide/hydroxide and only 

traces of C3FH6 were observed. However, silica substitution raised the stability, 

portlandite disappeared and a stable hydroandradite was formed (Fig. 63). 

Fig. 63 XRD pattern of the solid solution series of Ca3Fe2(SiO4)3-y(OH)4y, + CH. The dotted lines 

indicate the peak shifts. *Main reflections of the hydroandradite end members. Note that 

Xsi = y = 3 

136


At higher silica contents a peak shift from XSi 0.25 to 1.0 was observed, indicating a 

continuous change from approximately C3FS0.25H5.5 to C3FS0.95H4.1 and the presence of a 

solid solution could be possible. Between XSi = 1 and XSi = 1.5 a peak broadening was 

observed due to the presence of two hydrogarnets with different silicon contents 

(C3FS0.95H4.1 and C3FS1.52H2.26) as discussed above in connection with the synchrotron 

diffraction data. This finding indicates the presence of a miscibility gap between this 

range. At XSi = 1.5 and at XSi = 3, different peak positions were observed. Whether a 

continuous solid solution exists between these two solids or whether further miscibility 

gaps occur cannot be determined from the data presented here. 

The change in the solubility products with increasing silica content at 25 °C are shown in 

Fig. 64. The solubility product of C3FS3 (logKs0 = -53.51 and -53.31) was calculated from 

fG° = -5427.0 kJ/mol (Robie et al. [121]) and fG° = -5425.89 kJ/mol (Sverjensky et al. 

[122]) using the following equilibrium reaction. 

Ca3Fe2Si3O12 + 2H2O ↔ 3Ca 2+ + 2FeO2 - + 3HSiO3 - + OH - 

The thermodynamic data of water and the dissolved species used in the above reaction 

were taken from the GEMS thermodynamic database [121-125]. As shown in Fig. 64, the 

solubility product decreases with an increase of the silica content indicating that the 

presence of silica stabilizes the hydroandradite system. 

137


Fig. 64 Solubility products as a function of Si content in between C3FH6 and C3FS3 end members 

at 25 °C. The dotted line connects the solubility products of C3FH6 and C3FS3. 

3.5.8. Solid solution between aluminum and iron siliceous hydrogarnet, 

C3(A,F)SH4 

As discussed above, Al and Fe may substitute each other. Between C3AH6 and C3FH6 no 

solid solution formation was observed (see section 3.5.5). Frank-Kamenetskaya et al. 

[128] demonstrated the possible solid solution formation in the grossular-andradite series. 

In this study, a Ca3(AlxFe1-x)2(SiO4)(OH)8 solid solution series was synthesized at 110 °C 

by varying the Al/(Al + Fe) ratio from 0 to 1 (Fig. 65). XRD and TGA indicated that the 

solids contained impurities like C-S-H and calcite. The peaks were broad and a peak 

138


splitting was observed at d-spacing ~ 2.75 Å with a slight peak shift with increasing Fe 

content. This finding attributed to the formation of at least two hydrogarnets with 

different compositions. XRD investigation shows that Al-Si-hydrogarnet and the Fe-Si- 

hydrogarnet deviate from their target compositions during preparation. Two hydrogarnets 

with different silica contents were observed, confirming the existence of miscibility gaps 

in the (OH)4 4- -SiO4 4- series. However, due to similarity of their structures the formation of 

ideal solid solution between Al- and Fe-Si-hydrogarnet is tentatively proposed. A 

possible substitution of Si by OH in the tetrahedral site of the hydrogarnet could occur 

depending on the Fe content. In general, a simultaneous substitution in both the 

octahedral and the tetrahedral site of hydrogarnet could be possible. 

Fig. 65 XRD pattern of the solid solution series of Ca3(AlxFe1-x)2(SiO4)(OH)8 synthesized at 110 

°C. 

The solution compositions measured for Ca3(AlxFe1-x)2(SiO4)(OH)8 at 20 °C and 50 °C 

from the dissolution (undersaturation) experiment are presented in Table 32. The 

139


concentration of aluminum and silicon increases at both 20 °C and 50 °C, while the 

calcium concentration and pH only slightly change as a function of Al/(Al + Fe) ratio. 

Table 32 Measured ion concentration of Ca3(AlxFe1-x)2(SiO4)(OH)8 equilibrated for four months 

from dissolution (undersaturation) experiment. 

Al/(Al+Fe) Temprature Al Ca K Si Fe 

+ 

pH log Ks0 log Ks0 log Ks0 

ratios [°C] [mmol/l] [mmol/l] [mmol/l] [mmol/l] [mmol/l] C3AS0.84H4.32 C3FS0.95H4.1 C3(A,F)S0.9H4.2 

0.0 20


The total solubility product (ΣΠ) can be calculated using the equation below. The 

standard solid solution modeling in GEMS is setup for one ion substituting another ion in 

one crystallographic site per end member. To fulfill this condition the logK values were 

downscaled by a factor 2 

ΣΠ= 2× ({Ca 2+ } 1.5 {AlO2 - + FeO2 - } 1 {HSiO3 - } 0.45 {OH - } 1.55 · {H2O} 1.1 ) 

Where {} denotes the activity. On the basis of the measured ion concentrations and target 

compositions of the solid phases, total solubility products were calculated as given Table 

32. The total solubility products change as a function of Al/(Al + Fe) ratio. In the 

presence of Fe, Al-siliceous hydrogarnet has a lower solubility product at both 20 °C and 

50 °C. 

A Lippmann phase diagram was drawn for the Ca3(AlxFe1-x)2(SiO4)0.9(OH)8.4 series 

assuming an ideal solid solution. The solidus and solutus curve describe the equilibrium 

state of the solid solution series of Al/Fe-siliceous hydrogarnet. Based on the limited 

amount of experimental data an ideal solid solution between Al and Fe-Si-hydrogarnet 

was fitted between C3AS0.84H4.32 and C3FS0.95H4.1. The solubility products of the end 

members are given in Table 32. 

The Lippmann phase diagram of the solidus curve as a function of the solid composition 

and the solutus curve as a function of the liquid compositions of an ideal solid solution 

series of Al/Fe-siliceous hydrogarnet was plotted at 20 °C and 50 °C as shown in Fig. 66a 

and b. 

If an ideal solid solution is assumed, the modeled total solubility products of solidus 

agrees reasonably with the experimentally determined total solubility products. The 

results indicate possible formation of an ideal solid solution between Al and Fe-siliceous 

hydrogarnet at 20 °C and 50 °C. However, the data might also be explained by the 

141


complete absence of any solid solution as indicated by the dotted lines and the empty 

symbols in Fig. 66. Due to the formation of two miscibility gaps and the simultaneous 

presence of C3AS0.41H5.18, C3AS0.84H4.32, C3FS0.95H4.1, C3FS1.52H2.96 and C-S-H in varying 

amounts in the experiment, more detailed investigations are needed to derive a consistent 

thermodynamic model for this system. 

142


Fig. 66 Lippmann diagram illustrating the total solubility products of Al/Fe-siliceous hydrogarnet 

solid solution series Ca3(AlxFe1-x)2(SiO4)0.9(OH)8.4 at a) 20 °C b) 50 °C: .experimentally 

determined total solubility products (filled symbols), modeled total solubility products 

assuming ideal solid solution (dashed lines). In addition also the solubility product of 

C3AS0.84H4.32 and C3FS0.95H4.1 derived from the experimental data (empty symbols) and 

the solubility products assuming no solid solution (dotted lines) are given. X-axis: Al/(Al 

+ Fe) ratio in the solid or liquid phases, respectively. 


C3AH6 is the stable phase in CaO-Al2O3-H2O system in the absence of other ions. The 

co-precipitation of C4AHx was consistent; this might be related to the high pH reaction 

during synthesis. In contrast, C3FH6 was found to be metastable. This suggests that in the 

143


absence of silica, portlandite and amorphous Fe-hydroxide and possibly iron containing 

carbonate phases will form. The formation of C4FHx was persistent in CaO-Fe2O3-H2O 

system at room temperature. No solid solution formation occurred between C3AH6 and 

C3FH6. 

The Al and Fe siliceous hydrogarnet was more complex due to the formation of two 

hydrogarnets with different silica contents. In addition, C-(A)-S-H co-precipitated during 

the preparation of Si-hydrogarnet. Al-containing siliceous hydrogarnet did not form at 

room temperature but only at 110 °C. A miscibility gap was observed between 

C3AS0.41H5.18 and C3AS0.84H4.32 for samples synthesized at 110 °C. Hydroandradite formed 

both at room temperature and at 110 °C. Again the formation of a solid solution was 

observed with a miscibility gap between C3FS0.95H4.1 and C3FS1.5H2.96. 

Possibly a simultaneous substitution of Al by Fe in the octahedral sites and Si by OH - in 

the tetrahedral sites occurred with a miscibility gaps for Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y. 

The CaO-Al2O3-Fe2O3-SiO2-H2O is a complex system that exhibits multi ion substitution 

in the structure and more detailed investigations are needed. 

Implications for cementitious systems 

In pure CaO-Al2O3-H2O systems C3AH6 is the stable phase. The addition of sulfate 

destabilizes C3AH6 and leads to the formation of monosulfate or ettringite depending on 

the amount of sulfate added [7]. The presence of carbonate leads to the formation of 

monocarbonate [2, 67]. Similarly, the presence of silica could destabilize C3AH6 and 

cause the formation of the thermodynamically more stable Si-hydrogarnet. However, this 

process is kinetically hindered and observed only at higher temperature around 100 °C. 

In the presence of Fe2O3 the phase assemblage of CaO-(Al,Fe)2O3-H2O is distorted to 

144


CH, C3AH6 and Fe-hydroxide while C3FH6 is metastable. The presence of carbonate, 

sulfate or silica in a CaO-Fe2O3-H2O system leads to the formation of Fe-monocarbonate 

[19], Fe-monosulfate/Fe-ettringite or hydroandradite [15, 129, 130]. 

In a CaO-Al2O3-SiO2-H2O system the presence of Fe2O3 leads to the formation of stable 

hydrogarnets with the composition Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y. At room temperature 

Al/Fe containing siliceous hydrogarnet have been observed in hydrated cement pastes 

[22, 35, 65]. The kinetics of formation of mixed Al/Fe containing siliceous hydrogarnet is 

slow at room temperature. The presence of silica in OPC tends to stabilize a poorly 

crystalline mixed Al/Fe containing siliceous hydrogarnet. High temperature curing (≥ 95 

°C) accelerates the development of an intermixed hydrogarnet in CaO-Al2O3-Fe2O3-SiO2- 

H2O system. However, it was observed that Si-hydrogarnet formed as a main phase at 

higher temperature during cement reaction or in autoclaved concrete [23, 131]. In general 

an intermixed hydrogarnet Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y could potentially form in 

cement pastes due to the stability of the phase. The chemistry of hydrogarnet can also be 

influenced by the presence of magnesium. This might be due to the possible substitution 

of Ca by Mg in the hydrogarnet structure [120]. 

3.6. Summary 

In this thesis a number of new solubility data for Fe-containing hydrates and their 

structure have been determined. Most of the solubility data have been determined for the 

first time. The structure of Fe-monocarbonate, Fe-monosulfate and hydroandradite series 

have been determined. The results obtained in this chapter are summarized in Table 33. 

145


Table 33 Summary of the results obtained in chapter 3 and comparison with their Al-analogues. 

Phases Thermodynamic Crystal structure structure of Solid solution with Al- 

stability at 20 °C 

Al-analogue analogues 

Fe-hemicarbonate 

unstable in the presence n.d n.d. n.d 

(C4Fc0.5H10) 

of carbonate 

Fe-monocarbonate 

(C4FcH12) 

stable rhombohedral monoclinic none 

Fe-monosulfate 

stable rhombohedral rhombohedral solid solution with 

(C4FsH12) 

miscibility gap 

Fe-Friedel's salt 

stable rhombohedral < 35 °C: solid solution tentatively 

(C4FCl2H10) 

monoclinic 

> 35 °C 

rhombohedral 

Fe-strätlingite 

unstable with respect to n.d trigonal n.d 

(C2FSH8) 

Fe-hydroxide, 

portlandite and C-S-H 

Fe-OH-AFm 

unstable in the presence n.d n.d n.d 

(C4FH13) 

of carbonate 

Fe-katoite 

unstable with respect to n.d cubic no solid solution 

(C3FH6) 

Fe-hydroxide, 

portlandite and possibly 

Fe-hemi/monocarbonate 

in the presence of 

carbonate 

Fe-siliceous hydrogarnet 

(C3FS0.95H4.1/ C3FS1.52H2.96) 

stable cubic cubic ideal solid solution 

n.d: not determined, strätlingite [132] 

146

CHAPTER 4 FE-CONTAINING HYDRATES IN HYDRATED CEMENT 

4. FE-CONTAINING HYDRATES IN HYDRATED CEMENT 

4.1. Identification of Fe-containing hydrates in hydrated cement 


Whether and to what extent the Fe-containing phases and their solid solutions could form 

in OPC is poorly understood. Any approach to elucidate the fate of Fe in cementitious 

materials is complicated by the fact that identification of the Fe-containing hydrates in 

hydrated cement using standard techniques (XRD, TGA, SEM) is difficult as their signals 

significantly overlap with those of the corresponding Al-containing phases. Furthermore, 

also the formation of amorphous Fe-containing phases in hydrated cement is difficult to 

detect using standard techniques. 

Synchrotron-based X-ray absorption spectroscopy (XAS) can be used as a 

complementary technique as it is able to provide molecular-level information of 

cementitious systems [28-30]. Most frequently used XAS techniques are X-ray 

absorption near edge structure (XANES) and extended X-ray absorption fine structure 

(EXAFS) spectroscopy. The former technique is mainly used to study the oxidation state 

of the absorber atom and for fingerprinting on the basis of a comparison of reference 

spectra with unknown spectra of the absorber atom, while the latter technique enables us 

to determine the coordination sphere (i.e., type of neighboring atoms, bond length and 

coordination numbers) of the X-ray absorber atom of interest. Both techniques allow 

dilute samples to be examined (concentration of the X-ray absorber down to a few tens of 

ppm). Most importantly, XAS can be used to study amorphous solids, surface adsorbed 

complexes, or species in solution in addition to crystalline materials. Furthermore, 

advanced high resolution synchrotron-based X-ray micro-probe allows spatially resolved 

147


information on the speciation of the X-ray absorber of interest in compact matrices, such 

as cementitious materials, to be obtained [28, 30]. 

Synchrotron XAS has previously been used to determine the oxidation state and 

coordination environment of Fe in minerals and natural sediments [133, 134]. These 

studies showed that distinct features in the XANES spectra are useful for qualitatively 

distinguishing among major mineral classes. Detection of a particular mineral within a 

structural class was found to be more difficult and depend on the spectral uniqueness of 

these minerals. It was suggested that EXAFS may be more sensitive to the detection of 

particular components in a mixture at low Fe concentrations than the application of 

XANES. To the best of our knowledge there is only one study, which used XAS to 

determine the Fe speciation in cementitious systems [77]. The authors investigated the 

Fe-containing hydrates forming during the hydration of C2(A,F). The use of XAS for 

speciation studies on Fe in hydrated cement, however, is novel. 

In this study the hydration process of two different cements OPC and HS have been 

investigated with the aim of identifying the new Fe-containing phases formed. Different 

standard analytical techniques (XRD, TGA, SEM) were used to identify the main phases. 

The above techniques were complemented by selective dissolution using the salicylic 

acid/methanol (SAM) method to identify the minor Fe-containing phases. EXAFS was 

further applied to identify in situ the Fe-containing phases in the complex cement matrix. 

The study was completed by predicting the hydration assemblage using thermodynamic 

modeling and an updated database including new solubility data for the Fe-containing 

phases. 

148


4.1.2. Characterization of hydrated cement using standard analytical 

techniques 

The hydration of two different cements, OPC and HS, was studied to determine the Fe 

speciation at various stages of the hydration process. Both cements contained roughly 5 

wt.% Al2O3 while their Fe2O3 content differed widely. The cement OPC had a Fe2O3 

content of 3 wt.%, while the HS contained about 7 wt.%. The progress of hydration and 

appearance/disappearance of the main cement phases can easily be followed using 

standard analytical techniques like XRD, TGA and SEM. During the hydration the peaks 

of the reactive clinkers phases C3S and C3A diminished fast, while C2(A,F) and C2S 

peaks persisted over longer periods of time (Fig. 67). 

Fig. 67 XRD patterns of OPC (+) and HS (*) cements hydrated at 20 °C. 

149


Gypsum and anhydrite were consumed within the first day. The main hydrates formed 

during hydration were C-S-H and portlandite. In addition, the formation of AFt and AFm 

(monosulfate in the case of OPC and monocarbonate for the calcite containing HS) was 

observed. The formation of ettringite and AFm phases was observed within 1 day (Fig. 

67 and Fig. 68). The weight loss in the TGA-DTG curve between 100°C and 240 °C 

indicates the presence of ettringite, C-S-H and AFm phases (Fig. 68). 



100 

90 

80 

70 

-0.1 

-0.2 

gypsum 

Ettringite 

AFm phases 

C-S-H 

portlandite 

100 200 300 400 500 600 700 800 


carbonate 

Fig. 68 TGA-DTG curves of OPC (+) and HS (*) cements hydrated at 20 °C. 

unhyd+ 

1 day+ 

28 days+ 

150 days+ 

3 years+ 

3 years* 

Temperature had a significant influence on the hydration process (Fig. 69 and Fig. 70). 

While C-S-H and portlandite were still the main phases formed in both OPC and HS after 

hydration for 1 year at 50 °C ettringite was not detected, but AFm phase was detected in 

both the OPC and HS at 2θ ~ 11.30° (Fig. 69). The AFm phase peak was in between the 

150


peak of monosulfate and hemicarbonate. Ettringite and monocarbonate are expected to be 

thermodynamically less stable than monosulfate in cements at 50 °C [3, 32, 135]. 

However, in addition to monosulfate, the presence of either ettringite or monocarbonate 

was reported at 50 °C, depending on the Al2O3/SO3 ratio of the cement and on the 

presence of calcite [3, 135]. Fig. 69 shows that AFm and calcite were present in the 

samples hydrated for 1 year at 50 °C. In addition, peaks at 2θ ~ 17.47°, 2θ ~ 20.21° and 

2θ ~ 29.90° were observed, which can be assigned to siliceous hydrogarnet. Moreover, 

the broad DTG curves between 120 and 270 °C in Fig. 70 is consistent with the presence 

of C-S-H, AFm and Si-hydrogarnet phases. 

Fig. 69 XRD patterns of OPC (+) and HS (*) cement hydrated for 1 year at 50 °C: The XRD peak 

at 2θ ~ 11.30 is between the monosulfate and monocarbonate peaks. 

151




100 

90 

80 

70 

-0.1 

-0.2 

C-S-H 

AFm 

Si-Hg 

portlandite 

1 year (+) 

1 year (*) 

carbonate 

100 200 300 400 500 600 700 800 


Fig. 70 TGA-DTG curves of OPC (+) and HS (*) cement hydrated for 1 year at 50 °C 

However, minor phases and poorly crystalline phases are difficult to identify 

unambiguously by techniques like XRD, TGA and SEM, due to the overlap of peaks with 

the main phases. Therefore, selective dissolution using SAM (salicylic acid/methanol) 

was applied to dissolve the main phases with the aim of identifying clearly the minor 

phases formed in the course of the hydration process. The SAM extraction used in this 

study dissolves the silicate containing clinkers (C3S, C2S), C-S-H, portlandite, AFm and 

AFt phases. Fig. 71 and Fig. 72 show the XRD pattern and TGA-DTG curves of the 

residue of cement pastes hydrated between 28 days and 3 years at 20 °C after treatment 

with SAM. The presence of siliceous hydrogarnet at 2θ ~ 17.47° (corroborated by the 

additional peaks at 2θ ~ 20.21° and 2θ ~ 26.90°) is indicated in all SAM treated OPC 

samples irrespective of the hydration time. Note, however, that the intensity of siliceous 

hydrogarnet in the SAM treated HS cement paste is much less than in the OPC samples. 

152


This suggests presence of poorly crystalline siliceous hydrogarnet in the HS samples. 

Presence of siliceous hydrogarnet is further supported by TGA-DTG, which shows the 

weight loss attributable to siliceous hydrogarnet at around 240 °C. The TGA-DTG curves 

also show the weight loss at 400 °C, which is assigned to the presence of hydrotalcite 

[136] in the hydrated cement samples (Fig. 72). SAM extraction was further done for the 

OPC and HS samples hydrated for 150 days at 50 °C. The residue contains ferrite and 

siliceous hydrogarnet (Fig. 73). 

Fig. 71 XRD patterns of OPC (+) and HS (*) hydrated at 20 °C after selective dissolution using 

SAM. Note that the samples suffered from carbonation during SAM extraction. 

153




100 

90 

80 

70 

-0.1 

-0.2 

C-S-H 

Si-Hg 

hydrotalcite 

100 200 300 400 500 600 700 800 


28 days+ 

150 days+ 

3 years+ 

3 years * 

carbonate 

Fig. 72 TGA-DTG curves of OPC (+) and HS (*) hydrated at 20 °C after selective dissolution 

with SAM. 

Fig. 73 XRD patterns of OPC (+) and HS (*) hydrated at 50 °C for 150 days after selective 

dissolution with SAM. 

154


SEM/EDX studies were carried out to identify minor phases in the SAM treated OPC 

pastes and to determine the composition of siliceous hydrogarnet. Fig. 74a and b show 

the SEM images of SAM treated OPC samples hydrated for 2 years at 20 °C or 50 °C, 

respectively. Unreacted C2(A,F) (bright areas) was found to be covered partially by less 

dense siliceous hydrogarnet (light grey level). EDX was used to verify the presence of the 

latter phase. Fig. 75 shows the atomic ratios of (Al+Fe)/Ca against Si/Ca determined on 

the spots indicated in the SEM images. The composition of siliceous hydrogarnet was 

found to be Ca3Al0.9Fe1.0Si0.85O6 and Ca3Al1.1Fe0.95Si1.0O6.7 at 20 °C and 50 °C, 

respectively. The composition may vary due to the substitution of Al by Fe in 

hydrogarnet [129]. An important finding from this SEM/EDX study is that Fe tends to 

accumulate in the siliceous hydrogarnet together with Al, possibly indicating a solid 

solution. However, the presence of other Fe-containing phases (which might have 

dissolved during SAM extraction) in the hydrated cement cannot be excluded. 

155

a). 

b). 


Fig. 74 SEM/EDX of hydrated OPC for 2 years at (a) 20 °C and (b) 50 °C after selective 


156

(Al+Fe)/Ca 

(Al+Fe)/Ca 

0.8 

0.6 

0.4 

0.2 

b. 

a. 

0.8 

0.6 

0.4 

0.2 

0 

0 


Ca 3Al 0.9Fe 1.0Si 0.85O 6 

0 0.2 0.4 0.6 

Si/Ca 

Ca 3Al 1.1Fe 0.95Si 1.0O 6.7 

0 0.2 0.4 0.6 

Si/Ca 

Fig. 75 Atomic ratio of hydrated OPC for 2 years at (a) 20 °C and (b) 50 °C after selective 


157


4.1.3. Spectroscopic investigation 

Experimental studies indicated that a number of Fe-containing phases like Fe(OH)3, 

C4FcH12, C4FsH12, C6Fs3H32, C3(A,F)SH4, C3FSH4 and Fe-hydrotalcite are stable at the 

high pH values of OPC pore solutions [15, 19, 129, 130, 137]. As shown above, it is 

difficult to identify the Fe-containing hydrates in cement pastes using common 

techniques like XRD, TGA and SEM due to overlap of the siginals from Fe- and Al- 

containing phases. Thus EXAFS, which allows element specific speciation studies, was 

used in addition with the aim of identifying the Fe-containing phases in the complex 

cement matrices. EXAFS spectra were also collected for synthetic Fe-bearing hydrates in 

order to use them as reference spectra for the analysis of the composed spectra obtained 

from hydrated OPC and HS. 

4.1.3.1. XANES and EXAFS spectra of Fe-containing reference 

compounds 

Fig. 76 and Fig. 77 show the Fe K-edge XANES and EXAFS spectra of Fe-containing 

hydrates used as reference compounds for identifying these phases in the cement paste. 

The XANES spectra show slight shifts in the energy of the white lines (absorption edges) 

of the reference compounds (Fig. 76). Furthermore, the peaks have different intensities. 

This is related to differences in structural symmetry and the kind of neighboring atoms in 

Fe-containing hydrates. The differences between the spectra of Fe-containing hydrates 

are more pronounced in the EXAFS region (Fig. 77). This difference is attributed to 

variations in the backscattering contribustions from the neighboring atoms. C2F and 

C4AF clinkers have the same spectral feature despite of small differences in the 

158


intensities of the spectra, thus indicating similar symmetry. Similarly, no siginificant 

difference were observed between C3FSH4 and C3(A,F)SH4. 

In this study, Fe-containing phases were identifyed in hydrated cement based on EXAFS 

spectra due to the capability of the technique to identify poorly crystalline phases and 

those with low Fe concentrations. The EXAFS spectra of possible stable Fe-containing 

phases like the clinkers (C4AF and C2F), Fe(OH)3, C4FcH12, C6Fs3H32, C3(A,F)SH4, 

C3FSH4 and Fe-hydrotalcite were used as a reference compounds. Other possible Fe- 

containing phases like Al/Fe-solid solution AFm, AFt and hydrogarnet were not 

considered as the feature of the spectra are nearly identical with those of the pure Fe- 

containing AFm, AFt and hydrogarnet phases. Also C4FsH12 was not used as reference as 

the features are nearly identical with those of C4FcH12. 

Fig. 76 Fe K-edge XANES spectra of Fe-containing hydrates. The broken lines indicate the 

position of related spectral features. 

159


Fig. 77 k 3 -weigthed experimental bulk-EXAFS spectra of Fe-containing phases used as reference 

compounds. The broken lines indicate the position of related spectral features. 

4.1.3.2. Identification of Fe-containing hydrates 

The EXAFS spectra of hydrated OPC and HS were collected at the Fe K-edge (7112 eV). 

Fig. 78 and Fig. 79 show the EXAFS spectra of hydrated OPC after different hydration 

times aged at 20 and 50 °C. The features of the spectra change as the hydration process 

progresses with time thus indicating the formation of different Fe-containing phase 

during hydration. The unhydrated cement spectra only contain the ferrite clinker 

(C2(A,F)). The spectra of OPC hydrated for 4 and 8 hours are similar which indicates that 

the same Fe-containing phases present in similar amounts contributed to the composed 

spectra. A small difference was observed between the unhydrated cement spectra and 

hydrated spectra (4 and 8 hours) at k ~ 6-7 Å. However, for OPC hydrated at 50° C, the 

difference is more significant at 16 hours hydration. Note that in all cases the spectral 

160


features are nearly similar between 1 day and 1 year. The changes in the spectral features 

reveal the formation of new Fe-containing hydrates which contributed the composed 

spectra. This shows that the coordination environment of iron changes during the 

hydration process. 

k 3 (k) 

100 

80 

60 

40 

20 

0 

2 4 6 8 10 

k( Å -1 

) 

1 years 

150 days 

28 days 

1 day 

16 hrs 

8 hrs 

4 hrs 

Unhy 

Fig. 78 EXAFS spectra of hydrated OPC at 20 °C and at different ages (line: experimental data; 

dots: modeled data). The broken lines outline selected spectral features. 

161

k 3 (k) 


100 

80 

60 

40 

20 

0 

2 4 6 8 10 

k( Å-1 ) 

1 year 

150 days 

28 days 

1 day 

16 hrs 

8 hrs 

4 hrs 

Unhy 

Fig. 79 EXAFS spectra of hydrated OPC at 50 °C and at different ages (line: experimental data; 


The EXAFS spectra of hydrated HS cement at 20 °C and 50 °C are shown in Fig. 80 and 

Fig. 81. Pronounced changes are visible in the spectra between 16 hours and 1 day. 

Between 1 day and 1 year, however, the spectral features are very similar. This finding 

reveals major changes in the type and amount of Fe-containing phases during the first day 

of hydration while changes are less pronounced after 1 day hydration. The latter finding 

suggests that the same phases may be present between 1 day and 1 year hydration while 

the ratio of the phases may change with time. 

162

k 3 (k) 


100 

80 

60 

40 

20 

0 

2 4 6 8 10 

k( Å -1 

) 

1 years 

150 days 

28 days 

1 day 

16 hrs 

4 hrs 

Unhy 

Fig. 80 EXAFS spectra of hydrated HS at 20 °C and at different ages (line: experimental data; 

k 3 (k) 


80 

60 

40 

20 

0 

2 4 6 8 10 

k( Å -1 

) 

1 year 

150 day 

16 hrs 

Fig. 81 EXAFS spectra of hydrated HS at 50 °C and at different ages (line: experimental data; 

4 hrs 

1 hrs 

Unhy 


163


The Fe K-edge EXAFS spectra of the cement paste is considered to be composed of 

spectra of different Fe-containing phases. Linear combination (LC) fitting in combination 

with principal component analysis (PCA) and target transformation (TT) was carried out 

to identify the Fe-containing hydrates in the hardened cement pastes (see chapter 2.2.8). 

PCA was applied to determine the number of components contained in the Fe K-edge 

EXAFS spectra of the pastes. PCA predicted that three components are required to 

reproduce the Fe K-edge EXAFS spectra of the pastes. Two components were considered 

to be real while the third component was present at small amounts, suggesting that its 

contribution to the spectra was at noise level. TT was implemented to test which Fe- 

reference compound contributes to the complex hydrated cement spectra. The above 

listed references can have a potential to reconstruct the complex spectra as their SPOIL 

factor was < 5 [27]. 

164


Table 34 Relative weights of Fe-containing phases in hydrated OPC at 20 °C and 50 °C obtained 

from LC fitting. 

At 20 °C 

Age C2(A,F) Fe(OH)3 C3FSH4 C3(A,F)SH4 Fe-Ht C4FcH12 C6Fs3H32 R-factor 

Unhyd 1.00* - - - - - - 0.15 

4 hrs 0.63(3) 0.37(3) - - - - - 0.12 

8 hrs 0.70(4) 0.30(4) - - - - - 0.12 

16 hrs 0.36(3) 0.64(3) - - - - - 0.10 

1 day 0.45(2) - - 0.55(2) - - - 0.08 

28 days 0.42(2) - - 0.58(2) - - - 0.07 

150 days 0.40(1) - - 0.60(1) - - - 0.05 

1 year 

At 50 °C 

0.39(1) - - 0.61(1) - - - 0.04 


Unhyd 1.00* - - - - - - 0.15 

4 hrs 0.63(3) 0.37(3) - - - - - 0.12 

8 hrs 0.40(3) 0.60(3) - - - - - 0.07 

16 hrs 0.48(1) - - 0.52(1) - - - 0.05 

1 day 0.44(1) - - 0.56(1) - - - 0.04 

28 days 0.43(1) - - 0.57(1) - - - 0.05 

150 days 0.41(2) - - 0.59(2) - - - 0.05 

1 year 0.42(2) - - 0.58(2) - - - 0.12 

R is an indicator for the goodness of the fit. *The fitting showed 0.78(5) C2(A,F) and 0.22(4) of C2F for unhydrated cement. 

-Indicates that the phase could be present but below the detection limit of the method (


Table 35 Relative weights of Fe-containing phases in HS hydrated at 20 °C and 50 °C obtained 

from LC fitting. 

At 20 °C 

Age C2(A,F) Fe(OH)3 C3FSH4 C3(A,F)SH4 Fe-Htl C4FcH12 C6Fs3H32 R-factor 

Unhyd 1.00* - - - - - - 0.12 

4 hrs 0.78(3) 0.22(3) - - - - - 0.08 

16 hrs 0.56(3) 0.44(3) - - - - - 0.09 

1 day 0.62(2) - - 0.38(2) - - - 0.18 

28 days 0.57(2) - - 0.43(2) - - - 0.09 

150 days 0.56(2) - - 0.44(2) - - - 0.09 

1 year 

At 50 °C 

0.56(2) - - 0.44(2) - - - 0.07 


Unhyd 1.00* 0.12 

1 hrs 0.70(4) 0.30(4) 0.14 

4 hrs 0.65(2) 0.35(2) 0.09 

16 hrs 0.64(2) 0.36(2) 0.18 

150 days 0.59(1) 0.41(1) 0.09 

1 year 0.58(2) 0.42(2) 0.07 

R is an indicator for the goodness of the fit. *The fitting showed 0.49(4) C2(A,F) and 0.51(4) of C2F for unhydrated cement. 

-Indicates that the phase could be present but below the detection limit of the method (


hydration time as the rate of reaction increases with increasing temperature. At longer age 

Fe existed in C3(A,F)SH4. The trend was the same for hydrated HS cement showing that 

Fe tended to form Fe(OH)3 at early ages and siliceous hydrogarnet at later stages of 

hydration (Table 35). 

The fitting was also done by assuming three components to represent the unknown 

spectra. Consistent with the two component fitting, the results showed that at early age Fe 

tends to be associated with C2(A,F) and Fe(OH)3 and at longer age with C2(A,F) and 

siliceous hydrogarnet: C3(A,F)SH4 or C3FSH4. Traces of Fe(OH)3 or Fe-hydrotalcite was 

observed as a third components at longer hydration ages. The weight of Fe(OH)3 or Fe- 

hydrotalcite was less than 10% which is considered to be the uncertainty range of the 

method. 

Ferrite clinker (C2(A,F)) reacts rapidly in the early stage of hydration and persists at later 

age as demonstrated by XRD/Rietveld analysis [8, 9, 65]. In agreement with these 

findings EXAFS spectroscopy showed that the contribution of C2(A,F) spectra to the 

overall spectra decreased at early age and was almost constant after 1 day hydration. 

4.1.4. Thermodynamic modeling 

Based on the measured composition of the cement (Table 1), the phases formed during 

the OPC hydration were modeled using GEMS [6, 65]. For the modeling the solubility of 

Fe-containing hydrates as determined in this study [19, 129, 130] was combined with the 

thermodynamic database for cement hydrates Cemdata07 [3, 7, 15]. The formation of 

hydrogrossular was excluded as it does not form at ambient temperature. The modeling 

indicates the formation of C-S-H, portlandite and the Al-containing hydrates: Al- 

ettringite, Al-monosulfate, Al-hemicarbonate and Al-hydrotalcite (Fig. 82). Iron released 

167


due to the hydration of the clinkers was calculated to be present during the first hours as 

Fe-ettringite and after the consumption of gypsum and anhydrite as C3F0.95SH4.1, which is 

calculated to be the only Fe-hydrate stable at longer hydration times. The long term 

prediction is consistent with the EXAFS results (Table 34). The experimentally observed 

formation of Fe(OH)3 (instead of the predicted Fe-ettringite) at early hydration time is 

probably related to the slow formation of Fe-ettringite. Fe(OH)3 formed as an 

intermediate phase in the early stage of hydration process was less stable than 

C3F0.95SH4.1 and therefore destabilized at later ages. 

Note that, in calculations, where the formation of aluminium containing siliceous 

hydrogarnet was not prevented, the formation of mixed Al- and Fe-containing siliceous 

hydrogarnet and Al-ettringite was calculated but no AFm phases, which does not agree 

with the experimental observation. Whether and to what extent Al can incorporated at 

room temperature in Fe-siliceous hydrogarnet is still unclear and should be the subject of 

future studies. 

cm 3 /100g 

90 

80 

70 

60 

50 

40 

30 

20 

10 

C (A,F) 2 

gypsum 

C A 3 

C S 2 

C 3 S 

pore solution 

Al-monocarbonate 

Al-ettringite 

Siliceous hydrogarnet 

Portlandite 

C-S-H 

0 

1E-3 0.01 0.1 1 

time [days] 

10 100 1000 

hydrotalcite 

Al-hemicarbonate 

Al-monosulfate 

Fig. 82 Volume changes of hydrated phases at different hydration ages during hydration of OPC 

at room temperature. 

168



The presence of siliceous hydrogarnet might occur in hydrated cement particularly at 

high temperature (Fig. 69) as early indicated from earlier studies [3, 23, 37]. A limited 

amount of siliceous hydrogarnet could have formed also at room temperature as the first 

XRD peak of siliceous hydrogarnet is located near the CH peak at 2θ ~ 17.47° which 

makes it difficult to identify this peak in the XRD pattern of hydrated cement (Fig. 67). 

Selective dissolution using SAM allows CH and other silicate phases to be dissolved 

which gives rise to clear XRD peaks of siliceous hydrogarnet as shown in Fig. 71. Its 

composition was determined to be Ca3Al0.9Fe1.0Si0.85O6 and Ca3Al1.1Fe0.95Si1.0O6 for the 

OPC hydrated at 20 and 50 °C. These siliceous hydrogarnets contain both Al and Fe 

which indicates substitution of Al by Fe in the structure of hydrogarnet. The finding 

agrees with the result of Taylor et al. [35] who suggested the formation of hydrogarnet in 

hydrated cement with a composition Ca3Al1.2Fe0.8 SiO1.2H8 using SEM microanalysis. 

EXAFS spectroscopy shows that in all cases Fe was associated with C2(A,F) and 

Fe(OH)3 gel at early age hydration. With time, however, Fe(OH)3 dissolved and 

precipitated as C3(A,F)SH4, which supported the findings by selective dissolution with 

SAM for hydrated OPC. In hydrated HS, however, no siliceous hydrogarnet could be 

detected by XRD possibly due its poor crystallinity, while EXAFS spectroscopy shows 

the presence of C3(A,F)SH4. 

The experimentally determined solubility products indicate that Fe-ettringite is less stable 

than Al-ettringite [16] and therefore this phase is hardly present in hydrated cement 

pastes, except at early ages in the presence of gypsum. The Fe-AFm phases are generally 

more stable than the Al-AFm phases [19, 130]. However, the very low Fe concentrations 

(


phases and therefore Fe is preferentially bound in siliceous hydrogarnet. Fe(OH)3 formed 

as an intermediate phase in the early stage of the hydration process while it convertes to 

C3(A,F)SH4 with time. EXAFS spectroscopy indicates that small amounts of Fe could be 

bound in hydrotalcite or remaining Fe(OH)3 at longer hydration times. Thermodynamic 

modeling indicated that C3F0.95SH4.1 is the stable Fe-containing phase in hydrated OPC. 

In conclusion, EXAFS can be used to identify minor, amorphous or poorly crystalline 

phases in a complex cementitious system. Furthermore, identification of phases and 

thermodynamic modeling of the hydration process are important with a view to assessing 

of the durability of cementitious materials and assessments made in connection with the 

safe disposal of radioactive and hazardous wastes. 

170


4.2. Synthetic Fe-cement 


As discussed in section 4.1, the identification of Fe-bearing hydrates in the complex 

matrix of hydrated cement is difficult with techniques like XRD, TGA and SEM/EDS as 

their signals overlap to a large extent with those of the Al-bearing phases present and also 

partially with those of other solids. In section 4.1, EXAFS was used as an alternative 

method to determine the fate of Fe-hydrates in Portland cements and mixed Al-Fe- 

siliceous hydrogarnet was identified as the main Fe-containing hydrate after more than 1 

day of hydration, while no Fe-containing ettringite of monocarbonate was observed. In 

contrast to the observation in the cement system, not only mixed Al-Fe-Si-hydrogarnet, 

but also Fe-ettringite, Fe-monocarbonate or Fe-monosulfate are easily synthesized in 

water or KOH solutions [11-17, 19]. 

Thus, Al-free synthetic cements were prepared to investigate the hydration assemblage in 

the absence of Al in order to understand the chemistry of Portland cement without Al and 

the role of Fe in cement. 

The synthetic cements contained ferrite (C2F), alite, additional alkalis and varying 

amounts of gypsum and calcite to mimic the composition of Portland cement as close as 

possible. Isothermal conduction calorimeter, XRD and TGA were used to characterize 

the hydration of the synthetic cements. Thermodynamic modeling was also performed to 

predict the hydration assemblage of synthetic Fe-cement. 

171

Early reaction 


4.2.2. Effects of gypsum on the of hydration of synthetic Fe-cement 

Fig. 83 shows that pure C2F cement (with alkali sulfates present) reacted fast. The main 

heat release was observed around 6 hours. Addition of gypsum to the C2F clinker 

retarded the reaction. The presence of gypsum seems to retard C2F such as it does C3A 

[138], possibly also due to blocking of the surface sites in the presence of high 

concentrations of dissolved sulfate. The calorimeter curve of synthetic Fe-cement 

hydrated with different amounts of gypsum (Gyp-0% - Gyp-26%) shows an acceleration 

of the main alite reaction in the presence of more gypsum. Sulfate is known to accelerate 

the alite reaction [139]. The peak of the ferrite reaction is not visible in the calorimetric 

data for the synthetic Fe-cement. 

heat flow / J/(g·h) 

35 

30 

25 

20 

15 

10 

5 

0 

0 8 16 24 32 40 48 

hydration time / h 

C2F-Pure 

C2F-Gyp 

Gyp-0% 

Gyp-6% 

Gyp-26% 

Fig. 83 Heat flow of the hydration of C2F and synthetic Fe-cement in the presence of different 

amounts of gypsum. 

Hydrates formed 

The XRD and TGA results of Fe-synthetic cement with different amounts of gypsum 

hydrated for 3 days are shown in Fig. 84. The hydration of pure ferrite resulted in iron 

172


containing OH-AFm phases, Fe-hydroxide and some hemi- and monocarbonate due to 

contamination with CO2. In the presence of gypsum, pure ferrite produced traces of Fe- 

ettringite. The hydration of synthetic Fe-cement produced C-S-H and portlandite as main 

hydrates after 3 days. 



100 

90 

80 

Fe-Ett 

-0.1 

-0.2 

-0.3 

CsH 2 

Fe-Mc* 

+ 

C-S-H 

C 2 F 

CH 

CsH 2 

10 15 20 25 30 

AFm 

CsH 2 

2CuK 

CH 

* 

C 2 F 

* 

CsH 2 

C 2 F 

C 2 F-Pure 

C 2 F-Gyp 

Gyp-0% 

Gyp-6% 

Gyp-26% 

CsH 2 

100 200 300 400 500 600 700 800 


carbonate 

* 

C 2 F 

C 2 F-Pure 

C2F-Gyp 

Gyp-0% 

Gyp-6% 

Gyp-26% 

Fig. 84 XRD (above) and TGA-DTG (below) analysis of C2F and synthetic Fe-cement after 3 

days of hydration in the presence of different amounts of gypsum *Fe-OH-AFm 

+unidentified. 

173


The phase assemblage of the hydrating synthetic Fe-cement changed with time as shown 

in Fig. 85 after 3 months. Again, Fe-containing OH-AFm phases were observed in the 

pure ferrite sample. Fe-containing monocarbonate was also clearly visible due to 

carbonation of the pure ferrite sample. The ferrite had reacted completely during this time 

and in addition to the AFm phases also significant amounts of amorphous Fe-hydroxide 

formed as it is visible in the water loss between 70 and 100 °C in TGA. In the presence of 

gypsum and ferrite more Fe-ettringite formed. However, even after 3 months, a 

significant part of ferrite had not reacted. The presence of gypsum seems to strongly 

retard the reaction of ferrite strongly. The hydration of synthetic Fe-cement without 

additional gypsum produced C-S-H and portlandite. The formation of Fe-siliceous 

hydrogarnet would be possible under these conditions; however no Fe-siliceous 

hydrogarnet was detected by XRD. This could be related to the formation of poorly 

crystalline Fe-siliceous hydrogarnet (see chapter 3.5.6). In the presence of gypsum, Fe- 

ettringite was produced as the main Fe-containing hydrate. The reaction of ferrite was not 

complete in any of the synthetic Fe-cements, neither in the absence nor the presence of 

gypsum. 

174



100 

90 

80 

70 

-0.1 

-0.2 

-0.3 


Fe-Ett 

Fe-Ett 

CsH 2 

Fe-Mc 

+ 

* 

C 2 F 

* 

Fe-Ett 

CH 

Fe-Mc 

CsH 2 

C F 2 

C F 2 C F Fe-Ett CH 

2 

CsH 2 

10 15 20 25 30 

C-S-H 

AFm 

CsH 2 

2CuK 

* 

CsH 2 

CH 

CsH 2 

* 

carbonate 

C 2 F-Pure 

C 2 F-Gyp 

Gyp-0% 

Gyp-6% 

Gyp-26% 

100 200 300 400 500 600 700 800 


C 2 F-Pure 

C2F-Gyp 

Gyp-0% 

Gyp-6% 

Gyp-26% 

Fig. 85 XRD (above) and TGA-DTG (below) analysis of C2F and synthetic Fe-cement after 3 

months of hydration in the presence of different amounts of gypsum *Fe-AFm hydroxyl. 

175


Thermodynamic calculations 

Thermodynamic modeling was performed to calculate the composition of hydrated 

synthetic Fe-cement in the presence of varying amounts of gypsum. The model predicted 

the formation of C-S-H, portlandite and Fe-siliceous hydrogarnet in the absence of 

gypsum (Fig. 86). It should be noted that in the presence of silica, no monosulfate but 

only ettringite formation was predicted if more sulfate is present in the system. Fe- 

siliceous hydrogarnet is calculated to be more stable than Fe-monosulfate under these 

conditions. As the amount of gypsum increased the formation of Fe-ettringite and finally 

the presence of unreacted gypsum was calculated. This is consistent with the XRD and 

TGA results (Fig. 85) though Fe-siliceous hydrogarnet was not observed. Fe-siliceous 

hydrogarnet was calculated to be unstable in the presence of gypsum and Fe-ettringite. 

cm 3 /100g 

80 

70 

60 

50 

40 

30 

20 

10 

Fe-Si-hydrogarnet 

portlandite 

Fe-ettringite 

C-S-H 

0 

0 10 20 30 

Gypsum % 

40 50 60 

Gypsum 

Fig. 86 Calculated phase diagram of thermodynamic stable hydrate assemblages of synthetic Fe- 

cement with different amounts of gypsum. 

176

Early reaction 


4.2.3. Effects of calcite on the of hydration of synthetic Fe-cement 

The effects of the presence of both calcite and gypsum during the hydration of ferrite and 

of the synthetic Fe-cement were studied. In the presence of calcite, the rate of the ferrite 

reaction did not change significantly from the previous experiment. In the presence of 

gypsum, however, the reaction of the ferrite was strongly retarded while the reaction of 

alite was accelerated, as observed before for the calcite free samples (Fig. 87). 

Fig. 87 Conduction calorimeter curve of the hydration of synthetic Fe-cement with different 

amounts of gypsum and calcite. 

Hydrates formed 

The presence of calcite changed the phase assemblage for the pure ferrite but not for the 

synthetic cements (Fig. 88). The formation of Fe-monocarbonate was clearly visible 

during the hydration of pure ferrite in the presence of calcite. The presence of gypsum 

strongly retarded the ferrite reaction; only a small amount of Fe-ettringite was detected 

177


together with unreacted gypsum and calcite, while no formation of Fe-monocarbonate 

was observed. 

The hydrates formed in the synthetic Fe-cement with calcite but without additional 

gypsum were C-S-H and portlandite and unreacted calcite, while no Fe-monocarbonate 

was identified. The reaction of ferrite was again strongly retarded in the presence of 

silica, as observed previously for the calcite free cements, while the reaction of alite was 

nearly complete after the 3 months of hydration. In the presence of gypsum and calcite, 

the formation of Fe-ettringite was clearly visible, although the ferrite reaction had been 

retarded. 

178



100 

90 

80 

70 

-0.1 

Fe-Ett 

C-S-H 

-0.2 

-0.3 


AFm 

CsH 2 

CH 

100 200 300 400 500 600 700 800 


C2F + Cc7 

C2F + Cc + Gyp 

7.8% Cc + 0% Gyp 

7.4% Cc + 6% Gyp 

6.2% Cc + 20% Gyp 

Fig. 88 XRD (above) and TGA-DTG (below) analysis of synthetic Fe-cement after 3 months of 

hydration with different amounts of gypsum and calcite 

Thermodynamic calculations 

The thermodynamic calculations for the synthetic Fe-cements predicted the formation of 

Fe-siliceous hydrogarnet instead of monocarbonate (Fig. 89). In the presence of more 

calcium sulfate, again Fe-ettringite was calculated to be stabilized as calculated also for 

the calcite free synthetic Fe-cements as discussed above. 

Cc 

179

cm 3 /100g 


70 

60 

50 

40 

30 

20 

10 

Fe-Si-hydrogarnet 

portlandite 

Fe-ettringite 

C-S-H 

calcite 

0 

0 10 20 30 40 

Gypsum % 

50 60 

Fig. 89 Calculated phase diagram of thermodynamic stable hydrate assemblages of Fe-synthetic 

cement with different amounts of gypsum and calcite. 


The experimental study show that pure ferrite phase (C2F) reacted fast and formed Fe- 

containing OH-AFm phases. If the samples carbonate the formation of monocarbonate 

was observed. The presence of gypsum strongly retarded the ferrite reaction and led to 

the formation of Fe-ettringite. 

The hydration of synthetic Fe-cement resulted in the formation of C-S-H, portlandite and 

possibly Fe-siliceous hydrogarnet. In the presence of gypsum Fe-ettringite was also 

formed, both in the absence and presence of calcite. Thermodynamic modeling was in 

agreement with the experimental observations and predicted the formation of C-S-H, 

portlandite and Fe-siliceous hydrogarnet at lower gypsum content. Upon the addition of 

Gypsum 

gypsum stable Fe-ettringite was predicted, again in the absence and presence of calcite. 

180


The observations made in these synthetic Fe-cements concerning on the fate of iron are 

comparable to the observations made in commercial Portland cements. In the OPC and 

HS cements investigated (see section 4.1) the main hydrates were C-S-H, portlandite, Al- 

AFm and AFt phases. The formation of Al/Fe-siliceous hydrogarnet was also observed 

but not Fe-containing ettringite. The presence of aluminium in the OPC and HS cements 

led to the formation of stable Al-ettringite and Al-containing AFm phases while all Fe 

released from the ferrite was present as Fe-siliceous hydrogarnet in the long term time. 

The experiments with the synthetic Fe-cements confirmed the validity of the 

thermodynamic calculations in the synthetic cements and confirmed that the presence of 

silica suppressed the formation of Fe-containing AFm phases. 

181

5. GENERAL CONCLUSION AND OUTLOOK 


5.1. General conclusion 

Thermodynamic modeling of the hydration process allows the compositions of the 

hydration assemblages to be predicted under different conditions and to be extrapolated 

to long time scales. However, an important limitation of thermodynamic modeling was 

the lack of knowledge on the Fe speciation in hydrating cement, as Portland cements 

contain generally around 4-7 % iron oxide. No experimentally determined solubility 

products have been available for most Fe-containing hydrates. It was unclear to what 

extent Fe substitutes Al in AFm and hydrogarnet phases. In addition, there was little 

experimental evidence which Fe-containing phases form in hydrated Portland cements. 

This thesis aimed to fill these gaps by: 

i) determining the solubility of different Fe-containing hydrates, studying their 

crystal structure and solid solution formation with Al-containing hydrates. 

ii) investigating the speciation of iron(III) in hydrated cements. 

Fe-containing hydrates 

Possible Fe-containing hydrates (Fe-hemicarbonate, Fe-monocarbonate, Fe-monosulfate, 

Fe-Friedel’s salt, Fe-strätlingite, Fe-hydrogarnet and Fe-siliceous hydrogarnet) and their 

solid solution with their Al-analogues were synthesized at different temperatures. The Fe- 

containing AFm phases (Fe-monocarbonate, Fe-monosulfate and Fe-Friedel’s salt) are 

thermodynamically stable at 20 °C and 50 °C. At 80 °C, the Fe-containing AFm hydrates 

are unstable with respect to Fe-hydroxide/hematite and portlandite. Fe-hemicarbonate and 

Fe-AFm hydroxyl (C4FH13) are unstable in the presence of carbonate and converted to 

Fe-monocarbonate with time in all experiments. Fe-strätlingite could not be synthesised 

182


at any conditions. Portlandite and Fe-hydroxide co-precipitated with the different AFm 

phases. Generally, the AFm phases form very slowly (over the course of a few years) at 

20 °C from C2F, while at 50 °C the formation is faster. 

The structure of the stable Fe-containing AFm hydrates (Fe-monocarbonate, Fe- 

monosulfate, Fe-Friedel’s salt) were determined and refined using synchrotron powder 

diffraction data. Fe-monocarbonate, Fe-monosulfate and Fe-Friedel’s salt show a 

rhombohedral R3 c symmetry. 

The solubility products obtained at standard conditions (25 °C, 1 atm) for the Fe- 

containing AFm phases are generally 2 to 3 log units lower than those of the Al- 

containing AFm phases, comparable to the difference between Fe(OH)3 and gibbsite. 

logKs0 (Fe) log Ks0 (Al) 

Monosulfate -31.57 -29.26 

Monocarbonate -34.59 -31.47 

Hemicarbonate -30.83 -29.13 

C4(A,F)H13 -30.64 -25.4 

Friedel’s salt -28.62 -27.69 

(Al,Fe)(OH)3 -4.1 -1.24 

Notable exceptions are Fe-Friedel’s salt and Fe-hemicarbonate which have relatively high 

solubility products compared to the Al-phases, indicating that Fe-Friedel’s salt and Fe- 

hemicarbonate are most probably not stable in hydrated cements. 

Also the formation of hydrogarnets (Al-katoite (C3AH6), Fe-katoite (C3FH6), Al-siliceous 

hydrogarnet and Fe-siliceous hydrogarnet) was studied. C3AH6 was the stable phase in 

the CaO-Al2O3-H2O system in the absence of other ions, in all cases co-precipitation of 

traces of C4AH13 was observed. In contrast, C3FH6 was found to be metastable while 

C4FH13, portlandite, amorphous Fe-hydroxide and carbonate containing AFm phases 

183


formed. The investigation of Al- and Fe-siliceous hydrogarnet was complicated due to 

the formation of two hydrogarnets with different silica contents in both systems. In 

addition, C(-A)-S-H co-precipitated during the preparation of Al-containing siliceous- 

hydrogarnet. Al-containing siliceous hydrogarnet did not form at room temperatures but 

only at 110 °C. In contrast, stable but poorly crystalline Fe-siliceous hydrogarnet formed 

slowly at room temperature. At 110 °C the formation of two well crystalline Fe-siliceous 

hydrogarnet (C3FS0.95H4.1 and C3FS1.5H2.96) was observed both with the typical cubic 

crystal structure. 

The solubility products of Al and Fe-containing Si-hydrogarnets decreases strongly with 

increasing silica content indicating that a strong stabilisation. The solubility products 

obtained at standard conditions (25 °C, 1 atm) for the Fe-containing hydrogarnets were 5 

to 7 log units lower than those of the Al-containing hydrogarnets. 

log Ks0 (Fe) log Ks0 (Al) 

C3(A,F)H6 -25.56 -20.56 

C3AS0.41H5.18 

C3AS0.84H4.32 

C3FS0.95H4.1 

-32.75 

C3FS1.52H2.96 -34.68 

-25.47 

-26.70 

The relatively low solubility products of the Fe-containing siliceous hydrogarnets 

compared to the Al-phases, together with the observation that Fe-siliceous hydrogarnets 

form at room temperature, indicates that Fe-containing siliceous hydrogarnets could be 

stable in hydrated cements. 

184

Speciation of iron(III) in hydrated cements 


Standard analytical techniques (XRD, TGA, SEM) shows C-S-H, CH, AFt and AFm 

phases formation during hydration. With selective dissolution using SAM the presence of 

Al- and Fe-containing siliceous hydrogarnet was confirmed in the hydrated cements. 

EXAFS spectroscopy further shows that during the first few hours of hydration Fe(OH)3 

formed. At later ages, Fe-containing siliceous hydrogarnet was identified as the main Fe- 

containing hydrate. Thermodynamic modeling of the hydration predicts that iron should 

be bound in siliceous hydrogarnet, possibly as a solid solution together with aluminum: 

Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y. These modeling results are in agreement with the 

experimental observations. 

The investigations have been complemented with a study on the hydration of synthetic, 

aluminum free cements. In the presence of sulfate, Fe-ettringite was found to be stable 

besides C-S-H and CH. Possibly Fe-siliceous hydrogarnets were also present, but they 

could not be clearly identified standard analytical techniques. The same hydrate 

assemblage (Fe-ettringite, C-S-H and CH) was observed for the calcite free and calcite 

containing synthetic Fe-cements. The experimental findings are supported by the 

modeling results. In Portland cements, however, the formation of Fe-ettringite is not 

expected as the more stable Al-ettringite preferentially forms. The presence of aluminum 

in the investigated Portland cements leads to the formation of stable Al-ettringite and Al- 

containing AFm phases while all Fe is released during ferrite dissolution present as Fe- 

siliceous hydrogarnet in the long term. 

In general, the solubility products of Fe-containing hydrates determined in this study will 

help in the future to establish whether and to what extent Fe-containing hydrates are 

stable in blended cementitious systems, and allow further considerations on the durability 

185


of cementitious materials and assessments related to the safe disposal of radioactive and 

hazardous wastes. 

A significant part of the aluminum in hydrated Portland cement might be incorporated in 

C-S-H. Recent experimental data indicates that up to 50% of aluminum present in a 

blended PC-slag system was incorporated in the C-A-S-H phase [140]. The incorporation 

in Portland cement is somewhat lower as the incorporation of aluminium in C-S-H 

decreases with increasing Ca/Si ratio of the C-S-H [141]. 

This poses the question whether C-S-H might also incorporate significant amounts of 

iron. NMR results, however, indicate that the uptake of iron in C-S-H is much lower than 

the uptake of aluminum. The presence of Fe 3+ within the structure of the solids distorts 

the NMR signal and leads to strong dipolar coupling between the spin of the element 

investigated and the spins of the two unpaired electrons from Fe 3+ [142]. Montheilet et al. 

[143] calculated on the basis of the distortion of H-NMR measurements that 2.5×10 15 Fe 

were present per m 2 of C-S-H in a white cement, which corresponds to 1 Fe per 25’000 

Si 2 . In a grey cement, the H-NMR measurements indicate that 2-3 times more Fe are 

bound in the C-S-H [144], so that in a grey cement there might be 1 Fe per 10’000 Si, 

which corresponds to 0.1 mg Fe2O3 per 1 g SiO2 and thus is a negligible quantity. 

5.2. Outlook 

The research presented in this thesis also reveals that in different areas more detailed 

studies should be carried out to address the following topics: 

Solid solutions in the interlayer structure of Fe-Friedel’s salt and of mixed 

hydrogarnets (Ca3(AlxFe1-x)2(SiO4)3-y(OH)4y). 

2 Surface area of C-S-H: 125 m 2 /cm 3 , Molar volume of C-S-H :78 cm 3 /mol: 2.5×10 15 Fe/m 2 × 125 m 2 /cm 3 

×78 cm 3 /mol /(6×10 23 parts per mol) = 0.000041 Fe/Si = 1 Fe/25’000 Si 

186


The conditions and kinetics of the formation of mixed hydrogarnets. 

The fate of iron under reducing conditions. Reducing conditions in cements may 

dominate in materials blended with blast furnace slags, upon the addition of 

chromate-reducing agents (often FeSO4 xH2O) or in the near field of a repository 

for radioactive waste where strongly reducing conditions are expected to prevail 

in the long term. Besides sulphur (and H2O and O2) Fe constitutes the most 

abundant redox couple (Fe(II)/Fe(III)) and thus plays a key role in controlling 

redox equilibria of cementitious systems. The effect of reducing conditions on the 

Fe speciation in cement is still poorly understood. 

The possible use of Mössbauer spectroscopy or other techniques complementary 

to the EXAFS technique. 

187

ABBREVATIONS 

Cement chemistry notations 

ABBREVIATIONS 

C= CaO A=Al2O3 S = SiO2 s = SO3 

c = CO2 F = Fe2O3 H = H2O 

Hydrated phases 

AFm Al2O3-Fe2O3-mono phase AFt Al2O3-Fe2O3-tri phase 

C-S-H Calcium silicate hydrate 

Portlandite=CH=P 

Gypsum=CsH2 

Al-hemicarbonate=Al-Hc=C4Ac0.5H10 

Al-monocarbonate=Al-Mc= C4AcH11 

Fe-hemicarbonate=Fe-Hc=C4Fc0.5H10 

Fe-monocarbonate=Fe-Mc= C4FcH12 

Al-monosulfate =Al-Ms= C4AsH12 Fe-monosulfate=Fe-Ms= C4FsH12 

Al-Friedel’s salt=Al-Fr Fe-Friedel’s salt=Fe-Fr 

Al-OH-AFm= C4AH13 Fe-OH-AFm= C4FH13 

Al-katoite= C3AH6 Fe-katoite= C3FH6 

Al-ettringite=Al-Ett= C4As3H32 Fe-ettringite=Fe-Ett= C4Fs3H32 

Al-siliceous hydrogarnet= Al-Si-Hg (C4ASH4) Fe-Siliceous hydrogarnet= Fe-Si-Hg = 

(C4FSH4) 

OPC Ordinary Portland cement 

HS High sulfate resistant cement 

188

Techniques 

XRD X-ray diffraction 

TGA Thermogravimetric analysis 

DTG Derivative thermogravimetric analysis 

SEM Scanning electron microscopy 

XAS X-ray absorption spectroscopy 

ABBREVIATIONS 

EXAFS Extended X-ray absorption fine structure 

IR Infrared spectroscopy 

ICP/OES Inductively-coupled plasma optical emission spectrometry 

ICP/MS Inductively-coupled plasma mass spectrometry 

GEMS Gibbs energy minimization selector 

MBSSAS Margules binary solid solution aqueous solution 

PDF Powder diffraction file 

Thermodynamic parameters 

CP 0 standard molar heat capacity of species at T, P(J K -1 mol -1 ) 

∆rCP 0 T 

∆rCP 0 T0 

∆fG 0 T0 

∆rG 0 T 

∆Gex 

standard molar heat capacity change of reaction at T (J K -1 mol -1 ) 

standard molar heat capacity change of reaction at T0 = 298 K(25°C)( J K -1 mol -1 ) 

standard molar Gibbs energy (of formation from elements)at T0 =298 K 

(25°C)(KJ mol -1 ) 

standard Gibbs energy change in a reaction (KJ mol -1 ) 

excess molar Gibbs energy of mixing for the solid solution series (KJ mol -1 ) 

∆fG 0 i standard molar Gibbs energy of formation of end member i of a solid solution series (KJ 

mol -1 ) 

189

ABBREVIATIONS 

∆Gid molar Gibbs energy of mixing of an ideal solid solution (KJ mol -1 ) 

∆GM molar Gibbs energy of mixing for end members i of the solid solution series (KJ mol -1 ) 

∆Gss molar Gibbs energy of a solution between different end members i (KJ mol -1 ) 

γi Activity coefficient of species i 

∆rH 0 T standard change of enthalpy of reaction at T (KJK -1 mol -1 ) 

∆rH 0 T0 standard change of enthalpy of reaction at T0=298 K (25°C)(KJK -1 mol -1 ) 

∆fH 0 T0 standard molar enthalpy at T0 = 298 K (25°C)(KJ K -1 mol -1 ) 

I effective molal ionic strength of aqueous solution 

KT 

thermodynamic equilibrium constant of reaction at T 

П total solubility product in Lippmann phase diagrams 

R universal gas constant (8.31451 J K -1 mol -1 ) 

∆rS 0 T standard entropy change in reaction at T (J K -1 mol-1) 

∆rS 0 T0 standard entropy change in reaction at T0 =298K (25°C) (JK -1 mol -1 ) 

S 0 T0 

standard molar absolute entropy at T0 =289K (25°C) (JK -1 mol -1 ) 

T temperature of interest (K) 

T0 

V 0 

reference temperature (298 K) 

standard molar volume (cm 3 mol -1 ) 

Xaq,I aqueous activity fractions of the substitutable species i 

Xi 

mole fraction of end member i in solid solution 

logKs0 solubility product 

190

APPENDIX 

APPENDIX 

Appendix A: Additional fitted structural parameters 

Appendix A1. EXAFS fitted structural parameters for Fe-containing compounds and comparison 

with XRD data 

Phases Atomic pair N σ(Å) R(Å) R-Factor(%) R(Å) (XRD) 

C4AF Fe-O 3.9 0.009 1.95 2 1.94 [145] 

Fe-Ca 5.1 0.011 3.12 3.04/3.22 

Fe-Fe 5 0.007 3.6 3.48 

Fe-Al 4.8 0.004 3.59 3.66 

C2F Fe-O 3.0 + 0.006 1.9 6.2 1.96 [146] 

Fe-Ca 3.0 + 0.018 3.14 3.06/3.22 

Fe-Fe 2.0 + 0.005 3.65 3.72 

Fe-Fe 4.0 + 0.011 3.99 3.9 

Fe(OH)3 Fe-O 6.0 + 0.012 2 3.3 1.96 [147] 

Fe-Fe 4.0 + 0.012 3.04 3.01 

Fe-Fe 4.0 + 0.013 3.41 3.44 

C4FcH12 Fe-O 6.0+ 0.006 2.02 6 2.04 

Fe-Ca 6.0+ 0.008 3.47 3.46 

C4AcH12 Al-Ca 6.0 - 3.40* 

C6Fs3H32 Fe-O 5.4 0.004 2.03 2.5 1.92 [148] * 

Fe-Ca 5.1 0.01 3.52 

C6AsH32 Al-Ca 6.0 - - 3.44* 

Fe-Hydrotalcite Fe-O 5.7 0.003 2.03 1.5 2.06 [149]] 

Fe-Mg 6.0 # 0.005 3.13 3.11 

C3FSH4 Fe-O 6.0 + 0.009 2.01 3.7 2.02 [106] * 

Fe-Si 6.0 + 0.005 3.39 

Fe-Ca 6.0 + 0.007 3.46 

C3ASH4 Al-Si 6 3.43* 

Al-Ca 6 - - 3.43* 

N: Coordination number of the neighboring atom (Uncertainty ± 20%) 

R: Distance to the neighboring atom (Uncertainty ± 0.02 Å) 

σ: Debye-Waller factor 

*: values from Al-containing phases 

+: fixed parameters during fitting 

191

Intensity (arb. units) 

Appendix B: Additional figures 

Fe-Mc 

Fe-Hc 

P 

Fe-Mc 

Fe-Hc 

APPENDIX 

10 15 20 

2CuK 

25 30 

R 

P C 

1 years 

120 days 

28 days 

7 days 

Appendix B1 XRD pattern of Fe-Mc synthesized at 50°C, P:portlandite, C:calcite 

Weight loss in % 


100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

-0.4 

Fe-Mc 

Fe-Mc 

200 400 


600 800 

Appendix B2 Thermal analysis (TGA and DTG) of Fe-Mc at 50°C 

P 

C 

120 days 

1 year 

192

Intensity (arb. Units) 

P 

Fe 2O 3 

R 

C 

P 

20 30 

2 CuK 

APPENDIX 

P 

1 year 

120 days 

28 days 

Appendix B3 XRD pattern of Fe-Mc synthesized at 80°C 

Weight loss in % 


100 

90 

80 

70 

60 

-0.1 

-0.2 

-0.3 

-0.4 

P 

7 days 

1 year 

200 400 600 800 


C 

120 days 

Appendix B4 Thermal analysis (TGA and DTG) of Fe-Mc formation at 50°C 

193


Fe-MS 

C 2F 

P 

APPENDIX 

Fe-MS 

Fe-MS 

8 13 18 23 28 33 

2θCuKα 

Appendix B5 XRD pattern of Fe-Ms synthesized at 50°C 


Fe-Ms 

P 

Fe-Ms 

C 2F 

P 

Fe-MS 

8 13 18 23 28 33 

2θCuKα 

Fe 2O 3 

Appendix B6 XRD pattern of Fe-Ms synthesized at 80°C 

P 

P 

Fe 2O 3 

P 

C 2F 

360 days 

120 days 

28 days 

7 days 

360days 

120 days 

28 days 

7 days 

194

APPENDIX 

Appendix B7 XRD patterns of Fe-Friedel’s salt formed at 20°C and sampled after 500 days 

equilibration times synthesized from C2F, CaCl2.2H2O and CaO in 0.1 M KOH 

Appendix B8 TGA-DTG curves of Fe-Friedel’s salt formed at 20°C and sampled after 500 days 

equilibration times synthesized from C2F, CaCl2.2H2O, and CaO in 0.1 M KOH 

195

(Al+Fe)/Ca 

0.7 

0.6 

0.5 

0.4 

0.3 

0.2 

0.1 

0 

Hg 

AFm 

CH 

Si‐Hg 

C‐S‐H 

APPENDIX 

0 0.2 0.4 0.6 0.8 1 

Si/Ca 

Appendix B9 Atomic ratio of 50 years old hydrated cement before selective dissolution. The 

0.8 

0.6 

a 

0.4 

F 

e 

)/C 

l+ 

(A 

0.2 

0 

cement is composed of 64.5%of C3S, 10%C2S, 12.1%C3A, 3.2%CaSO4, 0.4CaO, 

0.8MgO, 0.08%Na2O, 0.2K2O, 0.27%TiO2. Note that this old samples obtained 

from Taylor. 

Ca 3Al 0.8Fe 0.7Si 0.98O 5 

0 0.1 0.2 0.3 0.4 0.5 0.6 

Si/Ca 

Appendix B10 Atomic ratio of 50 years old hydrated cement after selective dissolution SAM 

196

REFERENCES 

REFRENCES 

[1] H.F.W. Taylor, Cement Chemistry, Thomas Telford Publishing, London (1997). 

[2] H.J. Kuzel, H. Pöllmann, Hydration of C3A in the presence of Ca(OH)2, CaSO4·2H2O 

and CaCO3, Cement and Concrete Research, 21 (1991) 885-895. 

[3] B. Lothenbach, T. Matschei, G. Möschner, F.P. Glasser, Thermodynamic modelling 

of the effect of temperature on the hydration and porosity of Portland cement, Cement 

and Concrete Research, 38 (2008) 1-18. 

[4] B. Lothenbach, E. Wieland, A thermodynamic approach to the hydration of sulphateresisting 

Portland cement, Waste Management, 26 (2006) 706-719. 

[5] D.P. Bentz, Influence of alkalis on porosity percolation in hydrating cement pastes, 

Cement and Concrete Composites, 28 (2006) 427-431. 

[6] B. Lothenbach, F. Winnefeld, Thermodynamic modelling of the hydration of Portland 

cement, Cement and Concrete Research, 36 (2006) 209-226. 

[7] T. Matschei, B. Lothenbach, F.P. Glasser, Thermodynamic properties of Portland 

cement hydrates in the system CaO-Al2O3-SiO2-CaSO4-CaCO3-H2O, Cement and 

Concrete Research, 37 (2007) 1379-1410. 

[8] K. Scrivener, P.L. Pratt, Cerma proc 35 (1984) 207. 

[9] K.L. Scrivener, T. Füllmann, E. Gallucci, G. Walenta, E. Bermejo, Quantitative study 

of Portland cement hydration by X-ray diffraction/Rietveld analysis and independent 

methods, Cement and Concrete Research, 34 (2004) 1541-1547. 

[10] L. Black, C. Breen, J. Yarwood, J. Phipps, G. Maitland, In situ Raman analysis of 

hydrating C3A and C4AF pastes in presence and absence of sulphate, Advances in 

Applied Ceramics, 105 (2006) 209-216. 

[11] M. Collepardi, S. Monosi, G. Moriconi, M. Corradi, Tetracalcium aluminoferrite 

hydration in the presence of lime and gypsum, Cement and Concrete Research, 9 (1979) 

431-437. 

[12] M. Ecker, Diadochiebeziehungen in Calciumaluminatferraten und deren 

Hydratationsprodukten, PhD thesis, Hallesches Jahrbuch für Geowissenschaften, Reihe 

B, Martin-Luther-Universität Halle-Wittenberg, (1998). 

[13] A. Emanuelson, S. Hansen, Distribution of iron among ferrite hydrates, Cement and 


[14] N. Meller, C. Hall, A.C. Jupe, Colston, S.L., S.D.M. Jacques, P. Barnes, J. Phipps, 

The paste hydration of brownmillerite with and without gypsum: a time resolved 

synchrotron diffraction study at 30, 70, 100 and 150 °C, Journal of Materials Chemistry, 

14 (2004) 428-435. 

197

REFRENCES 

[15] G. Möschner, B. Lothenbach, J. Rose, A. Ulrich, R. Figi, R. Kretzschmar, Solubility 

of Fe-ettringite (Ca6[Fe(OH)6]2(SO4)3 · 26H2O), Geochimica et Cosmochimica Acta, 72 

(2008) 1-18. 

[16] G. Möschner, B. Lothenbach, A. Ulrich, R. Figi, R. Kretschmar, Solid solution 

between Al-ettringite and Fe-ettringite (Ca6[Al1-xFex(OH)6]2(SO4)3 . 26H2O), Cement and 


[17] W. Schwarz, Novel cement matrices by accelerated hydration of the ferrite phase in 

Portland cement via chemical activation: kinetics and cementitious properties, Advn Cem 

Bas Mat, 2 (1995) 189-200. 

[18] P.W. Brown, Early hydration of tetracalcium aluminoferrite in gypsum and lime– 

gypsum solutions., Journal of the American Ceramic Society, 70 (1987) 493–496. 

[19] B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Mesbah, Y. Filinchuk, 

A. Wichser, E. Wieland, Iron in carbonate containing AFm phases, Cement and Concrete 

Research, 41 (2011) 311-323. 

[20] M. Fukuhara, S. Goto, K. Asaga, M. Daimon, R. Kondo, Mechanisms and kinetics 

of C4AF hydration with gypsum, Cement and Concrete Research, 11 (1981) 407-414. 

[21] N.C. Collier, N.B. Milestone, J. Hill, I.H. Godfrey, The disposal of radioactive ferric 

floc, Waste Management, 26 (2006) 769-775. 

[22] R.S. Gollop, H.F.W. Taylor, Microstructural and microanalytical studies of sulfate 

attack. II. Sulfate-resisting Portland cement: Ferrite composition and hydration chemistry, 

Cement and Concrete Research, 24 (1994) 1347-1358. 

[23] M. Paul, F.P. Glasser, Impact of prolonged warm (85°C) moist cure on Portland 

cement paste, Cement and Concrete Research, 30 (2000) 1869-1877. 

[24] K.S. Harchand, Vishwamittar, K. Chandra, Infrared and Mössbauer study of two 

Indian cement, Cement and Concrete Research, 10 (1980) 243-252. 

[25] F. Jalilehvand, X-ray Absorption Spectroscopy. University of Calgary. 

(http://www.chem.ucalgary.ca/research/groups/faridehj/xas.pdf) Accessed June 1,2009. 

[26] M. Newville, Fundamentals of XAFS, Consortium for advanced radiation sources 

University of Chicago, IL, (2004). 

[27] A. Manceau;, M.A. Marcus;, N. Tamura;, Quantitative speciation of heavy metals in 

soils and sediments by synchrotron X-ray techniques, Reviews in Mineralogy and 

Geochemistry, 49 (2002) 341-428. 

[28] A.M. Scheidegger, M. Vespa, D. Grolimund, E. Wieland, M. Harfouche, I. 

Bonhoure, R. Dähn, The use of (micro)-X-ray absorption spectroscopy in cement 

research, Waste Management, 26 (2006) 699-705. 

198

REFRENCES 

[29] M. Vespa, R. Dähn, T.E. Gallucci, D. Grolimund, E. Wieland, A.M. Scheidegger, 

Microscale investigations of Ni uptake by cement using a combination of scanning 

electron microscopy and synchrotron-based techniques, Environmental Science and 

Technology, 40 (2006) 7702-7709. 

[30] E. Wieland, R. Dähn, M. Vespa, B. Lothenbach, Micro-spectroscopic investigation 

of Al and S speciation in hardened cement paste, Cement and Concrete Research, 40 

(2010) 885-891. 

[31] G. Möschner, A thermodynamic approach to cement hydration: the role of retarding 

admixtures and Fe-minerals during the hydration of cements, PhD thesis, ETH Zürich, 

(2007). 

[32] A.N. Christensen, T.R. Jensen, J.C. Hanson, Formation of ettringite, 

Ca6Al2(SO4)3(OH)12·26H2O, AFt, and monosulfate, Ca4Al2O6(SO4)·14H2O, AFm-14, in 

hydrothermal hydration of Portland cement and of calcium aluminum oxide—calcium 

sulfate dihydrate mixtures studied by in situ synchrotron X-ray powder diffraction, 

Journal of Solid State Chemistry, 177 (2004) 1944-1951. 

[33] E. Gallucci, P. Mathur, K. Scrivener, Microstructural development of early age 

hydration shells around cement grains, Cement and Concrete Research, 40 (2010) 4-13. 

[34] K.S. Harchand, R. Kumar, K. Chandra, Vishwamittar, Mössbauer and X-ray 

investigations of some portland cements, Cement and Concrete Research, 14 (1984) 170- 

176. 

[35] H.F.W. Taylor, D.E. Newbury, An electron microprobe study of a mature cement 

paste, Cement and Concrete Research, 14 (1984) 565-573. 

[36] A.P. Hammersley, S.O. Svensson, M. Hanfland, A.N. Fitch, D. Hausermann, Twodimensional 

detector software: From real detector to idealised image or two-theta scan, 

High Pressure Research: An International Journal, 14 (1996) 235 - 248. 

[37] G. Le Saout, E. Lécolier, A. Rivereau, H. Zanni, Chemical structure of cement aged 

at normal and elevated temperatures and pressures: Part I. Class G oilwell cement, 


[38] M. Neville, IFEFFIT: interactive EXAFS analysis and FEFF fitting, J Synchrotron 

Radiat 8(2005) 322–324. 

[39] B. Ravel, M. Neville, ATHENA, ARTEMIS, HEPHAESTUS: data analysis for Xray 

absorption spectroscopy using IFEFFIT, Journal of Synchrotron Radiation, 12 (2005) 

537-541. 

[40] M.A. Marcus, A.A. MacDowell, R. Celestre, A. Manceau, T. Miller, H.A. Padmore, 

R.E. Sublett, Beamline 10.3.2 at ALS: A hard X-ray microprobe for environmental and 

materials sciences, Journal of Synchrotron Radiation, 11 (2004) 239-247. 

199

REFRENCES 

[41] T. Ressler, J. Wong, J. Roos, I.L. Smith, Quantitative speciation of mn-bearing 

particulates emitted from autos burning (methylcyclopentadienyl) manganese tricarbonyladded 

gasolines using XANES spectroscopy, Environmental Science and Technology, 34 

(2000) 950-958. 

[42] E.R. Malinowski, Determination of number of factors and expermental error in a 

data matrix, Anal chem, 49 (1977) 612-617. 

[43] D.A. Kulik, GEMS-PSI 23, PSI Villigen, Switzerland; available at 

http://les.web.psi.ch/Software/GEMS-PSI/index.html, (2009). 

[44] W. Hummel, U. Berner, E. Curti, F.J. Pearson, T. Thoenen, Nagra/PSI Chemical 

Thermodynamic Data Base 01/0, Universal Publishers/uPublishcom, Parkland, Florida, 

(2002). 

[45] T. Thoenen, D.A. Kulik, Nagra/PSI Chemical Thermodynamic Data Base 01/01 for 

the GEM-Selektor (V.2- PSI) Geochemical Modeling Code: Release 28-02-03. Internal 

Report TM-44-03-04, Available from: http://leswebpsich/software/GEMS- 

PSI/thermodata/indexhtml 2003. 

[46] J. Ederova, V. Satava, Heat capacities of C3AH6, C4ASH12 and C3AS3H32, , 

Thermochim Acta, 31 (1979) 126-128. 

[47] G.M. Anderson, D.A. Crerar, Thermodynamics in Geochemistry: the Equilibrium 

Model, Oxford University Press, Oxford, 1993. 

[48] H.C. Helgeson, J.M. Delany, H.W. Nesbitt, D.K. Bird, Summary and critique of the 

thermodynamic properties of rock forming minerals, American Journal of Science,, 278- 

A (1978) 229. 

[49] D.A. Kulik, Thermodynamic properties of surface species at the mineral-water 

interface under hydrothermal conditions: a Gibbs energy minimization single-site 2pKA 

triple-layer model of rutile in NaCl electrolyte to 250°C, Geochimica et Cosmochimica 

Acta, 64 (2000) 3161-3179. 

[50] D.A. Kulik, Minimising uncertainty induced by temperature extrapolations of 

thermodynamic data: A pragmatic view on the integration of thermodynamic databases 

into geochemical computer codes. The use of thermodynamic databases in performance 

assessment. OECD, Paris, Barcelona, Spain., (2002). 

[51] D. Damidot, B. Lothenbach, D. Herfort, F.P. Glasser, Thermodynamics and cement 

science, Cement and Concrete Research, 41 (2011) 679-695. 

[52] T. Matschei, B. Lothenbach, F.P. Glasser, The AFm phase in Portland cement, 


[53] D. Kulik, GEMS-PSI 2.1, in, available at http://les.web.psi.ch/Software/GEMS-PSI/, 

PSI-Villigen, Switzerland, 2006. 

200

REFRENCES 

[54] M. Balonis, B. Lothenbach, G. Le Saout, F.P. Glasser, Impact of chloride on the 

mineralogy of hydrated Portland cement systems, Cement and Concrete Research, 40 

(2010) 1009-1022. 

[55] D.A. Kulik, M. Kersten, Aqueous solubility diagrams for cementitious waste 

stabilization systems: II, End-member stoichiometries of ideal calcium silicates hydrate 

solid solutions, Journal of the American Ceramic Society, 84 (2001) 3017-3026. 

[56] D.E. Macphee, S.J. Barnett, Solution properties of solids in the ettringite-thaumasite 

solid solution series, Cement and Concrete Research, 34 (2004) 1591-1598. 

[57] J. Bruno, D. Bosbach, D. Kulik, A. Navrotsky, Chemical Thermodynamics Vol. 10. 

Chemical Thermodynamics of Solid Solutions of Interest in Nuclear Waste Management, 

North Holland Elsevier Science Publishers B.V., Amsterdam, The Netherlands, 2007. 

[58] E. Curti, Coprecipitation of radionuclides with calcite: estimation of partition 

coefficients based on a review of laboratory investigations and geochemical data, Applied 

Geochemistry, 14 (1999) 433-445. 

[59] P.D. Glynn, MBSSAS: A code for the computation of Margules parameters and 

equilibrium relations in binary solid-solution aqueous-solution systems, Computers and 

Geoscience, 17 (1991) 907-966. 

[60] O. Redlich, A.T. Kister, Algebraic representation of the thermodynamic properties 

and the classification of solutions, Ind Eng Chem, 40 (1948) 345-348. 

[61] S.M. Leisinger, B. Lothenbach, G. Le Saout, C.A. Johnson, Thermodynamic 

modeling of solid solutions between monosulfate and monochromate 3CaO • Al2O3 • 

Ca[(CrO4)x(SO4)1-x] • nH2O, Cement and Concrete Research in press, (2011). 

[62] S.M. Leisinger, B. Lothenbach, G.L. Saout, R. Kägi, B. Wehrli, C.A. Johnson, Solid 

Solutions between CrO4- and SO4-Ettringite Ca6(Al(OH)6)2[(CrO4)x(SO4)1-x]3*26 H2O, 

Environmental Science & Technology, 44 (2010) 8983-8988. 

[63] T. Matschei, Thermodynamics of Cement Hydration, University of Aberdeen, 

(2007). 

[64] L.J. Parrot, D.C. Killoh, PREDICTION OF CEMENT HYDRATION, in, 1984, pp. 

41-53. 

[65] B. Lothenbach, G. Le Saout, E. Gallucci, K. Scrivener, Influence of limestone on the 

hydration of Portland cements, Cement and Concrete Research, 38 (2008) 848-860. 

[66] M. François, G. Renaudin, O. Evrard, A cementitious compound with composition 

3CaO.Al2O3.CaCO3.11H2O, Acta Crystallographica Section C: Crystal Structure 

Communications, 54 (1998) 1214-1217. 

[67] T. Matschei, B. Lothenbach, F.P. Glasser, The role of calcium carbonate in cement 

hydration, Cement and Concrete Research, 37 (2007) 551-558. 

201

REFRENCES 

[68] F.P. Glasser, A. Kindness, S.A. Stronach, Stability and solubility relationships in 

AFm phases: Part I. Chloride, sulfate and hydroxide, Cement and Concrete Research, 29 

(1999) 861-866. 

[69] H. Poellmann, Solid solution in the system 3CaO·Al2O3·CaSO4·aq– 

3CaO·Al2O3·Ca(OH)2·aq, Neues Jahrbuch fur Mineralogie, Abhandlungen 161 (1989) 

27-41. 

[70] M. Ecker, H.J. Pöllmann, Synthesis and characterisation of CO3-containing Ca-Al- 

Fe-hydrates, International symposium of chemistry of cement, Göthborg, (1997) 2ii032. 

[71] R. Fischer, H.J. Kuzel, Reinvestigation of the system C4A.nH2O-C4A.CO2.nH2O, 


[72] D. Damidot, S. Stronach, A. Kindness, M. Atkins, F.P. Glasser, Thermodynamic 

investigation of the CaO-Al2O3-CaCO3-H2O closed system at 25°C and the influence of 

Na2O, Cement and Concrete Research, 24 (1994) 563-572. 

[73] H. Pöllmann, Die Kristallchemie der Neubildungen bei Einwirkung von 

Schadstoffen auf hydrauliche Bindemittel, PhD thesis, Erlangen-Nürnberg, (1984). 

[74] J.P. Rapin, G. Renaudin, E. Elkaim, M. Francois, Structural transition of Friedel's 

salt 3CaO·Al2O3·CaCl2·10H2O studied by synchrotron powder diffraction, Cement and 


[75] G. Renaudin, F. Kubel, J.P. Rivera, M. Francois, Structural phase transition and high 

temperature phase structure of Friedels salt, 3CaO · Al2O3 · CaCl2 · 10H2O, Cement and 


[76] G. Renaudin, M. Francois, O. Evrard, Order and disorder in the lamellar hydrated 

tetracalcium monocarboaluminate compound, Cement and Concrete Research, 29 (1999) 

63-69. 

[77] J. Rose, A. Bénard, S. El Mrabet, A. Masion, I. Moulin, V. Briois, L. Olivi, J.Y. 

Bottero, Evolution of iron speciation during hydration of C4AF, Waste Management, 26 

(2006) 720-724. 

[78] A. Mesbah, C. Cau-dit-Coumes, F. Frizon, F. Leroux, J. Ravaux, G. Renaudin, A 

new investigation of the Cl − –CO3 2− substitution in AFm phases, Journal of the American 

Ceramic Society, 94 (2011) 1901-1910. 

[79] G. Renaudin, R. Segni, D. Mentel, J.-M. Nedelec, F. Leroux, C. Taviot-Gueho, A 

raman study of the sulfated cement hydrates: ettringite and monosulfoaluminate, Journal 

of Advanced Concrete Technology, 5 (2007) 299-312. 

[80] G. Renaudin, J.P. Rapin, B. Humbert, M. François, Thermal behaviour of the 

nitrated AFm phase Ca4Al2(OH)12(NO3)2 - 4H2O and structure determination of the 

intermediate hydrate Ca4Al2(OH)12(NO3)2 - 2H2O, Cement and Concrete Research, 30 

(2000) 307-314. 

202

REFRENCES 

[81] H.F.W. Taylor, Crystal structures of some double hydroxide minerals, Min Mag, 39 

(1973) 377-389. 

[82] A.F. Hollemann, E. Wiberg, Lehrbuch der anorganischen Chemie. de Gruyter, 

Berlin, (1985). 

[83] M.A. Trezza, A.E. Lavat, Analysis of the system 3CaO.Al2O3-CaSO4.2H2O-CaCO3- 

H2O by FT-IR spectroscopy, Cement and Concrete Research, 31 (2001) 869-872. 

[84] J. Neubauer, F. Götz-Neunhoeffer, D. Schmitt, M. Degenkolb, U. Holland, In-situ 

Untersuchung der frühen PZ-Hydratation., Internationale Baustofftagung (ibausi), 

Weimar, Germany, 16 (2006). 

[85] B.A. Clark, P.W. Brown, The formation of calcium sulfoaluminate hydrate 

compounds: Part II, Cement and Concrete Research, 30 (2000) 233-240. 

[86] J. Plank, D. Zhimin, H. Keller, F.v. Hössle, W. Seidl, Fundamental mechanisms for 

polycarboxylate intercalation into C3A hydrate phases and the role of sulfate present in 

cement, Cement and Concrete Research, 40 (2010) 45-57. 

[87] R. Allman, Refinement of the hydrid layer structure Ca2Al(OH)6.1/2SO4.3H2O, 

Neues Jahrbuch fur Mineralogie Monatshefte, 217 (1977) 136-144. 

[88] H.J. Kuzel, Ersatz von Al 3+ durch Cr 3+ und Fe 3+ in 3CaO.Al2O3.CaCl2.nH2O und 

3CaO.Al2O3.CaSO4.nH2O Zem-Kalk-Gips 21 (1968) 493. 

[89] W. Dosch, H. Zur Strassen, Untersuchungen in den Alkali enthaltenen Systemen, 

gebildet aus den Komponenten CaO, Al2O3, Fe2O3, SiO2, SO3, Na2O, K2O H2O, 

Dissertation University of Mainz: Mainz (GER), (1962). 

[90] A. Mesbah, M. François, C. Cau-dit-Coumes, F. Frizon, Y. Filinchuk, F. Leroux, J. 

Ravaux, G. Renaudin, Crystal structure of Kuzel's salt 

3CaO·Al2O3·1/2CaSO4·1/2CaCl2·11H2O determined by synchrotron powder diffraction, 


[91] P. Blanc, X. Bourbon, A. Lassin, E.C. Gaucher, Chemical model for cement-based 

materials: Thermodynamic data assessment for phases other than C-S-H, Cement and 


[92] F.E. Jones, The quaternary system CaO-Al2O3-CaSO4-H2O at 25 °C Journal of 

Physical Chemistry 48 (1944) 311-350. 

[93] F. Zhang, Z. Zhou, Z. Lou, Solubility product and stability of ettringite, Proceedings 

of the 7th Intern Congress on the Chemistry of Cement vol (II), (1980) 88-93. 

[94] U.A. Birnin-Yauri, F.P. Glasser, Friedel's salt, Ca2Al(OH)6(Cl,OH)·2HO: its solid 

solutions and their role in chloride binding, Cement and Concrete Research, 28 (1998) 

1713-1723. 

203

REFRENCES 

[95] J. Csizmadia, G. Balázs, F.D. Tamás, Chloride ion binding capacity of 

aluminoferrites, Cement and Concrete Research, 31 (2001) 577-588. 

[96] I. Rousselot, C. Taviot-Guého, F. Leroux, P. Léone, P. Palvadeau, J.-P. Besse, 

Insights on the Structural Chemistry of Hydrocalumite and Hydrotalcite-like Materials: 

Investigation of the Series Ca2M 3+ (OH)6Cl·2H2O (M 3+ : Al 3+ , Ga 3+ , Fe 3+ , and Sc 3+ ) by X- 

Ray Powder Diffraction, Journal of Solid State Chemistry, 167 (2002) 137-144. 

[97] F. Götz-Neunhoeffer, H. Poellmann, Synthesis and mineralogical characterization of 

calcium ferrate hydrates and their application as reservoir minerals in waste dumps, 

Applied Mineralogy in Research, Economy, Technology, Ecology and Culture, D 

Rammlmair et al (ed), 2 (2000) 551-555. 

[98] H.J. Kuzel, X-ray investigation of some complex calcium aluminate hydrates and 

related compounds, 5th International Symposium on the Chemistry of Cements, II-9 

Tokyo, (1968) 92-97. 

[99] J.-P. Rapin, Synthèse et étude cristallochimique de quelques aluminates et ferrites 

calciques hydratés de formule [Ca2M(OH)6] + , [X, nH2O] - avec X =Cl, Br, l, CI04, 1/2 

CO3 1/2 CrO4 et 1/2 SO4 et M =Al et Fe, PhD thesis, Nancy, (2001). 

[100] F. Götz-Neunhoeffer, Synthese und mineralogische Charakterisierung von 

Calciumferrathydraten-Speicherminerale für die Anwendung in Reststoffdeponien, PhD 

thesis, Erlangen, (1996). 

[101] C. Abate, B.E. Scheetz, Aqueous Phase Equilibria in the System CaO-Al2O3- 

CaCl2-H2O: The Significance and Stability of Friedel's Salt, Journal of the American 


[102] J.V. Bothe Jr, P.W. Brown, PhreeqC modeling of Friedel's salt equilibria at 23±1 

°C, Cement and Concrete Research, 34 (2004) 1057-1063. 

[103] M. Hobbs, Solubilities and ion exchange properties of solid solutions between the 

OH, Cl and CO3 end members of the monocalcium aluminate hydrates, PhD-Thesis, 

University of Waterloo, Canada, (2001). 

[104] N.C. Collier, N.B. Milestone, J. Hill, I.H. Godfrey, Immobilisation of Fe floc: Part 

2, encapsulation of floc in composite cement, Journal of Nuclear Materials, 393 (2009) 

92-101. 

[105] N. Neuville, E. Lécolier, G. Aouad, A. Rivereau, D. Damidot, Effect of curing 

conditions on oilwell cement paste behaviour during leaching: Experimental and 

modelling approaches, Comptes Rendus Chimie, 12 511-520. 

[106] O. Ferro, E. Galli, G. Papp, S. Quartieri, S. Szakall, G. Vezzalini, A new 

occurrence of katoite and re-examination of the hydrogrossular group, Eur J Mineral, 15 

(2003) 419-426. 

204

REFRENCES 

[107] K. Kyritsis, N. Meller, C. Hall, Chemistry and morphology of hydrogarnets formed 

in cement-based CASH hydroceramics cured at 200° to 350°C, Journal of the American 


[108] P. Barnes, J. Bensted, Structure and performance of cements, Spon press, London, 

(2002) 114-139. 

[109] E. Passaglia, R. Rinlnald, Katoite, a new member of the Ca3Al2(SiO4)3- 

Ca3Al2(OH)12 series and a new nomenclature for the hydrogrossular group of minerals, 

Bull Minéral, 107 (1984) 605-618. 

[110] T.G. Jappy, F.P. Glasser, Synthesis and stability of silica substituted hydrogarnet, 

Ca3Al2Si3-xO12-4x(OH)4x, Adv Cem Res, 4 (1991) 1-8. 

[111] M. Atkins, F.P. Glasser, A. Kindness, D.E. Macphee, Solubility data for cement 

hydrate phases (25 °C), DOE Report No: DoE/HMIP/RR/91/032,University of Aberdeen, 

(1991). 

[112] D.G. Bennett, D. Read, M. Atkins, F.P. Glasser, A thermodynamic model for 

blended cements. II: Cement hydrate phases; thermodynamic values and modelling 

studies, Journal of Nuclear Materials, 190 (1992) 315-325. 

[113] F.G. Butler, H.F.W. Taylor, The system CaO–Al2O3–H2O at 5 °C, Journal of the 

Chemical Society, (1958) 2103–2110. 

[114] E.T. Carlson, The system lime-Alumina-water at 1 °C, Journal of Research of the 

National Bureau of Standards, 16 (1958) 1-11. 

[115] R.B. Peppler, L.S. Wells, The system of lime, alumina, water from 50° to 250 ° C, 

J Res Natl Bur Stand, 52 (1954) 75-92. 

[116] L.S. Wells, W.F. Clarke, H.F. McMurdie, Study of the system CaO–Al2O3–H2O at 

temperatures of 21° and 90 °C, Journal of Research of the National Bureau of Standards, 

(1943) 367–407. 

[117] M. Atkins, D.G. Bennett, A.C. Dawes, F.P. Glasser, A. Kindness, D. Read, A 

thermodynamic model for blended cements, Cement and Concrete Research, 22 (1992) 

497-502. 

[118] V.J. Babushkin, G.M. Matveyev, O.P. Mchedlov-Petrossyan, Thermodynamics of 

Silicates. Springer-Verlag, Berlin, H, (1985). 

[119] D.E. Rogers, L.P. Aldridge, Hydrates of calcium ferrites and calcium 

aluminoferrites, Cement and Concrete Research, 7 (1977) 399-409. 

[120] E.P. Flint, H.F. McMurdie, L.S. Wells, Hydrothermal and X-Ray Studies of the 

Garnet-Hydrogarnet Series and Relationship of the Series to Hydration Products of 

Portland Cement, Jour Res Nat Bur Stand, 26 (1941) 13–33. 

205

REFRENCES 

[121] R.A. Robie, B.S Hemingway, Thermodynamic properties of minerals and related 

substances at 298.15 K and 1 bar (105 Pascals) pressures and at higher temperatures, US 

Geol Surv Bull, 2131 (1995) 461. 

[122] D. Sverjensky, E. Shock, H. Helgeson, Prediction of the thermodynamic properties 

of aqueous metal complexes to 1000 C and 5 kb, Geochimica et Cosmochimica Acta, 61 

(1997) 1359-1412. 

[123] J. Johnson, E. Oelkers, H. Helgeson, SUPCRT92: A software package for 

calculating the standard molal thermodynamic properties of minerals, gases, aqueous 

species, and reactions from 1 to 5000 bars and 0 to 1000 C, Computers & Geosciences, 

(1992) 899-947. 

[124] E. Shock, D. Sassani, M. Willis, D. Sverjensky, Inorganic species in geologic 

fluids: Correlations among standard molal thermodynamic properties of aqueous ions and 

hydroxide complexes, Geochimica et Cosmochimica Acta, 61 (1997) 907-950. 

[125] E.L. Shock, Database from E.L.Shock's GEOPIG web site 

http://epsc.wustl.edu/geopig (see references on Page 2) Imported into GEMS-PSI 

database on 10 Jan 2003 (KD44, BU44, TT44) Washington Univ St Louis MO, (1998). 

[126] C. Cohen-Addad, Etude du compose Ca3Fe2(SiO4)1.15(OH)7.4 par absorption 

infrarouge et diffraction des rayons X et des neutrons, Acta Crystallographica Section A, 

27 (1971) 68-70. 

[127] Jun Ito, Clifford Frondel, NEW SYNTHETIC HYDROGARNETS, THE 

AMERICAN MINERALOGIST,, 52 (1967). 

[128] O. Frank-Kamenetskaya, I. Rozhdestvenskaya, A. Shtukenberg, I. Bannova, Y. 

Skalkina, Dissymmetrization of crystal structures of grossular-andradite garnets Ca3(Al, 

Fe)2(SiO4)3, Structural Chemistry, 18 (2007) 493-503. 

[129] B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Wichser, E. Wieland, 

Sytheisis and characterization of CaO-Al2O3-Fe2O3-SiO2-H2O, in preparation (2011a). 

[130] B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Wichser, E. Wieland, 

Fe-containing AFm phases, in preparation, (2011b). 

[131] H. Michiaki, H. Yoshikazu, The reaction between aggregate and cement paste in 

autoclaved concrete, Bulletin of Engineering Geology and the Environment, 30 (1984) 

241-244. 

[132] R. Rinaldi, M. Sacerdoti, E. Passaglia, Strätlingite: crystal structure, chemistry, and 

a reexamination of its polytype vertumnite, European Journal of Mineralogy, 2 (1990) 

841-849 

[133] P.A. O'Day, N. Rivera, Jr., R. Root, S.A. Carroll, X-ray absorption spectroscopic 

study of Fe reference compounds for the analysis of natural sediments, American 

Mineralogist, 89 (2004) 572-585. 

206

REFRENCES 

[134] M. Wilke, F. Farges, P.-E. Petit, G.E. Brown, Jr., F. Martin, Oxidation state and 

coordination of Fe in minerals: An Fe K-XANES spectroscopic study, American 

Mineralogist, 86 (2001) 714-730. 

[135] B. Lothenbach, F. Winnefeld, C. Alder, E. Wieland, P. Lunk, Effect of temperature 

on the pore solution, microstructure and hydration products of Portland cement pastes, 


[136] M.B. Haha, B. Lothenbach, G. Le Saout, F. Winnefeld, Influence of slag chemistry 

on the hydration of alkali-activated blast-furnace slag — Part II: Effect of Al2O3, Cement 

and Concrete Research. 

[137] K. Rozov, U. Berner, C. Taviot-Gueho, F. Leroux, G. Renaudin, D. Kulik, L.W. 

Diamond, Synthesis and characterization of the LDH hydrotalcite-pyroaurite solidsolution 

series, Cement and Concrete Research, 40 (2010) 1248-1254. 

[138] H. Minard, S. Garrault, L. Regnaud, A. Nonat, Mechanisms and parameters 

controlling the tricalcium aluminate reactivity in the presence of gypsum, Cement and 


[139] S. Gunay, S. Garrault, A. Nonat, P. Termkhajornkit, Influnce of calcium sulphate 

on hydration and mechanical strength of tricalcium silicate, International congress on the 

chemistry of cement (ICCC), (2011). 

[140] V. Fernandez, A. Errien, G. Le Saout, K. Scrivener, Availability of Al2O3 in 

Blended Cements, Proc Final Conference Marie Curie Research Training Network, 

Villars-sur-Ollon, Switzerland, (September 2009) 1-3. 

[141] I.G. Richardson, G.W. Groves, The incorporation of minor and trace elements into 

calcium silicate hydrate (C-S-H) gel in hardened cement pastes, Cement and Concrete 

Research, 23 (1993) 131-138. 

[142] J. Skibsted, H.J. Jakobsen, C. Hall, Quantitative Aspects of 27Al MAS NMR of 

Calcium Aluminoferrites, Advanced Cement Based Materials, 7 (1998) 57-59. 

[143] L. Monteilhet, J.P. Korb, J. Mitchell, P.J. McDonald, Observation of exchange of 

micropore water in cement pastes by two-dimensional T_{2}-T_{2} nuclear magnetic 

resonance relaxometry, Physical Review E, 74 (2006) 061404. 

[144] P. Macdonald, personal communication, (2011). 

[145] F. Bertaut, P. Blum, A. Sagnières, La structure du ferrite dicalcique et de la 

brownmillerite, C R Hebd Seances Acad Sci, 244 (1957) 2944–2946. 

[146] A.L. Shaula, Y.V. Pivak, J.C. Waerenborgh, P. Gaczynski, A.A. Yaremchenko, 

V.V. Kharton, Solid Sate ionics, n°33-34 (2006) 2923-2930. 

207

REFRENCES 

[147] A. Szytuła, A. Burewicz, Ž. Dimitrijević, S. Kraśnicki, H. Rżany, J. Todorović, A. 

Wanic, W. Wolski, Neutron Diffraction Studies of α-FeOOH, physica status solidi (b), 26 

(1968) 429-434. 

[148] A. Moore, H.F.W. Taylor, Crystal Structure of Ettringite, Nature, 218 (1968) 1048- 

1049. 

[149] A. Olowe, Advances in X Ray Analysis, 39 (1998). 

208

CURRCULUM VITAE 

Curriculum Vitae 

Name : Belay Zeleke Dilnesa 

Date of Birth : 23 January 1983 

Nationality : Ethiopian 

Marital status : Married 

Language : Amharic (mother tongue), English (Fluent), 

German (medium), Spanish (understand) 

Academic Qualification 

Doctoral study PhD study in Material science 

École polytechnique fédérale de Lausanne EPFL, Switzerland 

(Jun. 2008-Dec. 2011) 

Post Graduate study 

Master study in Chemical and Process Engineering 

Rovira I Virgili University, Spain 

(Oct. 12, 2006 – Jul., 2007) 

Undergraduate study Bachelor Degree in Chemistry 

Alemaya University, Ethiopia 

(Sep. 2000-Jul. 2004) 

Professional experience 

February 2008 – Current, Laboratory Concrete / Construction Chemistry Empa 

(Swiss Federal Laboratories for Materials Science and Technology) 

PhD student and working on Hydration of cement and thermodynamic 

modeling 

October 2007 – January, 2008, Rovira I Virigili University, Department de 

Química Analítica i Quimica Orgànica, Spain 

Synthesis of epoxy resins for the application of thermosetting polymers 

August 2004 – August, 2006 Assistant Lecturer in Arba Minch University, 

Ethiopia in the Department of Applied Chemistry 

Computer skills 

Microsoft office, Origin 

Fortran program 

Hobbies 

Running 

Football

Publications 

CURRCULUM VITAE 

B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Mesbah, Y. Filinchuk, A. 

Wichser and E. Wieland, Iron in carbonate containing AFm phases. Cement and Concrete 

Research, 2011. 41(3): p. 311-323. 

B.Z. Dilnesa, B. Lothenbach, A. Wichser and E. Wieland, Synthesis and characterization 

of CaO-Al2O3-Fe2O3-SiO2 system.. (in preparation) 

B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Mesbah,, A. Wichser and E. 

Wieland, Stablity of monosulfate in the presence of iron.. (in preparation) 

B.Z. Dilnesa, B. Lothenbach, G. Le Saout, G. Renaudin, A. Mesbah,, A. Wichser and E. 

Wieland, Stablity of Fe-Friedel’s salt (in preparation) 

Dilnesa, B.Z., B. Lothenbach, E. Wieland, R. Dähn, and K.L. Scrivener. Identification of 

iron in hydrated cement. (in preparation) 

Dilnesa, B.Z., B. Lothenbach, E. Wieland, and K.L. Scrivener. Synthetic Fe-cement. (in 

preparation) 

Conference Proceedings 

Dilnesa, B.Z., B. Lothenbach, E. Wieland, R. Dähn, A.I. Wichser and K.L. Scrivener 

(2010) Preliminary investigation on the fate of iron during cement hydration, Proceedings 

of CONMOD 2010,Symposium on Concrete Modelling, Lausanne, Switzerland, 22 – 25 

June. 

Dilnesa B.Z., Lothenbach B., Le Saout G., Wieland E., Scrivener K.L.(2011) Fe- 

Containing Hydrates in Cementitious System. ICCC 13th Madrid, 3-8 July 

Wieland E., Dähn R., Dilnesa B.Z., Lothenbach B. (2011) Synchrotron-based micro 

spectroscopic investigations on Al, S, and Fe speciation in cementitious materials. ICCC 

13th Madrid, 3-8 July

Belay Zeleke Dilnesa - Eawag-Empa Library

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?