thermal cracking of ethylene, propylene and light hydrocarbon

Thermal cracking of ethylene, propylene and
light hydrocarbon mixtures
.?roment G.F., Van De Steene B.O.
Laboratorium voor Petrochenische Sechniek, Rijksuniversiteit,
3ent, Belgium.
;oossens A.G.
Kinetics Technology International, B.Y., The Hague,The Netherlands.
1. Introduction
There are quite a few publications dealing with the thermal
cracking of ethylene, of propylene and of their mixtures with
alkanes (1,2,3,4,5), but most of them are confined to rather
narrow ranges of the process variables. Further, there is little
agreement as to the product distribution and the kinetics of the
overall reaction, although it would seem that a gradual transition
between polymerization and decomposition processes could explain
some of the contradictions.
The thermal cracking of ethane, propane, n and i-butane has been
investigated rather intensively, but the thermal cracking of binary
and ternary nixtures has received far less attention. Although
industrial practice generally deals with mixtures, there are no
guiding lines to day for the prediction of reaction rates and
product distributions of mixtures cracking from data on the
cracking of individual components. The question v~hetlier or not
the reaction partners sioificantly interact to alter the product
distributions from those calculated from pure additivity has not
been settled yet.
The work reported in t!:e present paper aiiaed at determining
kinetics and product distributions of the thermal cracking of
ethylene, propylene, of mixtures of ethane-prooylene and of binary
.3nd ternary mixtures of ethane, propane, n- and i-butane from an
experimental program covering wide ranges of process variables
and carried out in a pilot plant. Further, it aimed at reconciling
some of the above mentioned contradictions in olefins cracking and
at derivin:: some general rules for mixtures cracking.
2. Sxperiniental program
The pilot plant 118s been described in detail by Van Damme et a1
(6) and Froinelit et a1 (7).
b . S thylene
The ethylene :7as high purity Lwade (Air Liquide CEI25) and contained
less than 0.2 wt :,1 c H
The yocess variobleb 3 \,ere 6' varied over the followin!: ranges :
Variable Ra.nge
ethylene flow rate (kg/hr) 0.5 - 3
dilution (lcz stc,?in/lcg hydrocarbon)
0.4 - 4
Xeynolds number 4,500-8,000
exit teriperature ("C) G25-850
exit $resmu-i: (atm.abs.) 1.2-2.3
pressure drop (atm.) 0.1-0.5
134

The majority of the 150 experi1:ients were grouped into 5 classes
depending upon the partial and the total pressure.
class exit pressure dilution inlet partial pressure
(atm.abs.) kg/ kg ) ( atm)
1.5 0.4
1.5 1
2.0 0.4
2.0 1
1 a5 4
0.92
0.59
1.36
0.77
0.235
B. Proleslene
The propylene was also high purity grade, containing less than
0.2 wt 5: C5+. The operating conditions were similar to those
described above, except for the inlet partial pressure, which was
varied between 0.164 and 1.16 atm.abs. and for the temperature
range, which extended from 625 to 87OOC.
~~I3inarz_an~-ternary-~~~r~~~r~~I3in-~~~~~~~
Ethane-propylene mixtures containing from 25 to 75 wt $ of propylene
were investigated under class l conditions in a temperature
range from 675 to 85OoC and with hydrocarbon flow rates ranging
from 2 to 3.6 Icg/hr. Similar conditions were chosen for the
lnvestigation of ethane-propane, n.butane-propane, n.butanei.butane
and ethane-n.butane mixtures.
Finally, the following ternary mixtures were investigated under
these conditions :
ethane propane
20. GI
41.39
30 * 63
42.23
70.61
67
37.53
11
18.40
18.45
n.butane (wt %)
11.65
20.78
58 32
39.35
10.93
3. Product Distributions.
A. Ther"al_crack~n~-o~-~~~y~~~~.
The main products of ethylene cracking are H2, CH4, C2H2, C2H6,
C3Hs, 1,3C H6 and a C +-fraction.
Figures 1 ?o 5 show tze yields of some of these products. The
results are plotted versus the ethylene conversion with the partial
pressure of the hydrocarbon and the total pressure as parameters.
Plotted in this way, there is practically no influence of temperature.
From these figures it is clear that the range of investigated
partial pressures has to be split in an area with $2H4<
0.6 atm, i.e. the class 5 experiments, and an area with pC2H >
0.6 atm, i.e. the experililents of the classes 1 to 4. This wit1 be
explained in the last section, in which the location of the border
line between polymerization and decomposition zone is discussed.
Table 1 surmiiaiuizes the influences of partial and total pressure
on the product yields. The Cz+ yield is favored by an increase in
partial and total pressure, ut there was also a stron
of the temperature : high temperatures lead to hi& c5 $ ylelds. inf1uence
At temperatures below 800°C most of the C5+ formed were heavier
thl toluene, but at the higher temperatures the product spectra
shifts towards lighter C5+ products.
135

TABLS 1
Influence of the partial pressure of ethylene and the total
pressurc on the yields.
component
H2
CH4
c2112
C2H6
C-€1 2 6
13C41-rg
-
increase of pastial pressure
increase of total
~ ~ 0 atm . 6 p>0.6 atm pressure
, I d d
r .1 none
/ i d
r no none
no none
f
f little or none slight decrease
3. Yliermal cracking of prg@zne
Figures 5 to 11 siiovr the yields of the different products in
propylene cracking.
The prinlary products of propylene cracking are H2, CH4? C2lI4,
1,x4~' ana C5+. Considering only classes 1 to 4, the initial
selectivities of these nroducts are indeDendent of partial and
total pressure. Tie foliowing values were found : %=.oj ;
CHqz.15 ; C2H - 15 . 1,3C H -.05 ; C5+=.6 .
In the class $-&ei*i,nent! the initial C5+ and 1,3c411~ selectivities
are lovrer and the initial H2 selectivity is higher, however.
Table 2 sumiiarizes the influence of partial - and total pressure
on the product spectrum.
TABLE 2
irilliuence of partial pressure of the hydrocarbon and of the total
gressure on product yields for propylene cracking.
component
C2€14
'2"6
C2H2
3'8
1C411g .
increase of partial increase of total
pressure pressure
none
f
c
t
none
f
r
I
none
no-ne
none
f
II
f
none
C . ;hemal cracking of hipzx and tepnary-hxdrocarbon mixtures
Figure 12 shows the experimental selectivitics for the different
products in ethane-propylene mixtures, versus the feed composition.
136
f
t

The selectivity of a component I in the cracking of a mixture
A-B is defined as :
mol9 of I from A + mols of I from B
YI = mols of A cracked + mois of B cracked
When there is no interaction, Froment et (7) showed that
equation 1 is identical with :
This predicted value of the selectivity will be called the non-
interaction selectivity. *
The pure additivity selectivity : yI = yI,*YA + yI,Bph 3), used
in the literature so far, is obviously a very special case of 2)
valid only for xA,jJ = x ~ , ~ ~ .
It is clear from Fig.rce 12 that 2) allows a far better prediction
of the experimental results than 3). Analogous figures were plotted
for different conversions and for all binary mixtures mentioned in
the experimental program. From all these curves, the following
rule can be deduced : "The experimental selectivities deviate from
the pure additivity lines in the same direction as the non-interaction
selectivitiesll. This means that the effect of interaction
can be predicted when the selectivities from the individual
Conponents and the relative rates of cracking are Icno~m.
The selectivities for cracking of ternary mixtures may be represented
in diagrams of the type shown in figure 13,. In this figure
the ethylene selectivity at conversions ~~,]\,~=40 F, $ ;
x ,-88 $ is plotted with respect to the ternary feed composition.
i'ttfi-ternary mixtures the pure additivity, non interaction and
experimental lines of Pig. 12 become surfaces. Here too the experimental
selectivities deviate from the pure additivity surface
in the same sense as the non-interaction surface.
4. Kinetics
The kinetics of the cracking of a component X, are derived from
the experiments by means of its continuity equation. For ethane
in a binary mixture ethane-propylene e.g., cracked in a tubular
reactor with plug flow conditions, the continuity equation may be
written :
n
E
FE,OdxE,p = A e d - m ) . ( C E , p ) dV 4)
E =expansion=
(total molar flow rate)exj-t-(mols G 2II 6 +molsC
mols C kI cracked + iiiols C3F16 cracked
2 6
6 = molar ratio propylene/etilylene
= dilation factor : mols 112O/mol hydrocarbon
In 4) the experimental temperature and total preswwe profile
are also accounted for.
Tlie kinetics can now be derived froin 4) as such or by inaking use
of the equivalent reactoz volume concept.
137

* The-es_ulval,en4_reac4or_volwne_conceEt
In this approach, the non-isothermal, non-isobaric data are first
reduced to isothermality and constant pressure. This leads to an
e uivalent reactorvolume V- rather than the physical volume V in
47,while T(z) is replaced gy a reference temperature and pt(Z) by
a reference pressure (6,7,8). The kinetic parameters of the
thermal cracking of individual alkanes were determined by means
of this concept. Figure 14 shows the Arrhenius plot of the first
order rate constants for ethane, propane, n.butane and i.butane.
The concept was also used to calculate the kinetic coefficients of
the cracking of the individual components in the mixture, assuming
first order. The integration of 4) then requires a relation
xp E'f(XE p). From the experimental data, this relation was found
in'all caees to be parabolic, so that equation 4 could be integra-
ted analytically. Figure 15 shows the value of kE p versus the
feed composition. The inhibiting effect of propylkne on the ethane
rate coefficient is very pronounced. The same procedure was
followed to calculate the individual rate coefficients for all the
components of the different feed mixtures.
Table 3 summarizes the relative influences, caused by cocrackinr.
TABLE 3
Effect of the addition of various components on the rate coeffi-
cient for the cracking of ethane, propane, n.butane, i.butane
and propylene.
I I Effect on rate coefficient
Bddi t ion
I Of
C&
',
C3HB
nC4H1 0
iC4H,0
C3H6
I kC2H6 kC3H8 I knC4H10 1 kiC4H10 1 kC3H6
- r r no data .r
4 - .1 no data f
.1 .1 .I no data
no data no data f - no data
J no data no data -
1
The individual rate coefficients in the ternary mixture ethane-
propane-n.butane were calculated in an analogous way. Figure 16
shows the effect on the rate coefficient for ethane cracking of
the addition of one or two components in various ratios. Analogous
diagrams could be shown for kp Ti and kN M. All three of them
clearly illustrate the effect bn the rate coefficients of inter-
action between reaction partners.
The equivalent reactorvolume concept requires an initial guess of
the activation energy, valid over the whole range of investigated
temperatures. This is no problem with alkanes but, as will be
discussed in the next section, the ethylene and propylene cracking
experiments cover a transition zone in which polymerization
processes with activation energies of 2 35 kcal/mol and decompo-
sition processes with activation energies of 2 70 kcal/mol
simultaneously occur. In such cases it is preferable to resort
138
N

to the non-isothermal non-isobaric approach used by Van Damme et
a1 (6). Equation
‘
4, with the experimental tenperature and pressure
Profile included, was integrate numerically and the residual sum
Of squares 9 (Xexp-xcalculated
)$ was minimized with A, E and n as
Parmeters, using Liarquardt’s search routine (22). The results are
shown in Table 4,5,6 and 7 respectively, from which it follows that
there is a strong influence of temperature, partial and total
Pressure on the kinetic parameters in both ethylene and Propylene
cracking.
TABm 4
Influence of the temperature on the kinetic parameters of the
overall ethylene disappearance.
1.61 2 0.025
1.46 + 0.122
1.43 z 0.042
800 1.39 10 1.19 + 0.034
825 1.22 0.042
~~ ~~~ ~
TABU 5
Influence of the temperature on the kinetic par-ametcrs of
propylene cracking
9.42 1012
4.43 IO1,
800
82 5 2.99 10
TABLE 6
Influence of inlet partial pressure and total pressure on the
kinetic parameters of ethylene cracking.
I A 1 E(kcal/mol) 1 n t
Class 1 I 2.48 10” 54.220 f 1.260 1.29 2 0.046 !
Class 2 1 .I1 10;; 81.910 f 1.340 1.54 0.01
Class 3 54.430 f 2.170 1.27 2 0.027
Class 4 5.47 IOl3 66.400 2 .211 1.35 2 0.023
Class 5 I .50 1015 72.840 2 .i91 1.36 2 0.018
iA ll classes 2.51 IOl3 TkELE 60.400 7 2 .975 1.30 2 0.027
Influence of inlet partial pressure ana total pressure on the
kinetic parameters of propylene cracking.
E( Iccal/mol
n
1 1 A
Class 2
Class 3
class 4 1.89 10
Class 5 5.18
51.320 2 2.600
66.020 2 28.000
56.690 2 2.900
63.960 2 1.080
47.710 2 2.150
50.650 1 A00
139
1.02 2 0.035
1.02 2 0.37
1.05 2 0.028
1.18 + 0.040
1.15 0.063
1.08 2 0.037

5. Decomposition and pol.wnerization zones in olefines crackins
i:. Propylene cracking
In the thermal cracking of propylene, polyierization and decomposition
processes are in competition (1,2,3,12,13,14). Low tenperatures
and high partial pressures favor polynerization, high tenperatures
and lovr partial pressures fa-ieu*.e 17.
To permit a direct cornp.rison with the literature, the propyhene
partial pressure is expressed in rm Me. From Anano's border line,
it follovis that the present results woulci be conpletely located
in the golpierization zone aid indeed no allene is found. However,
the order of the global disappezrance reaction is not 5/2 but
varies froiii 1/2 to 1 as the teriperature is increased, while the
activation enerfiie varj.es from 31; tn 66 kc~.l/~?:&!.- C!%C r;-ggcr;ts
that the present dat.?. would be located in the trznsition area and
this in turn requires the border line to be shifted towards the
left of t'ne figure. Such a shift could result from a decomposition
of allyl into acetylene instead of allene :
An isothermal, unsteady state radical siniulation of the propylene
experiments (10) led to a frequenc factor of 1.108 and an activation
energy of 32.4 kcal/mol for 77. These values are j?lausj.bl.e,
when compared with Allarals data (9) for the decomposition of c
radicals into C 11 :',ken reaction 7 instead of reaction 6 accoukts
for the/t-behav$o$'of the allylradical, the border line between the
two areas io, calculated from : IC C I1 ) = I: . This is the full
line on Figure 17. Both Kunugi's5t2j Bnd O U ~ experiments are now
located in the transition zone. Tile experiments of Laidler (1)
and Ingold (12) lie in the polyuerization area and these of Szwarc:
(13) and Salcakibarz (14) in the decomposition zone. This is in
agreement with the fact that neither Laidler nor Ingold found
acetylene, while Kunugi and Sakakibara d.id. The results of Szviarc
are doubtful bccause of his rudimentary analytical technique.
At 1200°C and higher, the decomposition of allyl into alleEe
becomes more and more important, became of the high activation
encrm, so that Sakalcibara, whose work is lrlainly situated in the
temperature range 12000c-1400°c, found inore allene and iiiethyl-
140

acetylene then acetylene.
B. :+Xl_gne crac+g
In CoWete amlogy with propylene cracking, the two major initial
reactions are :
%
C2fijo+C 2FI 4 -+ C~EI?O -+ c 4M 6 +HO
~ ( rno1-lsec-l) 1
7 107
~(~ccal/mol)
7.3 8)
c 2H 3 o+i:l--tC k 9 2H 2 +Ho+l,i 5 loq1 31.5 9)
The vinylradical acts as a p radical. in 8) and as a
9). The kinetic parameters of Z;'nese reactions were
extensive literature survey ( 1 1,15,16,17).
The border line between the two zrea's is now given hy :
k8(C21130)(C2H4) = k (C H O )(l.i)
9 2 3
This line is the flit1 line in Figure 18. ';;itti a second orcier
initiation 1i.ke 2C H 422I.i- "+C21150 and a teir;iinatj.on involving
two vinylradicals ;i2&Ii3 -+ .? , !C4Hg), an order of 2 is predicted
for Conditions in the polymerization zone and a.n order of 1 for
conditions in the dccoi~iposition zone.
Si1:lon and Iijaclc (1 5) and Silcocks (1 8) observed seconci order
kinetics, while l.:olera and Stuhbs (16) found second order 1cineti.c.s
in the region 5 of Fi,mre 18 and first order kinetics in region TI,
in agreement with tie proposed border line. R.m.uzi's data (11 )
ivhi.cli arc located in the transition area lead to an order of 312
for tiic ~:lolial. disa?xarrr.nc:e reaction.
The pilot plant experiinento cover the transition zone. The order
!ea.s shown in 'I'al:~lea 4 and 6 to evolve fron! 1 .6 to 1 .2 as the
temperature is increased and the activation eiiergy from 40 to
2.boIit 75 kce?.l/iilol.
I '
J?'Iie pol.yilir~i.iarttion can also lie of a J!lOleCILlar nature :
.A(I mol-' sec-l) :,'(1cca,l/rnQl) iter
C,ii +I ,'jC I L ---> polymer
r 4 4
3 107 27 19 IC)
Th.e ratio of the ra.c?ica.l. reaction and molecular condensation rates
was ca3.cula.ted fron 2.n isotlierraal. non cteady state radi.ca1 siimlation
of the ethylene experiinents by Sunderan & PToment (IO), to be
71 at 80GV and 1.; conversion, 15 at OOO°C and 3:~ conversion ana
10 at 825°C and 8 ;i conversion.
It follows that molecular condensation cannot be neglected a.t
8CO°C and higher. This explains vlhy the experi.inenta1 Cg+ frsction
was SO stroiicly correla.ted aitli the temperature. In propylene
crrtcitiiie the ratios of the rates of the radical po1yl;ierisation and
the moleculcir conderisation at OGCV would Le of the ordcr of IC00
because the allyl rzdical is hich and butadiene is low.
It is clear also froiii Piwre 16 for ethylene ax1 from Figure 17
for propylene that tile influence of the pnr-tial and total pressure
on fne order anci activation encrt27 vrill be inore or less pronounced
depending u?on the distance frnrn the border line. The higher acti-
vation enerzies Given in lnbles 6 and 7 for the conditions of the
classes 2, 4 and, 5 may be related to t
Pinally, il; ShOii.ld be added that even
less tila (1.2 wt ,:, in the etiiyleiie feed, the initiation reaction
C 11 i2CI11° cannot lje ne!:lecterl, as concluded already by
2 6
1:m~ugi (I I ) unc~ BY yowell and liartin (20) .
141

From the mechanistic point of view the initiation then becomes of
first order and the order of the global disappearance should vary
between 3/2 and 1/2. On the other hand, the molecular reactions
vtill increase the overall order. All these trends substantiate the
-rariation of the order from 1.6 to 1.2 observed in the present viorl;.
k
n
Pt
R
r
T
VE
X
Y
z
Y
b
E
Y
Notation
-1
frequency factor sec or 1 rno1-lsec-l
activation energy kc a1 /mol
concentration IUOl/l
concentration, total hydro-
carbons
molar flow rate of the craclced
coaiponent at the inlet
rate Coefficient
order of reaction
lJOl/l
mollsec
sec-’ or 1 mol-’ sec-’
total pressure atm.abs.
gas constant
kc a1 /mol ‘C
rate of reaction mol/l sec
temperature OK or ‘C
equivalent reactor volume 1
conversion
selectivity mol/nol
tube length rn
mol ratio mol/mol
dilution rat io
inol steam/mol hydrocarbor.
expansion factor mol products/mol hydrocarbon
cracked
mol fraction mol /mol
ethane
progylene
mixture
initial value
total
mean
Subscrints
Table captions
Table 1 : Xthylene cracking. Influence of the total pressure and
the partial pressure of ethylene on the yields of the
dj ffercnt groducts.
Table 2 : Propylene craclrint;. Influence of the total pressure and
the pnrtial pressure of propylene on the yields of the
different products.
Table 3 : Effect of the addition of various colnponcnts on the rate
coefficient €or the cracking of ethane, propane,
n-butane , i -butrine and propylene.
142
rn
m
m

w
w
m
H
m
m
m
H
m
m
m
H
Table 4 : Influence of temperature on the kinetic parameters of the
overall ethylene disappearance.
Table 5 : Influence of temperature on the kinetic parameters of the
overall propylene disappearance.
Table 6 : Influence of inlet partial pressure and total pressure on
the kinetic parameters of ethylene cracking.
Table 7 : Influence of inlet partial pressure and total pressure on
the kinetic parameters of propylene cracking.
Figure captions
Figure 1 : Ethylene cracking : hydrogen yield
Figure 2 : Zthylene cracking : methane yield
Figure 3 : Ethylene cracking : acetylene yield
Figure 4 : Ethylene cracking : propylene yield
Figure 5 : Ethylene cracking : butadiene yield
Figure 6 : Propylene cracking : hydrogen yield
Fiwe 7 : Propylene cracking : methane yield
Figure 8 : Propylene cracking : acetylene yield
Figure 9 : Propylene cracking : ethylene yield
Figure 10 : Propylene cracking : butadiene yield
Fi,gre 11 : Propylene cracking : C5+ yield
Figure 12 : Selectivity of H2, CH4 and C2H4 versus mol fraction
propylene in the mixture
Figure I3 : Selectivity of C H versus feed composition in the
ternary mixture &hane-propane-n.butane
- - lines on non-interaction surface ; + non inter-
action points ; . experimental
Figure 14 : Arrhenius plot of rate coefficients for the cracking of
light hydrocarbons
Figure 15 : Rate coefficient for the cracking of ethane in an
ethane-propylene mixture versus feed composition
Figure 16 : Rate coefficient for the cracking of ethane in ternary
mixtures of ethane-propane and n-butane
Figure 17 : Polymerization and decomposition zones in propylene
cracking
Figure 18 : Polymerization and decomposition zones in ethylene
cracking.
References
1) Laidler K.J., Vojciechowski B.TI., Proc.Roy;Soc. (London),
A259, 257 (1960)
2) Kunugi T., Sakai T., Soma K. and Sasaki Y., 1nd.Eng.Chem.
* Fundam., v01.9, NO3 (1970).
3) Aman0 A., Uchujana M., J.Phys.Chem., 68, 1133 (1964).
4) s&ai T., Soma K., Sasaki Y., Tominaga H. and Kunugi T.,
Refining Petroleum for Chemicals, 68 (1 970).
5) Kallend A.S., Wnell J.H., Shurlock B.C., Proc.Roy.Soc.
(London), A300, 120 (1967).
6) van Damme P.S., Narayanan S. and Froment G.F., A.1.Ch.E.
Journal, 1701.21, N06, 1065 (1975).
14 3

7) Froment G.F., Van de Steene B.O., Van Dame Y., marayanan S.
and Goossens A.G., 1nd.Eng.Chem.Process Des.&
Develop. (to be published)
8) Froriient G.F., Pycke Ii., Goethals G., Chem.Eng.Sci.,l3, 173,
180 (1961)
9) Allara D.L., i~ co!ripilation of kinetic parameters for the
thcrinal degradation of n-alkane molecules
IO) Sundaram id., Froment G.F., (to be published)
11 ) Icunugi T., Sakai T., Soma K., Sasaki Y., 1nd.Eng.Chem.Fundam.
Vo1.8, 374 (1969)
12) Ingold K.U., Stubbs F.J., J.Chem.Soc., 1749 (1951)
13) Szwarc U., J.Chem.Phys., I%, 234 (1949)
14) Sakakibara Y., Bull.Chem.Soc. Japan, 37, 1262 (1964).
15) Simon ki., Back X.H., Can.J.Chern., 47, 251 (1969)
16) Xolera M.J., Stubbs F.J., J.Chem.Soc., 301 (1952)
17) Benson S.Y., Hougen G.R., J.Phys.Chem., 71, 1735 (1967)
16) Silcocks C.G., Eroc.Roy.Soc., A233, 465 (1955)
.is) Bowley D., Steiner H., diss.!L'rans.Faraday SOC., 10,198(1951)
2O)Towell G .D :, 1,lartin J. J. , L. I ,Ch.3. Journal, vol. 7 ,1T04, 693 ( 1961 )
:?I) Froriient G.P., Van De Steene B.O. and Goossens A.G.
(to be published)
22) Marquardt D.?., Soc.Ind.ilp-ol.l,iath.J., 11, 431 (1963).
144
W
m
H
B,
W
m
W
m
W
m
m
m

N
I
.2
CLASS 1 0
CLASS 2 A
CLASS3 -k
CLASS 4 X
CLASS 5 0
Figure 1.
C
0 10 20
CONVERSION %
THERMAL CRACKING OF ETHYLENE
145
X
30

.l 2
I
u
1
CLASS 1 0
CLASS 2 A
CLASS 3 +
CLASS L X
CLASS 5 0
Figure 2.
0 10 20 30
CONVERSION O h
THERMAL CRACKING OF ETHYLENE
146
+

-
8"
N
0
m
Y
0 .
a Y
I
N
N
0
2
I 1
0
J
w
>
0
CLASS 1 0
CLASS 2 A
CLASS 3 +
CLASS L X
CLASS 5 0
Figure 3.
I I I
10. 20. 30
CON V E R S I 0 N %
THERMAL CRACKING OF ETHYLENE
147
,

N
u
.
I?
Y -
CLASS 1 0
CLASS 2 A
CLASS 3 +
CLASS L X
CLASS 5 0
Figure 4.
0. 10 20 30
CON V ER S IO N %
THERMAL CRACKING OF ETHYLENE H
148
H

I
.I
I
6
N
0
0 L
Y
0
0
- .
0
Y
u3
I
.l
0
0
_1
W
-
t-
2
0
CLASS 1 0
CLASS 2 A
CLASS 3 +
CLASS I x
CLASS 5 0
Figure 5.
X+
0
+ +
I 1 I I
0 10 20 30
CONVERSION '/o
THERMAL CRACKING OF ETHYLENE
149

CLASS 1
CLASS 2
CLASS 3
CLASS L
CLASS 5
Figure 6.
0
a
+
X
0
CONVERSION (% 1
THERMAL CRACKING OF PROPYLENE
150

.I
.'
B'
.
.I
.' B'
M'
. U:
.'
m:
E
m
.'
rn
.: W
.'
.,
.'
.I
25.
29
CLASS
CLASS
CLASS
CLASS
CLACS
1
2
3
L,
5
Figure 7.
0 20 LO 60 80 100
CONVERSION ( %)
THERMAL CRACKING OF PROPYLENE
151

CLA5S 1 0
CLASS 2 A
I,LA-\sS 3 +
CLASS L X
CLAZS 5 0
Figure 8.
2 i) LO 60 80 1oc
CONVERSION l%l
THERMAL CRACKING OF PROPYLENE
152
a
a
a
a
I
a
rn
PI
H
IC'
H
a'
H
I

30
20
10
0
Figure 9.
CLASS 1 0
CLASS 2 a 0
CLASS 3 +
CLASS 4 X
CLASS 5 0
20 LO 60 80 100
CONVERSION I % )
THERMAL CRACKING OF PROPYLENE
153

CLASS 1 0
CLASS 2 a
CLASS 3 +
CLASS L X
CLASS 5 0
Figure 10.
X
0 20. LO 60. 80 100
0
CONVERSION 1%)
THERMAL CRACKING OF PROPYLENE
154

CLASS 1 0
CLASS 2 fl
CLASS 3 +
CLASS L X
CLASS5 0
Figure 11.
20 40 60 80 100
CONVERSION [%I
THERMAL CRACKING OF PROPYLENE
155

SE
1
.a
.6
.4
.2
a
ECTl V IT Y
Figure 12.
ETHANE-PROPYLENE MIXTURES
CLASS 1
X z 49% X = 7Q%
E P
156

Figure 13.
TERNARY MIXTURES
ETHANE -PROPANE-N.BUTANE
CLASS 1 XE,M= 40% Xp,M= 73% XN,M= 88%
SELECTiVITY
157

-Nb
3
-1
2
1
0
-2
(
Figure 14.
I I I *
I I I 1 I I I
T(OC)
*
8 50 800 750 7 h
3 0.90 0.92 0.94 0.96 a98 1 1.02 lOOO/TR
L FIG L
158

k
E,P'
( 8
Figure 15.
ETHANE -PROPYLENE MIXTURES
CLASS 1
T =825OC
T = 800 OC
T 775OC e
O l
0
T = 75OoC - w
I
1 1 molsC3H6
mol C2H6
159
0
0

I
Figure 16.
TERNARY MIXTURES
N.BUTANE - PROPANE - ETHANE
CLASS 1 T = 8OO0C
.kEJMl
-
.3
R

*:
600 800 1000 T ("C)
161

2 ,
2 D ECOM PO S I T ION ZONE
.. .. .
1 : PRESENT WORK
2: SIMON ,
3+3': MOLERA
4: KUNUGI
5: TANlEWSKl
A: SILCOCKS
THERMAL CRACKING -
OF ETHYLENE
I I I I -+
600
162
8 00 lobo T ("C)