General Principles of HPLC Method Development

General Principles 

of HPLC Method 

Development 

Philip J. Koerner, Ph.D. 

Thailand 

September 2013 

Part 1. General Chromatographic Theory 

Part 2. The Stationary Phase: 

An Overview of HPLC Media 

Part 3. Role of the Mobile Phase in Selectivity

The Liquid Chromatographic Process 

3 

Column Chromatography: 

The Beginning of Liquid 

Chromatography 

Empty 

Column 

Adsorbent 

particles 

added 

Sample is 

loaded onto 

the top of 

the column 

Solvent is 

added to the 

top of the 

column 

Separation 

occurs as 

the bands 

move down 

the column 

4 

Mikhail Tsvet, 1872-1919

Basis of 

Chromatographic Separation 

Separation of compounds by HPLC depends on the interaction of 

analyte molecules with the stationary phase and the mobile phase. 

Mobile Phase 

Stationary Phase 

5 

The Liquid 

Chromatographic Process 

Analytes 

Mobile Phase 


6

The Liquid 


Analytes 

Mobile Phase 


7 

The Liquid 


Polar/Aqueous 

Mobile Phase 

H 

O 

N 

O H 

H 

N 

N 

H 

O H 

N 

O H 

H 

N 

O H 

H 

H 

Benzo(a)pyrene 

Ethyl sulfate 

Non-Polar Stationary 

Phase (e.g. C18) 

8

In any separation, almost never get a pure, single mode of separation. In RP, 

performance will be dictated by mixture of: 

1. Hydrophobic interactions 

2. Polar interactions 

3. Ionic interactions 

Mechanisms of Interaction 

In RP Chromatography 

Method Development = modulating stationary phase and mobile phase 

conditions to optimize these interactions and achieve a specific separation 

goal. 

OH 

Tapentadol 

CH 3 

CH 3 

N 

9 

CH 3 

CH 3 

Hydrophobic Interactions 

Hydrophobic 

C H 3 

C H 3 C H 3 C 

OH 

• Weak, transient interactions 

between a non-polar stationary 

phase and the molecules 

O 

C H 3 

H 3 

C H 3 

• Hydrophobic & van Der 

Waals interactions 

CH 3 

CH 3 

N 

H3C 

CH 3 

• Retention will be predicted by 

Log P values 

S i H 3 C S i 

H C S i 

H 3 C 

S i 

C H 3 

O C H 3 

C H 3 

O 3 

O - O H C H O H 

O O 3 

O 

C H 3 

S i 

C H 3 

C H 3 

S i S i S i S i S i S i S i 

O O O O O O 

H O O H H O O H O H O H H O 

O 

S i S i S i 

O O O H 

O H O H O H 

10

Polar Interactions 

C H 3 

C H 3 C H 3 C 

Polar 

• Interactions between polar 

functions groups of analyte 

and residual silanols or polar 

groups on media 

O 

C H 3 

H 3 

C H 3 

CH 

H C 3 

3 

N 

H 3 C 

H 3 C 

• Hydrogen bonding, dipoledipole 

interactions 

• Relatively weak and transient 

HO 

S i H 3 C S i 

H C S i 

H 3 C 

S i 

C H 3 

O C H 3 

C H 3 

O 3 


O O 3 

O 

C H 3 

S i 

C H 3 

C H 3 


O O O O O O 


O 

S i S i S i 

O O O H 

O H O H O H 

11 

Ion-exchange Interactions 

• Interactions between 

ionizable functional groups on 

analyte and counter-charged 

moiety on stationary phase 

Ion-Exchange 

• Ion-exchange 

C H 3 

C H 3 C H 3 C 

OH 

• Strong, slow interactions 

O 

C H 3 

H 3 

C H 3 

CH 3 

H 3 C 

H N 

+ 

H 

S i S i 

3 C 

H C 

CH 3 

3 H C S i 

H 3 C 

S i 

C H 3 

O C H 3 

C H 3 

O 3 


O O 3 

O 

C H 3 

S i 

C H 3 

C H 3 


O O O O O O 


O 

S i S i S i 

O O O H 

O H O H O H 

12

Chromatographic Measurements 

13 

Chromatographic 

Measurements 

14 

k’ Asym N α

Void Volume 

Analytes which do not interact with the adsorbent elute from the column in a 

volume equal to the void volume in the column. The void volume of a 

column is the amount of mobile phase in the column between the adsorbent 

particles and in the pores of the porous adsorbent particles. 

Mobile phase occupies the space 

between the particles or the 

interstitial volume. 

Mobile phase fills the pores of the 

porous adsorbent particles. 

15 

Void Volume 

A compound which does not interact with the adsorbent at all elutes at 

what is termed the void volume or the solvent front. The time that it 

takes for non-retained components to elute is the void time or t 0 

. 

• void volume 

• solvent front 

• t 0 

16

Retention Factor (k’) 

The retention factor of the eluting compound is its elution volume (time) 

relative to the elution volume (time) of an unretained compound. The k’ 

value for a given analytes will be determined by its relative affinity for the 

stationary phase and mobile phase. 

Retention factor (k’)= 

t R – t 0 

t 0 

t 0 

17 


For any given analyte, the k’ value will be most readily modified by changing 

the % of strong mobile phase (e.g. methanol or ACN). 

Example: The Separation of Steroids: 

Column used: Phenomenex Luna C18(2) 150 x 4.6 mm 

18

Retention Factor (k’): 

Effect of Changing % ACN 

19 

Peak Asymmetry (Asym) 

The asymmetry value (Asym) for a peak is a measure of the divergence of 

the peak from a perfect Gaussian peak shape. Peaks with asymmetry 

values > 1.0 are said to be “tailing”, those with asymmetry values < 1.0 are 

said to be “fronting”. 

Asymmetrical peaks can be 

attributed to a number of factors: 

(1) Strong secondary interactions (e.g. 

ionic interactions between basic 

compounds and residual silanols) 

(2) Sample overloading 

(3) Sample solvent incompatibility 

(4) Poor packing 

20

Peak Tailing due to 

Secondary Interactions 

Classical peak tailing in reversed-phase methods is most commonly caused 

by strong ionic interactions between basic analytes and residual silanols 

on the surface of the silica. 

C H 3 

C H 3 C H 3 

C 

O 

C H 3 

H 3 

C H 3 

Ion-Exchange 

O H 

C H 3 

H 3 C 

H 3 C 

H N 

+ 

C H 3 

S i H S i 3 C 

S i S i 

O C H 3 

C H 

O 3 

O - O H H 3 C 

H C C H 3 3 

C H O H 

O O 3 

O 

C H 3 

S i 

C H 3 

C H 3 


O O O O O O 


O 

S i S i S i 

O O O H 

O H O H O H 

21 



Example: The Separation of Tricyclic Antidepressants: 

Column used: Kinetex 2.6µ XB-C18 50 x 2.1 mm 

Brand H 2.7µ C18 50 x 2.1mm 

Amitriptyline (pKa 9.4) Nortriptyline (pKa 9.7) Protriptyline (pKa 8.2) 

22

30 

25 

20 

15 

10 

5 

50 

40 

30 

20 

10 

100 

80 

60 

40 

20 

DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000006.D) 

12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 


12.5 13 13.5 14 14.5 15 15.5 16 16.5 


12.5 13 13.5 14 14.5 15 15.5 16 16.5 

min 

min 

min 



Brand H 2.7µ C18 50x2.1mm 

Kinetex 2.6µ XB-C18 50x2.1mm 

XIC of +MRM (8 pairs): 264.2/191.1 Da ID: Protrip from Sample 28 (Halo C18-50x2, 2.7 um, MeOH:0.1% ... XIC of +MRM (8 pairs): 278.2/191.2 Max. Da 1.1e5 ID: Amitrip cps. from Sample 7 (TCA-Kin-XB-C18, 50x2, 2.6, MeOH 0.... 

1 5 

5 

5 

4 

1 

1 

4 

Peak tailing 

Max. 1.2e5 cps. 

6 

6 

2 3 

7 8 

2 3 

7 8 

3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 

Time, min 

Time, min 

XIC of +MRM (8 pairs): 264.2/191.1 Da ID: Protrip from Sample 28 (Halo C18-50x2, 2.7 um, MeOH:0.1% ... XIC of +MRM (8 pairs): 264.2/191.1 Max. Da 1.1e5 ID: Protrip cps. from Sample 13 (Kinetex XB-C18-50x2, 2.6 um, MeO... 

Max. 1.2e5 cps. 

4 

4 

6 

6 

3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 

Time, min 

3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 

Time, min 

23 

Mobile phase: A = 0.1% Formic acid in water, B = 100% Methanol, 

Gradient: 15 to 60% B in 5 min, hold for 1 min 

Flow rate: 400 µL/min 

1. Amoxapine 

2. Imipramine 

3. Desipramine 

4. Protriptyline* 

5. Amitriptyline 

6. Nortriptyline* 

7. Clomipramine 

8. Norclomipramine 


Sample Overloading 

Strongly basic analytes are very sensitive to sample loading, and will display 

peak tailing effects as a function of increasing loading (µg on column): 

mAU0 

2.5 µg on column; 

USP Tailing = 1.17 

14.208 

pKa 9.7 

5 µg on column; 


14.169 

mAU0 

12 

12 µg on column; 


14.157 

mAU0 

12 

24

1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 

25 

min 

Peak Fronting due to 

Sample Overloading 

Neutral and acidic compounds will typical show peak fronting when the 

column is overloaded. 

Detector saturation! 

DAD1 A, Sig=240,10 Ref=off (MT042211\DJGSK 2011-04-22 09-12-57\MUPIROCIN000001.D) 

2250 

2.495 

2000 

1750 

1500 

mAU 

Fronting due to overload 

1250 

1000 

750 

500 

250 

1.905 

1.745 0 

2.145 

Peak Fronting due to 

Sample Solvent Effects 

Peak fronting can also be due to sample solvent effects: 

(1) Sample insolubility 

(2) Sample solvent is too strong: 

Sample in 50% Methanol 

7.0e4 

6.5e4 

6.0e4 

5.5e4 

5.0e4 

4.5e4 

Hydromorphone 

Sample in 100% Methanol 

2.2e5 

2.0e5 

1.8e5 

1.6e5 

Breakthrough! 

1.4e5 

4.0e4 

3.5e4 

3.0e4 

1.2e5 

1.0e5 

0.24 

Fronting 

2.5e4 

8.0e4 

2.0e4 

1.5e4 

1.0e4 

5000.0 

0.0 

Morphine 

1.06 

1.77 

Norhydrocodone 

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 

6.0e4 

4.0e4 

2.0e4 

0.0 

1.36 

0.97 

0.51 1.71 

1.79 

1.85 2.15 3.213.36 3.97 

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 

26

Selectivity (α) 

Selectivity is a measure of the difference in the interactions of two compounds 

with the stationary phase. Selectivity is a function of both the stationary phase 

and the mobile phase. 

α = k 2 /k 1 

27 


The choice of stationary phase will often have a dramatic effect on the 

selectivity of analytes. 

O 

CH 3 

OH 

CH 3 

H 

H 

HO 

H 

H 

HO 

H 

H 

Estrone 

Estradiol 

m A U 

1 7 5 

C18 

Column 

1+2 

1 5 0 

1 2 5 

1 0 0 

α = 1.0 

7 5 

5 0 

2 5 

0 

1 2 3 4 5 

m 

i n 

m A U 

1 0 0 

Phenyl Column 1 

8 0 

6 0 

2 

α = 2.3 

4 0 

2 0 

0 

28 

1 2 3 4 5 

m 

i n


But mobile phase is also a very powerful tool in modulating selectivity. 

Column: Gemini 5 µm C6-Phenyl 

150 x 4.6mm 

Mobile phase: 20mM KH2PO4, pH 2.5; 

% organic as noted 

Flow rate: 1.0 mL/min 

Detection: UV at 220nm 

35% Methanol 

1. Saccharin 

2. p-Hydroxybenzoic Acid 

3. Sorbic Acid 

4. Dehydroacetic Acid 

5. Methylparaben 

20% Acetonitrile 

29 

Column Efficiency (N) 

The amount of band (peak) broadening or dispersion that occurs in the column 

is measured by calculating the column efficiency (N) expressed as the 

number of theoretical plates in the column: 

tR 

W 1/2 

• Columns that cause a lot of peak 

broadening have few theoretical 

plates. 

• Columns that produce very narrow 

peaks (little peak broadening) have 

a large number of theoretical plates. 

W 

30 

Peak Width (W)

Column Efficiency (N) is a 

Function of Particle Size 

Efficiency of a well-packed column will be a function of several factors, one of 

which will be particle size. As particle size decreases, efficiency will increase. 

In addition, back-pressure also increases as particle size decreases. 

10 µm 

5 µm 

3 µm 

Core-Shell 

sub-2 µm 

50,000 P/m 

100,000 P/m 

150,000 P/m 

300,000 P/m 

300,000 P/m 

Efficiency 

Back-Pressure 

31 

The Core-Shell Advantage 

Fully Porous Particles 

Columns packed with core-shell particles will deliver 

significantly higher efficiency (N) than columns packed with 

fully-porous particles of the same diameter.* 

Fully Porous Particles 

w 1/2 

Injected 

Sample 

Band 

* Gritti et al., Journal of Chromatography A, 1217 (2010) 1589 

t R 

6


Core-Shell Particles 




Kinetex Core-Shell Particles 

w 1/2 

5 6 

t R 



Comparison 




Fully Porous 3 µm C18; 150 x 4.6 mm 

N = 166,500 p/m 

Kinetex 2.6 µm C18; 150 x 4.6 mm 

N = 295,340 p/m 

78% Increase 

in Efficiency! 


34


Efficiency vs. Diameter 




450 

Efficiency( p/m) 

400 

350 

300 

250 

200 

150 

100 

50 

Fully Porous 

Core-Shell 

0 

5 3.5/3.6 2.5/2.6 1.7 1.3 

Particle Diameter (µm) 

*Farcas et al., HPLC 2012 Conference 

The van Deemter Equation 

The three principle factors that cause band broadening and a decrease in 

column efficiency are described by the van Deemter equation: 

e 

e 

* 

Simplified version: 

H = 

A · d p + B/µ + C · d e 

2 · µ 

Particle size 

Particle size 

Linear velocity (flow rate) 

Mass Transfer 

36 

Linear velocity (flow rate) 

Mobile phase viscosity


A-term 

H = 

A · d p + B/u + C · d e 

2 · u 

Eddy Diffusion 

Multi-path Effect 

w 1/2 

t R 

6 


B-term 

H = 

A · d p + B/u + C · d e 

2 · u 

Longitudinal diffusion 

w 1/2 

t R 

6


Fully Porous C-term 

H = 

A · d p + B/µ + C · d e 

2 · µ 

Mass Transfer (Kinetics) 

Dispersion due to resistance to mass transfer 

A 

B 

39 


Core-Shell C-term 

H = 

A · d p + B/µ + C · d e 

2 · µ 

Dispersion due to resistance to mass transfer 

Mass Transfer (Kinetics) 

A 

B


h vs. v Comparison 

H = 

A · d p + B/µ + C · d e 

2 · µ 

Reduced plate height h 

15 

10 

5 

4.6 x 100 mm Luna-C 18 

(2) 

Eddy dispersion 


Solid-liquid mass transfer 

Reduced plate height h 

15 

10 

5 

4.6 x 100 mm Kinetex-C 18 

Eddy dispersion 


Solid-liquid mass transfer 

0 

0 5 10 15 20 

Reduced velocity ν 

0 

0 5 10 15 20 

Reduced velocity ν 

*Gritti and Guiochon, 2012. LC-GC 30(7) 586-595. 

Effect of Column Length 

on Efficiency 

For a given particle size, column efficiency will be directly proportional to the 

length of the column. However, pressure and elution time will also be 

directly proportional to the column length. 

5cm 

10cm 

15cm 

25cm 

5,000 Plates 

8 min 

50 Bar 

10,000 Plates 

16 min 

100 Bar 

15,000 Plates 

24 min 

150 Bar 

25,000 Plates 

40 min 

250 Bar 

42

Balancing Column Length 

and Particle Size 

Column 

Length 

(mm) 

Efficiency dp 

5µm 

250 25,000 

150 15,000 

100 10,000 

50 5,000 

43 



Column 

Length 

(mm) 

Efficiency dp 

5µm 

Efficiency dp 

3µm 

250 25,000 37,500 

150 15,000 22,500 

100 10,000 15,000 

50 5,000 7,500 

44



Column 

Length 

(mm) 

Efficiency dp 

5µm 

Efficiency dp 

3µm 

Efficiency 

sub-2µm / 

Core-shell 

250 25,000 37,500 

150 15,000 22,500 45,000 

100 10,000 15,000 30,000 

50 5,000 7,500 15,000 

45 



Column 

Length 

(mm) 

Efficiency dp 

5µm 

Efficiency dp 

3µm 

Efficiency 

sub-2µm / 

Core-shell 

%Reduction in 

Analysis Time 

250 25,000 37,500 

150 15,000 22,500 45,000 33 

100 10,000 15,000 30,000 60 

50 5,000 7,500 15,000 80 

46 

Use shorter columns packed with smaller particles 

to reduce analysis time while maintaining/improving 

efficiency!!

Effect of Flow Rate on 

Efficiency 

Effect of Flow Rate on Column Efficiency (100x4.6mm) 

35000 

30000 

Core-Shell ~2 mL/min 

Core-Shell 2.6 

Luna 3u 

N (Plates/column) 

25000 

20000 

15000 

3µ ~1.5 mL/min 

Luna 5u 

10000 

5000 

5µ ~1 mL/min 

47 

0 

0 0.5 1 1.5 2 2.5 3 3.5 

Flow rate (ml/min) 

Quick Review 

The chromatographic measurements that we have discussed so far will all play 

a significant role in method development. 

1. Retention factor (k’) – retention of analyte relative to void t 0 

• Controlled by modulating % strong mobile phase 

2. Peak asymmetry – peak shape (fronting, tailing, symmetrical) 

• Result of secondary interactions (e.g. Ionic in RP mode) 

• Sensitive to sample loading & sample solvent effects 

3. Selectivity (α) – difference in the k’ of two analytes 

• Will be determined by mobile phase composition and nature of 

stationary phase 

4. Efficiency (N) – function of peak width and retention 

• Determined by particle size, column length, flow rate 

• Column packing will affect efficiency (vendor) 

48

Any Questions? 

49 

Resolution: The Goal of Chromatography 

50

Resolution: The Goal of 

Liquid Chromatography 

The goal and the purpose of 

liquid chromatography is to 

resolve the individual 

components of a sample from 

each other so that they may be 

identified and/or quantitated. 

51 



Resolution is a measure of how well two peaks are 

separated from each other. 

It is calculated as the difference in retention time of two eluting peaks divided by 

the average width of the two peaks at the baseline. 

R (resolution) = RT B - RT A / .5 (width A + width B ) 

52



α 

(1) Retention time for the two 

peaks will be a function of 

retention factor (k’). 

k’ 

(2) The selectivity (α) will also 

affect the retention time 

values for the two peaks. 

(3) Peak width will be a function 

of column efficiency (N) and 

asymmetry (Asym). 

53 

N, Asym 



The equation below allows us to calculate the relative importance of each of 

these three factors in overall resolution: 

Efficiency 

Retention factor 

Selectivity 

It is important to note that you (the analyst) have control over each of those 

factors through your choice of HPLC column and running conditions: 

1. Efficiency (N)→Particle size/morphology and column length 

2. Selectivity (α) → Stationary phase and mobile phase 

3. Retention factor (k’) → Stationary phase and mobile phase 

54

Resolution: The Relative 

Effectiveness of k’, α, and N 

Most important 

determinant of 

resolution!! 

Constant increase in 

resolution 

Ineffective after k’ ~10 

55 

The Impact of Efficiency 

on Resolution 

Modulating column efficiency is a very effective way to optimize resolution. 

There is a strong, linear correlation between N and R s 

, but it is not a 1:1 ratio. 

Column efficiency is a flexible tool because we can easily modify it via changes 

in particle size and column length. 

Column 

Length 

(mm) Efficiency 5µm Efficiency 3µm 

Efficiency 

sub-2µm / 

Core-shell 

250 25,000 37,500 

56 

More Pressure 

Longer Run 

Time 

150 15,000 22,500 45,000 

100 10,000 15,000 30,000 

50 5,000 7,500 15,000 

More efficiency/resolution 

More Back-Pressure


on Resolution 

Relative Resolution 

2.5 

2 

1.5 

1 

0.5 

Doubling column 

efficiency increases 

R s by a factor of 1.4x 

57 

0 

0 10000 20000 30000 40000 

Column Efficiency (Plates) 

mAU 


on Resolution 

100 

80 

60 

40 

5 µm 

80,000 P/m 

20 

0 

0 0.5 1 1.5 2 2.5 3 3.5 

min 

mAU 

140 

120 

100 

80 

3 µm 

150,000 P/m 

60 

40 

20 

0 

0 0.5 1 1.5 2 2.5 3 3.5 

min 

58

Optimizing Efficiency for 

Maximum Resolution 

1. For method development, start with an intermediate column length, packed 

with the smallest particle size that system pressure limitations will allow. 

• Conventional HPLC → 3µm 150x4.6mm or core-shell 100x4.6mm 

• UPLC → sub-2µm or core-shell particle 

• Work at optimal flow rate for that particle size 

2. Fine-tune for maximum productivity: 

• Excessive resolution → shorter column, increase flow rate 

• Insufficient resolution → longer column; modify flow rate to compensate 

for pressure 

59 

The Impact of Retention 

Factor on Resolution 

Retention factor is the most important, yet limited, factor in determining 

resolution. It is crucial to have a reasonable k’ value because analytes must be 

retained in order to separate them. The drawback is that at high k’ values, 

passive diffusion causes extensive band broadening and loss of performance. 

60

Optimizing Retention Factor 

for Maximum Resolution 

1. Adjust k’ value to be between 2 and 10 

• In RP, adjust % of organic (acetonitrile or methanol) 

• Altering nature of stationary phase/media can modulate k’ as well 

2. At k’ < 2, have sub-optimal resolution 

• May also have interference from solvent, non-retained components 

3. At k’ values > 10, band broadening due to diffusion limits resolution gain 

• In RP, complex mixtures of polar and non-polar components will require 

gradient for optimal performance/run time balance 

• Polar stationary phases can the “total elution window” of complex 

mixtures in isocratic mode 

61 

The Impact of Selectivity 

on Resolution 

Small changes in selectivity can have a dramatic effect on retention. This 

is one of the reason why the same stationary phases from different 

manufacturers can sometimes give very different results, and also why 

changes to mobile phase composition can alter the results so strongly. 

62

Method Development Exercise 1: 

Optimization to Reduce Analysis Time and 

Increase Productivity 

63 

Mupirocin Impurity Profile 

Column: C8 250 x 4.6mm 5µm 

Mobile phase: 

Flow rate: 

Components: 

70 / 30 0.1M Ammonium acetate / THF 

1.0 mL/min 

1-6 = Impurities A - G 

7. Mupirocin 

Mupirocin 

64

0 

mAU 

Mupirocin: Original Method 

DAD1 A, Sig=240,10 Ref=off (Z:\1\DATA\MT050211\DJGSK 2011-05-02 15-34-44\MUPIROCIN000001.D) 

175 

7 

150 

125 

100 

75 

50 

25 

5 

R s 

3/4 = 0.63; k’ = 2.3 

1 

2 3 6 

4 

~16 min 

min 

0 2 4 6 8 10 12 14 16 

Column: 

Luna 5µ C8(2) 250 x 4.6mm 

Mobile phase: 70/30 0.1M Ammonium acetate pH 5.7/THF 

Flow rate: 

1.0 mL/min 

65 

Step 1. Adjust k’ for better 

Resolution 

66 

Step 1. Reduce % organic to increase k’: 

• Increases R s 

• Increases run time

Step 2. Optimize Efficiency 

and Length 

Column 

Length 

(mm) 

Efficiency dp 

5µm 

Efficiency dp 

3µm 

Efficiency 

sub-2µm / 

Core-shell 

%Reduction in 

Analysis Time 

250 25,000 37,500 

150 15,000 22,500 45,000 33 

100 10,000 15,000 30,000 60 

50 5,000 7,500 15,000 80 

67 

Step 2. Switch to 150x4.6mm 3 µm media: 

• Reduces analysis time 

• Maintains efficiency 

Step 3. Optimize the Flow 

Rate 

35000 

30000 

25000 

Core-Shell 2.6 

Luna 3u 

Luna 5u 

N (Plates/column) 

20000 

15000 

10000 

5000 

68 

0 

0 0.5 1 1.5 2 2.5 3 3.5 

Flow rate (ml/min) 

Step 3. Increase flow rate to 1.5 mL/min: 

• Optimizes efficiency for 3 µm

mAU 

80 

VWD1 A, Wavelength=240 nm (JL050211\MUPI0003.D) 

Mupirocin: Intermediate 

Method 

5 

7 

60 

40 

1 

k’ = 9; R s 

3/4 = 1.6 

20 

2 3 4 

6 

20 min 

0 

0 2 4 6 8 10 12 14 16 18 

min 

Column: Luna 3µ C8(2) 150 x 4.6mm 

Mobile phase: 80/20 0.1M Ammonium acetate pH 5.7/THF 

Flow rate: 1.5 mL/min 

69 

R s increased from 0.63 to 1.6 

Run time increased from 16 to 20 minutes 

Step 4. Switch to Core-Shell Media 

Column 

Length 

(mm) 

Efficiency dp 

5µm 

Efficiency dp 

3µm 

Efficiency 

sub-2µm / 

Core-shell 

%Reduction in 

Analysis Time 

250 25,000 37,500 

150 15,000 22,500 45,000 33 

100 10,000 15,000 30,000 60 

50 5,000 7,500 15,000 80 

70 

Step 4. Switch to 100x4.6mm Core-Shell media: 

• Reduce analysis time 

• Increase efficiency

0 

mAU 

Mupirocin: Final Optimized Method 

mAU 

250 


5 

7 

200 

150 

1 

R s 

3/4 = 2.3 

100 

8 min 

50 

2 3 4 

6 

0 

0 1 2 3 4 5 6 7 8 9 

min 

Column: Kinetex 2.6µ C8 100 x 4.6mm 

Mobile phase: 80/20 0.1M Ammonium acetate pH 5.7:THF 

Flow rate: 1.5 mL/min (2mL/min if pressure allows) 

R s 

increased from 1.6 to 2.3 

Run time decreased from 20 to 8 minutes 

71 

Initial Method: 

Mupirocin: Final Optimized Method 

D AD 1 A, Sig= 240,1 0 R ef=o ff ( Z:\1 \DAT A\MT 050 211\ DJ G SK 201 1-0 5-0 2 15 -34 -44 \MU PIRO C IN00 0001 .D) 

1 75 

1 50 

1 25 

1 00 

75 

50 

25 

1 

5 

R s 

3/4 = 0.63 

2 3 4 

6 

7 

16 min 

0 2 4 6 8 1 0 12 14 16 

Final Optimized Method: 


mAU 

250 

5 

7 

200 

150 

100 

1 

R s 

3/4 = 2.3 

8 min 

50 

2 3 4 

6 

72 

0 

0 1 2 3 4 5 6 7 8 9 

min 

min


73 


Part 2. The Stationary Phase: 

An Overview of HPLC Media 

Part 3. Role of the Mobile Phase in Selectivity 

74

Fully Porous Silica 

Polymerization 

Alkaline 

Tetraethoxysilane 

Silica Sol-Gel 

99.9% of reactive 

surface area is 

internal 

75 

Fully Porous Silica 

Advantages: 

• Ability to derivatize with numerous 

bonded phases 

• High mechanical strength 

• Excellent efficiency 

• Highly amenable to modulation of 

material characteristics (pore size, 

surface area, etc.) 

Disadvantages: 

• Dissolution of silica at pH > ~7.5 (may 

extend with bonded phase) 

• Hydrolysis of bonded phase at pH

Organosilica Hybrid Particle 

Conventional Silica Particle 


Siloxane 

Bridge 

Ethane 

linkage 

Dissolution at pH > 7.5 Stable to pH ~12 

77 


Advantages: 

• Extended pH range from 1-12 

• Performance and strength of 

conventional silica particle 

• Unique selectivity 


• Fewer stationary phases available 

compared to conventional silica (e.g. 

cyano, amino) 

78

Core-Shell Particle 

0.35 µm Porous Shell 

2.6 µm Core-Shell 

Particle 

1.9 µm Solid Core 

79 

Core-Shell Particle 

Advantages: 

• 3x the efficiency of 5 µm fullyporous 

media & 2x the efficiency of 

3 µm media 

• Pressures compatible with 

conventional HPLC systems* 


• Pressure is still higher than 3 µm 

media 

• More sensitive to system extracolumn 

volumes 

• More sensitive to overload in 

some cases 

80

RP Stationary Phase Classes 

Alkyl bonded phases (C18, C8, C4): 

Polar-embedded phases: 

Fusion 

Phenyl phases (Phenyl, PFP): 

Polar-endcapped phases: 

F 

F 

F 

F 

F 

Hydro 

81 

Methylene Selectivity 

We use the methylene selectivity test to determine the ability of stationary 

phase to separate molecules based upon differences in their hydrophobic 

character. In general, very hydrophobic bonded phases (e.g. C18) will display 

higher levels of methylene selectivity than less hydrophobic phases. 

m A U 

2 5 0 

C18 

2 0 0 

1 5 0 

1 0 0 

5 0 

0 

m A U 

2 4 6 8 1 0 

m i n 

4 0 0 

Phenyl 

3 0 0 

2 0 0 

1 0 0 

0 

2 4 6 8 1 0 

m i n 

82


0.200 

C18 > C8 > C5 ≥ Phenyl > CN > Amino 

0.180 

0.160 

Slope of log k vs. # -CH2- units 

0.140 

0.120 

0.100 

0.080 

0.060 

0.040 

0.020 

0.000 

Luna 

C18(2) 

Synergi 

Hydro-RP 

Jupiter 

C18 

Synergi 

Max-RP 

Luna 

C8(2) 

Luna C5 

Luna 

Phe-Hex 

Jupiter 

C4 

Synergi 

Polar-RP 

Prodigy 

Phenyl 

Luna 

Cyano 

Luna 

Amino 

83 


Columns: 

Mobile phase: 

Flow rate: 

Components: 

5µm C18 150x4.6mm 

5µm C8 150x4.6mm 

5µm Phenyl 150x4.6mm 

65:35 Acetonitrile:Water 

1 mL/min 

Two steroids: 

1. Testosterone 

2. Methyltestosterone 

CH 3 

OH 

CH 3 

H 

CH 3 

H 

CH 3 

OH 

CH 3 

O 

H 

H 

O 

H 

H 

84 

Testosterone 

Methyltestosterone

Phenyl Selectivity 

Columns: 

Dimensions: 

Mobile phase: 

Flow rate: 

C18 

Phenyl 

150 x 4.6 mm 

75:25 Methanol:water 

1 mL/min 

OH 

CH 3 

O 

CH 3 

Components: 

1. Estrone 

2. Estradiol 

HO 

H 

H 

H 

HO 

H 

H 

H 

Estradiol 

Estrone 

85 

Phenyl Selectivity 

OH 

CH 3 

CH 3 

O 

H 

H 

C18 

m A U 

1 7 5 

1 5 0 

C18 

HO 

H 

Estradiol 

H 

HO 

H H 

Estrone 

1+2 

1 2 5 

1 0 0 

7 5 

5 0 

2 5 

0 

1 2 3 4 5 

m in 

Phenyl 

m A U 

1 0 0 

Phenyl 

1 

8 0 

2 

6 0 

4 0 

2 0 

0 

1 2 3 4 5 

m in 

86

Aqueous Stability of 

Embedded Phases 

Nucleic Acid Bases: 

Luna C18(2) 

Day 1 

Polar-Endcapped C18 

Day 1 

0 

2 4 6 8 10 12 

Day 2 

0 

0 2 4 6 8 10 12 

Day 6 

Phase Collapse! 

0 

2 4 6 8 10 12 14 

min 

0 

2 4 6 8 10 12 14 

87 

Aqueous Stability of 

Embedded Phases 

LC/MS/MS Analysis of ETG & ETS in Urine: 

Polar-Endcapped 2.5 µm C18 

100x3.0mm 

XIC of -MRM (6 pairs): 221.200/75.000 Da ID: ETG-1 from Sample 6 (P-2_Hyro-RP_100x4.6_4u_FR600... 

1.00e5 

9.50e4 

9.00e4 

8.50e4 

Max. 9020.0 cps. 

10mM Ammonium formate 

8.00e4 

7.50e4 

7.00e4 

6.50e4 

ETG 

ETS 

6.00e4 

1. Ethyl glucuronide 

2. Ethyl sulfate 

Intensity, cps 

5.50e4 

5.00e4 

4.50e4 

4.00e4 

3.50e4 

3.00e4 

2.50e4 

2.00e4 

1.50e4 

1.00e4 

5000.00 

0.00 

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 

49 96 144 191 239 287 334 382 429 477 525 572 620 

Time, min 

XIC of -MRM (6 pairs): 221.200/75.000 Da ID: ETG-1 from Sample 6 (P-2_Hyro-RP_100x4.6_4u_FR600... 

Max. 9020.0 cps. 

1.10e4 

ETG 

1.00e4 

9000.00 

ETG 

ETS 

8000.00 

7000.00 

Intensity, cps 

6000.00 

5000.00 

4000.00 

ETS 

3000.00 

2000.00 

1000.00 

0.00 

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 

49 96 144 191 239 287 334 382 429 477 525 572 620 

Time, min 

88


Part 2. Overview of HPLC Media 

Part 3. Role of the Mobile Phase in Selectivity 

Solvents for RP 

Chromatography 

Mobile phase selection is much more challenging that stationary phase 

selection because the options are limitless. However, in practical method 

development, we can dramatically narrow down the options to focus on those 

conditions which will give us the highest likelihood of success. 

Typical RP Solvents: 

Weak Solvent: 

Water/Buffer 

Strong Solvent: Acetonitrile (64) 

Methanol (34) 

Composite mixtures (1) 

THF (1) 

Frequency of use

Solvent Selectivity 

The elution strength of a given solvent is determined by its hydrophobicity (e.g. 

heptane would be stronger than hexane because it is more hydrophobic). The 

selectivity of a solvent, however, is determined by its polar characteristics 

(e.g. heptane and hexane would have the same solvent selectivity). 

Methanol is a strong proton 

donor and a strong proton 

acceptor in hydrogen bonding. 

C H 3 

OH 

Acetonitrile has a dipole 

moment but is only a very weak 

proton acceptor in hydrogen 

bonding. 

N CH 3 

δ − δ + 

Tetrahyrofuran is able to 

accept a proton in hydrogen 

bonding but cannot donate a 

proton. 

O 

Solvent Selectivity 

Optimum Separation of 4 Steroids in Different Solvents:

Solvent Screening for 

Isocratic Methods 

1. Start at high % acetonitrile and work backwards until k’ is 2-10 (if possible) 

80% ACN 

40% ACN 

25% ACN 

21% ACN 

k’ = 0 

k’ ~ 0.8 

k’ ~ 6 

k’ ~ 11 

Solvent Screening for 

Isocratic Methods 

2. Repeat with alternative solvent:

Buffer Selection for 

RP-HPLC 

= Typical for LC/UV 

= Typical for LC/MS 

Effect of pH on 

Base Silica 

Any silica-based RP material will have some residual silanols left after 

bonding and end-capping. These Si-OH groups can be deprotonated at 

values above pH ~3.5. The deprotonated silanols are more likely to engage in 

ion-exchange with basic analytes, leading to peak tailing. 

pH 3.5 

Si 

Si 

Si 

O 

O 

Si 

O 

O Si 

O 

O Si 

OH 

OH 

OH 

Si 

Si 

Si 

O 

O 

Si 

O 

O Si 

O 

O Si 

O 

OH 

O 

• Silanols protonated 

• Less ion-exchange 

• Less peak tailing 

• Silanols deprotonated 

• Increased ion-exchange 

• Increased peak tailing

Effect of pH on Analyte 

Ionization 

The primary mechanism of retention in RP chromatography is hydrophobic 

interaction. Ionizing compounds will cause them to behave as more polar 

species, and reduce their hydrophobic interaction with the stationary phase, 

leading to decreased retention. 

• More hydrophobic 

• More strongly retained 

• Less hydrophobic 

• Less strongly retained 

• More hydrophobic 

• More strongly retained 

• Less hydrophobic 

• Less strongly retained 

The ionization state of a molecule will be determined by the pH of the mobile 

phase. Therefore, mobile phase pH will dictate retention behavior of 

analytes with ionizable functional groups. 


Ionization 

Acidic Compounds: 

Basic Compounds: 


Retention Factor (k’)


Ionization 

Alkaline 

H + 

Acidic 

Alkyl Stationary Phase 

Aqueous Mobile Phase 


Ionization 

Alkaline 

H + 

Acidic 

H + 

Alkyl Stationary Phase 

Aqueous Mobile Phase


Retention 

Amitriptyline (pKa 9.4) = (B)ase Toluene = (N)eutral Naproxen (pKa 4.5) = (A)cid 

B 

A 

A 

B 

A 

B 

N 

N 

N 

Gradient Analysis 

The gradient slope is analogous to solvent strength in isocratic elution. 

Isocratic Solvent Strength: 

Increasing the solvent strength 

reduces analysis time but also 

reduces resolution. 

Decreasing the solvent strength 

increases resolution at the cost 

of increased analysis time. 

Solvent strength sometimes 

affects selectivity 

Gradient Slope: 

Increasing the gradient slope 

reduces analysis time but also 

reduces resolution. 

Decreasing the gradient slope 

increases the resolution at the 

cost of increased analysis time. 

Gradient slope sometimes 

affects selectivity 

The goal of gradient elution is to optimize resolution while 

minimizing analysis time.

Temperature in 

HPLC Methods 

The use of temperature in HPLC method development presents a challenge 

because it can have unpredictable effects on selectivity. 

The use of elevated temperatures will: 

1. Reduce mobile phase viscosity and back-pressure. This can allow you to 

operate at higher flow rates, or to use longer columns/smaller particle sizes. 

2. Reduce elution time. 

3. Improve method reproducibility (as opposed to operating at room 

temperature). 

However, it is impossible to determine if the use of elevated temperatures will 

help or hinder a specific separation. For complex separations, improvements 

in one portion of the chromatogram are almost always accompanied by 

decreases in another part of the same chromatogram. 

End of Section II 


104

End of the HPLC Method Development Portion

General Principles of HPLC Method Development

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?