General Principles of HPLC Method Development PDF

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
The Beginning of Liquid

Chromatography
Column Chromatography:
Empty Adsorbent Sample is Solvent is Separation

Column particles loaded onto added to the occurs as
added the top of top of the the bands
the column column 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
The Liquid
Chromatographic Process
Analytes
Mobile Phase
Stationary Phase
6
The Liquid
Analytes
Mobile Phase
Stationary Phase
The Liquid
H O H
H O
H N
Polar/Aqueous O
H
H
H O H
Mobile Phase
N
N N
O
H
H
Benzo(a)pyrene Ethyl sulfate
Non-Polar Stationary
Phase (e.g. C18)
8
Mechanisms of Interaction
In RP Chromatography
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
Method Development = modulating stationary phase and mobile phase
conditions to optimize these interactions and achieve a specific separation
goal.
Tapentadol
OH
CH3
CH3
N
CH3
CH3
9
Hydrophobic Interactions
CH3 Hydrophobic CH3 H C

CH3 • Weak, transient 3interactions
OH between a non-polar stationary
phase and the molecules
CH3
• Hydrophobic & van Der
CH3
Waals interactions
N
H3C CH3 • Retention will be predicted by
Log P values
H 3C S i H 3C S i H 3C CH3 CH3
H 3C S i Si
C H 3O C H 3 Si
O - OH C HO
3 CH3
O OH O O O CH3
Si Si Si Si Si Si Si Si Si Si
H 3C O O O O O O O O O OH
HO OH HO OH OH OH HO OH OH OH
10
Polar Interactions
CH3 CH3 3 H C
CH3 • Interactions between polar
functions groups of analyte
and residual silanols or polar
Polar groups on media
CH3 • Hydrogen bonding, dipole-

H3C
N dipole interactions
H3C
• Relatively weak and transient

H3C
HO
H 3C S i H 3C S i H 3C CH3 CH3
H 3C S i Si
CH3 Si
O C H 3O OH O
- OH
O
C HO
3 CH3
O O CH3
11
Ion-exchange Interactions
CH3 CH3 3 H C
CH3 • Interactions between
ionizable functional groups on
analyte and counter-charged
moiety on stationary phase
Ion-Exchange
• Ion-exchange
OH
• Strong, slow interactions

CH3
H3C
+
HN
H3C
H 3C S i H 3C S i CH3 H 3C CH3 CH3
H 3C S i Si
CH3 Si
O C H 3O OH O
- OH
O
C HO
3 CH3
O O CH3
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 Mobile phase fills the pores of the
between the particles or the porous adsorbent particles.
interstitial volume.
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 t0.
• void volume
• solvent front
• t0
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’)=

tR – t0
t0
t0
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.
CH3 CH3 H 3C
CH 3
Ion-Exchange
OH
CH3
H 3C
+
HN
H 3C
CH3
H 3C Si H C Si H 3C CH3 CH3
3 H 3C S i Si
CH3 Si
O C H 3O OH O
- OH
O
C HO
3 CH3
O O CH3
21
Peak Tailing due to

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
Peak Tailing due to
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% ... Max.
XIC of +MRM (8 pairs): 278.2/191.2 1.1e5
Da cps.
ID: Amitrip from Sample 7 (TCA-Kin-XB-C18, 50x2, 2.6, MeOH50.... Max. 1.2e5 cps.
1
5 5
4 1
1 4
Peak tailing
6 6
2 3 3 7
7 8 2
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% ... Max.
XIC of +MRM (8 pairs): 264.2/191.1 1.1e5
Da cps. from Sample 13 (Kinetex XB-C18-50x2, 2.6 um, MeO...
ID: Protrip 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 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
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*
23 7. Clomipramine
8. Norclomipramine
Peak Tailing due to

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):
mAU
DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000006.D)
14.208
2.5 µg on column;
30
25
20
USP Tailing = 1.17
15
10
pKa 9.7
5
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min
mAU
14.169
50 5 µg on column;
40
USP Tailing = 1.34
30
20
10
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min
mAU
14.157
100 12 µg on column;
80
USP Tailing = 1.87
60
40
20
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min
24
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)

mAU
2250 2.495
2000
1750
1500 Fronting due to overload
1250
1000
750
500
250 1.905
1.745 2.145
0
1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 min
25
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 Sample in 100% Methanol

7.0e4 2.2e5
6.5e4
Hydromorphone 2.0e5
6.0e4
1.8e5
5.5e4
5.0e4 1.6e5
4.5e4
Breakthrough!
1.4e5
4.0e4
1.2e5
3.5e4
1.0e5
Fronting
3.0e4 0.24
2.5e4 8.0e4
2.0e4
6.0e4
1.5e4
Morphine 4.0e4
1.06
1.0e4
1.77
Norhydrocodone 1.36 1.79
2.0e4
5000.0
0.97
0.51 1.71 1.85 2.15 3.213.36 3.97
0.0 0.0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
26
Selectivity (α
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 = k2/k1
27
Selectivity (α)
The choice of stationary phase will often have a dramatic effect on the
selectivity of analytes. CH
O
3
CH
OH
3
H H
H H H H
HO HO
Estrone Estradiol
m A U
1+2
1 7 5
C18 Column
1 5 0
α = 1.0
1 2 5
1 0 0
7 5
5 0
2 5
1 2 3 4 5 m in
m A U
1 0 0
Phenyl Column 1
8 0
6 0
2 α = 2.3
4 0
2 0
1 2 3 4 5 m in
28
Selectivity (α
α)
But mobile phase is also a very powerful tool in modulating selectivity.
Column: Gemini 5 µm C6-Phenyl

150 x 4.6mm 35% Methanol
Mobile phase: 20mM KH2PO4, pH 2.5;
% organic as noted
Flow rate: 1.0 mL/min
Detection: UV at 220nm
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:
• Columns that cause a lot of peak

tR
broadening have few theoretical
plates.
• Columns that produce very narrow
peaks (little peak broadening) have
a large number of theoretical plates.
W1/2
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-
core-shell particles will deliver
significantly higher efficiency (N) than columns packed with
fully-
fully-porous particles of the same diameter.*
Fully Porous Particles
w1/2
tR
Injected
Sample
Band
* Gritti et al., Journal of Chromatography A, 1217 (2010) 1589

Core-Shell Particles
fully-
Kinetex™ Core-Shell Particles
w1/2
5 6
tR

Comparison
fully-
Fully Porous 3 µm C18; 150 x 4.6 mm Kinetex™ 2.6 µm C18; 150 x 4.6 mm
N = 166,500 p/m N = 295,340 p/m
78% Increase
in Efficiency!
34
Efficiency vs. Diameter
fully-
450
400 Fully Porous Core-Shell
350
300
Efficiency( p/m)
250
200
150
100
50
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 · dp + B/µ + C · de2 · µ
Particle size
Particle size Linear velocity (flow rate)
Mass Transfer
Linear velocity (flow rate)

Mobile phase viscosity
36
A-term
H = A · dp + B/u + C · de2 · u
Eddy Diffusion
Multi-path Effect
w1/2
6
tR

B-term
H = A · dp + B/u + C · de2 · u
Longitudinal diffusion
w1/2
6
tR
Fully Porous C-term
H = A · dp + B/µ + C · de2 · µ
Mass Transfer (Kinetics)
Dispersion due to resistance to mass transfer
39

Core-Shell C-term
H = A · dp + B/µ + C · de2 · µ
Mass Transfer (Kinetics)
Dispersion due to resistance to mass transfer
A
B
h vs. v Comparison
H = A · dp + B/µ + C · de2 · µ
15 15
4.6 x 100 mm Luna-C18 (2) 4.6 x 100 mm Kinetex-C18
Reduced plate height h

Reduced plate height h
10 10 Eddy dispersion
Eddy dispersion Longitudinal diffusion
Longitudinal diffusion Solid-liquid mass transfer
Solid-liquid mass transfer
5 5
0 0
0 5 10 15 20 0 5 10 15 20
Reduced velocity ν 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 5,000 Plates

8 min
50 Bar
10cm 10,000 Plates

16 min
100 Bar
15cm 15,000 Plates
24 min
150 Bar
25cm 25,000 Plates
40 min
250 Bar
42
Balancing Column Length
and Particle Size
Column
Length Efficiency dp
(mm) 5µµm
250 25,000
150 15,000
100 10,000
50 5,000
43

and Particle Size
Column
Length Efficiency dp Efficiency dp
(mm) 5µµm 3µ
µm
250 25,000 37,500
150 15,000 22,500
100 10,000 15,000
50 5,000 7,500
44
and Particle Size
Column Efficiency
Length Efficiency dp Efficiency dp sub-2µµm /
(mm) 5µµm 3µ
µm 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

and Particle Size
Column Efficiency
Length Efficiency dp Efficiency dp sub-2µµm / %Reduction in
(mm) 5µµm 3µ
µm Core-shell 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
Use shorter columns packed with smaller particles

to reduce analysis time while maintaining/improving
efficiency!!
46
Effect of Flow Rate on
Efficiency
Effect of Flow Rate on Column Efficiency (100x4.6mm)
35000
Core-Shell ~2 mL/min Core-Shell 2.6
30000
Luna 3u
25000 Luna 5u
3µ ~1.5 mL/min
N (Plates/column)
20000
15000
10000
5µ ~1 mL/min
5000
0
0 0.5 1 1.5 2 2.5 3 3.5
Flow rate (ml/min)
47
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 t0

• 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
α
3. Selectivity ( ) – 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) = RTB - RTA / .5 (widthA + widthB)
52
α
(1) Retention time for the two k’

peaks will be a function of
retention factor (k’).
α
(2) The selectivity ( ) 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. α
Selectivity ( ) → 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 Rs, 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 Efficiency
Length sub-2µµm /
(mm) Efficiency 5µ
µm Efficiency 3µ
µm Core-shell
250 25,000 37,500
150 15,000 22,500 45,000

More Pressure
Longer Run
100 10,000 15,000 30,000

Time
50 5,000 7,500 15,000

More efficiency/resolution
56
More Back-Pressure
on Resolution
2.5
Doubling column
Relative Resolution
2 efficiency increases
Rs by a factor of 1.4x
1.5
0.5
0
0 10000 20000 30000 40000
Column Efficiency (Plates)

57

on Resolution
mA U
10 0
80
60
5 µm
40
80,000 P/m
20
0 0.5 1 1.5 2 2.5 3 3. 5 min
mA U
14 0
12 0
10 0
3 µm
80
150,000 P/m
60
40
20
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: 70 / 30 0.1M Ammonium acetate / THF

Components: 1-6 = Impurities A - G

7. Mupirocin
Mupirocin
64
Mupirocin: Original Method
DAD1 A, Sig=240,10 Ref=off (Z:\1\DATA\MT050211\DJGSK 2011-05-02 15-34-44\MUPIROCIN000001.D)
mAU
175 7
150 5
125
100
75
Rs 3/4 = 0.63; k’ = 2.3
50
1 ~16 min
6
2 34
25
0 2 4 6 8 10 12 14 16 min
Column: Luna 5µ C8(2) 250 x 4.6mm
Mobile phase: 70/30 0.1M Ammonium acetate pH 5.7/THF
65
Step 1. Adjust k’ for better

Resolution
Step 1. Reduce % organic to increase k’:

• Increases Rs
• Increases run time
66
Step 2. Optimize Efficiency
and Length
Column Efficiency
(mm) 5µµm 3µ
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
Step 2. Switch to 150x4.6mm 3 µm media:

• Reduces analysis time
• Maintains efficiency
67
Step 3. Optimize the Flow

Rate
35000
Core-Shell 2.6
30000
Luna 3u
25000 Luna 5u
N (Plates/column)
20000
15000
10000
5000
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
68
Mupirocin: Intermediate
Method
VW D1 A, W avelength=240 nm (JL050211\MUPI0003.D)
mAU
7
80 5
60
40 1
k’ = 9; Rs 3/4 = 1.6
20 min
20 3
2 6
4
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
Rs increased from 0.63 to 1.6

Run time increased from 16 to 20 minutes
69
Step 4. Switch to Core-Shell Media
Column Efficiency
(mm) 5µµm 3µ
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
Step 4. Switch to 100x4.6mm Core-Shell media:

• Reduce analysis time
• Increase efficiency
70
Mupirocin: Final Optimized Method
VW D1 A, W avelength=240 nm (JL050211\MUPI0006.D)
mAU
7
250
5
200
150
Rs 3/4 = 2.3
1
100
8 min
50
3
2 6
4
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)
Rs increased from 1.6 to 2.3

Run time decreased from 20 to 8 minutes
71
Mupirocin: Final Optimized Method

Initial Method:
D A D 1 A , S ig= 240,1 0 R ef=o ff ( Z:\1 \DA T A \M T 050 211\ DJ G S K 201 1-0 5-0 2 15 -34 -44 \M U P IRO C IN00 0001 .D)
m AU
1 75
7
1 50 5
1 25
1 00
Rs 3/4 = 0.63
75
16 min
50
1
25 2 34 6
0
0 2 4 6 8 10 12 14 16 m in
Final Optimized Method:

VWD1 A, Wavelength=240 nm (JL050211\MUPI0006.D)
mAU
250 5 7
200
150
1 Rs 3/4 = 2.3
100 8 min
3
50
2 6
4
0
72 0 1 2 3 4 5 6 7 8 9 min
Any Questions?
73
Part 2. The Stationary Phase:

An Overview of HPLC Media
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 <1.5
76
Organosilica Hybrid Particle
Conventional Silica Particle Organosilica Hybrid Particle
Ethane
Siloxane linkage
Bridge
Dissolution at pH > 7.5 Stable to pH ~12
77
Organosilica Hybrid Particle
Advantages:
• Extended pH range from 1-12
• Performance and strength of
conventional silica particle
• Unique selectivity
Disadvantages:
• 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
1.9 µm Solid Core
Particle
79
Core-Shell Particle
Advantages:
• 3x the efficiency of 5 µm fully-
porous media & 2x the efficiency of
3 µm media
• Pressures compatible with
conventional HPLC systems*
Disadvantages:
• Pressure is still higher than 3 µm
media
• More sensitive to system extra-
column 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
Hydro
F
F F
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 AU
250
C18
200
150
100
50
2 4 6 8 10 m in
m A U
4 00
Phenyl
3 00
2 00
1 00
2 4 6 8 1 0 m in
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 Synergi Jupiter Synergi Luna Luna C5 Luna Jupiter Synergi Prodigy Luna Luna
C18(2) Hydro-RP C18 Max-RP C8(2) Phe-Hex C4 Polar-RP Phenyl Cyano Amino
83
Columns: 5µm C18 150x4.6mm

5µm C8 150x4.6mm
5µm Phenyl 150x4.6mm
Mobile phase: 65:35 Acetonitrile:Water

Flow rate: 1 mL/min
Components: Two steroids:

1. Testosterone
2. Methyltestosterone
OH OH
CH3 CH3 CH
3
CH3 H
CH3 H
H H
H H
O
O
Testosterone
Methyltestosterone
84
Phenyl Selectivity
Columns: C18
Phenyl
Dimensions: 150 x 4.6 mm

Mobile phase: 75:25 Methanol:water
Flow rate: 1 mL/min
OH O
CH3 CH3
H H
Components: 1. Estrone
H H H H
2. Estradiol
HO HO
Estradiol Estrone
85
Phenyl Selectivity
OH O
CH3 CH3
H H
H H H H
HO HO
C18 mAU
Estradiol Estrone
1 75 C18
1 50
1+2
1 25
1 00
75
50
25
1 2 3 4 5 m in
Phenyl mAU
1 00
Phenyl 1
2
80
60
40
20
1 2 3 4 5 m in
86
Aqueous Stability of
Embedded Phases
Nucleic Acid Bases:
Luna C18(2) Polar-Endcapped C18

Day 1
Day 1
0
0
2 4 6 8 10 12
Day 2 0 2 4 6 8 10 12
Day 6
Phase Collapse!
0
0
2 4 6 8 10 12 14 min
2 4 6 8 10 12 14
87
Aqueous Stability of
Embedded Phases
LC/MS/MS Analysis of ETG & ETS in Urine:
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.
Polar-Endcapped 2.5 µm C18 1.00e5

9.50e4
100x3.0mm 9.00e4
8.50e4
8.00e4
7.50e4 ETS
10mM Ammonium formate 7.00e4
6.50e4 ETG
6.00e4
tensity, cps
5.50e4
5.00e4
1. Ethyl glucuronide 4.50e4

In
4.00e4
2. Ethyl sulfate 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
1.00e4
ETG 9000.00
ETG
ETS
8000.00
7000.00
tensity, cps
6000.00
In
5000.00
4000.00
3000.00
ETS
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
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) Frequency of use
Composite mixtures (1)
THF (1)
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 Acetonitrile has a dipole Tetrahyrofuran is able to

donor and a strong proton moment but is only a very weak accept a proton in hydrogen
acceptor in hydrogen bonding. proton acceptor in hydrogen bonding but cannot donate a
bonding. proton.
O
H3C OH N CH3
δ− δ+
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 pH >3.5
O O
OH O
Si O Si Si O Si
O O
Si O Si OH Si O Si OH
O O
Si O Si OH Si O Si O
• Silanols protonated • Silanols deprotonated

• Less ion-exchange • Increased ion-exchange
• Less peak tailing • 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 • Less hydrophobic • More hydrophobic • Less hydrophobic
• More strongly retained • Less strongly retained • More strongly retained • 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:
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 B A B
A
A
N
N N
Gradient Analysis
The gradient slope is analogous to solvent strength in isocratic elution.
Isocratic Solvent Strength: Gradient Slope:

Increasing the solvent strength Increasing the gradient slope
reduces analysis time but also reduces analysis time but also
reduces resolution. reduces resolution.
Decreasing the solvent strength Decreasing the gradient slope
increases resolution at the cost increases the resolution at the
of increased analysis time. cost of increased analysis time.
Solvent strength sometimes Gradient slope sometimes
affects selectivity 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
Any Questions?
104
End of the HPLC Method Development Portion

General Principles of HPLC Method Development PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

General Principles of HPLC Method Development PDF

Uploaded by

Copyright:

Available Formats

General Principles

Philip J. Koerner, Ph.D.

Part 1. General Chromatographic Theory

Part 3. Role of the Mobile Phase in Selectivity

The Beginning of Liquid

Empty Adsorbent Sample is Solvent is Separation

Separation of compounds by HPLC depends on the interaction of

Mobile Phase Stationary Phase

Benzo(a)pyrene Ethyl sulfate

CH3 Hydrophobic CH3 H C

CH3 • Hydrogen bonding, dipole-

• Relatively weak and transient

• Strong, slow interactions

Retention factor (k’)=

Retention Factor (k’)

Example: The Separation of Steroids:

Peak Asymmetry (Asym)

Asymmetrical peaks can be

Peak Tailing due to

Example: The Separation of Tricyclic Antidepressants:

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

Brand H 2.7µ C18 50x2.1mm Kinetex 2.6µ XB-C18 50x2.1mm

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

Peak Tailing due to

12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min

12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min

12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 min

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

1500 Fronting due to overload

Peak Fronting due to

Sample in 50% Methanol Sample in 100% Methanol

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

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

Column: Gemini 5 µm C6-Phenyl

Column Efficiency (N)

• Columns that cause a lot of peak

30 Peak Width (W)

The Core-Shell Advantage

Fully Porous Particles

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

Kinetex™ Core-Shell Particles

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

The Core-Shell Advantage

400 Fully Porous Core-Shell

Particle Diameter (µm)

*Farcas et al., HPLC 2012 Conference

The van Deemter Equation

Linear velocity (flow rate)

The Core-Shell Advantage

Mass Transfer (Kinetics)

Dispersion due to resistance to mass transfer

The Core-Shell Advantage

Mass Transfer (Kinetics)

Dispersion due to resistance to mass transfer

4.6 x 100 mm Luna-C18 (2) 4.6 x 100 mm Kinetex-C18

Reduced plate height h

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

Effect of Column Length

5cm 5,000 Plates

10cm 10,000 Plates

Balancing Column Length

250 25,000 37,500

150 15,000 22,500