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
Column
4
Mikhail Tsvet, 1872-1919
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
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
The Liquid
Analytes
Mobile Phase
Stationary Phase
The Liquid
H
O
N
O
Polar/Aqueous
Mobile Phase
N
O
H
H
H
N
H
Ethyl sulfate
Benzo(a)pyrene
Non-Polar Stationary
Phase (e.g. C18)
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
CH3
CH3
Hydrophobic Interactions
CH3
CH3
Hydrophobic
H C
CH3
Weak,
transient 3interactions
between a non-polar stationary
phase and the molecules
OH
CH3
Hydrophobic & van Der

Waals interactions
CH3
N
H3C
S i H 3C S i
C H 3O C H 3
H 3C
O
H 3C
Si
HO
10
Si
O
Si
O
OH
Si
O
HO
OH
Si
O
OH
Retention will be predicted by

Log P values
CH3
Si
O
OH
OH
Si
O
OH
H 3C
Si
C HO
3
Si
O
HO
H 3C S i
O
OH
CH3
O
Si
O
OH
CH3
Si
CH3
CH3
Si
OH
OH
Polar Interactions
CH3
CH3
H C
CH3
3
Interactions
between
polar
functions groups of analyte
and residual silanols or polar
groups on media
Polar
Hydrogen bonding, dipoledipole interactions
CH3
H3C
N
H3C
Relatively weak and transient

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

CH3
H3C
+
S i H 3C S i
CH3
O C H 3O
H 3C
O
H 3C
Si
HO
12
Si
O
Si
O
OH
Si
O
HO
H3C
CH3
OH
Si
O
OH
HN
-
Si
O
OH
OH
Si
O
OH
H 3C
Si
C HO
3
Si
O
HO
H 3C S i
O
OH
CH3
O
Si
O
OH
CH3
Si
CH3
CH3
Si
OH
OH
Chromatographic Measurements
13
Chromatographic
Measurements
14
Asym
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 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.
CH
CH3
H 3C
CH3
Ion-Exchange
OH
CH3
H 3C
HN
H 3C
Si H C Si
3
CH3
O C H 3O
H 3C
O
H 3C
Si
Si
O
HO
Si
O
OH
Si
O
HO
Si
O
OH
CH3
OH
Si
O
OH
OH
Si
O
OH
H 3C S i
O
H 3C
Si
O
HO
Si
C HO
3
Si
O
OH
CH3
O
OH
CH3
Si
CH3
CH3
Si
OH
OH
21
Peak Tailing due to

Example: The Separation of Tricyclic Antidepressants:
Column used: Kinetex 2.6 XB-C18 50 x 2.1 mm
Brand H 2.7 C18 50 x 2.1mm
Amitriptyline (pKa 9.4)
22
Nortriptyline (pKa 9.7)
Protriptyline (pKa 8.2)
Peak Tailing due to

Brand H 2.7 C18 50x2.1mm
Kinetex 2.6 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.
1.1e5
cps.
XIC of +MRM (8 pairs): 278.2/191.2
Da
ID: Amitrip
from Sample 7 (TCA-Kin-XB-C18, 50x2, 2.6, MeOH50....
1
5
5
4
1
4
Peak tailing
2 3
3.4
3.6
3.8
4.0
4.2
Time, min
4.4
4.6
4.8
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. 1.2e5 cps.
5.0
5.2
8
5.4
5.6
5.8
6.0
3.4
3.6
3.8
4.0
4.2
Time, min
4.4
7
4.6
4.8
8
5.0
5.2
5.4
5.6
Max.
1.1e5
cps. from Sample 13 (Kinetex XB-C18-50x2, 2.6 um, MeO...
XIC of +MRM (8 pairs): 264.2/191.1
Da
ID: Protrip
5.8
6.0
Max. 1.2e5 cps.
6
6
3.4
23
3.6
3.8
4.0
4.2
Time, min
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
Time, min
4.4
4.6
4.8
5.0
5.2
5.4
5.6
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 L/min
1. Amoxapine
2. Imipramine
3. Desipramine
4. Protriptyline*
5. Amitriptyline
6. Nortriptyline*
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 (g on column):
mAU
14.208
DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000006.D)
2.5 g on column;
USP Tailing = 1.17
30
25
20
15
pKa 9.7
10
0
12
mAU
12.5
13
13.5
14
14.5
15
15.5
16
16.5
14.5
15
15.5
16
16.5
14.5
15
15.5
16
16.5
min
14.169
5 g on column;
USP Tailing = 1.34
50
40
30
20
10
0
12
mAU
12.5
13
13.5
14
min
14.157
12 g on column;
USP Tailing = 1.87
100
80
60
40
20
0
12
24
12.5
13
13.5
14
min
5.8
6.0
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
Fronting due to overload
1500
1250
1000
750
500
1.905
250
2.145
1.745
0
1.8
1.9
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
1.6e5
5.0e4
Breakthrough!
4.5e4
1.4e5
4.0e4
1.2e5
3.5e4
1.0e5
3.0e4
8.0e4
2.5e4
2.0e4
1.5e4
Fronting
0.24
6.0e4
Morphine
4.0e4
1.06
1.0e4
1.77
Norhydrocodone
1.36
0.51
0.0
26
0.5
1.0
1.5
1.79
2.0e4
5000.0
2.0
2.5
3.0
3.5
4.0
0.0
0.5
0.97
1.0
1.71
1.5
1.85 2.15
2.0
3.213.36
2.5
3.0
3.97
3.5
4.0
Selectivity (
)
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.
= k2/k1
27
Selectivity ()
The choice of stationary phase will often have a dramatic effect on the
OH
O
selectivity of analytes.
CH
CH
3
H
H
HO
HO
Estrone
Estradiol
m A U
1 7 5
1+2
C18 Column
1 5 0
= 1.0
1 2 5
1 0 0
7 5
5 0
2 5
0
1
m in
m A U
1 0 0
Phenyl Column
8 0
= 2.3
6 0
4 0
2 0
0
1
28
m in
Selectivity (
)
But mobile phase is also a very powerful tool in modulating selectivity.
Column:
Gemini 5 m 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
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.
W1/2
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 m
5 m
3 m
Core-Shell
sub-2 m
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 corecore-shell particles will deliver
significantly higher efficiency (N) than columns packed with
fullyfully-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
Kinetex Core-Shell Particles
w1/2
tR

Comparison
Fully Porous 3 m C18; 150 x 4.6 mm
N = 166,500 p/m
Kinetex 2.6 m C18; 150 x 4.6 mm

N = 295,340 p/m
78% Increase
in Efficiency!
34

Efficiency vs. Diameter
450
400
Fully Porous
Core-Shell
Efficiency( p/m)
350
300
250
200
150
100
50
0
5
3.5/3.6
2.5/2.6
1.7
1.3
Particle Diameter (m)
*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
Simplified version:
H = A dp + B/ + C de2
Particle size
Linear velocity (flow rate)
Mass Transfer
Particle size
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/ + C de2
Mass Transfer (Kinetics)
Dispersion due to resistance to mass transfer
A
B
39

Core-Shell C-term
H = A dp + B/ + C de2
Mass Transfer (Kinetics)
Dispersion due to resistance to mass transfer
A
B

h vs. v Comparison
H = A dp + B/ + C de2
15
4.6 x 100 mm Luna-C18 (2)

10
Reduced plate height h
Reduced plate height h
15
Eddy dispersion
Solid-liquid mass transfer
4.6 x 100 mm Kinetex-C18

Eddy dispersion
Solid-liquid mass transfer
10
0
0
10
15
20
10
15
Reduced velocity
Reduced velocity
*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
42
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
20
Balancing Column Length

and Particle Size
Column
Length
(mm)
Efficiency dp
5
m
250
25,000
150
15,000
100
10,000
50
5,000
43

and Particle Size
44
Column
Length
(mm)
Efficiency dp
5
m
Efficiency dp
3
m
250
25,000
37,500
150
15,000
22,500
100
10,000
15,000
50
5,000
7,500

and Particle Size
Column
Length
(mm)
Efficiency dp
5
m
Efficiency dp
3
m
Efficiency
sub-2
m /
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
46
Column
Length
(mm)
Efficiency dp
5
m
Efficiency dp
3
m
Efficiency
sub-2
m /
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
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
Core-Shell ~2 mL/min
Core-Shell 2.6
30000
Luna 3u
N (Plates/column)
25000
Luna 5u
3 ~1.5 mL/min
20000
15000
10000
5 ~1 mL/min
5000
0
0
0.5
1.5
2.5
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
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

(1) Retention time for the two

peaks will be a function of
retention factor (k).
(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).
N, Asym
53

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.
2.
3.
54
Efficiency (N)
Particle size/morphology and column length
Selectivity ( ) Stationary phase and mobile phase
Retention factor (k) Stationary phase and mobile phase
Resolution: The Relative

Effectiveness of k, , 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.
Longer Run
Time
More Pressure
Column
Length
(mm)
Efficiency 5
m Efficiency 3
m
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
More efficiency/resolution
56
Efficiency
sub-2
m /
Core-shell
More Back-Pressure

on Resolution
Relative Resolution
2.5
Doubling column
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
5 m
80,000 P/m
60
40
20
0
0
0.5
1.5
2.5
3. 5
min
mA U
14 0
12 0
3 m
150,000 P/m
10 0
80
60
40
20
0
0
58
0.5
1.5
2.5
3. 5
min
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 3m 150x4.6mm or core-shell 100x4.6mm
UPLC sub-2m 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 5m
Mobile phase:
Flow rate:
70 / 30 0.1M Ammonium acetate / THF

1.0 mL/min
Components:
1-6 = Impurities A - G
7. Mupirocin
Mupirocin
64
Mupirocin: Original Method

mAU
DAD1 A, Sig=240,10 Ref=off (Z:\1\DATA\MT050211\DJGSK 2011-05-02 15-34-44\MUPIROCIN000001.D)
175
150
125
100
Rs 3/4 = 0.63; k = 2.3
75
50
~16 min
6
2 34
25
0
0
Column:
10
12
14
Luna 5 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
Step 1. Reduce % organic to increase k:

Increases Rs
Increases run time
66
16
min
Step 2. Optimize Efficiency

and Length
Column
Length
(mm)
Efficiency dp
5
m
Efficiency dp
3
m
Efficiency
sub-2
m /
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
Step 2. Switch to 150x4.6mm 3 m media:

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

Rate
35000
Core-Shell 2.6
30000
Luna 3u
N (Plates/column)
25000
Luna 5u
20000
15000
10000
5000
0
0
0.5
1.5
2.5
3.5
Flow rate (ml/min)
Step 3. Increase flow rate to 1.5 mL/min:

Optimizes efficiency for 3 m
68
Mupirocin: Intermediate
Method
VW D1 A, W avelength=240 nm (JL050211\MUPI0003.D)
mAU
7
5
80
60
40
k = 9; Rs 3/4 = 1.6
20
20 min
3
6
0
0
10
12
14
16
18
min
Column:
Luna 3 C8(2) 150 x 4.6mm
Mobile phase: 80/20 0.1M Ammonium acetate pH 5.7/THF
Flow rate:
1.5 mL/min
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
Length
(mm)
Efficiency dp
5
m
Efficiency dp
3
m
Efficiency
sub-2
m /
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
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
200
150
Rs 3/4 = 2.3
1
100
50
8 min
0
0
min
Column:
Kinetex 2.6 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:
m AU
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)
1 75
1 50
1 25
1 00
Rs 3/4 = 0.63
16 min
75
50
2 34
25
0
0
10
12
14
16
m in
Final Optimized Method:

VWD1 A, Wavelength=240 nm (JL050211\MUPI0006.D)
mAU
250
200
150
Rs 3/4 = 2.3
100
50
8 min
72
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
Siloxane
Bridge
Dissolution at pH > 7.5

Ethane
linkage
Stable to pH ~12
77
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 m Porous Shell
2.6 m Core-Shell
Particle
1.9 m Solid Core
79
Core-Shell Particle
Advantages:
3x the efficiency of 5 m fullyporous media & 2x the efficiency of
3 m media
Pressures compatible with
conventional HPLC systems*
Disadvantages:
Pressure is still higher than 3 m
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
Polar-endcapped phases:
Phenyl phases (Phenyl, PFP):

F
Hydro
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
C18
250
200
150
100
50
0
m A U
10
m in
1 0
m in
Phenyl
4 00
3 00
2 00
1 00
82
0.200
C18 > C8 > C5 Phenyl > CN > Amino
0.180
Slope of log k vs. # -CH2- units
0.160
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
83
Columns:
5m C18 150x4.6mm
5m C8 150x4.6mm
5m Phenyl 150x4.6mm
Mobile phase:
Flow rate:
65:35 Acetonitrile:Water
1 mL/min
Components:
Two steroids:
1. Testosterone
2. Methyltestosterone
OH
CH3
CH3
H
O
CH3
H
H
H
O
Testosterone
84
OH
CH3 CH
Methyltestosterone
Prodigy
Phenyl
Luna
Cyano
Luna
Amino
Phenyl Selectivity
Columns:
C18
Phenyl
Dimensions:
Mobile phase:
Flow rate:
150 x 4.6 mm
75:25 Methanol:water
1 mL/min
OH
CH3
Components:
1. Estrone
2. Estradiol
O
CH3
H
H
H
H
HO
HO
Estrone
Estradiol
85
Phenyl Selectivity
OH
CH3
O
CH3
H
H
H
H
C18
HO
HO
Estrone
Estradiol
mAU
1 75
C18
1+2
1 50
1 25
1 00
75
50
25
0
1
Phenyl
m in
mAU
1 00
Phenyl
80
60
40
20
0
1
86
m in
Aqueous Stability of
Embedded Phases
Nucleic Acid Bases:
Luna C18(2)
Polar-Endcapped C18
Day 1
Day 1
0
2
10
12
Day 2
10
12
Day 6
Phase Collapse!
0
2
10
12
14
min
0
2
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...
Polar-Endcapped 2.5 m C18

100x3.0mm
Max. 9020.0 cps.
1.00e5
9.50e4
9.00e4
8.50e4
8.00e4
ETS
7.50e4
10mM Ammonium formate
7.00e4
ETG
6.50e4
1.
2.
Ethyl glucuronide
Ethyl sulfate
In
tensity, cps
6.00e4
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
49
1.0
96
1.5
144
2.0
191
2.5
239
3.0
287
3.5
334
Time, min
4.0
382
4.5
429
5.0
477
5.5
525
6.0
572
XIC of -MRM (6 pairs): 221.200/75.000 Da ID: ETG-1 from Sample 6 (P-2_Hyro-RP_100x4.6_4u_FR600...
6.5
620
Max. 9020.0 cps.
1.10e4
1.00e4
ETG
ETG
9000.00
ETS
8000.00
In
tensity, cps
7000.00
6000.00
5000.00
4000.00
3000.00
ETS
2000.00
1000.00
0.00
88
0.5
49
1.0
96
1.5
144
2.0
191
2.5
239
3.0
287
3.5
334
Time, min
4.0
382
4.5
429
5.0
477
5.5
525
6.0
572
6.5
620

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)
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.
H3C
OH
Acetonitrile has a dipole

moment but is only a very weak
proton acceptor in hydrogen
bonding.
Tetrahyrofuran is able to
accept a proton in hydrogen
bonding but cannot donate a
proton.
CH3
+
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
k = 0
40% ACN
k ~ 0.8
25% ACN
k ~ 6
Solvent Screening for

Isocratic Methods
2. Repeat with alternative solvent:
21% ACN
k ~ 11
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
OH
Si O Si
O
O
Si O Si
O
Si O Si OH
Si O Si OH
Si O Si OH
Silanols protonated
Less ion-exchange
Less peak tailing
Si O Si O
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
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
Basic Compounds:
Acidic 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
A
A
N
N
Gradient Analysis
The gradient slope is analogous to solvent strength in isocratic elution.
Isocratic Solvent Strength:
Gradient Slope:
Increasing the solvent strength

reduces analysis time but also
reduces resolution.
Increasing the gradient slope

reduces analysis time but also
reduces resolution.
Decreasing the solvent strength

increases resolution at the cost
of increased analysis time.
Decreasing the gradient slope

increases the resolution at the
cost of increased analysis time.
Solvent strength sometimes

affects selectivity
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
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

The Liquid Chromatographic Process

The Beginning of Liquid

Separation of compounds by HPLC depends on the interaction of

Hydrophobic & van Der

Retention will be predicted by

Hydrogen bonding, dipoledipole interactions

Relatively weak and transient

Strong, slow interactions

Mobile phase occupies the space

Mobile phase fills the pores of the

Retention Factor (k)

Retention Factor (k)

Example: The Separation of Steroids:

Retention Factor (k):

Peak Asymmetry (Asym)

Peak Tailing due to

Peak Tailing due to

Amitriptyline (pKa 9.4)

Nortriptyline (pKa 9.7)

Protriptyline (pKa 8.2)

Peak Tailing due to

Kinetex 2.6 XB-C18 50x2.1mm

Max. 1.2e5 cps.

Max. 1.2e5 cps.

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

Peak Tailing due to

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

DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000008.D)

DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000010.D)

Peak Fronting due to

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

Fronting due to overload

Peak Fronting due to

Sample in 100% Methanol

Column Efficiency (N)

Columns that cause a lot of peak

Peak Width (W)

Column Efficiency (N) is a

The Core-Shell Advantage

The Core-Shell Advantage

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

The Core-Shell Advantage

Kinetex 2.6 m C18; 150 x 4.6 mm

The Core-Shell Advantage

Particle Diameter (m)

*Farcas et al., HPLC 2012 Conference

The van Deemter Equation

Linear velocity (flow rate)

The Core-Shell Advantage

The Core-Shell Advantage

The Core-Shell Advantage

The Core-Shell Advantage

The Core-Shell Advantage

4.6 x 100 mm Luna-C18 (2)

Reduced plate height h

Reduced plate height h

4.6 x 100 mm Kinetex-C18

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

Effect of Column Length

Balancing Column Length