Philip J. Koerner, Ph.D.
ThailandSeptember 2013
General Principles General Principles of HPLC Method of HPLC Method
DevelopmentDevelopment
Part 1. General Chromatographic Theory
Part 2. The Stationary Phase:
An Overview of HPLC Media
Part 3. Role of the Mobile Phase in Selectivity
3
The Liquid Chromatographic Process
4Mikhail Tsvet, 1872-1919
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
Column Chromatography:
5
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
6
The Liquid
Chromatographic Process
Mobile Phase
Stationary Phase
Analytes
7
Mobile Phase
Stationary Phase
Analytes
The Liquid
Chromatographic Process
8
Non-Polar Stationary Phase (e.g. C18)
O H
H
OH
H
O
HH
N
N
N
N
N
O H
H
O
HH
Polar/AqueousMobile Phase
Benzo(a)pyrene Ethyl sulfate
The Liquid
Chromatographic Process
9
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.
OH
N
CH3
CH3
CH3
CH3
Tapentadol
Mechanisms of Interaction
In RP Chromatography
10
Hydrophobic Interactions
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO HCH 3
OH O H OH O H O H O H OH O H O H O H
O O O- O H
O O OO
S iS i S i
S iS iCH 3
C H 3
CH 3C H 3
CH 3C H 3
CH 3 C H 3
C H 3C H 3 C H 3 CH 3
C H 3
C H 3
C H 3O H
OH
N
CH3
CH3
CH3CH3
Hydrophobic• Weak, transient interactions between a non-polar stationary phase and the molecules
• Hydrophobic & van DerWaals interactions
• Retention will be predicted by
Log P values
11
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO HCH 3
OH O H OH O H O H O H OH O H O H O H
O O O- O H
O O OO
S iS i S i
S iS iCH 3
C H 3
CH 3C H 3
CH 3C H 3
CH 3 C H 3
C H 3C H 3 C H 3 CH 3
C H 3
C H 3
C H 3O H
OH
N
CH3
CH3
CH3CH3
Polar
• Interactions between polar functions groups of analyteand residual silanols or polar groups on media
• Hydrogen bonding, dipole-dipole interactions
• Relatively weak and transient
Polar Interactions
12
Ion-exchange Interactions
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO HCH 3
OH O H OH O H O H O H OH O H O H O H
O O O- O H
O O OO
S iS i S i
S iS iCH 3
C H 3
CH 3C H 3
CH 3C H 3
CH 3 C H 3
C H 3C H 3 C H 3 CH 3
C H 3
C H 3
C H 3O H
OH
NH+
CH3
CH3
CH3CH3
Ion-Exchange
• Interactions between ionizable functional groups on analyte and counter-charged moiety on stationary phase
• Ion-exchange
• Strong, slow interactions
13
Chromatographic Measurements
14
Chromatographic
Measurements
k’ Asym N α
15
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.
16
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
17
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.
t0
Retention factor (k’)=
tR – t0
t0
Retention Factor (k’)
18
Retention Factor (k’)
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
19
Retention Factor (k’):
Effect of Changing % ACN
20
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
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 silanolson the surface of the silica.
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO
S iO HCH 3
OH O H OH O H O H O H OHO H O H O H
O O O- O H
O O OO
S iS i S i
S i
S iCH 3
C H 3
CH 3C H 3
CH 3C H 3
CH 3 C H 3
C H 3C H 3 C H 3
CH 3
C H 3
C H 3
C H 3O H
O H
NH+
CH 3
C H 3
C H 3CH 3
Ion-Exchange
21
2222
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
Peak Tailing due to
Secondary Interactions
Amitriptyline (pKa 9.4) Nortriptyline (pKa 9.7) Protriptyline (pKa 8.2)
23
Peak Tailing due to
Secondary Interactions
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.
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 6.0Time, min
1
2 3
45
6
7 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. 1.1e5 cps.
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 6.0Time, min
4
6
Mobile phase: A = 0.1% Formic acid in water, B = 100% Methanol,Gradient: 15 to 60% B in 5 min, hold for 1 minFlow rate: 400 µL/min1. Amoxapine2. Imipramine3. Desipramine4. Protriptyline*5. Amitriptyline6. Nortriptyline*7. Clomipramine8. Norclomipramine
XIC of +MRM (8 pairs): 278.2/191.2 Da ID: Amitrip from Sample 7 (TCA-Kin-XB-C18, 50x2, 2.6, MeOH 0.... Max. 1.2e5 cps.
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 6.0Time, min
2 3
1
4
5
6
78
1
5
XIC of +MRM (8 pairs): 264.2/191.1 Da ID: Protrip from Sample 13 (Kinetex XB-C18-50x2, 2.6 um, MeO... Max. 1.2e5 cps.
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 6.0Time, min
4
6
Kinetex 2.6µ XB-C18 50x2.1mmBrand H 2.7µ C18 50x2.1mm
Peak tailing
24
Strongly basic analytes are very sensitive to sample loading, and will display peak tailing effects as a function of increasing loading (µg on column):
min12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
mAU0
20
40
60
80
100
DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000010.D)
14.157
12 µg on column; USP Tailing = 1.87
min12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
mAU0
10
20
30
40
50
DAD1 C, Sig=254,4 Ref=300,100 (DJ040611\DJGSK 2011-04-06 12-28-48\GSK000008.D)
14.169
5 µg on column;
USP Tailing = 1.34
min12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
mAU0
5
10
15
20
25
30
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;USP Tailing = 1.17
Peak Tailing due to
Sample Overloading
pKa 9.7
25
min1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7
mAU0
250
500
750
1000
1250
1500
1750
2000
2250
DAD1 A, Sig=240,10 Ref=off (MT042211\DJGSK 2011-04-22 09-12-57\MUPIROCIN000001.D)
1.745
1.905
2.145
2.495
Fronting due to overload
Peak Fronting due to
Sample Overloading
Neutral and acidic compounds will typical show peak fronting when the column is overloaded.
Detector saturation!
26
Peak fronting can also be due to sample solvent effects:(1) Sample insolubility
(2) Sample solvent is too strong:
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.00.0
2.0e4
4.0e4
6.0e4
8.0e4
1.0e5
1.2e5
1.4e5
1.6e5
1.8e5
2.0e5
2.2e5
0.24
1.791.36
0.970.51 1.71 1.85 3.972.15 3.213.36
Sample in 100% Methanol
Peak Fronting due to
Sample Solvent Effects
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.00.0
5000.0
1.0e4
1.5e4
2.0e4
2.5e4
3.0e4
3.5e4
4.0e4
4.5e4
5.0e4
5.5e4
6.0e4
6.5e4
7.0e4
1.06
1.77
Sample in 50% Methanol
Morphine
Hydromorphone
Norhydrocodone
Breakthrough!
Fronting
27
Selectivity (αααα)
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
2828
The choice of stationary phase will often have a dramatic effect on the selectivity of analytes.
OH
H H
OCH
3
H
Estrone
OH
H H
OHCH
3
H
Estradiol
Selectivity (α)
m i n1 2 3 4 5
m A U
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 7 5 C18 Column1+2
α = 1.0
Phenyl Column 1
2
m i n1 2 3 4 5
m A U
0
2 0
4 0
6 0
8 0
1 0 0
α = 2.3
29
Selectivity (αααα)
But mobile phase is also a very powerful tool in modulating selectivity.
35% Methanol
20% Acetonitrile
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/minDetection: UV at 220nm
1. Saccharin2. p-Hydroxybenzoic Acid3. Sorbic Acid4. Dehydroacetic Acid5. Methylparaben
30
Column Efficiency (N)
W
W1/2
tR
Peak Width (W)
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 broadening have few theoretical plates.
• Columns that produce very narrow peaks (little peak broadening) have a large number of theoretical plates.
31
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
50,000 P/m
5 µm
100,000 P/m
3 µm
150,000 P/m
sub-2 µm
300,000 P/m
Efficiency
Back-Pressure
Core-Shell
300,000 P/m
Columns packed with coreColumns packed with core--shell particles will deliver shell particles will deliver
significantly higher efficiency (N) than columns packed with significantly higher efficiency (N) than columns packed with
fullyfully--porous particles of the same diameter.*porous particles of the same diameter.*
6
w1/2
tRInjected
Sample Band
Fully Porous Particles
* Gritti et al., Journal of Chromatography A, 1217 (2010) 1589
The Core-Shell Advantage
Fully Porous Particles
Columns packed with coreColumns packed with core--shell particles will deliver shell particles will deliver
significantly higher efficiency (N) than columns packed with significantly higher efficiency (N) than columns packed with
fullyfully--porous particles of the same diameter.*porous particles of the same diameter.*
Kinetex™ Core-Shell Particles
5 6
w1/2
tR
* Gritti et al., Journal of Chromatography A, 1217 (2010) 1589
The Core-Shell Advantage
Core-Shell Particles
34
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!
Columns packed with coreColumns packed with core--shell particles will deliver shell particles will deliver
significantly higher efficiency (N) than columns packed with significantly higher efficiency (N) than columns packed with
fullyfully--porous particles of the same diameter.*porous particles of the same diameter.*
* Gritti et al., Journal of Chromatography A, 1217 (2010) 1589
The Core-Shell Advantage
Comparison
*Farcas et al., HPLC 2012 Conference
Columns packed with coreColumns packed with core--shell particles will deliver shell particles will deliver
significantly higher efficiency (N) than columns packed with significantly higher efficiency (N) than columns packed with
fullyfully--porous particles of the same diameter.*porous particles of the same diameter.*
0
50
100
150
200
250
300
350
400
450
5 3.5/3.6 2.5/2.6 1.7 1.3
Particle Diameter (µm)
Eff
icie
ncy(
p/m
)
Fully Porous Core-Shell
The Core-Shell Advantage
Efficiency vs. Diameter
36
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
A · dp + B/µ + C · de2 · µH =
Simplified version:
Particle size
Linear velocity (flow rate)Mobile phase viscosity
Particle sizeLinear velocity (flow rate)Mass Transfer
A · dp + B/u + C · de2 · uH =
Eddy Diffusion
Multi-path Effect
6
w1/2
tR
The Core-Shell Advantage
A-term
A · dp + B/u + C · de2 · uH =
Longitudinal diffusion
6
w1/2
tR
The Core-Shell Advantage
B-term
39
Dispersion due to resistance to mass transfer
A
B
A · dp + B/µ + C · de2 · µH =
Mass Transfer (Kinetics)
The Core-Shell Advantage
Fully Porous C-term
Dispersion due to resistance to mass transfer
A
B
A · dp + B/µ + C · de2 · µH =
Mass Transfer (Kinetics)
The Core-Shell Advantage
Core-Shell C-term
A · dp + B/µ + C · de2 · µH =
0 5 10 15 200
5
10
15
4.6 x 100 mm Kinetex-C18
Re
du
ce
d p
late
he
igh
t h
Reduced velocity νννν
Eddy dispersion
Longitudinal diffusion
Solid-liquid mass transfer
*Gritti and Guiochon, 2012. LC-GC 30(7) 586-595.
0 5 10 15 200
5
10
15
4.6 x 100 mm Luna-C18
(2)
Eddy dispersion
Longitudinal diffusion
Solid-liquid mass transfer
Re
du
ce
d p
late
he
igh
t h
Reduced velocity νννν
The Core-Shell Advantage
h vs. v Comparison
42
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.
Effect of Column Length
on Efficiency
25cm 25,000 Plates40 min250 Bar
15cm 15,000 Plates24 min150 Bar
10cm 10,000 Plates16 min100 Bar
5cm 5,000 Plates8 min50 Bar
43
Balancing Column Length
and Particle Size
Column
Length
(mm)
Efficiency dp
5µµµµm
250 25,000
150 15,000
100 10,000
50 5,000
44
Column
Length
(mm)
Efficiency dp
5µµµµm
Efficiency dp
3µµµµm
250 25,000 37,500
150 15,000 22,500
100 10,000 15,000
50 5,000 7,500
Balancing Column Length
and Particle Size
45
Column
Length
(mm)
Efficiency dp
5µµµµm
Efficiency dp
3µµµµm
Efficiency
sub-2µµµµm /
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
Balancing Column Length
and Particle Size
46
Column
Length
(mm)
Efficiency dp
5µµµµm
Efficiency dp
3µµµµm
Efficiency
sub-2µµµµm /
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!!
Balancing Column Length
and Particle Size
47
Effect of Flow Rate on Column Efficiency (100x4.6mm)
Effect of Flow Rate on
Efficiency
0
5000
10000
15000
20000
25000
30000
35000
0 0.5 1 1.5 2 2.5 3 3.5
Flow rate (ml/min)
N (
Pla
tes/c
olu
mn
)
Core-Shell 2.6
Luna 3u
Luna 5u
5µ ~1 mL/min
3µ ~1.5 mL/min
Core-Shell ~2 mL/min
48
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 (αααα) – 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)
49
Any Questions?Any Questions?
50
Resolution: The Goal of Chromatography
51
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.
52
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)
Resolution: The Goal of
Liquid Chromatography
53
(1) Retention time for the two peaks will be a function of retention factor (k’).
(2) The selectivity (αααα) 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).
k’
α
N, Asym
Resolution: The Goal of
Liquid Chromatography
54
Selectivity
Retention factorEfficiency
The equation below allows us to calculate the relative importance of each of
these three factors in overall resolution:
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 (αααα) → Stationary phase and mobile phase3. Retention factor (k’) → Stationary phase and mobile phase
Resolution: The Goal of
Liquid Chromatography
55
Resolution: The Relative
Effectiveness of k’, αααα, and N
Ineffective after k’ ~10
Constant increase in resolution
Most important determinant of resolution!!
56
The Impact of Efficiency
on ResolutionModulating 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
Length
(mm) Efficiency 5µµµµm Efficiency 3µµµµm
Efficiency
sub-2µµµµm /
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
More efficiency/resolution
Lo
ng
er
Ru
nT
ime
More Back-Pressure
Mo
re P
res
su
re
57
The Impact of Efficiency
on Resolution
0
0.5
1
1.5
2
2.5
0 10000 20000 30000 40000
Rel
ativ
e R
esol
utio
n
Column Efficiency (Plates)
Doubling column
efficiency increases
Rs by a factor of 1.4x
The Impact of Efficiency
on Resolution
min0 0.5 1 1.5 2 2.5 3 3.5
mAU
0
20
40
60
80
100
5 µm
80,000 P/m
min0 0.5 1 1.5 2 2.5 3 3.5
mAU
0
20
40
60
80
100
120
140
3 µm
150,000 P/m
58
59
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
60
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.
61
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
62
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.
63
Method Development Exercise 1:
Optimization to Reduce Analysis Time and
Increase Productivity
64
Column: C8 250 x 4.6mm 5µm
Mobile phase: 70 / 30 0.1M Ammonium acetate / THF
Flow rate: 1.0 mL/min
Components: 1-6 = Impurities A - G
7. Mupirocin
Mupirocin Impurity Profile
Mupirocin
min0 2 4 6 8 10 12 14 16
mAU0
25
50
75
100
125
150
175
DAD1 A, Sig=240,10 Ref=off (Z:\1\DATA\MT050211\DJGSK 2011-05-02 15-34-44\MUPIROCIN000001.D)
Mupirocin: Original Method
1
2 3
5
6
7
~16 min
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
4
Rs 3/4 = 0.63; k’ = 2.3
65
66
Step 1. Adjust k’ for better
Resolution
Step 1. Reduce % organic to increase k’:• Increases Rs
• Increases run time
67
Step 2. Optimize Efficiency
and Length
Column
Length
(mm)
Efficiency dp
5µµµµm
Efficiency dp
3µµµµm
Efficiency
sub-2µµµµm /
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 µm media:• Reduces analysis time• Maintains efficiency
0
5000
10000
15000
20000
25000
30000
35000
0 0.5 1 1.5 2 2.5 3 3.5
Flow rate (ml/min)
N (
Pla
tes/c
olu
mn
)
Core-Shell 2.6
Luna 3u
Luna 5u
68
Step 3. Optimize the Flow
Rate
Step 3. Increase flow rate to 1.5 mL/min:• Optimizes efficiency for 3 µm
min0 2 4 6 8 10 12 14 16 18
mAU
0
20
40
60
80
VWD1 A, Wavelength=240 nm (JL050211\MUPI0003.D)
Mupirocin: Intermediate
Method
1
23
4
5
6
7
Column: Luna 3µ C8(2) 150 x 4.6mm
Mobile phase: 80/20 0.1M Ammonium acetate pH 5.7/THFFlow rate: 1.5 mL/min
Rs increased from 0.63 to 1.6
Run time increased from 16 to 20 minutes
20 min
k’ = 9; Rs 3/4 = 1.6
69
70
Step 4. Switch to Core-Shell Media
Column
Length
(mm)
Efficiency dp
5µµµµm
Efficiency dp
3µµµµm
Efficiency
sub-2µµµµm /
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
min0 1 2 3 4 5 6 7 8 9
mAU
0
50
100
150
200
250
VWD1 A, Wavelength=240 nm (JL050211\MUPI0006.D)
Mupirocin: Final Optimized Method
Column: Kinetex 2.6µ C8 100 x 4.6mm
Mobile phase: 80/20 0.1M Ammonium acetate pH 5.7:THFFlow rate: 1.5 mL/min (2mL/min if pressure allows)
Rs increased from 1.6 to 2.3Run time decreased from 20 to 8 minutes
8 min
1
5
6
7
23
4
Rs 3/4 = 2.3
71
Mupirocin: Final Optimized Method
min0 1 2 3 4 5 6 7 8 9
mAU
0
50
100
150
200
250
VWD1 A, Wavelength=240 nm (JL050211\MUPI0006.D)
8 min1
5
6
7
23
4
Rs 3/4 = 2.3
Final Optimized Method:
m in0 2 4 6 8 1 0 12 14 16
m AU0
25
50
75
1 00
1 25
1 50
1 75
D AD 1 A , S ig= 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
2 3
5
6
7
16 min
4
Rs 3/4 = 0.63
Initial Method:
72
73
Any Questions?Any Questions?
74
Part 1. General Chromatographic Theory
Part 2. The Stationary Phase:
An Overview of HPLC Media
Part 3. Role of the Mobile Phase in Selectivity
75
Tetraethoxysilane Silica Sol-Gel
Polymerization
Alkaline
75
99.9% of reactive surface area is internal
Fully Porous Silica
76
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
Fully Porous Silica
7777
Conventional Silica Particle Organosilica Hybrid Particle
Dissolution at pH > 7.5 Stable to pH ~12
SiloxaneBridge
Ethane linkage
Organosilica Hybrid Particle
78
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
Organosilica Hybrid Particle
79
1.9 µm Solid Core
0.35 µm Porous Shell
2.6 µm Core-Shell
Particle
79
Core-Shell Particle
80
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
Core-Shell Particle
81
Alkyl bonded phases (C18, C8, C4):
Phenyl phases (Phenyl, PFP):
F F
F
FF
Polar-embedded phases:
Fusion
Polar-endcapped phases:
Hydro
RP Stationary Phase Classes
82
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 i n2 4 6 8 1 0
m A U
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
m i n2 4 6 8 1 0
m A U
0
1 0 0
2 0 0
3 0 0
4 0 0
C18
Ph
en
yl
Methylene Selectivity
83
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
0.180
0.200
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
Slo
pe
of l
og
k v
s. #
-C
H2-
un
its
C18 > C8 > C5 ≥ Phenyl > CN > Amino
Methylene Selectivity
84
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
O
H H
CH3
H
CH3
OH
CH3
Methyltestosterone
O
H H
CH3 H
CH3
OH
Testosterone
Methylene Selectivity
Columns: C18
Phenyl
Dimensions: 150 x 4.6 mm
Mobile phase: 75:25 Methanol:water
Flow rate: 1 mL/min
Components: 1. Estrone
2. EstradiolOH
H H
OHCH
3
H
Estradiol
OH
H H
OCH
3
H
Estrone
85
Phenyl Selectivity
86
m in1 2 3 4 5
m A U
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 7 5
m in1 2 3 4 5
m A U
0
2 0
4 0
6 0
8 0
1 0 0
C18
Phenyl
1+2
1
2
OH
H H
OHCH
3
H
Estradiol
OH
H H
OCH
3
H
EstroneC18
Phenyl
Phenyl Selectivity
87
Nucleic Acid Bases:
Day 2
min2 4 6 8 10 12 14
0
Day 1
2 4 6 8 10 12
0
Phase Collapse!
Luna C18(2)
Day 1
0 2 4 6 8 10 12
0
Day 6
2 4 6 8 10 12 14
0
Polar-Endcapped C18
Aqueous Stability of
Embedded Phases
88
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.
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.549 96 144 191 239 287 334 382 429 477 525 572 620
Time, min
0.00
1000.00
2000.00
3000.00
4000.00
5000.00
6000.00
7000.00
8000.00
9000.00
1.00e4
1.10e4
Intensity, cps
ETG
ETS
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.
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.549 96 144 191 239 287 334 382 429 477 525 572 620
Time, min
0.00
5000.00
1.00e4
1.50e4
2.00e4
2.50e4
3.00e4
3.50e4
4.00e4
4.50e4
5.00e4
5.50e4
6.00e4
6.50e4
7.00e4
7.50e4
8.00e4
8.50e4
9.00e4
9.50e4
1.00e5
Intensity, cps
ETG
ETS
Polar-Endcapped 2.5 µm C18 100x3.0mm
10mM Ammonium formate
1. Ethyl glucuronide2. Ethyl sulfate
ETG
ETS
Aqueous Stability of
Embedded Phases
Part 1. General Chromatographic Theory
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
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).
Acetonitrile has a dipole
moment but is only a very weak proton acceptor in hydrogen
bonding.
N CH3δ+δ−
Tetrahyrofuran is able to
accept a proton in hydrogen bonding but cannot donate a
proton.
O
Methanol is a strong proton
donor and a strong proton
acceptor in hydrogen bonding.
CH3 OH
Solvent Selectivity
Optimum Separation of 4 Steroids in Different Solvents:
Solvent Selectivity
1. Start at high % acetonitrile and work backwards until k’ is 2-10 (if possible)
80% ACN
k’ = 0
25% ACN
k’ ~ 6
21% ACN
k’ ~ 11
40% ACN
k’ ~ 0.8
Solvent Screening for
Isocratic Methods
2. Repeat with alternative solvent:
Solvent Screening for
Isocratic Methods
= Typical for LC/MS
= Typical for LC/UV
Buffer Selection for
RP-HPLC
O
Si
O
Si
O
Si
OH
OH
OH
O
O
O
Si
Si
Si
pH <3.5
O
Si
O
Si
O
Si
O
OH
O
O
O
O
Si
Si
Si
pH >3.5
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.
• Silanols protonated• Less ion-exchange
• Less peak tailing
• Silanols deprotonated• Increased ion-exchange
• Increased peak tailing
Effect of pH on
Base Silica
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.
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.
• More hydrophobic
• More strongly retained
• Less hydrophobic
• Less strongly retained
• More hydrophobic
• More strongly retained
• Less hydrophobic
• Less strongly retained
Effect of pH on Analyte
IonizationR
ete
nti
on
Facto
r (k
’)
Acidic Compounds:
Rete
nti
on
Facto
r (k
’)
Basic Compounds:
Effect of pH on Analyte
Ionization
H+
Alkaline
Acidic
Aqueous Mobile PhaseAlkyl Stationary Phase
Effect of pH on Analyte
Ionization
H+
Alkaline
Acidic
Aqueous Mobile PhaseAlkyl Stationary Phase
H+
Effect of pH on Analyte
Ionization
B
NN
A
B
NN
AA
NN
B
Amitriptyline (pKa 9.4) = (B)ase Toluene = (N)eutral Naproxen (pKa 4.5) = (A)cid
Effect of pH on Analyte
Retention
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.
Gradient Analysis
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.
Temperature in
HPLC Methods
104
End of Section II
Any Questions?Any Questions?
End of the HPLC Method Development Portion