Vous êtes sur la page 1sur 15

Increasing Sensitivity in HPLC

Dr. Shulamit Levin


Medtechnica
levins@medtechnica.co.il
shulal@zahav.net.il
www.forumsci.co.il/HPLC

Noise and drift

PROPERTIES OF DETECTORS

Noise, drift, and smallest detectable peak.

DETECTION LIMIT

hsignal = 2 x h noise

h signal
h noise

Chromatographic Sensitivity

Increase Signal-to-Noise Ratio

Signal-to-Noise Ratio
6:1

3:1

Signal-to-noise (S/N)
is peak height to
noise

8:1

Increase S/N by
increasing peak
height

0.001
AU

0.2 AU

No
apparent
noise

Increase S/N by
decreasing noise

2.8

3.0

3.2

Minutes

Dr. Shulamit Levin, Medtechnica

3.4

Noise

2.00

3.00

4.00

Minutes

HPLC SYSTEM

Triazine herbicides at detection limit

Control &
Data
Processing

mV

Detector

1. Fucose
2. Galactosamine
3. Glucosamine
4. Galactose
5. Glucose
6. Mannose

300

Conditions:
Gradient: Phosphate-Acetonitrile
Column Novapak 2 x 300 mm
Sample: 2 ppb each pesticide
Injection: 150 l
(0.3 ng on column)

4
6
5

0.00
5.00

Minutes

20.00

Waste

0.0010
0.0008
AU

Simazine
0.0006

Desethylatrazine

0.0004

Fraction
Collector

a b
cd

0.0002

Pump
10.00 15.00

20.00

25.00 30.00
Minutes

35.00

40.00

45.00

flows 50-5000L/min)

Auto
Sampler

HPLC Column
in Oven

Pulse Dampers

Chromatographic System

Flow and pressure profiles for different types of


pumps and cam shape

Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

Dr. Shulamit Levin, Medtechnica

Composition accuracy and consistency

Chromatographic System

High-Pressure Mixing

Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

Waters

1.0

Waters

1.0

Low-Pressure mixing
A

Injector
Mixers

Pump

Pump Head Volume

Compositional Ripple

Larger steps

100 L

Small steps

0.16

0.12

%B
L

0%B

5%B

50 L

Ripple

Ideal

Small heads can deliver


more accurate gradients
when proportioning is
appropriate (RPS).

AU 0.08

Older technology
Baseline "noise" seen
(200 nm)
Insufficient mixing or
non-optimized
gradient formation
New technology

0.04

0.00
36.0

37.0

38.0

39.0

40.0

Minutes

Dr. Shulamit Levin, Medtechnica

Detectors

Chromatographic System

The most common HPLC detectors:


UV/Vis
Fixed wavelength
Variable wavelength
Diode array
Refractive index
Fluorescence
Electrochemical

Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

Less common:
Conductivity
Mass-spectrometric (LC/MS)
Evaporative light scattering

996 and 474 Comparisons


for Aflatoxin Analysis

Detector Criteria

SampleName: Aflatoxin Mix Vial: 2 Inj: 1 Ch: SATIN Type: Standard

55.00

Selectivity
Sensitivity and detection limit
Stability
Linear range
Dynamic Range
Reproducibility
Effect on peak shape
Maintenance

UV at 360 nm 31 point smoothed


UV at 360 nm
Fluorescence 365 ex 455 em

50.00
45.00
40.00

Aflatoxin
G1

35.00
Aflatoxin
G2

mV30.00

Aflatoxin B2

Aflatoxin
B1

25.00
20.00
15.00
10.00
5.00
0.00
4.00

Dr. Shulamit Levin, Medtechnica

5.00

6.00

7.00
8.00
Minutes

9.00

10.00

11.00

12.00

432 Conductivity and 996 PDA


Detectors in Series
1.60

Detection:

Direct Conductivity after


Suppression

1.20
4

5
7

0.80
S

0.40
0.00
3

0.05

Fluoride
Chloride
Nitrite
Bromide
Nitrate
Phosphate
Sulfate

1 ppm
2 ppm
4 ppm
4 ppm
4 ppm
6 ppm
4 ppm

Detection: UV (PDA) at 214 nm

0.04

Column:
Eluent:

0.03
AU
0.02
0.01

1.
2.
3.
4.
5.
6.
7.

Flow rate:
Injection
vol.:

Waters IC-Pak Anion HR


1.2 mM Sodium Carbonate/
1.2 mM Sodium Bicarbonate
1.0 mL/min
50 L

Factors Increasing Signal


Increase sample concentration
Increase injection volume
Choice of wavelength (s)
Low volume flow cell
Flow cell pathlength

Factors Affecting Noise in HPLC Detectors


Optics bench design
Lamp energy
Wavelengths
Mobile phase composition
Pump pulsation
Electronics

0.00
0.00

4.00

8.00

12.00
16.00
Minutes

20.00

24.00

PROPERTIES OF DETECTORS

Decrease Noise

CONTRIBUTION TO BAND
BROADENING

BASELINE STABILITY
SHORT
RANGE
NOISE

RESPONSE TIME

FLOW-CELL VOLUME

LONG
RANGE
DRIFT

Dr. Shulamit Levin, Medtechnica

Increase Response
Effect of Concentration on Spectra

Photodiode Array Detection: Detector Design

Pure Benzoic Acid

2.0

 Transverse Illumination
 Short pathlength
 Low sensitivity

Photodiode

Absorbance

Light source

Different
benzoic acid
concentrations

1.5

Photodiode

1.0

Spectral shape
changes

0.5

Light source
Flow Out

Photodiode

0
190.00

210.00

230.00

250.00

270.00

Light source

290.00

 Shaped Cell
 Long pathlength
 Better sensitivity
 Stray light
 Poor linearity
 Light Guided
 Long pathlength
 Good linearity
 Near theoretical S/N

Fiber Optics

nm

Flow In

Photodiode Array Detection: Flow Cell Performance

Chromatographic System

3.00

Linearity at 205 nm
2.50
CapLC Flow Cell

AU

2.00

Shaped Cell

1.50
1.00
0.50
0.00

Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

Light Guided PDA Flow Cell


0

Concentration

Dr. Shulamit Levin, Medtechnica

LOQ and LOD Relationship to


Column Performances

Peak Broadening

LOQ and LOD can be expressed by:

Cmax =

[1 +

Peak Shape

V inj c 0

(T a 1 ) ]

L d c (1 + k' )
2

Column Dimension
and Performance

Spherical and Irregular particles

Retention

Challenge of Making 2 m Packings


Centered at 3 m
Wider distribution

Centered at 2.4 m
Narrower distribution

Electron microphotograph of spherical and irregular silica particles.


[W.R.Melander, C.Horvath, Reversed-Phase Chromatography, in HPLC
Advances and Perspectives, V2, Academic Press, 1980]

Dr. Shulamit Levin, Medtechnica

Impact of Particle Size (dp)


on Resolution
Conditions:
1.20
1.10
1.00
0.90
0.80
0.70
0.60
AU

0.50
0.40
0.30

Symmetry C18, 3.5 m, 2.1 x 50 mm


Rs (peak A and B) = 11.97
Rs (peak B and C) = 8.79
Rs (peak C and D) = 7.35
C
Theoretical plates (peak D), 24,355
B
A

Columns: Symmetry C18, 5 m, 4.6 X 50


mm and Symmetry C18, 3.5 m, 4.6 X 50
mm
Mobile phase: A=0.1% TFA in water,
B=0.1% TFA in acetonitrile
Gradient: 0-60% B in 4 minutes
Column temperature: 30.0 C
Detector: 254 nm
Injection volume: 1 L

0.20

0.00
0.00

0.90
0.80
0.70
0.60
0.50
AU
0.40
0.30
0.20

Relative Sensitivity
4
5 m, 3.9 mm x 150
mm
3.5 m, 4.6 mm x 100
mm
5 m, 2.1 mm x 150
mm

3.5
3
2.5

0.10

1.00

Sensitivity as a Function of Column


Dimensions

0.50

1.00

1.50

2.00

Minutes

2.50

3.00

3.50

4.00

4.50

Symmetry C18, 5 m, 2.1 x 50 mm


Rs (peak A and B) = 6.71
Rs (peak B and C) = 7.32
Rs (peak C and D) = 6.45
C
Theoretical plates (peak D), 17,872
B
A

1.5
1
0.5

0.10
0.00
0.00

0.50

1.00

1.50

2.00

2.50
Minutes

3.00

3.50

4.00

4.50

5.00

F [mL/min]

Effect of Column Diameter and


Injection Volume on Detectability

Scaling Injection Volume for


Equivalent Resolution
14 L
4.6 mm ID

1.20
Conditions:

00

10.00

Columns:

10 L

2.1 mm 3.0 mm
a) Symmetry C

184.6 mm x 150 mm
b) Symmetry C 183.9 mm x 150 mm

c) Symmetry C
3.9 mm ID

Mobile Phase:

00

183.0 mm x 150 mm

d) Symmetry C

b)

182.1 mm x 150 mm
water/methanol/glacial acetic

10.00

acid 79:20:1

6 L

Flow Rates:
3.0 mm ID

a) 1.4 ml/min
b) 1.0 ml/min
c) 0.6 ml/min

10.00

00

3 L

d) 0.29 ml/min
Sample:

mixture of 6 sulfa drugs

d)
Minutes

10.00

0.80
0.60
0.40
0.20

2.1 mm ID

00

k' = 8, N = 10,000
L = 15 cm

3.9 mm 4.6 mm

1.00
Relative Sensitivity

a)

Waters
23,143

0.00
20

40

60

80

100

120

Injection Volume [L]

Dr. Shulamit Levin, Medtechnica

Tamoxifen: Influence of
Asymmetry on Impurity Profile

Sensitivity as a Function
of Column Diameter

Symmetry C

0.0015
1

12
3

AU

Conditions:
Columns:
Mobile Phase:

Minutes 10.00

5.00
0.0015
12
3

b. 3.9 mm i.d. Flow Rates:

0.0000
Minutes 10.00

5.00
0.0015

Detection:
Sample:

12
3

AU

18

USP Plates

a. 4.6 mm i. d.

0.0000

AU

2 3

Symmetry C

18 150 mm length
50 mM potassium phosphate,
pH 3.0/acetonitrile 55:45

.)
5.00
Minutes

a. 1.4 mL/min
b. 1.0 mL/min
c. 0.60 mL/min
d. 0.29 mL/min
240 nm
tamoxifen, 600 g/mL

Conditions:
Columns:
18

acetonitrile 55:45

.)
Minutes 10.00

0.0015

a.)
b.)
c.)
d.)

AU
d. 2.1 mm i.d.
0.0000
5.00

Resolution
Peaks 1 and 2
1.43
1.46
1.30
0.92

Minutes 10.00

USP Tailing
Peak 3
1.1
1.1
1.1
n.c.

Impact of Column Length


on Resolution (Approach 1)
Symmetry C18, 5 m, 2.1 x 20 mm
Gradient time: 1 minute

Conditions:

Symmetry C18, 5 m, 2.1 x 30 mm


Gradient time: 3 minutes

Symmetry C18, 5 m, 2.1 x 50 mm


Gradient time: 5 minutes

1.00

Mobile phase: A=0.1% TFA in water,


B=0.1% TFA in acetonitrile
Gradient: 0-60% B in noted time
Column temperature: 30.0 C
Detector: 254 nm
Injection volume: 1 L
Flow rate: 1 mL/min.

0.90

0.80

0.70

-Maintain resolution
when scaling gradient
volume proportionally to
column volume.

0.60

0.50
AU
0.40

a) Symmetry C 183.9 mm x 150 mm

b) Zorbax Rx C 184.6 mm x 150 mm


Mobile Phase: 50 mM potassium phosphate, pH 3/

Flow Rates:

0.0000
5.00
12
3

1.4
3.0

7 L injection

c. 3.0 mm i.d.

Conventional
C

a) 6500
b) 1080

10.00

Tailing
Factors

0.30

a) 1.0 ml/min
b) 1.4 ml/min

5.00

10.00
Minutes

Detector:

240 nm

Sample:

600 g/ml, 10 l injection

Water
23,14

Chromatographic System
Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

-Reduce analysis time by


>50%.

0.20

0.10

0.00
0.00

0.50

1.00

1.50

2.00

2.50
Minutes

3.00

3.50

4.00

4.50

5.00

Dr. Shulamit Levin, Medtechnica

PERFORMANCE BY ONE PEAK

Surface of a Reversed- Phase Packing

CH2
H3C

CH3

Si
H

H3C

Si

Si

Si

O
O

Si

O
O

H3C
CH3
H

H3C

Si

Si
O

Si
H

Si

Si

O
O

Or:

t0

N = 5.54 (

Si

Si
O

CH3

O
O

CH3
CH3 H3C
H

Si

Si
O

CH2
H3C
CH3
H

O
O

Si

Si

Si

Si

CH3

O
O

CH2

CH2

CH3

H2C

H2C

H2C

H2C

CH2

CH 2

CH2

CH2

N = 16

H2C

H2C

H2C

H2C

CH2

CH 2

CH2

CH2

tR

H2C

H2C

H2C

H2C

CH2

CH 2

CH2

CH2

NUMBER OF THEORETICAL PLATES

H3C

H3C

H3C

H3C

w1/2

O
O

Example of Lo we re d Pe ak-He ight

T=1.25
AUFS

T=1.66

0.0018

0.0019

0.0013

0.0014

AUFS

1.000

0.800

Symmetry 40nG Procainamide

Zorbax 40 nG Procainamide

T=1.00

0.0008

0.0009

0.600
0.0004

0.0003

0.400
-0.0002
0.00

5.00 10.00 15.00 20.00


T ime

0.200

-0.0001
0.00

5.00

10.00

15.00

20.00

T ime

0.000
0

10

Dr. Shulamit Levin, Medtechnica

20

10

Improvement in Peak Shape for


Bases
neutral
Conventional C
18

Peak Forms of Complexing Agent


(Hinokitiol)
0.2000
0.1900
0.1800
0.1700

Mobile Phase: 50% 20 mM Phosphate


Buffer pH 3.6 with 0.05% EDTA,
50% Acetonitrile

0.1600

base

0.1500
0.1400

1970's
ne utral

0.1300
0.1200

base

Metal-Free

0.1100

Modern C

0.1000

18

AU 0.0900
0.0800
0.0700

Hinokitiol

0.0600

base

0.0500

neutral

0.0400
0.0300
0.0200

Contaminated Column

0.0100

neutral

base

Hybride Silica
Technology -- Today

Time (min)

50

0.0000
-0.0100
2.00

4.00

Minutes

6.00

8.00

Waters Corporation

Chromatographic System
Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

Dr. Shulamit Levin, Medtechnica

11

Solvent Effects
Wavelength nm
200

205

210

215

220

230

240

250

Acetonitrile

0.05

0.03

0.02

0.01

0.01

<.01

Methanol
Degassed

2.06
1.91

1.00
0.76

0.53
0.35

0.37
0.21

0.24
0.15

0.11
0.06

0.05
0.02

0.02
<.01

Isopropyl

1.89

0.68

0.34

0.24

0.19

0.08

0.04

0.03

New THF
Old THF

2.44
>2.5

2.57
>2.5

2.31
>2.5

1.80
>2.5

1.54
>2.5

0.94
>2.5

0.42
>2.5

0.21
>2.5

1% HOAc

2.50

2.54

2.47

2.37

2.16

1.01

0.17

0.04

0.1% HPO4

0.01

0.01

0.01

<0.01

0.1% TFA

1.82

0.87

0.68

0.36

0.22

0.07

<0.01

1% TEA

2.33

2.42

2.50

2.45

2.37

1.96

0.50

Absorbance AU

0.12

432 Conductivity Detection of Seven


Anions with Chemical Suppression

432 Conductivity Detection of


Seven Anion Standard
1.
2.
3.
4.
5.
6.
7.

1.40
3
2

Fluoride
Chloride
Nitrite
Bromide
Nitrate
Phosphate
Sulfate

1 ppm
2 ppm
4 ppm
4 ppm
4 ppm
6 ppm
4 ppm

Column:
Eluent:
Flow rate:
Injection vol.:
Detection:

Waters IC-Pak Anion HC


Borate/Gluconate
2.0 mL/min
100 L
Direct Conductivity

1.60

1.
2.
3.
4.
5.
6.
7.

2
3

1.20

Fluoride
Chloride
Nitrite
Bromide
Nitrate
Phosphate
Sulfate

1 ppm
2 ppm
4 ppm
4 ppm
4 ppm
6 ppm
4 ppm

4
5
1

1.05

0.80
4

S
7

6
0.40

0.00
0.70
0.00

5.00

10.00

Minutes

15.00

20.00

25.00

0.00

4.00

8.00

12.00

16.00

20.00

24.00

Minutes

Basic HPLC Review


pH Effects on UV Spectra (L-Ascorbic Acid)

270
Lambda
Max

pH 12.0

266
pH 8.7
258
pH 4.9

245

pH 4.0
245
pH 2.0

Chromatographic System
Pump
Solvent proportioning
Detector
Column Perfromance & Geometry
Stationary phases' chemistry
Mobile phases' chemistry
Integration events

200 220 240 260 280


Wavelength

Dr. Shulamit Levin, Medtechnica

12

Integration Errors
Caused by Tailing

Peak Detection:
Peak Apex

T = 1.00
Recovered Peak Areas

T = 1.58
Recovered Peak Areas

99.9 %

97.8 %

99.8 %

95.3 %

99.6 %

92.3 %

Peak Height

Constructed
Baseline

Start of
chromatogram

Peak End

Peak Start

Retention Time

Exponential Skim

Tangential Skim

Resolves rider or shoulder peaks from a parent peak.

Integration of Small Peaks


PW=60
Area=25660

PW=15
Area=23011

-10%

AUFS = 0.003
Peak width changed
Threshold set at 30

3.80
4.20
Minutes

4.60

Thresh.=300
Area=23087

3.80
4.20
4.60
Minutes
Thresh.=
20
Area=25660

AUFS = 0.003

-10%
Peak width set at 30 sec

The Tangential and Exponential Skim events require the use of a


Force Drop Line within the parent peak. For best results, add
only a single drop line per peak that is to be skimmed .

Dr. Shulamit Levin, Medtechnica

Threshold changed
3.80
4.20
Minutes

4.60

3.80

4.20
Minutes

4.60

13