controlling the big bang in uhplc separations! · the big bang! dispersion begins at injection of...

Controlling the Big Bang in UHPLC Separations!

Agilent Technologies

2013

What Are the Goals of the Separation Improvement?

Speed

Resolution

Increased Sensitivity

A Combination of the Above

Any Improvement in Performance May be Compromised if the

Separation System is not Optimized and Controlled

What Needs to be Controlled in the System?

• Injection

• Column Efficiency

• Detection

• Sample Path Volume

What Is the Sample Path?

.

.

Any Volume the Sample “Sees”

• Injection Volume

• Connecting Tubing

• Column

• Detector

System Volume and Dispersion

• Since the Sample is in Solution ALL Volume Components

Increase Peak Dispersion

• Volume is your enemy in High Performance LC!

• A flow path is necessary to move the sample through the

system so Extra Column Volume (ECV) is necessary

• ECV in the system can be controlled and dispersion effects

minimized

Group/Presentation Title

Agilent Restricted

Month ##, 200X

Column and Extra Column Dispersion

• The Van deemter equation and graph describes dispersion in column during the separation process

• There are other factors that contribute to dispersion:

• σ2in - dispersion in the injector

• σ2 tu - dispersion in the

connecting tubes

• σ2 det - dispersion in the detector

• These are often grouped together under the term

Extra Column Volume (ECV)

The Big Bang!

Dispersion Begins at Injection of the Sample

Bonded phase

Bonded phase

• Differential partitioning of the components into the stationary and

mobile phases.

• Separation controlled by chemical interaction of mobile

phase/sample/bonded phase

Mobile phase

The Separation Process


Agilent Restricted

Month ##, 200X

Chromatographic Separation and

Band Broadening

• Peaks Enter the column in a Single Band

• Through Differential Partitioning Separate Bands Form

• As Bands Move Through Column Length More Separation Takes Place

• Bands Also Become Broader

• Band Width on Exiting the Column is Effected by Column Efficiency and In-Column Peak Dispersion (σ2)

W1 W2 ΔL L

Dispersion Within the LC Column

a) Longitudinal Diffusion (dispersion)

b) Radial Diffusion (dispersion)

Direction of Flow

Learn How to Optimize Your Auto Sampler Keep Control of Sample Injection

Delay Volume Reduction and Overlapped Injections

Main Pass Bypass

Simultaneous Speed and CO optimization by

• Simple „one-valve“ flow-though design

• Automatic delay volume reduction (6µl)

• Overlapped injection

• Programmable sample flush-out time

• External needle wash

• MCO by additional valve switches

Too Strong Sample Solvent Causes Peaks

to Move Too Fast, Too Far into Column

• Solvent Stronger than Mobile Phase prevents a

significant portion of Sample from adsorbing onto

packing

• Bands become spread out relatively far into column

• By the time the Mobile Phase flushes the strong

solvent the bands are very broad and may form split

(two) peaks

Use Care in Choosing the Sample Solvent Example of Too Strong Sample Solvent

Column: StableBond SB-C8, 4.6 x 150 mm, Mobile Phase: 82% H2O : 18% ACN

Injection Volume: 30 mL Sample: 1. Caffeine 2. Salicylamide

A. Injection Solvent

100% Acetonitrile B. Injection Solvent

Mobile Phase

0 10

Time (min)

0 10

Time (min)

1

2

1

2

• Use Sample Solvent No Stronger than Mobile Phase

• Ideal Solvent is 100% Aqueous in RP LC

Standard Analytical Solvent Saver Narrow Bore

Column

Cross Section

Column Internal

Diameter 4.6 mm 3.0 mm 2.1 mm

Flow Rate/min. 1.00 mL 0.40 mL 0.20 mL

Inj. Vol. Ratio 1 0.4 0.2

Relationship of Column i.d. to

Flow Rate and Injection Volume

Confidentiality Label

July 31, 2013 15

No Change in Flow as Column i.d. Decreases

Leads to Reduced Resolution

0 1 2 3 4 5 6

Rs = 1.32

Rs = 1.65

Time (min)

4.6 x 150 mm

3.0 x 150 mm

2.1 x 150 mm

R s = 0.95

483 psi

1135 psi

2316 psi

Flow Rate: 1.0 mL/min


July 31, 2013 16

Maintaining Linear Velocity on Columns of

Different i.d. Holds Resolution

0 40Time (min)

1

2 3

4

56

0 40

1

23

4

56

0 40

1

23

4

5

6

Time (min)Time (min)

4.6 x 150 mm 3.0 x 150 mm

2.1 x 150 mm


Injected: 3 uL

Detector Cell Volume: 8 uL


Injected: 2 uL



Injected: 1 uL


Width and Volume Lowest for Earliest Eluting Peaks Narrower Peaks Reduce Allowable Dispersion Volume

Peak Volume (mL)

Column

Dimension

Void Volume (mL)

k=1 k=3 k=5

4.6 x 150 mm

1.0 mL/min 1500 114 229 343

3.0 x 150 mm

0.4 mL/min 640 46 92 137

2.1 x 150 mm

0.2 mL/min 280 23 46 69

k = (tr-to)/to

N = 11,000 (constant)

• Peak volumes below 60 mL require optimized instrumentation for maximum efficiency.

As Efficiency Increases Peak Width Decreases System Dispersion Becomes More of a Factor

Columns: Eclipse Plus C18, as described below. Mobile Phase: A: water, B: MeOH, (15:85) Injection volume: 6uL

Temperature: 25°C Flow: 1 mL/min. Detection: 310, 4 nm, 0.5 s response time, semi-micro flow cell, Sample: Sunscreens

min 0 2 4 6 8 10 12 14

mAU

0

20

40

60

min 0 2 4 6 8 10 12 14

mAU

0

20

40

60

80

100

min 0 2 4 6 8 10 12 14

mAU

0

50

100

150

4.6 x 100 mm, 3.5 µm

1

2 3

4

Rs3,2= 6.65

Rs3,2= 6.51

Rs3,2= 6.41

4.6 x 50 mm, 1.8 µm

4.6 x 150 mm, 5 µm

Approximate Extra Column Volume Limits

for This 4.6mm i.d. Column Example

5um 3.5um 1.8um

Peak (1) Width in uL 181 136 45

Injection (1/10 P.V.) 18 14 5

Extra Column (1/3 P.V.) 60 45 15

ECV minus Inj. Vol. 42 21 10

ECV Estimate for

2.1mm i.d. Column

8 4 2

Injection and Flow Parameters are Optimized

What Now?

• Weakest Possible Sample Solvent

• Flow Rate/ Linear Velocity Appropriate to Column

Diameter

• Injection Volume Scaled to Peak Width

• Next Step is to optimize ECV

How Does Instrument Design Effect Performance?

• The Flow Path the Sample “Sees” from injection to Detection

Contributes to Dispersion and Peak Broadening

• Limiting excess volume will greatly reduce dispersion impact

on performance

• UHPLCs (Agilent 1290 Infinity) are already optimized for Sub-

2u , Poroshell and HPLC column performance

• What about other instruments?

1290 Infinity UHPLC

//nwadmc01/Field_Info/2_Analytical_LC/4_Video_Audio_Visuals/3_Photos/1_Systems

Optimizing the LC SYSTEM

• Connecting Tubing

• Proper Connection Technique

• Data Collection Rate for Detectors

Instrument Impact on Column Performance

.

Extracolumn Volume

Data Sampling ( or Acquisition) Rate

.

Number of

Scans or points

0 1 2 3 4 5 6 7 8 9 10 11

Effect of Extra-Column Volume

Extra-

Column

Volume

10 µL

20 µL

Time, min.

2.1 x 150 mm

F = 0.2 mL /

min.

001015P1.PPT

Loss of

Resolution


Agilent Restricted

Month ##, 200X

Dispersion in the Tubing

Dispersion Calculation

σt2 = π2r6Lu/24Dm

Dispersion in the tubing is proportional to the

• Length of tubing

• 6th power of the tube radius

Shortest tubing lengths possible minimize dispersion

Small changes in tubing i.d. have major effects on

peak width and efficiency

Tubing Volume

Tubing Length 10mm 50mm 100mm 150mm

Tubing i.d. Volume Volume Volume Volume

0.17mm (green) 0.227 uL 1.1uL 2.27 uL 3.3 uL

0.12mm (red) 0.113 uL 0.55uL 1.13 uL 1.65 uL

Conversion for Fast and Ultra-Fast HPLC 1200 through 1260 Series LC Systems

High pressure

Gradient pump

Std or Well

Plate sampler

Diode Array

Detector

Standard assembly

without standard mixer

0.12 x 400 mm capillary

Mass

Spectrometer

0.12 x XX mm PEEK Capillary

High pressure

Gradient pump

Std or Well

Plate sampler

Rapid Resolution

HT Column

Diode Array

detector

Waste




3 mL

heat exchanger

Thermostatted

Column

compartment

Traditional LC Columns Fast LC/UHPLC Columns

Rapid Resolution

HT Column



3 mL

heat exchanger

Thermostatted

Column

compartment

Optimizing Connecting Tubing Volume

For UHPLC Columns

min 0.5 1 1.5 2 2.5

mAU

0

100

200

350

400

550

600 System Tubing Volume Optimized

0.12mm i.d. tubing

Peak width 0.018 min


Resolution 1.902

min 0.5 1 1.5 2 2.5

mAU

0

100

200

300

400

System Tubing Volume Not Optimized

0.17mm i.d. tubing



Resolution 0.961

Smaller Column i.d. Requires a Lower

Detector Cell Volume

July 31, 2013


29

Black – 14 uL flow cell

3 x 100mm Column

Blue – 5 uL flow cell

Agilent Method Translator – Advanced Mode Can be used to model Extra Column Volume Effects on Efficiency

What Happens If a Connection is Poorly Made ?

If Dimension X is too long, leaks will occur

Ferrule cannot seat properly

Mixing Chamber

If Dimension X is too short, a dead-volume,

or mixing chamber, will occur

Wrong … too long, but obvious

It Leaks!

Wrong … too short, not so

obvious and Causes Dispersion! X

X

min 0 0.1 0.2 0.3 0.4

mAU

0

20

40

60

80

100

120

140 One bad capillary connection!

min 0 0.1 0.2 0.3 0.4

mAU

0

30

60

90

120

150

180

210 Fixed!

130 mAU

160 mAU

Influence of Bad Post-Column Connection

Effect of Data Acquisition Rate (time constant) High definition UHPLC Requires High Definition Chromatogram

min 0.1 0.2 0.3 0.4 0.5 0

80Hz

PW=0.30sec

40Hz

PW = 0.33 sec

20 Hz

PW=0.42sec

10Hz

PW=0.67sec

5Hz PW=1.24sec

• Increased Data Rate

• More Accurate “ Picture”

• Make Sure Rate is Adequate

• Faster Rates Generate More

Noise and Take up More Memory

Solve Apparent Rs Problems with Optimum

Data Collection Rate

min 0 0.1 0.2 0.3 0.4 0.5

mAU

0

20

40

60

80

100

min 0 0.1 0.2 0.3 0.4 0.5

mAU

0

10

20

30

40

50

Peak width = 0.017min at 80Hz

Peak width = 0.021min at 10Hz

Max Performance RRHD - UHPLC/TOF (1290/6230)

Identify More Compounds in Very Short Run Time

1 .5

min

Time Composition

0.0 10% ACN

1.5 100% ACN

224 pesticides at 50 pg each

217 ionized & detected in positive

mode

(97%, Find by Formula) 2.1 x 50 mm x 1.8 micron

Eclipse Plus C-18

900 bar

1.5 mL/min

1290 Infinity

Very Narrow Peaks in MS Require More

Scans/Second –Optimize Scan Speed!

1290 Infinity Applications

Peak Width 0.7 sec

Very Fast Gradients Benefit From Very Fast Scan MS with 40 Cycle/s, 5-90%B Gradient in 0.65min

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55

0.60 Time, min

0.0

5.0e4

1.0e5

1.5e5

2.0e5

2.5e5

3.0e5

3.5e5

4.0e5

4.5e5

5.0e5

5.5e5

6.0e5

6.5e5

7.0e5

Inte

nsity

, cps

0.34s

0.36s

0.36s 0.42s

0.36s

Atenolol

Metoprolol

Primidone

Verapamil

Beclomethasone-

dipropionate

H2O/ACN

Flow =1.8ml/min

5-90%B in 0.5min

Stop time =0.65min

80°C, ACR

MS 40Hz

100-1000Da

Peak capacity of >40 in 39 sec in the MS chromatogram

What Does an Optimized System Allow You to Do?

After you have optimized

• Injection Parameters

• Flow Rate/ Linear Velocity

• Extra Column Volume

• Detector Parameters

What is Possible?

• Much Faster Separations

• Much more Resolution

• More Reliable Data

• More Accurate and Precise Answers

Chemical antidiabetic drug substances of the sulfonylurea

class

Glipizide Gliclazide

Glibenclamide Glimepiride

Gliquidone Repaglinide

July 31, 2013 39

Agilent Profile

Converting a Method from HPLC to UHPLC

Reduce analysis time

Maintain linearity

Improve LOD and LOQ with UHPLC

July 31, 2013 40

Agilent Profile

Six consecutive runs of Antidiabetic drug standards Agilent ZORBAX Eclipse Plus C18, 4.6 × 150 mm, 5 μm

min 2.5 5 7.5 10 12.5 15 17.5 20

mAU

0

100

200

300

400

500

1

2

3

4

5

6

July 31, 2013 41

Agilent Profile

Six consecutive runs of Antidiabetic drug standards Agilent ZORBAX RRHD Eclipse Plus C18, 3×50 mm, 1.8μm

min 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

mAU

0

50

100

150

200

250

1

2

3

4

5

6

July 31, 2013 42

Agilent Profile

Linearity and LOD/LOQ

Comparison between HPLC and UHPLC conditions

July 31, 2013 43

Agilent Profile

Comparison of Chromatographic Conditions

HPLC vs. UHPLC

Summary

• UHPLC columns are more efficient than HPLC columns

• Peaks generated are much narrower and more effected by non-

optimized separation parameters and LC instruments

• 2.1mm i.d. columns are effected more than 3.0 and 4.6 mm i.d.

• Extra Column Volume causes dispersion and needs to be

minimized

• Maximum performance will be realized on new design LCs

• Very complex samples will require UHPLC instruments and

1.8u or Poroshell columns

• Optimized Systems and Columns Allow Faster, Better Data

APPENDIX

Additional Slides

Instrument Impact on Column Performance

Gradient Delay or Dwell Volume .

Extracolumn Volume

Data Sampling ( or Acquistion) Rate

.

Number of

Scans or points

Typical Performance Comparisons of Various

Particles and Column Lengths

Length N

5um

N

3.5um

N

1.8um

Bar

1.8um

N

Poro

120

Bar

Poro

120

150mm 12,500 21,000 32,500 560 ~28,000 280

100mm 8,500 14,000 24,000 420 ~20,000 210

50mm 4,200 7,000 12,000 210 ~10,000 105

• pressure determined with 60:40 MeOH/water, 1ml/min, 4.6mm ID

ABSTRACT

The process of sample introduction in UHPLC seems simple enough-a dissolved sample is transferred into the LC, the sample separates in the ultra-high efficiency column and flows to the detector in record time. Simple. Not really-it is more like controlled chaos. Bang! The sample is met with a fast moving flow of mobile phase that may or may not be compatible with your sample solvent. The sample molecules are chaotically diluted by the mobile phase before it reaches the column. The sample slams into the column inlet and needs to be in a tight band as the separation starts on impact with the column bed. It leaves the column in discrete bands which again suffer dilution effects before it reaches the detector - all in less than 10 minutes. From the moment of injection to the final detection, the volume in the system, and how you control it, has serious effects on the quality of the chromatographic peak and data generated. This presentation deals with these injection and volume effects, what you need to know for better control and making the best use of your UHPLC instrument. You will learn how to properly scale UHPLC conditions, set efficient injection volume and concentration, optimize the injection process, minimize the extra-column volume in the system, and set optimum data collection parameters to produce the best chromatography using the new generation of UHPLC columns.

controlling the big bang in uhplc separations! · the big bang! dispersion begins at injection of...

Documents