GC Method Development

A Decision Tree for

Gas Chromatography Method Development

BUILD BETTER GC METHODS WITH CRAWFORD SCIENTIFIC

A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 1 – Is analyte volatile enough for GC analysis?

Not sure N

Analyte Below 1000Da? Consider Derivatisation to increase volatility if

analytes have polar functional groups
Check for derivatisation reaction schemes – i.e.
use of BSTFA / MSTFA for acidic species etc.
Y N

Analyte Boling Point Consider liquid phase

below 350oC?* separation techniques

Y N
*Remember that analytes often
elute below their boiling point as
the high linear gas flow reduces the
vapour pressure within the system.
Probably suitable for GC Vapour Pressure >0 mmHg at 25oC? Even if the inlet is at 250oC this may
Check vapour pressure or Possibly volatile enough for GC analysis be enough to vaporise analytes
move to next stage – move to next stage with boiling points much higher
than this.

GC Method Development Page 2

A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 2 – Nature of the sample and the requirement for sample preparation
Simple Sample Complex Sample
(few components) (many components)

Solid Liquid Gas Solid Liquid Gas

Choose appropriate Choose appropriate Choose appropriate Choose appropriate Consider The need for Choose appropriate

sample solvent based sample solvent based sampling and sample solvent based sample preparation sampling and

on analyte (and on analyte (and sample introduction on analyte (and in the following order sample introduction

matrix) polarity matrix) polarity technique matrix) polarity of selectivity technique

Consider filtering the • Tedlar (gas • Solid Phase • Tedlar (gas

dissolved sample sampling) bag Extraction (SPE) sampling) bag

• Gas sampling • Solid Phase • Gas sampling

valve Microextraction valve

• Thermal (SPME) / Stir Bar • Thermal

desorption tube Sorbtive Extraction desorption tube

(SBSE)

• Dynamic Liquid /

Liquid Extraction
If the final sample is in liquid form, consider boiling point
(DLLME)
and expansion co-efficient when optimising splitless
• QuEChERS
injection conditions (where appropriate).
• Liquid / Liquid

Extraction (LLE)
The final analyte concentration should be considered if
• Filtration*
analyte dilution is employed.

If the final sample is in liquid form, consider boiling point

and expansion co-efficient when optimising splitless

Move to next stage injection conditions (where appropriate).

The final analyte concentration should be considered if

analyte dilution is employed.

*Check the chemical compatibility of filter materials

and housings when filtering organic solvents. Some
assessment should also be made of the absorptivity
Move to next stage
of the filter material towards target analytes to
ensure good accuracy and reproducibility.

GC Method Development Page 3

A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 3 (I) – Select appropriate detection technique: Quantitative Analysis

FID* ECD TCD NPD FPD CLD

Default choice Analyte contains Permanent Gas Analyte contains Analyte contains Analyte contains
Oxygen / Halogens Analysis N/P S/P S/N

• Response Type: • Response Type: • Response Type: • Response Type: • Response Type: • Response Type:
Universal (C) Selective Universal (C) Universal (C) Universal (C) Universal (C)

• Responds To: • Responds To: • Responds To: • Responds To: • Responds To: • Responds To:
Carbon Halogen or Oxygen Thermal Nitrogen or Phosphorous or Nitrogen or Sulphur
containing analytes Conductivity Phosphorous Sulphur
• Mass or • Mass or
Concentration • Mass or • Mass or • Mass or • Mass or Concentration
Dependant: Concentration Concentration Concentration Concentration Dependant:
Mass Dependant: Dependant: Dependant: Dependant: Mass
Concentration Concentration Mass Mass
• Destructive: • Destructive:
Y • Destructive: • Destructive: • Destructive: • Destructive: Y
N N Y Y
• LOD: • LOD:
10-12 gC/sec • LOD: • LOD: • LOD: • LOD: 10-12 gS/sec
10-14 g/mL 10-9 g/mL 1012 gN/sec 10-13 gP/sec 10-12 gN/sec
• Linear Range: 10-12 gS/sec
107 • Linear Range: • Linear Range: • Linear Range: • Linear Range:
104 up to 105 105 • Linear Range: >104 (S & N)
• Relative 104 (P)
Selectivity: • Relative • Relative • Relative Non-Linear (S) • Relative
none Selectivity: Selectivity: Selectivity: Selectivity:
Up to 106 versus none 2 x 104 v C (N) • Relative 107 v C (S)
• Uses: carbon 7 x 104 v C (P) Selectivity: 107 v C (N)
Default detector • Uses: 106 v C (P)
choice for organic • Uses: Permanent gas • Uses: 106 v C (S) • Uses:
compounds Environmental analysis or where Environmental Petroleum
analysis for non-destructive analysis / • Uses: characterisation,
• Optimise: pesticides & general detector is Petrochemical Any analysis flavour analysis,
Temperature herbicides required analysis requiring trace environmental
(typically 300 – determination of monitoring
350oC) • Optimise: • Optimise: • Optimise: P or S containing
Air to hydrogen Make up Gas Type Temperature (250 - Temperature compounds • Optimise:
ratio (typically & Flow (Nitrogen, 350oC typical) (lowest works best) Temperature
100:1) Methane or Methane Bead Power / Reference flow • Optimise: Furnace
Make-up gas flow with 5% Argon) Voltage (follow (typically 3x column Temperature (up to Temperature
(match air flow and Temperature (250 - manufacturers + makeup flow) 250oC typical) (800oC is typical)
optimise in +/-20% 350oC typical) guidelines) Filament Filter Ozone Flow
steps) Current & Mode Hydorgen Flow resistance (select correct filter- Hydrogen Flow
Magnitude of (low enough not (follow either P or S) Air Flow
applied current / avoid a flame manufacturer As per FID (follow
pulsed or constant formation) guidelines) manufacturer
current mode guidelines)

Move to next stage

*FID – Flame Ionisation Detector / ECD – Electron Capture Detector / TCD – Thermal Conductivity Detector / NPD – Nitrogen Phosphorous Detector / FPD – Flame Photometric Detector /
CLD – Chemiluminesence Detector. Also available – Photoionisation detector (PID) / Atomic Emission Detector (AED) / Electrolytic Conductivity Detector (ELCD) / Infrared Detector (IRD)

GC Method Development Page 4

A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 3 (II) – Select appropriate detection technique:

Qualitative and / or Quantitative Analysis

Mass Spectrometric Detection

Electron Ionisation (EI) Mode

• Response Type:
Selective & Universal

• Responds To:
Radical Cations created in ion source

• Mass or Concentration Dependant:

Mass

• Destructive:
Y Poor Molecular Ion Formation or

Sensitivity in EI mode
• LOD:
106

• Linear Range:
107
Try EI at 25eV electron energy to reduce degree of
• Relative Selectivity: fragmentation and promote intensity of molecular ion
0.1Da in selected ion or higher in single or
multiple reaction monitoring (SRM / MRM) modes

• Uses: Mass Spectrometric Detection

Universal for qualitative analysis Chemical Ionisation (CI) Mode

• Optimise: +ve -ve

Temperature of Transfer Line, Ion Source and
Mass Analyser
The MSD is a complex detector which requires
both calibration and optimisation using onboard
tuning algorithms • Chose Reagent Gas: • Chose Reagent Gas:
Isobutane Isobutane
Methane Methane
Ammonia Ammonia
(in order of decreasing (in order of decreasing
molecular ion intensity) molecular ion intensity)

• LOD: • LOD:
10-6 g 10-15 g

Move to next stage

GC Method Development Page 5

A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 4 – Select appropriate sample introduction technique1,2

Split / Splitless Programmed Thermal Static Headspace Purge & Trap Thermal Desorption
Injection Vaporising (Large Volume) Sampling Sampling Sampling
Injection

Default choice for Thermally labile Volatile analytes in For increased For increased
volatile / non-labile analytes or ultra-trace involatile liquids or selectivity and analyte selectivity and analyte
samples analysis solids preconcentration (liquid preconcentration
or solid samples) (gaseous samples)

Split Injection Optimise: Optimise: Optimise: Optimise:

Optimise: Mode Sample to headspace Purge Temperature Trap Sorbent Material
Liner Type (Solvent Vent / Cold Split / ratio Chemistry
(Straight through / Cold Splitless) (Determined by the sample Purge Flow
deactivated / no packing / volume to headspace vial Desorb Temperature
no restriction is typical) Initial Inlet Temperature size) Purge Volume
Inlet Temperature Equilibration Temp. Desorb Flow Rate
(250oC is typical – optimise Initial Inlet Time Equilibration Time Trap Sorbent Material
in steps of +/-50oC checking Shaking / Agitation On/ Chemistry Inlet Split Ratio
%RSD of repeat injections Vent flow pressure Off & Speed
and speak shape) Sample loop size Desorb Pre-heat Dry Purge Temperature /
Injection Volume Vent Flow Time (1mL or 3mL loops are Time / Flow
(1μL is typical – calculate typical) Desorb Time
sample expansion volume Purge Flow
doesn’t exceed liner Vial Equilibration Dry Purge Temperature /
volume)3 Purge Time Pressure and Time Time / Flow
Split Ratio (Sometimes required to
(100:1 is typical – optimise Injection Volume Loop Filling Temperature reduce carry-over between
in +/- 20:1 steps and Time samples)
Injection Speed
Spitless Injection Cryogenic Cooling Time Relative temperature of Sample size and purge
Optimise: Oven Temperature sample transfer line and solvent volume for solid
Liner Type Tracking GC inlet samples
(Packed / deactivated /
bottom restriction is typical) Split flow to promote
Injection Volume efficient gas plug transfer
(1μL is typical – calculate
sample expansion volume Initial GC oven
doesn’t exceed liner volume) temperature or cryo-
Splitless Time focussing time
(based on sample transfer
time from inlet to column)
Initial Oven Isothermal 1. In each case we have highlighted the broad principles behind optimisation of these techniques.
Temperature For more in-depth information visit the relevant topic on www.CHROMacademy.com
(should be 20oC lower than
2. Whilst we have proposed the parameters to be optimised in what we believe to be priority order, some
sample diluent boiling
iteration may be required to find the optimum parameter set
point)
Initial Oven Isothermal 3. For more information on calculators for vapour volume and pressure flow relationships – please visit;
Time https://www.agilent.com/en/support/gas-chromatography/gccalculators
(at least as long as the https://www.restek.com/Technical-Resources/Chromatography-Calculators
Splitless time) or several of the other freely available calculators from GC column manufacturers

Move to next stage

GC Method Development Page 6
A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 5 – Select appropriate GC Column

Non-polar Medium Polar Polar Permanent Gas or

Highly Volatile

100% methyl 5% phenyl Polyethylene Glycol Porous Layer Open

polysiloxane 95% methyl polysiloxane (PEG) or WAX Tubular (PLOT)

Low Retention Low Retention

Low Resolution

14% cyanopropyl 35% phenyl

phenylmethyl 65% methyl

polysiloxane polysiloxane

Low Retention Low Retention

Low Resolution

Volatiles phase 50% phenyl

(DB624 or similar) or 50% methyl

PLOT / WAX polysiloxane

Select Column Length

<10 components 10-50 components >50 components

10 - 15 m 30 m 60 - 120 m

Select Column i.d.

Fast GC / Highly Default Less Complex

Complex Samples Choice Samples

0.15 – 0.18 mm 0.25 – 0.32 mm 0.53 mm

Select Film Thickness

Select Carrier Gas & Linear Velocity
Low Volatility Default High Volatility
Analytes Choice Analytes
Helium Hydrogen Nitrogen

0.1 μm 0.25 – 0.5 μm 1-2 μm

30 – 35 cm/sec 40 – 45 cm/sec 10 - 15 cm/sec

Move to next stage

GC Method Development Page 7
A DECISION TREE FOR
GAS CHROMATOGRAPHY METHOD DEVELOPMENT

Stage 6 – Develop & Optimise Oven Temperature Program 1,2

Splitless Injection Split Injection

Set Initial Gradient Temperature & Hold Time

Follow screen method for Split injection but insert initial Screen sample using the following conditions
hold time of 1 min.
Initial Temp.: 40oC
Maintain Initial Temperature at 20oC below boiling point of Initial Time: 0 mins.
sample diluent Programming Rate: 10oC/min.
Final Temp.: 330oC
Optimise hold time empirically in steps of +/- 10 seconds (or column gradient max. temp.)
until peak area of early eluting peaks is <1% RSD Final Time: 10 mins

Peaks elute in a window of 5 mins or less?

Y N
Follow Split Temperature
Program Optimisation
Set Initial Gradient Temperature

Isothermal Analysis Possible Calculate T(i) (retention

Calculate T(f) (retention temperature of temperature of first analyte peak
final peak in the screening chromatogram) in the screening chromatogram)

Tiso = T(f) – 45oC Tinitial = T(i) – 45oC

Optimise Temperature Ramp Rate

Calculate or measure t0 (hold-up time)

Optimum Ramp Rate = 10oC per t0
Further optimise in steps of +/- 10oC/min.

Optimise Final Temperature & Time

Final temperature is
T(final) = T(f) + 20oC
Final Hold Time is t0 x 5 mins (optimise empirically)

Optimise resolution for critical pairs

1. The various relationships shown here are ‘rules’ of thumb’ and need to be
using mid-ramp holds
used as such to guide next steps. There are more finessed calculations
which can be found at www.CHROMacademy.com
Determine elution temperature of critical pair T(crit)
2. It is likely that some empirical optimisation will be required after each rule and insert mid ramp hold at
of thumb is employed and the results of the calculations will not deliver T(iso-hold) = T(crit) - 45oC
the optimum possible value for that variable.
Optimise the mid-ramp hold time empirically

GC Method Development Page 8

Crawford Scientific Ltd.
Holm Street, Strathaven
ML10 6NB
Scotland

Contact Us
phone: 01357 522 961
email: [email protected]
web: www.crawfordscientific.com

ABOUT THE AUTHOR

Tony Taylor has over 30 years of experience in developing chromatographic methods.
As the Technical Director of three varied contract and application development laboratories, he
understands what frustrates analytical chemists and how to help them overcome problems.
He has helped thousands of budding chromatography method developers using his own
experiences and insights, working with students to improve knowledge and understanding of
chromatographic processes and their application.