Converting Helium Jim McCurry Carrier Gas GC Agilent Technologies Methods Nitrogen and Wilmington, DE 18951 USA Hydrogen

1 October 12, 2012 Market Situation

. The world of He supply is not reliable, prices are increasing and customers are seeking alternative carrier gases

Agilent Restricted

Page2 2 October 12, 2012 The Carrier Gas Decision

No Is the chemist willing to convert Yes to alternative gasses?

GC Is the Application based on GC/MS GC or GC/MS?

Does the current GC method No have too much resolution?

Yes

He Conservation Consider migration to N2 Consider migration to H2

GC/MS specific H2 considerations

Agilent Restricted

Page3 3 October 12, 2012 Migration to H2: Considerations for GC & GC/MS

No Is the chemist willing to convert Yes to alternative gasses?

GC Is the Application based on GC/MS GC or GC/MS?

Does the current GC method No have too much resolution?

Yes

He Conservation Consider migration to N2 Consider migration to H2

GC/MS specific H2 considerations

Agilent Restricted

Page4 4 October 12, 2012 H2: Considerations for GC & GC/MS Application

H2 safety Sources of hydrogen Plumbing modifications Chromatographic method migration Method revalidation

Agilent Restricted

Page5 5 October 12, 2012 H2 Safety

GC, GC/MS: Both offer H2 enabled features

• Agilent H2 safety letter and safety manuals available • GC four levels of safety design • Safety Shutdown • Flow Limiting Frit • Oven ON/OFF Sequence • Explosion Test • Newer version 6890, 7890 GC and 5773, 5975 MSD offer greater safety than older versions of GC and GC/MSD.

Agilent Restricted

Page6 6 October 12, 2012 Source of Hydrogen

H2 Generator – preferred

• Very clean H2, >99.9999% available • More consistent purity • Built-in safety considerations • Make sure to buy a good one with a low spec for water and oxygen • Parker’s H-MD are used in LFS and SCS

H2 Cylinder • Consider gas clean filter • Possible to add safety device

Agilent Restricted

Page7 7 October 12, 2012 H2: Plumbing Modification Considerations

Tubing: • Use chromatographic quality stainless steel tubing

• Do not use old tuning (H2 is known as scrubbing agent) • Especially don’t use old copper tubing (brittleness is a safety concern)

Venting: • Connect split vent and septum purge vent to exhaust

Leak checking: • Recommend G3388B leak detector

Agilent Restricted

Page8 8 October 12, 2012 Migration to N2 for GC Application

No Is the chemist willing to convert Yes to alternative gasses?

GC Is the Application based on GC/MS GC or GC/MS?

Does the current GC method No have too much resolution?

Yes

He Conservation Consider migration to N2 Consider migration to H2

GC/MS specific H2 considerations

Agilent Restricted

Page9 9 October 12, 2012 Use N2 as carrier gas

Many HPI methods suited to nitrogen • Readily available and less expensive gas • No safety concern • Suitable for simple routine analysis (w. enough resolution) • 2-D GC ideally suited to nitrogen – Resolution issues solved by using 2 different columns

Potential issues • Reduced separation resolution • Not suitable for GC/MSD and certain GC detector applications

Agilent Restricted

Page10 10 October 12, 2012 Helium Carrier Gas Alternatives • Hydrogen, nitrogen, argon, argon/methane • Detector considerations can influence choice • Hydrogen and nitrogen are the most common alternatives

11 October 12, 2012 Helium Carrier Gas Alternatives Important Theoretical Considerations Relating To Peak Efficiency

Calculating Efficiency Efficiency & Carrier Gas Linear Velocity

We would like to know the actual time the Efficiency is a function of the component spends in the stationary carrier gas linear velocity or flow B Solute phase. + Cµ rate. HETP = A µ

t ' 2 + Start t ' = t - R R R t m n = The minimum of the curve 5.545 W h represents the smallest HETP (or Inert Gas largest plates per meter) and thus tR = Corrected Retention Time. t m t ' ' the best efficiency. "A" term is not R Time HETP t MOLECULAR present for capillary columns. R DIFFUSION B{ } C RESISTANCE TO MASS TRANSFER Let's relate “n” to the length of the column. n = effective theoretical n plates. Plates per meter (N) = or } A EDDY DIFFUSION L L Height to a theoretical plate (HETP) = µ ( opt equivalent n µ) Thus, the more efficient the column, the bigger the "N" the smaller the • Plot of HETP versus linear velocity is know as the Van Deemter "HETP“. plot. • The linear velocity value at the minimum of the curve is the optimum value for achieving the best efficiency.

12 Helium Carrier Gas Alternatives Let’s Make This Easy

Calculating Efficiency Efficiency & Carrier Gas Linear Velocity

We would like to know the actual time the Efficiency is a function of the component spends in the stationary carrier gas linear velocity or flow B Solute phase. + Cµ rate. HETP = A µ

t ' 2 + Start t ' = t - R R R t m n = The minimum of the curve 5.545 W h represents the smallest HETP (or Inert Gas largest plates per meter) and thus tR = Corrected Retention Time. t m t ' ' the best efficiency. "A" term is not R Time HETP t MOLECULAR present for capillary columns. R DIFFUSION B{ } C RESISTANCE TO MASS TRANSFER Let's relate “n” to the length of the column. n = effective theoretical n plates. Plates per meter (N) = or } A EDDY DIFFUSION L L Height to a theoretical plate (HETP) = µ ( opt equivalent n µ) Thus, the more efficient the column, the bigger the "N" the smaller the • Plot of HETP versus linear velocity is know as the Van Deemter "HETP“. plot. • The linear velocity value at the minimum of the curve is the optimum value for achieving the best efficiency.

13 Helium Carrier Gas Alternatives Let’s Make This Easy • Goal: change carrier gas while keeping other method conditions the same – use the same column – use the same oven program – adjust column flow or holdup time velocity to: • same peak elution order • same peak retention times

• For N2, test resolution of key components • adjust GC conditions (temp, flow) if needed

• Easier method revalidation – minimal changes to timed integration events – minimal changes to peak identification table

• Use tools built into the Agilent Chemstation to guide us through the process

14 October 12, 2012 The Tyranny of Van Deemter Why Nitrogen Gets a Bad Rap for Capillary GC

N HETP (mm) 2 C17 at 175º C k' = 4.95 Nitrogen Helium Hydrogen 1.2 OV-101 58 cm/s 58 cm/s 58 cm/s 2.4 mL/min 2.5 mL/min 1.7 mL/min 25m x 0.25 mm x 0.4µ 1.0 .8 He

.6 H .4 2

.2

10 20 30 40 50 60 70 80 90 Average Linear Velocity (cm/sec) R = 1.17 R = 1.37

• N2 actually provides the best efficiency, but at a slower speed • Most helium methods have too much resolution

– Lower N2 efficiency at higher flows can still provide “good enough” resolution • Most GC methods now use constant flow

– N2 efficiency losses with temp programming are not as severe

15 October 12, 2012 Helium Carrier Gas Alternative Test Case: ASTM D6584 for Free and Total Glycerin in Biodiesel Istd 2

1. Glycerol Istd 1 2. Monoglycerides 3. Diglycerides 2 4. triglycerides 3 1 4

5 10 15 20 25 30 35

COC Inlet: Oven Track Mode Pre-column: Ultimetal 2m x 0.53mm ID Column: Ultimetal DB5HT, 15m x 0.32mm ID x 0.1 df Column Flow: Helium at 3.0 mL/min (50 deg C) Column Pressure: 7.63 psi constant pressure mode Initial Column Temp: 50 oC for 1 min. Oven Ramp 1: 15 oC/min to 180 oC Oven Ramp 2: 7 oC/min to 230 oC Oven Ramp 3: 30 oC/min to 380 oC, hold 10 min.

Detector: FID with 25 mL/min N2 makeup

16 Configure Inlet for Carrier Gas in Chemstation

17 Set the Control Mode: Flow or Holdup Time

18 October 12, 2012 Wider Retention Time Variation Using the Same Flow 23.818 min

Helium Flow: 3.00 mL/min P: 7.63 psi Tr: 0.472 min. µ: 52.97 cm/s

23.469 min

Hydrogen Flow: 3.00 mL/min P: 3.85 psi Tr: 0.420 min. µ: 59.50 cm/s

23.705 min

Nitrogen Flow: 3.00 mL/min P: 7.09 psi Tr: 0.464 min. µ: 53.84 cm/s

18 19 20 21 22 23 24

19 October 12, 2012 Same Holdup Time (Tr) Gives Consistent Retention Times 23.818 min

Helium Flow: 3.00 mL/min P: 7.63 psi

Tr: 0.472 min. µ: 52.97 cm/s

23.862 min

Hydrogen Flow: 2.64 mL/min P: 3.43 psi

Tr: 0.472 min. µ: 52.97 cm/s

23.776 min

Nitrogen Flow: 2.94 mL/min P: 6.98 psi

Tr: 0.472 min. µ: 52.97 cm/s

18 19 20 21 22 23 24

20 October 12, 2012 Monoglyceride Resolution “Good Enough” Using Nitrogen Carrier

Mono2 area: 1049 Helium Flow: 3.00 mL/min Mono1 Mono3 Tr: 0.472 min.

Mono2 area: 1050 Hydrogen Flow: 2.64 mL/min Mono1 Mono3 Tr: 0.472 min.

Mono2 area: 1049 Nitrogen Flow: 2.94 mL/min Mono1 Mono3 Tr: 0.472 min.

16.5 17 17.5 18 18.5 19 19.5

21 October 12, 2012 ASTM D6584 - Quantitative Results For Alternative Carrier Gas

Weight Percent Helium Hydrogen Nitrogen Glycerin 0.015 0.014 0.013 Monoglycerides 0.226 0.216 0.223 Diglycerides 0.114 0.115 0.110 Triglycerides 0.071 0.085 0.098 Total Glycerin 0.097 0.095 0.098

22 October 12, 2012 Analysis of Oxygenates and Aromatics in Gasoline Using 2-D Gas Chromatography ASTM Method D4815 – Oxygenated Additives – Ethers and alcohols from 0.1 wt% to 15 wt% – Usually only one or two additives in a sample ASTM Method D5580 – Aromatics in Gasoline

– Measures benzene, toluene, C8 aromatics and C9 plus aromatics – Two injections per sample for complete analysis Preliminary separation removes light hydrocarbons from sample – Polar TCEP micro-packed columns retains polars and aromatics – Back flush TCEP column to non-polar capillary column (HP-1) to complete analysis

23 October 12, 2012 GC System for D4815 and D5580

Similarities between each method – Same GC hardware requirements • Inlets, detectors, plumbing and valve configuration – Same separation scheme • 2-D separation using polar TCEP micro-packed primary column • 20% TCEP on 80/100 Chromosorb PAW, 22“ x 1/16“ stainless • Only one difference in instrument requirements – non-polar capillary column – D4815 – 2.65 µm methyl silicone, 30m x 0.53mm – D5580 – 5 µm methyl silicone, 30m x 0.53mm

24 October 12, 2012 Configuration and Operation for D4815 and D5580

Primary flow split/splitless inlet S/S TCEP column EPC HP-1 capillary column

Secondary flow 7 6 5 8 4 FID PCM 9 3 EPC 10 1 2 variable TCD restrictor

25 October 12, 2012 Instrument Conditions

D4815 Method D5580 Method carrier gas nitrogen nitrogen Inlet Split/Splitless Split/Splitless inlet temperature 200 Deg C 200 Deg C inlet presuure 9 psi (constant P) 25 psi (constant P) TCEP column flow 5 mL/min 10 mL/min split vent flow 70 mL/min 100 mL/min split ratio 15:1 100:1 PCM pressure program 13 psi for 14 min 23 psi for 12.1 min 99 psi/min to 40 psi 99 psi/min to 40 psi HP-1 column flow 3 mL/min 10 mL/min FID temperature 250 deg C 250 deg C oven temperature 80 deg C isothermal 60 C for 6 min 2 C/min to 115 C 115 C for 1.5 min Run time 16 min 35 min

26 October 12, 2012 Analysis of MtBE and Ethanol in Gasoline using N2 Carrier Gas

MtBE

valve reset

DME(IS) benzene

ethanol

aromatic and heavy non-aromatic DME(IS) benzene hydrocarbons

2 4 6 8 10 12 14 16

27 October 12, 2012 ASTM Precision Specifications

D4815 Precision Measures Repeatability Reproducibility Compound Mass % Spec Observed Spec Observed Ethanol 0.99 0.06 0.01 0.23 0.01 Ethanol 6.63 0.19 0.03 0.68 0.04 MtBE 2.10 0.08 0.01 0.20 0.01 MtBE 11.29 0.19 0.05 0.61 0.08

Accuracy Evaluation MtBE mass % Sample known found SRM2294 #1 10.97 10.61 SRM2294 #2 10.97 10.60 AccuStd Check 12.00 11.81

28 October 12, 2012 D5580 Requires Two Runs per Sample for Complete Aromatic Analysis

Good resolution with nitrogen carrier gas 3 4 Method D5580A 1. benzene 2 2. toluene 3. 2-hexanone (IS) T1 T3 4. C8 aromatics, C9 plus aromatics, 1 and non-aromatic hydrocarbons 1 Method D5580B 1. 2-hexanone (IS) 2. ethylbenzene 3 T4 5 3. m,p-xylene T2 4. o-xylene 4 2 5. C9 plus aromatics

5 10 15 20 25

29 October 12, 2012 Detector Response Precision for a Pump Gasoline Sample – 5 Runs Over 5 Days FID Response (area counts) Run # Method MtBE Benzene Toluene Ethylbenzene m,p-Xylene o-Xylene 1 4815 24077 5580A 3306 19427 5580B 5402 21825 8069 2 4815 23384 5580A 3244 19436 5580B 5402 21956 8105 3 4815 22807 5580A 3178 19580 5580B 5422 22342 8232 4 4815 22915 5580A 3048 19344 5580B 5432 22263 8201 5 4815 23053 5580A 3099 19529 5580B 5422 22253 8199 Avg 23247 3175 19463 5416 22128 8161 Std. Dev 512 105 92 14 224 70 %RSD 2.2 3.3 0.5 0.2 1.0 0.9

30 October 12, 2012 Correcting Occasional Resolution Problems

Initial oven temp = 60 oC m,p-xylene

T2 = 1.85 min. ethylbenzene

Initial oven temp = 50 oC m,p-xylene

T2 = 1.85 min. ethylbenzene

5 10 15 20 25

– Some combinations of TCEP and HP-1 columns show less resolution using N2 carrier – Fastest and easiest solution is to lower the initial oven temperature

31 October 12, 2012 Using Nitrogen With High Resolution Capillary Columns

Test Case: EN14103 for Total FAME and Methyl Linolenate Content in B100 Biodiesel • Expand method to include animal fat derived biodiesel – Replace C17 FAMEs ISTD with C19 FAME ISTD

• Expand method to include tropical oil (palm) derived FAMEs – Quantify C6 to C24 FAMEs and methyl linolenate (C18:3) – Requires column temperature program to resolve all FAMEs

• Method updated to provide better precision – Verify C19:0 ISTD purity

32 October 12, 2012 EN14103:2011 GC Configuration and Operating Conditions

Standard 7890A GC Hardware G3440A Agilent 7890A Series GC Option 112 100 psi split/splitless Inlet with EPC control Option 211 Capillary FID with EPC control G4513A Agilent 7693 Autoinjector 19091N-133 HP-INNOWax Column, 30 m x 0.25 mm ID x 0.25 mm

Split/Splitless Inlet Temperature 250 oC Split flow 100 mL/min.

Column Flow He or H2 at 1 mL/min. constant flow 60 oC for 2 min. 10 oC /min. to 200 oC Column temperature 5 oC/min. to 240 oC Hold 240 oC for 7 min. Flame ionization detector 250 oC

33 October 12, 2012 EN14103 – FAME Identification Standard Helium at 1 mL/min Constant Flow 1 4

7 15 19 2 12 5 6 8 16 14 20 3 17 18 9 10 11 13

0 5 10 15 20 25 30

Peak # Name RT (min.) Peak # Name RT (min.) 1 methyl hexanoate C6:0 6.031 11 methyl arachidate C20:0 22.857 2 methyl myristate C14:0 15.878 12 methyl eicosonate C20:1 23.166 3 methyl myristoleate C14:1 16.275 13 methyl eicosadienoate C20:2 23.808 4 methyl palmitate C16:0 17.996 14 methyl arachidonate C20:4 24.551 5 methyl palmitoleate C16:1 18.311 15 methyl eicosatrienoate C20:3 24.730 6 methyl stearate C18:0 20.332 16 methyl behenate & C22:0 25.582 7 methyl oleate (9) C18:1 20.617 methyl eicosapentaenoate C20:5 8 methyl oleate (11) C18:1 20.697 17 methyl erucate C22:1 26.031 9 methyl linoleate C18:2 21.205 18 methyl lignocerate C24:0 29.574 10 methyl linolenate C18:3 22.052 19 methyl nervonate C24:1 30.203 20 methyl docosahexaenoate C22:6 30.365

34 October 12, 2012 EN14103 – FAME Retention Time Standard Comparison of He and N2 Carrier Gas pA 160 Helium at 1 mL/min 140 Constant Flow 120 100 80 60 40 20

16 18 20 22 24 26 28 30 min

140 120 Nitrogen at 1 mL/min 100 Constant Flow 80 60 40 20

16 18 20 22 24 26 28 30 min

35 October 12, 2012 EN14103 – Soya B100 Biodiesel Comparison of Helium and N2 Carrier Gas

C19:0 (IS) Helium at 1 mL/min Constant Flow

16 18 20 22 24 26 28 30 min

Nitrogen at 1 mL/min Constant Flow

16 18 20 22 24 26 28 30 min

36 October 12, 2012 EN14103 – FAME Content in Biodiesel Comparison of He and N2 Carrier Gas

FAME Content (wt%) Helium Nitrogen Sample Run 1 Run 2 r Run 1 Run 2 r r (spec) SRM 2772 96.9 96.7 0.2 97.2 97.2 0.0 1.01 Rapeseed 96.7 96.0 0.7 95.5 95.7 0.2 1.01 Coconut 87.4 86.9 0.5 86.6 86.9 0.3 1.01 Rape/Coco (50/50) 91.2 91.2 0.0 91.4 91.0 0.4 1.01

C18:3 (wt%) Helium Nitrogen Sample Run 1 Run 2 r Run 1 Run 2 r r (spec) SRM 2772 7.3 7.3 0.0 7.3 7.3 0.0 0.2 Rapeseed 8.5 8.4 0.1 8.3 8.3 0.0 0.2 Coconut 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Rape/Coco (50/50) 4.2 4.2 0.0 4.2 4.1 0.1 0.1

37 October 12, 2012 If You MUST Use Helium – Consider Conservation

No Is the chemist willing to convert Yes to alternative gasses?

GC Is the Application based on GC/MS GC or GC/MS?

Does the current GC method No have too much resolution?

Yes

He Conservation Consider migration to N2 Consider migration to H2

GC/MS specific H2 considerations

Agilent Restricted

Page38 38 October 12, 2012 Reduce Helium Consumption

• 7890 GC Gas Saver – reduces split flow to conserve helium consumption • Leak checking • Turn off the system to avoid long term idle – Not always recommended for some GC columns or traps

Agilent Little Falls Site was able to reduce

consumption by 50%

Agilent Restricted

Page39 39 October 12, 2012 Summary Migrating your GC method from helium carrier gas can be easily done

• For resolution critical methods, H2 offer the best alternative – Agilent GC and GC/MS systems have many built-in safety features

• For many GC applications, N2 offers a cheap, easy alternative without any safety worries – Most existing helium methods have too much resolution – N2 can be used without changing any of the existing GC conditions • keep the holdup time the same as the original method

– 2-D methods have high resolution built-in, so N2 is ideally suited as a carrier gas • Valve-based or Deans switch, not GCxGC

• If helium alternative is not an option, consider conservation

40 October 12, 2012