Conversion of Agilent
EI GC/MSD Systems
From Helium To
Hydrogen Carrier Gas
Bruce D. Quimby, Ph.D.
GPD Solutions Group
and
Harry F. Prest, Ph.D.
GC/MS Research and Development
October 9, 2012
C&EN Webinar
Question: Have You Ever Tried Converting Your
GC/MS From Helium To Hydrogen Carrier Gas?
Answer 1: “No. Never thought of trying it. Too busy.”
Answer 2: “We did that several years ago. It took some effort,
but it has been working fine.”
Answer 3: “We tried it once. The results were horrible and
confusing! We had a very high background, high
noise, and very bad peak shapes. After 4 days, we
went back to helium.”
C&EN Webinar
2
October 9, 2012
Introduction: Converting from He to H2 Carrier Gas
Many GC/MS users are considering changing from helium to hydrogen
carrier gas due to price/availability problems with helium.
– Read Chemical and Engineering News - July 16, 2012 (Page 32-34)
This talk describes the steps recommended for converting EI GC/MS
methods.
It is important to recognize the differences with using hydrogen carrier. Time
should be allotted for adapting the method, optimization, and resolving
potential problems. Areas that will need attention include:
–
–
–
–
–
–
–
choice of supply of H2
remove/bypass carrier gas filter (may be re-installed later if needed)
GC/MSD hardware changes
choosing new chromatographic conditions
potential reduction in signal-to-noise ratio (2-5x)
changes in spectra and abundance ratios for some compounds
activity and reactivity with some analytes
C&EN Webinar
3
October 9, 2012
Introduction: Converting from He to H2 Carrier Gas
Methods that will generally require less optimization include analytes that
are:
– “durable” compounds
– at higher concentrations
– analyzed with split injections
– derivatized
Methods that will generally require more optimization include analytes that
are:
– “fragile” compounds
– at trace concentrations
Allow time for necessary updates to SOPs and validation
C&EN Webinar
4
October 9, 2012
H2 Safety
C&EN Webinar
5
October 9, 2012
Letter from Agilent Safety Engineer: Describes H2
Safety Features of Agilent GCs
C&EN Webinar
6
October 9, 2012
Designed for Reliability – H2 Safety
Safety Shutdown
When gas pressure set points are not met, the valve and heater are shut off
to prevent explosion
Flow Limiting Frit
If valve fails in open position, inlet frit limits the flow
Oven ON/OFF Sequence
Fan purges the oven before turning on heater to remove any collected H2
Explosion Test
GC and MS designed to contain parts in case of explosion
C&EN Webinar
7
October 9, 2012
Safety First: Read This Before You Start
G3170-90010
C&EN Webinar
8
October 9, 2012
Source of H2 Carrier
and
Plumbing
C&EN Webinar
9
October 9, 2012
Source of Hydrogen: Use a Hydrogen Generator
Higher initial expense than cylinders, much lower cost over time
Very clean H2, >99.9999% available
More consistent purity
Safety considerations
– H2 is only generated at needed pressure (like 40 psig)
– Flow is limited (like 250 mL/min)
– Auto-shutdown if set point pressure cannot be maintained
– Minimal stored gas (like 50 mL at 40 psig) of H2 at any one time
Make sure to buy a good one with a low spec for water and oxygen
C&EN Webinar
10
October 9, 2012
Hydrogen Generation
This is an example of the
type of generator to use
for H2 carrier gas.
Look for generators with a
>99.9999% specification
and low individual specs
on water and oxygen.
C&EN Webinar
11
October 9, 2012
Plumbing the Instrument
Chromatographic quality stainless steel tubing is often recommended for H2
plumbing and is probably the best choice if available. Users may have to
follow local codes or internal company guidelines.
We have also used new 1/8th copper that has been cleaned for GC use.
Dirty tubing will cause huge contamination problems, as H2 appears to carry
dirt out of metal more than He does.
Don’t use really old copper tubing, as it becomes brittle and can break.
Note that MSD leak checks will not always find big outgoing leaks. Leak
check when complete with electronic leak detector.
When plumbing a H2 generator, start out with no traps. Only add traps if
needed. Make sure the water and oxygen levels are low enough.
C&EN Webinar
12
October 9, 2012
Split Vent and Septum Purge Vent Should be
Connected to Exhaust
C&EN Webinar
13
October 9, 2012
MS Components Required for
Conversion:
Magnet and Draw
Out Lens
C&EN Webinar
14
October 9, 2012
Check the Magnet in the 5975
If the magnet in the
5975 does NOT have
a serial number
along the left edge
(as viewed through
the source window),
it is suitable for use
with Helium, but not
Hydrogen.
If not, please contact
your Agilent service
engineer to change it
to a Helium and
Hydrogen compatible
magnet before
converting the
system to H2 carrier.
Note: All 5973 magnets are compatible with both He and H2
C&EN Webinar
15
October 9, 2012
Replace the Standard 3 mm Draw Out Lens With
the Optional Hydrogen Draw Out Lens
G2589-20045
Hydrogen Draw
Out Plate, Inert
Standard 3 mm
Draw Out Plate,
Inert
C&EN Webinar
16
October 9, 2012
Choosing a Column and
Method Conditions
C&EN Webinar
17
October 9, 2012
MSD Pumping Capacity for H2 is Less than for He
This Limits Column Choice
These are approximate values for the 5975. Maximum flow of H2 to maintain
reasonable source pressure:
– Performance turbo: 2 mL/min
– Standard turbo: 1 mL/min
– Diffusion pump: 0.75 mL/min
– Pressure pulsing: turbo < 3 mL/min, diffusion < 2.5 mL/min
It is very helpful to have an ion gauge on the MS to monitor the vacuum vs.
column flow.
Try to avoid flows that produce pressures higher than 5 x 10-5 torr. You can
get useful data above this pressure, but performance may start to degrade
C&EN Webinar
18
October 9, 2012
Guidelines for Choosing Column Dimensions and
Flows for Method Converted from He to H2
Determine max flow of H2 into MS that will give source pressure of 5 x 10-5
torr or less source pressure. This is your max column flow.
Choose column dimensions at initial oven temp of method to give:
− a flow < max column flow for vacuum pump
− a flow > min column flow for efficiency
− an inlet pressure of at least 5 psig
Keeping a temperature ramp of the same number of °C/void time will give
similar elution order. Use the Agilent method translator for this.
These are only approximate guidelines. Sometimes you may have to
deviate from them.
C&EN Webinar
19
October 9, 2012
Van Deemter Curves
If the flow of He or H2 is too low, you loose efficiency MUCH FASTER
than if it is too high. This is why operating at or above minimum
linear velocity is important.
Use 35 cm/sec as minimum for H2
C&EN Webinar
20
October 9, 2012
Column Choice
When in doubt (or a hurry) use:
− Use 20m x 180 um column and set constant flow at 0.3-0.5 mL/min.
− Try to get a linear velocity of at least 35 cm/sec
Try to use the same phase and phase ratio to get similar elution order.
Smaller bore columns have lower capacity. You may have overloading on
high level calibration standards, or samples
You may need to use pulsed injection to load sample into small bore
columns when using a low flow rate.
The flow range for your H2 setup is limited on the low side by the flow to get
35 cm/sec and on the high side by the vacuum pump capacity
C&EN Webinar
21
October 9, 2012
Using the Agilent Flow Calculator: Set Linear
Velocity to 35 for H2 Carrier
If Inlet Pressure Too Low, Raise Flow
Inlet pressure is too low
Set inlet pressure to 5 psig
C&EN Webinar
22
October 9, 2012
Using the Agilent Method Translator: 30m x 0.25
mm id x 0.25 um Checkout Column with Perf Turbo
Meets flow requirements
for pump and column
efficiency with 2x speed
gain
C&EN Webinar
23
October 9, 2012
Using the Agilent Method Translator: With Perf
Turbo and Std Turbo
For 20m x 0.18 mm id x
0.18 um Column:
− Meets flow
requirements for
standard turbo pump.
− Flow here is chosen
to give a 2.5 x speed
gain.
C&EN Webinar
24
October 9, 2012
Initial Startup with
Hydrogen
C&EN Webinar
25
October 9, 2012
Initial Problems with Switching GC/MS Methods to
from He to H2 Carrier Gas
High background that looks like hydrocarbons 
Reduced signal to noise (worse MDL) 
Significant tailing for many compounds 
C&EN Webinar
26
October 9, 2012
Switch to H2 1st Problem: High Background
You will see this throughout work
with H2.
• It will get much smaller with time and
use.
• It used to takes days or weeks for
background to fall to acceptable
levels.
Agilent developed a new
conditioning protocol to clean-up
the background much faster.
(see later slide)
C&EN Webinar
27
October 9, 2012
GC/MS Toxicology Analyzer Checkout Sample
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Amphetamine
Phentermine
5 ng/uL
Methamphetamine
Nicotine
Methylenedioxyamphetamine(MDA)
Methylenedioxymethamphetamine(MDMA)
Methylenedioxyethylamphetamine
Meperidine
Phencyclidine
Methadone
Cocaine
SKF-525a (RTL Compound)
9
Oxazepam
Codeine
8
4
2
10
each compound
17
15 18 22
14
20
19
1
4
23
24
26
27
28
13
5
2
25
21
6
Lorazepam
Diazepam
Hydrocodone
Tetrahydrocannabinol
Oxycodone
Temazepam
Flunitrazepam
Diacetylmorphine
Nitrazepam
Clonazepam
Alprazolam
Verapamil
Strychnine
Trazodone
16
12
11
7
3
15
16
17
18
19
20
21
22
23
24
25
26
27
28
6
8
10
12
14
16
18
C&EN Webinar
28
October 9, 2012
Switch to H2 2nd Problem – Peak Tailing
Abundance
2000000
1800000
1600000
20m x 0.18mm id x 0.18 um DB-5MSUI
1400000
1200000
Hydrogen
1000000
800000
600000
400000
200000
Time-->
0
Abundance
3000000
2800000
2600000
2400000
2200000
2000000
1800000
1600000
1400000
1200000
1000000
800000
600000
400000
200000
0
Time-->
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
Helium
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
Immediately after conversion to H2, peak shapes on TIC
are poor and tailing badly.
C&EN Webinar
29
October 9, 2012
Switch to H2 3rd Problem: Spectrum Changes over
Peak Profile
There is More Happening Here
than Simple Peak Tailing
Cocaine
7.20
7.40
7.60
7.80
8.00
8.20
Peak Spectrum
Abundance
8.40
8.60
8.80
9.00
Tail Spectrum
Abundance
Average of 8.018 to 8.036 min.: ToxCKO_H2_4.D\data.ms
82
182
24000
22000
Average of 8.048 to 8.085 min.: ToxCKO_H2_4.D\data.ms
91
14000
20000
12000
18000
16000
10000
Cocaine
14000
12000
10000
105
8000
41
64
55
4000
68
2000
m/z-->
78
6000
4000
6000
0
Not Cocaine
8000
42
40
60
80
303
198
122
152 167
272
135
215 230 244 259
288
327 341
100 120 140 160 180 200 220 240 260 280 300 320 340
m/z-->
115128
167
141
154
2000
0
40
60
191 207221 242 265 281 300314327341
80 100 120 140 160 180 200 220 240 260 280 300 320 340
C&EN Webinar
30
October 9, 2012
Main Ions of Cocaine Do Not Tail. Tail Made Up of
Other Ions
Tail is different than peak. May be formed from catalytically shredded main
component. This problem is reduced with time
Cocaine
Tail is not cocaine ions
C&EN Webinar
31
October 9, 2012
H2 Conversion
Considerations for Success
C&EN Webinar
32
October 9, 2012
Agilent’s Recipe for H2 Success: Minimizing the Time for
Chromatography and Background to Stabilize
1. Have a good source of clean hydrogen like a >99.9999% purity hydrogen
generator.
2. Keep H2 flow limited to value suitable for your pumping system
3. Use optional Hydrogen draw out lens
4. After setup, purging and pump down:
– Set the source to max temp for your source (check to see what yours is)
– Reduce the EMV to 800V
– Leave the FILAMENT ON overnight. This cleans up background rapidly.
5. Peak shape will be much better and background will be much lower in
the morning.
6. Lower the source temp to method value, retune, and run some samples
7. It is a good idea to have an extra set of filaments on hand in case one
burns out. This hasn’t been a problem, but a good idea.
C&EN Webinar
33
October 9, 2012
Agilent’s Recipe for H2: Success
• Peak Shape Much Better
• Background is Lower
• Better Signal-to-Noise
Abundance
2600000
2400000
2200000
2000000
1800000
1600000
1400000
1200000
1000000
800000
600000
400000
200000
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
Time-->
C&EN Webinar
34
October 9, 2012
Agilent Toxicology Analyzer
Features:
• Configured, tuned and tested for the application
• Retention time locked (RTL) with DRS database
• Constant flow to MS
• Backflush to reduce cycle time and remove matrix artifact
Challenge:
• Convert from He to H2 carrier without altering performance
criteria
C&EN Webinar
35
October 9, 2012
He Constant Flow Toxicology Analyzer
Parameters:
• Constant Flow to MS
• RTL DRS Method
Liquid
Injector
AUX EPC
Vent
S/S Inlet
~12psi
1m x 0.15 mm id
1.7 mL/min
Column
PUU
7890A
GC
5975C
MSD
15 m X 0.25 mm id X 0.25 um DB-5MSUI
C&EN Webinar
36
October 9, 2012
H2 Constant Flow Toxicology Analyzer
Parameters:
 Constant Flow to MS
 RTL DRS Method
Liquid
Injector
No Change in Performance
AUX EPC
Vent
S/S Inlet
~8psi
0.81m x 0.12 mm id
1.7 mL/min
Column
PUU
7890A
GC
5975C
MSD
15 m X 0.25 mm id X 0.25 um DB-5MSUI
C&EN Webinar
37
October 9, 2012
Converting Toxicology Analyzer from He to H2
Abundance
Toxicology Check Out Mix
1200000
1000000
Helium
800000
600000
400000
200000
0
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
Time-->
Abundance
1200000
1000000
800000
Hydrogen
600000
400000
200000
0
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
Time-->
C&EN Webinar
38
October 9, 2012
Switching GC/MS Methods to from He to H2 Carrier
Gas
Reduced high background that looks like hydrocarbons 
Improved signal-to-noise relative to pre-conditioned state 
Reduced tailing for many compounds 
C&EN Webinar
39
October 9, 2012
Performance Expectations
C&EN Webinar
40
October 9, 2012
Performance Expectations
Signal-to-noise ratio is often worse by about 2-5 x. This obviously varies from
compound to compound
While most spectra remain same, there are always exceptions. Users should
check the reference spectra for important targets to make sure they have not
changed
Same comment for target/qualifier ratios
Some compounds may disappear at low levels. Examples include: some
nitrogen and oxygen containing compounds (alcohols, aldehydes, ketones)
like those found in flavor samples
C&EN Webinar
41
October 9, 2012
Library Match Factors – Helium and Hydrogen
Carrier Gas Comparison
Dichlorvos
Mevinphos
Ethalfluralin
Trifluralin
Atrazine
BHC-gamma (Lindane, gamma HCH)
Chlorpyrifos-methyl
Heptachlor
Malathion
Chlorpyrifos
DDE, p,p'Dieldrin
Hexazinone
Propargite
Leptophos
Mirex
Fenarimol
Coumaphos
Etofenprox (Ethofenprox)
NIST (Forward)
Helium
Hydrogen
924
867
925
852
927
897
897
856
914
901
925
858
922
897
926
851
916
836
901
890
934
903
926
903
901
832
851
821
884
847
903
881
905
872
864
804
906
842
NIST (Reverse)
Helium
Hydrogen
929
904
973
967
927
897
897
857
915
908
931
878
926
908
929
856
953
846
902
892
946
940
927
903
906
912
857
829
884
847
914
934
908
877
870
813
916
843
In general, these 20 spectra in hydrogen carrier show lower match factors than in
helium carrier gas. They can still be properly identified in library search.
C&EN Webinar
42
October 9, 2012
Lindane
Helium Carrier Gas
Observations
• Ion ratios changed
• Extra peaks in H2
NIST Forward Match = 925
Hydrogen Carrier Gas
NIST Forward Match = 858
C&EN Webinar
43
October 9, 2012
Heptachlor
Observations
• Ion ratios changed
Helium Carrier Gas
NIST Forward Match = 926
Hydrogen Carrier Gas
NIST Forward Match = 851
C&EN Webinar
44
October 9, 2012
Other Considerations:
H2 is not an inert gas, and therefore inertness problems will still exist
or be worse.
Use the lowest inlet temp that works (to reduce reactions with H2)
Use pulsed injection, especially with small bore columns
Consider using an MMI in cold splitless mode for fragile compounds
Using a deactivated S/SL weldments might help (but don’t ever scrub it)
Avoid using methylene chloride as a solvent (especially wet).
– At higher inlet temps (like >280C), HCl is formed and causes problems.
– If must be used, use lowest inlet temp and maybe a deactivated S/SL
weldment, or Multimode inlet
Avoid carbon disulfide as a solvent
Use liners with a taper at bottom to minimize sample contact with gold seal
– Use Agilent Ultra Inert Liners.
C&EN Webinar
45
October 9, 2012
Hydrogen Carrier Summary
1. Get a GOOD hydrogen generator and plumb it properly
2. Pick a column that meets the flow needs of method and
MSD pump
3. Do magnet update (if necessary) and install Hydrogen draw
out lens
4. After purging and pump down, set source to max temp and
EM to 800V. Leave filaments on all night.
5. Reset source temp, retune, and run samples. Run practice
samples for a few days to make sure system has stabilized
6. Source cleaning will be needed much less frequently
C&EN Webinar
46
October 9, 2012
Thank You for Your Attention
For additional information please visit:
http://www.agilent.com/chem/heliumupdate
C&EN Webinar
47
October 9, 2012