Journal of Analytical Toxicology 2012;36:112 –115
doi:10.1093/jat/bks002
Article
A Gas Chromatography– Thermal Conductivity Detection Method for Helium Detection
in Postmortem Blood and Tissue Specimens†
Jason E. Schaff1*, Roman P. Karas1 and Laureen Marinetti2
1
Federal Bureau of Investigation Laboratory, Quantico, Virginia, and 2Montgomery County Coroner’s Office, Dayton, Ohio
* Author to whom correspondence should be addressed: Jason E. Schaff, FBI Laboratory Chemistry Unit, 2501 Investigation Parkway, Rm 4220,
Quantico, Virginia, 22135. Email: [email protected].
In cases of death by inert gas asphyxiation, it can be difficult to
obtain toxicological evidence supporting assignment of a cause of
death. Because of its low mass and high diffusivity, and its
common use as a carrier gas, helium presents a particular challenge in this respect. We describe a rapid and simple gas chromatography –thermal conductivity detection method to qualitatively
screen a variety of postmortem biological specimens for the presence of helium. Application of this method is demonstrated with
three case examples, encompassing an array of different biological
matrices.
Experimental
Introduction
Standards and controls
Standards of pure helium and pure air were prepared by filling
250-mL volumetric flasks with DI water, inverting in a DI water
bath, and then bubbling in the appropriate gas to completely
displace the water. The flasks were capped with rubber “turnover septum” stoppers, providing an airtight seal, before being
removed from the water bath. Standards of helium in air were
prepared by filling 100-mL volumetric flasks with air, as
described, just to the volume mark and then injecting appropriate amounts of the pure helium standard through the
stopper into each flask with a 3-cc plastic syringe. Standards
were prepared at concentrations of 0.5%, 1%, 2%, 3%, and 4%
(v/v), each in a separate flask. The 4% standard was prepared
with two consecutive 2-mL injections.
Negative control samples of DI water and of whole blood
were prepared by measuring 10 mL of the appropriate matrix
into 16- 100-mm culture tubes, capping with rubber septa,
and then venting the headspace in each tube to vacuum for 5 s
through a syringe needle. Positive control samples of DI water
and of whole blood were prepared from 10-mL aliquots of
matrix measured into 16- 100-mm culture tubes which were
then capped with rubber septa. Each sample was two-needle
sparged with a gentle flow of helium for 30 min to saturate the
matrix. The vent needle was fitted with a 0.2-mm PTFE syringe
filter to contain the resulting blood foam, with approximately
0.5 mL of the blood sample lost to foaming. After settling, the
headspace of each sample was vented to vacuum for 10 s
through a syringe needle to remove any undissolved helium.
Though asphyxiation has been a well-known means of suicide
for decades, deliberate use of inert gases as a means of asphyxia has become a known practice only fairly recently. The
first public discussion of inert gases as a means to commit
suicide seems to have been at a 1999 conference of the Self
Deliverance New Technology Group, with detailed instructions published three years later in the third edition of Final
Exit (1). The usual method recommended by various sources
is to feed a tube from the gas source into a bag secured over
the victim’s head, although some internet sources recommend use of a breathing mask. Helium is generally recommended as the “ideal” gas for suicide by asphyxia both
because of its easy availability in party balloon kits and its
low narcotic and hallucinogenic potential relative to other
inert gases (2). Despite this fact, there are only a handful of
cases of deliberate helium asphyxia reported in the medicolegal literature (3 –8). In all but one of these reports, the cause
of death was determined from physical evidence found at the
site of death and/or physical findings from autopsy without
supporting toxicological data. Auwaerter et al. (3) reported
analysis of helium by headspace gas chromatography–mass
spectrometry (GC–MS), but their method requires a complex
procedure for sampling gas from the lungs at the time of
autopsy. Yohitome et al. (9) reported GC– thermal conductivity detection (TCD) analysis for helium in a case of accidental
asphyxiation requiring a similar procedure for obtaining
gaseous samples at autopsy.
†
Materials
Carrier gas grade helium was obtained from ARCET
(Fredericksburg, VA), dry air was taken from the laboratory
building compressed air system, and high purity nitrogen was
produced in-house via a Domnick-Hunter (Gateshead, U.K.)
MaxiGas system. Drug-free human whole blood was obtained
from Clinical Controls International (Los Osos, CA), and deionized (DI) water was produced in-house with a Millipore
(Millford, MA) Synergy reverse osmosis system.
This is publication 11-17 of the Laboratory Division of the Federal Bureau of Investigation (FBI). Names of commercial manufacturers are provided
for identification purposes only, and inclusion does not imply endorsement of the manufacturer, or its products or services by the FBI. The views
expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the FBI or the U.S. Government. This
work was prepared as part of their official duties. Title 17 U.S.C. 105 provides that ‘Copyright protection under this title is not available for any
work of the United States Government’. Title 17 U.S.C. 101 defines a United States Government work as a work prepared by an employee of the
United States Government as part of that person’s official duties.
Published by Oxford University Press 2012.
through the electrical tape, and a second piece of electrical tape
was placed over the puncture after sampling. When not under
analysis, liquid samples were stored at approximately 48C, and
solid samples were stored at approximately –208C.
Results and Discussion
Figure 1. Experimental setups for producing gas standards (A) and matrix positive
controls (B).
Figure 1 illustrates the process for preparation of standards and
positive controls.
Analytical procedure
CG experiments were performed on an Agilent Technologies
(Wilmington, DE) 6890-N system equipped with a 30-m 0.32-mm 12-mm J&W (Wilmington, DE) HP-molesieve capillary
column. Carrier gas was high purity nitrogen at 1.0 mL/min. The
oven was maintained at a constant temperature of 358C with a
10 min run time. The thermal conductivity detector was maintained at a temperature of 2508C with makeup flow of 5 mL/min
and reference flow of 20 mL/min, both of high purity nitrogen.
The detector was set for negative polarity operation with a 5 Hz
data sampling rate. For the first reported case, a purged-packed
inlet was used with a temperature of 2008C. Prior to submission
of the second two cases, the instrument was converted to a
split/splitless injector, also operated at 2008C, with a 2:1 split
ratio. All injections were performed manually using a standard
Agilent 10-mL GC syringe with 10-mL injection volume.
Standards were directly sampled from the flasks in which they
were prepared, drawing gas aliquots through the rubber septa.
The controls and samples submitted in either blood collection
tubes or crimp-top vials were centrifuged 90 s at 1000 rpm to
remove any biological residue from the caps and then warmed
for 30 min at 35–378C. Gas aliquots for analysis were drawn
through the container septa or caps. Samples submitted in metal
specimen cans were allowed to stand for at least 1 h at room temperature. The lid on each can was then punctured with a
hammer and metal probe, and the resulting hole was immediately
covered with a piece of electrical tape. Gas aliquots were taken
Method validation and performance
Nitrogen was chosen as the carrier gas for this method
because its thermal conductivity is much lower than that of
helium while still being greater than that of most common
gases. Because the thermal conductivity of the analyte of interest is higher than that of the carrier gas, it was necessary to
operate the detector in negative polarity mode. Given that the
response factor for a TCD is proportional to the difference
between the thermal conductivity of the carrier gas and that of
the eluting analyte (10), the only common gases expected to
produce a positive signal response, along with their approximate response factor relative to helium at 300 K, are hydrogen
(120%), neon (20%), methane (6%), and oxygen (, 1%). These
response factors are only approximate because detector conditions do have some effect upon analyte response.
A 2% standard of helium in air was used for optimization of
instrumental method parameters. Initial development and validation was carried out with the purged-packed inlet. Two peaks
were observed: one, based on the work of Yoshitome et al. (9),
for oxygen and one, seen in the helium-fortified samples but
not in any blank samples, for helium. Flow rates between 1.0
and 2.5 mL/min and isothermal column temperatures between
30 and 508C were investigated, with optimum peak separation
obtained at a flow of 1.0 mL/min and a temperature of 358C.
Upon conversion to a split-splitless inlet, these flow and temperature conditions were retained, and split ratios from 0:1
(splitless) to 10:1 were investigated. An optimal balance
between peak shape and analyte response was obtained at a
split ratio of 2:1. Inlet temperature and detector conditions,
except for polarity, were taken from default parameters used
with other methods on this instrument.
The set of five standards of helium in air and an air blank
were analyzed three times, twice with the purged-packed inlet
and once with the split-splitless inlet. We did not attempt to
quantitate any case samples because of an inability to procure
or produce adequate quantitative controls of helium in biological specimens. However, results from these curves strongly
suggest that accurate quantitation would be possible if this
problem could be overcome. For all three curves, R 2 values of
0.99 or greater were observed, and the standard error estimate
of the intercept was larger than the calculated value of the
intercept. All samples except for the 0.5% standard yielded calculated values within 10%, relative to the target values. For the
0.5% standard, calculated values were all within 20%, relative to
the target value. Figure 2 shows chromatograms for the series
of standards on both the purged-packed and split-splitless configurations. Retention times for helium and the presumptive
oxygen peak were, respectively, 3.15 + 0.03 min and 3.62 +
0.11 min for the purged-packed configuration and 2.79 + 0.01
and 3.34 + 0.01 min for the split-splitless configuration.
The negative control water and blood samples showed no
detectable signal at the retention time for helium, while the
A Gas Chromatography –Thermal Conductivity Detection Method for Helium Detection in Postmortem Blood and Tissue Specimens 113
Figure 2. Chromatograms of standards of helium in air with a purge-packed injector
(A) and with a split-splitless injector (B).
Figure 3. Chromatograms of water and blood helium controls analyzed with
split-splitless injection. The 4% standard of helium in air is included for reference.
positive control samples showed strong helium response,
much greater than the 4% helium standard. Figure 3 shows
sample chromatograms for controls and the 4% helium standard. Note that there was no significant difference in helium
response between the water and blood positive control
samples, suggesting that there is little or no matrix effect on
analyte recovery. No significant change in helium signal, measured as chromatographic peak area, was observed in either
the positive controls or the 1% helium standard over the
course of four days, indicating that helium is reasonably
stable in samples stored under the conditions given in the
Experimental section.
Example cases
Case 1. The decedent was a white male, 39 years old, who was
estranged from his wife. His wife received a letter from him
two days after the last time he was known to be alive. The
letter stated he would be dead by the time she received it. The
wife called the local police and requested an officer go to her
husband’s residence to check on him.
114 Schaff et al.
Figure 4. Chromatograms of a 4% standard of helium in air, a negative control
blood sample, and the blood, brain, and lung specimens from case example #2.
Responding officers found that he was deceased, seated in a
chair. He had a plastic bag over his head, and the bag was
secured with a Velcro strap. The decedent had a flexible hose
in his mouth and the other end of the hose was attached to a
helium tank. The valve on the tank was open. Empty beer cans
and an empty Jack Daniels bottle were found near the body.
The box for the helium tank was found in a trash can. A purchase receipt was not found. The autopsy results were as
follows; generalized visceral congestion with severe congestion
and edema of lungs (right lung 725 g, left lung 635 g) and
hypertensive and arteriosclerotic cardiovascular disease with
cardiomegaly (405 g).
The atherosclerosis disease involved the left anterior descending coronary artery, which was 20 –30% narrowed by
plaque in its mid section, the circumflex coronary artery,
which was 20– 30% narrowed by plaque in its mid section, and
the right coronary artery, which was 20 –30% narrowed by
plaque in its mid section. Blood in a red-top Vacutainer tube,
liver and lung in metal specimen cans, and a portion of brain in
a 20-mL crimp-top vial were submitted for helium analysis.
Toxicology results showed the presumptive presence of
helium in the lung, brain, blood, and liver. Toxicology results
were negative for volatiles and drugs of abuse. The cause of
death was determined to be asphyxia due to helium inhalation.
Case 2. The decedent was a 56-year-old male with a recent
job loss and in the process of forced eviction. He was found
with a bag over his head with a hose attached to a helium tank.
The decedent’s pet cat was found in a similar situation. A copy
of Final Exit opened to the proper page was also found at the
scene. The autopsy showed signs consistent with asphyxia. The
toxicology results were as follows: ethanol in the blood and
urine both with a concentration of 0.14 g%, doxylamine in the
blood at less than 0.05 mg/mL and positive in the urine, and
35 ng/mL carboxy THC in blood and positive in the urine with
no parent THC in blood. Blood in a grey-top Vacutainer tube
and brain and lung in metal specimen cans were submitted for
helium analysis. Toxicology results showed the presumptive
presence of helium in the lung but not in the blood or brain, as
shown in Figure 4.
Case 3. The decedent was a 20-year-old white male college
student. He had some psychiatric problems and had attempted
suicide by means of an overdose in the past. The decedent had
to drop his classes because of his mental status, and his roommate had also told him he did not want to room with him any
longer. He was taking lithium and clonazepam and had been
seeing a psychiatrist since his past overdose attempt.
He told another student he was going to use a helium tank
to take his life. That student called the decedent’s roommate,
who was off on spring break, and he in turn called campus
police. Campus police checked on the decedent in the
morning and spoke with him, and he denied making any
threats. Later in the afternoon, campus housing personnel
called the decedent’s mother and requested that she ask her
son to come home because of his threats and past conduct.
She asked if they could explain the situation to him. They went
to his apartment on campus, and when they received no response, they used a key to gain entry, found him lying unresponsive in bed, and called 911. Emergency medical services
and campus police responded and pronounced him dead.
The decedent had a helium tank between his legs with an
attached hose that extended to his face. His head was wrapped
in a plastic bag with a piece of clothing wrapped around his
neck to effect some type of seal. There were several notes
found at the scene, one addressed to his parents and one to his
roommate. The autopsy results were as follows: an abrasion to
the frontal scalp, atherosclerotic cardiovascular disease, and
cardiomegaly (460 g) with left ventricular hypertrophy. The
atherosclerosis consisted of left anterior descending and right
main coronary arteries with 10– 70% narrowing. There was also
papillary thyroid carcinoma. Toxicology results included
lithium in blood of 0.29 mEQ/L. Blood in a grey-top
Vaccutainer tube and portions of liver and brain in 20-mL
crimp-top vials were submitted for helium analysis. Helium was
not detected in the blood, brain, or liver. The cause of death
was determined to be asphyxia.
Conclusions
A simple and rapid GC –TCD method to screen for the presence of helium in biological samples was developed and
applied to a variety of matrices from several postmortem cases.
This method does not require that any specialized sample
collection techniques be used at autopsy, only that samples be
promptly and completely sealed and that they be stored refrigerated or frozen, as appropriate. Not surprisingly, based upon
the cases examined, lung tissue appears to be the best matrix
for analysis, although detection is clearly possible in other
samples. This method is purely for screening purposes, to
provide analytical evidence that may or may not support physical evidence found at a death scene. With only a chromatographic retention time, it is not possible to establish an
incontrovertible identification for helium. The fact that only a
very small number of gases are physically capable of producing
a chromatographic peak in this analysis does, however, mean
that false-positive results for unputrefied specimens are extremely unlikely.
References
1. Humphry, D. Final Exit, 3rd ed. Random House: New York, NY,
2002; pp 130–140.
2. Bennett, P.B.; Roxtain, J.C. Inert gas narcosis. In Physiology and
Medicine of Diving, Brubank, A.O.; Neuman, T.S., Eds.; Elsevier
Science Ltd., Edinburgh U.K., 2003; pp 300–322.
3. Auwaerter, V.; Perdekamp, M.G.; Kempf, J.; Schmidt, U.;
Weinmann, W.; Pollak, S. Toxicological Analysis after Asphyxial
Suicide with Helium in a Plastic Bag. Forensic Sci. Int. 2007, 170,
139–141.
4. Schon, C.A.; Ketterer, T. Asphyxial Suicide by Inhalation of Helium
Inside a Plastic Bag. Am. J. Forensic Med. Pathol. 2007, 28 (4),
364–367.
5. Gallagher, K.E.; Smith, D.M.; Mellen, P.F. Suicidal Asphyxiation by
Using Pure Helium Gas: Case Report, Review, and Discussion of the
Influence of the Internet. Am. J. Forensic Med. Pathol. 2003, 24
(4), 361–363.
6. Ogden, R.D.; Wooten, R.H. Asphyxial Suicide with Helium and a
Plastic Bag. Am. J. Forensic Med. Pathol. 2002, 23 (3), 234–237.
7. Grassberger, M.; Krauskopf, A. Suicidal Asphyxiation with Helium:
Report of Three Cases. Wien Klin. Wochenschr. 2007, 119 (9),
323–325.
8. Ogden, R.D.; Hamilton, W.K.; Whitcher, C. Assisted Suicide by
Oxygen Deprivation with Helium at a Swiss Right-to-Die
Organization. J. Med. Ethics 2010, 36, 174– 179.
9. Yoshitome, K.; Ishikawa, T.; Inagaki, S.; Yamamoto, Y.; Miyaishi, S;
Ishizu, H. A Case of Suffocation by an Advertising Balloon Filled
with Pure Helium Gas. Acta Med. Okayama 2002, 56 (1), 53 –55.
10. Colon, L.A.; Baird, L.J. Detectors in Modern Gas Chromatography. In
Modern Practices of Gas Chromatography, Grob, R.L.; Barry, E.F.,
Eds.; John Wiley & Sons: Hoboken, NJ, 2004; pp 289– 298.
A Gas Chromatography –Thermal Conductivity Detection Method for Helium Detection in Postmortem Blood and Tissue Specimens 115