Author's personal copy
3He mass spectrometry for very low-level measurement of organic tritium in
P. Jean-Baptistea,*, E. Fourre ´a, A. Dapoignya, D. Baumiera, N. Baglanb, G. Alanicb
aLSCE, CEA-CNRS-UVSQ, CEA/Saclay, 91191, Gif-sur-Yvette, France
bCEA/DAM/DIF, 91297 Arpajon, France
a r t i c l e i n f o
Received 14 September 2009
Received in revised form
14 October 2009
Accepted 16 October 2009
Available online 10 November 2009
a b s t r a c t
The design, setup and performance of a mass spectrometric system for the analysis of low to very low-
level tritium in environmental samples are described. The tritium concentration is measured indirectly
by the3He ingrowth from radioactive decay after complete initial degassing of the sample. The analytical
system is fully computer-controlled and consists in a commercial helium isotope mass spectrometer
coupled with a high vacuum inlet system. A detection limit of 0.15 Bq/kg is routinely obtainable for
sample sizes of 20 g of water equivalent and an accumulation time of three months. Larger samples (and/
or longer accumulation time) can be used to obtain lower detection limits. In addition to the benefit of
a lower detection limit, another advantage of this non-destructive method lies in the simplicity of the
analytical procedure which strongly limits the risk of contamination. An inter-comparison was
successfully performed with the conventional beta counting technique on lyophilized grass samples, in
a range of tritium concentrations of environmental interest. It shows that the3He mass spectrometry
method yields results that are fully consistent with the conventional liquid scintillation technique over
a wide range of tritium concentrations.
? 2009 Elsevier Ltd. All rights reserved.
Tritium (3H) is present in the environment as a result of both
natural and anthropic sources. Owing to its low natural production
by cosmic radiations in the upper atmosphere (Table 1) and to its
relatively short half-life of 4500 days (Lucas and Unterweger,
2000), natural tritium levels in precipitation and surface waters, as
determined from early tritium (pre-nuclear) measurements (Libby,
1955) and from polar ice cores (Jouzel et al., 1982; Fourre ´ et al.,
2006 and references therein) were quite low, not exceeding a few
Bq/L. In the fifties and early sixties however, this low background
has been multiplied by a factor of one thousand due to the release
of tritium in considerable amounts by the atmospheric tests of
nuclear weapons. Tritium is also released in various amount by
industrial activities (Table 1), including the nuclear industry,
compounds industry (tritium being used as a radiation source in
luminous paints and GTLS – Gaseous Tritium Light Sources – for
clocks, watches and various devices including emergency signs and
military equipments). Although these tritium sources may be
significant locally, at the global scale tritium levels are steadily
declining (IAEA, 2009), so that present-day levels are usually at or
near minimum detectable concentrations by conventional liquid-
scintillation counting systems.
Over the last three decades, an alternative method based on the
detection of its radioactive daughter3He by mass spectrometry (the
so-called3He ingrowth method – Clarke et al., 1976) has been
extensively used for routine measurements of verylow to ultra-low
levels of tritium (in the range 0.1–0.01 Bq/L) in oceanography and
hydrology (see Jenkins, 2007; Phillips and Castro, 2007; Jenkins,
2009 and references therein).
Our group has been involved since the early 80’s in the
measurement of tritium in oceanic and continental waters by the
3He ingrowth method (Jean-Baptiste et al., 1992). The principle of
the method is to remove the3He initially dissolved in the water
sample by degassing under vacuum, then to store it in a closed
container (usually a glass bulb) to allow for the accumulation of
tritiogenic3He. The tritium content of the sample is subsequently
deduced from the mass spectrometric determination of the amount
of3He produced during the storage period. Since3He is sparinly
soluble in organic compounds and readily enters the gas phase
(Tremblay et al., 2006), the technique is adaptable to a wide variety
of sample matrices such as food items, vegetation, etc. Hence,3He
mass spectrometryappears asa promisingoption for
* Corresponding author. Tel.: þ33 169087714; fax: þ33 169087716.
E-mail address: firstname.lastname@example.org (P. Jean-Baptiste).
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity
journal homepage: www.elsevier.com/locate/jenvrad
0265-931X/$ – see front matter ? 2009 Elsevier Ltd. All rights reserved.
Journal of Environmental Radioactivity 101 (2010) 185–190
Author's personal copy
environmental tritium studies and monitoring. However, in spite of
this, this technique has remained largely ignored so far by the
environmental tritium community, with only a few studies repor-
ted in the litterature based on this technique (Surano et al., 1992;
Brown,1995; Kotzer et al.,1998). Here, we describe the application
of this mass spectrometric method to biological samples, with the
aim of providing essential elements for those interested in evalu-
ating or comparing measurement techniques for environmental
2. Experimental procedure
2.1. Samples collection, packaging and preparation
Since Tissus Free Water Tritium (TFWT) and exchangeable
Organically Bound Tritium (OBT) tend to equilibrate with their local
environment in a matter of hours or less (Mann, 1971), measure-
ment of environmental tritium samples requires precaution to
In the field, the samples (sediment, soil, grass, moss, vegetable,
fish, etc.) are stored in plastic boxes sealed in polyethylene bags
and then placed in a refrigerator at ?20?C within 24 h after
In the lab, each frozen sample (still in its box and polyethylene
bag) is placed in a vacuum-oven (at ambient temperature) con-
nected to a lyophilizer (Alpha 1–2 LD, Martin-Christ Gesellschaft,
Germany) already under vacuum at ?50?C. While the oven is being
flushed with argon, the polyethylene bag and the box cover are
quickly removed and the oven is immediately evacuated with
a primary pump to avoid contamination by ambient air moisture.
Then the valve between the oven and the lyophilizer is opened to
start lyophilization. At the end of drying procedure (typically 48 h),
the temperature of the oven is raised to 40?C for 12 h to eliminate
any remaining trace of water. Then the oven is flushed with argon
and the plastic box is quickly closed and sealed again in a poly-
ethylene bag filled with argon. The lyophilizer is stopped and filled
with argon as well. Once melted, the water is collected through
a valve at the bottom of the lyophilizer in a 500 ml Pyrex bottle for
Tissus Free Water Tritium (TFWT) analysis using our standard
procedure for tritium in water (Jean-Baptiste et al., 1992). To avoid
cross-contamination, the lyophilizer is rinced with tritium-‘‘free’’
water and dried with an hair drier between each sample (for
tritium-‘‘free’’ water, we use groundwater from the Paris basin
aquifer- commercial name ‘‘Chantereine’’, whose tritium content is
Dried sediments and soils are first sieved at 2 mm, and biolog-
ical samples are finely ground before being transferred into
a 100 ml low helium diffusivity Corning 1724 glass bulb (previously
weighed). To minimize3He blank, the bulbs are previously baked in
a flow of argon at 600?C for 24 h to remove the helium dissolved in
the glass. All manipulations are undertaken in a glove-box flushed
with argon to minimize contamination by ambient air moisture.
The bulb is attached to a high vacuum line and evacuated down to
<10?5Torr, then it is flame-sealed and weighed again to determine
the mass of the sample. After sealing, the samples are stored at
?20?C to further minimize helium diffusion through glass (Jean-
Baptiste et al., 1989).
Whenever larger samples are needed, either due to the very low
organic matter content of the sample (such as soils and sediments
whose organic content can be less than 1%) or because the time
cylinders closed with two high vacuumvalves mounted in series.
2.2. Mass spectrometry analysis
After a period of storage of typically 100–150 days, the bulbs are
connected to the inlet line of a mass spectrometer for3He analysis.
The instrument is a MAP-215-50 noble gas mass spectrometer
equipped with a stainless steel low blank inlet system (3He
blank<3?10?20mol) – Fig. 1. The pressure gauges, pneumatic
Nupro valves and breaking devices for glass bulb reopening
are computer-controlled so that up to 12 samples can be processed
in a row without any manual intervention. The measurements
3He/4He ¼1.38?10?6) drawn from a 5 L tank filled with clean air at
known pressure, temperature and relative humidity conditions,
through a precisely calibrated volume V¼122.2?0.1 (2s) mm3.
The3Heþion beam (typically in the range 1–250 ions/s) passes
through an electrostatic filter before impinging the detector (a 16-
stages electron multiplier connected to a pulse counting system).
Thanks to this filter, the3He background, which constitutes the
ultimate limit for3He detection, is verylow(<0.05 count/sec).4Heþ
is measured on a Faraday cup. The4He signal is usually very low
(<4?10?14mol) and corresponds to the small helium residue left
behind at the end of the degassing step (plus the blank of the mass
spectrometer inletsystem). The
(<5?10?20mol) can be calculated by applying the3He/4He ratio of
the blank component (see below).
The mass spectrometer is operated in a static mode. Five series
of ten 10-seconds integrations are performed on the3Heþpeak and
airstandard (He¼5.24 ppm,
Tritium production and inventory.
Release rate (PBq/an) Tritium inventory (PBq) References
Nuclear power plantsa
Total nuclear industry
Craig and Lal (1961), Nir et al. (1966), Jouzel et al. (1982)
Traub and Jensen (1995)
–Krejci and Zeller (1979), Combs and Doda (1979)
and Okada and Momoshima (1993)
bCumulated releases of atmospheric tests.
cCumulated releases (1950–1997).
dEstimate based on available data for the year 1978.
P. Jean-Baptiste et al. / Journal of Environmental Radioactivity 101 (2010) 185–190 186
Author's personal copy
on the baseline. Each series is interspaced by five measurements of
the4Heþpeak and baseline. Each sample measurement is followed
by a standard air aliquot measured in the same conditions.
However, whereas the sample contains some 10?14mol of gas only,
the air aliquot contains about 10?10mol of helium and neon (neon
is not quantitatively retained by the charcoal trap at 77 K – Fig. 1).
Hence, the pressure in the ion source is different for the standard
and for the sample, with potential effect on the ion source sensi-
tivity (Burnard and Farley, 2000). This means that although the3He
response of the mass spectrometer remains linear, the slope will
depend on the pressure in the source. The magnitude of this
pressure effect is determined by comparing the3He signals for an
aliquot of a3He–N2mixture (with no4He and no neon) and for the
same aliquot added to the standard air aliquot. The measured
pressure correction factor, FP, is in the range 0.85–0.95, depending
on the tuning of the mass spectrometer (a correction factor of 0.968
was reported in Clarke et al.,1976), and has to be closely monitored
through time (every week).
The amount of3He accumulated in the sample (typically in the
range 5?10?19–10?17 3He mol) is calculated by directcomparison to
the standard air aliquot. The
measurement of the
measured3He signal to give the tritiogenic component3HeT:
3He blank is inferred from the
4Heþresidue and is subtracted from the
where3Hestandardis the3He amount in the air standard aliquot,
N3sample, N4sample, N3standardand N4standardare the3He and4He
peaks height for the sample and for the standard air aliquot
respectively. Rais the atmospheric3He/4He ratio and Rblankis the
3He/4He ratio of the blank component. In theory, Rblankcan differ
from Raby up to ?15% (¼4/30.5–1) due to isotope fractionation
during the degassing step. Our own statistics of blank measure-
ments on samples analysed immediately following degassing (i.e.,
with no tritiogenic3He contribution) shows that Rblankis identical
to Rawithin 10%.
3. Activity and T/H isotopic ratio calculations
The specific tritium activity of the sample, AT(in Bq/kg of dry
sample), is directly deduced from the tritiogenic3He component
(3He)T(in mol) by the following formula:
m?1 ? e?ls?
where m represents the mass of the dry sample (in kg), l is the
tritium radioactive constant, N is the Avogadro number, and s the
It follows that the T/H ratio of the sample is given by:
where [H] is the hydrogen mass fraction of the analysed dry
material and MH
is the mass of one mole of hydrogen
(MH¼10?3kg). For TFWT, the hydrogen mass fraction [H] is
directly deduced from the composition of the water molecule
(hence [H]¼2/18¼0.11). The hydrogen mass fraction [H] of
organic materials is measured on a CHN analyser at the Micro-
Analysis Laboratory (CNRS/ICSN – Gif/Yvette). Plants usually have
an organically-bound hydrogen mass fraction in the range 0.05–
0.07. Fats have higher [H] contents up to w0.12. A comprehensive
database on elemental composition of food items and biological
tissus is available in the Handbook of Biological Data (1956) and in
the Geigy Scientific Tables (1981).
3.1. Tritium activity of the non-exchangeable OBT
Depending on authors, OBT values reported in the literature
may either represent the total organically-bound tritium or the
exchangeable hydrogen pool is the only hydrogen fraction that
faithfully records the history of environmental tritium seen by
a living organism during its growing period, a consensus seems to
Fig. 1. Schematic diagram of the mass spectrometer inlet system (PP, primary pump; TP, turbomolecular pump; IP, ion pump; GP, getter pump).
P. Jean-Baptiste et al. / Journal of Environmental Radioactivity 101 (2010) 185–190187
Author's personal copy
exist from now on over the idea that reported data should corre-
spond to this non-exchangeable fraction. If the TFWTand total OBT
content of a sample are known, the T/H value of the non-
exchangeable hydrogen pool (T/H)nexcan be deduced from the T/H
ratio of the totalOBT, (T/H)OBTand of the tissue-free water (T/H)TFWT
by assuming that the exchangeable hydrogen pool was at equilib-
rium with the tissue free-water:
where a is the fraction of exchangeable hydrogen (typically
between 0.2 and 0.3). (T/H)*TFWTis the T/H value of the tissue free-
water corrected for the tritium enrichment by Raleigh distillation
during lyophilization (Kim and Baumgartner, 1991):
ð1 ? aÞ
where r is the ratio of the tissue free-water hydrogen pool to
þexchangeable OBT). The value of the isotope fractionation factor
b corresponding to the lyophilization conditions was determined to
be b¼1.1 from special experiments performed on crushed ice of
known tritium concentration.
In a similar way, the specific activity of the non-exchangeable
tritium, (AT)nex(in Bq/kg of dry sample), can be determined from
the total specific tritium activity of the sample (AT)OBT(in Bq/kg of
dry sample) and the specific tritium activity of the Tissue Free
Water (AT)TFWT(in Bq/kg of water):
TFWT¼ ðT=HÞTFWT?ð1 ? rÞð1?bÞ=b
ðATÞnex¼ ðATÞOBT?a ? WEF ? ðATÞTFWT?ð1 ? rÞð1?bÞ=b
where WEF is the water equivalent factor of the dry material (i.e.,
the ratio of the hydrogen mass fraction of the sample to the
hydrogen mass fraction of water).
In our opinion, this procedure is preferable to the alternative
method which consists in eliminating the exchangeable organic
tritium by isotopic exchange with tritium-free water for about 48 h
(Pointurier et al., 2003). For non-aquatic samples, the isotopic
causes significant losses of organic matter by dissolution and/or
degradation of the sample (Pointurier et al., 2004) and leads to less
altered when soaked in water, from our own experience, both
methods are suitable and give consistent results. Note howeverthat
less time-consuming and avoids risks of contamination. In the case
the T/H value of the tritium-‘‘free’’ water.
The mainproblemwith non-exchangeable OBT determination is
that the exact proportion of the exchangeable hydrogen pool,
determined from equilibration experiments with deuterium or
tritium, is not always known with the desired accuracy. Although
the fraction of tritium that is readily exchangeable is known with
a good precision for some well documented terrestrial species from
replicated experiments (Lang and Mason, 1960; Guenot and Belot,
1984; Kim and Baumgartner, 1991), we do not know which values
could be used in the case of aquatic sedimentary and biota samples
for instance, due to the lack of experimental work that could serve
of reference in that matter. Also, experimental results for a given
species often reveal substantial discrepancies, depending on the
experimental procedure (Mann, 1971; Grinsted and Wilson, 1979).
In fact, organic hydrogen appears to be distributed over several
sub-fractions with different kinetics of isotopic exchange, thus
raising the question of the definition of the exchangeable pool of
organic hydrogen (Baumgartner and Gonhaerl, 2004). Note that
this serious limitation to the overall precision of the non-
exchangeable OBT determination is independent of the technique
used for measuring tritium, and is a general concern for the entire
environmental tritium community.
4. Results and discussion
In the3He ingrowth method, the main source of uncertainty
relates to the3Heþdetection. Fig. 2 shows a plot of the statistical
erroron3Heþcounting as afunction of the size of the3Heþpeak, for
a number of analysis performed during the past few months. For
4Heþ, statistical error is much smaller, ranging from 0.01% for the
standard air aliquot to 1% at the blank level. The overall uncertainty
in the tritium concentration, including counting statistics and all
other factors (3He and4He contentof the air standard aliquot, blank
correction, pressure correction, mass of the sample, etc.) is shown
in Fig. 3 for various sample sizes. Fora typical sample size of 20 g (of
water equivalent) and a storage time of three months, it follows
that the limit of detection (defined as the minimum concentration
of a substance being analysed that has a 99% probability of being
identified – EPA, 1997) is w0.15 Bq/kg. Fig. 3 shows that the tech-
nique is essentially ‘‘open-ended’’ in that larger samples (and/or
longer accumulation time) can be used to obtain lower detection
3He+ statistical counting error (%)
s / s t n
c ( k
Fig. 2. Plot of the3He statistical counting error (in %) vs3He peak height (in counts/s)
for mass spectrometer data collected over several months.
Tritium concentration (Bq/kg)
l a t o
k / q
r r e
0.001 0.01 0.1 1 10 100
Fig. 3. Monte-Carlo simulation of the total error (1s þsystematic) vs tritium
concentrations for various sample sizes (5,10, 20, 40, 80 and 160 g of water equivalent)
and a storage time of three months.
P. Jean-Baptiste et al. / Journal of Environmental Radioactivity 101 (2010) 185–190188
Author's personal copy
As an illustration of the overall reproductibility of the method,
the results of total OBT concentrations measured on two series of
replicates from two different lyophilized grass samples are shown
in Fig. 4. The standard deviations of the tritium concentrations are
almost identical to the total errors of the individual measurements,
indicating that the error calculation of the individual measurement
includes all major sources of uncertainty.
An intercomparison exercise (Fig. 5a) was carried out on
lyophilized grass samples with a conventional liquid scintillation
system (Pointurier et al., 2004) in a range of tritium concentrations
typical of environmental samples (between 1 and 40 Bq/kg of dry
material). Since the exercise was specifically designed to compare
the two analytical techniques, the measurements were carried out
on total OBT. For higher tritium concentrations, we participated to
an intercalibration with seven other labs of the french CETAMA
group (Comite ´ d’ETAblissement des Me ´thodes d’Analyse) operating
conventional beta counting systems. The sample distributed among
the participants was a lyophilized grass sample collected down-
wind of a stack releasing tritiated effluents. Fig. 5a and b shows that
the tritium results obtained by3He mass spectrometry are consis-
tent with the conventional liquid scintillation technique over
a wide range of tritium concentrations.
5. Concluding remarks
The3He ingrowthmethodis veryeffectivein determining lowto
very low-level tritium concentrations in environmental samples.
The main advantage of this non-destructive technique over the
methods of beta counting is its lower detection limit. A typical
detection limit of 0.15 Bq/kg is routinely obtainable for a sample
size of 20 g (of water equivalent) and an accumulation time of 3
months. This tritium detection limit can be pushed further down
(or the accumulation time reduced) with larger samples. Counting
time of the3Heþpeak can also be extended to improve counting
Another major advantage lies in the simplicity of the analytical
procedure which requires only two steps, e.g., complete degassing
of the sample and3He measurement. Processes such as combus-
tion, distillation, isotopic enrichment are not necessary, therefore it
is less susceptible to contamination. The only disadvantage is that
every measurement involves a mandatory storage period to allow
3He to accumulate. However, this accumulation time can be
reduced using larger samples and can also be adjusted based on
desired detection limits.
Therefore, in face of the globally declining tritium levels,3He
mass spectrometry may become the method of choice for routine
low-level environmental tritium monitoring and research in the
future, applicable to a wide variety of samples of radio-ecological
Baumgartner, F., Gonhaerl, W., 2004. Non-exchangeable organically bound tritium
(OBT): its real nature. Anal. Bioanal. Chem. 379, 204–209.
Brown, R.M., 1995. The Measurement of Tritium in Canadian Food Items. AECL-
CNRL/INFO-0499. Atomic Energy Control Board, Ottawa, Canada.
Burnard, P.G., Farley, K.A., 2000. Calibration of pressure-dependent sensitivity and
discrimination in Nier-type noble gas ion sources. Geochem. Geophys. Geosyst.
Clarke, W.B., Jenkins, W.J., Top, Z., 1976. Determination of tritium by mass-spec-
trometric measurement of He-3. Appl. Radiat. Isot. 27, 515–522.
Combs, F., Doda, R.J., 1979. Large-scale distribution of tritium in a commercial
product. Behaviour of Tritium in the Environment. IAEA-SM-232/57, Vienna, pp.
Craig, H., Lal, D., 1961. The production rate of natural tritium. Tellus 13, 85–105.
EPA, 1997. Terms of Environment. Environmental Protection Agency, United States.
Fourre ´, E., Jean-Baptiste, P., Dapoigny, A., Baumier, D., Petit, J.R., Jouzel, J., 2006. Past
and recent tritium levels in Arctic and Antarctic polar caps. Earth Planet. Sci.
Lett. 245, 56–64.
Geigy Scientific Tables, 1981. In: Lentner, C. (Ed.), Units of Measurements, Body
Fluids, Composition of the Body, Nutrition, vol. 1. Ciba-Geigy Ltd, Basle,
Grinsted, M.J., Wilson, A.T., 1979. Hydrogen isotopic chemistry of cellulose and
other organic materials of geochemical interest. New Zeal. J. Sci. 22, 281–287.
Guenot, J., Belot, Y.,1984. Assimilation of3H in photosynthesizing leaves exposed to
HTO. Health Phys. 47, 849–855.
IAEA, 2009. Global network of isotopes in precipitation. Isotope Hydrology Infor-
mation System (ISOHIS). International Atomic Energy Agency, Vienna. http://
Tritium concentration (Bq/kg)
3He mass spectrometry
n i t n
c a t e
k / q
o i t a
r t n
u i t i r
k / q
o i t a
r t n
u i t i r
y r t e
r t c
Fig. 5. (a) Comparison of total OBT concentrations in lyophilized grass samples
measured by liquid scintillation (Pointurier et al., 2004) and3He mass spectrometry.
The error bars correspond to 1-sigma errors of the individual measurements. (b) Result
of an inter-comparison exercise on a lyophilized grass sample between seven
conventional beta counting laboratories and our3He mass spectrometric system. The
error bars correspond to 1-sigma errors of the individual measurements.
k / q
o i t a
r t n
u i t i r
Fig. 4. Reproducibility of measured tritium concentrations for two lyophilized grass
samples of w30 g of dry matter (equivalent to w16 g of water) stored 44 days (sample
A) and 90 days (sample B). The error bars correspond to 1-sigma errors of the indi-
P. Jean-Baptiste et al. / Journal of Environmental Radioactivity 101 (2010) 185–190189
Author's personal copy Download full-text
Jean-Baptiste, P., Andrie ´, C., Lelu, M., 1989. Helium diffusion through glass. Glass
Technol. 30, 228–230.
Jean-Baptiste, P., Mantisi, F., Dapoigny, A., Stievenard, M., 1992. Design and perfor-
mance of a mass spectrometric facility for measuring helium isotopes in natural
waters and for low-level tritium determination by the3He ingrowth method.
Appl. Radiat. Isot. 43, 881–891.
Jenkins, W.J., 2007. Tracers of ocean mixing. In: Holland, H.D., Turekian, K.
(Eds.), Treatise on Geochemistry.
Jenkins, W.J., 2009. Tritium–helium dating. In: Steele, J., Thorpe, S., Turekian, K.
(Eds.), Encyclopedia of Ocean Sciences. Elsevier, Amsterdam, pp. 3048–3056.
Jouzel, J., Merlivat, L.D., Mazaudier, D., Pourchet, M., Lorius, C., 1982. Natural tritium
deposition over Antarctica and estimation of the mean global production rate.
Geophys. Res. Lett. 9, 1191–1194.
Kim, M.A., Baumgartner, F., 1991. Tritium fractionation in biological systems and in
analytical procedures. Radiochim. Acta 54, 121–128.
Kotzer, T., Trivedi, A., Waito, G., Workman, W., 1998. Ultralow-level measure-
ments of organically-bound tritium in bioassay samples. AECL-11956, COG-
98-131-1, 9 pp.
Krejci, K., Zeller, A., 1979. Tritium pollution in the Swiss luminous compounds
industry. Behaviour of Tritium in the Environment. IAEA-SM-232/11, Vienna,
Lang, A.R.G., Mason, S.G., 1960. Tritium exchange between cellulose and water:
accessibility measurements and effects of cycling drying. Can. J. Chem. 38,
Libby, W.F., 1955. Tritium in nature. J. Washington Acad. Sci. 45, 301–314.
Lucas, L.L., Unterweger, M.P., 2000. Comprehensive review and critical evaluation of
the half-life of tritium. J. Res. Natl. Inst. Stand. Technol. 105, 541–549.
Elsevier, Amsterdam,pp. 223–246.
Mann, J., 1971. Deuteration and tritiation. In: Bikales, N.M., Segal, L. (Eds.), Cellulose
and Cellulose Derivatives, Part IV. Wiley-Interscience, New York, pp. 89–116.
Nir, A., Kruger, S.T., Lingenfelter, R.E., Flamm, E.J., 1966. Natural tritium. Rev. Geo-
phys. 4, 441–456.
Okada, S., Momoshima, N., 1993. Overview of tritium: characteristics, sources and
problems. Health Phys. 65, 595–609.
Phillips, F.M., Castro, M.C., 2007. Groundwater dating and residence-time
measurements. In: Holland, H.D., Turekian, K. (Eds.), Treatise on Geochemistry.
Elsevier, Amsterdam, pp. 451–497. Chapter 5.15.
Pointurier, F., Baglan, N., Alanic, G., Chiappini, R., 2003. Determination of organically
bound tritium background level in biological samples from a wide area in the
south-west of France. J. Environ. Radioact. 68, 171–189.
Pointurier, F., Baglan, N., Alanic, 2004. A method for the determination of low-level
Spector, W.S. (Ed.), 1956. Handbook of Biological Data. National Academy of
Sciences, W.B. Saunders Company, Philadelphia.
Surano, K.A., Hudson, G.B., Failor, R.A., Sims, J.M., Holland, R.C., MacLean, S.C.,
Garrison, J.C., 1992. Helium-3 mass spectrometry for low-level tritium analysis
of environmental samples. J. Radioanal. Nucl. Chem. 161, 443–453.
Traub, R.J., Jensen, G.A., 1995. Tritium radioluminescent devices health and safety
manual. Pacific Northwest Laboratory Report PNL-10620/UC610. Richland,
Tremblay, P., Savard, M.M., Vermette, J., Paquin, R., 2006. Gas permeability, diffu-
sivity and solubility of nitrogen, helium, methane, carbon dioxide and formal-
dehyde in dense polymeric membranes using a new on-line permeation
apparatus. J. Membr. Sci. 282, 245–256.
UNSCEAR, 2000. Report of the United Nation Scientific Committee on the Effects of
Atomic Radiations to the General Assembly. United Nations, New York.
P. Jean-Baptiste et al. / Journal of Environmental Radioactivity 101 (2010) 185–190 190