Non-statistical formation of J=1 T2 (ortho- T2 ) in recombination reaction of T+T+M→T2+M in liquid helium at 1.42-2.50 K

Article abstract of DOI:10.1016/S0009-2614(01)01233-7

During the course of an investigation on chemical behaviors of hydrogen isotopes in quantum media, non-statistical formation of ortho-T₂ was observed in recombination reaction of T+T+^3,4He→T₂+^3,4He in liquid

Full text of DOI:10.1016/S0009-2614(01)01233-7

7 December 2001
Chemical Physics Letters 349 ꢀ2001) 421±425
www.elsevier.com/locate/cplett
Non-statistical formation of J ˆ 1 T2 ꢀortho-T2) in
recombination reaction of T ‡ T ‡ M ! T2 ‡ M
in liquid helium at 1.42±2.50 K
a
Kazunari Iguchi , Takayuki Kumada , Kenji Okuno , Yasuyuki Aratono
b
a
b,
*
a
Faculty of Science, Shizuoka University, Ohya, Shizuoka 422-8529, Japan
Advanced Science Research Center, Japan Atomic Energy Research Institute, Tokai-mura, Naka-gun, Ibaraki 319-1195, Japan
b
Received 20 August 2001; in ®nal form 1 October 2001
Abstract
During the course of an investigation on chemical behaviors of hydrogen isotopes in quantum media, non-statistical
formation of ortho-T2 was observed in recombination reaction of T ‡ T ‡ 3;4He ! T2 ‡ 3;4He in liquid 3He±4He
mixture at 1.42±2.50 K. The experimental ortho-T2 fractions are larger than 0.9. The values are considerably higher
than the equilibrium fraction of 0.75 at room temperature. Similar results have been reported in H±H recombination at
low temperature. These facts suggest that similar reaction mechanism would be applied to both H and T atoms, which
have the same nuclear spin number of 1/2. Ó 2001 Elsevier Science B.V. All rights reserved.
1. Introduction
because vibrationally and rotationally excited hy-
drogen molecule ꢀHÃ2) cannot be stabilized via ra-
diative process of the recombination energy,
DE ˆ 4:476 eV [9]. A collision with a third body
causes the quasi-bound state to relax to a stable
truly bound state.
Since T has nuclear spin number of 1/2, T2 has
two nuclear spin isomers of para- and ortho-states
[10]. The former has the rotational quantum
number ꢀJ) of even and the latter the odd number.
Though the theoretical investigation of hydrogen
recombination reaction has been devoted primar-
ily to calculation of rate constant and to under-
standing of reaction mechanism, it can predict
ortho to para ratio in the recombined molecular
hydrogen [3±5]. Though some discrepancies be-
tween theoretical prediction and experimental
An atom±atom recombination reaction between
hydrogen atoms is the simplest and prototype
chemical process and thus many theoretical and
experimental works have been carried out [1±8].
According to the resonant recombination theory
[1±3,5], the reaction proceeds through collisional
deactivation with third body,
H ‡ H ! HÃ2
HÃ2 ‡ M ! H2 ‡ M ‡ DE ꢀM : third body†
* Corresponding author. Fax: +81-29-282-5927.
E-mail address: aratono@popsvr.tokai.jaeri.go.jp ꢀY. Ara-
tono).
0009-2614/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved.
PII: S0009-2614ꢀ01)01233-7

422
K. Iguchi et al. / Chemical Physics Letters 349 22001) 421±425
results have been reported on the ortho to para
ratio, non-statistical and preferential formation of
ortho-H2 ꢀo-H2† at low temperature predicted
theoretically has been qualitatively proved by ex-
perimental studies. Most of the investigations are
on H atom, one on D atom [2] but no report has
been published on T atom, to the best of our
knowledge.
Two of the authors ꢀKumada and Aratono)
have been studying tunneling abstraction reaction
of H and D in solid hydrogen and T in liquid
3He±4He mixture at very low temperatures [11,12].
Similar radiochemical methods used in the previ-
ous paper, i.e., nuclear activation for T formation
and radio-gaschromatographic analysis of reac-
tion products, were successfully applied in the
present experiments. Here we report non-statistical
ratio of o-T2 against p-T2 in the recombination
reaction of T in the liquid 3He±4He mixture at
1.42±2.50 K.
droxide [FeOꢀOH)] from Nakarai Tesque, which
has been used for conversion of o-H2 to p-H2 as
catalyst [13], was also utilized for the conversion of
o-T2 to p-T2 at room temperature. The fraction of
o-T2, fo-T , was calculated as follows:
2
f
o-T2
radioactivity of o-T2
radioactivity of p-T2 ‡ radioactivity of o-T2
ˆ
:
3. Results and discussion
Tritium atom is formed with a very high recoil
energy of 192 keV by the neutron absorption of
3He with a concomitant release of 576 keV proton
ꢀp) as 3He ‡ n ! TÃà ‡ pÃÃ, where TÃà and pÃÃ
mean translationally excited tritium atom and
proton.
The possible reaction pathways leading to the
formation of H2, HT and T2 can be considered as
follows:
2. Experimental
ꢀ1) Thermalization of TÃÃ and pÃÃ by successive
collisions with liquid 3;4He and neutralization of p
to H.
Helium-3 supplied by Isotec, has a listed iso-
topic enrichment of more than 99.9 at.%. Tritium
content in the gas is less than 1 Â 10À11%, which is
below detection limit of radioactivity counting
TÃÃ
‡
‡
3;4He ! T ‡ 3;4He
ꢀ1†
ꢀ2†
4
system. The purity of He is higher than 99.999%.
pÃÃ
3;4He ‡ eÀ ! H ‡ 3;4He
According to the phase diagram of the liquid
3He±4He mixture [12], the solution takes on a
normal¯uid or super¯uid state depending on
temperature and composition. The atomic fraction
Thermalization of TÃÃ and pÃÃ in liquid 3;4He oc-
curs very eciently due to very close mass to col-
lision entity. The kinetic energy of T or H after N
head-on collisions, EN , can be written as
3
of He in lique®ed mixture solution was measured
by quadrupole mass spectrometer and ranged
from 0.3 to 1. Cold neutron irradiation was per-
formed at neutron beam guide of JRR-3M ꢀup-
graded Japan Research Reactor No. 3) for about
40 h at 1.42±2.50 K. Neutron ¯ux measured
by Au-foil activation method was about 5 Â 1011
N
EN ˆ Ei  fm2=ꢀm1 ‡ m2†g ;
where Ei is the initial kinetic energy of T or p and
m1 and m2 are the masses of 3He or 4He and T or p,
respectively. Thermal energy at reaction tempera-
tures is 1:22 Â 10À4 eV ꢀ1.42 K) to 2:15 Â 10À4 eV
ꢀ2.50 K). Thus, N can be calculated to be less than
100 for all the experimental conditions. Even if the
collision is not always head-on, energy loss process
proceeds in condensed phase and thus recoil T and
p is thermalized suciently prior to recombination
reaction.
mÀ2 À1. Since the irradiation port is 30 m far from
s
reactor core, c-ray dose rate is very low and c-ray
radiolysis of T2 can be neglected. After the irra-
diated sample was recovered from irradiation
vessel to analyzing system, o-T2 and p-T2 were
separated by radio-gaschromatograph equipped
with a gas-¯ow proportional counter. Gamma
alumina column was used at 77 K. IronꢀIII) hy-
ꢀ2) Recombination reactions of the thermalized
H and T.

K. Iguchi et al. / Chemical Physics Letters 349 22001) 421±425
423
H ‡ H ! HÃ2
H ‡ T ! HTÃ
T ‡ T ! TÃ2
ꢀ3†
ꢀ4†
ꢀ5†
ꢀ6†
arises from tunneling abstraction and hot atom
reactions of T with surface impurity hydrogen of
the reaction vessel, will be responsible for the
formation of HT. The detailed reaction mecha-
nism of the reactions ꢀ3)±ꢀ6) including the wall
e€ect will be discussed after theoretical calculation
now in progress by our group. In the present
Letter, we focus on nuclear spin isomer ratio in T2
formation. The second peak in Fig. 1a is small and
was not observed in some experiments, as is seen in
Fig. 1b. The retention time of the second peak is
close to that of DT. However formation of DT
is negligible because natural abundance of D is
0.015%. Second and third peaks were considered
to be originated from p- and o-T2 and identi®ed as
follows. One of the samples irradiated at 2.50 K
was divided into two parts. One was analyzed as
such. Catalytic conversion of o-T2 to p-T2 was
applied to the second part at room temperature for
12.5 h. Radio-gaschromatographic results are
shown in Fig. 2. Two peaks are seen before con-
version ꢀFig. 2a). After treatment by catalyst, new
peak grows between these two peaks ꢀFig. 2b). The
fraction of the third peak ꢀradioactivity of third
peak/total radioactivity of second and third peaks)
is 0.89 and di€erent from equilibrium value ꢀ0.75)
at room temperature [15]. This is due to relatively
short contacting time of sample with catalyst. The
new peak was assigned to p-T2 on the basis of the
following considerations. ꢀa) In the present radio-
gaschromatographic condition, only tritiated hy-
drogen isotopes can be detected. ꢀb) The tritiated
hydrogen isotopes produced by ꢀ3)±ꢀ6) and wall
e€ect are HT and T2. ꢀc) HT is not responsible for
the formation of the new peak because chemical
reaction of HT with catalyst does not occur. These
suggest that the new peak originates from T2. ꢀd)
The retention time of the new peak is shorter than
that of the third peak. In H2 and T2 molecules,
ortho-state is more strongly bound to the surface
of absorbant because para-state is spherically
symmetric and ortho-state is in p orbital-like in
rotation [14]. This means p-T2 elutes faster than
o-T2. Same order was observed also for p-H2 and
o-H2 in the present experiments.
HÃ2; HTÃ; T2à ‡ 3;4He ! H2; HT; T2 ‡ 3;4He
HÃ2; HTÃ and TÃ2 are vibrationally and rotationally
excited hydrogen molecules and stabilized by col-
lisional energy transfer with liquid helium ꢀthird
body).
Fig. 1 shows radio-gaschromatograms at 2.25 K
ꢀa) and 1.42 K ꢀb). The ®rst peak was assigned to
HT by standard HT/DT mixture gas. In addition
to the reaction ꢀ4), so-called wall e€ect, which
(a) 2.25 K
HT 29093 cpm
(b) 1.42 K
HT 15028 cpm
Table 1 shows fraction of o-T2 as functions of
temperature and atomic fraction of He in normal
and super¯uid states. The values are the average of
3
Fig. 1. Radio-gaschromatograms of cold neutron irradiated
3He±4He mixture solutions at 2.25 K ꢀa) and 1.42 K ꢀb).

424
K. Iguchi et al. / Chemical Physics Letters 349 22001) 421±425
atures. If p-T2 or mixture of p-T2 and o-T2 is
produced in initial stage and is converted during
neutron irradiation, storage and separation, the
fraction of o-T2 should be less than 0.75. The
present results exceeding 0.9 indicate that con-
version does not occur under the present experi-
mental conditions and hence o-T2 is preferentially
formed at the reaction temperatures. No appre-
ciable di€erence is seen between super¯uid and
normal¯uid states.
Some theoretical studies based on orbiting
resonance complex mechanism have been carried
out in 1970s [1±3,5]. According to the theory, the
resonance complex is in a quasi-bound state with
lifetime longer than vibrational period of the
complex and has the characteristic rotational and
vibrational quantum numbers [1,2,5]. Because of
conservation of the overall symmetry, the rota-
tional selection rule for H2 is DJ ˆ À2 so that
odd J resonance complex contributes to the for-
mation of o-H2 while even J-state will results in
the formation of p-H2. Calculations show the
recombination of H atoms occurs through six
resonance states with vibrational and rotational
quantum numbers ꢀv; J) of ꢀ14, 5), ꢀ14, 4), ꢀ13,
8), ꢀ12, 12), ꢀ12, 11) and ꢀ11, 13) in descending
order of importance and predict preferential for-
mation of o-H2 [1±5]. This is qualitatively in
consistent with experimental results. The prefer-
ential formation of o-T2 obtained in the present
experiments seems to support the similar reaction
mechanism to that for hydrogen molecule. We
believe that comparison of experimental recom-
bination behaviors with theoretical calculation
among three hydrogen isotopes, H, D and T, is
important for better understanding of the reac-
Fig. 2. Assignment of peaks by catalytic conversion of o-T2 to
p-T2. ꢀa) Before conversion of o-T2 to p-T2. ꢀb) After conversion
of o-T2 to p-T2.
twice experiments except at 2.25 K ꢀthree times).
As is seen from Table 1, the o-T2 fractions are
higher than 0.9 over the entire range of temper-
Table 1
Fractions of o-T2 as functions of reaction temperature and atomic fraction of 3He
3
a
Temp.
ꢀK)
3He=ꢀ He ‡ 4He†
o-T2=ꢀo-T2 ‡ p-T2†
Phase
1.42
1.75
2.00
2.25
2.50
0:38 Æ 0:05
1:00 Æ 0:00
0:42 Æ 0:00
0:40 Æ 0:12
0:43 Æ 0:00
1:00 Æ 0:00
1:00 Æ 0:00
0:98 Æ 0:02
0:91 Æ 0:06
0:98 Æ 0:02
Super¯uid
Normal¯uid
Normal¯uid
Normal¯uid
Normal¯uid
a The values are the average of twice experiments except at 2.25 K ꢀthree times).

K. Iguchi et al. / Chemical Physics Letters 349 22001) 421±425
425
References
tion mechanism. From such a point of view,
theoretical calculation is in progress.
[1] R.E. Roberts, R.B. Bernstein, C.F. Curtiss, J. Chem. Phys.
50 ꢀ1969) 5163.
[2] R.E. Roberts, R.B. Bernstein, Chem. Phys. Lett. 6 ꢀ1970)
282.
4. Conclusion
[3] M. Menzinger, Chem. Phys. Lett. 10 ꢀ1971) 507.
[4] D.W. Trainor, D.O. Ham, F. Kaufman, J. Chem. Phys. 58
ꢀ1973) 4599.
Preferential formation of o-T2 in tritium atom±
tritium atom recombination reaction has been
observed in liquid helium at very low temperature.
The non-statistical formation of ortho-state in the
present experiments are qualitatively in consistent
with the theoretical prediction and experimental
results in H±H recombination reaction. Details on
the reaction mechanism will be reported in the
near future based on the theoretical calculation
which is currently under way.
[5] P.A. Whitilock, J.T. Muckerman, R.E. Roberts, J. Chem.
Phys. 60 ꢀ1974) 3658.
[6] L.P. Walkauskas, F. Kaufman, J. Chem. Phys. 64 ꢀ1976)
3885.
[7] D.N. Mitchell, D.J. LeRoy, J. Chem. Phys. 67 ꢀ1977)
1042.
[8] Y.M. Xiao, S. Buchman, L. Pollack, D. Kleppner, T.J.
Greytak, J. Chem. Phys. 96 ꢀ1992) 4032.
[9] T.J. Greytak, D. Kleppner, in: G. Grynberg, R. Stora
ꢀEds.), New Trends in Atomic Physics, North-Holland,
Amsterdom, 1984, p. 1186.
[10] I.F. Silvera, Rev. Mod. Phys. 52 ꢀ1980) 393.
[11] T. Kumada, K. Komaguchi, Y. Aratono, T. Miyazaki,
Chem. Phys. Lett. 261 ꢀ1996) 463.
Acknowledgements
[12] Y. Aratono, T. Matsumoto, T. Takayanagi, T. Kumada,
K. Komaguchi, T. Miyazaki, J. Phys. Chem. A 102 ꢀ1998)
1501.
The authors are grateful to personels of Me-
chanical Engineering Division of Japan Atomic
Energy Research Institute for their technical sup-
ports and to Drs. C. Yonezawa and H. Matsue
and Mr. F. Sasajima for their cooperation in
construction of the irradiation apparatus.
[13] T. Kumada, J. Kumagai, T. Miyazaki, J. Chem. Phys. 114
ꢀ2001) 10024.
[14] Y.L. Sandler, J. Chem. Phys. 29 ꢀ1958) 97.
[15] P.C. Souers, in: Hydrogen Properties for Fusion Energy,
University of California Press, Berkeley, 1986, p. 26.