Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration
Double doping of solid normal hydrogen with CH₃F and O₂ at about 4.2 K gives evidence of (ortho-H₂)n:CH₃F clusters and of O₂:CH₃F complex formation. A FTIR analysis of the time evolution of the spectra, in the ν₃ C–F stretching mode region, points out a behavior of the clusters very different from...
Saved in:
| Published in: | Физика низких температур |
|---|---|
| Date: | 2011 |
| Main Authors: | , |
| Format: | Article |
| Language: | English |
| Published: |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
2011
|
| Subjects: | |
| Online Access: | https://nasplib.isofts.kiev.ua/handle/123456789/118542 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Cite this: | Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration / L. Abouaf-Marguin, A.-M. Vasserot // Физика низких температур. — 2011. — Т. 37, № 4. — С. 456-462. — Бібліогр.: 31 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
nasplib_isofts_kiev_ua-123456789-118542 |
|---|---|
| record_format |
dspace |
| spelling |
Abouaf-Marguin, L. Vasserot, A.-M. 2017-05-30T15:22:07Z 2017-05-30T15:22:07Z 2011 Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration / L. Abouaf-Marguin, A.-M. Vasserot // Физика низких температур. — 2011. — Т. 37, № 4. — С. 456-462. — Бібліогр.: 31 назв. — англ. 0132-6414 PACS: 33.20.Ea, 36.40.–с https://nasplib.isofts.kiev.ua/handle/123456789/118542 Double doping of solid normal hydrogen with CH₃F and O₂ at about 4.2 K gives evidence of (ortho-H₂)n:CH₃F clusters and of O₂:CH₃F complex formation. A FTIR analysis of the time evolution of the spectra, in the ν₃ C–F stretching mode region, points out a behavior of the clusters very different from that of (ortho-H₂)n:H₂O clusters. The main point is the observation of CH₃F molecules migration in solid para-H₂ at 4.2 K, which is a behavior different from H₂O in identical experimental conditions. This is proved by the increase with time of the CH₃F:O₂ complex integrated intensity with a rate constant K=2.7(2) ⋅10⁻⁴s⁻¹. The authors are grateful to Professor David T. Anderson for helpful discussions and constructive comments. en Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України Физика низких температур Физические свойства криокристаллов Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration Article published earlier |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| title |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration |
| spellingShingle |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration Abouaf-Marguin, L. Vasserot, A.-M. Физические свойства криокристаллов |
| title_short |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration |
| title_full |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration |
| title_fullStr |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration |
| title_full_unstemmed |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration |
| title_sort |
infrared spectroscopy of solid normal hydrogen doped with ch₃f and o₂ at 4.2 k: ch₃f:o₂ complex and ch₃f migration |
| author |
Abouaf-Marguin, L. Vasserot, A.-M. |
| author_facet |
Abouaf-Marguin, L. Vasserot, A.-M. |
| topic |
Физические свойства криокристаллов |
| topic_facet |
Физические свойства криокристаллов |
| publishDate |
2011 |
| language |
English |
| container_title |
Физика низких температур |
| publisher |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| format |
Article |
| description |
Double doping of solid normal hydrogen with CH₃F and O₂ at about 4.2 K gives evidence of (ortho-H₂)n:CH₃F clusters and of O₂:CH₃F complex formation. A FTIR analysis of the time evolution of the spectra, in the ν₃ C–F stretching mode region, points out a behavior of the clusters very different from that of (ortho-H₂)n:H₂O clusters. The main point is the observation of CH₃F molecules migration in solid para-H₂ at 4.2 K, which is a behavior different from H₂O in identical experimental conditions. This is proved by the increase with time of the CH₃F:O₂ complex integrated intensity with a rate constant K=2.7(2) ⋅10⁻⁴s⁻¹.
|
| issn |
0132-6414 |
| url |
https://nasplib.isofts.kiev.ua/handle/123456789/118542 |
| citation_txt |
Infrared spectroscopy of solid normal hydrogen doped with CH₃F and O₂ at 4.2 K: CH₃F:O₂ complex and CH₃F migration / L. Abouaf-Marguin, A.-M. Vasserot // Физика низких температур. — 2011. — Т. 37, № 4. — С. 456-462. — Бібліогр.: 31 назв. — англ. |
| work_keys_str_mv |
AT abouafmarguinl infraredspectroscopyofsolidnormalhydrogendopedwithch3fando2at42kch3fo2complexandch3fmigration AT vasserotam infraredspectroscopyofsolidnormalhydrogendopedwithch3fando2at42kch3fo2complexandch3fmigration |
| first_indexed |
2025-11-25T20:29:39Z |
| last_indexed |
2025-11-25T20:29:39Z |
| _version_ |
1850521360882728960 |
| fulltext |
© L. Abouaf-Marguin and A.-M. Vasserot, 2011
Fizika Nizkikh Temperatur, 2011, v. 37, No. 4, p. 456–462
Infrared spectroscopy of solid normal hydrogen doped
with CH3F and O2 at 4.2 K: CH3F:O2 complex
and CH3F migration
L. Abouaf-Marguin and A.-M. Vasserot
Laboratoire de Physique Moléculaire pour l’Atmosphère et l’Astrophysique,
Université Pierre et Marie Curie-Paris 6, CNRS UMR7092, Paris F-75005, France
E-mail: luce.abouaf@upmc.fr
Received August 26, 2010
Double doping of solid normal hydrogen with CH3F and O2 at about 4.2 K gives evidence of (ortho-H2)n:CH3F
clusters and of O2:CH3F complex formation. A FTIR analysis of the time evolution of the spectra, in the ν3 C–F
stretching mode region, points out a behavior of the clusters very different from that of (ortho-H2)n:H2O clusters. The
main point is the observation of CH3F molecules migration in solid para-H2 at 4.2 K, which is a behavior different
from H2O in identical experimental conditions. This is proved by the increase with time of the CH3F:O2 complex in-
tegrated intensity with a rate constant K = 2.7(2) ⋅10–4 s–1.
PACS: 33.20.Ea Infrared spectra;
36.40.–с Atomic and molecular clusters.
Keywords: matrix isolation, solid hydrogen, CH3F:O2 complex, nuclear spin conversion, infrared spectroscopy.
1. Introduction
If room temperature normal hydrogen (nH2, 75% or-
tho-H2 with total nuclear spin I = 1, and 25% para-H2
with total nuclear spin I = 0) is condensed on a cold sub-
strate at cryogenic temperatures (< 5 K), the ortho-H2
concentration within the solid remains nearly constant on
the order of days. Indeed, in solid hydrogen mainly hcp
structured, H2 molecules are almost free rotors and nuc-
lear spin conversion between ortho and para states is for-
bidden [1,2]. Therefore, solid normal hydrogen freshly
condensed is composed of 25% para-H2 molecules (pH2)
(J = 0, total spin I = 0, degeneracy 1) with no electric
quadrupole moment (spherical symmetry), and 75% or-
tho-H2 molecules (oH2) (J = 1, total spin I = 1, degenera-
cy 3) with a quadrupole moment. Furthermore, ortho mo-
lecules can migrate inside the crystal. The nuclear spin
transition requires a magnetic field gradient, and in solid
normal hydrogen a very slow conversion ortho to para of
the H2 molecules is observed due to interaction between
neighboring oH2 molecules. This self-conversion is as-
sisted by the diffusion of ortho molecules, which are de-
localized inside the crystal [3,4]. A doping with a para-
magnetic molecule such as O2 enhances very efficiently
the conversion (catalyzed conversion).
In solid pH2, dopant molecules such as H2O [5,6], CH4
[7–10] and HCl [11], trapped in high-symmetry single
substitutional sites with 12 nearest neighbors, undergo al-
most free rotation. The CH3F molecule rotates only about
its C3 axis [12], and always carries a nonzero total nuclear
spin (I = 3/2 or 1/2).
Via their quadrupole moment, oH2 molecules interact
more strongly than pH2 ones with dopant molecules, and
are particularly able to form clusters with H2O and CH3F
in solid H2. For H2O [13], a nonstatistical cluster distribu-
tion of (oH2)n:H2O clusters have been identified (n = 1 to
11) using IR spectroscopy, and their evolution with time in
O2 doped solid normal hydrogen has been simulated [14].
(oH2)n:CH3F clusters have been identified (n = 1 to 12) in
pH2 crystals where the amount of oH2 was systematically
varied from low (300 ppm) to high concentrations (> 3%)
[15–17]. In pH2 solids containing high levels of oH2, even
larger (oH2)n:CH3F clusters (n > 12) are observed, with
oH2 molecules in the second solvation shell.
The present paper is devoted to an analysis of the time
evolution of the (oH2)n:CH3F clusters, hereafter labeled as
CLn, observed in the ν3 vibrational range (C–F stretching)
via IR spectroscopy, in solid normal H2 at 4.2 K. Doping
with O2 is used to accelerate the o/p conversion process in
solid nH2. A brief description of the experimental condi-
tions is given in Sec. 2. The spectroscopic observations are
Infrared spectroscopy of solid normal hydrogen doped with CH3F and O2 at 4.2 K
Fizika Nizkikh Temperatur, 2011, v. 37, No. 4 457
detailed in Sec. 3, for concentrations CH3F/O2/H2 =
= 1/0/4000, 1/10/6000 and 1/20/4000, with kinetic analysis
using spectral decompositions. The discussion in Sec. 4, is
centered on the CH3F:O2 complex assignment and time
evolution, and analyzes the differences between H2O and
CH3F behaviors. In the conclusion, Sec. 5, we point out
some questions raised by the present results.
2. Experimental procedures
The experimental setup has already been used exten-
sively for rare gas matrices [18,19]. Briefly, hydrogen
(purity N55, supplied by Alphagaz), H2O (doubly distilled
and degassed), O2, CH3F and CH4 are premixed at various
concentrations to ensure the homogeneity of the gaseous
mixture, and premixing favors complex formation [20]. A
solid polycrystalline film is obtained by deposition of the
gas mixture onto a cold gold-coated mirror, at a rate of
about 9 mmol/h, with deposition times ranging from 30 to
60 min, depending on dopant concentration and desired
absorbance. The temperature of 4.2 K, is given by a cali-
brated silicon diode sensor embedded in the OFHC copper
block of the mirror, and monitored by a S.E.R.2i BHT2400
temperature regulator, to get rid of slow drifts or oscilla-
tions, which are finally below ±0.02 deg [21]. The samples
thickness is approximately 40 to 60 µm, and its homogene-
ity is controlled by scanning a 2 scans spectrum every
5 torr deposited. The temperature gradient due to the spec-
trometer light is weaker than 1 deg. Indeed, the nuclear
spin conversion time of H2O diluted in solid Ar (1/2000) is
identical when irradiated permanently by the globar light
and irradiated only 15% of the time [21,22]. Infrared spec-
tra are recorded with a Bruker IFS113v FTIR spectrometer,
with ~ 0.15 cm–1 and ~ 0.2 cm–1 resolution at 2000 and
4000 cm–1, respectively. The time dependence of absorp-
tions is obtained by calculating integrated intensities of the
relevant IR features in each spectrum recorded with 40
scans.
3. Spectroscopic analysis
3.1. General observations
The oH2 concentration as a function of time is studied
using solid hydrogen IR absorption features recorded be-
tween 4000 and 5000 cm–1 and the empirical law of
Ref. 23. The time dependence of the oH2 concentration for
a CH3F/H2 = 1/4000 doped nH2 solid has been reported
previously [24]. A fit of the data to a first-order rate law
gives a time constant of 2520 min, which is slightly less
than the self-conversion time constant of 3570 min indicat-
ing CH3F can catalyze o/p conversion within the nH2 solid.
For the samples co-doped with O2, CH3F/O2/H2 =
= 1/10/6000 and 1/20/4000, the data again fit well to a
first-order rate law with time constants of about 1960 and
680 min, respectively [24]. Under these conditions the oH2
decay is predominantly catalyzed by O2 since it is present
at much higher concentrations than CH3F.
At time t = 0, for an oH2 concentration above 60% [23],
the CH3F spectrum exhibits several structures between 600
and 4000 cm–1 related to the different vibrations of
12CH3F, mainly ν3 (~1033 cm–1), ν6 (~1183 cm–1), ν2
(~1460 cm–1), ν5 (~1470 cm–1), 2ν5 (~2861 cm–1), ν1
(~2965 cm–1), and ν4 (~3014 cm–1). The most intense vib-
rations of isotopic 13CH3F also appear. A time dependence
is measured for all the CH3F absorptions, but the most
dramatic and useful spectroscopic changes are observed in
the region of the ν3 12C–F stretching vibration. Indeed, in
previous works [15,16], multiple resolved features appear
between 1030 and 1040 cm–1 for the ν3 stretching vibra-
tion, which have been assigned to (oH2)n:CH3F clusters
that form even in solid hydrogen matrices enriched in the
para nuclear spin state.
3.2. Spectra at time t = 0
In freshly deposited samples, whatever the O2 con-
centration, all the CH3F molecules are embedded in
large CLn (n > 12) clusters. The ν3 C–F stretching mode
(Fig. 1) exhibits a single asymmetric peak which can be
assigned to (oH2)n:CH3F clusters, with n > 12, at a fre-
quency (~1033 cm–1) lower than that of CL12 (1034.5 cm–1),
in agreement with previous works [15,16]. The ν3 peak ob-
served at 1032.9 cm–1 without O2 is asymmetric with a
sharp edge on the low-energy side and tails toward the
high-energy side. For increasing O2 concentrations of
1/600 and 1/200, respectively, the ν3 feature becomes
broader, more and more asymmetrical, and shifts to the
blue at 1033.1 and 1033.2 cm–1. These spectroscopic ob-
servations are consistent with small amounts of o/p con-
Fig. 1. Absorption spectra of CH3F in the v3 mode region at
time t = 0 for different O2 concentrations. The frequencies of
(oH2)n:CH3F clusters [15,16] are indicated at the top of the
figure.
1042 1040 1038 1036 1034 1032
Wavenumber, cm–1
0.2
1/0/4000
1/10/6000
1/20/4000
T = 4.2 K
CH F/O /H3 2 2
A
b
so
rb
an
ce
C
L
0
C
L
1
C
L
2
C
L
3
C
L
4
C
L
5
C
L
6
C
L
1
2
C
L
(
>
1
2
)
n
L. Abouaf-Marguin and A.-M.Vasserot
458 Fizika Nizkikh Temperatur, 2011, v. 37, No. 4
version occurring during deposition in the samples contain-
ing O2. In addition, two relatively broad peaks appear at
1038 and 1038.8 cm–1, and the intensity of these features is
obviously correlated with the O2 concentration.
3.3. Time effects
The time dependence of the ν3 spectra is illustrated in
Fig. 2. For both samples containing O2, the absorption as-
signed to clusters with n > 12 near 1033 cm–1 decreases in
intensity and becomes broader while shifting to higher
frequencies. These spectroscopic observations are consis-
tent with gradual conversion of the larger CLn clusters to
smaller cluster sizes as oH2 is converted irreversibly to
pH2. The 1038.8 cm–1 absorption present at time t = 0 in-
creases and narrows with time, while shifting also to high-
er frequencies. This last feature is assigned to the CL1 clus-
ter that seems to be present even at early times for samples
doped with O2. It appears that clusters CL0 and CL1 grow
directly from the decrease of clusters with n > 6. Clusters
with n between 3 and 6 can be clearly identified for the
CH3F/O2/H2 = 1/10/6000 concentration sample (Fig. 2,a).
A broad and weak feature labeled B, also observed in
Fig. 2 of Ref. 12, grows temporarily but then stops. A
small peak labeled S is observed, which has never been
mentioned previously, and grows in with time at about
1038.2 cm–1. This feature does not seem to belong to the
CLn cluster absorptions, and its integrated intensity is re-
lated to the O2 concentration. It continues to grow up to
the last spectrum recorded in Fig. 2,b.
3.4. Spectral decompositions and kinetics
Kinetics studies based on integrated intensities calculated
with fixed integration limits are not relevant here. Indeed, the
CLn cluster features are broadened by the presence of O2 and
by the high oH2 concentrations present at early times, and
thus the individual cluster features are not resolved but rather
are strongly overlapping. Furthermore, each CLn (n = 1–12)
cluster peak frequency shifts with time to higher values by at
least 0.5 cm–1. Therefore, kinetics are based on spectral de-
compositions (Fig. 3) carried out with similar continuity rules
as for H2O [14], but without details for n > 6. The kinetic re-
sults based on these spectral decompositions are presented in
Fig. 4 (integrated intensities in cm–1) for the structures which
Fig. 2. Time evolution of the spectra at T = 4.2 K: CH3F/O2/H2 =
= 1/10/6000 (a); CH3F/O2/H2 = 1/20/4000 (b). For B and S defi-
nitions, see text Sec. 3.3.
B S
0.3
1.2
3.3
8.1
14.4
25.3
41.9
64.4
12795
9228
6884
4696
3310
1450
794
134
73.80
oH2
(%)
t, min
0.3a
A
b
so
rb
an
ce
C
L
0
C
L
1
C
L
2
C
L
3
C
L
4
C
L
5
C
L
6
C
L
1
2
CH F/O /H = 1/10/60003 2 2
1042 1040 1038 1036 1034 1032
B S
0.0
0.0
0.0
0.1
1.3
3.5
8.4
14.6
27.2
49.6
64.3
18839
14014
10017
5020
2925
2098
1367
901
415
127
0
oH2
(%)
t, min
0.2b
Wavenumber, cm–1
CH F/O /H = 1/20/40003 2 2
A
b
so
rb
an
ce
C
L
(
>
1
2
)
n
Wavenumber, cm–1
1042 1040 1038 1036 1034 1032
Fig. 3. Example of spectral decomposition: CH3F/O2/H2 =
= 1/10/6000. Structure S appears between CL2 and CL3. For sCL
(clusters with n > 6), the decomposition (grey dotted lines) is just
used to reproduce the absorption profile.
SB
0.1
A
b
so
rb
an
ce
C
L
0
C
L
2
C
l 3
C
L
1
C
L
4
C
L
5
C
L
6
sC
L
Experiment
Total fit
Decomposition
Arbitrary lines
CH F/O /H = 1/10/60003 2 2
t = 5814 min
1042 1040 1038 1036 1034 1032
Wavenumber, cm–1
Infrared spectroscopy of solid normal hydrogen doped with CH3F and O2 at 4.2 K
Fizika Nizkikh Temperatur, 2011, v. 37, No. 4 459
can be clearly identified: CL0, CL1, CL2, S, sCL (clusters
with n > 6), and the total integrated intensity Σtot (from 1031.5
to 1045 cm–1). Intensities are corrected by the factor
[1 + 5.5⋅10–6 t] (t unit: min) to account for sublimation of the
sample. This sublimation correction is determined using the
slow decrease of the 4505 cm–1 structure of H2 observed in
the CH3F/O2/H2 = 1/20/4000 experiment after 6000 min (oH2
concentration < 0.1%). It is weaker than the correction deter-
mined for H2O experiments (8.7 ⋅ 10–6) which corresponds to
sample about 4 times thicker [23]. Figures 4,a (CH3F/O2/H2 =
= 1/10/6000) and 4,b (CH3F/O2/H2 = 1/20/4000) show similar
behaviors, with evidence of an acceleration in nuclear spin
conversion due to higher O2 concentration: features at
13000 min of Fig. 4,a are somewhat equivalent to features
at 4000 min of Fig. 4,b (oH2 ~ 0.3%).
The total integrated intensity Σtot of the ν3 mode decreas-
es during the conversion process until about 1% oH2 concen-
trations (about 10000 min for 1/10/6000, and about 5000 min
for 1/20/4000). At later times, only CL0, CL1, and structure S
remain, with a constant total integrated intensity (Fig. 4,b,
Fit 1). Therefore, the vibrational transition strength of clus-
ters with n > 6 is larger than that of CL0, CL1 and S, and we
consider that the oscillator strengths of CL0, CL1 and S are
similar. After 6000 min, CL1 remains almost constant
(Fig. 4,b, Fit 2), and we can conclude that S (Fit 3) grows
mainly at the expense of CL0.
At an oH2 concentration below 10%, in the approxima-
tion of a first-order decay, sCL vanishes with time con-
stants of about 1600 min (Fig. 4,a, Fit 0) and 1170 min
(Fig. 4,b, Fit 4) for concentrations 1/10/6000 and
1/20/4000, respectively.
4. Discussion
4.1. Assignments
The (oH2)n:CH3F cluster assignment is straightforward
from previous studies [15,16]. The broad transitory bump
at 1041 cm–1 labeled B, seems to be an intermediate be-
tween CL1 and CL0. It may be CH3F solvated in a cage of
12 pH2 molecules (CL0) with oH2 molecules in the second
or higher solvation shells, as observed for H2O [14]. It dis-
appears when the oH2 concentration vanishes (Fig. 2,b).
At time t = 0, the large absorption at 1038.8 cm–1 is as-
signed to CL1. The structure S is obviously related to the
O2 concentration. A complete analysis of the spectrum
from 400 to 5000 cm–1 did not allow to detect any unex-
pected impurity, so that S can be assigned confidently to
the complex CH3F:O2, hereafter labeled as
2OCL . It origi-
nates in the asymmetric feature at ~1038 cm–1 (Fig. 1)
most pronounced at long times.
4.2. CH3F diffusion: the CH3F:O2 complex
Based on the integrated intensities of feature S and CL0,
it appears that the concentration of CH3F:O2 is growing at
the expense of CH3F solvated completely by pH2 mole-
cules. This can be explained by a migration of CH3F mole-
cules when not clustered to oH2. Indeed, O2 is not sup-
posed to migrate under these conditions, as the H2O:O2
complex does not grow with time under analogous condi-
tions. Therefore, the growth of the CH3F:O2 complex can
be modeled by a diffusion controlled mechanism since the
two molecules probably stick to each other to form irre-
versibly the complex as soon as they are in nearest neigh-
bor positions, with almost no activation energy. It is not in
the scope of this work to develop a detailed analysis of the
diffusion, and we only draw up a balance of the process
Fig. 4. (Color online) Time dependence of the integrated intensity
of some structures (unit: cm–1). a — CH3F/O2/H2 = 1/10/6000,
Fit 0: exponential decay of sCL after 4000 min, time constant τ =
= 1600(90) min; b — CH3F/O2/H2 = 1/20/4000, Fit 1: linear re-
gression of Σtot, slope 5(3) ⋅10–7 min–1; Fit 2: linear regression of
CL1, slope 1.8(2) ⋅10–6 min–1; Fit 3: exponential growth of S,
integrated intensity ( )2OCLI fitted by the law: ( )2OCLI =
0[1 exp ( / )] ,I t I= Δ − τ + with I0 = 0.03(1), ΔI = 0.29(1), τ =
= 12200(600) min; Fit 4: exponential decay of sCL, time constant
τ = 1170(55) min.
0 2000 4000 6000 8000 10000 12000
0
0.2
0.4
0.6
0.8
1.0
1/10/6000
Fit 0
a
CL0 CL1 CL2
S sCL �tot
CH F/O /H3 2 2
In
te
g
ra
te
d
i
n
te
n
si
ty
0 3000 6000 9000 12000 15000 18000
0
0.2
0.4
0.6
0.8
1.0
1/20/4000
Fit 4
Fit 3Fit 2
Fit 1
b
t, min
CH F/O /H3 2 2
In
te
g
ra
te
d
i
n
te
n
si
ty
t, min
L. Abouaf-Marguin and A.-M.Vasserot
460 Fizika Nizkikh Temperatur, 2011, v. 37, No. 4
through the law of mass action, corresponding to the ele-
mentary reaction:
3 2 3 2CH F O CH F : O+ → ,
where the O2 concentration (
2O 1/ 600C = or 1/200) is
much higher than that of the CH3F molecules (
3CH F =C
= 1/4000). Under these conditions, the rate coefficient of
this reaction describes the mass transport at which reac-
tants encounter each other [25].
The free concentrations of the two partners (CH3F and
O2) obviously decrease with time, however, since
2 3O CH FC C>> , the O2 concentration can be approximated
as constant. If K is the reaction rate, the rate equation is
3
3 2
CH F
CH F O
dC
KC C
dt
= −
and
3 2CH F Oexp ( )C A KC t B= − + .
The decrease in the number of free CH3F molecules is ba-
lanced by the increase in CH3F:O2 clusters. Therefore, the
function to be fitted for the increase of structure S can be
written
O2CL 1 exp ,tC D E⎡ ⎤⎛ ⎞= − − +⎜ ⎟⎢ ⎥τ⎝ ⎠⎣ ⎦
where the parameter E corresponds to the concentration of
CH3F:O2 clusters formed during deposition, before time
t = 0; D represents the concentration of CH3F molecules
free to form CH3F:O2 clusters, and
2O1/ .KCτ =
Therefore in absorption spectra the time dependence
of the CH3F:O2 complex integrated intensity ( )2OCLI
should be fitted by the law:
( )2O 0CL 1 exp tI I I⎡ ⎤⎛ ⎞= Δ − − +⎜ ⎟⎢ ⎥τ⎝ ⎠⎣ ⎦
with integrated intensities I0 and (ΔI + I0) at times t = 0
and t → ∞, respectively.
The analysis of the increase of S with time, for concen-
trations 1/20/4000 (Fig. 4,b, Fit 3), leads to the following
values: I0 = 0.03(1), ΔI = 0.29(1), τ = 12200(600) min. It
means that at 4.2 K the rate of CH3F:O2 cluster formation
is K = 2.7(2)⋅10–4 s–1. At the end of the experiment, all the
CH3F molecules are either isolated in pH2 (CL0), clustered
with oH2 (CL1), or complexed with O2. We can suppose
that at infinite time all CH3F molecules will be clustered
with O2 with a total integrated intensity of 0.32.
For concentrations 1/10/6000, such a kinetic analysis is
not possible. Indeed, as observed in Fig. 2,a and in Fig. 3,
the
2OCL absorption feature is much weaker, and lies
almost completely obscured below the CL2 and CL3 ab-
sorptions at least up to 7000 min. It grows very slowly
(Fig. 4,a, black stars), as expected for this more dilute
sample. Figure 5 presents the result of a careful analysis of
S with integration limits adapted to the end of the experi-
ment (1037.54 and 1038.50 cm–1), when the frequencies
do not shift anymore. Because the
2OCL peak overlaps
with CL2 and CL3, the integrated intensity in this range
exhibits a maximum at about 6000 min. Examination of
Fig. 2,a shows that S can only be measured reliably after
~10000 min, when the CL2 and CL3 absorptions vanish.
Therefore, its time dependence is significant at the end of
the experiment between 10000 and 13000 min, which is
too short of duration for a fit.
However, both solid samples contain the same total
amount of CH3F, as confirmed by a same value of the total
integrated intensity at time t = 0 (Fig. 4), with almost all
CH3F molecules in clusters with n > 6. There is then a pos-
sibility to figure out what could be the time dependence of
2OCL in the more dilute sample, with a similar law:
( )2O 0CL 1 exp .tI I I⎡ ⎤⎛ ⎞′ ′ ′= Δ − − +⎜ ⎟⎢ ⎥′τ⎝ ⎠⎣ ⎦
Indeed 0,I I′ ′Δ and the time constant ,′τ can be esti-
mated knowing that the same CH3F amount, in a deposi-
tion 50% times longer at similar rate, implies a thickness of
the 1/10/6000 sample e′ = 1.5 e, if e is the thickness of the
1/20/4000 sample. Consequently:
• at time t = 0, some
2OCL clusters are formed upon de-
position with an integrated intensity I0 ~ 0.03 for concentra-
tions 1/20/4000. In the dilute sample, the CH3F and O2 con-
centrations are 1/1.5 and 1/3 times that of the concentrated
sample, respectively, with e′ = 1.5 e. Considering in a first
approach that the number of CH3F:O2 clusters formed during
Fig. 5. (Color on line) Time dependence of the integrated intensi-
ty of structure S (CH3F:O2 complex) for concentrations
CH3F/O2/H2 = 1/10/6000. Black stars: experimental data with
integration limits at 1037.54 and 1038.5 cm–1, relevant behavior
after 10000 min; dotted line: calculation with the law
( )2OCL 0.31 [1 exp ( / 54900)] 0.01.I t′ = − + The uncertainty area
is obtained at ± 20% of the parameters values.
0.12
0.10
0.08
0.06
0.04
0
2000 4000 6000 8000 10000 12000
t, min
In
te
g
ra
te
d
i
n
te
n
si
ty
Experiment
Calculation
Uncertainty
area
CH F/O /H3 2 2
1/10/6000
T = 4.2 K
0.02
Infrared spectroscopy of solid normal hydrogen doped with CH3F and O2 at 4.2 K
Fizika Nizkikh Temperatur, 2011, v. 37, No. 4 461
deposition is roughly proportional to the concentrations and
the thickness, it comes 0 0~ / 3 ~ 0.01;I I′
• at infinite time, all the CH3F molecules are clustered,
and the total
2OCL absorption should have the same val-
ue: 0 0~ ~ 0.32.I I I I′ ′Δ + Δ + It comes: ~ 0.31;I ′Δ
• for 1/10/6000, as the O2 and CH3F concentrations are
lower than for 1/20/4000, by 1/3 and 2/3 respectively, we can
suppose roughly that the time constant for the cluster forma-
tion may be about 4.5 times longer: ~ 4.5 ~ 54900′τ τ min.
The kinetic equation for the dilute sample is then sup-
posed to be roughly:
( )2OCL 0.31 1 exp 0.01 .
54900
tI ⎡ ⎤⎛ ⎞′ = − − +⎜ ⎟⎢ ⎥⎝ ⎠⎣ ⎦
This approximate behavior is represented in Fig. 5 with
a dotted line. All our measurements are accurate to within
± 20%. This leads us to define an uncertainty area between
the 2 grey curves for the increase of the CH3F:O2 cluster in
the dilute sample (Fig. 5). Finally, with all the overlapping
problems, and knowing that integrated intensity measure-
ments are somewhat dependent on the choice of the base-
line (which is able to shift all the data), we can conclude
that the increase of structure S is consistent in both expe-
riments. It confirms the assignment of S to the CH3F:O2
cluster, and the diffusion of CH3F molecules in solid pH2
under our experimental conditions.
4.3. Time dependence of sCL
The long observation periods presented here for these
CH3F/O2 double doped nH2 samples allows the various oH2
to pH2 conversion mechanisms to be studied. During the
course of the experiment there is a competition between oH2
self-conversion (very slow except at very high oH2 concen-
trations), O2 catalyzed conversion which is typically more
efficient than self-conversion but relies on oH2 migration to
the O2 catalytic center, and conversion catalyzed by CH3F
which is only active inside CLn clusters which are at compa-
rably low concentrations compared to O2. For example, the
shapes of the decay curves for sCL are qualitatively different
for the two CH3F/O2 samples studied. For the 1/10/6000
sample, initially the decay in the sCL integrated intensity is
very slow however it switches over to single exponential
decay in the 3000–4000 min time range and then decays
steadily with a time constant of 1600(90) min. Thus, for this
sample the balance between oH2 to pH2 conversion pro-
cesses that lead to growth and decay of (oH2)n:CH3F clusters
with n > 6 turns over within this time range. In contrast, for
the 1/20/4000 sample with the higher O2 concentration, the
sCL integrated intensity decays from t = 0 with a single ex-
ponential time constant of 1170(55) min. This suggest that
for the 1/20/4000 sample the higher O2 concentration rapidly
catalyzes oH2 to pH2 conversion within the (oH2)n:CH3F
clusters while instead for the 1/10/6000 sample, intracluster
conversion by CH3F is the dominant process.
4.4. Comparison with H20
Experiments with CH3F and H2O have been performed
under identical conditions, with the same amount of CO2
atmospheric impurity at 2300 cm–1. However for H2O,
clusters with n < 12 are only observed, and they grow and
decrease in a cascade-like process [14]. The aptitude of a
molecule to cluster with oH2 can be considered at first-
order to be due to its permanent dipole moment, via a di-
pole-quadrupole attraction. For H2O, the dipole moment is
1.841 D [26], and for CH3F, it is 1.8584 D [27], only about
1% higher. It is clear that this small difference cannot ex-
plain the much stronger attraction between CH3F and oH2.
To go further, the London dispersion interactions, induced
dipole-induced dipole interactions between instantaneous
transient dipoles have to be considered. They are propor-
tional to molecules polarizability. The polarizability of
CH3F, about 70% higher than that of H2O [28], could be
responsible for the higher ability of CH3F to catch oH2 at
4.2 K. Furthermore, the ability of CH3F to catalyze oH2
conversion may explain the weak growing of intermediate
clusters.
Contrary to CH3F, no H2O migration has been observed
[13,14]. At first sight, molecular migration within solid
hydrogen should be related to the size and mass of the mo-
lecule. To get an order of magnitude prediction of the mo-
lecular volumes of CH3F and H2O, we consider the bond
lengths, angles, and van der Waals radii of the atoms. For
H2O (RO–H = 0.9599 Å, θHOH = 104.45°) [29], we can
suppose that the freely rotating molecule occupies a sphere
with a diameter
2H OD ~ 3.92 Å. The symmetric top CH3F
molecule rotating about its C3 axis, with almost tetrahedric
angles (RC–H = 1.0870 Å, RC–F = 1.3884 Å, θHCF ~ 110°)
[30,31], occupies a cylinder with diameter
3CH Fd ~ 4.44 Å
and length
3CH Fl ~ 4.31 Å. Simple geometrical considera-
tions do not answer the question, as CH3F seems larger and
is more massive than H2O.
5. Conclusion
In the present work, we have shown that CH3F is a
much stronger sequester of migrating oH2 molecules than
H2O. It may be due to London dispersion interactions, but
theoretical developments are needed.
Furthermore, differently from what we have observed
for H2O [13,14], a migration of CH3F occurs in solid pH2
at about 4.2 K. It has been proved in our experimental con-
ditions, through the increase with time of the O2:CH3F
complex. In previous works, the very low concentrations,
the short duration of the experiments, and also the low
temperature, did not allow the detection of CH3F migra-
tion. A complete study of the comparative migration of
CH3F and H2O in solid pH2 would be welcome in an at-
tempt to identify which physical (or chemical) properties
of the molecules could be involved in such a process. In-
L. Abouaf-Marguin and A.-M.Vasserot
462 Fizika Nizkikh Temperatur, 2011, v. 37, No. 4
deed the potential for small molecules to migrate in solid
pH2 could be a very useful property for controlled chemi-
cal reactions.
Aknowledgment
The authors are grateful to Professor David T. Ander-
son for helpful discussions and constructive comments.
1. I.F. Silvera, Rev. Mod. Phys. 52, 393 (1980), and references
therein.
2. T. Oka, Ann. Rev. Phys. Chem. 44, 299 (1993).
3. V. Shevtsov, A. Scherbakov, P. Malmi, E. Ylinen, and M.
Punkkinen, J. Low Temp. Phys. 104, 211 (1996).
4. V. Shevtsov, P. Malmi, E. Ylinen, and M. Punkkinen,
Physica B284–288, 385 (2000).
5. M.E. Fajardo, S. Tam, and M.E. DeRose, J. Mol. Struct.
695–696, 111 (2004).
6. M.E. Fajardo and C.M. Lindsay, J. Chem. Phys. 128, 014505
(2008).
7. T. Momose, M. Miki, T. Wakabayashi, T. Shida, M.-C.
Chan, S.S. Lee, and T. Oka, J. Chem. Phys. 107, 7707
(1997).
8. S. Tam, M.E. Fajardo, H. Katsuki, H. Hoshina, T. Waka-
bayashi, and T. Momose, J. Chem. Phys. 111, 4191 (1999).
9. M. Miki and T. Momose, Fiz. Nizk. Temp. 26, 899 (2000)
[Low Temp. Phys. 26, 661 (2000)].
10. Y. Miyamoto, M. Fushitani, D. Ando, and T. Momose, J.
Chem. Phys. 128, 114502 (2008).
11. D.T. Anderson, R.J. Hinde, S. Tam, and M.E. Fajardo, J.
Chem. Phys. 116, 594 (2002).
12. Y.-P. Lee, Y.-J. Wu, and J.T. Hougen, J. Chem. Phys. 129,
104502 (2008).
13. L. Abouaf-Marguin, A.-M. Vasserot, C. Pardanaud, J. Stien-
let, and X. Michaut, Chem. Phys. Lett. 454, 61 (2008).
14. L. Abouaf-Marguin, A.-M.Vasserot, and C. Pardanaud, J.
Chem. Phys. 130, 054503 (2009).
15. K. Yoshioka and D.T. Anderson, J. Chem. Phys. 119, 4731
(2003).
16. K. Yoshioka and D.T. Anderson, J. Mol. Struct. 786, 123
(2006).
17. K. Yoshioka, P.L. Raston, and D.T. Anderson, Int. Rev.
Phys. Chem. 25, 469 (2006).
18. X. Michaut, A.-M. Vasserot, and L. Abouaf-Marguin, Vib-
rat. Spectr. 34, 83 (2004).
19. C. Pardanaud, A.-M. Vasserot, X. Michaut, and L. Abouaf-
Marguin, J. Mol. Struct. 873, 181 (2008).
20. V. Raducu, D. Jasmin, R. Dahoo, P. Brosset, B. Gauthier-Roy,
and L. Abouaf-Marguin, J. Chem. Phys. 102, 9235 (1995).
21. C. Pardanaud, thesis, Université Pierre et Marie Curie-Paris
6, pp. 75, 160 (2007).
22. L. Abouaf-Marguin, A.-M. Vasserot, C. Pardanaud, and X.
Michaut, Chem. Phys. Lett. 447, 232 (2007).
23. L. Abouaf-Marguin and A.-M. Vasserot, Chem. Phys. Lett.
460, 82 (2008).
24. L. Abouaf-Marguin, A.-M.Vasserot, and A. Lekic, Chem.
Phys. Lett. 470, 228 (2009).
25. D.F. Calef and J.M. Deutch, Ann. Rev. Phys. Chem. 34, 493
(1983).
26. M.T. do N.Varella, M.H.F. Bettega, M.A.P. Lima, and L.G.
Ferreira, J. Chem. Phys. 111, 6396 (1999).
27. M.D. Marshall and J.S. Muenter, J. Mol. Spectr. 83, 279
(1980).
28. T.N. Olney, N.M. Cann, G. Cooper, and C.E. Brion, Chem.
Phys. 223, 59 (1997).
29. A.F.A. Vilela, P.R.P. Barreto, R. Gargano, and C.R.M.
Cunha, Chem. Phys. Lett. 427, 29 (2006).
30. A. Karpfen and E.S. Kryachko, Chem. Phys. 310, 77 (2005).
31. O.R. Gilliam, H.D. Edwards, and W. Gordy, Phys. Rev. 75,
1014 (1949).
|