Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms
Hydrogen atoms were trapped in a quench condensed Kr matrix and investigated by EPR. Each hyperfine component is a superposition of broad and narrow line. The spectrum of narrow lines shows an axial anisotropy of the hyperfine structure constant. The extent of the anisotropy is found to depend on...
Gespeichert in:
| Datum: | 2007 |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
2007
|
| Schriftenreihe: | Физика низких температур |
| Schlagworte: | |
| Online Zugang: | http://dspace.nbuv.gov.ua/handle/123456789/121782 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms / Yu.A. Dmitriev // Физика низких температур. — 2007. — Т. 33, № 6-7. — С. 661-667. — Бібліогр.: 21 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-121782 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1217822017-06-17T03:03:14Z Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms Dmitriev, Yu.A. Quantum Crystals Hydrogen atoms were trapped in a quench condensed Kr matrix and investigated by EPR. Each hyperfine component is a superposition of broad and narrow line. The spectrum of narrow lines shows an axial anisotropy of the hyperfine structure constant. The extent of the anisotropy is found to depend on both the deposition temperature, Tdep, and the temperature of the solid sample, Tsample. As Tdep increases, the broad lines diminish while the anisotropy of the spectrum of narrow lines becomes less pronounced. The spectrum of narrow lines originate from H atoms in well defined environments and is attributed to a superposition of two spectra given by the atoms in substitutional fcc and hcp sites. The spectrum of broad lines is assumed to originate from the atoms trapped in highly disordered regions in the lattice. These regions are found to start relaxing at Tsample as low as 12 K. 2007 Article Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms / Yu.A. Dmitriev // Физика низких температур. — 2007. — Т. 33, № 6-7. — С. 661-667. — Бібліогр.: 21 назв. — англ. 0132-6414 PACS: 68.55.–a; 68.55.Ln; 76.30.–v; 61.72.Nn http://dspace.nbuv.gov.ua/handle/123456789/121782 en Физика низких температур Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
Quantum Crystals Quantum Crystals |
| spellingShingle |
Quantum Crystals Quantum Crystals Dmitriev, Yu.A. Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms Физика низких температур |
| description |
Hydrogen atoms were trapped in a quench condensed Kr matrix and investigated by EPR. Each hyperfine
component is a superposition of broad and narrow line. The spectrum of narrow lines shows an axial anisotropy
of the hyperfine structure constant. The extent of the anisotropy is found to depend on both the deposition
temperature, Tdep, and the temperature of the solid sample, Tsample. As Tdep increases, the broad lines diminish
while the anisotropy of the spectrum of narrow lines becomes less pronounced. The spectrum of
narrow lines originate from H atoms in well defined environments and is attributed to a superposition of two
spectra given by the atoms in substitutional fcc and hcp sites. The spectrum of broad lines is assumed to originate
from the atoms trapped in highly disordered regions in the lattice. These regions are found to start relaxing
at Tsample as low as 12 K. |
| format |
Article |
| author |
Dmitriev, Yu.A. |
| author_facet |
Dmitriev, Yu.A. |
| author_sort |
Dmitriev, Yu.A. |
| title |
Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms |
| title_short |
Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms |
| title_full |
Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms |
| title_fullStr |
Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms |
| title_full_unstemmed |
Structural formation and thermal relaxation of quench-condensed Kr films: effect on EPR spectrum of trapped hydrogen atoms |
| title_sort |
structural formation and thermal relaxation of quench-condensed kr films: effect on epr spectrum of trapped hydrogen atoms |
| publisher |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| publishDate |
2007 |
| topic_facet |
Quantum Crystals |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/121782 |
| citation_txt |
Structural formation and thermal relaxation of
quench-condensed Kr films: effect on EPR spectrum of
trapped hydrogen atoms / Yu.A. Dmitriev // Физика низких температур. — 2007. — Т. 33, № 6-7. — С. 661-667. — Бібліогр.: 21 назв. — англ. |
| series |
Физика низких температур |
| work_keys_str_mv |
AT dmitrievyua structuralformationandthermalrelaxationofquenchcondensedkrfilmseffectoneprspectrumoftrappedhydrogenatoms |
| first_indexed |
2025-07-08T20:30:56Z |
| last_indexed |
2025-07-08T20:30:56Z |
| _version_ |
1837112133772378112 |
| fulltext |
Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7, p. 661–667
Structural formation and thermal relaxation of
quench-condensed Kr films: effect on EPR spectrum of
trapped hydrogen atoms
Yu.A. Dmitriev
A.F. Ioffe Physico-Technical Institute, 26 Politekhnicheskaya Str., St. Petersburg 194021, Russia
E-mail: dmitriev.mares@pop.ioffe.rssi.ru
Received October 20, 2006
Hydrogen atoms were trapped in a quench condensed Kr matrix and investigated by EPR. Each hyperfine
component is a superposition of broad and narrow line. The spectrum of narrow lines shows an axial aniso-
tropy of the hyperfine structure constant. The extent of the anisotropy is found to depend on both the deposi-
tion temperature, Tdep, and the temperature of the solid sample, Tsample. As Tdep increases, the broad lines di-
minish while the anisotropy of the spectrum of narrow lines becomes less pronounced. The spectrum of
narrow lines originate from H atoms in well defined environments and is attributed to a superposition of two
spectra given by the atoms in substitutional fcc and hcp sites. The spectrum of broad lines is assumed to orig-
inate from the atoms trapped in highly disordered regions in the lattice. These regions are found to start re-
laxing at Tsample as low as 12 K.
PACS: 68.55.–a Thin film structure and morphology;
68.55.Ln Defects and impurities: doping, implantation, distribution, concentration;
76.30.–v Electron paramagnetic resonance and relaxation;
61.72.Nn Stacking faults and other planar or extended defects.
Keywords: thin films, trapped atoms, atomic spectra.
1. Introduction
In the last decades a large amount of the experimental
and theoretical work has been focused on the properties
of small clusters of inert-gas elements and their solids
with pores on the nanometer scale. A highly porous media
(gels) were obtained in experiments when a jet of helium
containing small amounts of impurity atoms and mole-
cules was injected into superfluid helium [1]. Synchro-
tron x-ray diffraction studies [2] provide evidence that
these materials are made of particles or small clusters sur-
rounded by, possibly, layers of solidified helium.
Another considerable effort has been spent on a
closely-related problem: obtaining and studying disordered
solids of pure rare gases the thermodynamically stable phase
of which shows a crystalline order. Strzhemechny and co-
workers [3] undertook x-ray diffraction investigations into
the structure and morphology of low-temperature quench-
condensed binary alloys of hydrogen with argon and kryp-
ton in search of possible highly amorphized states in both
systems. They concluded that the states formed in as-grown
samples are in many respect analogs of helium-impurity me-
dia. Atoms and small molecules are often used as probes to
study microstructure of inhomogeneities in quench-depos-
ited rare gave solids. Methane molecules matrix-isolated in
rare gas solids have been used as such a probe [4–7]. High-
resolution neutron scattering spectra of quench-condensed
nonequilibrium CH4/Kr samples showed two rotational
transitions for lowest levels [4]. The authors attributed one
of them to J � �0 1transition of the almost free CH4 quan-
tum rotor on a fcc substitutional site of the Kr lattice, while
the other was interpreted as J � �0 1 transition at hcp ma-
trix sites caused by the presence of stacking faults or hcp
crystallites. They also stressed that the different width of the
lines makes it probable that fcc and hcp sites do not exist
within the same crystallite as a result of individual stacking
faults but are more likely originating from fcc and hcp crys-
tallites. Two dominant sites for guest molecules were ob-
served in Ar, Kr, and Xe hosts in experiments where
high-resolution infrared absorption spectra were recorded
for the v4 vibrational mode of CH4 trapped in these rare gas
solids [5]. Having analyzed the fine structure of the spectra,
© Yu.A. Dmitriev, 2007
annealing studies and effects of deposition temperature, the
authors came to the conclusion that one of these two sites
was for CH4 molecules trapped in the normal cu-
bic-close-packed environment while the other site was for
CH4 trapped in hexagonal-close-packed pockets induced by
stacking faults formed during deposition.
Neutron scattering experiments on vapor deposited
CH4/Ar films [6] revealed sharp transitions which origi-
nated from CH4 molecules in well defined environments.
They were attributed to molecules in an undisturbed fcc site,
a fcc site neighboring a stacking fault, an hcp site in hcp
crystallites, and an hcp site within a single stacking fault.
CH4/Ne samples prepared by matrix condensation showed
addition diffraction peaks of hexagonal regions in addition
to broadened diffraction lines of the equilibrium fcc cubic
lattice. This also means that stacking faults, creating locally
hexagonal environments, must be present at high concentra-
tion [7]. Based on neutron and x-ray diffraction data on va-
por deposited pure Kr [8], the authors reported the basic
sample structures to be fcc and in some cases also hcp,
which originated from fcc due to stacking faults.
The structure of condensed pure Kr films was also in-
vestigated in x-ray diffraction experiments by Menges
and L�hneysen [9]. They observed only peaks which could
be attributed to the fcc structure and stressed that no hints
were found for the existence of the hcp structure.
Strzhemechny et al. [3] examined quench-deposited Kr/H2
and Ar/H2 mixtures by x-ray diffraction. For lower nominal
hydrogen fractions (c � 10%) they saw only reflections that
could be attributed to the (111) line from a krypton-rich cu-
bic lattice.
Pulsed NMR was used to study ortho-H2 as a dilute im-
purity in solid Ne, Ar, Kr, and para-H2 [10]. Having ana-
lyzed a temperature dependence of the intramolecular nu-
clear-spin relaxation of isolated o-H2 molecules, the authors
came to the conclusion that the introduction of an o-H2 mol-
ecule into solid p-H2 is not accompanied by large distortions
and random static crystal fields, while the Ne and Ar results,
on the other hand, indicate lack of symmetry for significant
crystal fields at o-H2 sites. The authors also point out that
these fields may reflect distortions that accompany the intro-
duction of an o-H2 impurity into the rare gas solids, either
substitutionally or otherwise. They also considered another
possibility of o-H2 molecules preferentially collecting
around disordered sites and found indirect evidences in fa-
vor of this. The case of H2/Kr is more complex, and the rele-
vant results indicate more symmetric fields, perhaps nearly
axial. The authors suggester an intramolecular relaxation
model based on a mixture of sites with large crystal fields of
no symmetry (70%) and axial symmetry (30%).
Thus, comparing the above results, there remain some
questions concerning the structure of nonequilibrium rare
gas solids and immediate surroundings of the impurity
particles in these solids. The aim of this study is to obtain
additional information in order to resolve the problem us-
ing the matrix isolation and EPR techniques. Solid Kr is
the most suitable object because of both the high sublima-
tion temperature favorable for sample growing into small
single crystals and the absence of the nuclear magnetic
moments resulting in the high resolution spectra.
2. Experimental
The discharge part of the setup as well as the microwave
cavity and the gas filling and purification system remained
unchanged [11]. The major distinction is that the bottom of
the quartz finger which served as the low temperature sub-
strate was cooled by liquid He vapor. The substrate temper-
ature was controlled by changing the vapor flow. This tem-
perature was measured by a thermocouple located on the
outside surface of the finger. To monitor the vapor tempe-
rature, a carbon-resistance thermometer was mounted in-
side the finger.
3. Results and discussion
Figure 1,a shows one of the spectra of trapped H atoms
we obtained in the present experiments. The substrate
temperature during deposition, Tdep , was about 6.5 K; the
commonly used gas flows were: 0.4 and 0.6 mmole/h
through channels A and B, respectively. The H2 concen-
tration in the mixture with Kr passed through the dis-
charge was 5%. The concentration of molecular hydrogen
in the solid sample estimated from geometry of the depo-
662 Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7
Yu.A. Dmitriev
50 dB
a
3087 3090 3593 B, G
b
26 dB
Fig. 1. The EPR spectrum of hydrogen atoms trapped in solid
krypton at 6.5 K. The attenuation of the microwave power is
50 dB (a), 26 dB (b). The sample temperature during registration
is 5 K. The microwave resonance frequency is 9410.82 MHz.
sition part of the set-up was 1.7�10
–4
. The anisotropy of
the lines clearly seen from Fig. 1,a and a superposition of
narrow and broad lines which is obvious from Fig. 1,b is a
characteristic feature of such spectra. We have found that
the narrow lines saturate more readily than the broad
ones. This behavior allows distinguishing any of these
spectra.
We have also found that the spectrum of broad lines
became less pronounced with increasing Tdep , while the
extent of anisotropy and the linewidth of the spectrum of
narrow lines decreased (Fig. 2,a). Figure 2,b shows a
spectrum obtained after an annealing of short duration.
Immediately after the annealing the sample temperature
was decreased to 12 K and isotropic lines were recorded.
Then the temperature was gradually increased to 23.5 K.
Based on a comparison of the spectra taken in different
runs, we came to the astonishing conclusion that the
broad line position changed slightly from run to run. The
only explanation is that the broad lines do not come from
one type of centers, but from centers of several types.
This assumption was verified by changing the sample
temperature, Tsample . The broad line shape changed irre-
versibly as the sample temperature increased (Fig. 3). The
lines not only appeared less intensive, but revealed a
structure.
A detailed investigation of the transformations of the
broad line shape showed that, as Tsample was increased in
a gradual manner, the transformation finished in general
before the temperature rose to about 20 K. Thus, we have
observed a lattice relaxation at quite low temperatures
when the Kr-atom self diffusion in the equilibrium lattice
is insignilicant. During such a relaxation, some of the H
atoms giving the broad lines were lost in the recombina-
tion process. Some transitions contributing to the broad
line disappeared while other transitions appeared nar-
rower becawse the surrounding became more regular. The
narrow lines showed only a minor decrease in the inten-
sity as the temperature increased to a moderate value. The
broad EPR lines observed for a freshly prepared sample
come from the H atoms trapped in the matrix in the highly
disordered or even amorphous state. The former assump-
tion seems to be more likely because it is generally agreed
that no pure amorphous rare gas solids can be produced
with experimentally attainable quenching rates [9]. On
the other hand, the amorphous state might appear as a lo-
cal structure not observed by x-ray or neutron diffraction
methods. We suppose that the broad line spectrum is not
associated with the porosity of the solid gas matrix. The
reason for this assumption is that the number of pores
starts to prevail at deposition temperatures below 35 K for
Structural formation and thermal relaxation of quench-condensed Kr films
Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7 663
a
b
3087 3089 3593 B, G
Fig. 2. EPR spectra of hydrogen atoms trapped in solid krypton
at 10.5 K: a — before annealing; the sample temperature during
recording is 11 K, the microwave resonance frequency is
9410.90 MHz, the attenuation of the microwave power is 50 dB;
b — after annealing of short duration at approximately 40 K; the
sample temperature during recording is 23.5 K, the attenuation
of the microwave power is 40 dB, the spectrometer gain is in-
creased.
a
b
d
c
Fig. 3. Effect of the sample temperature, Tsample, on the shape of
the broad line. Only the low-field component is shown: a — after
deposition on the substrate at 4.2 K, Tsample = 9 K; b — after a
short annealing at 29 K, Tsample = 9 K; c — the sample tempera-
ture is gradually increased up to 25 K, Tsample = 25 K; d — the
sample temperature is then decreased to 4.2 K, Tsample = 4.2 K.
the Kr solid [12]. The pore concentration increases, the
pore size distribution widens and its maximum shifts to-
wards smaller sizes as the Tdep decreases in a rather grad-
ual manner [12]. On the other hand, our results suggest a
sharp increase of the intensity of the broad lines when
Tdep is below 7 K. It is interesting that a diffuse intensity
was observed in the neutron diffraction study of the
quench condensed solid Kr [8]. The authors assumed that
the diffuse intensity was due to continuously deformed
unit cells. Thus, a local perturbation of the lattice relaxes
over a large number of unit cells into unperturbed zones
of the crystal. Our observation of the broad EPR lines
supports the presence of such continuously distorted re-
gions.
Another unusual result of the present study is the aniso-
tropy of narrow lines. The spectrum parameters measured
here, A iso � 1408.97(21) MHz and g iso � 2.00164(12), are
close to the values reported earlier [13,14] for the substitu-
tionally trapped H atoms. One can see from Fig. 1, that the
axial anisotropy of the hyperfine structure constant A is
well pronounced, while the g-factor anisotropy is not evi-
dent. We estimate the difference between perpendicular,
A� , and parallel, A | |, components to be about 3 MHz. This
rather small anisotropy was possible to be observed be-
cause of the small linewidth which implies well defined
surroundings of the trapped H atoms. The linewidth de-
pends on both Tdep and Tsample and on the ranges from
about 0.12 to 0.3 G. The axial anisotropy of the spectrum is
a fingerprint of the axial crystal field at the H atom loca-
tion. Let a and b be two amplitudes measured between the
base line and the two peaks of the EPR line. Here a is the
greater one. Then a b/ ratio may be considered as a mea-
sure of the line anisotropy. In this way we found the aniso-
tropy to decrease as the Tsample increased. This is evident
from Fig. 4.
The extent of the anisotropy at a given Tsample depends
on Tdep . For samples prepared at substrate temperatures
below approximately 5 K and, especially, when the broad
line spectrum was pronounced, the anisotropy to might
persist up to temperatures of about 40–48 K (depending
on the conditions during deposition). At these tempera-
tures the solid gas layer started to evaporate in our
experiments. Very recently, a characteristic desorption
temperature of Tdes � 42 K has been found for quench-
condensed Kr films by means of high-frequency surface
acoustic waves [15]. Thus, our estimation of Tdes checks
with the earlier observation. Even after a deep annealing
of short duration, the spectrum might stay anisotropic,
Fig. 4. As Tsample decreased, the line showed an increased
anisotropy. The anisotropy may regain its initial extent.
This depends in part on the maximum Tsample reached du-
ring annealing and on the annealing duration. The not
fully reversible line shape suggests some structural relax-
ation or H atoms diffusion into other sites. On the other
hand, samples obtained at Tdep above approximately 10 K
show spectra closer to the isotropic EPR spectrum,
Fig. 2,a. After a short annealing, the spectrum intensity
went down and the line shape changed to isotropic,
Fig. 2,b. The transformation has occurred. This result is in-
herent in the spectra of samples obtained at high Tdep .
Figure 5 shows a reversible change in the line shape.
The sample was prepared at Tdep of about 7 K and was
subjected to intermediate annealing at 25 K, resulting in
the broad lines going down. Being initially anisotropic,
the lines grew isotropic at high temperatures and regained
its initial shape after the sample was again cooled down to
low temperatures. In the process, the amount of EPR cen-
ters dropped essentially, suggesting recombination of dif-
fusing atoms. Interestingly, the shape variation had no ef-
fect on the line parameters, A iso and g iso . We suppose that
the spectrum behavior described above may be explained
assuming this spectrum to be a superposition of two spec-
tra: one is due to H atoms trapped in fcc regions and the
other, by atoms in hcp regions. Indeed, it was shown
[16,17] that the hyperfine structure constant and the
g-factor are governed mainly by the pair interaction be-
tween a trapped atom and host atoms and molecules. In
this case, A and g are sensitive to the distance between the
H atom and the host particles. These distances are the same
664 Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7
Yu.A. Dmitriev
a
b
9.5 K 15 K
18.5 K
26 K
6.5 K
282624222018161412108
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
a/
b
T, K
Fig. 4. The temperature dependence of the a b/ ratio in one of
the runs. The ratio serves as a tentative measure of the aniso-
tropy (Tdep = 7 K, microwave power attenuation is 50 dB): a —
Tsample is decreased (solid circles), Tsample is again lowered
down (a star); b — tracing the low-field line through the tem-
perature cycling; the last two lines are recorded at an amplifi-
cation 1.5 times hidher.
for the immediate surroundings in fcc and hcp structures.
The number of matrix atoms nearest to a trapped H atom is
also equal for these two structures. So, the lines of the cor-
responding spectra are expected to coincide. The changes
in the extent of the anisotropy are then explained by varia-
tions in the relative number of fcc (surroundings of cubic
symmetry) and hcp (surroundings of axial symmetry) sites.
In accordance with earlier observations [4–8] the hexago-
nal environment appears only under quench condensation
conditions. The present results correlate well with the re-
ported high stability of the hcp structure [6].
It was found in neutron experiments on CH4 trapped in
solid Ar [6] that the hcp crystallites, which were assumed
to be the cause of the line at 0.59 meV, required 1 h annea-
ling at Tsample = 80 K to disappear through recrystalli-
zation into fcc structure. That is why we have still ob-
served, in some experiments, the EPR spectrum anisotropy
even after a deep short annealing. On the other hand, the
transformation at very low temperatures of a part of the hcp
regions into fcc is evidenced by the irreversible anisotropy
in Fig. 4. The irreversible decrease in the anisotropy upon a
considerable change in Tsample , also observed in the pres-
ent investigation, is most likely explained by the diffusion
of H atoms occuring at a sufficiently high temperature with
subsequent trapping. Since the major part of the lattice has
the fcc structure, the extent of the spectrum anisotropy de-
creases again after trapping. In the present experiments,
such a decrease is frequently accompanied by a decline in
line intensity, which confirms the assumed diffusion. In
turn, the reversible change in the anisotropy may be ex-
plained by: a) an effect which oscillations of the neighbor-
ing matrix atoms probably have on the anisotropy extent,
b) an effect of fast motion of H atoms on the EPR line
shape. With increasing temperature, the oscillations be-
come more intensive resulting in a decreasing anisotropy.
It was pointed [18] (molecular dynamics simulations of O2
rotations in gas solids) that a small cage distortions around
the impurity in the classical solids are diminished by lattice
vibrations, which may lead to a nearly spherical cage ge-
ometry. Indeed, no anisotropy was reported for H atoms in
the H2 matrix despite the hcp equilibrium structure of solid
hydrogen. Even in para-H2 solid, where the EPR linewidth
for trapped H atoms is about 0.1 G, the spectrum stays iso-
tropic. Although as-prepared vapor deposited H2 solids
contain fcc regions which convert irreversibly to hcp upon
annealing [19], it is commonly believed that the fcc struc-
ture constitutes only a minor part of the sample.
The isotropic spectra of H in H2 may arise for two rea-
sons: a smaller matrix shift of the hyperfine constant as
compared with solid Kr; and large zero-point vibrations of
the host molecules. The considerable change of the line
shape at temperatures near 32 K (Fig. 5) was accompanied
by a rapid decrease in the number of trapped H atoms. On
the other hand, it has been mentioned above that for sam-
ples grown at Tdep closer to 4.2 K, the anisotropy may per-
sist even at temperatures above 32 K. On those samples we
observed a substantially slower decrease in line intensity at
Tsample near 32 K. These results suggest that the rapid mo-
tion of diffusing H atoms averages the interaction with
their surroundings and removes the EPR line anisotropy.
It is interesting to compare the present results with
those obtained earlier [14]. In EPR studies of the thermal
mobility of atomic hydrogen in solid Kr matrix, Vaskonen
and co-authors observed no diffusion of substitutionally
trapped H atoms up to the evaporation temperature of the
matrix. Not indicated, this temperature is possibly above
30–40 K, for which we observed H-atom diffusion. The
reason for this discrepancy is that the solids obtained in
Ref. 14 cannot be considered as quench-condensed.
Indeed, the deposition temperature was 22 K and not fa-
vorable for vacancy formation needed for the substitu-
tional atoms to diffuse. The authors stressed that H atoms
generated by UV-photolysis of HBr and HCl precursor
molecules in solid Kr showed generally an octahedral
trapping rather than a substitutional one. The H-atom
Structural formation and thermal relaxation of quench-condensed Kr films
Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7 665
5.5 K 30.5 K
34.5 K
33.5 K
b
a
c
Fig. 5. The sample was prepared at Tdep of about 7 K and expe-
rienced intermediate annealing at Tsample = 25 K with the re-
sult of the broad lines going down: a — anisotropic EPR spec-
trum of H atoms in solid Kr, Tsample = 5.5 K; b — high-field
component of the doublet showing a change with time of the
isotropic line intensity; Tsample is indicated at the top; c — the
spectrum regains its anisotropy after the sample temperature
has been decreased again to Tsample = 5.5 K. The microwave
power attenuation is 50 dB (a) and 40 dB (b, c).
population in the octahedral sites of the as-deposited sam-
ples exceeded that in the substitutional sites by about two
orders of magnitude [14]. Classen et al. [20] investigated
acoustic properties of quench-condensed films of Ne and
Ar and observed structural relaxation in a rather broad
temperature region. For an Ar film condensed at 3.6, the
relaxation was found [20] at temperatures in the range
from 3.6 to 7.5 K and at 10, 12.7, and 13.9 K. These are
well below Tdes � 30 K for Ar [15]. The authors came to
the conclusion that there is a rather broad spectrum of ac-
tivation energies involved in the irreversible structural re-
arrangement. Hence, when the temperature is increased,
more and more atomic rearrangements can take place,
which requires overcoming higher energy barriers.
Using the double-paddle oscillator, White et al. [21]
measured the shear modulus of thin films of Ar and Ne
during annealing at cryogenic temperatures. For Ar and
Ne films deposited at 1.2 and 1.3 K, respectively, they ob-
served and studied a stiffening of the films at 10.407 and
20.155 K in Ar, and at 3.049 K in Ne. The desorption
temperature for Ne is 8 K [15]. Obviously, one may ex-
pect a similar behavior of quench-condensed Kr films,
i.e., structural relaxations at low temperatures in a broad
temperature range. This is partly supported by the broad
line transformation below 20 K observed in the present
study. Hence, with a plenty of vacancies in the quench-
condensed Kr, these structural rearrangements may suffi-
ciently increase the atomic hydrogen diffusion to be re-
corded in our experiment. The EPR lines for H in Kr pre-
sented in Ref. 14 are isotropic. Based on the present
findings, this result is expectable for a sample deposited
at 22 K.
Prager and Langel [6] pointed out that a local hcp
environment may be produced by stacking faults along
the (111) direction of the fcc structure of the undisturbed
matrix. One stacking fault creates two (111) planes with
hcp surrounding and two planes of fcc symmetry facing
the defect. An H-atom trapped in such a hcp surrounding
would experience an axial crystal field. However, it takes
several lattice constants for a distortion to relax. This
would have effect on the hyperfine constant of the
trapped atom and, possibly, broaden the line. Therefore, it
seems more likely that the surrounding of the axial sym-
metry is associated with atoms trapped in hcp crystallites.
4. Conclusions
EPR spectra have been obtained for H atoms trapped in
quench-condensed solid Kr. It was shown that the sub-
strate temperature during deposition had a decisive effect
on the shape of the hyperfine components. The experi-
ments were conducted at Tdep ranging from 1.5 to 11 K.
Mostly, each hyperfine component was a superposition of
broad and narrow lines. At higher temperatures of the
range, Tdep � 10 K, the broad lines were extremely weak
or even unobservable. The intensity of these lines in-
creased rapidly as the Tdep decreased from about 7 K. The
broad lines were attributed to atoms trapped in a strongly
disordered phase. As the temperature was raised, these
lines revealed the structure suggesting a relaxation of this
phase which happened at Tsample < 20 K. The narrow lines
showed an axial anisotropy of the hyperfine structure
constant. We presented experimental evidence suggesting
that these lines are most likely caused by atoms trapped in
hcp crystallites of the nonequilibrium solid Kr. The extent
of the anisotropy decreased with increasing Tdep . The ani-
sotropy was also found to depend on the sample tempera-
ture. In our experiments we obtained a rather large frac-
tion of H atoms trapped in hcp crystallites, which are
believed to be only a small fraction of the solid sample.
One of possible reasons for this findings is attributed to
the trapping process of H atoms and H2 molecules. The
incident hydrogen atoms and molecules lost only a small
portion of their kinetic energy in collisions with the sur-
face Kr atoms because of the large difference between the
masses of the colliding particles. On the other hand this
loss is much greater in H–H2 and H2–H2 collisions. This
may cause a nonuniform distribution of hydrogen over
the solid sample and appearance of regions with suffi-
ciently large hydrogen content. Since the energy differ-
ence between the hcp and fcc lattices is small, such
hydrogen-rich regions being under nonequilibrium con-
ditions might possibly bring about the hcp structure.
5. Acknowledgments
Thanks are due to M.E. Kaimakov for active participa-
tion in part of these investigations and for reading the
manuscript.
1. E.B. Gordon, L.P. Mezhov-Deglin, and O.F. Pugachev,
JETP Lett. 19, 63 (1974).
2. V. Kiruykhin, B. Keimer, R.E. Boltnev, V.V. Khmelenko,
and E.B. Gordon, Phys. Rev. Lett. 79, 1174 (1997).
3. M.A. Strzhemechny, N.N. Galtsov, and A.I. Prokhvatilov,
Fiz. Nizk. Temp. 29, 699 (2003) [Low Temp. Phys. 29, 522
(2003)].
4. M. Prager and W. Langel, J. Chem. Phys. 90, 5889 (1989).
5. L.H. Jones, S.A. Ekberg, and B.I. Swanson, J. Chem. Phys.
85, 3203 (1986).
6. M. Prager and W. Langel, J. Chem. Phys. 88, 7995 (1988).
7. M. Prager, B. Asmussen, and C.J. Carlile, J. Chem. Phys.
100, 247 (1994).
8. W. Langel, W. Schuller, E. Kn�zinger, H.-W. Fleger, and
H.J. Lauter, J. Chem. Phys. 89, 1741 (1988).
9. H. Menges and H. v. L�hneysen, J. Low Temp. Phys. 84,
237 (1991).
10. M.S. Conradi, K. Luszczynski, and R.E. Norberg, Phys.
Rev. B20, 2594 (1979).
666 Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7
Yu.A. Dmitriev
11. R.A. Zhitnikov and Yu.A. Dmitriev, Fiz. Nizk. Temp. 24,
923 (1998) [Low Temp. Phys. 24, 693 (1998)].
12. S.A. Nepijko, I. Rabin, and W. Schulze, Chem. Phys. Chem.
6, 235 (2005).
13. S.N. Foner, E.L. Cochran, V.A. Bowers, and C.K. Jen, J.
Chem. Phys. 32, 963 (1960).
14. K. Vaskonen, J. Eloranta, T. Kiljunen, and H. Kunttu, J.
Chem. Phys. 110, 2122 (1999).
15. M. Layer, A. Netsch, M. Heitz, J. Meier, and S. Hunklinger,
Phys. Rev. B73, 184116 (2006).
16. F.J. Adrian, J. Chem. Phys. 32, 972 (1960).
17. T. Kiljunen, J. Eloranta, and H. Kunttu, J. Chem. Phys.
110, 11814 (1999).
18. Z. Li and V. A. Apkarian, J. Chem. Phys. 107, 1544 (1997).
19. S. Tam and M.E. Fajardo, Fiz. Nizk. Temp. 26, 889 (2000)
[Low Temp. Phys. 26, 653 (2000)].
20. J. Classen, J. Meier, M. Heitz, and S. Hunklinger, Physica
B263–264, 163 (1999).
21. B.E White, Jr, J. Hessinger, and R.O. Pohl, J. Low. Temp.
Phys. 111, 233 (1998).
Structural formation and thermal relaxation of quench-condensed Kr films
Fizika Nizkikh Temperatur, 2007, v. 33, Nos. 6/7 667
|