Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite

The effective thermal conductivity of samples of cryocrystal nanocomposite obtained from argon and SiO2 nanopowder was determined in the temperature interval 2–35 K using the steady-state method. The thermal conductivity of crystalline argon with nanoparticles of amorphous silica oxide embedded in...

Повний опис

Збережено в:
Бібліографічні деталі
Дата:2016
Автори: Nikonkov, R.V., Stachowiak, P., Jeżowski, A., Krivchikov, A.I.
Формат: Стаття
Мова:English
Опубліковано: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2016
Назва видання:Физика низких температур
Теми:
Онлайн доступ:http://dspace.nbuv.gov.ua/handle/123456789/128508
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Цитувати:Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite / R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov // Физика низких температур. — 2016. — Т. 42, № 4. — С. 403–406 . — Бібліогр.: 31 назв. — англ.

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id irk-123456789-128508
record_format dspace
spelling irk-123456789-1285082018-01-11T03:02:39Z Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite Nikonkov, R.V. Stachowiak, P. Jeżowski, A. Krivchikov, A.I. Наноструктуры при низких температурах The effective thermal conductivity of samples of cryocrystal nanocomposite obtained from argon and SiO2 nanopowder was determined in the temperature interval 2–35 K using the steady-state method. The thermal conductivity of crystalline argon with nanoparticles of amorphous silica oxide embedded in its structure shows a weak dependence on particle linear dimension in the interval 5–42 nm. The temperature dependence of the thermal conductivity of the nanocomposites can be well approximated by taking into account only the two mechanisms of heat carrier scattering: phonon-phonon interaction in U-processes and scattering of phonons by dislocations. 2016 Article Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite / R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov // Физика низких температур. — 2016. — Т. 42, № 4. — С. 403–406 . — Бібліогр.: 31 назв. — англ. 0132-6414 PACS: 65.60 +a, 66.70.–f, 63.20.–e http://dspace.nbuv.gov.ua/handle/123456789/128508 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 Наноструктуры при низких температурах
Наноструктуры при низких температурах
spellingShingle Наноструктуры при низких температурах
Наноструктуры при низких температурах
Nikonkov, R.V.
Stachowiak, P.
Jeżowski, A.
Krivchikov, A.I.
Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
Физика низких температур
description The effective thermal conductivity of samples of cryocrystal nanocomposite obtained from argon and SiO2 nanopowder was determined in the temperature interval 2–35 K using the steady-state method. The thermal conductivity of crystalline argon with nanoparticles of amorphous silica oxide embedded in its structure shows a weak dependence on particle linear dimension in the interval 5–42 nm. The temperature dependence of the thermal conductivity of the nanocomposites can be well approximated by taking into account only the two mechanisms of heat carrier scattering: phonon-phonon interaction in U-processes and scattering of phonons by dislocations.
format Article
author Nikonkov, R.V.
Stachowiak, P.
Jeżowski, A.
Krivchikov, A.I.
author_facet Nikonkov, R.V.
Stachowiak, P.
Jeżowski, A.
Krivchikov, A.I.
author_sort Nikonkov, R.V.
title Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
title_short Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
title_full Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
title_fullStr Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
title_full_unstemmed Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite
title_sort thermal conductivity of argon–sio₂ cryocrystal nanocomposite
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
publishDate 2016
topic_facet Наноструктуры при низких температурах
url http://dspace.nbuv.gov.ua/handle/123456789/128508
citation_txt Thermal conductivity of argon–SiO₂ cryocrystal nanocomposite / R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov // Физика низких температур. — 2016. — Т. 42, № 4. — С. 403–406 . — Бібліогр.: 31 назв. — англ.
series Физика низких температур
work_keys_str_mv AT nikonkovrv thermalconductivityofargonsio2cryocrystalnanocomposite
AT stachowiakp thermalconductivityofargonsio2cryocrystalnanocomposite
AT jezowskia thermalconductivityofargonsio2cryocrystalnanocomposite
AT krivchikovai thermalconductivityofargonsio2cryocrystalnanocomposite
first_indexed 2025-07-09T09:13:35Z
last_indexed 2025-07-09T09:13:35Z
_version_ 1837160115855163392
fulltext Low Temperature Physics/Fizika Nizkikh Temperatur, 2016, v. 42, No. 4, pp. 403–406 Thermal conductivity of argon–SiO2 cryocrystal nanocomposite R.V. Nikonkov, P. Stachowiak, and A. Jeżowski Institute for Low Temperature and Structure Research, Polish Academy of Sciences, PN 1410, 50-950 Wroclaw, Poland E-mail: P.Stachowiak@int.pan.wroc.pl A.I. Krivchikov B. Verkin Institute for Low Temperature Physics and Engineering of NAS Ukraine 47 Prospekt Nauky, Kharkiv 61103, Ukraine Received December 19, 2015, published online February 24, 2016 The effective thermal conductivity of samples of cryocrystal nanocomposite obtained from argon and SiO2 nanopowder was determined in the temperature interval 2–35 K using the steady-state method. The thermal conduc- tivity of crystalline argon with nanoparticles of amorphous silica oxide embedded in its structure shows a weak de- pendence on particle linear dimension in the interval 5–42 nm. The temperature dependence of the thermal conduc- tivity of the nanocomposites can be well approximated by taking into account only the two mechanisms of heat carrier scattering: phonon-phonon interaction in U-processes and scattering of phonons by dislocations. PACS: 65.60 +a Thermal properties of amorphous solids and glasses: heat capacity, thermal expansion, etc.; 66.70.–f Nonelectronic thermal conduction and heat-pulse propagation in solids; thermal waves; 63.20.–e Phonons in crystal lattices. Keywords: thermal conductivity, cryocrystals, nanocomposites, phonon-phonon processes, dislocation scattering. Introduction Not very long ago a new chapter in technology has been crack opened. It seems that the bulk materials using nano- and microparticles due to their particular thermal proper- ties are going to play an important role in many future ap- plications [1–4]. Along with the advent of the new materi- als, new basic problems and questions regarding the thermal transport mechanisms in bulk solids containing numerous nanoparticles in their crystalline matrices have emerged. Many thermal conduction mechanisms of the new materials were negligible or just absent from the clas- sical ones. Therefore, the analysis of phonon transfer in crystals containing nanoparticles requires a new approach. For example, when we consider a phonon of a wave vector q, which encounters on its pathway a nanoparticle of linear dimensions R, it gets scattered. The result of the scattering depends on the so called size parameter ,χ which is de- fined as Rχ = q . When the size parameter is small (χ 1)<< , the scattering rule obeys Rayleigh law, i.e., the scattering probability varies as frequency to the fourth power. At the other limit, where the size parameter reaches big numbers ( 1)χ >> the scattering probability is inde- pendent of frequency of the phonon and the phonon scat- tering cross section depends on the path length through which the phonon travels across the nanoparticle and the associated phase lag [5]. Thus, the size of the nanoparticle becomes a very important factor for scattering cross sec- tion in this regime. In the Rayleigh scattering regime, the scattering cross section depends both on the difference of masses of the nanoparticle and the host crystal and on the difference of the force constancies acting in the two con- stituencies [5]. In the case of polydispersion of nanoparti- cles, when the size of nanoparticles deviates from its mean value, the scattering cross section based on mean diameter increases with increasing standard deviation of the linear dimension of the nanoparticles [5]. The considerations sketched above do not exhaust the complexity of problems related to the thermal energy transfer in the discussed nanostructured objects. They can be helpful for the analy- sis of thermal transport processes in case of the nanoparti- cles being far from each other. The term “far from each other” is considered here with respect to the wavelength of the phonons. Therefore, if the ratio of distance between the nanoparticles and the wavelength is much large than 1, © R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov, 2016 R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov then independent phonon scattering (the scattering by one nanoparticle without reference to the others) is a good ap- proach. However, if the ratio σ of the interparticle distance d and the phonon wavelength 2 /π q is close to 1, for cor- rect description of the thermal processes multiple and de- pendent phonon scatterings have to be taken into account. Here the multiple phonon scattering means that a scattered wave from one nanoparticle is incident on another particle to be scattered again while the dependent scattering is far- field interference of waves scattered by the different nano- particles due to phase difference, which is ignored in the independent scattering regime. When 1σ ≈ , the multiple elastic scattering due to nanoparticles modifies the original crystalline matrix phonons velocity, density of states and their equilibrium intensity while the dependent scattering results in change of original cross section for phonon scat- tering by the nanoparticles. In this indirect way the multi- ple and dependent processes influence the crystal thermal conductivity [6]. In a real crystal — with nanoparticles embedded in its structure — the abovementioned condition 1σ ≈ may be fulfilled when the volume density of the na- noparticles is either high enough or at low temperatures. As for the former, for a given volume fraction, the interparticle distance is smaller for smaller linear dimen- sions of the nanoparticles. The effect of nanoparticles on the thermal conductivity of nanocomposites was quantitatively confirmed by nu- merous experiments [7–10]. Here we report our prelimi- nary results of investigations of thermal conductivity of argon crystals containing in their volume silica nanoparti- cles of different linear dimensions. We have chosen the crystal because of its simplicity, both in terms of the crys- tallographic structure and interatomic interactions. As with other rare gas solids, it is bound by van der Waals forces and has an fcc structure in the whole range of temperatures and pressures. The argon crystal is also the best experimentally known noble gas one. The first measurements of its thermal con- ductivity were made in the 1950’s by White and Woods [11]. Up to now, a number of measurements have been made by other experimentators, both at constant volume and at equilibrium vapor pressure, in various temperature intervals, see e.g. [12–14]. Experiment In our experiment the thermal conductivity of cryocrystal nanocomposites was determined by steady-state heat flow method in the temperature range from 2.2 to 35 K. The meth- odology of the measurement and its technical aspect have been described in details in our previous paper [15]. The thermal conductivity cell was made from glass tube of an inner diameter of 6 mm, a wall thickness of 1 mm and a length of 50 mm. To the ends of the tube two caps made of copper were fixed with epoxy. The cell was equipped with two germanium resistance thermometers spaced 10 mm from each other. The lower one was mount- ed 10 mm above the bottom of the ampoule. To the top cap an electric gradient heater was attached. Through the cap a thin-wall stainless steel capillary ran. The capillary allowed to pump out the cell or fill it with argon gas and thermal exchange gas of helium. During the experiment the bottom cap rested in a copper base of controlled temperature. For obtaining the nanocomposite samples, argon gas of 99.999% purity produced by Messer Group GmbH and amorphous silica oxide nanoparticles of linear dimension of about 5 nm, 15 nm and 42 nm were used. The SiO2 5 and 42 nm nanoparticle powders were synthesized in the Insti- tute for Low Temperatures and Structure Research PAS, Wrocław, Poland by combustion method while the one of 15 nm was obtained from Sigma-Aldrich Ltd. The 15 nm nanoparticles showed higher dispersion than those ob- tained in the ILTSR PAS. The size of the particles was determined by Brunauer–Emmett–Teller method. The sur- face area of the powders were 533.6, 143.2 and 64.1 m2/g for 5, 15 and 42 nm nanoparticles, respectively. The powder of an amount of 0.15 g was placed inside the cell after the bottom cap was fixed to the glass tube. Then the upper cap was also glued and the assembled cell was installed in the cryostat. The nanopowder volume den- sity in the cell was 7%. The cell prepared in the way de- scribed above was placed in the measuring chamber of the cryostat. At the beginning of the experiment the tempera- ture of the cell was lowered to a little bit above the triple point temperature of argon and argon gas was let to the cell, whereupon the condensation to its liquid phase began. During the condensation the temperature of the upper part of the cell was kept a few Kelvins higher than the tempera- ture of the bottom so that the liquid gradually filled the cell from its bottom to the top. In this way we obtained an ar- gon–SiO2 liquid nanocomposite. Finally, the temperature of the bottom of the ampoule was slowly lowered — the liquid solidified forming an argon–SiO2 cryocrystal nanocomposite. Cooling rate during the crystal growth was 3 K/h. After crystallization the nanocomposite was cooled down to the temperature of the thermal conductivity meas- urement at the cooling rate of 6 K/h. In the process of determination of the thermal conduc- tivity two distorting factors were taken into account: 1. The parasitic temperature gradient being a result of the heat radiation due to the temperature mismatch of the LHe thermal shield of the measuring cell and the sample and 2. The heat transported by the cell glass wall. To determine the first one, the measurements of the parasitic temperature gradient was carried out at various temperatures for each of the samples. As for the second one, the measurement of the dependence of the thermal conductivity coefficient of emp- ty cell was performed in a separate experiment. The random error of the thermal conductivity measure- ment in low temperatures did not exceed 1.5%, whereas 404 Low Temperature Physics/Fizika Nizkikh Temperatur, 2016, v. 42, No. 4 Thermal conductivity of argon–SiO2 cryocrystal nanocomposite above 20 K it increased to 3%, mostly due to effects con- nected with spurious heat leaks. The systematic error did not exceed 3%. Results and discussion Thermal conductivity ( )Tκ of the investigated nanocomposites and pure solid argon [16] are both pre- sented in Fig. 1. For all argon–SiO2 cryocrystal nanocom- posites their thermal conductivity is lower than that of the bulk polycrystal argon [16–20]. The dependence ( )Tκ shows a behavior typical for polycrystaline atomic crystals. With decreasing temperature the thermal conductivity ini- tially increases according to the relationship ( ) TTκ ∝ then it attains a maximum and finally decreases as T2. The be- havior of 2( ) TTκ ∝ is specified by dominant scattering of phonons by static strain fields surrounding dislocations [19]. The thermal conductivity of Ar–15 nm SiO2 nano- composite shows the smallest thermal conductivity. In the temperatures below 6 K the dependence can be approxi- mated by ( ) nTTκ ∝ , where n = 1.55. Here it should be not- ed that the exponent n < 2 is, in some cases, the manifesta- tion of a complex fractal structure of a substance [21–23]. For the preliminary analysis of the temperature depend- ence of thermal conductivity coefficient of the investigated samples we used a simple thermal conductivity model. Assuming that the heat is conducted along parallel paths, the thermal conductivity coefficient of components of a medium is additive. It follows that the thermal conductivity of two-component medium consisting of polycrystalline argon and nano-SiO2 material can be approximated as a weighted sum of the thermal conductivities of solid argon Ar ( )Tκ and the nanoparticles of SiO2 2SiO ( )Tκ according to the following expression [24]: 2Ar SiO( ) ( ) ( )T T Tκ = κ + κ . (1) The value of thermal conductivity of the argon–SiO2 nanocomposite consists of the thermal conductivity Ar ( )Tκ of solid argon which is a medium for SiO2 parti- cles. The thermal conductivity Ar ( )Tκ is determined by phonon transport of the heat [25–29]. The contribution of thermal conductivity coefficient of SiO2 nanoparticle, 2SiO ( )Tκ , to the sum ( )Tκ is small — in Fig. 1 such low- thermal conductivity 2SiO ( )Tκ obtained for the SiO2 nanopowder of 15 nm alone was depicted. The thermal con- ductivity 2SiO ( )Tκ shows a dependence which is almost not influenced by the size of SiO2 nanoparticles in the interval from 5 up to 42 nm. In Fig. 1 the dependence 2SiO ( )Tκ for 15 nm nanopowder is shown. The data obtained for the nano- particles of linear dimensions of 5 and 42 nm agree with the shown ones within the experiment accuracy. Since the contribution of nanoparticles to the resultant thermal conductivity of the investigated nanocomposite is negligible, in further analysis we focus exclusively on ar- gon matrix phonons. Let us take into account only two mechanisms of heat carrier scattering: phonon-phonon interactions in U-processes and scattering of phonons by dislocations. The influence of grain boundary scattering is small in the thermal conductivity of the current sam- ples [25]. Therefore, the dependence Ar ( )Tκ can be writ- ten as a reciprocal of the sum of thermal resistance due to phonon dislocation scattering and phonon-phonon scatter- ing in U-processes with the following expression Ar dis ph1/ ( ) 1/ ( ) 1/ ( )T T Tκ = κ + κ , (2) where 1 / ph ( ) eE Tk T AT −= describes the three-phonon processes and 2 dis )(T BT=κ is the contribution of disloca- tion scattering. We fitted equation (2) to all the data ob- tained in the described here experiment. The solid lines shown in Fig.1 are results of the fitting. As one can see from the figure, the fittings describe the experimental data well, both for pure argon and argon–SiO2 nanocomposites. The best fit parameters A, B and E are given in Table 1. Table 1. Value of the parameters A, B and Е of equation (2), for which the experimentally obtained dependence of the thermal conductivity of the investigated nanocomposites is best approxi- mated. Sample Three-phonon U-processes Dislocation scattering A, Wm–1 E, K В, Wm–1⋅K–3 Pure Ar 21 11.5 0.25 Ar–5 nm SiO2 21 8 0.07 Ar–15 nm SiO2 21 5 0.036 Ar–42 nm SiO2 20 7 0.06 Fig. 1. Thermal conductivity of: pure solid argon () [16], SiO2 nanoparticles of 15 nm linear dimension (), argon with SiO2 nanoparticles of linear dimentions 5 nm ( ) and 15 nm (). For the clarity of the picture, the data obtained for solid argon–SiO2 nanoparticles of linear dimension of 42 nm were not shown. Three solid lines are model approximation of temperature de- pendence of the thermal conductivity of the solid argon and the investigated nanocomposite samples. Low Temperature Physics/Fizika Nizkikh Temperatur, 2016, v. 42, No. 4 405 R.V. Nikonkov, P. Stachowiak, A. Jeżowski, and A.I. Krivchikov In the framework of the model described above, the re- duction of the thermal conductivity of the investigated nanocomposite is a result of an increase, in comparison to the pure argon polycrystal, of intensity of phonon scatter- ing by dislocations. This may be an effect of an increase of the dislocation density in the nanocomposite, which in turn may be a combined result of heterogeneous growth of pol- ycrystalline argon with the nanoparticles being the crystal- lization centers and the cooling process of the composite in pre-melting temperature region, wherein solid argon shows a large value of the coefficient of linear expansion [30,31] A deviation from the dependence 2( )T Tκ ∝ for the sample of argon–15 nm SiO2 at temperatures T < 6 K should be noted. The deviation may be a result of contribution of the thermal conductivity of nanopowder to the total ther- mal conductivity. It can be observed due to low thermal conductivity of the investigated sample. Summary The dependence of thermal conductivity coefficient on temperature of Ar–SiO2 nanocomposites was experimen- tally investigated in the temperature range 2–35 K by steady-state heat flow method. The investigated samples consisted of solid argon with SiO2 amorphous nanoparti- cles embedded in the argon crystalline matrix. The fraction of silica was 7% of the volume of the investigated samples. It was found that the thermal conductivity of the Ar–SiO2 nanocomposites can be described by taking into account merely two argon crystal phonon scattering mechanism: phonon-phonon scattering in U-processes and scattering of phonons by the crystal dislocations. The authors want to thank Dr. Robert Pązik for provid- ing SiO2 nanopowders. This work was supported by the National Science Cen- tre (Poland) grant nr. UMO-2013/08/M/ST3/00934. 1. M.T. Hung, C.C. Wang, J.C. Hsu, J.Y. Chiou, S.W. Lee, T.M. Hsu, and P.W. Li, Appl. Phys. Lett. 101, 251913 (2012). 2. Y. Ma, R. Heijl, and A.E.C. Palmqvist, J. Mater. Sci. 48, 2767 (2013). 3. Y. Ezzahri and K. Joulain, J. Appl. Phys. 113, 043510 (2013). 4. M.L. Lee and R. Venkatasubramanian, Appl. Phys. Lett. 92, 053112 (2008). 5. W. Kim and A. Majumdar, J. Appl. Phys. 99, 084306 (2006). 6. R. Prasher, J. Heat Transfer 128, 627 (2006). 7. N. Mingo, D. Hauser, N.P. Kobayashi, M. Plissonnier, and A. Shakouri, Nano Lett. 9, 711 (2009). 8. W. Kim, S.L. Singer, A. Majumdar, J.M.O. Zide, D. Klenov, A.C. Gossard, and S. Stemmer, Nano Lett. 8, 2097 (2008). 9. W. Kim, J. Zide, A. Gossard, D. Klenov, S. Stemmer, A. Shakouri, and A. Majumdar, Phys. Rev. Lett. 96, 045901 (2006). 10. W. Kim, S.L. Singer, and A. Majumdar, J. Phys. Conf. Ser. 92, 012085 (2007). 11. G.K. White and S.B. Woods, Philos. Mag. 3, 785 (1958). 12. A. Berne, G. Boato, and M. De Paz, Cimento B Series 10, 182 (1966). 13. F. Clayton and D.N. Batchelder, J. Phys. C 6, 1213 (1973). 14. D.K. Christen and G.L. Pollack, Phys. Rev. B 12, 3380 (1975). 15. R.V. Nikonkov, P. Stachowiak, T.V. Romanova, A. Jeżowski, and V.V. Sumarokov, Fiz. Nizk. Temp. 41, 625 (2015) [Low Temp. Phys. 41, 492 (2015)]. 16. P. Stachowiak, V.V. Sumarokov, J. Mucha, and A. Jeżowski, Phys. Rev. B 58, 2380 (1998). 17. I.N. Krupskii and V.G. Manzhelii, Sov. Phys. JETP 2, 1097 (1969). 18. L. Finegold and N.E. Phillips, Phys. Rev. B 177, 1383 (1969). 19. F. Clayton and D.N. Batchelder, J. Phys. C: Solid State Phys. 6, 1213 (1973). 20. D.K. Christen and G.L. Pollack, Phys. Rev. B 12, 3380 (1975). 21. D. Szewczyk, A. Jezowski, G.A. Vdovichenko, A.I. Krivchikov, F.J. Bermejo, J.L. Tamarit, andJ.W. Taylor, J. Phys. Chem. B 119, 8468 (2015). 22. M. Hassaine, R.J. Jiménez-Riobóo, I.V. Sharapova, O.A. Korolyuk, A.I. Krivchikov, and M.A. Ramos, J. Chem. Phys. 131, 174508 (2009). 23. A.I. Krivchikov, O.A. Korolyuk, and I.V. Sharapova, Fiz. Nizk. Temp. 38, 95 (2012) [Low Temp. Phys. 38, 74 (2012)]. 24. A.I. Krivchikov, B.Y. Gorodilov, O.A. Korolyuk, V.G. Manzhelii, O.O. Romantsova, H. Conrad, and D.D. Klug, Phys. Rev. B 73, 064203 (2006). 25. S. Ju and X. Liang, J. Appl. Phys. 108, 104307 (2010). 26. J.E. Turney, E.S. Landry, A.J.H. McGaughey, and C.H. Amon, Phys. Rev. B 79, 064301 (2009). 27. H. Kaburaki, J. Li, S. Yip, and H. Kimizuka, J. Appl. Phys. 102, 043514 (2007). 28. Z. Zhong and X. Wang, J. Appl. Phys. 100, 044310 (2006); P. Heino, Phys. Rev. B 71, 144302 (2005). 29. K.V. Tretiakov and S. Scandolo, J. Chem. Phys. 121, 11177 (2004). 30. O.G. Peterson, D.N. Batchelder, and R.O. Simmons, Phys. Rev. B 150, 703 (1966). 31. C.R. Tilford and C.A. Swenson, Phys. Rev. B 5, 719 (1972). 406 Low Temperature Physics/Fizika Nizkikh Temperatur, 2016, v. 42, No. 4 Introduction Experiment Results and discussion Summary