The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures

Orientational glasses with CO molecules occupying 26 and 90% of the octahedral interstitial sites in the C₆₀ lattice have been investigated by the dilatometric method in a temperature interval of 2.5–22 K. At temperatures 4–6 K the glasses undergo a first-order phase transition which is evident fr...

Повний опис

Збережено в:
Бібліографічні деталі
Опубліковано в: :Физика низких температур
Дата:2008
Автори: Dolbin, A.V., Esel’son, V.B., Gavrilko, V.G., Manzhelii, V.G., Vinnikov, N.A., Gadd, G.E., Moricca, S., Cassidy, D., Sundqvist, B.
Формат: Стаття
Мова:English
Опубліковано: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2008
Теми:
Онлайн доступ:https://nasplib.isofts.kiev.ua/handle/123456789/116979
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:The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures / A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Manzhelii, N.A. Vinnikov, G.E. Gadd, S. Moricca, D. Cassidy, B. Sundqvist // Физика низких температур. — 2008. — Т. 34, № 6. — С. 592–598. — Бібліогр.: 23 назв. — англ.

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-116979
record_format dspace
spelling Dolbin, A.V.
Esel’son, V.B.
Gavrilko, V.G.
Manzhelii, V.G.
Vinnikov, N.A.
Gadd, G.E.
Moricca, S.
Cassidy, D.
Sundqvist, B.
2017-05-18T17:18:35Z
2017-05-18T17:18:35Z
2008
The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures / A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Manzhelii, N.A. Vinnikov, G.E. Gadd, S. Moricca, D. Cassidy, B. Sundqvist // Физика низких температур. — 2008. — Т. 34, № 6. — С. 592–598. — Бібліогр.: 23 назв. — англ.
0132-6414
PACS: 74.70.Wz
https://nasplib.isofts.kiev.ua/handle/123456789/116979
Orientational glasses with CO molecules occupying 26 and 90% of the octahedral interstitial sites in the C₆₀ lattice have been investigated by the dilatometric method in a temperature interval of 2.5–22 K. At temperatures 4–6 K the glasses undergo a first-order phase transition which is evident from the hysteresis of the thermal expansion and the maxima in the temperature dependences of the linear thermal expansion coefficients α (T), and the thermalization times τ₁(T) of the samples. The effect of the noncentral CO–C₆₀ interaction upon the thermal expansion and the phase transition in these glasses was clarified by comparing the behavior of the properties of the CO–C₆₀ and N₂–C₆₀ solutions.
The authors thank Prof. A.S. Bakai for helpful discussions. The authors are also indebted to the Science and Technology Center of Ukraine (STCU) for the financial support of this study (Project Uz-116).
en
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
Физика низких температур
Низкоразмерные и неупорядоченные системы
The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
spellingShingle The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
Dolbin, A.V.
Esel’son, V.B.
Gavrilko, V.G.
Manzhelii, V.G.
Vinnikov, N.A.
Gadd, G.E.
Moricca, S.
Cassidy, D.
Sundqvist, B.
Низкоразмерные и неупорядоченные системы
title_short The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
title_full The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
title_fullStr The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
title_full_unstemmed The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures
title_sort effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid co–c₆₀ solutions at low temperatures
author Dolbin, A.V.
Esel’son, V.B.
Gavrilko, V.G.
Manzhelii, V.G.
Vinnikov, N.A.
Gadd, G.E.
Moricca, S.
Cassidy, D.
Sundqvist, B.
author_facet Dolbin, A.V.
Esel’son, V.B.
Gavrilko, V.G.
Manzhelii, V.G.
Vinnikov, N.A.
Gadd, G.E.
Moricca, S.
Cassidy, D.
Sundqvist, B.
topic Низкоразмерные и неупорядоченные системы
topic_facet Низкоразмерные и неупорядоченные системы
publishDate 2008
language English
container_title Физика низких температур
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
format Article
description Orientational glasses with CO molecules occupying 26 and 90% of the octahedral interstitial sites in the C₆₀ lattice have been investigated by the dilatometric method in a temperature interval of 2.5–22 K. At temperatures 4–6 K the glasses undergo a first-order phase transition which is evident from the hysteresis of the thermal expansion and the maxima in the temperature dependences of the linear thermal expansion coefficients α (T), and the thermalization times τ₁(T) of the samples. The effect of the noncentral CO–C₆₀ interaction upon the thermal expansion and the phase transition in these glasses was clarified by comparing the behavior of the properties of the CO–C₆₀ and N₂–C₆₀ solutions.
issn 0132-6414
url https://nasplib.isofts.kiev.ua/handle/123456789/116979
citation_txt The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C₆₀ solutions at low temperatures / A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Manzhelii, N.A. Vinnikov, G.E. Gadd, S. Moricca, D. Cassidy, B. Sundqvist // Физика низких температур. — 2008. — Т. 34, № 6. — С. 592–598. — Бібліогр.: 23 назв. — англ.
work_keys_str_mv AT dolbinav theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT eselsonvb theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT gavrilkovg theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT manzheliivg theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT vinnikovna theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT gaddge theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT moriccas theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT cassidyd theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT sundqvistb theeffectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT dolbinav effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT eselsonvb effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT gavrilkovg effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT manzheliivg effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT vinnikovna effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT gaddge effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT moriccas effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT cassidyd effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
AT sundqvistb effectofthenoncentralimpuritymatrixinteractionuponthethermalexpansionandpolyamorphismofsolidcoc60solutionsatlowtemperatures
first_indexed 2025-11-27T04:18:07Z
last_indexed 2025-11-27T04:18:07Z
_version_ 1850796099083698176
fulltext Fizika Nizkikh Temperatur, 2008, v. 34, No. 6, p. 592–598 The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C60 solutions at low temperatures A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Manzhelii, and N.A. Vinnikov B. Verkin Institute for Low Temperature Physics and Engineering of the National Academy of Sciences of Ukraine 47 Lenin Ave., Kharkov 61103, Ukraine E-mail: dolbin@ilt.kharkov.ua G.E. Gadd, S. Moricca, and D. Cassidy Australian Nuclear Science and Technology Organization, NSW 2234, Australia B. Sundqvist Department of Physics, Umea University, SE - 901 87 Umea, Sweden Received January 4, 2008 Orientational glasses with CO molecules occupying 26 and 90% of the octahedral interstitial sites in the C60 lattice have been investigated by the dilatometric method in a temperature interval of 2.5–22 K. At tem- peratures 4–6 K the glasses undergo a first-order phase transition which is evident from the hysteresis of the thermal expansion and the maxima in the temperature dependences of the linear thermal expansion coeffi- cients �(T), and the thermalization times �1(T) of the samples. The effect of the noncentral CO–C60 interac- tion upon the thermal expansion and the phase transition in these glasses was clarified by comparing the be- havior of the properties of the CO–C60 and N2–C60 solutions. PACS: 74.70.Wz Fullerenes and related materials. Keywords: fullerite C60, thermal expansion, polyamorphism, ñarbon oxide (CO). Introduction Below T = 90 K fullerite C60 transforms into an orientational glass. According to dilatometric and x-ray structural data [1–7], the gases dissolved in C60 produce a significant effect on the thermal expansion of the glass and cause a first-order phase transition (polyamorphism) in it. It is interesting to find out how particular molecular parameters of the admixture gas can influence the proper- ties of C60 lattice as a result of a impurity–matrix interac- tion. To judge accurately the effect of varying a certain molecular parameter, the gas impurities should be chosen so that they differ mainly in this particular parameter, whilst other molecular parameters that may have an effect on the C60 lattice are essentially kept the same. For exam- ple, we would like to probe the effect on the impurity–ma- trix interaction of altering the electronic charge distribu- tion within a diatomic gas. The choice of a homo and hetero diatomic gas with similar molecular bond lengths would be a good starting point for investigating this im- portant question. We have conveniently chosen CO–C60 and N2–C60 solutions. In contrast to O2, CO and N2 mole- cules do not react chemically with C60 at the temperatures to which C60 has to be heated to desorb volatile impuri- ties. These molecules also have practically identical mo- lecular weights (M(CO) = 28.0105, M(N2) = 28.0134) as well as comparable gas-kinetic diameters (�(CO) = = 3.766 �, �(N2) = 3.756 �) [8], but they differ signifi- cantly in electric quadrupole moments Q (Q(CO) = = –2.839·10–26 esu, Q(N2) = – 1.394·10–26 esu) [9]. N2 also does not have a dipole moment whereas CO does. However, as will be discussed further on, it is the quad- rupole moment and not the dipole moment, that contri- butes most to the effect that these impurities have upon © A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Manzhelii, N.A.Vinnikov, G.E. Gadd, S. Moricca, D. Cassidy, and B. Sundqvist, 2008 the low-temperature thermal expansion and polyamor- phism of C60. The dilatometric data on orientational C60 glasses with molar N2 concentrations (N2–to–C60 molecule ratio) of 9.9 and 100% has previously been published in Ref. 3, so that this paper extends the studies to include those from CO–C60 solutions, followed by comparison of the two data sets. In this study, we investigate the impurity effect of CO on the properties and phase transformations of orien- tational C60 glasses. Solutions of CO–C60 with both 26 and 90 mol. % CO, were investigated. The impurity (N2, CO) molecules occupy the octahe- dral interstitial cavities in the C60 lattice, of which there is effectively one octahedral cavity per C60 molecule. As a result of this, the molar CO and N2 concentrations are equal to the N2 and CO occupancies of the octahedral sites in the C60 lattice. Samples and measuring technique The C60 sample with 26 mol. % CO was prepared as follows. Prior to saturation with CO, the sample, which was a pressed cylinder of solid C60 powder, 9 mm high and 10 mm in diameter (prepared by a procedure as de- scribed in Ref. 2), was kept for 72 hours under dynamic evacuation to remove gas impurities (P = 1·10–3mm Hg, T = 400 °C). The outgassed sample and cell, was filled with CO gas at room temperature to a pressure of 760 mm Hg and sealed. The sample was kept under these sealed conditions for 105 days. The thermal expansion of the CO–C60 solutions was investigated using a low-temperature capacitance dilato- meter. Its design and the measurement technique are de- tailed in Ref. 14. Immediately before the dilatometric measurement, the measuring cell with the CO–C60 sample and which was filled with CO, was cooled slowly to 65 K, which is just below the freezing point of CO at 68 K. The cell was eva- cuated at this temperature to remove the condensed CO, that was CO that had not been absorbed by the sample. The sample was pumped on further until a base pressure of 1·10–5 mm Hg was attained, followed by cooling of the sample to the base temperature of 4.2 K. The thermal ex- pansion of the CO–C60 sample was measured after a four- hour exposure to this temperature. After measuring the thermal expansion of the sample, the amount of gas impurities and their compositions were determined qualitatively and quantitatively using a vac- uum desorption gas analyzer [12]. It was found that about 26% of the octahedral cavities of the C60 lattice were oc- cupied by CO. Most of the CO was desorbed on heating the sample to 300 °C (Fig. 1). The preparation and analy- sis techniques for the C60 sample with 90 mol. % CO are described in Ref. 13. Results and discussion The temperature dependences of the linear thermal ex- pansion coefficient (LTEC) �(T) of pure C60 and of the C60 samples with different contents of the CO impurity are shown in Fig. 2. The �(T) values are averaged over several measurement series. Owing to the cubic symme- try of their lattices, the thermal expansion of the samples can be described with a single LTEC. The thermal expansion of the investigated samples ex- hibited a number of specific features. On heating (curves 1, 2) and subsequent cooling (curves 3, 4) the thermal ex- pansion coefficient has a hysteresis which points to a first-order phase transition in the orientational CO–C60 glasses. No hysteresis was observed for pure C60 (curve 5). From the two different CO–C60 samples, it appears that the onset of the hysteresis shifts towards higher tempera- ture with increased impurity concentration, shifting from 3 K for the 26 mol. % CO to 4 K for the 90 mol. % CO. Within the temperature range starting from the lowest measured temperature of 2.5 K to the respective hyste- resis onset temperatures, it is found that the �(T) for a particular sample are practically identical for both the heating and cooling curves. Moreover, the LTECs of both the CO–C60 samples (26 and 90 mol. % CO) and that of pure C60 also coincide within the measurement error. On heating in the interval 4–6 K there is a region of insta- bility with higher experimental errors and local LTEC maxima (Fig. 2), however, the errors are appreciably lower than the maxima heights observed. It is assumed [3–5] that within this interval of temperatures, occurs the first-order phase transition between the two differing im- purity doped orientational glasses. Previous investigations [1–5] show that the thermal expansion of gas-doped C60 contains positive and nega- tive components with different characteristic relaxation times (�1 and �2, respectively). With a temperature The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C60 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 593 0 2 4 6 8 10 12 20 200 300 450 T, C° CO2 CO 0.4 n , % 3.4 10.7 4.7 7.3 Fig. 1. The composition of the gas mixture (in percentage of occupation of the octahedral cavities) desorbed from the C60 sample with 26 mol. % CO on stepwise heating the sample to T = 450 °C. change of the sample, the positive component is attrib- uted [1–5] to the process of temperature equalization over the bulk sample (thermalization), whilst the negative component accounts orientational changes of the C60 molecules induced by the temperature change of the sam- ple. Since a C60 crystal is perceived as consisting of do- mains of different orientational orders of the C60 mole- cules, and with these domains in a particular crystal separated by interlayers of C60, it has been concluded the- oretically [15–17] that the negative component of the thermal expansion observed at these low temperatures studied, results from the C60 reorientation within these actual interlayers and not in the domains themselves. The thermal expansion of C60 samples doped with CO also has two components. They were separated in a simi- lar fashion to the techniques described in Ref. 1. The tem- perature dependences of the positive and negative compo- nents for samples with different CO concentration are illustrated in Fig. 3. It is seen from Fig. 2,b that all the LTEC curves are lower than the LTEC of pure C60 over this temperature range, with the heating and cooling curves of the 90 mol. % 594 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 A.V. Dolbin et al. 2 4 6 8 10 12 14 16 18 20 22 –0.2 0 0.2 0.4 0.6 0.8 1.0 T, K 1 2 3 4 5 2 3 4 5 6 7 8 –0.1 0 0.1 0.2 T, K 1 2 3 4 5 � , 1 0 – 5 – 1 K � , 1 0 – 5 – 1 K a b Fig. 3. The temperature dependences of the positive and nega- tive components of the thermal expansion coefficient of CO–C60 solutions as studied in the intervals of 2.5–22 (a) and 2.5–8 (b) K. The positive contributions are the labelled curves 1 — 26 mol. % CO and 2 — 90 mol. % CO, whilst the negative contributions are labelled as the curves 3 — 26 mol. % CO and 4 — 90 mol. % CO. Pure C60 which only exhibits a positive contribution is shown as a dotted line (5). 4 6 8 10 12 14 16 18 20 220 0.2 0.4 0.6 0.8 1.0 � , 1 0 – 5 – 1 K � , 1 0 – 5 – 1 K T, K 1 3 2 4 5 2 3 4 5 6 7 8 0 0.1 0.2 T, K 1 3 2 4 5 a b Fig. 2. The temperature dependence of the linear thermal ex- pansion coefficient of CO–C60 solutions in temperature inter- vals 2.5–22 (a) and 2.5–8 (b) K. The curves 1 and 2 are from heating the 26 and 90 mol. % CO–C60 samples, respectively, whilst curves 3 and 4 are from cooling the 26 and 90 mol. % CO–C60 samples, respectively. The dotted line (5) is from pure C60 by either heating or cooling the sample. CO and the heating curve of the 26 mol. % CO, being markedly lowered. From Fig. 2,b we can conclude that this lowering scales with increases in the concentration of the CO impurity. Above 8 K, although the heating and cooling LTEC curves for the 90 mol. % CO and the heat- ing curve of the 26 mol. % CO appear lowered even fur- ther than the corresponding LTEC curve of pure C60, the cooling curve of the 26 mol. % CO is more or less identi- cal with that of pure C60. However if we consider just the positive component to this curve, as shown in Fig. 3, it is seen to also be lower than that of pure C60 (which only ex- hibits a positive component). In N2–C60 solutions this lowering effect of the LTEC curves, exists only at high N2 concentrations and is much less [3]. This lowering effect is explained as follows. Over the temperature range spanned in our experiments (2.5–22 K), the thermal ex- pansion of pure C60 is determined by the changes with temperature that occur in a range of phenomena, which predominantly include the translational lattice vibrations, the C60 librations and the soft modes and two-level sys- tems of the C60 glasses, and in particular those associated with the changing of the relative orientations of the C60 molecules with respect to each other. The admixed gas molecules within the octahedral sites can affect the above contributors as well as making their own contribution to the thermal expansion of the solid CO–C60 solution, through its own thermal motions. As noted above, the thermal expansion coefficients of pure C60 and the CO–C60 solutions coincide at the lowest temperatures of the experiment. This means that at these temperatures the CO impurity has little effect on the dom- inant contributors to the thermal expansion that being the translational lattice vibrations, the two-level systems and the soft modes [18]. The weak effect of the impurity on the translational vibrations of the C60 lattice is quite natu- ral because CO adds little to the effective molecular weight of the CO–C60 solutions and changes the lattice constant of C60 at most by 0.15 % [19]. As the tempera- ture rises, the contributions of the C60 librations and the motions associated with the CO molecule (translational, rotational and internal vibrational) increase significantly. But any contribution from the CO through its increased thermal motion can only lead to higher LTEC values. Therefore, the lower LTECs of the CO–C60 solutions in comparison with those of pure C60 must be attributed to the diminished contribution of the C60 librations. This arises from the fact that the CO molecules at T � 77 K are oriented in a particular fashion within the octahedral in- terstitial sites of C60 [11,20,21] so that there is a noncentral interaction between the impurity and the sur- rounding C60 molecules and which is not nullified by any rotation of the CO molecules within the sites. The CO molecule has both dipole and quadruple electrical mo- ments although the dipole moment is rather weak [22], so that the noncentral CO–C60 interaction is mainly deter- mined by the quadrupole moment of the CO molecule. This CO orientational induced noncentral force interac- tion acting on the C60 molecules tends to increases the fre- quency of their librations. As a result, the contribution of the C60 librations to the positive component of the ther- mal expansion for these CO–C60 solutions, and over the studied temperature range, is reduced as compared to that of the pure C60 sample. Only at higher temperatures will their effect be realized as their contribution increases with temperature as they become more and more thermally ac- tivated. This effect is weaker for the N2–C60 solutions as the quadrupole moment of N2 molecules is much smaller and this is clearly seen in Fig. 4 that compares the positive and negative components of the thermal expansion for samples with 90 mol. % CO and 100 mol. % N2. We have chosen these two samples to compare, because the effects of impurities upon the thermal expansion of doped fullerites are most evident at their high concentrations. In the context of the above consideration, the negative component of the thermal expansion is determined by the probability of reorientation of the C60 molecules in the domain interlayers. It is found that the magnitude of the negative component of the LTEC for the CO–C60 solu- tions decreases considerably as the CO concentration in- creases from 26 to 90% (Fig. 3). The absence of a nega- tive component in the thermal expansion of pure C60 prompts us to conclude that on dissolution of CO in the C60 lattice, the probability of C60 reorientation in the do- The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C60 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 595 5 10 15 20 – 0.5 0 0.5 1.0 T, K 1 2 3 4 5 � , 1 0 – 5 – 1 K Fig. 4. The temperature dependences of the positive and nega- tive components of the linear thermal expansion coefficient are shown for both CO–C60 (solid lines) and N2–C60 (dashed lines) [3] solutions. The positive components are the curves 1 — 90 mol. % CO and 2 — 100 mol. % N2, whilst the nega- tive components are the curves 3 — 90 mol. % CO and 4 — 100 mol. % N2. Again pure C60 which only has a positive con- tribution is shown as the dotted line (5). main interlayers must first increase with impurity concen- tration up to a certain impurity concentration but increas- ing the impurity concentration past this, it must start to decrease again and Fig. 3 even suggests it contribution is reduced to zero contribution at 100% occupancy with CO. By contrast, for N2–C60 solutions, in which the non- central interaction between the N2 and the C60 molecules is weaker, there is an opposite trend with the negative contribution to the thermal expansion being much greater at higher N2 concentrations than at lower ones. It should be noted that a change from a low to a high impurity con- centration will reduce the temperature interval of the ne- gative contribution for the CO–C60 solution but in con- trast increases it for the N2–C60 solution [3]. It is of our opinion that these observations of the ther- mal expansion behavior for the CO–C60 and N2–C60 solu- tions suggests that in both cases, there is a competition between two contrasting mechanisms. On the one hand, we have the CO and N2 impurities introduced into the in- terstitial sites of C60 pushing the neighboring C60 mole- cules farther apart. This mechanism tends to reduce the effect from the noncentral interaction between the C60 molecules but promote their reorientation. This increases the negative contribution to the thermal expansion. On the other hand, their introduction also results in a noncentral interaction between the impurity and the neighboring C60 molecules that reduces the probability of C60 reorientation and decreases the negative contribution to the thermal expansion. The first mechanism dominates in the N2–C60 solutions while the other prevails in the CO–C60 solutions with high CO concentrations. It is expected that the noncentral interaction between the impurity and C60 matrix can affect the characteristic time of C60 reorientation (�2) within the interlayers be- tween the domains. As seen in Fig. 5, the �2 values have been extracted from the negative component of the LTEC are much longer for the CO–C60 solution than for the N2–C60 one, indicating that the CO molecules with the substantially larger quadrupole moment than that of N2 greatly depress on account of this enhanced noncentral CO–C60 interaction, the probability of C60 reorientation. The local maxima in the temperature dependences of the positive components of the LTECs for the CO–C60 samples may indicate that within the interval of 4–5,5 K occurs the temperatures associated with the phase trans- formations between the orientational CO–C60 glasses. This assumption is supported by the analysis of the tem- perature dependences of the relaxation time �1(T) associ- ated with thermal equilibration of the CO–C60 solution, and obtained from the positive component of the LTEC. As shown in Fig. 6, this extracted thermalization time �1 of the sample, increases sharply in the temperature inter- val of the local LTEC maxima because the heat supplied to the sample during heating is partially consumed by the phase transformation in the orientational glass. In contrast to the CO–C60 solutions, the dependences �(T) and �1(T) of the N2–C60 solution have no distinct maxima. The glasses coexisting in gas–fullerites solu- tions differ only in the orientational order of the C60 mol- ecules [2,15–17]. Since the noncentral interaction be- tween the impurity and matrix molecules is stronger in the CO–C60 solution, we can assume that the latent heat of the phase transformation between the glasses and associ- ated with this change in the orientational order is much larger for this solution, up as maxima in the �(T) and �1(T) plots. It is known that gas impurities of high concentrations can often cause microcracking and even fracture of the C60 samples [2,3,5,23]. The higher �1 values for the sam- ple with the high CO concentration can be attributed to 596 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 A.V. Dolbin et al. 5 10 15 20 25 1100 � 2 , s T, K 20 55 150 400 Fig. 5. The characteristic time �2 for C60 reorientation ex- tracted from the negative components of the thermal expan- sion: CO–C60 with 90 (�) and 26 (�) mol. % CO; and N2–C60 with 100 mol. % N2 (�). 0 5 10 15 20 25 50 100 150 200 250 T, K � 1 , s Fig. 6. The characteristic times �1 of the positive compo- nents of thermal expansion from 90 (�) and 26 (�) mol. % CO–C60. evidence for microcracks occurring within such samples. Such an occurrence will increase the thermal resistance and hence the characteristic time of thermalization �1. The phase transformation in the CO–C60 samples was investigated by conducting a series of experiments in- volving thermocycling of the samples at T > 5.5 K. The thermocycling was performed in several narrow intervals, with the step being no more than 2 K. The experimental technique and data processing have been detailed else- where [2,3]. In the course of the thermocycling, the hys- teresis loop was found to narrow gradually (from cycle to cycle) and in doing so the negative component of the ther- mal expansion decreased until the LTECs measured on heating were approaching those obtained on cooling. The process of thermocycling thus brought the system to a more advantageous thermodynamic state between the co- existing glasses. The characteristic times �� of this process are shown in Fig. 7 as a function of the average thermo- cycling temperature. It is interesting that the dependences ��(T) are similar qualitatively for all three samples, although the ��(T) max- ima for the CO–C60 solutions are shifted towards a higher temperatures. Currently, the lack of information concern- ing the actual distinctions between the orientational glasses coexisting in the CO–C60 and N2–C60 solutions, impedes any further analysis of the temperature depen- dences of ��(T) for these systems. Conclusions A first order phase transition was observed in the orientational C60 glasses at liquid helium temperatures, during dilatometric investigations on two CO–C60 solu- tions with 26 and 90 mol. % CO. The phase transforma- tion revealed itself through observation of a hysteresis in the thermal expansion, the occurrence of local maxima in the temperature dependence of the linear thermal expan- sion coefficients, and lastly by a maximum in the temper- ature dependence of the thermalization time �1 of the in- vestigated systems. From the temperature range of the observed maxima in the thermal expansion, the phase transitions in the orientational glasses of the CO–C60 so- lutions is believed to occur in the interval 4–6 K. The thermal expansion of the CO–C60 solutions is a sum of positive and negative components and each with the characteristic relaxation times �1 and �2, respectively. �1 is the time of temperature equalization over the sample (thermalization) whilst �2 specifies the time of C60 reori- entation in the interdomain space within the CO–C60 crystallites. We compared the thermal expansion of CO–C60 and N2–C60 solutions in which the impurity molecules have close gas-kinetic diameters and molecular weights, but where CO has a considerably larger quadrupole moment than N2. Because of the stronger noncentral interaction bet- ween the interstitial CO and the neighboring C60 mole- cules, the CO–C60 solution has some specific features that distinguish it from the N2–C60 solution. These are: i) The linear thermal expansion coefficients (LTECs) are lower in the «high-temperature» phase in comparison with the LTECs of pure C60. This is because the frequen- cies of C60 librations are increased through this non- central interaction and their contribution to the LTECs shifts to temperatures above the T interval of the experi- ment. ii) The dependences �(T) and �1(T) have maxima in the temperature interval of phase transformation. No maxima were detected in the N2–C60 solutions. iii) The C60 molecules have much longer reorientation times �2, which is an obvious consequence of the en- hancement of the noncentral interaction between the im- purity and matrix molecules. iv) There is a change in the concentration dependence of the negative contribution to the LTEC. Two contrasting mechanisms are responsible for these observations asso- ciated with the negative LTEC contribution. On the one hand, impurities increase the spacings between the C60 molecules which depresses their noncentral interaction and increases the probability of their reorientation. On the other hand, the noncentral interaction between the impu- rity and matrix molecules decreases the probability of C60 reorientation. The first mechanism is dominant in the N2–C60 solutions, whilst the other predominates in the CO–C60 solutions with higher CO concentrations. The authors thank Prof. A.S. Bakai for helpful discus- sions. The authors are also indebted to the Science and The effect of the noncentral impurity-matrix interaction upon the thermal expansion and polyamorphism of solid CO–C60 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 597 4 6 8 10 12 14 16 18 20 22 0 1000 2000 3000 4000 5000 6000 7000 � ', s T, K Fig. 7. The temperature dependences of the characteristic time �� for the phase transformation between the orientational glasses for N2–C60 with 100 mol. % N2 (�, [3]) and for CO–C60 with 26 (�) and 90 (�) mol. % CO. Technology Center of Ukraine (STCU) for the financial support of this study (Project Uz-116). 1. A.N. Aleksandrovskii, A.S Bakai, A.V. Dolbin, V.B. Esel’son, G.E. Gadd, V.G. Gavrilko, V.G. Manzhelii, S. Moricca, B. Sundqvist, and B.G. Udovidchenko, Fiz. Nizk. Temp. 29, 432 (2003) [Low Temp. Phys. 29, 324 (2003)]. 2. A.N. Aleksandrovskii, A.S. Bakai, D. Cassidy, A.V. Dol- bin, V.B. Esel’son, G.E. Gadd, V.G. Gavrilko, V.G. Man- zhelii, S. Moricca, and B. Sundqvist, Fiz. Nizk. Temp. 31, 565 (2005) [Low Temp. Phys. 31, 429 (2005)]. 3. V.G. Manzhelii, A.V. Dolbin, V.B. Esel’son, V.G. Gavril- ko, D. Cassidy, G.E. Gadd, S. Moricca, and B. Sundqvist, Fiz. Nizk. Temp. 32, 913 (2006) [Low Temp. Phys. 32, 695 (2006)]. 4. A.V. Dolbin, N.A.Vinnikov, V.G. Gavrilko, V.B. Esel’son, V.G. Manzhelii, and B. Sundqvist, Fiz. Nizk. Temp. 33, 618 (2007) [Low Temp. Phys. 33, 465 (2007)]. 5. A.V. Dolbin, V.B. Esel’son, V.G. Gavrilko, V.G. Man- zhelii, N.A.Vinnikov, G.E. Gadd, S. Moricca, D. Cassidy, and B. Sundqvist, Fiz. Nizk. Temp. 33, 1401 (2007) [Low Temp. Phys. 33, 1068 (2007)]. 6. A.I. Prokhvatilov, N.N. Galtsov, I.V. Legchenkova, M.A. Strzhemechny, D. Cassidy, G.E. Gadd, S. Moricca, B. Sun- dqvist, and N.A. Aksenova, Fiz. Nizk. Temp. 31, 585 (2005) [Low Temp. Phys. 31, 445 (2005)]. 7. B. Sundqvist, Adv. Physics 48, 1 (1999). 8. S. Chapmen and T.G. Cowling, The Mathematical Theory of Nonuniform Gases, Cambridge (1939), pp. 225 and 229. 9. C. Graham, D.A. Imrie, and R.E. Raab, Mol. Phys. 93, 49 (1998). 10. B. Renker, G. Roth, H. Schober, P. Nagel, R. Lortz, C. Meingast, D. Ernst, M.T. Fernandez-Diaz, and M. Koza, Phys. Rev. B64, 205417 (2001). 11. S. Smaalen, R. Dinnebier, I. Holleman, G. Helden, and G. Meijer, Phys. Rev. B57, 6321 (1998). 12. A.M. Aleksandrovskii, M.A. Vinnikov, V.G. Gavrilko, O.V. Dolbin, V.B. Esel’son, and. V.P. Maletskiy, Ukr. Fiz. Zh. 51, 1150 (2006). 13. G.E. Gadd, S. Moricca, S.J. Kennedy, M.M. Elcombe, P.J. Evans, M. Blackford, D. Cassidy, C.J. Howard, P. Prasad, J.V. Hanna, A. Burchwood, and D. Levy, J. Phys. Chem. Solids 58, 1823 (1997). 14. A.N. Aleksandrovskii, V.B. Esel’son, V.G. Manzhelii, A.V. Soldatov, B. Sundqvist, and B.G. Udovidchenko, Fiz. Nizk. Temp. 23, 1256 (1997) [Low Temp. Phys. 23, 943 (1997)]. 15. V.M. Loktev, J.N. Khalack, and Yu.G. Pogorelov, Fiz. Nizk. Temp. 27, 539 (2001) [Low Temp. Phys. 27, 397 (2001)]. 16. J.M. Khalack and V.M. Loktev, Fiz. Nizk. Temp. 29, 577 (2003) [Low Temp. Phys. 29, 429 (2003)]. 17. A.S. Bakai, Fiz. Nizk. Temp. 32, 1143 (2006) [Low Temp. Phys. 32, 1033 (2006)]. 18. M.A. Ramos, C. Talon, R.J. Jimenez-Rioboo, and S. Viei- ra, J. Phys.: Condens. Matter 15, 1007 (2003). 19. N.N. Galtsov, A.I. Prokhvatilov, G.N. Dolgova, D. Cas- sidy, G.E. Gadd, S. Moricca, and B. Sundqvist, Fiz. Nizk. Temp. 33, 1159 (2007) [Low Temp. Phys. 33, 881 (2007)]. 20. I. Holleman, G. von Helden, E.H.T. Oithof, P.J.M. van Bentum, R. Engeln, G.H. Nachtegaal, A.P.M. Kentgens, B.H. Meier, A. van der Avoird, and G. Meijer, Phys. Rev. Lett. 79, 1138 (1997). 21. I. Holleman, G. von Helden, A. van der Avoird, and G. Meijer, Phys. Rev. Lett. 80, 4899 (1998). 22. M.W. Melhuish and R.L. Scott, J. Phys. Chem. 68, 2301 (1964). 23. L.S. Fomenko, S.V. Lubenets, V.D. Natsik, Yu.E. Stetsen- ko, K.A. Yagotintsev, M.A. Strzhemechny, A.I. Prokhvati- lov, Yu.A. Osipyan, A.N. Izotov, and N.S. Sidorov, Fiz. Nizk. Temp. 34, 86 (2008) [Low Temp. Phys. 34, 69 (2008)]. 598 Fizika Nizkikh Temperatur, 2008, v. 34, No. 6 A.V. Dolbin et al.