J. Chem.

Thermodynamics 118 (2018) 21–25

Contents lists available at ScienceDirect

J. Chem. Thermodynamics
journal homepage: www.elsevier.com/locate/jct

Low-temperature thermodynamic properties of holmium selenide (2:3)

M.A. Bespyatov a,⇑, A.E. Musikhin a, V.N. Naumov a, L.N. Zelenina a,b, T.P. Chusova a, R.E. Nikolaev a,
N.G. Naumov a,b
a
Nikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russia
b
Novosibirsk State University, Pirogova Street 2, 630090 Novosibirsk, Russia

a r t i c l e i n f o a b s t r a c t

Article history: Holmium selenide Ho2Se3(cr) was synthesized and characterized by powder X-ray diffraction and energy
Received 26 May 2017 dispersive analysis. The heat capacity of Ho2Se3(cr) has been measured by adiabatic method in the tem-
Received in revised form 7 September 2017 perature range of (4.75–302.60) K. The pronounced k-anomalies were found in Cp,m(T) dependence with
Accepted 26 October 2017
the maxima at temperatures TC1 = 6.03 K and TC2 = 228.40 K. A conclusion was drawn that the phase tran-
Available online 27 October 2017
sition associated with antiferromagnetic ordering is observed at TC1. The values of thermodynamic func-
tion (entropy, enthalpy and Gibbs reduced energy) were calculated from the experimental Cp(T)
Keywords:
dependence. Absolute entropy value was used for calculation of Ho2Se3(cr) formation entropy at T =
Heat capacity
Thermodynamic functions
298.15 K.
Holmium selenide Ó 2017 Elsevier Ltd.

1. Introduction
Rare-earth chalcogenides are the perspective materials for prac-
tical application in thermoelectric converters of energy, laser tech-
nique, precise metallurgy and other fields of technology [1–4]. At
the same time many properties of these compounds, including
thermodynamic, are studied insufficiently. For calculating chemi-
cal equilibrium constants it is necessary to know the exact values
of such quantities as entropy and enthalpy under standard condi-
tions. The most reliable data of these quantities can be obtained
from the results of the low-temperature heat capacity measure-
ment [5–7]. At present, there are no data on the heat capacity of
Ho2Se3 at low temperatures.
Holmium selenide Ho2Se3 is a semiconductor material with p-
type conduction [8]. It crystallizes in orthorhombic structure type
of Sc2S3 which is a superstructure of the NaCl type based on the
ordering of cation vacancies [9]. Magnetic susceptibility of Ho2Se3
was investigated in the temperature interval 78–400 K in [10]. In
this temperature range, the magnetic susceptibility obeys the
Curie-Weiss law with a negative Weiss constant (HP =
5 K). This
means that a phase transition from the paramagnetic phase to the
antiferromagnetic one is possible for Ho2Se3 at low (close to 0 K)
temperatures. This transition is known to manifest itself in heat
capacity as k-anomaly.
This work presents the results of experimental study of Ho2Se3
heat capacity by adiabatic calorimetry in the temperature range
⇑ Corresponding author.
E-mail address: bespyatov@niic.nsc.ru (M.A. Bespyatov).
from 4.75 to 302.60 K. The entropy, enthalpy and Gibbs reduced
energy of Ho2Se3(cr) were calculated on the base of data obtained.
Absolute entropy data were used for calculation of Ho2Se3(cr) for-
mation entropy at T = 298.15 K.
2. Experimental
2.1. Preparation and characterization of the sample
The sample of Ho2Se3 investigated in this work was prepared
from the elements of mass fraction purity as follows (Ho: 0.999;
Se: 0.999) which were taken in stoichiometric ratio for obtaining
11.3 g of final product in the presence of a small amount (<0.05
mol.%) of iodine (I2: 0.9999). Holmium ingot was thoroughly
mechanically cleaned from traces of oxide and put together with
selenium in a silica ampoule with a graphite crucible inside to pre-
vent a direct contact between metal and SiO2. Iodine was added to
activate Ho and to prevent the formation of a dense protective
layer of monoselenide or sesquiselenide on the surface of unre-
acted metal [11]. All manipulations were made in dry box filled
with Ar. Evacuated ampoule was filled with Ar, frozen after addi-
tion of I2 evacuated again and sealed.
Sealed ampoule was put in a two-zone vertical resistance fur-
nace and was heated of to a temperature of 1118 K with a rate of
15 K/h. After holding the ampoule in an isothermal section of
furnace at a temperature of (1118 ± 5) K for 620 h, a temperature
gradient was set along the length of the ampoule. The end of the
ampoule where the substances were located was kept at a

https://doi.org/10.1016/j.jct.2017.10.013
0021-9614/Ó 2017 Elsevier Ltd.
22 M.A. Bespyatov et al. / J. Chem. Thermodynamics 118 (2018) 21–25
temperature of (1118 ± 10) K, while the temperature of the empty
end of the ampoule was set to (823 ± 10) K. This procedure allowed
us to remove iodine in the form of holmium triiodide from the final
product and to verify the completeness of the reaction. After hold-
ing in this mode for 170 h, the ampoule was removed from the hot
furnace and cooled in air. A small amount of light yellow conden-
sate (HoI3) was observed at the cold end of the ampoule. The
absence of selenium condensate at the cold end of the ampoule
testified to the completeness of the synthesis reaction. The final
product was brown crystalline powder.
The synthesized sample was studied by powder X-ray diffrac-
tion and energy dispersive analysis (EDS).
X-ray powder diffraction pattern was collected on a Philips APD
1700 diffractometer (Cu-Ka radiation, graphite monochromator,
silicon pellet as an external standard, room temperature) in the
2h range of (10–80)° with a step of 0.02°. According powder diffrac-
tion data the sample is a single phase. All reflections are indexed in
orthorhombic Fddd space group (No. 70) with refined cell param-
eters a = (1.1416 ± 0.0002) nm, b = (0.8131 ± 0.0002) nm, c =
(2.4270 ± 0.0003) nm that correspond to literature data for Ho2Se3
[9] (Fig. 1).
The composition of the sample obtained was identified by
microprobe analysis on a TM 3000 HITACHI device (EDS System
for Hitachi Tabletop Microscope TM 3000 Quantax). Polished sam-
ple for EDS analyses was prepared according standard procedure.
Elemental intensities for Ho and Se were determined relative to
single crystal standards using HoPO4 and Bi2Se3. The analysis
results are averaged over 10–15 measurements. According to the
analysis results the composition of the sample corresponds to the
chemical formula HoSe1.50±0.02 (Table 1) and it does not contain
any impurities within the error of the method (atomic content of
holmium and selenium <0.2%).
2.2. Heat capacity measurements
The heat capacity Cp,m(T) of the sample of Ho2Se3 was measured
by the adiabatic method on the equipment described in [12]. A dis-
mountable calorimeter ampoule and the methodology of filling an
ampoule with a substance, as well as the filling of the ampoule
with heat-exchange gas (helium), are all described in our previous
works [13,14]. The temperature of the calorimeter was measured
by a standard platinum resistance thermometer (R100/R0 =
1.3925). This thermometer was calibrated at FSUE ‘‘VNIIFTRI”
(Moscow region, Russia) against the ITS-90 for T > 13 K, and by
Fig. 1. Experimental (black) and theoretical (grey) powder pattern for Ho2Se3.
Table 1
Characterization of the chemicals used.
Chemicals Source/supplier State Mass Fractional Purity
Ho ‘‘Metallon”, Russia Solid 0.999
Se ‘‘Metallon”, Russia Solid 0.999
HoSe1.50±0.02 Synthesis Solid >0.99 (microprobe analysis)
the method of McCracken and Chang [15] for T < 13 K. The stan-
dard uncertainty for the temperature was u(T) = 0.01 K. The resis-
tance thermometer was measured by a potentiometric method
with cyclic inversion of the thermometer current direction for
excluding the thermal electromotive forces. Resolution of the ther-
mometric apparatus was 510
5 K above 50 K falling to 1.510
3 K
at 11 K. The adiabatic control system gave the temperature stabil-
ity of the calorimeter vessel within 110
5 Kmin
1. The measure-
ment results of standard substance (benzoic acid) heat capacity
which characterize a calibration uncertainty deviate from the stan-
dard data [16] by less than 1% in the range of (5
2 0) K, less than
0.3% in the range of (20–80) K, and less than 0.15% in the range of
(80–300) K. The sample of Ho2Se3 used for the heat capacity mea-
surements was weighed in vacuo. The sample mass is (5.1255 ±
0.0001) g. The molar mass used in the calculation of the molar heat
capacity was determined from the formula Ho2Se3 as 566.74
gmol
1.
3. Results
The experimental heat capacity values for Ho2Se3 are presented
in chronological sequence in Table 2. The heat capacity Cp,m(T) has
been measured at 82 points in the range (4.75–302.60) K. The mea-
surements of Series III were with small temperature rise to deter-
mine the shape of the heat capacity curve near the peak.
In the experimental heat capacity two pronounced k-anomalies
are observed (Figs. 2 and 3). The type of anomalies indicates phase
transitions with critical temperatures TC1 = 6.03 K and TC2 = 228.40
K.
It was established in [10] that Ho2Se3 is a paramagnetic and at
temperatures close to 0 K it can undergo a ‘‘paramagnetic-anti
ferromagnetic” phase transition. In addition, similar transitions to
the magnetically ordered state were observed for various rare-
earth chalcogenides at temperatures below 10 K [8]. Proceeding
from this, it can be concluded that the nature of the phase transi-
tion at TCl is related to the antiferromagnetic ordering of the mag-
netic moments of holmium ions.
The magnetic susceptibility obeys the Curie-Weiss law in the
range of (80–400) K [10]. This means that the nature of the anoma-
lies manifested in the heat capacity in the interval (190–224) K is
nonmagnetic. Two series of experiments with different heating
rates were performed in this interval. Neither evidence of thermal
hysteresis nor slowness in attainment of equilibrium was found.
The absence of a temperature hysteresis and the k-shaped anomaly
in the heat capacity are signs second-order phase transitions. Ano-
malous contributions to entropy and enthalpy in the interval (190–
224) K are DSan = (0.15 ± 0.03) JK
1mol
1 and DHan = (29 ± 5)
Jmol
1. For detailed analysis the nature of the observed phase
transition, one needs to conduct a special research.
The electric crystal field in which the Ho3+ ion is situated in the
compound Ho2Se3 splits the seventeen fold degeneracy of the state
of the multiplet term 5I8 into sublevels. New levels with higher
energies are formed as a result of the splitting. The process of ther-
mally activated transitions of electrons to the excited states begins
with an increase in the temperature. During this process the
energy is absorbed, which causes a contribution to the heat capac-
ity Ho2Se3 (the Schottky contribution CSch(T)).
 M.A. Bespyatov et al. / J. Chem. Thermodynamics 118 (2018) 21–25 23

Table 2
Experimental values of heat capacitya for crystalline Ho2Se3 (molar mass: 566.74 gmol
1) at pressure p = .1 MPa (u(p) = 0.05p).

T/K Cp,m/Jmol
1K
1 T/K Cp,m/Jmol
1K
1 T/K Cp,m/Jmol
1K
1

Series 1 72.45 86.70 249.67 123.4
4.75 4.433 76.57 90.48 256.45 123.6
5.01 4.889 Series 2 263.41 123.5
5.32 7.582 80.18 93.23 270.54 123.7
6.03 11.86 85.83 97.43 Series 3
7.23 9.963 91.80 101.0 217.14 123.4
8.62 9.268 97.35 103.9 220.38 123.3
10.31 9.187 102.86 106.2 222.96 123.9
10.68 9.794 108.72 108.5 224.88 124.3
11.33 10.14 114.93 110.9 226.80 124.9
12.42 11.01 121.18 112.4 228.40 125.0
13.07 11.54 127.91 114.2 229.68 124.4
14.47 12.75 135.11 115.7 230.95 123.6
16.86 14.62 142.05 116.8 232.22 122.9
20.03 17.51 148.79 117.9 233.80 123.6
22.71 19.72 155.35 118.6 235.69 122.7
25.13 22.18 161.76 119.2 238.19 123.1
28.04 25.54 168.04 119.9 241.31 123.4
31.20 29.55 174.80 120.3 244.42 123.3
34.35 34.03 182.04 120.9 Series 4
37.49 38.79 189.15 121.4 274.01 123.6
40.37 43.37 196.14 122.1 276.98 124.0
43.16 48.00 203.04 122.2 280.45 123.7
46.66 53.47 209.83 122.9 284.65 124.0
51.18 60.48 216.54 123.2 289.20 123.9
55.76 66.98 223.16 124.0 293.71 123.6
59.76 72.46 229.70 124.4 298.18 123.9
63.99 77.76 236.16 123.1 302.60 124.2
68.39 82.55 242.82 122.7
a
Standard uncertainty for temperature u(T) = 0.01 K; relative combined standard uncertainty for the heat capacities uc,r(Cp,m): 0.04 at T/K < 20, 0.004 at 20  T/K < 80,
0.0018 at 80  T/K < 190, 0.003 at 190  T/K < 245, 0.0019 at T/K  245.

Fig. 3. Experimental heat capacity for Ho2Se3 in the range from 150 K to 302.60 K.
Fig. 2. Experimental heat capacity for Ho2Se3 in the range from 4.75 K to 302.60 K.

The total heat capacity Cp,m(T) of Ho2Se3, in the range 5–190 K

can be represented in the form:
C p;m ðTÞ ¼ C M ðTÞ þ C L ðTÞ þ C Sch ðTÞ;
where CM(T) and CL(T) are the magnetic (below 10 K) and lattice
components respectively. At present, there is insufficient informa-
tion to separate these contributions. Data on the heat capacity for
isostructural nonmagnetic compounds, necessary for the calcula-
tion of the lattice component, are not presented in the literature.
The experimental data on heat capacity have been smoothed by
the Rumshiskii method [17]. This method approximates the exper-
imental curve by spline functions. Each spline function of the third,
fourth or a higher order approximates a pre-selected number of
points k (k = 5, 6, 7 or more) depending on the degree of curvature
of the function to be smoothed. In so doing, spline functions with
matching at the connection point of the first derivatives for the i-
ð1Þ th and (i + 1)-th spline functions can be used.
Deviation of the experimental values from the smoothed curve
is shown in Fig. 4. The standard deviation of the experimental
points from the smoothed curve is equal to: 3.4% (4.75–20 K),
0.19% (20–80 K), 0.10% (80–190 K), 0.20% (190–245 K) and 0.13%
(245–302.60 K).
The information on the heat capacity from 4.75 to 0 K is neces-
sary for the calculation of thermodynamic functions. Holmium
selenide is an antiferromagnetic below 6 K. Therefore, the thermal
excitation of magnons near 0 K contributes to the heat capacity
proportional to T3 and is added to the phonon contribution, which
is also proportional to T3 at low temperatures [18]. Accordingly, the
24 M.A. Bespyatov et al. / J. Chem. Thermodynamics 118 (2018) 21–25

temperature dependence of the heat capacity near 0 K can be

represented:

C p;m ¼ ða1 þ a2 Þ  T 3 ; ð2Þ

where a1 is the constant associated with the Debye temperature,

Fig. 4. Relative deviations (DC = Cp,m(T)
Cs(T)) of the experimental values of heat
capacity Cp,m(T) from the smoothed curve Cs(T) for Ho2Se3.
Table 3
Molar thermodynamic functionsa (heat capacity C°p,m, entropy D0TS°m, enthalpy D0TH
°m, reduced Gibbs energy U°m = D0TS°m
D0TH°m/T) for crystalline Ho2Se3 at pressure
p = .1 MPa (u(p) = 0.05p); molar mass: 566.74 gmol
1.
and a2 is the constant associated with the Neel temperature. The
Schottky contribution, which is present in such compounds, was
assumed to be negligibly small near 0 K [18]. In this connection,
the data about a heat capacity below 4.75 K were obtained by
graphically extrapolating the experimental heat capacity to zero
by the equation of the straight line in the coordinates of Cp,m vs. T3.
The values of entropy D0TS°m, enthalpy D0TH°m and reduced
Gibbs’ energy U°m were obtained by numerical integration of the
smoothed dependence Cs(T) in the range of (0–302) K. The
smoothed values of the heat capacity and the thermodynamic
functions are listed in Table 3.
Using the value of the absolute entropy of holmium selenide
and holmium [19] and selenium [20], the entropy of formation at
T = 298.15 K was calculated to be DfS(298.15, Ho2Se3, cr) =

(42.4 ± 0.3) Jmol
1K
1. This value corresponds to the following
reaction:
2HoðcrÞ þ 3SeðcrÞ ! Ho2 Se3 ðcrÞ; ð3Þ
where cr – crystalline.
4. Conclusions

T/ C°p,m/ D0TS°m/ D0TH°m/ U°m/ The isobaric heat capacity of Ho2Se3 crystalline sample has been
K Jmol
1K
1 Jmol
1K
1 Jmol
1 Jmol
1K
1 studied for the first time in the temperature interval from 4.75 to
0 0 0 0 0 302.60 K. The k-anomaly, which is associated with antiferromag-
5 5.229 1.640 6.157 0.409 netic ordering, was revealed in the temperature range of (5
1
6.03 12.00 3.189 14.78 0.737
0) K with maximum at Tc1 = 6.03 K. One more k-anomaly with
10 9.321 8.227 53.81 2.846
15 13.14 12.67 109.3 5.387 maximum at Tc2 = 228.40 K which points to the second order phase
20 17.39 17.04 185.6 7.755 transition was revealed in the temperature interval from 190 to
25 22.07 21.40 283.9 10.05 245 K. Standard thermodynamic functions (entropy, enthalpy and
30 27.86 25.92 408.3 12.31 Gibbs reduced energy) of Ho2Se3(cr) as well as its formation
35 34.94 30.74 564.9 14.60
40 42.81 35.91 759.0 16.93
entropy at T = 298.15 K were calculated on the base of heat capac-
45 50.85 41.42 993.2 19.34 ity data.
50 58.65 47.18 1267 21.84
60 72.86 59.16 1927 27.05 Acknowledgements
70 84.33 71.30 2715 32.51
80 93.16 83.16 3604 38.10
90 99.99 94.54 4572 43.75 The reported study was supported by RFBR, research project No.
100 105.2 105.4 5600 49.37 14-03-00619.
110 109.2 115.6 6671 54.93
120 112.4 125.2 7780 60.39
130 114.8 134.3 8916 65.73 References
140 116.5 142.9 10,070 70.94
150 117.9 151.0 11,250 76.01 [1] C. Wood, High-temperature thermoelectric energy conversion-II, Materials
160 119.0 158.6 12,430 80.94 Surv Energy Convers. Manage. 24 (1984) 331–343.
[2] P.N. Kumta, S.H. Risbud, Rare-earth chalcogenides – an emerging class of
170 119.9 165.9 13,630 85.72
optical materials, J. Mater. Sci. 29 (1994) 1135–1158.
180 120.8 172.7 14,830 90.37
[3] A.V. Prokofiev, A.I. Shelykh, A.V. Golubkov, I.A. Smirnov, Crystal growth and
190 121.5 179.3 16,040 94.88
optical properties of rare earth sesquiselenides and sesquisulphides – new
200 122.1 185.5 17,260 99.25 magneto-optic materials, J. Alloys Comp. 219 (1995) 172–175.
210 122.8 191.5 18,480 103.5 [4] J.S. Sanghera, L.B. Shaw, I.D. Aggarwal, Applications of chalcogenide glass
220 123.6 197.3 19,710 107.6 optical fibers, C. R. Chim. 5 (2002) 873–883.
228.4 125.2 201.9 20,760 111.0 [5] I.V. Korchemkin, V.I. Pet’kov, A.V. Markin, N.N. Smirnova, A.M. Kovalskii,
230 124.3 202.8 20,960 111.7 Thermodynamic properties of caesium–cobalt phosphate CsCoPO4, J. Chem.
240 123.1 208.0 22,190 115.6 Thermodyn. 96 (2016) 34–40.
250 123.4 213.0 23,420 119.4 [6] M.R. Bissengaliyeva, N.S. Bekturganov, D.B. Gogol, S.T. Taimassova, Low-
260 123.6 217.9 24,660 123.1 temperature heat capacity and thermodynamic functions of natural
270 123.7 222.6 25,890 126.7 chalcanthite, J. Chem. Thermodyn. 111 (2017) 199–206.
280 123.8 227.1 27,130 130.2 [7] J.J. Calvin, M. Asplund, Y. Zhang, B. Huang, B.F. Woodfield, Heat capacity and
290 123.9 231.4 28,370 133.6 thermodynamic functions of c-Al2O3, J. Chem. Thermodyn. 112 (2017) 77–85.
[8] H. Bergmann, H. Hein, P. Kuhn, U. Vetter, Rare earth elements and selenium in
298.15 123.9 ± 0.2 234.8 ± 1.2 29,380 ± 80 136.3 ± 0.8
book, in: Sc, Y, La-Lu rare earth elements C 9, Compounds with Se, 1985, pp. 1–
300 123.9 235.6 29,610 136.9
526.
302 123.9 236.4 29,860 137.6 [9] W. Urland, H. Person, Zur Kristallstruktur von Ho2Se3, Z. Naturforsch. B Chem.
a Sci. 53 (1998) 900–902.
Standard uncertainty for temperature u(T) = 0.01 K; relative combined standard
[10] G.V. Lashkarev, V.P. Fedorchenko, V.A. Obolonchlk, I.P. Skripka, Magnetic
uncertainty for the heat capacities uc,r(Cp,m): 0.04 at T/K < 20, 0.004 at 20  T/K < 80,
susceptibility of the selenides of some yttrium-group rare-earth elements,
0.0018 at 80  T/K < 190, 0.003 at 190  T/K < 245, 0.0019 at T/K  245.
Powder Metall. Met. Ceram. 15 (1976) 394–396.
 M.A. Bespyatov et al. / J. Chem. Thermodynamics 118 (2018) 21–25
[11] K.A. Gschneidner Jr., Preparation and processing of rare earth chalcogenides, J.
Mater. Eng. Perform. 7 (1998) 656–660.
[12] V.A. Drebushchak, V.N. Naumov, V.V. Nogteva, I.A. Belitsky, I.E. Paukov,
Thermochim. Acta 348 (2000) 33–40.
[13] V.N. Naumov, V.V. Nogteva, Calorimeter with dismountable seal for low
temperature research, Instrum. Exp. Tech. 28 (1985) 1194–1199.
[14] M.A. Bespyatov, I.S. Cherniaikin, K.V. Zherikova, V.N. Naumov, I.K. Igumenov,
N.V. Gelfond, N.B. Morozova, Low-temperature heat capacity of tetrakis
(2,2,6,6-tetramethyl-3,5-heptanedionato)zirconium, J. Chem. Thermodyn.
110 (2017) 171–174.
[15] F.L. McCracken, S.S. Chang, Simple calibration procedures for platinum
resistance thermometers from 2.5 to 14 K, Rev. Sci. Instrum. 46 (1975) 550–
553.
[16] N.P. Rybkin, M.P. Orlova, A.K. Baranyuk, N.G. Nurullaev, L.N. Rozhnovskaya,
Precision calorimetry at low temperatures, Meas. Tech. 17 (1974) 1021–1026.
25
[17] A.E. Musikhin, V.N. Naumov, M.A. Bespyatov, N.V. Ivannikova, The heat
capacity of Li2MoO4 in the temperature range 6–310 K, J. Alloys Compd. 639
(2015) 145–148.
[18] E. Gmelin, Modern low-temperature calorimetry, Thermochim. Acta 29 (1979)
1–39.
[19] B.C. Gerstein, M. Griffel, L.D. Jennings, R.E. Miller, R.E. Skochdopole, F.H.
Spedding, Heat capacity of holmium from 15 to 300 K, J. Chem. Phys. 27 (1957)
394.
[20] S.S. Chang, A.B. Bestul, Heat capacities of selenium crystal (trigonal), glass, and
liquid from 5 to 360 K, J. Chem. Thermodyn. 6 (1974) 325–344.
JCT 17-449