1638 J. Chem. Eng.

Data 2001, 46, 1638-1642

Measurement of Heat Capacities for 12 Organic Substances by

Tian-Calvet Calorimetry

1 rgen Gmehling*
Lars Becker and Ju
University Oldenburg, Technische Chemie, Fachbereich 9, Postfach 2503, 26111 Oldenburg, Germany

Heat capacities for 1,2-dimethoxybenzene, 2,5,8,11-triethylene glycol dimethyl ether, N-ethyl-2-pyrrolidone,

isopentyl acetate, 2-ethylhexanal, toluene, 1-phenylethanol, propyl acetate, octyl acetate, dimethyl
sulfoxide, diethyl phthalate, and diethyl carbonate were measured with the “three-step” method using a
Tian-Calvet batch calorimeter. The measurements of saturated liquid heat capacity have an approximate
uncertainty of (0.5% and cover a temperature range between 310 K and 420 K. Furthermore, the
experimental results were compared with three group contribution methods for liquid heat capacities.

Introduction and electrical heating and is monitored using a Pt 100

resistance thermometer located between the two cells. Only
To complete the Dortmund Data Bank, which contains
the temperature of the calorimeter block is recorded. It is
nearly all worldwide available phase equilibrium data and
not possible to measure the temperature of the sample in
excess properties, work on pure component properties was
the cell directly. For heat capacity measurements the
started in 1992 at the University of Oldenburg. The main
temperature of the calorimeter block is varied linearly with
objectives of the project, sponsored by the German Federal
time. This means that there is a temperature difference
Ministry for Research and Technology (BMBF), are, in
between the calorimeter block and the sample during the
addition to the basic aim of collecting and correlating the
measurement. To consider this, the melting points of
data, the testing and further development of estimation
heptane and benzoic acid were measured using three
methods for pure component properties with a view to the
different temperature programs with heating rates of 0.15,
development of software tools for CAMD. Although the
0.3, and 0.5 K‚min-1 (in total six “melting” points).
Department of Industrial Chemistry at the University of
The parameters of the following equation were fitted to
Oldenburg has built up the world’s largest factual data
the differences ∆Tm between the measured and published
bank for pure component properties, it is still required to
melting temperatures (Tm ) 182.57 K for heptane6 and
measure data for compounds with missing or questionable
Tm ) 395.49 K for benzoic acid7) and the corresponding
values or for which data are available only for a limited
heating rates Z:
temperature range.
The aim of this work is to measure heat capacities with Tcorr ) b0 + b1∆Tm + b2Z + b3Z2 (1)
the help of a SETARAM Tian-Calvet calorimeter using
the “three-step” method. Furthermore, a comparison of the Tcorr is the corrected temperature. The temperature is
experimental data with calculated values of the group computer controlled, and the signals (temperature and heat
contribution methods for isobaric heat capacities developed flux) are automatically recorded.
by Ruzicka and Domalski,1,2 Missenard,3 and Luria and The cells used (volume ) 8.5 cm3) are made of stainless
Benson4 was performed. Additionally, the reproducibility steel with aluminum or Teflon seals. The two cells were
of the Tian-Calvet calorimeter for the measurement of placed in the calorimeter, one charged with the sample and
heat capacities was checked. the other cell empty. To determine the mass of the samples
a SATORIUS A 200 S balance with an accuracy of (0.2
Experimental Section mg was used. As the cells were not totally filled with the
The measurements reported here were performed using sample, the measured heat capacities csat are measured at
a Tian-Calvet heat flow batch calorimeter from SETARAM saturated pressure. At low vapor pressure, as in this case,
(model BT2.15), applicable in a temperature range from the heat of vaporization can be neglected. Therefore, it is
77 K to 473 K. The principle of the measurement method not necessary to distinguish between isobaric heat capaci-
has been described in detail by Calvet and Prat.5 The ties Cp at the saturation curve and csat below 0.9Tb (normal
calorimeter consists of two thermal fluxmeters, each con- boiling point) because the differences are smaller than the
structed by a series of 480 thermocouples surrounding a uncertainty of high-precision measurements.8
cylindrical cavity. The fluxmeters are arranged sym- The temperature program of the three-step method
metrically around the two cells in an aluminum block (performed between 20 °C and 160 °C) keeps the temper-
located in the cavity. The signal delivered by the power ature of the cells constant for 4 h at 20 °C before the
difference of the two fluxmeters is proportional to the heat temperature is increased to 160 °C with a heating rate of
effects occurring in the cells. The temperature of the 0.15 K‚min-1. This procedure corresponds to Figure 1.
calorimeter block can be regulated using liquid nitrogen The temperature program has to be repeated with empty
cells and with the reference sample (because of the refer-
* Author to whom correspondence should be addressed (e-mail ence sample no electrical calibration is required). In our
Gmehling@tech.chem.uni-oldenburg.de). investigation, sapphire was used as reference substance,
10.1021/je010177o CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/24/2001
 Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001 1639

Registry No. 112-49-2), N-ethyl-2-pyrrolidone (C6H11NO,

113.159 g/mol, CAS Registry No. 2687-91-4), isopentyl
acetate (C7H14O2, 130,19 g/mol, CAS Registry No. 123-92-
2), 2-ethylhexanal (C8H16O, 128.214 g/mol, CAS Registry
No. 123-05-7), toluene (C7H8, 92.14 g/mol, CAS Registry
No. 108-88-3), 1-phenylethanol (C8H10O, 122.17 g/mol, CAS
Registry No. 98-85-1), propyl acetate (C5H10O2, 102.13
g/mol, CAS Registry No. 109-60-4), octyl acetate (C10H20O2,
172.27 g/mol, CAS Registry No. 112-14-1), dimethyl sul-
foxide (C2H6OS, 78.13 g/mol, CAS Registry No. 67-68-5),
diethyl phthalate (C12H14O4, 222.24 g/mol, CAS Registry
No. 84-66-2), and diethyl carbonate (C5H10O3, 118.13 g/mol,
CAS Registry No. 105-58-8) were measured. (CAS Registry
No. were provided by the author.) All chemicals were
obtained from Aldrich and stored over molecular sieve 3
Å. All compounds were distilled under vacuum using a
Vigreux column with a height of 1.5 m. The purities
(determined using gas chromatography) of all compounds
were >99.95% by mass.
Figure 1. Three-step method for determination of heat capacities
(the course of temperature is not true to scale).
Results
because a large number of reliable heat capacity data are The results are listed in Table 1 and presented in Figure
available for it. For each run (temperature, heat flow, and 2 together with the published data. For a clear presentation
time) 5000 heat-flow data points are recorded. It is not not all 1500 data points but the data in 5 K steps are given.
possible to calculate heat capacities from each data point
because a large number of data points are necessary to get Discussion
reliable baselines at the start and the end of a measure- One of the aims of the pure component properties project
ment and to reach equilibrium during the temperature is the test of estimation methods. For this study substances
increase. Therefore, only 1500 data points can be used for are suitable that were not used for fitting the parameters,
calculating the heat capacities with the following equation:7 for example, compounds for which no experimental data
are available so far. For 1,2-dimethoxybenzene, N-ethyl-
Q̇sample - Q̇empty nref 2-pyrrolidone, isopentyl acetate, 2-ethylhexanal, and 1-phen-
cprobe ) × × cref (2)
Q̇ref - Q̇empty nsample ylethanol no experimental heat capacity data are available.
For 2,5,8,11-triethylene glycol dimethyl ether, propyl ac-
n describes the number of moles of sample (nsample) or of etate, octyl acetate, dimethyl sulfoxide, diethyl phthalate,
the reference sample (nref). and diethyl carbonate only a small temperature range was
Heat capacities for 1,2-dimethoxybenzene (C8H10O2, covered. The measured heat capacities were compared with
138.17 g/mol, CAS Registry No. 91-16-7), 2,5,8,11-trieth- heat capacities predicted by the group contribution meth-
ylene glycol dimethyl ether (C8H18O4, 178.23 g/mol, CAS ods of Missenard,3 Luria and Benson,4 and Ruzicka and

Figure 2. Experimental saturated heat capacity data csat measured with the three-step method compared with published data: b, 1,2-
dimethoxybenzene, this work; -, 2,5,8,11-tetraoxadodecane, this work; - - -, 2,5,8,11-tetraoxadodecane, Domalski et al.;8 4, N-ethyl-2-
pyrrolidone, this work; +, isopentyl acetate, this work; 0, 2-ethylhexanal, this work; 9, toluene, this work; - ‚ ‚ - ‚, toluene, Domalski et
al.;8 ], 1-phenylethanol, this work; /, propyl acetate, this work; - ‚ - ‚ ‚, propyl acetate, Domalski et al.;8 2, octyl acetate, this work; O,
dimethyl sulfoxide, this work; ), dimethyl sulfoxide, Domalski et al.;8 [, diethyl phthalate, this work; s, diethylphthalate, Domalski et
al.;8 ×, diethyl carbonate, this work; ‚ ‚ ‚, diethyl carbonate, Domalski et al.8
1640 Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001

Table 1. Experimental Molar Saturated Heat Capacity Data csat for 1,2-Dimethoxybenzene, 2,5,8,11-Triethylene Glycol
Dimethyl Ether, N-Ethyl-2-pyrrolidone, Isopentyl Acetate, 2-Ethylhexanal, Toluene, 1-Phenylethanol, Propyl Acetate,
Octyl Acetate, Dimethyl Sulfoxide, Diethyl Phthalate, and Diethyl Carbonate Measured Using the Three-Step Method
T/K csat/J‚mol-1‚K-1 T/K csat/J‚mol-1‚K-1 T/K csat/J‚mol-1‚K-1 T/K csat/J‚mol-1‚K-1
1,2-dimethoxybenzene triethylene glycol dimethyl ether N-ethyl-2-pyrrolidone isopentyl acetate
310.02 280.98 310.07 365.50 310.04 205.42 310.03 256.57
315.00 283.87 315.06 367.10 315.02 207.41 315.01 258.80
320.05 286.03 320.04 368.60 320.07 208.99 320.06 260.43
325.00 288.04 325.07 371.00 325.01 210.60 325.01 262.27
330.06 290.18 330.05 372.70 330.07 212.16 330.06 265.07
335.04 292.60 335.03 375.19 335.05 214.47 335.05 266.73
340.02 294.20 340.02 376.00 340.04 216.26 340.03 269.00
345.02 295.90 345.01 378.29 345.03 218.09 345.02 271.23
350.01 297.22 350.00 378.77 350.02 219.53 350.01 272.62
355.08 298.98 355.07 380.90 355.01 221.59 355.00 275.04
360.07 300.63 360.07 382.73 360.00 223.52 360.07 277.22
365.06 302.04 365.05 384.48 365.08 225.51 365.07 279.31
370.05 303.81 370.04 387.00 370.07 227.63 370.05 281.86
375.04 304.68 375.03 388.87 375.05 229.44 375.04 283.79
380.03 305.82 380.02 390.49 380.04 231.21 380.03 285.49
385.01 307.16 385.01 393.02 385.03 233.02 385.01 287.83
390.08 308.39 390.08 395.75 390.01 234.84
395.06 310.01 395.06 398.32 395.08 236.84
400.04 311.74 400.05 400.78 400.06 239.29
405.02 312.89 405.03 402.70 405.04 241.03
410.00 315.25 410.00 405.88 410.02 243.62
415.06 317.06 415.06 407.90 415.08 245.37

2-ethylhexanal toluene 1-phenylethanol propyl acetate

310.05 272.67 310.02 159.66 310.06 266.85 310.08 198.18
315.02 274.69 315.04 161.59 315.04 270.78 315.06 199.86
320.08 276.32 320.03 163.25 320.01 273.91 320.03 201.31
325.02 277.93 325.00 164.90 325.03 276.94 325.06 202.80
330.07 279.59 330.07 166.76 330.00 280.19 330.03 204.60
335.06 282.51 335.07 168.19 335.07 281.86 335.01 205.92
340.04 284.52 340.06 170.03 340.06 285.19 340.08 207.71
345.03 286.67 345.06 171.80 345.05 287.39 345.07 208.79
350.03 288.23 350.05 172.96 350.04 288.84 350.06 211.01
355.01 290.51 355.21 175.02 355.03 290.67 355.06 212.55
360.01 292.95 360.04 176.60 360.02 292.43 360.05 214.09
365.08 295.19 365.03 178.22 365.04 215.74
370.07 298.01 370.02 180.08 370.03 217.51
375.07 299.72 375.02 219.01
380.05 301.25 380.00 220.48

octyl acetate dimethyl sulfoxide diethyl phthalate diethyl carbonate

310.03 347.47 310.04 153.81 311.29 371.48 310.00 221.18
315.01 350.69 315.02 154.67 315.02 373.92 315.06 223.35
320.06 352.97 320.06 155.24 320.07 376.56 320.03 225.09
325.01 355.62 325.01 155.75 325.02 379.14 325.06 226.49
330.06 358.78 330.07 156.20 330.08 382.31 330.03 229.02
335.05 360.94 335.05 156.48 335.06 383.71 335.02 230.39
340.03 364.13 340.04 157.10 340.05 388.58 340.00 232.42
345.02 367.02 345.03 157.64 345.04 391.15 345.08 234.30
350.01 369.11 350.02 158.02 350.03 393.40 350.07 235.66
355.00 372.32 355.01 158.78 355.02 396.69 355.06 237.81
360.08 375.39 360.00 159.40 360.02 399.91 360.05 239.73
365.06 378.29 365.08 160.06 365.00 402.70 365.04 241.46
370.05 381.55 370.06 160.89 370.08 406.38 370.03 243.52
375.0 384.19 375.05 161.45 375.07 409.05 375.02 244.12
380.03 387.01 380.04 161.95 380.05 411.70 380.01 246.70
385.01 390.16 385.03 162.64 385.04 415.26 385.07 247.90
390.08 393.29 390.01 163.34 390.02 417.92 390.06 249.30
395.07 396.18 395.08 164.01 395.01 420.15 395.05 250.97
400.05 399.44 400.06 164.75 400.07 421.58 400.03 253.90
405.03 402.44 405.04 165.25 405.06 424.20 405.01 256.13
410.00 406.02 410.02 165.96 410.04 427.02 410.07 257.90
415.06 408.29 415.07 166.63 415.01 431.22
420.02 412.30

Domalski in the valid temperature range.1,2 The results are values are listed. Column 5 displays the average deviation
shown in Table 2. from the recommended values of Domalski et al.8 together
Column 1 of Table 2 displays the temperature range with the temperature range covered by the parameters of
covered by the experimental data. In columns 2-4 the the recommended values listed in column 6. The compari-
average absolute relative deviations, given by 100/ son with the recommended values of Domalski et al.8 shows
npΣ|(csat,exptl - csat,calcd)/csat,exptl| (whereby np is the number that the deviations are small (<0.2%) for toluene, dimethyl
of points), between the calculated and the experimental sulfoxide, and diethyl phthalate. Domalski et al.8 state that
 Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001 1641

Table 2. Comparison of the Experimental Data with Results of the Group Contribution Method of Missenard,3 Ruzicka
and Domalski,1,2 and Luria and Benson4 and Recommended Values
av deviation av deviation av deviation av deviation temp range
temp range of of of Ruzicka of Luria and of Domalski covered by
exptl data/K Missenard/% and Domalski/% Benson/% (recommended values)/% recommended values/K
1,2-Dimethoxybenzene
310-415 n.a. 3.28 n.a. n.a. n.a.
Triethylene Glycol Dimethyl Ether
310-415 4.02 1.23 n.a. 0.52 230-350
N-Ethylpyrrolidone
310-415 n.a. n.a. n.a. n.a. n.a.
Isopentyl Acetate
310-385 n.a. 0.2 n.a. n.a. n.a.
2-Ethylhexanal
310-380 n.a. 3.03 n.a. n.a. n.a.
Toluene
310-370 1.70 0.28 1.02 0.20 310-370
1-Phenylethanol
310-360 9.71 n.a. n.a. n.a. n.a.
Propyl Acetate
310-380 n.a. 1.43 n.a. 1.78 298-338
Octyl Acetate
310-420 n.a. 0.32 n.a. n.a. n.a.
Dimethyl Sulfoxide
310-415 n.a. n.a. n.a. 0.08 296-348
Diethyl Phthalate
311-415 n.a. 0.34 n.a. 0.20 273-353
Diethyl Carbonate
310-410 n.a. n.a. n.a. 0.82 295-359

the parameters for these compounds agree with the ex- Table 3. Comparison of the Experimental Data of
Toluene in This Work and by Becker et al.9 with
perimental data within (0.5%, but for triethylene glycol Recommended Values of Domalski et al.8
dimethyl ether, diethylene carbonate, and especially propyl
acetate the deviations are larger. However, for all of these csat/J‚mol-1‚K-1
compounds Domalski et al.8 give an error of (3%. This this Becker, recommended
means that the deviation of the experimental values in T/K work 2000 values deviation/%
comparison to the recommended values of Domalski et al.8 310.02 159.66 160.51 0.529
is within the uncertainty of the parameters. The deviations 312.05 161.01 161.11 0.061
for the three group contribution methods are quite differ- 315.04 161.59 161.99 0.248
ent. Only three components can be calculated by using the 317.00 162.44 162.58 0.085
320.03 163.25 163.49 0.148
Missenard3 method, and also four components cannot be 321.95 164.05 164.07 0.011
calculated with the Ruzicka and Domalski1,2 method, 325.00 164.90 165.00 0.059
although this is a quite new group contribution method 326.95 165.81 165.59 -0.130
with a large assortment of groups. The group contribution 330.07 166.76 166.56 -0.124
method of Luria and Benson4 can be used only for hydro- 331.95 167.12 167.14 0.013
carbons, for example, toluene. The deviations for the 335.07 168.19 168.12 -0.041
336.95 169.00 168.71 -0.172
Missenard3 method are quite large. In particular, the 340.06 170.03 169.70 -0.194
deviations for toluene are remarkably high, although a 345.06 171.80 171.30 -0.293
great number of data for this component are available. The 350.05 172.96 172.92 -0.027
average deviations of the Ruzicka and Domalski1,2 method 355.21 175.02 174.61 -0.230
are all within reasonable limits. The deviations for propyl 360.04 176.60 176.22 -0.215
acetate and isopentyl acetate are especially small consider- 365.03 178.22 177.90 -0.177
370.02 180.08 179.60 -0.265
ing that these components were not included during the
development of the group contribution method.
Also, the reproducibility of the results is an important capacity measurement above 100 K with a SETARAM
indicator for the quality of the measured data. For this Tian-Calvet calorimeter (BT type) found by Handa et al.10
reason (and to check the calibration) toluene was measured
Conclusion
in this work a second time. Table 3 shows the experimental
data of toluene measured in this work together with data Liquid molar heat capacities of 12 organic substances
published by Becker et al.,9 both measured with the Tian- were measured from 310 K to 420 K. The experimental
Calvet calorimeter. In column 4 the relative deviations, data were compared with recommended values of Domalski
given by 100(csat,lit. - csat)/csat,lit., from the recommended et al.8 and with data calculated using the group contribu-
values of Domalski et al. and the experimental data are tion methods of Ruzicka and Domalski,1,2 Missenard,3 and
shown. In both cases the deviations are within (0.3%. This Luria and Benson.4 Additionally, the reproducibility of the
demonstrates the high reproducibility of the heat capacity Tian-Calvet calorimeter was verified. Comparison with the
measurements obtained by the Tian-Calvet calorimeter. recommended values of Domalski et al.8 (as far as data are
This result also confirms the precision of (0.5% in heat available) shows the high accuracy of the heat capacity
1642 Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001

measurements. Comparison of the group contribution (2) Ruzicka, V.; Domalski, E. S. Estimation of the Heat Capacities
methods with the experimental data shows that the method of Organic Liquids as a Function of Temperature Using Group
Additivity. II. Compounds of Carbon, Hydrogen, Halogens, Ni-
of Missenard is not suitable to predict the heat capacities trogen, Oxygen, and Sulfur. J. Phys. Chem. Ref. Data 1993, 22
of the measured components. The main problem with the (3), 619-650.
Missenard method is that the number of structural groups (3) Missenard, F. A. Methode Additive pour la Determination de la
Chaleur Molaire de Liquides. C. R. Acad. Sci. Paris 1965, 260,
is very limited. Only three of the measured components 5521-5523.
could be calculated. The deviations of the method of (4) Luria, M.; Benson, S. W. Heat Capacities of Liquid Hydrocarbons.
Ruzicka and Domalski, on the other hand, are within Estimation of Heat Capacities at Constant Pressure as a Tem-
perature Function, Using Additivity Rules. J. Chem. Eng. Data
reasonable limits. For propyl acetate and isopentyl acetate 1977, 22, 90-100.
the deviations are remarkably small. However, also for this (5) Calvet, E.; Prat, H. Recent Progress in Microcalorimetry; Perga-
method the necessary parameters for 4 of the 12 compo- mon Press: Oxford, U.K., 1963.
(6) van Miltenburg, J. C.; van den Berg, G. J. K.; van Bommel, M. J.
nents are not available. Comparison of the heat capacities Construction of an Adiabatic Calorimeter, Measurements of the
measured in this work and by Becker et al.9 show that the Molar Heat Capacity of Synthetic Sapphire and of n-Heptane. J.
reproducibility of the Tian-Calvet calorimeter is very high. Chem. Thermodyn. 1987, 19, 1129-1137.
(7) Höhne, G.; Hemminger, W.; Flammersheimer, H.-J. Differential
This means that Tian-Calvet calorimetry is a very effective Scanning Calorimetry, An Introduction for Practitioners; Springer-
technique to measure heat capacities. Heat capacity mea- Verlag: Berlin, Germany, 1996.
surements are necessary not only for the design of chemical (8) Domalski, E. S.; Zabransky, M.; Ruzicka, V.; Majer, V. Heat
processes but also for the further development of group Capacity of Liquids: Volume I + II, Critical Review and Recom-
mended Values. J. Phys. Chem. Ref. Data 1996, Monograph 6.
contribution methods. Comparison of the experimental data (9) Becker, L.; Aufderhaar, O.; Gmehling, J. Measurement of Heat
with the three group contribution methods shows that for Capacities for Nine Organic Substances by Tian-Calvet Calorim-
all prediction methods the number of available structural etry. J. Chem. Eng. Data 2000, 45, 661-664.
(10) Handa, Y. P.; Hawkins, R. E.; Murray, J. J. Calibration and
groups is very limited. This demonstrates that an extension Testing of a Tian-Calvet Heat-Flow Calorimeter, Enthalpies of
of these methods is desirable. Fusion and Heat Capacities for Ice and Tetrahydrofuran Hydrate
in the Range 85 to 270 K. J. Chem. Thermodyn. 1984, 16, 623-
Literature Cited 632.

(1) Ruzicka, V.; Domalski, E. S. Estimation of the Heat Capacities

of Organic Liquids as a Function of Temperature Using Group Received for review June 18, 2001. Accepted September 19, 2001.
Additivity. I. Hydrocarbon Compounds. J. Phys. Chem. Ref. Data
1993, 22 (3), 597-618. JE010177O