Fluid Phase Equilibria 303 (2011) 205–216

Contents lists available at ScienceDirect

Fluid Phase Equilibria

journal homepage: www.elsevier.com/locate/fluid

Recommended vapor pressures for thiophene, sulfolane, and dimethyl sulfoxide

Michal Fulem a,b,∗ , Květoslav Růžička a , Martin Růžička a
a
Department of Physical Chemistry, Institute of Chemical Technology, Prague, Technická 5, CZ-166 28 Prague 6, Czech Republic
b
Department of Semiconductors, Institute of Physics, Academy of Sciences of the Czech Republic, v. v. i., Cukrovarnická 10, CZ-162 00 Prague 6, Czech Republic

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

Article history: Recommended vapor pressure data for important industrial solvents, thiophene (CAS RN: 110-02-1),
Received 14 October 2010 sulfolane (CAS RN: 126-33-0), and dimethyl sulfoxide (CAS RN: 67-68-5), were developed by the simul-
Received in revised form 1 February 2011 taneous correlation of vapor pressure and related thermal data (heat capacities of condensed phases,
Accepted 3 February 2011
ideal gas heat capacities and calorimetrically determined enthalpies of vaporization). For sulfolane and
dimethyl sulfoxide, new vapor pressure data were obtained using the static method in the temperature
Keywords:
interval from 273 to 308 K. Liquid heat capacities and calorimetric enthalpies of vaporization were taken
Thiophene
from the literature and/or determined by Calvet calorimetry. The thermodynamic properties in the ideal
Sulfolane
Dimethyl sulfoxide
gaseous state were calculated using the methods of statistical thermodynamics based on experimental as
Vapor pressure well as calculated fundamental vibrational frequencies and molecular structure data. Comparisons with
Heat capacity literature values are shown for all measured and derived properties.
Ideal gas thermodynamic properties © 2011 Elsevier B.V. All rights reserved.
Vaporization enthalpy
Recommended vapor pressure equation

1. Introduction sulfolane, and dimethyl sulfoxide were developed by the simulta-

neous correlation of selected vapor pressure and related thermal
Thiophene, sulfolane, and dimethyl sulfoxide are important data [1].
industrial solvents produced on a large scale. Despite of this fact,
their vapor pressures at ambient temperatures are not known with 2. Simultaneous treatment of vapor pressures and related
sufficient accuracy. Reliable vapor pressure data are indispensible thermal data (SimCor method)
for processes involving phase equilibria and for phase equilibrium
studies on systems containing these compounds. g
Vapor pressure p, enthalpy of vaporization  Hm and the dif-
In this work, we first assessed all available literature vapor ference between ideal gas heat capacity and heat capacity of liquid
pressure data and examined their consistency with calorimetri- g 0 g0 
 Cp,m = Cp,m − Cp,m are linked by exact thermodynamic relation-
cally determined vaporization enthalpies and heat capacities of ships
condensed phases and ideal gas. Literature review and consis-
 d ln p  g
 Hm
tency tests revealed that new vapor pressure data in the low
RT 2 = g = H  (1)
pressure region for sulfolane and dimethyl sulfoxide and new liq- dT sat  z
uid heat capacities for sulfolane were needed. These data were
experimentally determined in this work. Also, heat capacities of  dH    
d[T 2 (d ln p/dT )]
ideal gas and thermodynamic properties in the ideal gaseous =R = C 
state for all the studied compounds were calculated using the dT sat dT
sat
methods of statistical thermodynamics. We employed both the 0g
=  Cp,m + pVT correction (2)
experimental and calculated fundamental vibrational frequencies
and compared the obtained results as part of our effort to assess where subscript ‘sat’ denotes a derivative along the saturation line,
the reliability of the calculations based purely on ab initio data. g
R is the molar gas constant (R = 8.314472 J K−1 mol−1 [2,3]),  z
Consequently, recommended vapor pressure data for thiophene, stands for the difference between the compressibility factors of the
coexisting phases. H and C are auxiliary quantities which can
be calculated using Eqs. (1) and (2) either from the vapor pres-
∗ Corresponding author at: Department of Physical Chemistry, Institute of Chem- sure correlating equation (by substituting the derivative dln p/dT
ical Technology, Prague, Technická 5, CZ-166 28 Prague 6, Czech Republic. into Eqs. (1) and (2)) or from experimental values of enthalpy of
g g 0
E-mail addresses: michal.fulem@vscht.cz, fulemm@vscht.cz (M. Fulem). vaporization  Hm and  Cp,m calculated from experimental data

0378-3812/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.fluid.2011.02.002
206 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216
g0
by combining heat capacity of ideal gas Cp,m and calorimetric val-
 . The possibility to calculate H
ues of heat capacity of liquid Cp,m
and C both from the experimental thermal data and vapor pres-
sure data and from a vapor pressure correlating equation means
that after selecting a suitable relationship describing p vs. T it is
possible to correlate simultaneously experimental vapor pressures
g
p, enthalpies of vaporization  Hm and heat capacity difference
g
0
 Cp,m as a function of temperature.
The Cox equation was found to be the most adequate for describ-
ing simultaneously vapor pressure and related thermal data as a
function of temperature down to the triple point [1]. The Cox equa-
tion has the form [4]
   n
upper limits are 133 and 1333 Pa, respectively. The temperature of
p T 0 /K i the pressure sensors is kept at T = 318 K by an internal temperature
ln = 1− exp Ai (T/K) , (3)
p0 T/K
i=0
where p is the vapor pressure, T is the temperature, T0 and p0 is the
temperature and pressure of an arbitrarily chosen reference point
and Ai are correlation parameters. n = 2 is usually adequate; less
or more parameters can be used when temperature range is too
g 0
narrow or too wide or when temperature dependence of  Cp,m is
complex.
Equations analogous to those for the equilibrium between the
liquid and vapor phases can be derived for the equilibrium between
the solid and vapor phases. The SimCor method was thoroughly
tested in our laboratory [1] and used to obtain recommended vapor
pressure for several crystalline and liquid compounds [5–12].
3. Experimental
3.1. Materials
Sulfolane was obtained from Novasol Belgium. The mass fraction
purity (determined by gas gromatography (GC)) and mass frac-
tion of water (determined by Coulometric titration, UOP 481) were
w = 0.9985 and w(H2 O) = 3 × 10−4 , respectively, as stated in the cer-
tificates of analysis provided by the supplier. The sample was stored
over 4 Å molecular sieves for approximately 2 weeks. The mass
fraction of water determined prior to taking vapor pressure and
heat capacity measurements was w(H2 O) = 6.5 × 10−5 (determined
by Coulometric titration with Metrohm 831). Dimethyl sulfoxide
was purchased from Aldrich. The mole fraction purity (determined
by GC) and mass fraction of water (determined by Coulomet-
ric titration) were x = 0.9999 and w(H2 O) = 2.7 × 10−5 , respectively,
as stated in the certificates of analysis provided by the supplier.
Considering the assessed purities, the compounds were studied
without further purification.
3.2. Vapor pressure measurements
Prior to vapor pressure measurements, the samples were thor-
oughly degassed using an apparatus shown in Fig. 1. The samples
were directly pumped with a turbomolecular vacuum pump for 2
weeks at approximately 298 K. The sample loss was prevented by
cooling the upper part of the measuring cell with Peltier thermo-
electric modules.
Vapor pressure measurements were performed using the static
method with an apparatus internally denoted as STAT6, which was
previously described in detail [13]. Therefore, only a short descrip-
tion is presented here. The apparatus is constructed of stainless
steel internally electrochemically polished tubing with ConFlat DN
16 CF and VCR connections and all-metal, pneumatically operated,
angle valves VAT series 57 (VAT Vacuumvalves AG, Switzerland) for
UHV. The pressure is measured simultaneously by two capacitance
diaphragm absolute gauges MKS Baratron 690A01TRA and MKS
Fig. 1. An apparatus for sample degassing.
Baratron 690A11TRA (MKS Instruments Inc., USA). Their measuring
controller which constrains the upper temperature limit of vapor
pressure measurements to about 310 K. Calibrations of the pres-
sure gauges at 318 K performed by the manufacturer at 11 equally
spaced pressures from 0 to 130 Pa and from 0 to 1300 Pa, respec-
tively, with a maximum relative deviation of 0.03%, are traceable
to the National Institute of Standards and Technology (NIST). Addi-
tional calibrations focusing on the beginning of a pressure span
(p < 10 Pa) were performed by the Czech Metrology Institute. The
resulting relative uncertainty of the pressure reading is less than
0.05% which is in accordance with the uncertainty stated by the
manufacturer. The sample temperature is measured by a platinum
resistance thermometer Burns Engineering 12001-A-12-6-2-A in
a four-wire connection calibrated by the manufacturer at the ice
point and/or by comparison to standard platinum resistance ther-
mometer (SPRT). This SPRT was calibrated to the ITS-90, and its
calibration was traceable to NIST. The uncertainty of the mea-
surement of sample temperature is 0.02 K, which is insignificant
in the pressure range investigated in this work. All temperatures
reported here are based on the international temperature scale ITS-
90. The performance of the STAT6 apparatus was checked by the
measurement with naphthalene, which is recommended for cali-
brating vapor pressure apparatuses [11]. The agreement with the
recommended data [11] was within the uncertainty of vapor pres-
sure measurement with the STAT6 apparatus, which is adequately
described by the expression (p/Pa) = 0.005(p/Pa) + 0.05.
3.3. Heat capacity measurements
The ␮DSC IIIa calorimeter (Setaram, France) was used for the
heat capacity determination of sulfolane in the temperature range
from 304 to 355 K. The measurements were carried out in the
incremental temperature scanning mode (step method) [14] with
a number of 5 K steps and a heating rate of 0.5 K min−1 followed by
isothermal delays of 3600 s. The mass of the sample was approxi-
mately 1 g. The relative uncertainty of heat capacity measurements
performed using the step method is estimated to be less than 1%. A
detailed description of the calorimeter and calibration can be found
in a recent paper by Straka et al. [15].
Three successive runs were performed in order to obtain heat
capacity of sulfolane. The measuring cell was empty in the first run
and filled with the reference material and the measured sample in
the second and third runs, respectively. Synthetic sapphire, NIST
SRM 720, was used as the reference material. The reference cell
was empty during all runs. Heat capacity was calculated from the
equation
msapp cp,sapp (As − AE )
csat,s = (4)
ms (Asapp − AE )
where csat,s is the saturation specific heat capacity of the measured
sample; cp,sapp is the specific heat capacity of the reference sub-
stance (synthetic sapphire); ms is the mass of the sample; msapp is
 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216 207

Table 1 Table 2
Experimental vapor pressures p for liquid sulfolane. Experimental vapor pressures p for dimethyl sulfoxide.

T/K p/Pa T/K p/Pa Crystalline phase Liquid phase

a
300.16 1.24 305.17 1.87 T/K p/Pa T/K p/Pa
300.16 1.26a 305.17 1.88
301.16 1.34a 305.17 1.87 273.65 7.89 288.15 37.63a
301.16 1.34a 306.17 2.03 273.65 7.89 288.15 37.62a
301.16 1.34a 306.17 2.03 273.65 7.89 288.15 37.62a
302.16 1.46 306.17 2.03 275.65 9.79 290.65 45.65a
302.16 1.46 307.17 2.23 275.65 9.79 290.65 45.65a
302.16 1.46 307.17 2.24 275.65 9.79 290.65 45.65a
303.17 1.55 307.17 2.20 278.15 12.78 293.15 55.12
303.17 1.58 308.17 2.41 278.15 12.77 293.15 55.09
303.17 1.58 308.17 2.41 278.15 12.75 293.15 55.10
305.17 1.88 308.17 2.40 280.64 16.55 295.65 66.31
280.65 16.56 295.65 66.32
a
Subcooled liquid. 280.65 16.56 295.65 66.33
283.15 21.38 298.15 79.65
283.15 21.37 298.15 79.66
the mass of the reference substance; As is the integrated value of the 285.65 27.52 298.15 79.64
differential heat flow when the measuring vessel contains the sam- 285.65 27.52 300.65 95.18
ple; Asapp is the integrated value of the differential heat flow when 285.65 27.51 300.65 95.17
288.15 35.31 300.65 95.17
the measuring vessel contains the reference substance; and AE is
288.15 35.25 303.16 113.38
the integrated value of differential heat flow in the case when the 288.15 35.25 303.16 113.40
measuring vessel is empty. The temperature dependence of heat 290.65 44.87 303.16 113.37
capacity for sulfolane was approximately linear. Thus, the average 290.65 44.87 305.66 134.57
heat capacity over individual temperature steps can be considered 290.65 44.86 305.66 134.59
305.66 134.61
as a true heat capacity relating to the mean temperature of the
308.16 159.24
interval. The correction for sample vaporization, as discussed in 308.16 159.22
Zábranský et al. [16], was not applied. Any correction made would 308.16 159.21
be much smaller than the experimental uncertainty of our mea- a
Subcooled liquid.
surements themselves and can be neglected. The correction does
not exceed 0.1% even for much more volatile compounds when the
volume of the condensed phase is about 3/4 of the total volume
of the measuring cell [15,17]. The saturated molar heat capaci-
ties Csat,m obtained in this work are identical to isobaric molar
heat capacities Cp,m in the temperature range studied as it is not
necessary to make a clear distinction between Cp,m along the satu-
ration curve and Csat,m below 0.9 Tb , where Tb is the normal boiling
temperature [16].
In addition to the measurements performed using the step
method, we performed experiments using the continuous method
in the temperature range from 253 to 350 K to investigate the phase
behavior of sulfolane. The measurements were performed with
both heating and cooling rate ranging from 0.05 to 0.5 K min−1 .

4. Results and discussion

4.1. Vapor pressures

The vapor pressure measurements of sulfolane and dimethyl

sulfoxide were performed in the temperature interval from 253 to
308 K by varying the temperature at random to detect systematic
errors caused by insufficient degassing of the sample. The experi-
ments were carried out repeatedly at selected temperatures. When
the pressure did not change with the number of measuring cycles,
the sample was considered completely degassed, and the final
set of data was recorded. At least two experimental points were
obtained for each temperature. The experimental vapor pressures
obtained in this work are given in Table 1 (sulfolane) and Table 2
(dimethyl sulfoxide). Our calorimetric measurements and several
literature studies [18–20] revealed that sulfolane can form several
solid phases in a narrow temperature range below its melting tem-
perature. The phase transitions were very difficult to separate on
a DSC thermogram. Furthermore, the thermal history and exper-
imental conditions including the mass of the sample significantly
affected the phase behavior of sulfolane as demonstrated in Fig. S1.
Therefore, it was not obvious which crystalline phase was formed
after cooling sulfolane below its melting temperature during the
Fig. 2. Arc representation for vapor pressure data of thiophene: () Milazzo [58];
() Waddington et al. [22] (static method); ( ) Waddington et al. [22] (ebulliom-
etry); () White et al. [59]; (夽) Eon et al. [60]; (䊉) Daubert and Hutchison [62]
(ebulliometry); ( ) Daubert and Hutchison [62] (static method); ( ) Evlanov [74].
The data points are connected with lines to show the trend.
vapor pressure experiments and thus to which phase the measured
vapor pressure data corresponded. Consequently, only the vapor
pressure data in the liquid state are given in Table 1. As sulfolane
undercools readily below its melting temperature, as evidenced by
both the vapor pressure and calorimetric measurements, Table 1
also includes data points corresponding to subcooled liquid. For
dimethyl sulfoxide, the vapor pressure data for both crystalline and
liquid phases are reported in Table 2.
Literature data on vapor pressure for thiophene, sulfolane and
dimethyl sulfoxide listed in Table 3 were first critically assessed
using the arc representation [21] (see Figs. 2–4). This allowed us
208 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216

Table 3
Overview of the literature vapor pressure data for thiophene, sulfolane and dimethyl sulfoxide.a , b

Reference Nc (Tmin − Tmax )/K (pmin − pmax )/kPa  T /Kd  p /Pae Purity/% Method

Thiophene
Milazzo [58] 10 228.20–289.40 0.107–6.599 nosp.f nosp. nosp. Static
Waddington et al. [22] 14 273.15–313.15 2.854–20.736 0.01 8 99.96 mol% Static
Waddington et al. [22] 15 312.21–392.94 19.920–270.111 0.01 13 99.96 mol% Ebulliometry
White et al. [59] 20 299.92–366.38 11.607–133.077 0.008 nosp. 99.90 mol% Ebulliometry
Eon et al. [60] 4 333.45–363.45 44.930–126.790 0.05 13 99 + % Static
Triday and Rodriguez [61] 3 308.15–318.15 16.865–25.638 0.1 66 nosp. Static
Daubert and Hutchison [62] 13 265.35–347.95 1.647–74.914 0.1 7 99 + % Ebulliometry
Daubert and Hutchison [62] 18 346.14–527.98 73.153–3613.8 0.1 3500 99 + % Static
Sulfolane
Benoit and Charbonneau [63] 3 303.15–323.15 0.003–0.011 nosp. nosp. nosp. Static
Ashcroft et al. [64] 6 303.15–328.15 0.009–0.021 0.02 3 nosp. Static
Shcherbina et al. [65] 40 364.45–528.15 0.200–54.00 0.1 13 nosp. Static
Negadi et al. [66] 10 307.71–362.29 0.003–0.105 0.02 0.2% 99 Static
Steele et al. [29] 15 423.82–542.06 2.000–70.128 0.01 0.1–3 99.95 Ebulliometry
This work 24 300.16–308.17 0.001–0.002 0.02 0.005p + 0.05 99.85 Static
Dimethyl sulfoxide
Douglas [67] 7S 293.15–323.15 0.056–0.409 0.03 1% 99 Gas saturation
Heinrich and Surový [56] 24 334.45–464.95 0.733–104.325 0.1 7 nosp., w(H2 O) = 4 × 10−4 Ebulliometry
Jose et al. [68] 21 293.14–343.15 0.055–1.255 0.01 nosp. nosp. Static
Jakli and Van Hook [57] 33 325.5–442.1 0.460–56.342 nosp. 7 nosp. Static
Nishimura et al. [69] 26 300.25–423.35 0.089–31.930 nosp. nosp. 99.5 + Static
Campbell [70] 20 317.30–462.00 0.013–10.413 nosp. 13 nosp. Static
Ul’yanov et al. [71] 10 322.80–389.30 0.400–13.589 nosp. nosp. nosp. Ebulliometry
Negadi et al. [72] 13 293.10–363.08 0.056–3.266 0.01 5% 99 + Static
Kaporalova and Chaikin [73] 11 294.75–463.45 0.080–101.192 nosp. 7 nosp. Static
This work 50 273.65–308.16 0.001–0.159 0.02 0.005p + 0.05 99.99 Static
a
The data from references written in bold were used in the SimCor method.
b
The references reporting single or two vapor pressure points obtained for example as part of the synthesis or VLE measurements on multi-component mixtures are not
considered.
c
N stands for number of experimental points.
d
 T stands for uncertainty in temperature.
e
 p stands for uncertainty in pressure.
f
nosp. stands for not specified.

to reject obvious outliers that were distant from the rest of the
data. Consequently, the consistency of the vapor pressure data with
ideal gas heat capacities and heat capacities of condensed phases
was tested. The selected vapor pressure data used in the SimCor
method are given in bold in Table 3.

Fig. 4. Arc representation for vapor pressure data of dimethyl sulfoxide in the liquid
state: () Douglas [67]; () Heinrich and Surový [56]; ( ) Jose et al. [68]; () Jakli
and van Hook [57]; (夽) Nishimura et al. [69]; ( ) Ul’yanov et al. [71]; ( ) Negadi
et al. [72]; () Kaporalova and Chaikin [73]; (䊉) this work. Data by Cambell [70]
are not displayed due to large deviations from other data sets. The data points are
connected with lines to show the trend.

4.2. Heat capacity of condensed phases

Literature data on heat capacities of condensed phases for thio-

Fig. 3. Arc representation for vapor pressure data of sulfolane: () Benoit and Char-
phene, sulfolane, and dimethyl sulfoxide are summarized in Table 4.
bonneau [63]; () Ashcroft et al. [64]; ( ) Shcherbina et al. [65]; () Negadi et al.
[66]; (夽) Steele et al. [29]; (䊉) this work. The data points are connected with lines
The selected heat capacity data used in the SimCor method are
to show the trend. given in bold in Table 4.
 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216 209

Table 4
Overview of the literature heat capacities of liquid for thiophene, sulfolane and dimethyl sulfoxide.a

Reference N (Tmin − Tmax )/K  C /%b Purity/% Method

Thiophene
Waddington et al. [22] 16 240.3–336.1 0.20 99.955 Adiabatic
Figuière et al. [23] 31 238.0–299.9 0.30 99.94 Adiabatic
Sulfolane
Castanoglo et al. [28] 1 303.1 1 nosp.c Calvet
Steele et al. [29] 13 315.1–555.1 1 99.95 DSC
Ho et al. [30] 11 303.2–353.2 1 99+ DSC
Mundhwa et al. [31] 11 303.2–353.2 99 Calvet
This work 22 304.0–355.1 1 99.85 Calvet
Dimethyl sulfoxided
Clever and Westrum [25] 12 295.8–348.0 0.15 99.96 Adiabatic
Becker and Gmehling [26] 22 310.0–415.1 0.50 99.95 Calvet
a
The data from references written in bold were used in the SimCor method.
b
 C stands for relative uncertainty in heat capacity.
c
nosp. stands for not specified.
d
References containing heat capacity at single temperature are not listed (they can be found in data collection by Zábranský et al. [24]).

Table 5

Experimental molar liquid heat capacities Cp,m of sulfolane.

First run Second run

−1 −1 −1
T/K 
Cp,m /(J K mol ) T/K 
Cp,m /(J K−1 mol )

304.04 179.5 304.04 179.9

309.15 181.6 309.14 181.3
314.25 182.6 314.25 183.8
319.35 184.7 319.35 184.5
324.46 185.3 324.46 184.4
329.56 187.4 329.56 186.6
334.66 188.7 334.67 187.9
339.77 189.7 339.77 188.5
344.88 190.2 344.88 190.2
349.98 192.5 349.98 192.0
355.09 193.4 355.09 192.3

For thiophene, the recommendation by Zábranský et al. [16] is and recommended values by Zábranský et al. [27] deviate by 1–2%
based on data by Waddington et al. [22] and Figuière et al. [23]. from Eq. (5) in the temperature range studied. The deviations are
Mutual agreement of the two data sets is within 1% and the uncer- systematically positive, except those of the data by Castagnolo
tainty assigned by Zábranský et al. [16] is 0.5%. The heat capacities et al. [28], which might be attributed to a higher water content
of thiophene were used in the temperature range from 240 to 340 K of sulfolane samples used in the previous studies [29–31]. No
in the SimCor method. For dimethylsulfoxide, a recent recommen- information about the water content is given in these studies.
dation by Zábranský et al. [24] is based on data by Clever and The water content w(H2 O) of the sample used in [28] was less
Westrum [25] and Becker and Gmehling [26]. These two datasets than 7 × 10−5 . For the SimCor method, the liquid heat capacities
are in very good agreement in the temperature range used in the
SimCor method (295–345 K) and the assigned uncertainty is 0.5%.
The heat capacities of crystalline dimethyl sulfoxide were taken
from [25] and used in the temperature range from 200 to 250 K
in the SimCor method. For sulfolane, the critical review from 2001
[27] contains four sources and the recommendation is based on
data by Castagnolo et al. [28] and Steele et al. [29]. The uncertainty
assigned by Zábranský et al. [27] is 1%. We found new liquid heat
capacities reported by Ho et al. [30] and by Mundhwa et al. [31],
which are not covered by the recommendation from 2010 [24].
As the mutual agreement of the abovementioned datasets is only
moderate, new measurements with carefully dried sulfolane sam-
ple were made. Liquid heat capacities of sulfolane determined in
this work are given in Table 5. The experimental isobaric molar heat
capacities for sulfolane in the temperature range from 304 to 355
were fitted with a polynomial expression using the least-squares
method

Cp,m /(J K−1 mol−1 ) = 3.67592 + 6.56725 × 10−1 (T/K)

− 6.10893 × 10−4 (T/K)

2
(5)

Fig. 5. Comparison of experimental liquid heat capacity Cp,m for sulfolane with
,calc
smoothed values Cp,m obtained from Eq. (5) in the temperature range studied:
The liquid heat capacities obtained in this work are compared (䊉) this work; () Mundhwa et al., [31]; (夽) Ho et al. [30]; ( ) Steele et al. [29]; (+)
with the literature values in Fig. 5. The literature data [28–31] Castagnolo et al. [28]; (−) Zábranský et al. [27].
210 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216

obtained in this work were used in the temperature range from

304 to 355 K.

4.3. Calorimetrically determined vaporization enthalpies

Calorimetrically determined enthalpy of vaporization was

found only for thiophene in the literature. Waddington et al. [22]
reported three values in the temperature range from 318 to 357 K.
These data were used in the SimCor method. Majer and Svoboda
[32] assigned the uncertainty less than 0.25% to these data.

4.4. Thermodynamic properties in the ideal gas state

The thermodynamic properties in the ideal gas state were cal-

culated by statistical mechanics using the rigid rotor – harmonic
oscillator (RRHO) approximation with correction for internal rota-
tion in the case of dimethyl sulfoxide. Contributions of internal
rotations of methyl tops were calculated from energy levels of
hindered rotations. We employed both the experimental and cal-
culated fundamental vibrational frequencies and compared the
obtained results as part of our effort to assess the reliability of the
calculations based purely on ab initio data. Molecular geometry
g0 g0,ref
Fig. 6. Comparison of molar ideal gas heat capacities Cp,m of thiophene. Cp,m was
calculated in this work using experimental vibrational frequencies [41]: (䊉) Frenkel
g0
et al. [42]; () calorimetric values by Waddington et al. [22]; () Cp,m calculated
using DFT B3LYP/6–311 + G(2df, p) vibrational frequencies in this work.

optimizations, vibrational frequencies calculations and potential

scans for internal rotations of methyl groups were performed using
the density functional theory (DFT) at the B3LYP/6–311 + G(2df, p)
level of theory with the Gaussian 03 software package [33]. The
basis set 6–311 + G(2df, p) was selected based on our preliminary
unpublished results for 29 rigid molecules and 14 molecules with
one internal rotation for which the comparison with reliable lit-
erature data for the thermodynamic properties in the ideal gas
state, fundamental vibrational frequencies and in some cases bar-
riers to internal rotations is possible. The calculations performed
with basis sets 6–31G(d), 6–31 + G(d, p), 6–31G(2df, p), 6–311 + G(d,
p), 6–311 + G(2df, p) and 6–311 + + G(3df,3pd) suggest that the
results improve with the increasing basis set up to 6–311 + G(2df,
p). The basis set 6–311 + + G(3df,3pd) yield slightly worse results
than 6–311 + G(2df, p). The calculated vibrational frequencies were
not scaled. We do not believe that constant scaling factors, i.e.
wave number and temperature independent, as mostly used today,
can bring any significant improvement. This is demonstrated in
Figs. S2 and S3, where the heat capacity and entropy for thiophene
g0
calculated using scaled and unscaled fundamental frequencies are Fig. 7. Comparison of standard molar gas entropies Sm (p = 105 Pa) of thiophene.
g0,ref
compared with those obtained using experimental frequencies. Sm was calculated in this work using experimental vibrational frequencies [41]:
g0
(䊉) Frenkel et al. [42]; () Sm calculated using DFT B3LYP/6–311 + G(2df, p) vibra-
This opinion is further supported by our results for other rigid
tional frequencies in this work.
molecules which show the same trend as depicted in Fig. S2. The rel-
ative deviations are about the same when using unscaled or scaled
frequencies but they have the opposite sign. Merrick et al. [34] have p) is in good agreement with that determined experimentally
recently suggested specific scaling factors for obtaining the vibra- [36–40] and is given in Table S1. The principal moments of inertia
tional component of the thermal contribution to enthalpy Hvib (T) used for the calculation of the contributions of the general rota-
and entropy Svib (T) at temperatures 298.15, 450 and 600 K. These tion were IA = 1.042 × 10−45 kg m2 , IB = 1.549 × 10−45 kg m2 , and
scaling factors were not used in our work for the following reasons IC = 2.591 × 10−45 kg m2 . The calculated vibrational frequencies
i) in our opinion using different scaling factors for different ther- together with the experimental fundamental vibrations reported
modynamic properties breaks the thermodynamic consistency, ii) by Klots et al. [41] are given and compared in Table S2.
given the uncertainty in scaling factors [35] the values of the scaling The thermodynamic properties in the ideal gas state were
factors for Hvib (T) and Svib (T) for the basis set 6–311 + G(2df, p) obtained using i) experimental fundamental vibrations and com-
can be considered as equal to 1.00, iii) the scaling factors are given plete vapor phase assignment reported by Klots et al. [41], ii)
at only three temperatures allowing only interpolation within the calculated vibrational frequencies. The obtained thermodynamic
temperature range from 298 to 600 K. A more throughout work properties in the temperature range from 100 to 1000 K are given
on scaling factors for calculating the thermodynamic properties in in Table 6 and compared in Figs. 6 and 7. As can be seen, the agree-
the ideal gas state is currently in progress in our research group. ment of ideal gas heat capacities and standard molar entropies
Computational details for each of the studied compounds are given calculated using experimental fundamental vibrations [41] with
below. TRC tables [42] and calorimetric values [22] is excellent. The TRC
Thiophene. The molar mass of thiophene used for the calculation calculations [42] were based on the spectral data published prior to
of the translation contribution was M = 84.141 g mol−1 . The sym- 1984 [36,38,43–45]. The values obtained using calculated frequen-
metry of its overall rotation was C2v with symmetry number  = 2. cies do not deviate by more than 1.6% for ideal gas heat capacities
The molecular structure calculated using the B3LYP/6–311 + G(2df, and by 0.6% for standard molar entropies. For the SimCor method,
 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216 211

Table 6
Standard molar thermodynamic functions of thiophene in the ideal gaseous state at p = 105 Pa.

T/K Experimental frequencies [41] DFT B3LYP/6–311 + G(2df, p) frequencies

g0 g0 g0 g0 g0 g0 g0 g0
Cp,m Sm T0 Hm /T T0 Gm /T Cp,m Sm T0 Hm /T T0 Gm /T
−1
(J K mol−1 ) (J K−1 mol−1 ) (J K−1
mol−1
) (J K−1
mol−1
) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 )

100 34.16 227.76 33.38 194.38 34.07 227.74 33.37 194.38

150 39.19 242.39 34.34 208.05 38.82 242.29 34.25 208.03
200 48.71 254.89 36.67 218.22 48.03 254.63 36.47 218.16
250 60.69 267.02 40.25 226.77 59.73 266.58 39.93 226.65
273.15 66.51 272.65 42.23 230.42 65.45 272.12 41.85 230.27
298.15 72.76 278.74 44.53 234.21 71.61 278.12 44.09 234.03
300 73.22 279.19 44.70 234.49 72.06 278.56 44.26 234.30
400 96.07 303.49 54.78 248.71 94.72 302.49 54.13 248.36
500 114.21 326.96 64.93 262.03 112.82 325.66 64.13 261.52
600 128.20 349.08 74.36 274.72 126.80 347.52 73.46 274.05
700 139.17 369.70 82.86 286.83 137.77 367.92 81.89 286.02
800 148.03 388.88 90.47 298.4 146.62 386.91 89.45 297.46
900 155.35 406.75 97.29 309.46 153.95 404.62 96.22 308.40
1000 161.51 423.44 103.41 320.03 160.12 421.17 102.31 318.85

the ideal heat capacities were calculated using experimental fun-

damental vibrations [41] in the temperature range from (200 to
600) K with a step of 5 K and fitted with the third-order polynomial
equation using the least-squares method. The parameters of this
equation are given in Table S3.
Sulfolane. The molar mass of sulfolane used for the calculation
of the translation contribution was M = 120.171 g mol−1 . The opti-
mized molecular structure of sulfolane corresponded to the twist
conformation in which sulfolane molecule was found to predomi-
nantly exist during the investigation of the structural stability of
sulfolane by DFT-B3LYP and MP2 calculations with 6–311 + G(d,
p) basis set [46]. The molecular geometry of the twist conformer
is given in Table S1. The symmetry of its overall rotation was C2
with symmetry number  = 2. The principal moments of inertia
used for the calculation of the contributions of the general rota-
tion were IA = 2.250 × 10−45 kg m2 , IB = 4.181 × 10−45 kg m2 , and
IC = 4.505 × 10−45 kg m2 . The calculated vibrational frequencies
along with the experimental fundamental vibrations reported by
Katon and Feairheller [47] are given and compared in Table S2.
g0 g0,ref
As can be seen from Table S2, the use of scaling factor of Fig. 8. Comparison of molar ideal gas heat capacities Cp,m of sulfolane. Cp,m was
g0
calculated in this work using experimental vibrational frequencies [47]. () Cp,m
0.9686 [34] suggested for scaling fundamental frequencies at the
calculated using DFT B3LYP/6–311 + G(2df, p) vibrational frequencies in this work.
B3LYP/6–311 + G(2df, p) level of theory would yield systematically
underestimated low frequencies.
The thermodynamic properties in the ideal gas state were although Vansteenkiste et al. [48] have recently suggested to treat
obtained using i) experimental vibrations reported by Katon and ring puckering modes using the one-dimensional hindered rotor
Feairheller [47], ii) calculated vibrational frequencies. As Katon (1D-HR) scheme [49]. The authors demonstrated the application of
and Feairheller [47] did not report the lowest frequency vibra- the 1D-HR model for four-membered rings. While for entropy the
tion corresponding to the ring puckering (the authors stated: “No application of 1D-HR scheme resulted in slightly closer agreement
bands could be found at lower frequency, despite careful search- of calculated data with experimental values when compared with
ing”), the calculated frequency for this mode was added to the the HO approximation (comparison was made for two compounds),
experimental vibrations. The value of the lowest frequency was for heat capacities the agreement with experimental data slightly
dependent on the basis set selected for the calculation of vibrational worsened (comparison was made for four compounds). As in this
frequencies. DFT-B3LYP with different basis sets yielded the follow- work heat capacities are of primary interest, the 1D-HR scheme was
ing values: 23 cm−1 (6–31G(d)), 45 cm−1 (6–31 + G(d, p)), 9 cm−1 not applied for the ring puckering mode.
(6–31G(2df, p)), 42 cm−1 (6–311 + G(d, p)), 42 cm−1 (6–311 + G(2df, The obtained thermodynamic properties in the temperature
pf, p)), and 36 cm−1 (6–311 + + G(3df,3pd)). Badawi et al. [46] using range from (100 to 1000) K are given in Table 7 and compared in
MP2/6–31G(d, p) obtained 24 cm−1 for the same mode. Since for the Figs. 8 and 9. The relative deviations between the values obtained
calculation of thermodynamic properties the accuracy of the low- using the calculated and experimental vibrational frequencies do
frequency vibrations is crucial, we performed the calculations with not exceed 2.5% for ideal gas heat capacities and 0.6% for standard
42 cm−1 and with values lowered and increased by 50% (21 and molar entropies. For the SimCor method, the ideal heat capacities
63 cm−1 ) to estimate the uncertainty in the thermodynamic prop- were calculated using experimental fundamental vibrations [47]
erties corresponding to the uncertainty of the lowest vibrational and calculated frequency for the lowest vibrational mode (42 cm−1 )
mode. For ideal gas heat capacities, the difference was less than 0.1% in the temperature range from 200 to 600 K with a step of 5 K.
above T = 200 K where ideal gas heat capacities are commonly used Obtained heat capacities were fitted with the third-order polyno-
for the SimCor method. For standard molar entropies, the differ- mial equation whose parameters are given in Table S3.
ence was slightly higher but did not exceed 2%. The ring puckering Dimethyl sulfoxide. The molar mass of dimethyl sulfoxide used
mode was treated in the harmonic oscillation (HO) approximation for the calculation of the translation contribution was M = 78.134
212 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216

Table 7
Standard molar thermodynamic functions of sulfolane in the ideal gaseous state at p = 105 Pa.

T/K Experimental frequencies [47] DFT B3LYP/6–311 + G(2df, p) frequencies

g0 g0 g0 g0 g0 g0 g0 g0
Cp,m Sm T0 Hm /T T0 Hm /T Cp,m Sm T0 Hm /T T0 Hm /T
−1
(J K mol−1 ) (J K−1 mol−1 ) (J K−1 −1
mol ) (J K−1
mol−1
) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 )

100 49.68 256.62 41.17 215.45 50.87 257.10 41.53 215.56

150 63.54 279.29 46.24 233.05 65.13 280.35 46.97 233.38
200 79.61 299.72 52.53 247.19 80.85 301.21 53.44 247.76
250 97.40 319.37 59.70 259.67 97.83 321.05 60.60 260.45
273.15 106.02 328.37 63.26 265.11 106.02 330.07 64.10 265.97
298.15 115.42 338.06 67.24 270.82 114.96 339.74 67.99 271.75
300 116.11 338.78 67.54 271.24 115.62 340.45 68.29 272.17
400 152.20 377.22 84.28 292.94 150.27 378.54 84.52 294.02
500 182.89 414.59 101.03 313.56 180.14 415.39 100.75 314.64
600 207.78 450.22 116.82 333.39 204.55 450.46 116.09 334.38
700 228.02 483.82 131.31 352.51 224.47 483.54 130.19 353.35
800 244.75 515.39 144.48 370.91 240.99 514.63 143.04 371.59
900 258.75 545.05 156.42 388.63 254.89 543.84 154.72 389.12
1000 270.59 572.94 167.26 405.68 266.71 571.32 165.34 405.98

tion were calculated using

1 −ε /kT
Qint rot = e j (6)
int
j

where  int is the internal symmetry number of rotating top (for

methyl tops  int = 3), k is the Boltzmann constant, T is the tem-
perature, and the energy levels εj were obtained by solving a
one-dimensional Schrödinger equation for hindered internal rota-
tion
2
h̄ d2 
− + V (ϕ) = E (7)
2Ir dϕ2

where h̄ is Planck constant h divided by 2 ,  is the wave func-

g0
Fig. 9. Comparison of standard molar gas entropies Sm (p = 105 Pa) of sulfolane.
g0,ref
Sm was calculated in this work using experimental vibrational frequencies [47].
g0
() Sm calculated using DFT B3LYP/6–311 + G(2df, p) vibrational frequencies in this
work.
g mol−1 . The symmetry of its overall rotation was Cs with sym-
metry number  = 1. The molecular structure calculated using the
B3LYP/6–311 + G(2df, p) is in good agreement with that determined
experimentally [50] and is given in Table S1. The principal moments
of inertia used for the calculation of the contributions of the general
rotation were IA = 1.212 × 10−45 kg m2 , IB = 1.240 × 10−45 kg m2 ,
and IC = 2.029 × 10−45 kg m2 . The calculated vibrational frequencies
together with experimental fundamental vibrations are given and
compared in Table S2. Again, the comparison between calculated
and experimental frequencies (Table S2) indicates that the use of
scaling factor of 0.9686 [34] suggested for scaling fundamental fre-
quencies at the B3LYP/6–311 + G(2df, pf, p) level of theory would
yield systematically underestimated low frequencies. The exper-
imental vibrations were taken from Geiseler and Hanschma [51]
( < 600 cm−1 : Raman spectra in the liquid phase;  > 600 cm−1 :
IR spectra in the gaseous phase). In addition, the frequencies of
the torsional modes, obtained from intensity measurements on
vibrational satellites in the microwave spectrum, were taken from
Dreizler and Dendl [52]. The frequencies of the torsional modes
(207 and 231 cm−1 ) were excluded from the vibrational contribu-
tion to the partition function and treated using the 1D-HR scheme
[49]. The contributions of the methyl rotations to the partition func-
tion, E is the energy, ϕ is the torsional angle, and V(ϕ) is the
potential function of internal rotation. Energy levels were calcu-
lated by the variation method using the program StatTD. The first
512 levels were used in Eq. (6). The reduced moments of iner-
tia Ir for the methyl rotations were calculated according to Pitzer
[53] from the optimized geometrical parameters of the molecule,
Ir = 5.139 × 10−47 kg m2 . The potential functions V(ϕ) of methyl
rotations were determined by scanning the dihedral angle ϕ from 0◦
to 360◦ at 10◦ increments and allowing all other structural parame-
ters to be optimized at the B3LYP/6–311 + G(2df, p) level of theory.
The potential energy profile shown in Fig. S4 was fitted to the
Fourier series of the form
V (ϕ)/(J mol−1 ) = V0 + V3 cos(3ϕ) + V6 cos(6ϕ) (8)
with parameters V0 = 4852 J mol−1 , V3 = −4969 J mol−1 and
V6 = 117 J mol−1 . The obtained potential barrier of methyl rota-
tion was 9.94 kJ mol−1 . This value compares reasonably with
two experimentally determined values found in the literature
9.20 kJ mol−1 [54] (determined from 13 C dipole relaxation times)
and 12.84 kJ mol−1 [52] (determined by microwave spectroscopy).
The thermodynamic properties obtained by the procedure
described above are given in Table 8. Comparison of the heat capac-
ity and entropy data calculated using the 1D-HR model with those
obtained using the HO approximation for methyl rotations is shown
in Figs. 10 and 11. The maximum relative deviation is 4.5% for heat
capacity and 1.5% for entropy. For the SimCor method, the ideal
heat capacities were calculated using experimental fundamental
vibrations [51] and the 1D-HR scheme for methyl rotations in tem-
perature range from 200 to 600 K with a step of 5 K. Obtained heat
capacities were fitted with the third-order polynomial equation
whose parameters are given in Table S4.
 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216 213

Table 8
Standard molar thermodynamic functions of dimethyl sulfoxide in the ideal gaseous state at p = 105 Pa.

T/K Experimental frequencies [51] DFT B3LYP/6–311 + G(2df, p) frequencies

g0 g0 g0 g0 g0 g0 g0 g0
Cp,m Sm T0 Hm /T T0 Hm /T Cp,m Sm T0 Hm /T T0 Hm /T
−1
(J K mol−1 ) (J K−1 mol−1 ) (J K−1 −1
mol ) (J K−1
mol−1
) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 ) (J K−1 mol−1 )

100 48.13 237.12 37.75 199.37 49.10 237.48 38.03 199.45

150 60.04 258.95 43.25 215.70 61.36 259.78 43.83 215.95
200 70.53 277.68 48.77 228.90 71.86 278.90 49.55 229.35
250 80.39 294.48 54.12 240.36 81.53 295.99 54.99 241.00
273.15 84.81 301.80 56.53 245.26 85.84 303.39 57.42 245.97
298.15 89.49 309.43 59.10 250.33 90.38 311.11 59.99 251.11
300 89.83 309.98 59.29 250.69 90.71 311.67 60.18 251.49
400 107.12 338.25 69.14 269.11 107.46 340.11 69.96 270.16
500 121.78 363.78 78.24 285.53 121.68 365.67 78.92 286.75
600 134.05 387.10 86.55 300.55 133.57 388.94 87.07 301.87
700 144.48 408.56 94.10 314.46 143.68 410.31 94.45 315.85
800 153.49 428.46 100.97 327.48 152.42 430.07 101.16 328.91
900 161.35 447.00 107.26 339.75 160.09 448.48 107.29 341.19
1000 168.22 464.37 113.02 351.35 166.84 465.70 112.92 352.79

g0
Fig. 10. Comparison of molar ideal gas heat capacities Cp,m of dimethyl sulfox-
g0,ref
ide. Cp,m was calculated in this work using experimental vibrational frequencies
g0
[51] and 1D-HR treatment of methyl rotations. () Cp,m calculated using DFT
B3LYP/6–311 + G(2df, p) vibrational frequencies and HO approximation for all vibra-
g0
tional modes; (䊉) Cp,m calculated using DFT B3LYP/6–311 + G(2df, p) vibrational
g0
frequencies and 1D-HR treatment of methyl rotations; () Cp,m calculated using
experimental vibrational frequencies [51] and HO approximation for all vibrational
modes.
4.5. Recommended vapor pressure data developed by the SimCor
method
Selected vapor pressure data (data given in bold in Table 3) were
treated simultaneously with selected liquid heat capacities (data
given in bold in Table 4), ideal gas heat capacities (Tables 6–8) and
vaporization enthalpies (data available only for thiophene) using
the SimCor method. PVT correction used in Eq. (1) and Eq. (2) was
evaluated from the virial equation with the second virial coefficient
estimated by the method of Tsonopoulos [55]. Critical parameters,
dipole moments and acentric factors needed in this method are
given in Table S4. The thermal data were used in the temperature
range in which the PVT correction does not play a significant role.
Molar volumes of the liquid phase along the saturation line play
a negligible role in the temperature range considered and were
therefore neglected.
Recommended vapor pressure data are represented by the Cox
equation, Eq. (3), whose parameters are given in Table 9. The
comparison with the literature values is shown using deviation
plots in Figs. 12–14. For thiophene, consistency of selected vapor
g0
Fig. 11. Comparison of molar ideal gas heat capacities Sm (p = 105 Pa) of dimethyl
g0,ref
sulfoxide. Sm was calculated in this work using experimental vibrational fre-
g0
quencies [51] and 1D-HR treatment of methyl rotations. () Sm calculated using
DFT B3LYP/6–311 + G(2df, p) vibrational frequencies and HO approximation for all
g0
vibrational modes; (䊉) Sm calculated using DFT B3LYP/6–311 + G(2df, p) vibrational
g0
frequencies and 1D-HR treatment of methyl rotations; () Sm calculated using
experimental vibrational frequencies [51] and HO approximation for all vibrational
modes.
pressures [22] with related thermal data is excellent as also evi-
g 0 g0 cd
denced by Fig. 15, where the differences cd Cp,m = Cp,m − Cp,m
g g0
and C  = Cp,m − Cp,m
cd obtained from C cd and C
p,m p,m with those
obtained from the Cox equation using Eqs. (1) and (2) are shown
g cd is the heat
(Cp,m is the heat capacity of saturated vapor and Cp,m
capacity of crystalline or liquid phase). For sulfolane, mutual con-
sistency of selected vapor pressure datasets with thermal data is
satisfactory, however, four adjustable parameters had to be used
to describe data within their experimental uncertainties. In the
case of dimethyl sulfoxide, the agreement of the two middle-range
vapor pressure datasets [56], [57] is reasonable at pressures higher
than 6 kPa. It should be however noted that these datasets are of
moderate quality only and neither of the two datasets would yield
acceptable values when extrapolated towards room temperature.
The simultaneous correlation with selected low-pressure datasets
and thermal data shows satisfactorily good consistency which
places confidence on the recommended values. Vapor pressure
equation was also developed for crystalline dimethyl sulfoxide;
in this case the Cox equation with two parameters was ade-
quate. The parameters of Cox equations for solid and liquid phase
were constrained to provide consistent values of vapor pressure,
214 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216

Table 9
Parameters of the Cox equation, Eq. (3).

Compound Phase A0 A1 (×103 ) A2 (×106 ) A3 (×109 ) T0 p0 /Pa (Tmin − Tmax )/K

Thiophene Liquid 2.790512 −1.717967 1.717448 0 357.28 101,325 235–393

Sulfolane Liquid 2.895527 −1.571109 2.046244 −1.161591 559.11 101,325 300–534
Dimethyl sulfoxide Crystalline 3.391908 −0.1939959 0 0 291.67 49.11 200–291.67
Dimethyl sulfoxide Liquid 2.891322 −1.557871 1.448432 0 463.75 101,325 291.67–465

Fig. 12. Comparison of experimental vapor pressures p with recommended val-

Fig. 14. Comparison of experimental vapor pressures p with recommended values
ues pSimCor developed by the SimCor method for thiophene: () Milazzo [58]; ()
pSimCor developed by the SimCor method for dimethyl sulfoxide in the liquid state:
Waddington et al. [22] (static method); ( ) Waddington et al. [22] (ebulliometry);
() Douglas [67]; () Heinrich and Surový [56]; ( ) Jose et al. [68]; () Jakli and van
() White et al. [59]; (夽) Eon et al. [60]; (䊉) Daubert and Hutchison [62] (ebulliom-
Hook [57]; (夽) Nishimura et al. [69]; ( ) Ul’yanov et al. [71] (partially displayed due
etry); ( ) Daubert and Hutchison [62] (static method, partially displayed due to
to large deviations from pSimCor ); ( ) Negadi et al. [72]; () Kaporalova and Chaikin
large deviations from pSimCor ); ( ) Evlanov [74]; (· · ·) absolute deviations of 10, 102 ,
[73] (partially displayed due to large deviations from pSimCor ); (䊉) this work; (· · ·)
103 , and 104 Pa.
absolute deviation of 1, 10, 102 , and 103 Pa. Data by Cambell [70] are not displayed
due to large deviations from pSimCor .

g g0 g
0
Fig. 15. Comparison of differences cd Cp,m = Cp,m − Cp,m
cd
and C  = Cp,m − Cp,m
cd

cd g0
obtained from Cp,m and Cp,m and pVT correction with those obtained from the Cox
g g0
0
equation, Eq. (3): () thiophene,  Cp,m 
from Cp,m and Cp,m ; () thiophene, C from
 g0 
Cp,m , Cp,m and pVT correction; (—) thiophene, C from the Cox equation, Eq. (3);
Fig. 13. Comparison of experimental vapor pressures p with recommended values 0g
() sulfolane,  Cp,m 
from Cp,m
g0
and Cp,m ; (䊉) sulfolane, C from Cp,m

, Cp,m and
g0

pSimCor developed by the SimCor method for sulfolane: ( ) Shcherbina et al. [65] pVT correction; (- - -) sulfolane, C from the Cox equation, Eq. (3); () dimethyl
(partially displayed due to large deviations from pSimCor ); () Negadi et al. [66] (par- g 0
sulfoxide in the crystalline phase,  Cp,m cr
from Cp,m
g0
and Cp,m ; () dimethyl sulfox-
tially displayed due to large deviations from pSimCor ); (夽) Steele et al. [29]; (䊉) this g0
ide in the crystalline phase, C from Cp,m
cr
, Cp,m and pVT correction; (· · ·) dimethyl
work; (· · ·) absolute deviation of 0.1, 1, 10, 102 , and 103 Pa. Data by Benoit and Char-
sulfoxide in the crystalline phase, C from the Cox equation, Eq. (3); () dimethyl
bonneau [63] and Ashcroft et al. [64] are not displayed due to large deviations from g 0  g0
sulfoxide in the liquid phase,  Cp,m from Cp,m and Cp,m ; () dimethyl sulfoxide in
pSimCor . g0
the liquid phase, C from Cp,m

, Cp,m and pVT correction; (– – –) dimethyl sulfoxide
in the liquid phase, C from the Cox equation, Eq. (3).
 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216 215

Table 10 cd
Cp,m molar heat capacity of condensed phases (crystalline or
g
Enthalpies of vaporization  Hm for thiophene, sulfolalane and dimethyl sulfoxide
liquid phase) at constant pressure
calculated using Clapeyron equation and Cox equation, Eq. (3).
cr
Cp,m molar heat of crystalline phase at constant pressure
g −1
T/K  Hm /(kJ mol ) 
Cp,m molar liquid heat capacity at constant pressure
g 0
Thiophene Sulfolane Dimethyl sulfoxide  Cp,m difference between ideal gas heat capacity and heat
240 37.84 capacity of liquid
g
250 37.28 C auxiliary quantity defined by Eq. (2), C  = Cp,m − Cp,m
cd
260 36.74 E energy
270 36.19 g0 g0
T0 Gm /T reduced Gibbs energy in the ideal gas state, T0 Gm /T ≡
280 35.65
g0 g0
290 35.12 (Gm (T ) − Gm (0))/T
298.15 34.69 66.88 53.30 g
 Hm enthalpy of vaporization
300 34.59 66.76 53.18 g
310 34.06 66.13 52.56
cr Hm enthalpy of sublimation
320 33.53 65.50 51.94 cr Hm enthalpy of fusion
330 33.00 64.89 51.33 H auxiliary quantity defined by Eq. (1)
340 32.48 64.29 50.73 Hvib (T) vibrational component of the thermal contribution to
350 31.96 63.70 50.14
enthalpy
360 31.44 63.12 49.55 g0 g0
370 30.93 62.55 48.97 T0 Hm /T reduced enthalpy in the ideal gas state, T0 Hm /T ≡
g0 g0
380 30.42 61.99 48.40 (Hm (T ) − Hm (0))/T
390 29.91 61.43 47.84
h̄ reduced Planck constant (h̄ = h/2 )
400 60.88 47.29
410 60.33 46.74
h Planck constant (h = 6.62606896 × 10−34 J s)
420 59.78 46.19 IA , IB , IC principal moments of inertia
430 59.24 45.65 Ir reduced moments of inertia for internal rotations
440 58.69 45.12 k Boltzmann constant (k = 1.3806504 × 10−23 J K−1 )
450 58.15 44.60
m mass
460 57.60 44.08
470 57.04 M molar mass
480 56.48 p vapor pressure
490 55.91 ptp triple point pressure
500 55.33 p0 reference pressure in Cox equation, Eq. (3)
510 54.74
520 54.15
Qint contribution of internal rotations to partition function
530 53.54 R molar gas constant (R = 8.314472 J K−1 mol−1 )
Svib (T) vibrational component of the thermal contribution to
entropy
enthalpy of sublimation and vaporization at the triple point tem- T temperature
perature. T0 reference temperature in Cox equation, Eq. (3)
Enthalpies of vaporization calculated using Clapeyron equation Ttp triple point temperature
are tabulated as a function of temperature in Table 10. In the case V(ϕ) potential function of internal rotation
of dimethylsulfoxide, there are no literature data for comparison. V0 , V3 , V6 parameters of Eq. (8)
Calculated vaporization enthalpies for thiophene are in perfect 0 z difference between the compressibility factors of satu-
agreement with calorimetric literature data by Waddingron et al. rated gas and liquid
[22]. For sulfolane, calculated enthalpies of vaporization are slightly
lower than literature values [29]. Note that the literature data [29]
Greek letters
were also obtained using Clapeyron equation, but without inclusion
εj energy levels
of thermal data.
ϕ torsional angle
 wave function
5. Conclusions
 symmetry number of molecule
 int internal symmetry number
Recommended vapor pressure equation for thiophene, sul-
folane, and dimethyl sulfoxide was developed by a multi-property
fit of selected experimental vapor pressure data, calorimetrically Subscripts
measured enthalpies of vaporization and differences in heat capac- max maximum
ities of condensed phases and ideal gas. New vapor pressure data min minimum
for sulfolane and dimethyl sulfoxide, and liquid heat capacities for sat property or derivative along saturation line
sulfolane were determined in this work. The thermodynamic prop- s sample
erties in the ideal gaseous state for all the studied compounds were sapp sapphire
calculated by the methods of statistical thermodynamics. E empty

List of symbols Superscripts

calc calculated
ref reference
A integrated values of differential heat flow
A0 , A1 , A2 , A3 parameters of Cox equation, Eq. (3) Acknowledgements
c specific heat capacity
g0
Cp,m molar ideal gas heat capacity at constant pressure We thank Vladimir Diky and Ala Bazyleva for providing the pro-
g
Cp,m molar heat capacity of saturated vapor at constant pres- gram StatTD and for valuable and helpful discussions on statistical
sure thermodynamic calculations of ideal gas properties, Pavel Morávek
216 M. Fulem et al. / Fluid Phase Equilibria 303 (2011) 205–216

for help with the sample degassing prior to vapor pressure determi- [32] V. Majer, V. Svoboda, Enthalpies of Vaporization of Organic Compounds. A
nations, and Novasol Belgium for providing us free of charge with Critical Review and Data Compilation, Blackwell, Oxford, 1985.
[33] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheese-
anhydrous sample of sulfolane. This work is supported by the Min- man, J. Montgomery, J.A., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S.
istry of Education of the Czech Republic under grant MSM 604 613 Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
7307 and the Czech science foundation project 203/09/1327. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P.
Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Strat-
Appendix A. Supplementary data mann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari,
Supplementary data associated with this article can be found, in J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Ste-
the online version, at doi:10.1016/j.fluid.2011.02.002. fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. John-
son, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian, Inc., Wallingford
References CT, 2004.
[34] J.P. Merrick, D. Moran, L. Radom, J. Phys. Chem. A 111 (2007) 11683–11700.
[1] K. Růžička, V. Majer, AIChE J. 42 (1996) 1723–1740. [35] K.K. Irikura, R.D. Johnson, R.N. Kacker, J. Phys. Chem. A 109 (2005) 8430–8437.
[2] P.J. Mohr, B.N. Taylor, D.B. Newell, Rev. Modern Phys. 80 (2008) 633–730. [36] B. Bak, D. Christensen, L. Hansen-Nygaard, J. Rastrup-Andersen, J. Mol. Spec-
[3] P.J. Mohr, B.N. Taylor, D.B. Neweo, J. Phys. Chem. Ref. Data 37 (2008) 1187–1284. trosc. 7 (1961) 58–63.
[4] E.R. Cox, Ind. Eng. Chem. 15 (1923) 592–593. [37] R.A. Bonham, F.A. Momany, J. Phys. Chem. 67 (1963) 2474–2477.
[5] V. Roháč, V. Růžička, K. Růžička, M. Poledníček, K. Aim, J. Jose, M. Zábranský, [38] J. Jokisaari, Y. Hiltunen, T. Väänänen, Mol. Phys. 51 (1984) 779–791.
Fluid Phase Equilib. 157 (1999) 121–142. [39] P.B. Liescheski, D.W.H. Rankin, J. Mol. Struct. 178 (1988) 227–241.
[6] V. Roháč, J.E. Musgrove, K. Růžička, V. Růžička, M. Zábranský, K. Aim, J. Chem. [40] I.V. Kochikov, Y.I. Tarasov, V.P. Spiridonov, G.M. Kuramshina, D.W.H. Rankin,
Thermodyn. 31 (1999) 971–986. A.S. Saakjan, A.G. Yagola, J. Mol. Struct. 567–568 (2001) 29–40.
[7] K. Růžička, I. Mokbel, V. Majer, V. Růžička, J. Jose, M. Zábranský, Fluid Phase [41] T.D. Klots, R.D. Chirico, W.V. Steele, Spectrosc. Acta A: Mol. Biomol. Spectrom.
Equilib. 148 (1998) 107–137. 50 (1994) 765–795.
[8] I. Mokbel, K. Růžička, V. Majer, V. Růžička, M. Ribeiro, J. Jose, M. Zábranský, Fluid [42] M.L. Frenkel, G.J. Kabo, K.N. Marsh, G.N. Roganov, R.C. Wilhoit, Thermodynamics
Phase Equilib. 169 (2000) 191–207. of Organic Compounds in the Gas State, Thermodynamics Research Center,
[9] V. Růžička, M. Zábranský, K. Růžička, V. Majer, Thermochim. Acta 245 (1994) College Station, Texas, 1994.
121–144. [43] J.M. Orza, M. Rico, An. Quim., Ser. B 58 (1962) 295–314.
[10] V. Roháč, K. Růžička, V. Růžička, D.H. Zaitsau, G.J. Kabo, V. Diky, K. Aim, J. Chem. [44] M. Rico, J.M. Orza, J. Morcillo, Spectrochim. Acta 21 (1965) 689–719.
Thermodyn. 36 (2004) 929–937. [45] D.W. Scott, J. Mol. Spectrosc. 31 (1969) 451–463.
[11] K. Růžička, M. Fulem, V. Růžička, J. Chem. Eng. Data 50 (2005) 1956–1970. [46] H.M. Badawi, W. Forner, B. El Ali, A. Al-Durais, Spectrosc. A: Mol. Biomol. Spec-
[12] M. Fulem, K. Růžička, V. Růžička, T. Šimeček, E. Hulicius, J. Pangrác, J. Chem. trom. 70 (2008) 983–990.
Thermodyn. 38 (2006) 312–322. [47] J.E. Katon, W.R. Feairheller, Spectrochim. Acta 21 (1965) 199–201.
[13] M. Fulem, K. Růžička, P. Morávek, J. Pangrác, E. Hulicius, B. Kozyrkin, V. [48] P. Vansteenkiste, V. Van Speybroeck, G. Verniest, N. De Kimpe, M. Waroquier,
Shatunov, J. Chem. Eng. Data 55 (2010) 4780–4784. J. Phys. Chem. A 110 (2006) 3838–3844.
[14] G.W.H. Höhne, W.F. Hemminger, H.-J. Flammersheim (Eds.), Differential Scan- [49] J. Pfaendtner, X. Yu, L.J. Broadbelt, Theor. Chem. Acc. 118 (2007) 881–898.
ning Calorimetry, Springer Verlag, Berlin, 2003. [50] V. Typke, M. Dakkouri, J. Mol. Struct. 599 (2001) 177–193.
[15] M. Straka, K. Růžička, V. Růžička, J. Chem. Eng. Data 52 (2007) 1375–1380. [51] G. Geiseler, G. Hanschma, J. Mol. Struct. 11 (1972) 283–296.
[16] M. Zábranský, V. Růžička, V. Majer, E.S. Domalski, Heat Capacity of Liquids. [52] H. Dreizler, G. Dendl, Z. Naturforsch. 20a (1965) 1431–1440.
Critical Review and Recommended Values, American Chemical Society, Wash- [53] K.S. Pitzer, W.D. Gwinn, J. Chem. Phys. 10 (1942) 428–440.
ington, DC, 1996. [54] J.R. Lyerla, D.M. Grant, J. Phys. Chem. 76 (1972) 3213–3216.
[17] M. Čenský, M. Lipovská, H.G. Schmidt, V. Růžička, G. Wolf, J. Therm. Anal. [55] C. Tsonopoulos, AIChE J. 20 (1974) 263–272.
Calorim. 63 (2001) 879–899. [56] J. Heinrich, J. Surový, Sborník Prác Chemickej Fakulty SVŠT (1966) 207–211.
[18] J. Combet, M. Bée, D. Djurado, G. D’Assenza, G. Commandeur, Phys. B 233 (1997) [57] G. Jakli, W.A. Van Hook, J. Chem. Thermodyn. 4 (1972) 857–864.
95–101. [58] G. Milazzo, Gazzetta Chimica Italiana 74 (1944) 58–62.
[19] M. Della Monica, L. Jannelli, U. Lamanna, J. Phys. Chem. 72 (1968) 1068–1071. [59] P.T. White, D.G. Barnard-Smith, F.A. Fidler, J. Ind. Eng. Chem. 44 (1952)
[20] D.W. Kydon, A.R. Sharp, M.E. Hale, A. Watton, J. Chem. Phys. 72 (1980) 1430–1438.
6153–6157. [60] C. Eon, C. Pommier, G. Guiochon, J. Chem. Eng. Data 16 (1971) 408–410.
[21] M. Čenský, V. Roháč, K. Růžička, M. Fulem, K. Aim, Fluid Phase Equilib. 298 [61] J.O. Triday, P. Rodriguez, J. Chem. Eng. Data 30 (1985) 112–116.
(2010) 192–198. [62] T.E. Daubert, G. Hutchison, AIChE Symp. Ser. 86 (1990) 93–114.
[22] G. Waddington, J.W. Knowlton, D.W. Scott, G.D. Oliver, S.S. Todd, W.N. Hubbard, [63] R.L. Benoit, J. Charbonneau, Can. J. Chem. 47 (1969) 4195–4198.
J.C. Smith, H.M. Huffman, J. Am. Chem. Soc. 71 (1949) 797–808. [64] S.J. Ashcroft, A.D. Clayton, R.B. Shearn, J. Chem. Eng. Data 24 (1979) 195–199.
[23] P. Figuière, H. Szwarc, M. Oguni, H. Suga, J. Chem. Thermodyn. 17 (1985) [65] A.E. Shcherbina, L.M. Kaporovskii, E.I. Shcherbina, Zh. Prikl. Khim. 57 (1984)
949–966. 1910–1912.
[24] M. Zábranský, Z. Kolská, V. Růžička, E.S. Domalski, J. Phys. Chem. Ref. Data 39 [66] L. Negadi, I. Mokbel, A. Ait-Kaci, J. Jose, ELDATA: Int. Electron. J. Physico-Chem.
(2010) 1–404. Data 1 (1995) 313–320.
[25] H.L. Clever, E.F. Westrum, J. Phys. Chem. 74 (1970) 1309–1317. [67] T.B. Douglas, J. Am. Chem. Soc. 70 (1948) 2001–2002.
[26] L. Becker, J. Gmehling, J. Chem. Eng. Data 46 (2001) 1638–1642. [68] J. Jose, R. Philippe, P. Clechet, B. Soc. Chim. Fr. (1971) 2860–2866.
[27] M. Zábranský, V. Růžička, E.S. Domalski, J. Phys. Chem. Ref. Data 30 (2001) [69] M. Nishimura, M. Nakayama, T. Yano, J. Chem. Eng. Jpn. 5 (1972) 223–226.
1199–1689. [70] A.N. Campbell, Can. J. Chem. 57 (1979) 705–707.
[28] M. Castagnolo, A. Inglese, G. Petrella, A. Sacco, Thermochim. Acta 44 (1981) [71] B.A. Ul’yanov, B.I. Shchelkunov, M.Y. Fereferov, A.I. Moiseev, V.K. Stankevich,
67–76. Zh. Prikl. Khim. 64 (1991) 1579–1580.
[29] W.V. Steele, R.D. Chirico, S.E. Knipmeyer, A. Nguyen, J. Chem. Eng. Data 42 (1997) [72] L. Negadi, T. Guetachew, A. Ait-Kaci, J. Jose, ELDATA: Int. Electron. J. Physico-
1021–1036. Chem. Data 3 (1997) 91–99.
[30] S.-C. Ho, J.-M. Chen, M.-H. Li, J. Chin. Inst. Chem. Eng. 38 (2007) 349–354. [73] G.A. Kaporalova, A.M. Chaikin, Khim. Fiz. 18 (1999) 68–69.
[31] M. Mundhwa, S. Elmahmudi, Y. Maham, A. Henni, J. Chem. Eng. Data 54 (2009) [74] S.F. Evlanov, Khim. Prom-st. (1995) 536–541.
2895–2901.