Embed Size (px)
Citation preview
OATAO is an open access repository that collects the work of Toulouseresearchers and makes it freely available over the web where possible
Any correspondence concerning this service should be sent
to the repository administrator: [email protected]
This is an author’s version published in: http://oatao.univ-toulouse.fr/25 859
To cite this version:
Menossi, Matías and Milanesio, Juan M. and Camy, Séverine and Harrisson, Simon and Strumia, Miriam and
Destarac, Mathias Volumetric properties of carbon dioxide + acrylic acid binary in the context of supercritical precipitationpolymerization. (2020) The Journal of Supercritical Fluids,160. 104787. ISSN 0896-8446.
Official URL:
https://doi.org/10.1016/j.supflu.2020.104787
Open Archive Toulouse Archive Ouverte
Volun1
contex
Matias MMiriam S
• lnstituto de lnX5016GCA, C6r• Laboratoire d'Laboratoire dFrance d Laboratoire d
Keywords: Carbon dioxAcrylic acid
Density Liquid-vaporHigh-pressu
* CorresponE-mail add
etric properties of carbon dioxide + acrylic acid binary in the
t of supercritical precipitation polyn1erization
enossi a .Juan M. Milanesio a, •. Séverine Camyb , Simon Harrissonc. trumia a, Mathias Destaracd
vestigaci6n y Desarrollo en lngenierfa de Procesos y Qufmica Aplicada (/PQI\-UNC-CON/CIT). Av. Vélez Sarsjield 1611, Ciudad Universitaria, doba,Argentina e Génie Chimique, Université de Toulouse, CNRS, /NPf, UPS, Toulouse, France e Chimie des Polymères Organiques (LCPO). CNRS, ENSCBP University of Bordeaux. UMR 5629, 16, av. Pey Ber/and, 33607 Pessac Cedex,
es /MRCP, Université de Toulouse, CNRS UMR 5623, Université Paul Sabatier, 118 route de Narbonne, 31062, Toulouse Cedex 9, France
ide
equilibrium
GRAP HICAL AB STRACT
!&,--------------------,
..
,.
• 3,32.c
O 3232 IC
• 33321(
3432 tC
o, 02 03 04 os 06 07 0.8 0,9
Acrtllc aoo more tracU<)!I
ABST RACT
Acrylic acid can be polymerized by precipitation in different solvents. Carbon dioxide is an interesting solvent given its tunable density and solvent power depending on the pressure and temperature. These physicochemical characteristics make it possible to solubilize some polar compounds in CO 2 depending on temperature and pressure. ln this work. we report pressure vs volume pseudo-continuous curves at
re constant temperature for the CO 2 + M binary mixture. They were determined in a fully computerized variable-volume high-pressure view cell capable of monitoring the position of the piston. The experimental data is simultaneously isoplethic and isothermal and it covers a wide range of pressures (up to 20MPa). Using this raw experimental data, other properties were determined such as liquid-vapor
s vols inc
phase boundaries, density and excesmole fraction of Min the mixture wa
ding author.
1. Introdu
ress: [email protected] U,M. Milanesio).
Carbon d
conversion
several cha
solvents by
umes. The temperature range was 313.15-363.15 K and the reased from 0.044 to 0.594.
ction
ioxide (CO2 ) capture, utilization and storage, including
to valuable chemicals, are contemporary issues with
llenges (1). The replacement of conventional organic
more benign compounds has a considerable interest
iCptqc=psvaoncsbCprph
ptTtfrtrdiabh
sstbnbbTimimstotTmwta
ddcapwpw
n new technologies with the aim of reducing waste streams [2].O2 has many unique properties and constitutes an increasinglyopular alternative in polymer processing [2]. Is a low dielec-ric constant compound and a weak Lewis acid with a significantuadrupole moment. Further, CO2 is a gas at normal conditions but,ompressed, can reach the supercritical domain (TC =304.25 K, PC7.38 MPa) at relatively moderate conditions. CO2 has interestingroperties in the supercritical region including tunable density andolvent power, intermediate viscosities between vapor and liquidiscosities and “liquid-like” heat capacity depending on pressurend temperature [3]. Particularly, in polymer synthesis, CO2 hasther interesting advantages, such as high mass transport rates,o Trommsdorff effect and no chain transfer to solvent [2]. Super-ritical CO2 can solubilize low molecular weight polar moleculesuch as ethanol, acetone, tetrahydrofuran [4], and even some car-oxylic acids such as acrylic acid (AA) [5]. Therefore, supercriticalO2 (scCO2) can also be used as a reaction medium in radicalolymer synthesis, as many unsaturated monomers are soluble atelatively moderate conditions [6–8]. However, the correspondingolymers are typically insoluble in scCO2, giving rise to a typicaleterogeneous polymerization.
Hydrophilic polymers can be obtained by various heterogeneousolymerization processes in organic solvent, including precipita-ion, inverse emulsion, suspension and dispersion polymerizations.hese systems differ in the initial phase behavior of the polymeriza-ion mixture, polymerization kinetics, and mechanism of particleormation [9]. Although widely used, these processes sometimesequire purification steps. Desorption of the surfactant from par-icles and removal of volatile organic compounds (VOCs), likeesidual monomer and/or solvent from the polymer, are among therawbacks of these processes [10]. Precipitation polymerization
s a stabilizer-free process in which both initiator and monomerre soluble in the reaction medium while the polymer is insolu-le and precipitates as it is formed and allows the preparation ofigh-purity polymer powders [11].
Poly(acrylic acid) (PAA) is used in many applications: as auperabsorbent polymer for hygiene products and agriculture,cale inhibitor for water treatment and dispersant for paints andhe paper industry. On an industrial scale, it is produced eithery solution free-radical polymerization in water or in heteroge-eous media [12]. AA has been polymerized successfully in CO2y precipitation polymerization in a batch reactor [2,11,13,14] andy continuous precipitation polymerization in Continuous Stirredank Reactors (CSTR) [7]. To properly design supercritical reactors,
t is important to consider the phase behavior of the initial reactiveixture. In this context, one of the main goals of the present work
s to experimentally find conditions of homogeneity for the binaryixture CO2 + AA at different temperatures to carry out the initial
tages of the polymerization in single fluid phase conditions. Onhe other hand, density data is important for the design of continu-us supercritical reactors, given that the volumetric flow rates andhe residence time are related to this property of the fluid mixture.o experimentally find conditions of homogeneity for the solventixtures, an isothermal/isoplethic method was used in this work,ith a continuous scan of the pressure and the density of the mix-
ure. This method was developed by Prof. Kiran’s group and therere several publications showing its importance [15–17].
The main goal of this work is to provide new experimentalata on phase boundaries at different temperature and volumetricata for the ‘CO2 + AA’ binary system. Pressure vs Volume pseudo-ontinuous curves at constant temperature were determined in
variable volume high-pressure view cell, in a wide range of
ressures. As part of these experiments, liquid-vapor boundariesere determined, and Density vs Pressure and Density vs Tem-erature plots were built. Six different mixture compositionsere loaded to the cell from 0.044 to 0.594 mol fraction of AA in
the 313.15–363.15 K temperature range. The maximum pressurerecorded was 20 MPa and the maximum density 980 kg/m3. Withthis volumetric data, excess volume plots at high pressure forthe mixture were obtained, resulting in negative deviations. Theexperimental window swept in this work is in accordance with thepolymerization conditions described previously in the literaturefor AA in CO2 [7,11,13,14]. The phase boundaries were correlatedwith a simple model. A three-parameter cubic equation of state(3P-EoS), the RKPR EoS, developed by Cismondi et al. [18], wasapplied in order to improve the representation of densities fordifferent types of compounds, while maintaining the relativesimplicity of cubic equations of state. This EoS has shown topredict very well the phase behavior of CO2-containing mixtures[19–21], but this is the first time, to our knowledge, that is appliedto a carboxylic acid binary system. This work forms part of awider project involving the copolymerization of AA and othercomonomers in compressed solvents.
2. Materials and methods
2.1. Materials
SEPPIC (Toulouse, France) provided acrylic acid (AA) (79-10-7CAS number) 97 % pure, according to gas chromatography (GC) con-taining 200 ppm of inhibitor (monomethylether of hydroquinone(MEHQ)) and it was used without further purification. Carbon diox-ide (CO2) (124-38-9 CAS number) with a purity of 99.95 % (GC)was purchased from Air Liquide (Toulouse, France) and was usedas received.
2.2. Experimental setup and operational procedures
The volumetric data and phase boundaries were obtained in avariable volume high pressure cell (Top Industrie No 2607 0000)that incorporates a linear variable differential transducer (LVDT) forcontinuous sensing of the position of the movable piston inside thevariable-volume part of the cell. A motorized screw is connected tothe back part of the piston and is used to change the inner volume ofthe system and, therefore, the pressure. The volume, pressure andtemperature of the system together with the rate of increasing ordecreasing volume can be controlled and recorded by the softwareprovided with the phase equilibrium cell. Temperature was mea-sured by a thermocouple (J type, precision of ±0.1 K) placed in thecenter of the cell. Pressure was measured using a pressure trans-ducer (Keller, precision of ±0.13 MPa) equipped with a pressurenumerical display. The body of the view cell is built in stainless steeland it has internal tubing for temperature control. The maximuminternal volume of the cell is 31.8 × 10−5 m3 while the minimumvolume is 9.8 × 10−5 m3. These values were confirmed by measur-ing the mass of a low surface tension solvent, i.e. n-hexane, that canbe loaded into the cell at the maximum (minimum piston stroke)and minimum volume (maximum piston stroke). A sapphire win-dow permits visual observations as well as the recording of thesystem (by using the camera (5), see Fig. 1) during a given experi-ment. The temperature of the cell is controlled using a circulationbath (Lauda Eco Silver) and tubing incorporated in the body of thecell as it was described above. Fig. 1 shows a diagram of the phaseequilibrium cell used in this work.
In a typical experiment, liquid AA is charged to the cell using aregular syringe through the inlet port (10) (see Fig. 1) The mass ofAA is weighted in a Sartorius analytical balance with a weighingrange of 1 × 10−5–0.2 kg and a weighing uncertainty of 1 × 10−7 kg.
The amount of acid transferred is measured gravimetrically. Car-bon dioxide is charged through the inlet valve (6), connected to ahigh pressure electronically controlled syringe pump (Top IndustriePMHP 100–500). The amount of carbon dioxide charged is deter-
(1)
(3)
(8)
(9)
(6)
(4) {10)
(5)
(2)
Fig. 1. Schema ump, (3): Motorized piston driver, (4): Magnetic stirrer, (5): Camera, (6): lnlet valve, (7) liquids.
mined from
pump. lnitiset at r• andstate. The v
of the syrinopened, lettin the head
( 6} is closedare stabilize
the carbon dThe dens
p-V-T inform
of carbon dplication ofin volume (V
charging th
piston is seing the con
mechanicalpressure or
forward un
volume of tture and voOnce the mthe screw m
a continuousmooth exp
ume. the th
changes ass
experimentconditions. isotherm. F
in an experidata pointsThere is a ga long the foeffect) of th
0:00:00 00:06:29 00:12:58
Time [min)
00:19:26 00:25:55
ypicalell. A.
feren
ibutef thes in t
ed e
perim prim
to th
d int
its ondu
erage
ation
al, st
tic diagram of the experimental setup: (1 ): High-pressure view cell, (2): Syringe p: Purge valve, (8): Pressure transducer, (9): Temperature sensor, (10): lnlet port for
the initial and final conditions in the head of the syringe
ally, the temperature and the pressure of the head are p•_ At those conditions. the CO2 must be in the liquid
olume v• occupied by the carbon dioxide in the head
ge is registered. Theo, the inlet valve (6) of the cell is ing the carbon dioxide flow into the cell. The pressure of the syringe pump decreases. Finally, the inlet valve
and the pressure and the temperature of the syringe d again at r and p•_ The new volume V .. occupied by
ioxide in the head of the syringe (V .. < v•) is registered. ity of carbon dioxide at r• and P' is computed from CO2
ation from the NISTweb book database (22). The mass
ioxide loaded into the cell is obtained simply by multi the density of CO2 obtained from NIST by the difference .-V .. }. Total charge was typically about 0.014 kg. After
e system. the cell is heated to the temperature and the
t at the maximum volume of the system, while mixtent with a magnetic stirring bar (4) (see Fig. 1). Once
and thermal equilibrium is observed (no change in temperature of the system}, the piston is slowly moved
til the maximum pressure (20 MPa) or the minimum
he cell (9.8 x 10-5 m3 ) is reached. Pressure, temperalume of the cell are recorded during the piston stroke. aximum pressure or the minimum volume is reached. oves the piston backward to its initial position. Under
s and slow pressure change, the fluid experiences a ansion or compression. During these changes in vol
ermal circulation bath compensates for any thermal
ociated with expansion or compression of the fluid. The
s can be considered to be under quasi equilibrium state The rate of volume change was 2.5 x 10-8 m3/s for eachig. 2 shows typical pressure vs time raw data obtained
120
110
� 100<Il B 90
� � 80
CL 70
60
500
Fig. 2. Tlibrium c343.151<
The dif
be attrwalls oducer i
describ
this exThe
linked
charge
ing andwere c
The av
propag
Mate ri
ment. This particular data set corresponds to about 800 with data being recorded every 2 s during the scans. ood reproducibility of the piston position and pressure rward and backward direction (i.e. there is no hysteresis e scans at the rate of volume change (2.5 x 10-8 m3/s).ture and CO
mole fractioThe sudd
to the chancompressib
pressure vs time raw data obtained in the high-pressure phase equicrylic acid mole fraction of the mixture is 0.214 and the temperature
ces in pressure were 0.5 MPa maximum and they can
d to the piston seal friction against the inner cylinder cell. lt is important to note that the pressure trans
he back of the piston. This methodology was previously
lsewhere (15-17) but for the first time was applied to
ental setup. ary sources of error in the determination of density are
e uncertainties on the exact amount of the fluid CO2
o the cell. and the errors associated with the LVDT read
relation to the position of the piston. Measurements cted using pure CO2 and compared with NIST database.
error in density was estimated to be less than 5%. A
of the error analysis is provided in the Supplementary
udying the effect of the error in the pressure. tempera
2 and M masses loaded to the cell in the informed M
n. en change in the pressure vs time slope corresponds
ge in the phase scenario inside the cell. The isothermal ility of the liquid mixture is much lower than the two-
1100 � ------------------- -�
1000
900
Ë 800
� 700
600
500
400
300 2 10 12
Pres1ure / MPa
313.2 K
-PureC02 -xM•0.044 -xM•0.175
xAA= 0.214 -xAA•0.3 -xAA=0.594
14 16 18 20
Fig. 3. Oensity vs pressure at constant overall composition (xM) and constant temperature (313.2K) for the system M+CO2 • Expcrimcntal data:
( -) x,u = 0.0 (data from NIST (22));
( -) XAA = 0.044 (this work);
( -- ) x,u = 0.175 (this work);
( ) XAA = 0.214 (this work);
( - ) XAA = 0.300 (this work);
( -) Xt1t1
(--) Axt1t1: AA mol
phase liquidin Fig. 3 corliquid-like compressioture. The bdensity of tmixture.
2.3. Modeli
EquationIate and prerange of contem were cRKPR-EoS e
P= /"!.b -
Where, Pisis the molaand the repuniversal gaSi,EoS, whicpure comporule. For moattractive pmixture areder Waals cinteraction
3. Results
3.1. Density
The densfor five diff0.044, 0.175
1100 ---------------------�
1000
900
500
400
323,2 K
-PureC02 -xAA=0.044 -><AA•0.175
xAA=0.214 -xAA•0.3 -><AA•0.594
300 '--_______ ..,_ ____________ _, 2 10 12 14 16 18 20
Pre11l.Ke / MPa
Fig. 4. Density vs pressure at constant overall composition (xM) and constant temperature (323.2 K) for the system M + CO2 • Experimental data:
( -) XM = 0.0 (data from NIST (221);
( -) XAA = 0.044 (this work);
(-) xt1t1=0.175(thiswork);
( - ) XAA = 0.214 (this work);
( - ) XM = 0.300 (this work);
) x,u = 0.594 (this work);
) AA saturated liquid density ac 323.2 K [23]. mole fraction.
---------------------�
--=
-xAA•0.()4,4 -�=0.175 -xAA=0.214
-><AA=0.3 -'XAA • 0.594
---------�----------�-' 2 10 12 14 16 18 20
Pressure/ MPa
ensity vs pressure at constant overall composition (xM) and constant tem (333.2K) for the system M+CO2 • ental data:
) x,u = 0.0 (data from NIST [221);
) XAA = 0.044 (this work);
) XAA = 0.175 (this work);
) Xt1t1 = 0.214 (this work);
) XAA = 0.300 (this work);
) XM
) A mol
res w, 333. In ad poseal
= 0.594 (this work); A saturated liquid density at 313.2 K [23). e fraction.
-vapor mixture. Therefore, the lower slope segments respond to compression or decompression of a single
phase, and the steeper slope segments correspond to n or decompression of a liquid-vapor two-phase mixreak point (pressure, temperature, composition and he mixture) is considered to be the bubble point of the
ng
s of state (EoS) are the proper type of models to corredict the fluid phase behavior of mixtures over a wide ditions. Experimental data for the M + C02 binary sys
orrelated using the RKPR-EoS (18]. The expression for xplicit in pressure, is as follows:
( 8 b) ( l-81EoSb) (l)
v+ 1,Eos v+ 1+8i:E.s
the absolute pressure, T is the absolute temperature, vr volume, a and b are the attractive energy parameter ulsive co-volume parameter respectively, and R is the s constant. The RKPR-EoS introduces a third parameter h allows to improve the reproduction of PvT data for unds. In a mixture, 81 ,Eos depends on a linear mixing re details about RKPR-EoS see Ref. (18]. In thiswork, the arameter a and the repulsive parameter b in the binary obtained using quadratic mixing rules with classic van ombining rules with both an attractive and a repulsive parameter k12 and 112 .
and discussion
( -
(--xt1t1: AA
1100
1000
900
1 800
� 700
� .. 600
!500
400
300
Fig. S. DperatureExperim
( -
( -
( -
( -
( -
( -
(--XAA: AA
peratu323.1520MPato avoipiston
ities for C02 + M binary system have been determined erent solutions containing increasing amounts of M:
, 0.214, 0.300 and 0.594mol fraction of M. The tem-
the mixturedatabase (2M at the cduced in thein density fo
= 0.594 (this work);
A saturated liquid density at 333.2 K (23].
e fraction.
here the experiments were carried out were: 313.15, .15, 343.15, 353.15 and 363.15 K and the pressure up to ll determinations, the scan rates were 2.5 x 10-8 m3/s, ssible errors in pressure caused by the friction of the s. Figs. 3-8 show the density isotherms for each of
s. The pure C02 isotherms were extracted from NIST 2]. Also, in dashed lines, the density for saturated pure orresponding temperature is shown. They were intro plots for illustration purposes and represent the limits r both pure compounds. The data shows the expected
1100 ,----------------------7
Aayte 8Cld WU8led l(Jld demlty 813432 K 1000 •••••••• • • • • •• • • • • • • • • • • • • • • • • •• • • • • • • •• • • • • • •• • • • • • • • • • • • •
900
È 800
t 700
� ·; 600
500
400
343.2 K
-PureC02
-MA•0.044 ->AA.= 0.175
xAA.• 0.214 -xAA.11!0.3
-xAA= 0.594
3002 L--------
8--
1-o
-<"'-1
-2--1-4--
1-6--
1-8---:'20
Pre11ure / MPa
Fig. 6. Density vs pressure at constant overall composition (x..,,) and constant temperature (343.2 K) for the system M +COi.
Experimental data: ( -) x,u = 0.0 (data from NIST (22)); ( -) XAA = 0.044 (this work); ( -- ) x,u = 0.175 (this work); ( -) XAA = 0.214 (this work); ( -) XAA = 0.300 (this work); ( - ) x,1,1
(- -) AAXAA: AA mole
trends: the pure CO2 toible when tgiven changrich mixturregion (i.e. ation, increashow the vsures for earaw experim
1100 r--
1000 -- -
900
l: 900
� 700
� ·.; 600
500
400
300 2
35
-P--.AA-xAA
xAA=(l--.AA=0--.AA=0
Fig. 7. Densityperature (3532Experimenta( -) x,u
( -) XAA
( -- ) x,u
( -) XAA
( -) XAA
( - ) x,1,1
(--) AAXAA: AA mole
1100 ,------------------------,
900
l: 600
� 700
�
i! 600
600
400
363.2 K
-P\iro002
--.AA•0.0437 --.AA= 0.175
xAA•0.214 --.AA=0.3
--.AA-0.594
300 L-------------.c....-------�20 2 6 10 12 14 16 18
Pretsure f MPa
Fig. 8. Density vs pressure at constant overall composition (x..,,) and constant temperature (3632 K) for the system M +COi.
Experimental data: ( -) XAA = 0.0 (data from NIST (22)); ( -) XAA = 0.044 (this work); ( -) x,u = 0.175 (this work); ( -) XAA = 0.214 (this work); ( -) XAA = 0.300 (this work);
) XM
) A mol
cons dataomp8 ateren
= 0.594 (this work); saturated liquid density at 343.2 K [23). fraction.
density, at a given temperature, increases in going from M richer mixtures and they become less compress
he M mole fraction is increased. This means that, for a e in pressure, the relative change in density is less in Mes than in CO2-rich mixtures. Also, in the single-phase
( -(--xAA: AA
sure atdensitystant cFigs. 3-for diff
t pressures above the break point), for a given composising pressure results in increasing densities. Figs. 9-14 ariation of density with temperature at selected presch mixture in the homogeneous liquid regions. As the ental density data was acquired as a function of pres-
--------------------7
AQ)tc illdd wtante<f iq\.iddonetty et 353,2 K ----- - - - - - - - - - ---- - - - - - - - - - ----- - - - - -- - - - ----- - - - - - - - -- -
3.2 K
ureC02 =0.0437 •0.175
,214 .3
.594
10 12 14 16 18 20 Pre1-1ure / MPa
vs pressure at constant overall composition (x..,,) and constant tem K) for the system M +COi.
l data: = 0.0 (data from NIST (22)); = 0.044 (this work); = 0.175 (this work); = 0.214 (this work); = 0.300 (this work); = 0.594 (this work); saturated liquid densiry at 353.2 K [23). fraction.
data pointsous lines wAs expectedincreases wtemperaturpronouncedwith xAA = 0the density
900
800
700
1600
- 500
i! •OO Pure
+11
�13W
200 .o,.1s
-0-17r
1�L3.- 2 -
Fig. 9. DensityExperimenta(-•-(-•-( -•-(-•-The straight li
= 0.594 (this work); A saturated liquid densiry at 363.2 K (23). e fraction.
tant composition from the phase equilibrium cell, the as a function of temperature (different isobars at con
osition) was obtained by intercepting the isopleths of a constant pressure and registering the density values t temperatures. This explains the fewer experimental in Figs. 9-14 compared with Figs. 3-8. The continuere added to facilitate the visualization of the trends. , at a given composition, the density of the mixture ith increasing pressure. For a given isobar, increasing
e decreases the mixture density, and this effect is less for M-rich mixtures. For example, for the mixture .594 (Fig. 14), at a pressure of 13 MPa, the variation in between 333.15 and 363.15 K is around 30 kg/m3• while
CO2
tvt>a
a
wa
vPa
-3-13-.2- --32-3.2---3-33-.2---34-3.2 ___ :isJ __ .2 __ 36-:-:-3.2---=37·3,2
Tem.,.rature / K
vs temperature at constant pressure for pure C02.
l data from NIST [221: )P= 11 MPa; )P= 13MPa; )P= 15 MPa; )P= 17MPa;
nes were added to facilitate visualization.
900
800
700
\ Q600
! 500
400 x..,. • 0.0-U
.. 0-11 MPa
300 -.13MPa
-0-15MPa
.. 0·17MPa 200
303.2 313 .2 323 .2 333.2 343.2 Temperature / K
363.2 363.2 373.2
Fig. 10. Oensity vs temperature at constant ove rail composition (xM = 0.044) and constant �ressure for the system AA +C02.
Expcrimcntal data (this work):
( -- • -) P = 11 MPa;
( -• -) P = 13 MPa;
( -• -) P = 15 MPa;
(-•-
x,u: AA mol
The straigbt li
for the mixttemperaturTo our knowdensities foparison cansection, phawith literatindicate thewith increa
Figs. 15 aM mole frasure. As canincreases th
\ t
900 ---
800
700
-eoo
f �
500 XM •
400
-0-11
...,.13MP
-a-15 MP
"o 17MP
300 L--303.2
Fig. 11. Oensiconstant pressExpcrimcnta
(-•-
(-•-
(-•-
(-•-
XAA: AA mol
The straight li
1000�---------------------7
X,v. = 0.594
--0-11tiPa
..,.13Wa
-o-1sw>a
.. o,-11w:ia
900 L----------------------' 303.2 313.2 323.2 333.2 343.2 363.2 363.2 373.2
Tempe rature / K
Fig. 14. Oensity vs temperature at constant ove rail composition (xM = 0.594) and constant pressure for the system AA +C02. Experimental data (this work):
( -• -) P = 11 MPa;
(-•-)P= J3MPa;
(-•-)P= 15MPa;
•-
mol
ight li
nonl incr
attri disc
ase b
se bes tim in pressure vs density curves corresponds to the onset of
)P= 17 MPa;
e fraction.
nes were added to facilitate visualization.
ure with xAA =0.175 (Fig. 11) at the same pressure and e range, the variation in density is more than 300 kg/m3
•
ledge, there are no previous publications that report r C02 +Mat high pressures. Therefore, no direct com be made with existing data. Nevertheless, in the next se boundaries determined in this work were compared ure data. The sudden slope changes in the isotherms phase boundaries. This break point is much clearer
sing M content in the mixture.
(-
XAA: AA
The stra
tian isdensitycan bewill be
3.2. Ph
Phasure vchange
nd 16 show the evolution of the density as a function of ction of the mixture, at a given temperature and pres be seen, increasing the content of M in the mixture e density. The dependence of density on M molar frac-
--------------------
0.175
MPa
a
a
a ---------------- - --�
313.2 323.2 333.2 343.2 363.2 363.2 373.2 Temp«ature J K
ty vs temperature at constant overall composition (xM =0.175) and ure for the system AA +C02. l data (this work):
)P=ll MPa;
)P= J3MPa;
)P= lSMPa;
)P= 17MPa;
e fraction.
nes were added tO facilitate visualization.
a new vapoing the voluliquid phasvolume of t
900 --
800
x..,. •
"D-1
500 -a.1
-0-1
-o-1
400 L-303.2
Fig. 12. Oensiconstant pressExperimenta
(-•-
(-•-
(-•-
( _ •
XM: AA mol
The straigbt li
)P= 17 MPa;
e fraction.
nes were added to facilitate visualization.
inear. At intermediate composition, there is a steeper ease with the addition of M to the mixture. This effect buted to the better packing of the liquid mixture and ussed in Section 3.3.
oundaries
havior data can also be obtained from continuous prese recordings. As mentioned before, the abrupt slope
r phase when moving the piston backward and increasme of the system, or to the collapse of the vapor into the e when moving the piston forward and decreasing the he system. The phase boundaries obtained in this work
----------------------,
0.214
1 MPa
3 MPa
5 MPa
7 MPa ---------------------..J
313.2 323.2 333.2 343.2 363.2 363.2 373.2 Temperature / K
ty vs temperature at constant overall composition (xM =0.214) and ure for the system AA +C02. l data (this work):
)P= li MPa;
)P= 13MPa;
)P= 15MPa;
) P = 17 MPa;
e fraction.
nes were added to facilitate visualization.
1000 ,--------------------�
900
700 ._.,. = 0.300
+13MPa
.--a 15 t.f>a
.. o,.17MPa 600 ........ -----�--�--�--�----...........
303.2 313.2 323.2 333.2 353.2 363.2 373.2 Temperature / K
Fig. 13. Density vs temperature at constant ove rail composition (x..,, = 0.300) and constant pressure for the system M+COi. Expcrimcntal data {this work):
(-•-)P= 13MPa; (-•-)P= ISMPa; ( - • ) P = 17 MPa; XAA: AA mole fraction.
The straight lines were added to facilitate visualization.
1,100.0 ,---------------------�
1,000.0
900.0
? f800.0
i100.o ..
600.0
500.0
13 MPa
333.2 K
400.0 ........ -----�--------------....0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Acrylic acld mole fraction
Fig. 15. Mass density vs composition (x..,.) at constant temperature (3332 K) and constant pressure (13 MPa) for the system M +C02.
x..,,: M mole fraction. The straight lines were added to facilitate visualization.
1,100.0 ,---------------------�
1,000.0
900.0
� 800.0
� i100.o . C .
0 600.0
500.0
15 MPa
343.2 K
400.0 '-------�--------------... 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Acrylic acld mole fraction
Fig. 16. Mass density vs composition (x..,.) at constant temperature (3432 K) and constant pressure (15 MPa) for the system M +C02.
x..,,: M mole fraction. The straight lines were added to facilitate visualization.
18
16
14
: 12 :E
5 10
i .t 8
<> <> <>
�<> � �-•
<> • •
•
• •
o<> $ � • � • <> 0 0 •
<1°o�g •
<>
<> • <> •
<> <> •
• <> <> 0
• <>
•
•
•
•
0
•
e313.2K 0323.2 K e333.2 K e343.2 K •363.2 K •36 3.2 K 0318.2 K Ref. 5 0338.2 K Ref. 5 o 368.2 K Ref. 5
o 373.2 K Ref. 5
<> <> <> <>
2 .___...._ ___________ _._ _ __. __ _._ ___ __, 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Acrylic ackl mole fraction
Fig. 17. Pressure vs composition (x..,.) at constant temperature for the system M+C(h. Experimental data (this work):
(•)T=3l3.2K; (o) T=323.2 K;
(•)T=333.2 K; (•)T=343.2K; (•)T=353.2K; (•) T=363.2 K;
(0) T=318.2 K (Ref rs]);
(0) T=338.2 K (Ref [5]);
(0) T=358.2
(0) T=373.2XAA: AA mol
are shown icompare thByun et al. ues and theboundaries of the mixtdensity of tshow the saperature. ThDIPPR databsities of theM
The phastwo adjusta11 2 for the ceters used interaction system are mode! fittincan be seenaries. but, goverpredicthigher M caries. Althocases (363.1as a simplethe phase bexperimentgle crowdedis shown indata obtainoverpredictmixture) anthe bounda
K (Ref[5]); K (Ref[5));e fraction.
n Fig. 17. Four more data series were added to Fig. 17 to e data obtained in this work with the values reported by [S]. There is good agreement between the present val literature data. Table 1 shows the experimental phase obtained in this work. including the global composition ure, the phase boundary pressure and the saturation he mixture. An additional line was added to Table 1 to turation pressure and density of pure Mat given temese values were calculated using correlations from the ase (23]. ln ail cases, as expected, the saturation den
mixtures are lower than the saturation density of pure
e boundaries obtained were fitted using RKPR EoS with ble parameters: k12 for the attractive parameter a and o-volume of the mixture b. The pure compound paramin RKPR EoS are presented in Table 2, and the binary parameters fitted for the phase boundaries of the binary presented in Table 3. Fig. 18 shows the results of the g compared with the data obtained in this work. As it . there is a good representation of the phase boundenerally, at low M content of the mixture, the EoS
s the pressure of appearance of the new phase, and at omposition, the EoS underpredicts the phase boundugh the differences can be as high as 2 MPa in some 5 K and xAA = 0.594), it is important to note that RKPR. cubic EoS mode!, can represent the main trends of ehavior of 'carboxylic acid + C02' binary mixture in the al window studied in this work. To avoid showing a sin Figure, the phase boundary data from Byun et al. [S]
Fig. 19 using the same parameters as in Fig. 18 for the ed in this work. Again, at low M concentration, the EoS s the phase boundaries (and the critical pressure of the d for high M content mixtures, the EoS underpredicts ry pressures. The percentual average relative deviation
Table 1Phase boundaries and saturation densities for CO2 + AA (this work). Grey lines: pure AA saturation data extracted from DIPPR [23]. The percentual average relative deviation(ARD) is shown at the bottom of each isotherm.
Temperature [K] AA mole fraction Pressure [MPa] Saturation density [kg/m3] Transition type*
313.15
0.044 ± 0.001 8.14 613 ± 10 DP0.175 ± 0.001 8.31 805 ± 13 BP0.214 ± 0.002 7.12 826 ± 16 BP0.300 ± 0.002 6.45 837 ± 9 BP0.594 ± 0.004 4.00 959 ± 10 BP
Ref. [23] 1.000 0.014 1028ARD (%) 6.24
323.15
0.044 ± 0.001 9.91 497 ± 8 DP0.175 ± 0.001 10.00 729 ± 12 BP0.214 ± 0.002 8.87 803 ± 15 BP0.300 ± 0.002 8.04 849 ± 9 BP0.594 ± 0.004 5.10 952 ± 10 BP
Ref. [23] 1.000 0.0024 1017ARD (%) 8.82
333.15
0.044 ± 0.001 11.12 485 ± 8 DP0.175 ± 0.001 11.06 678 ± 11 BP0.214 ± 0.002 10.19 784 ± 15 BP0.300 ± 0.002 9.18 840 ± 9 BP0.594 ± 0.004 5.77 938 ± 10 BP
Ref. [23] 1.000 0.0041 1005ARD (%) 6.74
343.15
0.044 ± 0.001 12.81 454 ± 8 DP0.175 ± 0.001 12.68 660 ± 11 BP0.214 ± 0.002 12.14 781 ± 15 BP0.300 ± 0.002 10.84 792 ± 8 BP0.594 ± 0.004 6.99 937 ± 10 BP
Ref. [23] 1.000 0.0067 994ARD (%) 7.76
353.15
0.044 ± 0.001 13.90 427 ± 7 DP0.175 ± 0.001 14.02 626 ± 10 BP0.214 ± 0.002 13.68 748 ± 14 BP0.300 ± 0.002 12.34 776 ± 8 BP0.594 ± 0.004 7.91 926 ± 10 BP
Ref. [23] 1.000 0.0105 982ARD (%) 7.02
363.15
0.044 ± 0.001 15.16 397 ± 7 DP0.175 ± 0.001 15.45 587 ± 10 BP0.214 ± 0.002 15.75 743 ± 14 BP0.300 ± 0.002 14.30 778 ± 8 BP0.594 ± 0.004 9.260 909 ± 9 BP
Ref. [23] 1.000 0.0161 970ARD (%) 8.31
u(T) = ±0.1 K. u(p) = ±0.13 MPa. u is the standard uncertainty in the measurement.* BP: Bubble Point; DP: Dew Point.
Table 2Pure compound parameters for the RKPR EOS used in this work.
Carbon dioxide Acrylic acid
Acentric factor 0.2236 0.5383ac (MPa. m6/mol2) 0.38751 × 10−7 2.20618 × 10−6
b (m3/mol) 2.76 × 10−5 6.52 × 10−5
ı1 1.9722 3.3181
Table 3RKPR EoS binary interaction parameters granted by the minimization of relativeerrors of bubble pressure.
kij lij
(eutttdm
density was found in the literature. The results are shown in Fig. 20.
CO2+AA −0.06 −0.1
%ARD) between experimental and predicted value is given at thend of each isotherm in Table 1. Although the prediction of the liq-id densities obtained with the EoS was not quantitatively accurate,he general trends were well represented by the model. In general,he predicted liquid densities of the mixtures were always lower
han the experimental values, mostly at high pressure. The modeleviations increase with increasing carbon dioxide content of theixture.3.3. Excess volume
The excess volumes for the mixtures, representing the differ-ence between real and ideal molar volume of the mixture, weredetermined from Eq. (2), where �m is the density of the mixtureand xi, Mi, and �i are the mole fraction, the molar mass, and thedensity of the component i, respectively:
VE = xCO2MCO2 + xAAMAA�m
− xCO2MCO2
�CO2− xAAMAA
�AA(2)
In this relationship, densities are expressed on mass basis,whereas the resulting excess volumes are dimensionally on molarbasis. Using the density data (in kg/m3) for the pure com-pounds and for the mixtures, and the molar mass values for CO2(0.04401 kg/mol) and AA (0.07206 kg/mol), the excess volumes(m3/mol of mixture) were evaluated as a function of the mole frac-tion of AA in the saturated liquid mixtures at different temperatures(333.15 and 343.15 K) at a given pressure of 15 MPa. The density ofAA was considered as the saturated liquid density at 333.15 and343.15 K, given that no experimental data of compressed liquid
The excess volumes are negative in the whole composition range.The minimum of the excess volume seems to be around xAA∼=0.25,a composition similar to those used for radical polymerization of
18 • 313.2 K O 323.2 K
18 • 333.2 K • 343.2 K
14 • 353.2 K
• 363.2 K
12 -313.2 K RKPR -323.2 KRKPR
-333.2 K RKPR 10 -343.2 KRKPR
-353.2 K RKPR
-363.2 K RKPR
2Li.....-���-����-���-��._;:,..:::,:,;s.:,......�...J
0 0.1 0.2 0.3 o., 0.5 0.6 0.7 0.8 0.9 Acrylc acid mole fraction
Fig. 18. Pressure vs composition (xM) at constant temperature for the system M+C(h. Experimental data (this work): (•)T=3l3.2K; (o) T=323.2 K;
(•) T=333.2 K;( ) T=343.2 K;
(•) T=353.2 K;(•) T=363.2 K;
Computed isothenns (Model: RKPR, kc02.AA=-0.06, lc02.AA=-O.J):
( -- ):T=313.2K; ( -- ):T=323.2K; ( -- ):T=333.2 K; ( - ):T=343.2K;( __ ):T=343.2K;( --):T=363.2K;
XAA: AA mole fraction.
18
18
14
: 12 :Il
! 10
Ë 8
o 318.2KR8'.5 o 338.2 K Rel. 5 o 358.2 K Ref. 5 o 373.2 K Ref.5
0 -318.2KRKPR
-338.2KRKPR
-358.2 K RKPR
-373.2 K RKPR
2 w......-���-����-���-���..::::,.....,.;ss....�...J
0 0.1 0.2 0.3 o., 0.5 0.6 0.7 0.8 0.9 Ac,ylic acid mole fraction
Fig. 19. Pressure vs composition (x,,,,) at constant temperature for the system M+C(h. Experimental data: (<>) T:31 8.2 K (Ref [5]);
(<>) T=338.2 K (Ref [5]);
(<>) T=358.2 K (Ref [5));
(<>) T=373.2 K (Ref [51); Computed isotherms (Model: RK PR, kco2.AA=-O .06, lco2.AA=·Ü .1 ):
( --):T=3l8.2 K ; ( -- ):T=338.2K; ( -- ):T=358.2 K ; ( -- ):T=373.2 K.
XAA: AA mole fraction.
o.œ+oo
-5.0E-06
•1.0E-05
-1.SE-05
•2.0E-05
15MPa
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Acrylic ackl mole fraction
Fig. 20. Excess volume vs composition (xM) at constant temperature and constant pressure (lSMPa) for the system M +C02• Experimental data (this work): (-•-) T=333.2K; (-•-XM: AA molThe straight li
Min carbosent the miAs introduceffect of thevalues of exing, often ardifferent sizliquid mixtferent sizesmolecules iticles. This particle mixthe state, e.molecular mAlthough ththe idea cansibility of dof the mixtis lowered density anderal referenin the wholhigh pressu'n-decane +umes for 'acexcess voluclosely relathat methavalues by SXMerhanol�0.
reported inxAA�0.25, a
Fig. 21 sthe M mol333.2 and 3solution is vexcess volucompositionshrinkage o
This respolymerizaproceeds, th
) T=343.2K. e fraction.
nes were added to facilitate visualization.
n dioxide [2,7,11 ]. This composition appears to reprexture composition with the highest degree of packing. ed in Section 3.1 (Figs. 15 and 16) there is a packing mixture at intermediate compositions. The negative cess volume in mixtures are indicative of better packising in asymmetric systems that involve molecules of es (24]. In the literature, the concept of "packing" in
ures was described for non-polar hydrocarbons of dif where small molecules fill the voids between larger n analogy to dynamics observed in packing of solid parapproach shows an analogy between the porosity of a ture at different states (i.e. the porosity depends on
g. packed or fluidized) and different temperatures for ixtures ( i.e. the 'porosity' depends on the temperature). e concept was developed for non-polar hydrocarbons, be extended to our mixture. Ease of rotation and accesifferent conformational states for the biggest molecule ure, i.e. M. will be more difficult when temperature at a given pressure, unfavorably affecting the packing leading to less negative excess volumes. There are sevces in the literature showing a negative excess volume e compositional range for COi-containing mixtures at re. Bessières et al. (25] measured excess volumes for CO2' mixtures, Kiran et al. (26] measured excess voletone + CO2 ' mixtures and Smithjr. et al. (27] measured mes for 'methanol + CO2 ' mixtures. The latter being the ted system compared with 'CO2 + M' mixtures, given nol and M are bath polar compounds. The reported mith Jr. et al. (27] are as low as -20 x 10-6 m3/mol at 25, at 323 .2 K and 11 MPa, while in the 'CO2 + M' data this work, the excess volume is -23 x 10-6 m3/mol at t 343.2 K and 15 MPa. hows the molar volume of the mixture as a function of ar fraction at 15 MPa for two different temperatures: 43.2 K. The negative deviation compared to the ideal ery clear. The ratio between the absolute value of the
me and the molar volume of the mixture at the same can be as high as 30 %. This represents a significant
f the mixture compared to an ideal solution. ult could be important in radical batch supercritical tions using C02 as the solvent, given that, as the reaction e Mis consumed to form oligomers first and finally the
9.0E--05
8.SE--05
L 8,0E--05 0
E
7.5E-05
ë
7,0E--05
�
2 6.SE--05
6,0E--05
S.SE--05
0 0.1 0.2
343.2K
0.3 0.4 0.6 0.6
Acrylic ackl mole fraction
0,7 0,8 0.9
Fig. 21. Molarpressure (15 Mfx18
polymer. Th
evolves insi
decrease or
as long as thxAA�0.25 [2
can only inc
4. Conclus
Pseudo-
carried out
313.15-363
increases w
ature. Also,
increases w
sure, increa
This effect
boundaries
the break p
with previo
ary data wak12 and 112.
boundaries
the pressur
At xAA high
sures comp
are less tha
Excess v
range indic
pure compo
M mole fra
tive mixtur
pressure, hiumes. This rpolymeriza
Declaration
None.
Acknowled
SEPPIC i
work. We a
ciones Cien
and Univers
Appendix A
Supplem
in the onlin
104787.
References
(1) E. Alper, Oprocesses2016.11.0
(2) T.J. Romaacrylic ae912-915,
(3) M. MukhChem. Te
2. . Gupt7. . Byu
tems C-192,oyèreymer /10.10iu,j.Mcrylic polym6-255
hen, Xrmole
nomecromo30055rshad
thodo/10.10 Tsavactureomat. Ollagared slic ae-378,erth. ann'
u, B. Hpress6-18iu,j.Mymeriymeri6/j.ce
. Milanane, an13) 65alco, Eary flu://dx
. Granoly( lad mix6/j.su
Cismo- PR e.org/1ima,J.na, Thne) m016/j
Cismoilibriaelopmm. Re
Cismose equs cub
volume vs composition (x,.,) at constant temperature and constant Pa) for the system M+COi.
erefore. the composition of the CO2 + M binary mixture
de the reactor. The pressure in a batch reactor should
increase depending on the initial reactive composition,
e reaction progresses. If the initial composition is near ,7,11 ], the pressure in a batch polymerization reactor
rease.
ions
continuous pressure vs density measurements were
for CO2 + M binary mixtures in the range of
.15 K and up to 20 MPa. The data show that the density
ith the M content of the mixture at a given temper
at a given composition, the density of the mixture
ith increasing pressure. As expected, for a given pres
sing temperature decreases the density of the mixture.
is less clear for mixtures with high M content. Phase
were obtained from the pressure vs density scans as
oint of a given isotherm. They show good agreement
us data from literature. Modeling of the phase bound
s made using RKPR EoS with two adjustable parameters: In general. there is a good representation of the phase
. At low M content of the mixture, the EoS overpredicts
e of appearance of the new phase and the critical point.
er than 0.2, the mode! underpredicts the bubble pres
ared with the experimental measurements. The %ARD
n 8.5 % for ail isotherms measured in this work.
olumes are highly negative in the whole composition
ating better packing of the mixtures compared to the
unds. The largest negative values are observed at 0.25
ction, a composition closely related to the initial reac
e for supercritical polymerization of M. At a given
gher temperatures lead to higher negative excess volesult cou Id have an impact on the pressure du ring batch tion in supercritical conditions.
of Competing Interest
gements
207(4) R.B
200(5) H.-S
sys179
(6) C. Bpolorg
(7) T. Lof athe254
(8) J. ClntemoMama
(9) R. Ameorg
(10) J.G.struNan
(11) J.-Ninfracry372
(12) G. HUllm
(13) Q. Xcom187
(14) T. Lpolpol101
(15) J.Moct(20
(16) N. Fbinhttp
(17) H.Eof pflui101
118) M. RK doi
(19) S. SGeaalka10.1
(20) M. equdevChe
(21) M. pharule
s gratefully acknowledged for financial support of this
lso wish to thank to Consejo Nacional de Investiga
tificas y Técnicas de la Republica Argentina (CONICET)
idad Nacional de C6rdoba (UNC).
671-681,(22) E.W. Lem
(Eds.), "ThWebBookof Standaorg/10.18
. Supplementary data
entary material related to this article can be found,
e version, at doi: https://doi.org/10.1016/j.supflu.2020.
. Yuskel Orhan, C02 Utilization: developments in conversion , Petroleum. 3 (2017) 109-126, http://dx.doi.org/10.1016/j.petlm. 03. ck, E.E. Maury,J.M. DeSimone, Preeipitation polymerization of id in supercritical carbon dioxide, Macromolecules 28 (1995) http://dx.doi.org/10.1021/ma00108a017. opadhyay, Extraction and processing with supercritical fluids,J. chnol. Biotechnol. 84 (2009) 6-12, http://dx.doi.org/10.1002/jctb.
a.j.-J. Shim. Solubility in Supercritical Carbon Dioxide, CRC Press,
n, B.M. Hasch, M.A. Mchugh, Phase behavior and modeling of the Oi-acetonitrile and C02-acrylic aeid, Fluid Phase Equilib. 115 (1996)
http://dx.doi.org/ 10.1016/0378-3812(95 )02830-7. , C. Jérôme, A Debuigne, Input of supercritical carbon dioxide ta synthesis: an overview, Eur. Polym.J. 61 (2014) 45-63, http://dx.doi. 16/ j.eurpolymj2014.07 .019. . Desimone, G.W. Roberts, Continuous preeipitation polymeri2ation aeid in supercritical carbon dioxide : the polymerization rate and er molecular weight,J. Polym. Sei. Part A: Polym. Chem. 43 (2005) 5, http://dx.doi.org/10.1002/pola20728.
. Liu, Z. Liu, O. Hu, C. Zhang, O. Xue,J. Xiao, Z. Liu, cular-interaction-dominated solvation behaviors of liquid
rs and polymers in gaseous and supercritical carbon dioxide, lecules 45 (2012) 4907-4919, http://dx.doi.org/10.1021/ 6z. y, Suspension, emulsion, and dispersion polymerization: a logical survey, Colloid Polym. Sei. 270 (1992) 717-732, http://dx.doi. 07/bf00776142. las. Emulsion copolymerization (also leading to core-shell s). in: S. Kobayashi, K. MO lien (Eds.), Encyclopedia of Polymeric erials, Springer. Berlin, Heidelberg. 2015. nier, T. Tassaing, S. Harrisson, M. Destarac, Application of online pectroscopy ta study the kinetics of preeipitation polymerization of id in supercritical carbon dioxide, React. Chem. Eng. 1 (2016) http://dx.doi.org/ 10.1039/C6RE00022C. G. Schomick, F.L. Buchholz, Polyacrylamides and Poly(Acrylic Aeids), s Encyclopedia of lndustrial Chemistry, 2015, pp. 1-16.
an, H. Yan, Preeipitation polymerization of acrylic aeid in ed carbon dioxide - cosolvent systems,]. Appl. Polym. Sei. 88 (2003) 80, http://dx.doi.org/10.1002/app.11908. . DeSimone, G.W. Roberts, Kinetics of the precipitation zation of acrylic aeid in supercritical carbon dioxide : the locus ofzation, Chem. Eng. Sei. 61 (2006) 3129-3139, http://dx.doi.org/10. s.2005.11.052. esio,j.C. Hassler, E. Kiran, Volumetric properties of propane, n -d their binary mixtures at high pressures, lnd. Eng. Chem. Res. 52 92-6609, http://dx.doi.org/10.1021/ie4007084. . Kiran, Volumetric properties of ethyl acetate + carbon dioxide id mixtures at high pressures,]. Supercrit. Fluids 61 (2012) 9-24,
.doi.org/ 10.1016/j.supflu.2011.09.016. de Ili, E. Kiran, High pressure density, Miscibility and compressibility ctide�o-glycolide) solutions in acetone and acetone + C02 binary tures,]. Supercrit. Fluids 75 (2013) 159-171, http://dx.doi.org/10. pflu.2012.12.034. ndi, J. Mollerup, Development and application of a three-parameter quation of state, Fluid Phase Equilib. 232 (2005) 74-89, http://dx. 0.1016/j.fluid.2005.03.020. M. Milanesio,J.I. Ramello, M. Cismondi, C. Secuianu, V. Feroiu, O. e effect of the naphthenic ring on the VLE of (Carbon dioxide + ixtures,]. Chem. Thermodyn. 93 (2016) 374-385, http://dx.doi.org/ .jct.2015.07 .018. ndi, S.B. Rodriguez-Reartes,j.M. Milanesio, M.S. Zabaloy, Phase of CO 2 + n -Alkane binary systems in wide ranges of conditions : ent of predictive correlations based on cubic mixing rules, lnd. Eng.
s. 51 (2012) 6232-6250, http://dx.doi.org/10.1021/ie2018806. ndi,j.M. Mollerup, M.S. Zabaloy, Equation of state modeling of the ilibria of asymmetric COi+n-alkane binary systems using mixing
ic with respect ta mole fraction,]. Supercrit. Fluids 55 (2010)
http://dx.doi.org/ 10.1016/j.supflu.2010.10.007. mon, M.O. Mclinden, D.G. Friend, in: P.J. Linstrom, W.G. Mallard ermophysical Properties of Fluid Systems" en NIST Chemistry . NIST Standard Reference Database Number 69, National lnstitute rds and Technology, Gaithersburg MD, 2020, p. 20899, http://dx.doi. 434/T40303 (September 17th, 2019).
[
[
[
[
doi.org/10.1021/je9501503.[27] R.L. Smith Jr., C. Saito, S. Suzuki, S.-B. Lee, H. Inomata, K. Arai, Temperature
23] DIPPR-801. Evaluated Process Design Data, Public Release. American Instituteof Chemical Engineers, Design Institute for Physical Property Data,BYU-DIPPR. Provo, Thermophysical Properties Laboratory, Utah, 2003.
24] H.J. Finkers, J.C. Bosma, A.A. Broekhuis, A novel approach to predict the excess
volume of hydrocarbon mixtures, Chem. Eng. Sci. 66 (2011) 2889–2897,http://dx.doi.org/10.1016/j.ces.2011.02.007.25] D. Bessières, H. Saint-Guirons, J.-L. Daridon, Volumetric behavior of decane +carbon dioxide at high pressures. measurement and calculation, J. Chem. Eng.Data 46 (2001) 1136–1139, http://dx.doi.org/10.1021/je010016k.
26] E. Kiran, H. Pöhler, Y. Xiong, Volumetric properties of pentane + carbondioxide at high pressures, J. Chem. Eng. Data 41 (1996) 158–165, http://dx.
dependence of dielectric spectra of carbon dioxide and methanol mixtures athigh-pressures, Fluid Phase Equilib. 194–197 (2002) 869–877, http://dx.doi.org/10.1016/S0378-3812(01)00714-2.
