Computational Evaluation of Mixtures of ... - ACS Publications

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Computational Evaluation of Mixtures of Hydrofluorocarbons and Deep Eutectic Solvents for Absorption Refrigeration Systems Rubaiyet Abedin, Sharareh Heidarian, John C. Flake, and Francisco R. Hung Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02003 • Publication Date (Web): 09 Aug 2017 Downloaded from http://pubs.acs.org on August 19, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Computational Evaluation of Mixtures of Hydrofluorocarbons and Deep Eutectic Solvents for Absorption Refrigeration Systems Rubaiyet Abedin,1,1 Sharareh Heidarian,2,1 John C. Flake2 and Francisco R. Hung1,2 1 2

Department of Chemical Engineering, Northeastern University, Boston, MA 02115

Cain Department of Chemical Engineering, Louisiana State University, Baton Rouge, LA 70803

Abstract We used computational tools to evaluate three working fluid mixtures for single-effect absorption refrigeration systems, where the generator (desorber) is powered by waste or solar heat. The mixtures studied here resulted from combining a widely used hydrofluorocarbon (HFC) refrigerant, R134a, with three common deep eutectic solvents (DESs) formed by mixing choline chloride (hydrogen bond acceptor, HBA) with either urea, glycerol or ethylene glycol as the hydrogen bond donor (HBD) species. The COSMOtherm/TmoleX software package was used in combination with refrigerant data from NIST/REFPROP, to perform a thermodynamic evaluation of absorption refrigeration cycles using the proposed working fluid mixtures. Afterwards, classical MD simulations of the three mixtures were performed to gain insights of these systems at the molecular level. Larger cycle efficiencies are obtained when R134a is combined with choline chloride and ethylene glycol, followed by the system where glycerol is the HBD, and finally that where the HBD is urea. MD simulations indicate that the local density profiles of all species exhibit very sharp variations in systems containing glycerol or urea; furthermore, the Henry’s law constants of R134a in these two systems are larger than those observed for the HFC in choline chloride and ethylene glycol, indicating that R134a is more soluble in the latter DES. Interaction energies indicate that the R134a-R134a interactions are weaker in the system where ethylene glycol is the HBD, as compared to in the other DES. Radial distribution functions confirm that in all systems, the DES species do not form strong directional interactions (e.g., hydrogen bonds) with the R134a molecules. Relatively strong interactions are observed between the Cl anions and the hydrogen atoms in R134a, however the atom-atom interactions between R134a and the cation and HBD species are weaker and do not play a 1 2

Equal contribution Corresponding author. Email: [email protected]

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 35

significant role in the solvation of the refrigerant. In all systems, R134a has the largest diffusion coefficients, followed by the HBD, the anion and the cation; the diffusion coefficients are the largest in the systems containing ethylene glycol, followed by those having glycerol and urea. This work is our first step towards our long-term goal of designing and demonstrating optimal working fluid mixtures for use in absorption refrigeration systems. Our results suggest that COSMO-RS can be used to perform a rapid screening of a large number of working fluid mixtures, and select a few candidates for further exploration using molecular simulations and experiments. These latter approaches can be used to refine the accuracy of the COSMO-RS predictions, and to optimize the selection of optimal working fluid mixtures for demonstration in absorption refrigeration systems powered by solar or waste heat sources.

1. Introduction It is hard to overstate the benefits of refrigeration and air conditioning, but the net effect of consuming so much energy to keep cool is concerning.1,2 In 2013, the total residential energy consumption was about 21.5 trillion Btu, representing ~22% of the primary total energy consumption in the US.3 About 48% of residential energy consumption was used for heating and cooling,4 mostly as electricity in vapor compression cycles using hydrofluorocarbons (HFCs) as working fluids. Furthermore, HFCs are long-lived greenhouse gases; the EPA estimated U.S. fluorinated gas emissions to be near the equivalent of 180 million metric tons of CO2 in 2014.5 When these numbers are combined with the CO2 emissions associated with the generation of electricity used in vapor compression systems, and considering worldwide usage, the net effects on greenhouse gas generation associated with cooling and heating are quite staggering. Therefore, it is critical from economic and environmental perspectives to develop more energyefficient refrigeration (and heat pump) systems. Single-effect absorption refrigeration systems (Figure 1) have found widespread applications in places where heat released from power generation or other industrial processes is available. These systems are similar to standard vapor compression cycles, but the compressor is replaced by an absorber, pump and generator (shown on the right side of Figure 1). In the absorber, an absorbent-refrigerant solution rich in the former (‘strong solution’) absorbs the lowpressure refrigerant gas coming from the evaporator, releasing heat in the process and diluting


2

Page 3 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

the concentration of absorbent. This ‘weak solution’ is pumped and preheated before it is sent into the desorber (or ‘generator’ in Figure 1), where the refrigerant is separated from the weak solution at high pressure by adding heat. The separated refrigerant, now in gas phase, then releases heat to the surroundings in the condenser and undergoes a throttling process that cools it to a very low temperature, going into the evaporator where heat is removed from the airconditioned space. Double- and triple-effect absorption refrigeration systems (which typically involve additional heat exchangers, expansion valves and generators, and possibly additional pumps and condensers) are also commercially available. Qcond

Solar / waste heat

Qgen 3 8

7

4 Expansion valve 1

Expansion valve

‘Strong’ DES solu on with R134a

9

6 Pump

10

5

‘Weak’ DES solu on with R134a

2

Qevap

Qabs

Figure 1. Scheme of an absorption refrigeration cycle using a mixture of R134a (refrigerant) with a DES (absorbent). Conventional absorption refrigeration systems (typically ammonia-water or water-LiBr) are commonly used in large-scale cooling applications where heat is readily available. In these systems, the coefficient of performance β is obtained by dividing the heat removed in the evaporator by the total energy input to the cycle (i.e., the pump work and heat supplied to the generator): (1)


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 35

If we use waste heat (i.e., heat at no additional cost) in the generator, the second metric used to quantify the efficiency of these systems is the ‘practical’ or “waste-heat” efficiency, or η, i.e. the heat removed in the evaporator divided by the pump work alone: (2)

It is important to note that much less energy is required to increase the pressure of a liquid (using a pump) than to raise the pressure of a refrigerant vapor (using a compressor as in the vapor-compression cycle). The value of β including generator heating is typically less than 0.5 for single effect absorption systems (triple effect systems can reach β = 2); however, η can be 10 to 100 times greater than β. Thus, absorption systems are typically used where waste heat is abundant, such as in cogeneration systems, large campuses, hospitals and industrial settings. Unfortunately, current absorption systems using ammonia-water, water-LiBr or other state of the art working fluids require that the source of heat used at the desorber (generator) to be at a fairly high temperature (~150 °C). This requirement is primarily due to the vapor-liquid equilibrium behavior of ammonia-water or LiBr-water mixtures at conditions considered ideal for reasonable system parameters (flow rates, equipment sizing).6-9 Lowering the operation temperatures in the generator to ~75 °C typically results in high absorbent concentrations in the vapor phase (and higher circulation ratios), which then requires relatively large absorbers, generators, and pumping systems. Here the advantages of using ionic liquids (ILs) or deep eutectic solvents (DESs) as absorbents become evident, as both ILs and DESs have virtually no vapor pressure and hence no IL/DES would be circulated outside the absorption loop. ILs are organic salts that are in liquid phase at or near room temperature, and because over 1015 different cations and anions can be combined to form different ILs,10,11 they have been deemed as “designer solvents” with very tunable properties. In turn, DESs are made by mixing two or more components, typically a hydrogen bond acceptor (HBA, e.g., a quaternary ammonium salt such as choline chloride), and a hydrogen bond donor (HBD, e.g., urea) that can form hydrogen bonds with each other.12,13 Formation of hydrogen bonds hinders the crystallization of the individual components, which then form a eutectic mixture that has a melting point that is much lower than those of the individual components. DES share similar physical and chemical properties with ILs but have


4

Page 5 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

their own advantages such as low price, chemical inertness in water, easy to prepare and biodegradability, biocompatibility, and non-toxicity. Using ILs as absorbents for HFCs was evaluated in several recent studies and patents.14-25 Shiflett and others formerly at DuPont first reported the application of ILs for use as solvents for refrigerants, publishing several reports detailing solubility measurements and phase equilibria calculations for several refrigerants in ~19 different ILs,19-23 and filed several patents, including one involving the use of ILs and water in absorption systems.25 Kohl et al. found very high cycle efficiencies when using mixtures of common imidazolium ILs and HFCs in low-wattage absorption refrigeration systems using waste heat sources at low temperatures (ηR134a>ηR32>ηR125>ηR143a. The efficiencies determined from the TZVPD_FINE parameterization follow almost exactly the same trends with the exception of the systems with the lowest efficiencies, for which this parameterization predicts that ηR143a>ηR125. COSMO-RS predicts a very small difference in solubilities between streams 5 and 8 for R125 when using the TZVPD_FINE parameterization, which leads to a large requirement in pumping power that drops the value of efficiency for this


14

Page 15 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

system. Overall, the results shown in Table S2 suggest that COSMO-RS calculations can qualitatively capture the trends in efficiency, and thus can be used to perform rapid evaluations of the thermodynamic efficiency of absorption refrigeration cycles. These calculations can be used to quickly screen a number of working fluid mixtures, and select a few candidates for further exploration using experiments and molecular simulations. These latter approaches can be used to refine the COSMO-RS models, as well as to further select a few optimal working fluid mixtures for demonstration in lab-scale absorption refrigeration systems powered by solar or waste heat sources.

3.2. Thermodynamic evaluation of absorption refrigeration cycles using R134a-DESs mixtures In Tables 3 and 4 we present our results for the thermodynamic evaluation of a singleeffect absorption refrigeration cycle (Figure 1) using a mixture of R134a and different DES as working fluids. Properties of the R134a-DES mixtures were obtained using the COSMO-RS parameterizations TZVP (Table 3) and TZVPD_FINE (Table 4), as discussed in Section 2.1. For a given system, the values of efficiencies η vary between the different COSMO-RS parameterizations, mainly due to the large variations observed in the computed solubilities of R134a in streams 5 and 8. The results discussed in the previous section (Figures S1 and S2, and Table S1) suggest that the parameterization TZVPD_FINE would lead to R134a solubilities that would be more accurate. Although the results presented in Table 1 suggest that the densities computed using the TZVP parameterization would be closer to experimental values, our calculations indicate that the solubility of R134a in the DESs have a larger effect on the computed efficiency of the cycles. As a result, the computed cycle efficiencies are not very accurate, and therefore we believe a discussion of uncertainty in our calculations is not very relevant here. However, qualitatively both parameterizations show that larger cycle efficiencies are obtained for systems where the HBD in the DES is ethylene glycol, followed by glycerol and urea. Here we note that we have kept fixed the operating conditions of the absorption refrigeration cycle (see Section 2.1) for all systems considered. In principle, for a given working fluid, one could tune the operating conditions as to maximize the efficiency η; however, for simplicity we have not attempted to perform such an optimization process for our systems. The viscosity of the DES is another important factor to consider for these systems. Increasing solvent


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 35

viscosities can lead to larger requirements for pumping power due to increased friction losses, and can also lead to mass transfer issues at the absorber and the desorber. Although we did not explicitly consider the viscosity in our calculations, we note that experimental values of the viscosity range between 37-41 mPa.s for choline chloride + ethylene glycol at 25 °C.47,54 The other DESs are more viscous: 188-247 mPa.s for choline chloride + glycerol at 30 °C;47,51,76 and 152-974 mPa.s for choline chloride + urea at 30 °C.47,50,77,78 These viscosity values also suggest that choline chloride + ethylene glycol is the best option among the three DES evaluated in this study.

Table 3. Thermodynamic evaluation of an absorption refrigeration cycle (Fig. 1) using a mixture of R134a and different DES as working fluids. Properties of the R134a-DES mixture were determined using the COSMO-RS parameterization TZVP. System

Choline chloride + urea Choline chloride + glycerol Choline chloride + ethylene glycol

X5,R (kg R134a) / (kg DES) 0.4807

ρ5 (kg/m3)

η

1221.2

X8,R (kg R134a) / (kg DES) 0.2715

31.73

23.91

41.82

0.9659

1206.6

0.4110

11.96

12.11

82.56

1.3618

1172.8

0.5382

8.059

10.09

99.16

(W) (g/s)

Table 4. Thermodynamic evaluation of an absorption refrigeration cycle (Fig. 1) using a mixture of R134a and different DES as working fluids. Properties of the R134a-DES mixture were determined using the COSMO-RS parameterization TZVPD_FINE. System

Choline chloride + urea Choline chloride + glycerol Choline

X5,R (kg R134a) / (kg DES) 0.03781

ρ5 (kg/m3)

η

1205.3

X8,R (kg R134a) / (kg DES) 0.03272

1304

697.6

1.434

0.1842

1162.0

0.1262

114.4

72.47

13.80

0.3380

1101.2

0.1821

42.57

32.14

31.11


(W) (g/s)

16

Page 17 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

chloride + ethylene glycol 3.3. Molecular dynamics simulations In Figure 3 we present representative simulation snapshots of our systems containing choline chloride, ethylene glycol and varying amounts of R134a, ranging from 5% to 20% molar basis. For both sets of conditions considered (298 K and 4.9 bar, or 343 K and 10.2 bar), the molecules of R134a tend to cluster together as the mole fraction of refrigerant increases; for 5% and 10% of R134a, the refrigerant molecules (blue) scatter throughout the simulation box, whereas small clusters of R134a molecules are noticeable at 20%. Similar snapshots are presented in Figures S3 and S4 (Supporting Information) for mixtures of the DESs formed by combining choline chloride with either urea (Fig. S3) or glycerol (Fig. S4) with varying amounts of R134a. These snapshots are qualitatively similar to those depicted in Figure 3, however clusters of R134a molecules are noticeable when the total amount of refrigerant is 15%, and the clusters are much larger than those observed in the DES containing ethylene glycol.

Figure 3. Representative simulation snapshots of systems containing the DES choline chloride + ethylene glycol and varying mole fractions of R134a (left to right: 5%, 10%, 15% and 20%), at T = 298 K and P = 4.9 bar (top row), and T = 343 K and P = 10.2 bar (bottom row). Red = choline


17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 35

(cation), orange = chlorine (anion), green = ethylene glycol (HBD), and blue = R134a (refrigerant). In order to further examine the behavior of our systems as we increase the content of R134a, we performed MD simulations placing our systems in the center of an elongated, orthorhombic box with gas space (modeled as vacuum regions, 25 nm3 volume) on each side along the horizontal x-axis. We ran simulations containing 10% and 20% molar basis of R134a for each of the three systems in the NVT ensemble at 298 K or 343 K. In Figure 4 we present representative simulation snapshots and local density profiles along the x-axis, for R134a in choline chloride + ethylene glycol at 298 K. As the total content of R134a increases from 10% (Fig. 4, left) to 20% (Fig. 4, right) in our systems, the simulation snapshots indicate that more molecules of refrigerant appear in the vacuum (gas) regions and accumulate in the gas-liquid interface. As the total amount of refrigerant increases from 10% to 20%, the local density profiles for R134a (Fig. 4, bottom) also show larger peaks at the interface and slightly larger values in the gas regions, as well as increased amounts of R134a within the liquid phase (i.e., within x ~ 5 nm and x ~ 15 nm). In contrast, the local density profiles for the cation, anion and HBD are qualitatively similar for both systems, exhibiting small variations in local density. Within the liquid phase, the variations in the local density of R134a are more pronounced than those observed for the rest of the species in both systems (Fig. 4, bottom). These trends contrast to what is observed for R134a in the DES choline chloride + urea (Figure 5). As the total amount of refrigerant increases to 20%, significantly larger variations in the local densities of all species are observed (Fig. 5, bottom) as compared to those found for R134a in choline chloride + ethylene glycol (Fig. 4, bottom). When HBD = urea, as we move within the liquid phase from one gas-liquid interface to the other one (Fig. 5, bottom right), we encounter regions having large amounts of R134a and low densities of choline cation and urea, and viceversa (low density of R134a and large quantities of choline and urea). Variations in the local density of chloride anion are also observed, but these fluctuations are smaller when compared to those of the other species (Fig. 5, bottom right). A cluster of R134a molecules within the liquid phase is also noticeable from the simulation snapshot (Fig. 5, top right). For systems containing 10% of R134a, sharper variations in the local density profiles are observed for all species in the system choline chloride + urea (Fig. 5, bottom left) as compared to those depicted for the system choline chloride + ethylene glycol (Fig. 4, bottom left). Similar results for R134a in choline chloride + glycerol are


18

Page 19 of 35

shown in Figure S5 (Supporting Information); this system exhibits a behavior that is qualitatively similar to those presented in Fig. 5 for R134a in choline chloride + urea. Results from NVT simulations using elongated boxes for the system R134a in choline chloride + ethylene glycol at T = 343 K are presented in Figure S6 (Supporting Information). When compared to Fig. 4, results from Fig. S6 show smaller local densities of R134a dissolved in the DES, and larger amounts of refrigerant present in the gas phase, which are the expected trends as increases in the temperature should lead to a reduction in R134a solubility in these systems. Results for the other two systems at T = 343 K are qualitatively similar to those shown in Figures 5 and S5, and are not shown for brevity.

900

900

800

800

700

700

Mass density (kg/m3)


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

600 500 400 300 200 100

600 500 400 300 200 100

0

0 0

5

10

15

20

0

x, nm

5

10

15

20

x, nm

Figure 4. Representative simulation snapshots (top) and local density profiles (bottom) of systems containing 10% (left) and 20% (right) molar basis of R134a in the DES choline chloride + ethylene glycol, from NVT simulations in elongated boxes with T = 298 K. Red = choline (cation), orange = chlorine (anion), green = ethylene glycol (HBD), and blue = R134a (refrigerant). The mole fractions represent the overall composition of R134a in the whole simulation box (i.e., not distinguishing between gas and liquid phases, or the gas/liquid interface)


19

Page 20 of 35

900

900

800

800

700

700


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Langmuir

600 500 400 300 200 100

600 500 400 300 200 100

0

0 0

5

10

15

20

0

x, nm

5

10

15

20

x, nm

Figure 5. Representative simulation snapshots (top) and local density profiles (bottom) of systems containing 10% (left) and 20% (right) molar basis of R134a in the DES choline chloride + urea, from NVT simulations in elongated boxes with T = 298 K. Red = choline (cation), orange = chlorine (anion), green = urea (HBD), and blue = R134a (refrigerant). The mole fractions represent the overall composition of R134a in the whole simulation box (i.e., not distinguishing between gas and liquid phases, or the gas/liquid interface) The results shown in Figs. 4, 5 and S5 suggest that the solubility of R134a is larger in DESs where ethylene glycol is the HBD, followed by systems where glycerol is the HBD and lastly those where HBD = urea. In Figure 6 we present the Henry’s law constant of R134a in the different DESs as a function of temperature and a pressure of 1 bar, as computed from MD simulations. We note that from these values at 1 bar, standard thermodynamic relations can be used to estimate the Henry’s law constant at other pressures.79 This property can be also defined as:

P x→0 x

K H = lim

(8)


20

Page 21 of 35

Where x is the mole fraction of the refrigerant dissolved in the liquid phase and P is the pressure of refrigerant in the gas phase. From Figure 6, the system where the HBD is urea exhibits the largest values of KH at all temperatures, and therefore the solubility of R134a is the lowest among all systems studied, followed by the DES where glycerol is the HBD. The DES containing choline chloride + ethylene glycol has the smallest value of the Henry’s law constant, and thus exhibits the largest solubility of R134a among the three systems. In all cases, KH increases with temperature in an approximately linear way. We have also computed the Henry’s law constant of R134a in these systems using COSMO-RS and the parameterization TZVPD_FINE, obtaining the following values at 298 K and 343 K: 202.3 bar and 416 bar (HBD = urea); 49.7 bar and 109.3 bar (HBD = glycerol); and 38.9 bar and 97.8 bar (HBD = ethylene glycol). These values are quantitatively different from those computed from our MD simulations, but the trends are in qualitative agreement (i.e., KH,Urea > KH,Glycerol > KH,EthyleneGlycol, and that KH increases with temperatures for all systems). These observations again suggest that COSMO-RS could be used as a tool to rapidly screen a large number of refrigerant-solvent mixtures for absorption refrigeration, provided that the selected systems are further examined using molecular simulations and experiments. Further refinements in the COSMO-RS models are required for this tool to be quantitatively accurate for these systems, for example following similar strategies as those used to improve COSMO-SAC models.80 We are not aware of any experimental data of the Henry’s law constant of R134a in the DESs examined here. 2000

Henry's law constant of R134a (bar)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

y = 24.471x - 6605.4 R² = 0.96101

1800 1600 1400 1200

y = 8.0036x - 1894.5 R² = 0.97482

1000 800 600 400

y = 6.7125x - 1735.8 R² = 0.9946

200 0 280

290

300

310

320

330

340

350

T, K

Figure 6. Henry’s law constant of R134a as a function of temperature in the DESs formed by combining choline chloride with urea (blue circles), glycerol (red squares) or ethylene glycol


21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 35

(green triangles), as determined from MD simulations at the corresponding temperatures and P = 1 bar. Lines represent linear fits to the data. Radial distribution functions g(r) can give insights about the local organization of specific molecules around each species at the nanoscopic level, helping to interpret the interactions between different species. For example, g(r) can be used to determine if some functional groups in the DESs have preferential interactions (e.g., hydrogen bonds) with the refrigerant, and how the different species organize around the refrigerant molecules, providing insights into the molecular mechanisms of solvation in these systems. In Figure 7 we show the center-of-mass (COM) g(r) between cation-refrigerant, anion-refrigerant, HBD-refrigerant, and refrigerant-refrigerant interactions for R134a in choline chloride + ethylene glycol at 298 K and 4.9 bar; similar results for R134a in choline chloride + urea, and for R134a in choline chloride + glycerol, are presented in Figures S7 and S8 (Supporting Information). Results at 343 K and 10.2 bar are qualitatively similar to those shown in Figs. 7, S7 and S8 and thus not shown for brevity. For all systems and all species considered in Figs. 7, S7 and S8, the first and second peaks appear around 0.5 nm and ~ 0.8-0.9 nm. For all systems, the first peak observed in the refrigerant-refrigerant g(r) is more intense and sharper than the first peak observed in the g(r) for cation-refrigerant, anion-refrigerant and HBD-refrigerant. The first peak in the refrigerantrefrigerant g(r) is more intense in the systems where HBD = urea or HBD = glycerol (Figs. S7d and S8d) than in the systems where HBD = ethylene glycol (Fig. 7d), especially at the highest concentrations where the R134a molecules cluster together. As the amount of R134a is varied between 5% and 20%, all g(r) functions in all systems show variations in peak intensity, but no new peaks/shoulders/features appear in the g(r) functions (Figs. 7, S7 and S8). From Figs. 7a-c, S7a-c and S8a-c, the intensity of the first peak decreases as the amount of R134a increases; in contrast, from Fig. 7d, S7d and S8d the intensity of the first peak increases as the refrigerant content reaches 15% but then decreases when the R134a concentration is 20% molar. The fact that the first peak in all of the COM g(r) functions appear around 0.5 nm and not at shorter distances (Figs. 7, S7 and S8), suggests that the DES species do not form strong directional interactions such as hydrogen bonds with the molecules of R134a. Examination of g(r) functions between different atoms of the refrigerant molecules and DES species (Figures 8 and S9-S10, Supporting Information) confirm this statement; the first peaks in all these atom-atom g(r) functions appear around 0.35-0.50 nm, and all have weak intensities. The sole exception is the


22

Page 23 of 35

first peak in the g(r) function involving the Cl anion and the H2 atoms in R134a (Figs. 8a, S9a and S10a), which has a relatively strong intensity in all systems, but appears at a distance of 0.4 nm. These results show that in all systems the hydrogen atoms of the refrigerant molecules interact strongly with the chlorine anion, but the atom-atom interactions between R134a and the cation and HBD species are weaker and do not seem to play a significant role in the solvation of the refrigerant. Therefore, we do not observe any preferential organization of the cation and HBD species around the R134a molecules of solute. Results at T = 343 K and 10.2 bar follow the same trends as those observed in Figs. 8, S9 and S10, and are not shown for brevity. 2.5

2.5

(a)

(b)

5% 10%

2

5% 10%

2

15%

15%

20%

20%

g(r)

1.5

g(r)

1.5

1

1

0.5

0.5

0

0 0

0.5

1

1.5

2

2.5

0

0.5

1

r, nm

1.5

2

2.5

r, nm

2.5

4.5

(c)

(d)

5%

5%

4

10%

2

10%

15%

3.5

15%

20%

3

20%

g(r)

1.5

g(r)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

2.5 2

1

1.5 1

0.5

0.5 0

0 0

0.5

1

1.5

2

2.5

0

0.5

r, nm

1

1.5

2

2.5

r, nm


23

Langmuir

Figure 7. Center-of-mass radial distribution functions g(r) for (a) Choline-R134a, (b) ChlorineR134a, (c) Ethylene glycol-R134a, and (d) R134a-R134a pairs in systems containing R134a (5%-20%) and choline chloride + ethylene glycol, at 298 K and 4.9 bar. 2.5

2.5

(a)

5% 10% 15% 20%

2

2.5

(b)

5% 10% 15% 20%

2

g(r)

g(r) 1

1

0.5

0.5

0.5

0 0

2.5

0.5

1

1.5 r, nm

2

2.5

0 0

2.5

(d)

5% 10% 15% 20%

2

0.5

1

1.5 r, nm

2

2.5

0

2.5

(e)

5% 10% 15% 20%

2

1

0.5

0.5

0.5

0

2.5

0.5

1

1.5 r, nm

2

2.5

5% 10% 15% 20%

2

0.5

1

1.5 r, nm

2

(h)

(f)

2.5

5% 10% 15% 20%

2.5

0

0.5

1

1.5 r, nm

2

2.5

5% 10% 15% 20%

2 1.5 g(r)

1.5

2

0 0

2.5

(g)

1.5 r, nm

g(r)

g(r) 1

0

1

1.5

1

0

0.5

2

1.5

g(r)

1.5

5% 10% 15% 20%

1.5

1

0

(c)

2

1.5

g(r)

1.5

g(r)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 35

1

1

0.5

0.5

0

0 0

0.5

1

1.5 r, nm

2

2.5

0

0.5

1

1.5 r, nm

2

2.5

Figure 8. Atom-atom radial distribution functions g(r) in systems containing R134a (5%-20%) and choline chloride + ethylene glycol, at 298 K and 4.9 bar, for the following pairs: (a) H2 (R134a) – Cl (anion), (b) F2 (R134a) – HO1 (cation), (c) H2 (R134a) – OAA (HBD), (d) F1 (R134a) – HAC (HBD), (e) F2 (R134a) – HAA (HBD), (f) F2 (R134a) – HAC (HBD), (g) F1 (R134a) – HAA (HBD), and (h) F1 (R134a) – HO1 (cation).


24

Page 25 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Average interaction energies (electrostatic + van der Waals) between molecules of R134a and the DES species in all our systems are reported in Tables 5 (for T = 298 K, P = 4.9 bar) and S3 (for T = 343 K and P = 10.2 bar, Supporting Information). In general, all interaction energies increase monotonically as the molar concentrations of R134a become larger in the different systems. For both sets of conditions of T and P, in the system where HBD = ethylene glycol, the interactions of the refrigerant with the cation are the strongest, followed by those with the HBD, then with other R134a molecules, and lastly with the anions. In contrast, in the DES containing urea, the strength of the refrigerant interactions follows the order cation-R134a-HBD-anion, whereas when HBD = glycerol, the order is HBD-cation-R134a-anion. These results suggest that the cations are the dominant species that dissolve R134a in DESs containing ethylene glycol or urea, whereas glycerol is the predominant dissolvent species of refrigerant in its DES. The small size of the anion makes it to have the weakest interactions with R134a, however Cl has the strongest g(r) peak intensity with the H2 atoms of R134a (Figs. 8a, S9a and S10a). At concentrations of refrigerant below 20%, the strongest R134a-R134a interactions are always observed when HBD = urea or glycerol, which is related to the fact that these systems show a stronger tendency for R134a molecules to cluster together in the liquid phase as its concentration is raised, as compared to the DES containing ethylene glycol. Table 5. Average interaction energies between refrigerant molecules and other species in our systems at 298 K and 4.9 bar

HBD Urea

R134a mole % 5% 10% 15% 20%

Average interaction energies (kJ/mol) R134aR134aR134aR134acation anion HBD R134a -1671.1 -280.8 -914.3 -407.7 -3078.3 -590.8 -1829.1 -902.4 -3335.2 -836.3 -1785.0 -2038.8 -3654.1 -1020.0 -1817.9 -3094.3

Glycerol

5% 10% 15% 20%

-1310.2 -2314.6 -2989.8 -3964.7

-349.8 -626.3 -892.0 -1237.7

-1419.7 -2633.9 -3496.1 -4386.5

-376.6 -931.6 -1680.2 -2304.5

Ethylene

5%

-1354.2

-356.9

-1212.8

-381.4


25

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

glycol

10% 15% 20%

-2500.8 -3297.5 -3868.5

Page 26 of 35

-708.0 -1014.0 -1270.8

-2185.4 -2958.6 -3672.1

-894.1 -1597.8 -2392.0

Dynamical properties such as viscosities and diffusivities play an important role in the potential application of the systems studied here in absorption refrigeration, as these properties can impact pumping operations as well as mass transfer processes in the absorber and desorber. In Figure 9 we show the diffusion coefficients of the cation, anion, HBD and refrigerant as a function of the refrigerant content for all systems studied, as computed from MD simulations at T = 298 K, P = 4.9 bar. These diffusivities were calculated using the following Einstein relation involving the slope of the mean-squared displacements (MSDs),81,82 as measured from our simulations: 2 1 d 1 N D = lim  ri ( t ) − ri ( 0 )  ∑ 6 t→∞ dt N i=1

(9)

We note that for the DES containing urea or glycerol, we could not reliably determine the diffusion coefficients of R134a molecules from the MSDs at the largest concentrations of refrigerant examined (15% and 20%), probably due to the clustering of refrigerant molecules observed in these systems; perhaps simulated times longer than 110 ns would be needed to accurately determine diffusivities of R134a in these systems. However, in general the diffusion coefficients of all species increase with increasing amounts of R134a, with the sole exception of choline chloride + glycerol at 20% of R134a, for which the diffusivities of refrigerant and cation seem to drop compared to the values found at 15% of R134a (here we note that the diffusivities computed for the refrigerant and the cation for 20% of R134a in this DES had significantly larger uncertainties, again suggesting that simulations longer than 110 ns are needed for this particular system). For all systems, the refrigerant has the largest mobility, followed by the HBD, the anion and the cation. For any given % of R134a, species have the largest diffusivities in the systems where the HBD is ethylene glycol, followed by those where HBD = glycerol and lastly those systems where HBD = urea. These trends are in agreement with the fact that the DES choline chloride + ethylene glycol has the smallest viscosity among the systems studied, followed by choline chloride + glycerol and choline chloride + urea. Results at T = 343 K follow similar trends as those shown in Fig. 9, with larger values of D (not shown for brevity)


26

Page 27 of 35

2.5

D (10-11 m2/s)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3.5

(a)

(b)

3

2

(c)

15

2.5 1.5

2

1

1.5

10

5

1 0.5

0.5

0

0 0

5

10

15

20

0 0

5

10

15

20

0

5

10

15

Mole % R134a Figure 9. Diffusion coefficients D of the different species in systems containing varying amounts of R134a in choline chloride + HBD, where HBD = urea (a); glycerol (b); or ethylene glycol (c). Red triangles = choline (cation); orange squares = chlorine (anion); green diamonds = HBD; and blue circles = R134a (refrigerant). T = 298 K, P = 4.9 bar.

4. Conclusions We have used computational tools to evaluate three working fluid mixtures for use in single-effect absorption refrigeration systems, where the generator (desorber) is powered by waste or solar heat. The working fluids studied here resulted from combining a widely used HFC refrigerant, R134a, with three common DESs formed by mixing choline chloride with either urea, glycerol or ethylene glycol as the HBD species. The COSMOtherm/TmoleX software package was used in combination with refrigerant data from NIST/REFPROP, to perform a thermodynamic evaluation of absorption refrigeration cycles using the proposed working fluid mixtures. Afterwards, classical MD simulations of the three mixtures were performed at the operation temperatures and pressures of the absorber (T = 298 K and P = 4.9 bar) and desorber (T = 343 K and P = 10.2 bar), to gain insights of these systems at the molecular level and fundamentally understand how the interactions between the different components affect macroscopic properties relevant to absorption refrigeration. In order to validate our COSMO-RS calculations, and due to the lack of experimental data for the R134a-DESs mixtures considered here, we compared results from COSMO-RS with reported data for (1) pure DESs, (2) HFCs in the IL [bmim+][PF6-], and (3) absorption refrigeration cycles using HFCs in [bmim+][PF6-]. These comparisons indicate that refinements in the COSMO-RS models are needed for the results to match the reported data; nevertheless, in all cases the trends from the COSMO-RS


27

20

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 35

calculations were in qualitatively agreement with results from previous studies. These observations suggest that COSMO-RS can be used to perform a rapid, preliminary screening of a large number of working fluid mixtures, and select a few candidates for further exploration using experiments and molecular simulations. These latter approaches can be used to refine the COSMO-RS models, as well as to further select a few optimal working fluid mixtures for demonstration in lab-scale absorption refrigeration systems powered by solar or waste heat sources. Our thermodynamic evaluation of absorption refrigeration cycles using R134a-DESs mixtures indicates that larger efficiencies are obtained when R134a is combined with choline chloride + ethylene glycol, followed by the system where glycerol is the HBD, and finally that where HBD = urea. MD simulations of these last two systems indicate that the local density profiles of all species exhibit very sharp variations, as compared to those computed for R134a in choline chloride + ethylene glycol; these observations suggest that R134a molecules tend to cluster together more easily when HBD = urea or HBD = glycerol, which is supported by measurements of the R134a-R134a interaction energies in our systems. In addition, results for the Henry’s law constants of refrigerant in these systems, as computed from COSMO-RS calculations and MD simulations, indicate that R134a has the largest solubility when ethylene glycol is the HBD, followed by the systems where HBD = glycerol and HBD = urea. Our results indicate that the R134a-R134a interactions are weaker when HBD = ethylene glycol than in the other DES examined. The g(r) functions suggest that in all systems, the DES species do not form strong directional interactions (e.g., hydrogen bonds) with the R134a molecules. Relatively strong interactions are observed between the Cl anions and the hydrogen atoms in R134a, but the atom-atom interactions between R134a and the cation and HBD species are weaker and do not play a significant role in the solvation of the refrigerant. Examination of interaction energies suggest that the cations are the dominant species that dissolve R134a in DESs containing ethylene glycol or urea, whereas glycerol is the predominant dissolvent species of refrigerant in its DES. Computation of the diffusion coefficients indicate that R134a has the largest mobility, followed by the HBD, the anion and the cation. The diffusion coefficients are the largest in the systems where HBD = ethylene glycol, followed by those where HBD = glycerol and HBD = urea; these trends agree with those observed for experimental measurements of the viscosity of these DESs. This work is our first step towards our long-term goal of designing and


28

Page 29 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

demonstrating optimal working fluid mixtures for use in solar- or waste heat-powered absorption refrigeration systems. We note that here we have studied a widely used HFC (R134a) in very common DESs (choline chloride + either urea, glycerol or ethylene glycol). We could use the approach described in this paper, to screen a large number of candidate refrigerant and absorption solvents, and select a few systems for further study using molecular simulations and experiments, and eventually demonstrate optimal mixtures in absorption refrigeration cycles. We will focus on these tasks in our following studies.

Acknowledgements This work was partially supported by funds from the National Science Foundation (CAREER Award CBET-1649455), Louisiana State University and Northeastern University. Highperformance computational resources for this research were provided by Research Computing at Northeastern University (https://www.northeastern.edu/rc/), High Performance Computing at Louisiana State University (http://www.hpc.lsu.edu), and Louisiana Optical Network Initiative (http://www.loni.org).

Supporting Information Tables S1 (comparison of Henry’s law constant of R134a in [bmim+][PF6-] from COSMO-RS with previous results), S2 (comparison of efficiency of absorption refrigeration cycles of HFCs in [bmim+][PF6-] from COSMO-RS with previous results) and S3 (average interaction energies between R134a and other DES species at 343 K and 10.2 bar); Figures S1-S2 (comparison of solubility of HFCs in [bmim+][PF6-] from COSMO-RS with previous results), S3-S4 (snapshots from MD simulations of R134a in choline chloride + urea or glycerol), S5-S6 (snapshots and local density profiles of R134a in choline chloride + glycerol at 298 K, and in choline chloride + ethylene glycol at 343 K), S7-S8 (COM g(r) of systems of R134a in choline chloride + urea or glycerol), S9-S10 (atom-atom g(r) of systems of R134a in choline chloride + urea or glycerol). This information is available free of charge via the Internet at http://pubs.acs.org/. References


29

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 35

(1) Basile, S.: Cool: How Air Conditioning Changed Everything; Fordham University Press: New York, 2014. (2) Cox, S.: Losing Our Cool: Uncomfortable Truths About Our Air-Conditioned World (and Finding New Ways to Get Through the Summer); New Press: New York, 2010. (3) EIA Total Energy Monthly Energy Review. https://www.eia.gov/totalenergy/data/monthly/index.php - consumption (accessed June 2017) (4) EPA Heating and cooling no longer majority of U.S. home energy use. https://www.eia.gov/todayinenergy/detail.php?id=10271&src=%E2%80%B9%20Consumption% 20%20%20%20%20%20Residential%20Energy%20Consumption%20Survey%20(RECS)-f3 (accessed June 2017) (5) EPA Greenhouse Gas Emmisions. https://www.epa.gov/ghgemissions (accessed June 2017). (6) Arora, R. C.: Refrigeration and Air Conditioning; PHI Learning: New Delhi, 2010. (7) Kang, Y.; Kunugi, Y.; Kashiwagi, T. Review of Advanced Absorption Cycles: Performance Improvement and Temperature Lift Enhancement. Int. J. Refrig. 2000, 23, 388-401. (8) Kim, D. S.; Infante Ferreira, C. A. Solar Refrigeration Options – A State-of-theArt Review. Int. J. Refrig. 2008, 31, 3-15. (9) Srikhirin, P.; Aphornratana, S.; Chungpaibulpatana, S. A Review of Absorption Refrigeration Technologies. Renewable Sustainable Energy Rev. 2001, 5, 343-372. (10) Plechkova, N. V.; Seddon, K. R. Applications of Ionic Liquids in the Chemical Industry. Chem. Soc. Rev. 2008, 37, 123-150. (11) Ionic Liquids in Synthesis 2nd ed; Wasserscheid, P.; Welton, T., Eds.; WileyVCH: Weinheim, 2008. (12) Zhang, Q.; De Oliveira Vigier, K.; Royer, S.; Jerome, F. Deep Eutectic Solvents: Syntheses, Properties and Applications. Chem. Soc. Rev. 2012, 41, 7108-7146. (13) Francisco, M.; van den Bruinhorst, A.; Kroon, M. C. Low-TransitionTemperature Mixtures (LTTMs): A New Generation of Designer Solvents. Angew. Chem., Int. Ed. 2013, 52, 3074-3085. (14) Dong, L.; Zheng, D.; Sun, G.; Wu, X. Vapor–Liquid Equilibrium Measurements of Difluoromethane + [Emim]OTf, Difluoromethane + [Bmim]OTf, Difluoroethane + [Emim]OTf, and Difluoroethane + [Bmim]OTf Systems. J. Chem. Eng. Data 2011, 56, 36633668. (15) Dong, L.; Zheng, D.; Wu, X. Working Pair Selection of Compression and Absorption Hybrid Cycles through Predicting the Activity Coefficients of Hydrofluorocarbon + Ionic Liquid Systems by the UNIFAC Model. Ind. Eng. Chem. Res. 2012, 51, 4741-4747. (16) Kim, S.; Kim, Y. J.; Joshi, Y. K.; Kohl, P. A.; Fedorov, A. G. Absorption Heat Pump/Refrigeration System Utilizing Ionic Liquid and Hydrofluorocarbon Refrigerants. J. Electron. Packag. 2012, 134, 031009-031009. (17) Kim, S.; Patel, N.; Kohl, P. A. Performance Simulation of Ionic Liquid and Hydrofluorocarbon Working Fluids for an Absorption Refrigeration System. Ind. Eng. Chem. Res. 2013, 52, 6329-6335. (18) Kim, Y. J.; Kim, S.; Joshi, Y. K.; Fedorov, A. G.; Kohl, P. A. Thermodynamic Analysis of an Absorption Refrigeration System with Ionic-Liquid/Refrigerant Mixture as a Working Fluid. Energy 2012, 44, 1005-1016.


30

Page 31 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(19) Shiflett, M. B.; Yokozeki, A. Vapor−Liquid−Liquid Equilibria of Pentafluoroethane and Ionic Liquid [bmim][PF6] Mixtures Studied with the Volumetric Method. J. Phys. Chem. B 2006, 110, 14436-14443. (20) Shiflett, M. B.; Yokozeki, A. Vapor−Liquid−Liquid Equilibria of Hydrofluorocarbons + 1-Butyl-3-methylimidazolium Hexafluorophosphate. J. Chem. Eng. Data 2006, 51, 1931-1939. (21) Shiflett, M. B.; Yokozeki, A. Solubility and Diffusivity of Hydrofluorocarbons in Room-Temperature Ionic Liquids. AIChE J. 2006, 52, 1205-1219. (22) Shiflett, M. B.; Yokozeki, A. Solubility Differences of Halocarbon Isomers in Ionic Liquid [emim][Tf2N]. J. Chem. Eng. Data 2007, 52, 2007-2015. (24) Zheng, D.; Dong, L.; Huang, W.; Wu, X.; Nie, N. A Review of Imidazolium Ionic Liquids Research and Development Towards Working Pair of Absorption Cycle. Renewable Sustainable Energy Rev. 2014, 37, 47-68. (25) Shiflett, M. B.; Yokozeki, A. Absorption Cycle Utilizing Ionic Liquids and Water as Working Fluids. US 8506839 B2, 2013. (26) Diedenhofen, M.; Klamt, A. COSMO-RS as a Tool for Property Prediction of IL Mixtures - A Review. Fluid Phase Equilib. 2010, 294, 31-38. (27) Eckert, F.; Klamt, A. Fast solvent screening via quantum chemistry: COSMO-RS approach. AIChE J. 2002, 48, 369-385. (28) Klamt, A. Conductor-Like Screening Model for Real Solvents - A New Approach to the Quantitative Calculation of Solvation Phenomena. J. Phys. Chem. 1995, 99, 2224-2235. (29) Klamt, A.; Eckert, F.; Arlt, W.: COSMO-RS: An Alternative to Simulation for Calculating Thermodynamic Properties of Liquid Mixtures. Annual Review of Chemical and Biomolecular Engineering, Vol 1; Annual Review of Chemical and Biomolecular Engineering; Annual Reviews: Palo Alto, 2010; Vol. 1; pp 101-122. (30) Klamt, A.; Jonas, V.; Burger, T.; Lohrenz, J. C. W. Refinement and Parametrization of COSMO-RS. J. Phys. Chem. A 1998, 102, 5074-5085. (31) COSMOtherm C3.0, release 1601. http://www.cosmologic.de/products/cosmotherm.html (accessed May 2017). (32) Tmolex V4.1. http://www.cosmologic.de/turbomole/tmolex.html (accessed May 2017). (33) Zhang, X. C.; Liu, Z. P.; Wang, W. C. Screening of Ionic Liquids to Capture CO2 by COSMO-RS and Experiments. AIChE J. 2008, 54, 2717-2728. (34) Farahipour, R.; Mehrkesh, A.; Karunanithi, A. T. A Systematic Screening Methodology Towards Exploration of Ionic Liquids for CO2 Capture Processes. Chem. Eng. Sci. 2016, 145, 126-132. (35) Bara, J. E.; Moon, J. D.; Reclusado, K. R.; Whitley, J. W. COSMOTherm as a Tool for Estimating the Thermophysical Properties of Alkylimidazoles as Solvents for CO2 Separations. Ind. Eng. Chem. Res. 2013, 52, 5498-5506. (36) Shannon, M. S.; Tedstone, J. M.; Danielsen, S. P. O.; Hindman, M. S.; Irvin, A. C.; Bara, J. E. Free Volume as the Basis of Gas Solubility and Selectivity in Imidazolium-Based Ionic Liquids. Ind. Eng. Chem. Res. 2012, 51, 5565-5576. (37) Thermophysical Properties of Fluid Systems. http://webbook.nist.gov/chemistry/fluid/ (accessed May 2017). (38) Reference Fluid Thermodynamic and Transport Properties Database (REFPROP) Version 9.1. https://www.nist.gov/srd/refprop (accessed May 2017)


31

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 35

(39) Deglmann, P.; May, K.; Furche, F.; Ahlrichs, R. Nuclear Second Analytical Derivative Calculations Using Auxiliary Basis Set Expansions. Chem. Phys. Lett. 2004, 384, 103-107. (40) Rappoport, D.; Furche, F. Property-Optimized Gaussian Basis Sets for Molecular Response Calculations. J. Chem. Phys. 2010, 133, 134105. (41) Sierka, M.; Hogekamp, A.; Ahlrichs, R. Fast Evaluation of the Coulomb Potential for Electron Densities Using Multipole Accelerated Resolution of Identity Approximation. J. Chem. Phys. 2003, 118, 9136-9148. (42) Weigend, F. Accurate Coulomb-fitting Basis Sets for H to Rn. Phys. Chem. Chem. Phys. 2006, 8, 1057-1065. (43) Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A 1988, 38, 3098-3100. (44) Perdew, J. P. Density-Functional Approximation for the Correlation Energy of the Inhomogeneous Electron Gas. Phys. Rev. B 1986, 33, 8822-8824. (45) Schäfer, A.; Huber, C.; Ahlrichs, R. Fully Optimized Contracted Gaussian Basis Sets of Triple Zeta Valence Quality for Atoms Li to Kr. J. Chem. Phys. 1994, 100, 5829-5835. (46) Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence, Triple Zeta Valence and Quadruple Zeta Valence Quality for H To Rn: Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297-3305. (47) D'Agostino, C.; Harris, R. C.; Abbott, A. P.; Gladden, L. F.; Mantle, M. D. Molecular Motion and Ion Diffusion in Choline Chloride Based Deep Eutectic Solvents Studied by 1H Pulsed Field Gradient NMR Spectroscopy. Phys. Chem. Chem. Phys. 2011, 13, 2138321391. (48) Shah, D.; Mjalli, F. S. Effect of Water on the Thermo-Physical Properties of Reline: an Experimental and Molecular Simulation Based Approach. Phys. Chem. Chem. Phys. 2014, 16, 23900-23907. (49) Ciocirlan, O.; Iulian, O.; Croitoru, O. Effect of Temperature on the Physicochemical Properties of Three Ionic Liquids Containing Choline Chloride. Rev. Chim. (Bucharest, Rom.) 2010, 61, 721-723. (50) Xie, Y.; Dong, H.; Zhang, S.; Lu, X.; Ji, X. Effect of Water on the Density, Viscosity, and CO2 Solubility in Choline Chloride/Urea. J. Chem. Eng. Data 2014, 59, 33443352. (51) Abbott, A. P.; Harris, R. C.; Ryder, K. S.; D'Agostino, C.; Gladden, L. F.; Mantle, M. D. Glycerol eutectics as sustainable solvent systems. Green Chem. 2011, 13, 82-90. (52) Leron, R. B.; Wong, D. S. H.; Li, M.-H. Densities of a deep eutectic solvent based on choline chloride and glycerol and its aqueous mixtures at elevated pressures. Fluid Phase Equilib. 2012, 335, 32-38. (53) Shahbaz, K.; Mjalli, F. S.; Hashim, M. A.; AlNashef, I. M. Prediction of the surface tension of deep eutectic solvents. Fluid Phase Equilib. 2012, 319, 48-54. (54) Mjalli, F. S.; Abdel Jabbar, N. M. Acoustic investigation of choline chloride based ionic liquids analogs. Fluid Phase Equilib. 2014, 381, 71-76. (55) Leron, R. B.; Li, M.-H. Molar heat capacities of choline chloride-based deep eutectic solvents and their binary mixtures with water. Thermochim. Acta 2012, 530, 52-57. (56) Perkins, S. L.; Painter, P.; Colina, C. M. Molecular Dynamic Simulations and Vibrational Analysis of an Ionic Liquid Analogue. J. Phys. Chem. B 2013, 117, 10250-10260.


32

Page 33 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(57) Perkins, S. L.; Painter, P.; Colina, C. M. Experimental and Computational Studies of Choline Chloride-Based Deep Eutectic Solvents. J. Chem. Eng. Data 2014, 59, 3652-3662. (58) Hess, B.; Kutzner, C.; Spoel, D. v. d.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435-447. (59) Lindahl, E.; Hess, B.; Spoel, D. v. d. GROMACS 3.0: A Package for Molecular Simulation and Trajectory Analysis. J. Mol. Model. 2001, 7, 306-317. (60) Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. GROMACS: Fast, Flexible, and Free. J. Comput. Chem. 2005, 26, 1701-1718. (61) GROMACS User Manual version 5.0.4. ftp://ftp.gromacs.org/pub/manual/manual-5.0.4.pdf (Accessed May 2017). (62) Berendsen, H. J. C.; Spoel, D. v. d.; Drunen, R. v. GROMACS: A MessagePassing Parallel Molecular Dynamics Implementation. Comput. Phys. Commun. 1995, 91, 43-56. (63) Peguin, R. P. S.; Kamath, G.; Potoff, J. J.; Rocha, S. R. P. d. All-Atom Force Field for the Prediction of Vapor-Liquid Equilibria and Interfacial Properties of HFA134a. J. Phys. Chem. B 2009, 113, 178-187. (64) Schuttelkopf, A. W.; Aalten, D. M. F. v. PRODRG: A Tool for High-Throughput Crystallography of Protein-Ligand Complexes. Acta Crystallogr., Sect. D: Struct. Biol. 2004, 60, 1355-1363. (65) Martinez, L.; Andrade, R.; Birgin, E. G.; Martinez, J. M. PACKMOL: A Package for Building Initial Configurations for Molecular Dynamics Simulations. J. Comput. Chem. 2009, 30, 2157-2164. (66) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual Molecular Dynamics. J. Mol. Graphics 1996, 14, 33-38. (67) Larsen, G. S.; Lin, P.; Hart, K. E.; Colina, C. M. Molecular Simulations of PIM1-like Polymers of Intrinsic Microporosity. Macromolecules 2011, 44, 6944-6951. (68) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G. A Smooth Particle Mesh Ewald Method. J. Chem. Phys. 1995, 103, 8577-8593. (69) Bussi, G.; Donadio, D.; Parrinello, M. Canonical Sampling through Velocity Rescaling. J. Chem. Phys. 2007, 126, 014101. (70) Hess, B. P-LINCS: A Parallel Linear Constraint Solver for Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 116-122. (71) Bennett, C. H. Efficient Estimation of Free-Energy Difference from Monte Carlo Data. J. Comput. Phys. 1976, 22, 245-268. (72) Shirts, M. R.; Pande, V. S. Comparison of Efficiency and Bias of Free Energies Computed by Exponential Averaging, the Bennett Acceptance Ratio, and Thermodynamic Integration. J. Chem. Phys. 2005, 122, 144107. (73) Lee, L. L.: Molecular Thermodynamics of Nonideal Fluids; Butterworth: Boston, 1988. (74) Shen, Y.; He, X.; Hung, F. R. Structural and Dynamical Properties of a Deep Eutectic Solvent Confined Inside a Slit Pore. J. Phys. Chem. C 2015. (75) Liyana-Arachchi, T. P.; Zhang, Z.; Vempati, H.; Hansel, A. K.; Stevens, C.; Pham, A. T.; Ehrenhauser, F. S.; Valsaraj, K. T.; Hung, F. R. Green Leaf Volatiles on Atmospheric Air/Water Interfaces: A Combined Experimental and Molecular Simulation Study. J. Chem. Eng. Data 2014, 59, 3025-3035.


33

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 35

(76) Yadav, A.; Trivedi, S.; Rai, R.; Pandey, S. Densities and Dynamic Viscosities of (Choline Chloride + Glycerol) Deep Eutectic Solvent and its Aqueous Mixtures in the Temperature Range (283.15–363.15) K. Fluid Phase Equilib. 2014, 367, 135-142. (77) Yadav, A.; Pandey, S. Densities and Viscosities of (Choline Chloride + Urea) Deep Eutectic Solvent and Its Aqueous Mixtures in the Temperature Range 293.15 K to 363.15 K. J. Chem. Eng. Data 2014, 59, 2221-2229. (78) Abbott, A. P.; Capper, G.; Davies, D. L.; Rasheed, R. K.; Tambyrajah, V. Novel Solvent Properties of Choline Chloride/Urea Mixtures. Chem. Commun. 2003, 70-71. (79) Prausnitz, J. M.; Lichtenthaler, R. N. Gomes de Azevedo, E. Molecular Thermodynamics of Fluid-Phase Equilibria; 3rd Ed.; Prentice Hall: Upper Saddle River, 1998 (80) Hsieh, C.-M.; Sandler, S. I.; Lin, S.-T. Improvements of COSMO-SAC for vapor– liquid and liquid–liquid equilibrium predictions. Fluid Phase Equilib. 2010, 297, 90-97. (81) Allen, M. P.; Tildesley, D. J.: Computer Simulation of Liquids; Clarendon Press: Oxford, 1987. (82) Frenkel, D.; Smit, B.: Understanding Molecular Simulation: From Algorithms to Applications; 2nd ed.; Academic Press: San Diego, 2002.


34

Page 35 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Condenser

Generator

Solution heat exchanger Valve Pump Valve

Evaporator

Absorber

Figure for the TOC


35

Computational Evaluation of Mixtures of ... - ACS Publications

Recommend Documents