Physicochemical interactions of ionic liquids with coal; the viability of ionic liquids for pre-treatments in coal liquefaction

Physicochemical interactions of ionic liquids with coal; the viability of ionic liquids for pre-treatments in coal liquefaction

Fuel 143 (2015) 244–252 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Physicochemical interactions ...

4MB Sizes 19 Downloads 305 Views

Fuel 143 (2015) 244–252

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Physicochemical interactions of ionic liquids with coal; the viability of ionic liquids for pre-treatments in coal liquefaction Joshua Cummings a, Kalpit Shah b,⇑, Rob Atkin a, Behdad Moghtaderi b a b

Discipline of Chemistry, The University of Newcastle, Callaghan, NSW 2308, Australia Chemical Engineering, Faculty of Engineering and Built Environment, The University of Newcastle, Callaghan, NSW 2308, Australia

h i g h l i g h t s  Interactions between coal and ionic liquids have been examined via SEM and TGA.  The viability of ionic liquids for pretreatments in liquefaction is investigated.  Morphological and thermal properties of the coals were altered with pretreatment.

a r t i c l e

i n f o

Article history: Received 26 August 2014 Accepted 13 November 2014 Available online 1 December 2014 Keywords: Ionic liquid Coal liquefaction Low temperature pre-treatment Efficiency improvement

a b s t r a c t Three Australian sub-bituminous coals were treated with three different ionic liquids (ILs) at a temperature of 100 °C. The thermal behaviour of these treated coals were compared against raw coals via pyrolysis experiments in a Thermogravimetric Analyser. Morphological comparisons were also made via Scanning Electron Microscopy. Among the studied ILs, 1-butyl-3-methylimidazolium chloride [Bmim][Cl] was found to perform the most consistently in being able to alter the thermal and morphological properties of most of the coals used. It is posited that this may be due to the large difference in charge density between the delocalised charge of the large bmim cation and the chloride anion which allows this IL to disrupt the cross linked network of coal. It was also found that the interactions of the ionic liquids are coal specific, for instance none of the ionic liquids were able to change the thermal properties of coal A. Moreover, the results indicated that among the studied coals, coal R showed the highest mass loss during pyrolysis in TGA and coal C showed the highest amount of swelling and fragmentation in SEM images. The results displayed in this study indicate that the potential for ionic liquids to be used as pre-treatments in coal liquefaction is promising. Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

1. Introduction Coal is a heterogeneous material made up of both organic and inorganic components. The differing components of coal influence aspects of its behaviour, such as gasification and combustion reactivity. It is Australia’s primary source of energy, making up roughly three quarters of Australia’s Electricity generated [1]. Additionally, coal plays a key economic role in Australia’s exports; Australia is the 4th largest producer of coal and the largest exporter in the world [1]. Conversely to this, Australia is heavily dependent on imports to satisfy our liquid fuel requirements, especially petroleum. Our dependence on imported liquid fuels has increased from 60% in 2000 to over 90% of our transport fuel demand today [2].

⇑ Corresponding author. Tel.: +61 240339332; fax: +61 240339383. E-mail address: [email protected] (K. Shah). http://dx.doi.org/10.1016/j.fuel.2014.11.042 0016-2361/Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

Because of this increasing reliance on imports, our economy is left susceptible to fluctuating market prices. [3]. Coal liquefaction is a process whereby coal is transferred into liquid fuel. There are two types of coal liquefaction; Indirect Coal Liquefaction (ICL), which involves the production of liquid fuels via an intermediate gasification step, where a mixture of Carbon monoxide and Hydrogen (syngas) are produced via the gasification of coal. This gas is then used to construct hydrocarbon chains of a range of lengths via condensation in order to produce liquid fuels [3–5]. Direct Coal liquefaction (DCL) involves the splitting of the convoluted, 3-dimensional cross-linked macrostructure of coal via high temperatures and pressures followed by hydrogenation in order to obtain shorter hydrocarbon chains of desired lengths with the intention of obtaining liquid fuel. This is usually done in the presence of a catalyst [3,6–8]. The mechanisms behind DCL are complex and not fully understood yet, however it is generally agreed upon that the following

245

J. Cummings et al. / Fuel 143 (2015) 244–252

steps take place; the coal macrostructure is broken down into various radicals, hydrogen then caps these radical moieties in order to form shorter hydrocarbon chains with a higher H/C ratio [7,9]. This process is usually done in the presence of a catalyst, which helps to distribute hydrogen into the coal structure using solvents. A significant issue that is a hindrance to the application of DCL is the large capital and operating costs [10]. The high pressures (15–20 MPa), temperatures (380–450 °C) and amount of hydrogen required for this process means that the initial investment costs associated with commercialising this process are quite high [6,11]. Increasing the efficiency of this process and thus decreasing the operational and capital costs required has been an area of interest for many [12,13]. This can be done by via improved catalysts, optimising process parameters and with various pretreatments. Pre-treatments are utilised in order to swell and fracture the coal before it undergoes DCL. This is advantageous as it enables the partial breakdown of the coal macro structure, and also allows the porosity of the coal to increase, which facilitates the dispersion of hydrogen donor solvent and catalyst onto the coal structure [9,14]. Numerous studies have focused on the use of organic solvents as pre-treatment of coal, such as toluene, hexane, NMP, pyridine and phenanthridine to fragment, swell or dissolve coal [4]. The organic solvents usually employed however have drawbacks associated with them, these include; cost, losses during their use, degradation, their recovery after use and long term performance. The recent push for industries to employ green processes also means that the toxicity of these organic solvents is another major issue hindering their industrial applicability [15]. A type of solvent that potentially might not have these disadvantages are ionic liquids (ILs). Ionic liquids are salts that have a melting point below 100 °C [16]. The first ionic liquid, ethylammonium nitrate (Fig. 1), was synthesised in 1914 [17]. In the last few decades there has been a surge of interest in these salts, this is largely due to their unique properties. ionic liquids are green solvents with negligible vapour pressure and they often exhibit exceptional thermal stability. It has been reported that some ILs can be thermally stable up to temperatures of 450 °C [18]. Furthermore, they are not flammable at room temperatures [19]. But the most promising property of ionic liquids is their tunability. ILs are often referred to as designer solvents due to the fact that their properties can be tuned for a specific purpose via changing their ion composition [20]. The amount of ion combinations available are vast; it has been estimated that there are as many as 1018 possible ionic liquids [21]. ILs have been studied as solvents extensively in recent times. Literature suggests their possible use in solvent extraction of biomass and biofuel production [15]. However, the studies on coal and IL interactions are quite limited in the open literature [15,22,23]. Firstly, Painter et al. [22] studied the dispersion and dissolution of Illinois No. 6 coal into different ILs. It was concluded in this study that only certain ILs were able to disperse and fragment some coals. In reply to that, Shah et al. [15] showed that the coal–IL interactions may be coal maceral specific. Shah et al. [15] observed that vitrinite rich coal was found to be swelled during the IL treatment whilst inertinite rich coal was severely fragmented.

Fig. 1. Molecular structure of the first synthesised ionic liquid (Ethyl Ammonium Nitrate).

Moreover, the dissolution of vitrinite rich coal in IL was found to be >30% compared to intertinite rich coal. Although few studies are published in this area, the science behind the interactions of different coals and ILs is still not properly understood. The number of coals and ILs studied in the literature are very limited and it is highly recommended that more samples of coals and ILs should be added in the test matrix in order to define/correlate the mechanisms of coal-IL interactions. Therefore, the current study is looking at interactions of three different Australian sub-bituminous coal with three different ILs. More specifically, the study of IL’s application in pre-treatment for direct coal liquefaction is quite limited in the existing literature. Therefore, the current paper investigates the viability of ILs for the coal pre-treatment in liquefaction.

2. Experimental Three types of Australian subbituminous coals were used in this investigation, which will be referred to henceforth as C, A and R. These were ground down to a size of roughly 150–212 lm. The proximate and ultimate analysis of these can be seen in Table 1 below. Three Ionic Liquids were used in this investigation (Table 2): 1. 1-Ethyl-3-methylimidazolium dicyanamide [Emim][DCA], 2. 1-Butyl-3-methylimidazolium chloride [Bmim][Cl] and 3. 1-Butyl-3-methylimidazolium trycyanomethanide [Bmim][TCM] The above selected ILs were able to dissolve multiple aromatics from several petrochemical streams and hence were chosen based on the COSMO-RS screening carried out by Hansmeier [24]. Samples were made up consisting of coal and ionic liquid at a volume ratio of 20:80 respectively. The IL and coal were placed in a jar with a magnetic stirrer; the jar was then tightly sealed and placed in an oil bath to be heated for 3 h at 100 °C as can be seen in Fig. 2. After 3 h the jar was taken out of the oil bath, this mixture was then washed with 100 ml of distilled water and filtered using filter paper (pore size 11 lm). The water/IL mixture was placed in the oven in order to recover the IL and the coal samples were placed in the oven at low temperatures (80 °C) to dry. However, based on the previous literature on graphite interactions with ionic liquid [19], it was hypothesised that a single water wash may not be sufficient enough to separate IL from coal, as some IL may still remain trapped in pores. Therefore, multiple distilled water washes (5–6 times) were utilised followed by conductivity measurements of the washed sample in order to check for the presence of any IL adsorbed to the coal surface. Two different types of

Table 1 Proximate and ultimate analysis of the coals used. Coals C

A

R

Proximate analysis Q (kJ/kg) Moisture (%) Ash (%) Volatile Matter (%) Fixed Carbon (%)

18,026 3.9 32.5 35.9 64.1

24,956 1.5 23.0 50.6 49.4

26,748 3.7 9.8 35.9 64.1

Ultimate analysis Carbon (%) Hydrogen (%) Nitrogen (%) Sulphur (%) Oxygen (%)

73.8 4.3 1.1 0.3 20.5

78.3 6.7 1.1 0.7 13.2

77.2 5.2 2.0 0.7 15.0

246

J. Cummings et al. / Fuel 143 (2015) 244–252

Table 2 The structures and abbreviations of the ionic liquids used in this study. Ionic liquid

Structure

Abbreviation

1-Ethyl-3methylimidazolium dicyanamide

Emim DCA

1-Butyl-3methylimidazolium chloride

Bmim Cl

1-Butyl-3methylimidazolium tricyanomethanide

Bmim TCM

Magnetic Stirrer

Olive oil 90-100 oC

Hot Plate

Fig. 2. Schematic representation of the experimental set up.

treated coal samples were obtained: 1. single distilled water wash without conductivity measurements and 2. multiple distilled water washes with conductivity measurements. Pyrolysis experiments were performed for all raw and IL treated coals in a Thermogravimetric Analyser TGA Q50 in order to observe any changes in the thermal behaviour and mineralogy of the coals after treatment with ILs. Multiple runs were performed in order to ensure that the results were reliable. Roughly 10 mg of sample was placed in a platinum pan. The experiments were conducted in a nitrogen atmosphere at a flow rate of 100 ml/min and a ramp rate of 10 °C/min from 0 to 650 °C. Alongside this, Scanning Electron Microscopy (SEM) was also employed in order to observe changes in coal size, swelling, fragmentation and morphology with IL treatment. 3. Results After the experiments, ILs were stored in vials. The recovered ILs from various experiments can be seen in Fig. 3. In general, it was observed that physical appearance of ILs such as colour and state

have been changed after the treatment. For example, [Bmim][Cl] is generally found to be in solid state at room temperature. However, after recovery, it did not return to a solid state. The physical alterations of ILs such as change of state and colour indicates that some coal components might have been dissolved in the ILs during the water wash. As a possibility, water soluble fractions of coal such as water soluble salts and organically bound minerals can be considered. Also, water soluble organic compounds of coal such as aliphatic dicarboxylic acids and carbonyls, amines and saccharides might have been disbanded during water wash. Moreover, if the coal-IL treatment at 100 °C has allowed the liberation of low–high molecular weight volatiles such as aliphatic dicarboxylic acids, phenols, aromatic acids, cyclic acid, humic acid, pyridine, benzoquinoline, cyclopentenopyridine, aniline, phenylpyridine, N-benzylaniline, quinolone and naphthylamine; they also might be dissolved in IL–water mixture. However, from our past experience, such coal tars have very low solubility in water. In the current study, the most of the ILs used are found to be water soluble. Therefore, it can be predicted that during multiple water washes ILs might have been effectively removed from the macro-porous structures present in coal, leaving the tars on the filter paper along with coal due to their insolubility in water. To investigate this further in detail, TGA pyrolysis runs and SEM images of the treated coals have been studied. TGA results of the treated coals are given in Figs. 4–6. For coal A, as shown in Fig. 4(a), the TG profile before conductivity wash shows massive amounts of mass loss, with the [Bmim][Cl] treated coal losing roughly 60% of its mass. [Bmim][TCM] shows a similar trend in sharp mass loss, but at roughly 50%. Under closer inspection it can be seen that the drastic mass loss takes place over at roughly 250 °C. This is the temperature where devolatilization of coal starts. Nevertheless, it was found that [Bmim][Cl] also has the same decomposition temperature of 250 °C [25]. Therefore, to rule out the possibility of [Bmim][Cl] interfering with the thermal profile of coal, the conductivity washes were carried out. The results of the treated coal after thermal conductivity washes are presented in Fig. 4(b). It can be seen that after conductivity washes the treated coals have a very similar thermal profile to that of raw coal A. A majority of the work published in the past [22,23] on coalIL interactions have performed a single water/solvent wash. However, multiple washes coupled with conductivity measurements shown in this work have confirmed that such technique is essential in giving reliable and repeatable data. For coal A, however, it was found that none of the tested ILs were significantly able to alter its thermal profile. Fig. 5(a) illustrates a similar trend for coal C without conductivity test. When referring to the TG profiles of the coals before the conductivity wash, large and dramatic mass losses can be observed. The [Bmim][Cl] washed coal, again, shows mass losses of roughly 60%, with the majority of it occurring at 250 °C. The other IL’s exhibit similar trends with sharp mass losses of 50% and 40%. However, the TG profiles of the treated coals after conductivity washing show a differing trend. Unlike coal A, the treated C coals show a difference in mass loss when compared to the raw coal as shown in Fig. 5(b). The [Bmim][Cl] and [Emim][DCA] washed coals have a mass loss of roughly 15% more than the raw coal beginning at 300 °C, this is 100 °C less than the raw coal. These results imply that the ionic liquids have changed the thermal profile of coal C. This alteration of thermal properties indicates that ILs may be capable of interacting with and altering the cross-linked macrostructure at low temperatures. These alterations may allow the coal to be fractured and fragmented. Fig. 6 for coal R also showed a differing trend to that of Figs. 4 and 5 with regards to conductivity wash. It can be seen when referring to TG profiles obtained before a conductivity wash was performed (Fig. 6(a)), that there is no sharp mass loss peaks occurring at low

247

J. Cummings et al. / Fuel 143 (2015) 244–252

Fig. 3. Pure Emim DCA (left) with two recovered vials of Emim DCA after washing coal (A), pure Bmim Cl (left) with two recovered vials of Bmim Cl after washing coal (B) and pure Bmim TCM (left) with two recovered vials of Bmim TCM after washing coal (C).

1

1

(a)

0.9 0.8

0.8

Mass %

Mass %

(b)

0.9

0.7 0.6

Emim DCA

0.7 0.6

Emim DCA

0.5

Bmim TCM

Bmim Cl

Bmim Cl

0.5 0.4

Bmim TCM Raw A

0

100

200

300

400

500

600

0.4

700

Raw A

0

100

200

300

T (oC)

400

500

600

700

T (oC)

Fig. 4. TG profiles of raw coal A and its IL washed counterparts before conductivity washing (a) and after conductivity washing (b).

1

1 0.9

(a)

0.9 0.8

Mass %

Mass %

0.8

0.7 Emim DCA

0.6

Bmim Cl

0.5 0.4

(b)

Raw C

100

200

Emim DCA

0.6

Bmim Cl

0.5

Bmim TCM

0

0.7

300

400

T (oC)

500

600

700

0.4

Bmim TCM Raw C

0

100

200

300

400

500

600

700

T (oC)

Fig. 5. TG profiles of raw coal C and its IL washed counterparts before conductivity washing (a) and after conductivity washing (b).

248

J. Cummings et al. / Fuel 143 (2015) 244–252

1

1

0.9

0.7 Emim DCA

0.6

Bmim Cl Bmim TCM

0.5

0.4

100

0.7 Emim DCA

0.6

Bmim Cl Bmim TCM

0.5

Raw R

0

(b)

0.8

Mass %

Mass %

0.8

0.9

(a)

200

300

400

T (o C)

500

600

700

0.4

Raw R

0

100

200

300

400

500

600

700

T (oC)

Fig. 6. TG profiles of raw coal R and its IL washed counterparts before conductivity washing (a) and after conductivity washing (b).

Fig. 7. Scanning electron micrographs of raw coal C at 125 (A), washed by [Bmim][Cl] at 125 (B), washed by [Emim][DCA] at 125 (C) washed by [Bmim][TCM] at 125 (D) raw coal C at 1000 (E) and washed by [Bmim][Cl] at 1000 (F).

J. Cummings et al. / Fuel 143 (2015) 244–252

249

Fig. 8. Scanning electron micrographs of raw coal R at 125 (A), washed by [Bmim][Cl] at 125 (B), washed by [Emim][DCA] at 125 (C) washed by [Bmim][TCM] at 125 (D) raw coal R at 1000 (E) and washed by [Bmim][Cl] at 1000 (F).

temperatures. Instead there are steady declines in mass% with temperature increase. The mass loss observed in the treated coals (Fig. 6(b)) is that of roughly 35% for the highest and 30% for the lowest, however the best performing IL has changed from [Emim][DCA] to [Bmim][Cl]. This shows that ILs might not have been remained on the coal surface after initial wash. This may be attributed to the physical and chemical properties of the coal such as particle porosity and surface charge. To verify this, BET surface area measurements were carried out on all three studied coals. It was found that raw coal R has the lowest BET surface area than coals A and C (i.e. coal R = 0.53 m2/g, coal A = 1.2 m2/g and coal C = 11.6 m2/g). Furthermore, Scanning Electron Microscopy (SEM) was employed in order to observe any morphological changes in the coal after IL treatment. The images obtained here are for each raw coal and their respective treated coals with multiple conductivity washes. SEM images obtained (Fig. 7) show a distinct change before and after ionic liquid pre-treatment for coal C. Fig. 7(a) shows the image of raw coal C. Most of the coal particles can be seen to be of 150–212 lm in size, the particles are not fractured. A large difference can be observed between this and its [Emim][DCA] (b)/

[Bmim][Cl] (c) treated counterparts. Fig. 7(b) shows far smaller particle sizes, this indicates that extensive fracturing and fragmentation has occurred as a result of ionic liquid treatment, this fragmentation is even more pronounced in Fig. 7(c), where the coal particles look even finer in size. This fracturing might have resulted from the high levels of stress placed on the cross-linked structure which can be a result of swelling [22]. Another difference that can be observed is that the raw coal appears hard whereas the [Bmim][Cl] and [Emim][DCA] treated coals appear soft and have a powder-like texture. When referring to Fig. 7(d) a different morphology is observed, the [Bmim][TCM] treated coal appears largely unchanged from its raw counterpart, particle size is similar and the particles appear to be hard. The extent to which [Bmim][Cl] is able to fragment can be further illustrated when comparing Fig. 7(e) and (f). The SEM photo of raw coal C at 1000 shows the particles have a rigid surface, which is flat with minimal cracking, compare this to the [Bmim][Cl] treated coal at the same magnification and a large difference is observed. The porosity of the particles can be seen to have increased greatly. Overall, the morphology of coal C has been completely changed by [Bmim][Cl] and less so by [Emim][DCA], but was unchanged by [Bmim][TCM].

250

J. Cummings et al. / Fuel 143 (2015) 244–252

Fig. 9. Scanning Electron Micrographs of raw coal A at 125 (A), washed by [Bmim][Cl] at 125 (B), washed by [Emim][DCA] at 125 (C) washed by [Bmim][TCM] at 125 (D) raw coal A at 1000 (E) and washed by [Bmim][Cl] at 1000 (F).

Similar morphological changes can be observed for coal R, referring to Fig. 8(a–d) a similar, albeit, less pronounced trend is observed. The raw coal is observed to be roughly 200 lm in size and fairly homogeneous, its surface appears to be hard with very minimal cracking. Referring to coal R treated with [Bmim][Cl] and [Emim][DCA], differences are observed, the [Emim][DCA] (Fig. 8(b)) treated coal appears to be cracked and slightly fractured, however nowhere near the extent of the [Emim][DCA] treated C coal. Similarly to coal C, this trend is more prominent in the [Bmim][Cl] treated coal (Fig. 8(c)), more fracturing and fragmentation is observed, with smaller particle size on average, but still this is nowhere near the extent of the [Bmim][Cl] treated C coal. When referring to Fig. 8(d), the [Bmim][TCM] treated coal again appears largely unchanged from its raw counterpart, the particle size is unchanged and the particles do not appear to be cracked or fractured to any extent. Morphology differences between the raw and [Bmim][Cl] treated R coal can be illustrated further by referring to Fig. 8(e and f). Fragmentation and swelling of these treated coals C and R, especially those treated by [Bmim][Cl], is indicative of the ability of

these ILs to disrupt intermolecular interactions in these coals. Also, as discussed earlier, water insoluble tars deposited on the treated coal surfaces is clearly visible in Fig. 7(b) and (c). This ability of IL is likely to be beneficial for liquefaction and many other coal processing applications. SEM photos of coal A and its IL washed counterparts were taken. However, in contrast to both coals C and R, no morphological changes were observed with IL treatment. This can be seen when comparing Fig. 9a to Fig. 9b and c. The particle size observed for the raw coal remains the same in the coals treated by [Emim][DCA], [Bmim][Cl] and [Bmim][TCM]. Along with particle size, surface morphology also remained unchanged after IL treatment. This can be observed in Fig. 9e and f comparing the [Bmim][Cl] washed coal to the raw coal at 1000 magnification, there is minimal cracking present in both the raw and treated coals. Overall, the results indicate that [Bmim][Cl] showed the most consistent performance among the studied ILs. It was also found that among all studied coals, coal R showed the largest change in its thermal behaviour and coal C showed the largest change in morphology.

J. Cummings et al. / Fuel 143 (2015) 244–252

4. Discussion The importance of non-covalent interactions in the cross-linked macro structure of coal, especially in that of lower-medium coals, has been established previously [26,27], these interactions include aromatic–aromatic ionic bonding, and hydrogen bonding. If these interactions can be disrupted, then the cross-linked macrostructure can be broken down. It has been shown previously via solvent extraction that low molecular weight species can exist trapped within the pores of the rigid cross-linked polyaromatic network of coal [7,28,29]. Through the disruption of this network, the mobility of these lower molecular weight fractions trapped in the cross-linked structure can be increased. The properties of the ILs used in this study may allow them to participate in p–p stacking, p–cation interactions, hydrogen bonding and electrostatic interactions. It is believed that these interactions might have allowed the liberation of lower molecular weight volatiles from the confines of the cross-linked network. This liberation explains the consistently larger amounts of mass lost for coals C and R with low temperature ionic liquid treatment as was shown in Figs. 5 and 6. This was also corroborated by the morphological changes observed in coals C and R via SEM. The increased fracturing and cracking observed for coal R when treated by [Emim][DCA] and more so by [Bmim][Cl] implies that structural changes in the cross-linked network have occurred due to coal/IL interactions. This effect was even more pronounced with coal C, where a large amount of fragmentation and fracturing was observed as a result of treatment by [Emim][DCA] and [Bmim][Cl], the surface morphology had also appeared to change from rigid to extremely porous and soft. This again implies that a breakdown of the macrostructure had occurred as a result of IL treatment. The results also display that the extent of the coal/IL interactions are not only coal specific, as none of the ILs were able to change coal A, but also IL specific, as [Bmim][Cl] was the most consistent in being able to alter the properties of coals C and R. Previous work carried out by Guo [30] showed that ILs with a large local non-neutrality in charge density between the cation and anion were more effectively able to inhibit the precipitation of asphaltenes, it was proposed that this allowed the ability of anions to undergo acid-base interactions to increase. The results presented in this paper support this trend. The larger charge density of the Cl anion in comparison to the TCM and DCA anions might mean that they are more able to engage in interactions with functional groups that are electron deficient, steric effects may also play a part in this due to the differing anion sizes. It is posited that this coupled with p–p stacking, p-cation interactions, and electrostatic interactions allows [Bmim][Cl] to significantly alter the crosslinked network of coal C and R. Also, when these ionic liquids interact with coal, breaking down the macrostructure, it is possible that they are also causing selfdonation of hydrogen to occur. As when the structure swells and fragments, disrupting the cross linked coal network, it causes hydrogen to be transferred in order to stabilize the partially broken up polyaromatic network. This causes the coal to be partially hydrogenated, making it far easier for the Hydrogenation step of DLC to occur. This liberation of lower molecular weight volatiles, and the breaking down of the large, cross-linked network is advantageous for coals undergoing DCL. The large increase in porosity is advantageous for catalyst impregnation and hydrogen dispersion during hydrogenation and the presence of smaller particles and fragments means that less time is required in order to breakdown the large polymerized aromatic macrostructure. These factors could lead to a significant reduction in the amount of time and thus energy required for hydrogenation to occur. This is important as it reduces

251

the energy and hydrogen consumption of the DCL process. This will help to lower the overall capital cost, which is one of the biggest hurdles impeding the commercialisation of this process. 5. Conclusions The viability of ionic liquids as a solvent pre-treatment for direct coal liquefaction was investigated. Three different ionic liquids were used to treat three sub bituminous Australian coals at temperatures as low as 100 °C. After the treatment, coals and ionic liquids were separated by number of water wash followed by conductivity measurements to ensure the complete separation. Thermal properties of the treated coals were compared against the raw coal using pyrolysis experiments in a Thermogravimetric Analyser (TGA). Scanning Electron Microscopy (SEM) was also used to investigate the morphological changes in the treated coals. Results indicate that the some of the ionic liquids used were able to alter the thermal and physical properties of some of the sub-bituminous coals used in this study. Coals R and C showed considerable mass loss during the pyrolysis experiments compared to the raw ones, whereas coal A was unchanged. Also, extensive fragmentation and swelling were observed for the coals R and C when treated by [Emim][DCA] and especially [Bmim][Cl]. Among the studied ionic liquids, [Bmim][Cl] was found to perform the most consistently in being able to alter the sub-bituminous coals structure. This may be due to the large disparity in charge density between cation and anion along with p–p stacking, p-cation interactions, hydrogen bonding and electrostatic interactions which allows this ionic liquid to disrupt the cross linked network of coal. Moreover, the coal and ionic liquid interactions were found to be more coal specific as indicated in the previous literature as none of the ionic liquid was able to change the thermal or physical properties of coal A. Acknowledgements The authors wish to acknowledge the financial support provided by the University of Newcastle Australia for the work presented in this paper. We also wish to acknowledge the contributions made by undergraduate students Andrew See and Timothy Law. References [1] Lynton Jaques M, Bradshaw LC, Budd Anthony, Michael Huleatt DH, Lambert Ian, Steve LePoidevin AM, et al. Australian energy resource assessment; 2010. p. 130–70. [2] Blackburn J. Australia’s liquid fuel security Part 2. In: NRMA motoring and services; 2014. [3] Höök M, Aleklett K. A review on coal-to-liquid fuels and its coal consumption. Int J Energy Res 2010;34:848–64. [4] Vasireddy S, Morreale B, Cugini A, Song C, Spivey JJ. Clean liquid fuels from direct coal liquefaction: chemistry, catalysis, technological status and challenges. Energy Environ Sci 2011;4:311. [5] Eric RT, Larson D. Synthetic fuel production by indirect coal liquefaction. Energy Sustain Dev 2003;7:79–102. [6] Mochida I, Okuma O, Yoon SH. Chemicals from direct coal liquefaction. Chem Rev 2014;114:1637–72. [7] Speight JG. The chemistry and technology of coal. third ed. 6000 Broken Sound Parkway NW, Suite 300: CRC Press; 2013. [8] Atsushi Ishihara EWQ, Putu Sutrisna I, Kabe Yaeko. Liquefaction of coal. In: Coal and coal-related compounds structures, reactivity and catalytic reactions. Elsevier Science and Technology Books; 2004. [9] Shui H, Cai Z, Xu C. Recent advances in direct coal liquefaction. Energies 2010;3:155–70. [10] John Stipanovich BGO, Strege Joshua R, Kurz Marc D, Snyder Anthony C, Jensen Melanie D. Feasibility of direct coal liquefaction in the modern economic climate. In: Energy and environmental research center university of North Dakota, U.S. Department of Energy; 2009. [11] Belgin Gözmen LA, Erbatur Gaye, Erbatur Oktay. Direct liquefaction of highsulfur coals: effects of the catalyst, the solvent, and the mineral matter. Energy Fuels 2002;16:1040–7.

252

J. Cummings et al. / Fuel 143 (2015) 244–252

[12] Jose M Rincon SC. Influence of preswelling on liquefaction of coal. Fuel 1988;67:1162–3. [13] Baldwin DRK RM, Nguanprasert O, Miller RL. Liquefaction reactivity enhancement of coal by mild alkylation and solvent swelling techniques. Fuel 1991;70:429–33. [14] Pintoa IGF, Lobob LS, Cabritaa I. Effect of coal pre-treatment with swelling solvents on coal liquefaction. Fuel 1999;78:629–34. [15] Shah K, Atkin R, Stanger R, Wall T, Moghtaderi B. Interactions between vitrinite and inertinite-rich coals and the ionic liquid – [Bmim][Cl]. Fuel 2014;119:214–8. [16] Rika Hagiwara YI. Room temperature ionic liquids of alkylimidazolium cations and fluoroanions. J Fluorine Chem 2000;105:221–7. [17] Walden P. Molecular weights and electrical conductivity of several fused salts. Bull Acad Imper Sci 1914:405–22. [18] Endres F, Zein El Abedin S. Air and water stable ionic liquids in physical chemistry. Phys Chem Chem Phys: PCCP 2006;8:2101–16. [19] Robert Hayes GGW, Atkin Rob. At the interface: solvation and designing ionic liquids. Phys Chem Chem Phys: PCCP 2010;12:1709–23. [20] Marsh JAB KN, Lichtenthaler R. Room temperature ionic liquids and their mixtures—a review. Fluid Phase Equilibria 2004;219:93–8. [21] Visser AES, Reichert RP, Mayton WM, Sheff R, Wierzbicki S, Davis A, et al. Taskspecific ionic liquids incorporating novel cations for the coordination and

[22] [23] [24] [25] [26] [27] [28] [29] [30]

extraction of Hg2+ and Cd2+: synthesis, characterization, and extraction studies. Environ Sci Technol 2002;36:2523–9. Painter P, Pulati N, Cetiner R, Sobkowiak M, Mitchell G, Mathews J. Dissolution and dispersion of coal in ionic liquids. Energy Fuels 2010;24:1848–53. Pulati N, Sobkowiak M, Mathews JP, Painter P. Low-temperature treatment of illinois No. 6 coal in ionic liquids. Energy Fuels 2012;26:3548–52. Hansmeier AR. Ionic liquids as alternative solvents for aromatics extraction. Eindhoven University of Technology; 2010. Niklas Meine FBARR. Thermal stability of ionic liquids assessed by potentiometric titration. Green Chem 2010. Haenel MW. Recent progress in coal structure research. Fuel 1992;71:1211–23. Painter POPC, Scaroni A. Ionomers and the structure of coal. Energy Fuels 2000;14:1115–8. Vahrman M. The smaller molecules derived from coal and their significance. Fuel 1970;49:5–16. Gotz WPAGKE. Investigations on the petroleum generation potential of bituminous coals from the Saar region. Org Geochem 1991;17:695–704. Guo Y-FHAT-M. Effect of the structures of ionic liquids and alkylbenzenederived amphiphiles on the inhibition of asphaltene precipitation from CO2injected reservoir oils. Langmuir 2005;21:8168–74.