The effect of densification on brown coal physical properties and its spontaneous combustion propensity

Fuel 193 (2017) 54–64

Contents lists available at ScienceDirect

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

Full Length Article

The effect of densification on brown coal physical properties and its spontaneous combustion propensity Mohammad Reza Parsa a, Yoshimitsu Tsukasaki a,b, Emily L. Perkins a,c, Alan L. Chaffee a,⇑ a

School of Chemistry, Monash University, Clayton, Victoria, Australia Nippon Steel and Sumitomo Metal Corporation, 6-1 Marunouchi 2-Chome, Chiyoda-ku, Tokyo 100-8071, Japan c Department of Chemical Engineering, Monash University, Clayton, Victoria, Australia b

h i g h l i g h t s
Densification of brown coal with NaOH reduces its tendency to spontaneously combust.
Surface area and pore volume of densified products directly correlate with Tcr.
Vickers hardness of densified products increases with NaOH concentration.
Surface morphology of products densified with NaOH becomes smoother.

a r t i c l e

i n f o

Article history: Received 24 May 2016 Received in revised form 1 December 2016 Accepted 7 December 2016

Keywords: Brown coal Lignite Densified coal Self-heating Surface area Pore volume

a b s t r a c t The process, where brown coal is extruded after mechanical kneading and then allowed to air dry slowly to form a product known as ‘densified coal’, was applied to reduce the moisture content of two Victorian brown coals. NaOH at different concentrations (0–1.5 M) was used as an additive in the kneading step. The spontaneous combustion propensity of the densified products was evaluated and compared against multiple physical properties and morphological features of materials. The densification process reduced the moisture content of the sample from around 60% to around 12%. NaOH addition led to a progressive reduction in the CO2 surface area, as well as the porosity determined by mercury intrusion, due to the development of a stronger electrostatic network within the coal structure. The reduced micropore volume limits the accessibility of O2 to internal surfaces of the coal leading to a significant increase in the critical ignition temperature (Tcr) measured by the wire basket test method. SEM imaging indicated that the coal particle surface changed from spongy and porous for nitrogen dried raw coal to very smooth and contiguous for densified coal. These trends also correlated with progressive reduction in the CO2 surface area, as well as the porosity determined by mercury intrusion porosimetry. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction The state of Victoria in Australia is said to possess 25% of the known world reserves of low rank coal. Victorian brown coal is a very significant source of energy because of its potential for open-cut mining (lower mining cost), its high reactivity and low ash content. Although it has the potential to remain a major energy source for the local economy into the future [1], a significant obstacle to further development of Victorian brown coal usage is its high moisture content, typically around 60%. Thus, it would be benefi-

⇑ Corresponding author. E-mail address: [email protected] (A.L. Chaffee). http://dx.doi.org/10.1016/j.fuel.2016.12.016 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.

cial to remove the water in order to decrease transportation costs, increase power plant efficiency and reduce net CO2 emissions. A further issue is that dried brown coal easily disintegrates into fine dust, which results in an increase of spontaneous combustion propensity [2,3]. Spontaneous combustion takes place when the accumulation of heat within carboniferous materials, resulting from some low temperature chemical and/or physical processes, is faster than the release of heat into the environment [4,5]. Spontaneous combustion of coal is a serious issue in the world’s coal industry causing many problems such as the difficulty of transportation of coal over large distances, fire safety concerns, storage issues and long-term environmental problems [6]. Self-heating needs to be prevented and/or controlled, but the processes responsible for self-heating are complex. The tendency

M.R. Parsa et al. / Fuel 193 (2017) 54–64

for spontaneous combustion in the field is affected not only by inherent factors, which were the subject of our study, but also by external factors such as ambient temperature and humidity, stockpile size and management protocols, moisture content and precipitation patterns, prevailing winds and, where relevant, transport conditions [7,8]. The complexity of chemical structure, the inhomogeneity of coal seams (and samples) and the variability of experimental methods has led to inconsistent, confusing and sometimes conflicting results in the literature [5]. Although much valuable research has been carried out to describe the spontaneous combustion of coal, studies focused on developing a fundamental scientific understanding of the impact of physical and chemical properties on low temperature reactions are rare [5,9]. Many techniques have been proposed for removing the water from brown coal by evaporative and non-evaporative techniques. Some of these techniques such as hydrothermal dewatering [10], mechanical thermal dewatering [11] and the densification process [12] have advantages that include a reduced energy requirement and less severe conditions to remove the water from the coal [10,13]. The optimum drying process would require minimal energy; and produce a product of sufficient strength such that it can be handled without attrition together with physical properties that provide it with a reduced propensity to spontaneously combustion. The densification process developed in the late 1980s [12] transforms run-of-mine brown coal into a dense, dry, hard product. The process begins by shear attrition using a batch or continuous kneading process which leads to the formation of a dough of plasticine consistency. Applied shearing stress releases water from the cellular structure of the coal, forming a relatively homogenous smooth plastic dough during the kneading stage. The dough is extruded in order to produce pellets of the desired shape and size. Finally, the pellets are air-dried at room temperature forming hard and dense coal pellets with high compressive strength [12,14]. However, the impact of densification on the spontaneous combustion propensity of the coal has not previously been investigated in any systematic way. Sujanti and Zhang studied the effect of different inorganic additives on brown coal spontaneous combustion after impregnation from aqueous solution [9,15,16]. Potassium chloride, Montan powder, and sodium chloride were shown to have the highest inhibitory effect, followed by calcium chloride and magnesium acetate, whereas the spontaneous combustion was promoted when calcium carbonate, sodium acetate, potassium acetate, and pyrite were added to the coal. The spontaneous combustion propensity of coal samples was not significantly affected by the addition of sodium nitrate and ammonium chloride [9]. However, it remains unknown just how these additives affect the coal properties so as to modify the spontaneous combustion propensity. The objective of this study was to develop an understanding of the relationship between densified brown coal’s physical properties and the spontaneous combustion behaviour of brown coal. We have systematically modified some individual brown coal properties (surface area, pore size distribution, hardness) in a controlled way which also overcomes the complexity inherent in some prior studies that have focussed on comparisons between different coals. The spontaneous combustion tendency of the products was evaluated using the wire basket method.

2. Experimental 2.1. Coal samples Two Victorian brown coals, Morwell (MW) and Loy Yang (LY), have been used in this study originating from the Latrobe Valley,

55

Victoria, Australia. Table 1 shows the proximate and ultimate analyses of both coals. The Campbell Microanalytical laboratory at the University of Otago carried out the ultimate analysis. The inorganic compositions of the coal samples, determined as described in Section 2.4, are given in Table 2. Received samples were milled to <3 mm and then homogenised prior to the densification process. All samples were ground and sieved to <0.018 mm before all other analytical tests. In order to remove the inorganic matter, an acid washing procedure was carried out by placing 175 g wet coal in 2 l side-arm flasks with 1.5 l sulfuric acid (0.1 M concentration) and stirred for 24 h. Acid washed (AW) samples were vacuum filtered and further washed with deionized water, and vacuum filtered to remove acid. This rinsing procedure was repeated until the pH of sample became constant. Water washed (WW) samples were also rinsed in the same way, so as to mimic any effects of further washing on the coal as experienced by acid washing. 2.2. Densification process The process begins by applying shear attrition to 100 g of wet coal and desired amount of water or NaOH solution, using an IKA HKD-T 0.6 laboratory kneader. The ratio of total water (added, plus the coal moisture) to solid coal (dry basis) employed was 95 ml/40 g, and the amount of NaOH to db coal was 0.7 g (0.5 M), 1.4 g (1 M) and 2.1 g (1.5 M) NaOH to 40 g db coal. These ratios were kept constant for washed samples (WW and AW) with higher initial moisture content. During the kneading stage, any residual component and cellular structures of the coal collapses under the applied shearing stress, the sample becomes thoroughly homogenised and water within the coal structure is released. After 1 h of kneading with a torque of 25 N cm and a speed of 120 rpm a smooth coal dough which is plastic-like in character, is produced. Next, the dough is extruded through an 8 mm diameter steel nozzle using a compressed air driven rod (620 kPa). The extruded material was collected in a tray and cut into 5–10 mm length pellets with a scalpel. Finally, the pellets were air-dried at ambient conditions for at least 72 h to form hard and dense coal pellets [12,14]. 2.3. The wire basket method The wire basket method was used to determine the Tcr of samples [17]. The wire basket test was established as a standard test method for characterising the inherent tendency of the coal to react in circumstances where heat insulation is minimal (sample size of 1 cubic in.) and air flow is not impeded (fan forced oven). For a given coal sample of known particle size, Tcr values are reproducible to within ±2 °C or less [5]. Briefly, 10–13 g of sample, which can vary according to the packing density of the sample, was loaded to fill a 1 in. stainlesssteel mesh cube basket. The loaded basket was fixed to a sample holder. Three K-type thermocouples were inserted vertically in the middle of the sample, one at the centre of the basket and the other two half-way between the centre and two opposite outside edges of the cube. A fourth thermocouple was positioned horizontally to measure the actual oven temperature. The wire basket was suspended in a fan forced oven, which could be heated to a pre-set temperature to observe the combustion behaviour of the sample. At the start of the experiment, the oven was set to the desired temperature. Fan forcing maintained the airflow during the experiment. The temperatures measured by the thermocouples were continuously logged. To obtain Tcr for a sample, the wire basket experiments were repeated at different oven temperatures, at 2 °C intervals. The Tcr of each sample was determined by averaging

56

M.R. Parsa et al. / Fuel 193 (2017) 54–64

Table 1 Ultimate analysis and proximate analysis of LY and MW run-of-mine coal. Sample

Moisture (% as received)

MW LY

60 59

Proximate analysis (% dry basis) Ash

Volatile

Fixed carbon

C

H

O (diff.)

N

S

Cl

1.9 3.5

48.4 49.4

49.7 47.2

67.4 65.7

4.5 4.7

27.3 28.2

0.5 0.6

0.24 0.66

0.06 0.11

Table 2 Inorganic matter analysis of LY and MW run-of-mine coal. Inorganic

SiO2 Al2O3 Fe2O3 TiO2 K2O MgO Na2O CaO SO3 P2O5 a

MW

LY

% of ash

g/100 g dba

% of ash

g/100 g db

2.9 1.0 14.8 0.1 0.2 20.1 2.3 36.0 22.2 0.0

0.10 0.02 0.28 0.002 0.004 0.38 0.044 0.68 0.42 0

56.6 19.2 2.3 8.0 0.2 2.4 4.1 1.0 5.4 0.2

1.98 0.67 0.08 0.28 0.007 0.084 0.14 0.035 0.19 0.007

Dry basis.

the lowest temperature at which the thermal runaway occurred and the highest temperature at which the thermal runaway did not occur. Thus, higher values of Tcr indicate lower spontaneous combustion propensity [9].

2.4. Physical analysis methods To measure the moisture content of the raw (as received) and washed coals (after filtration) and of the densified products (after standing at ambient conditions for at least 72 h, crushing and sieving to <0.018 mm) a 5 g sample was dried in an oven at 105 °C under nitrogen flow for 3 h [1] and re-weighed. Tests were carried out in duplicate and moisture content was calculated according to Eq. (1).

moisture contentð% wet basisÞ ¼ 100 

Ultimate analysis (% dry basis)

mass of coal before drying  mass of coal after drying mass of coal before drying ð1Þ

Equilibrium moisture contents at relative vapour pressures (RVP) of 11.3%, 51.4% and 92.3% were determined by putting 4.5– 5.5 g representative samples into glass dishes. The open dishes were placed in desiccators, each containing a different saturated salt–water mixture chosen to give these specific RVP values at 30 °C [18,19]. A minimum equilibration time of two weeks was allowed for all samples. The total ash content of the samples was measured following the standard ashing procedure [20]. Coal samples were analysed for inorganic content by inductively coupled plasma optical emission spectrometry (ICP-OES) using a Perkin-Elmer Optima 7000 DV ICP-OES spectrometer. Samples of coal or pellets were dissolved by microwave digestion at 1200 W using a Multiwave 3000, Anton Paar, according to the method developed by Low and Zhang [21]. The Vickers hardness tests method was carried out using a Wolpert Wilson Vickers hardness apparatus (model: 432SVA) under ambient laboratory conditions. A pyramid shape diamond with a square base and an angle of 136 degrees between opposite faces applied a load of 9.8–98.0 N (1–10 kgf) on densified pellets of brown coals. The full load was maintained for 15 s [22].

The surface area was calculated from CO2 adsorption measurements at 0 °C analysed using the Dubinin–Radushkevich equation [23]. The molecular area of CO2 was taken to be 0.254 nm2 [23]. 100 mg of powder sample <0.018 mm was placed into the analysis tube and moisture removed from the pre-dried sample by degassing at 105 °C overnight before the analysis. The analysis was carried out on a Micromeritics Tristar II instrument. Mercury intrusion and extrusion curves were measured on a Micromeritics AUTOPORE IV, model 1327 instrument. Pellet samples were crushed and sieved to have a particle size <0.018 mm and, after drying at 105 °C, approximately 0.25 g of individual samples was added to a calibrated 3 ml powder penetrometer. Measurement was performed by applying pressures between 3.6 kPa and 413 MPa which, according to the Washburn equation for cylindrical pores [7], equates to pore diameters, ranging from 0.0030 to 341 lm at a surface tension of 0.485 N m1 and a contact angle of 130°. Pore size distributions, macro, meso and micro pore were determined from the intrusion curves, as per the method described previously by Bergins et al. [24]. In order to avoid any error caused by compressibility at high mercury intrusion pressures; all obtained data was corrected using the extrusion curved for each data set [24]. All results are expressed on a volume per gram of dry coal basis. The macropore volume and mesopore volume were obtained by summing volumes with diameters ranging between 50 and 0.05 lm and 0.05 and 0.0036 lm respectively [24]. The micropore volume (Vmicro) was calculated from the skeletal density (calculated from maximum mercury intrusion after compressibility correction) and the He density [24].

Vmicro ¼

1

qskeletal



1

qHe

ð2Þ

The morphology of the samples was examined by scanning electron microscopy (FEG/SEM), Nova NanoSEM 450 operated with a field emission gun at 5 kV. Samples were dispersed in ethanol and mounted on conductive carbon tape coated with 1–2 nm platinum. Typically, approximately 10 particles of each sample were examined in detail. 3. Results and discussion 3.1. Wire basket results The wire basket test results for the densified products from raw and AWLY coal, raw and AWMW coals and the densified products prepared them with NaOH are given in Fig. 1. It can be seen that acid washing of raw brown coal generally resulted in a higher critical ignition temperatures (Tcr) presumably due to the removal of acid-extractable metals (see Section 3.3). Nitrogen drying leads to the reduction of Tcr for all samples. In addition, the densification of all coals without NaOH resulted in slightly higher propensity for spontaneous combustion compared to their non-densified counterparts by showing lower Tcr values. In both series of LY coals (raw and AW), spontaneous combustion propensity first increased (lower Tcr) with the addition of 0.5 M NaOH and thereafter decreased with the addition of 1.0

M.R. Parsa et al. / Fuel 193 (2017) 54–64

and 1.5 M NaOH. The wire basket results for raw LY and AWLY showed that all AW samples gave higher Tcr values than the analogous samples derived from the raw coal. The trend of Tcr for raw and AWLY densified products first decreased with the addition of 0.5 M NaOH and then increased with the addition of 1.0 and 1.5 M NaOH, similar to AWMW coal series. On the other hand, the Tcr value for the raw MW coal products increased incrementally from 147 °C for densified product with no NaOH addition to 167 °C for densified product with 1.5 M NaOH addition. 3.2. Moisture content The moisture contents for raw, WW and AW coals before and after densification are given in Fig. 2. The densification of brown coals followed by air drying at ambient conditions significantly reduces the moisture content of all products, relative to the parent coals. The original moisture contents for raw, WW and AWLY were 57%, 71% and 69% respectively, after densification the moisture content dropped to approximately 15%, 20% and 15%. It should be noted that the densified products’ moisture contents depend on ambient temperature and relative humidity at ambient conditions. Similar decreases in moisture content were observed for raw and AWMW coal, decreasing from 60% and 66% to 15% and 14% respectively. These values are comparable to those previously measured for densified coal prepared from brown coal from the MW seam, 15.9% [12] and those from the LY and MW seams of 16% [25]. AW and unwashed coal densified products appear to have slightly lower moisture content than that of WW coal. It seems that the densification of brown coal with different concentrations of NaOH did not have a significant effect on the moisture content of the densified products. The results of studies reporting the effect of moisture content on the coal spontaneous combustion behaviour are complicated and sometimes contradictory [26–35]. It is known that partial drying of coal decreases the Tcr until a minimum is reached, as a result of higher exposure of the coal to oxygen and the formation of more active sites for oxidation. Tcr then increases with the further reduction of the moisture content, due to the ‘‘ageing effect” on the coal structure and the low amount of water available for rapid formation of peroxide complexes during oxidation [2,27,36]. The presence of a large quantity of water may also reduce the tendency of self-heating of coal by reducing the number of radical sites where oxidation can take place and obstructing the oxygen

Fig. 1. Wire basket test results for densified products of LY and MW raw and AW coals.

57

molecules access to the coal structure. In addition, the high moisture content postpones the self-heating by adsorbing the heat released by oxidation [31]. In this study, reductions in the moisture content by drying the raw coal at 105 °C under nitrogen and by densification without NaOH, accelerated low temperature oxidation reactions, resulting in lower Tcr values. For coals densified with NaOH addition, any effect of moisture on Tcr is masked by the effect of NaOH addition. Although the moisture contents of densified coals with NaOH were similar to densified coals without NaOH, their Tcr values were significantly different (higher), inferring that NaOH addition is a more dominant factor on influencing Tcr. 3.3. Inorganic matter and ash content One of the important factors determining the spontaneous combustion behaviour of brown coal is the inorganic content. Fig. 3a and b provides details of the major inorganic components of raw, AW and WWLY and raw and AWMW. It can be seen from Fig. 3 and Table 2, that the amount of Fe, Mg, Ca and S in MW coal is more than in LY, whereas Na, Al and Si exist in higher amounts in LY coal. It has been reported [9] that Fe and S increased the propensity of spontaneous combustion of brown coal thus the generally the lower Tcr value for the MW coal samples relative to the LY samples. In addition, aluminium and silica are known to suppress spontaneous combustion, helping to postpone the ignition of LY, leading to higher Tcr values [37]. The acid washing of both coals led to the substantive removal of the elements examined (Fig. 3). The higher Tcr of the AW samples compared to raw coals can be attributed to removal of those inorganic components from the coals. Both water washing and acid washing reduce the inorganic contents and therefore the ash contents of the non-densified washed coals are lower than raw coals (see Fig. 4). The ash content of the densified products increased with increasing addition of NaOH, as expected, since the NaOH addition provides additional Na ions. This observation is consistent for samples made from both coals and all wash types. 3.4. Vickers hardness determination The Vickers hardness index (Fig. 5) indicates an increasing trend with an increased addition of alkali during the kneading process. This is in agreement with previous work that measured the crush strength of densified brown coal [25]. Pandolfo and Johns [25] reported that the addition of an alkali during the kneading process affects the coal structure in two major ways. Firstly it solubilizes part of the coal and redistributes these soluble humic components within the internal pore structure of the coal. Secondly the surface functionalities of the coal are changed to have more phenolic (AO) and carboxylate (ACOO) groups. In other words, by increasing the concentration of NaOH, the pH increases and this consequently changes the form of acidic functional groups from AOH to AO and ACOOH to ACOO. This causes the weak hydrogen-bonding interactions within the coal structure that exist in an acidic natural environment to be replaced with stronger electrostatic bonds in high pH environments leading to the generation of a firmer network structure. Whilst the densified products of non-washed and WW coals have similar values of Vicker’s hardness, the densified products from the AW coal have lower hardness index values than the non-washed and WW analogous. Thus, it appears that removing the acid extractable inorganic matter has also had an impact on mechanical properties of the pellets. This is consistent with prior the results found for crush strength [25], where acid washing created a less dense pellet that had reduced crush strength. It is inter-

58

M.R. Parsa et al. / Fuel 193 (2017) 54–64

Fig. 2. Moisture content of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series.

Fig. 3. Inorganic content of raw, WW and AW coals; (a) LY series and (b) MW series.

Fig. 4. Ash contents of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series.

esting that, the removal of inherent cations by acid washing and water washing processes cannot be compensated for by addition of NaOH. 3.5. CO2 surface area results The CO2 surface areas of densified products are given in Fig. 6. Water washing and acid washing the parent coals reduced the surface areas slightly, in agreement with earlier work [38]. The results

also revealed that densification of raw and AW coal with NaOH decreased the CO2 surface area significantly. Indeed, reduction of CO2 surface area was observed to directly correlate with NaOH concentration. Previous work has shown that the addition of sodium ions can have an effect in reducing the surface area measured by CO2 adsorption [38], but a significantly greater reduction was observed in the present work. A reason for the decreasing surface area is likely to be that the progressive addition of NaOH progressively generates stronger

M.R. Parsa et al. / Fuel 193 (2017) 54–64

59

Fig. 5. Vickers hardness index of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series.

electrostatic bonds, as mentioned in the previous section, so as to ‘tighten’ the coal structure together. The CO2 surface area value is actually calculated from the adsorption of CO2 molecules in the micropores of the coal structure, therefore a decreased surface area value is indicative of a decreased micropore volume [23]. The reduction in micropore volume will reduce the opportunity for O2 to access internal surfaces of the coal [5,39], and this will lead to a significant effect on the self-heating phenomena of coal by controlling the rate of oxidation [40–42]. The reduction of surface area and micropore volumes also reduces the concentration of active sites, such as hydroperoxides, available to interact with atmospheric oxygen [41]. Both factors probably contribute to the reduced spontaneous combustion tendency of the brown coal. 3.6. Mercury intrusion porosimetry results The pore size distributions determined by mercury intrusion porosimetry for LY samples are shown in Fig. 7. The inter/intra particle boundary (at 50 lm) and the macro- and mesoporosity ranges are marked on the graphs to aid interpretation. Fig. 7 shows that the major portion of the pore volume of all samples is in the macropore region and that with increasing the concentration of NaOH addition the pore size distribution curves flatten significantly, such that the volume due to mesopores and small macropores is substantially reduced. The same trend was observed for MW coal products (Fig. 8).

A summary of the mercury surface areas and mean pore diameters determined from mercury intrusion porosimetry for all samples tested is given in Table 3. It can be seen that by increasing the concentration of NaOH, a shift in the corresponding mean pore diameter to larger diameters occurs. For LY coal, the AW sample densified with NaOH showed higher mean pore diameters than the corresponding densified WW and raw coal samples series. In contrast, For MW coal, the mean pore diameters of raw MW densified products series were higher than those of the AW series. By increasing the addition of NaOH, the mercury surface area dropped in accordance with the corresponding decrease in macro, meso, micro and total pore volumes (Fig. 9a) from around 22 m2/g for non-densified LY to around 12 m2/g for densified products with 1.5 M NaOH. No significant difference among mercury surface area of densified products made from raw, AW and WW coals was observed (Table 4). The water washing process produces an increase the microporosity of LY sample relative to raw LY and in contrast acid washing of LY coal caused a decrease in micropore volume. However, the WW coal densified with NaOH showed a more dramatic decrease of micropore volume compared to the densified raw and AW products with NaOH. In general, as a function of increasing the concentration of NaOH, the macro, meso, micro and total pore volumes show a decreasing trend for all three LY coal samples (Fig. 9a). This decrease in total pore volume was associated with a decrease in mercury surface area which was consistent with the decrease in CO2 surface area.

Fig. 6. The CO2 surface area of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series.

60

M.R. Parsa et al. / Fuel 193 (2017) 54–64

Fig. 7. Logarithmic differential mercury intrusion of raw, WW and AW LY coals and their densified products prepared with varying additions of NaOH.

Fig. 8. Logarithmic differential intrusion of raw and AW MW coals and their densified products prepared with varying additions of NaOH.

Table 3 Mercury surface area and mean pore diameter of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series. LY

Raw

WW

AW

Non-densified Densified with NaOH

Non-densified Densified with NaOH

Non-densified Densified with NaOH

MW

Mercury surface area (m2 g1)

Mean pore diameter (lm)

Mercury surface area (m2 g1)

Mean pore diameter (lm)

0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

21.7 ± 0.5 18.0 ± 0.3 17.0 ± 0.2 15.0 ± 0.0 12.2 ± 0.2

10.4 ± 0.2 8.8 ± 0.3 13.7 ± 0.5 11.5 ± 0.2 12.9 ± 0.3

19.3 ± 0.7 18.7 ± 0.4 17.1 ± 0.4 11.4 ± 0.3 12.5 ± 0.5

24.3 ± 0.3 9.2 ± 0.2 17.2 ± 0.5 24.2 ± 0.3 28.1 ± 0.5

0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

21.2 ± 0.5 18.6 ± 0.1 15.8 ± 0.3 14.9 ± 0.3 10.8 ± 0.4

5.9 ± 0.2 6.4 ± 0.4 7.6 ± 0.3 12.3 ± 0.5 13.5 ± 0.5

0.0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

23.4 ± 0.3 18.7 ± 0.5 15.6 ± 0.2 15.5 ± 0.2 13.4 ± 0.3

5.3 ± 0.4 15.2 ± 0.5 25.2 ± 0.7 27.3 ± 0.5 29.0 ± 0.3

21.8 ± 0.6 20.2 ± 0.2 13.9 ± 0.0 13.8 ± 0.2 10.5 ± 0.1

6.2 ± 0.1 8.9 ± 0.4 11.1 ± 0.3 12.7 ± 0.2 19.3 ± 0.4

No significant change in micropore volume occurred for AWMW series compared to raw MW series. Even though the micro- and meso pore volumes of the AWMW samples series of densified coals decreased more dramatically with increasing NaOH concentration than for the densified coals in the raw MW series, the decrease of macropore volumes in the raw MW series with increasing NaOH addition resulted in lower total pore volumes in this (raw) series. These trends are in good agreement with the CO2 surface area results for these samples.

Thus, it is apparent that the reductions of pore volume and mercury surface area with increasing the NaOH concentration correlate inversely with Tcr for all series of samples. This is in agreement with previous results from the group [2]. However, in this case some of the pore volumes are significantly lower. 3.7. Equilibrium moisture content The equilibrium moisture contents of samples, measured in desiccators at 30 °C, are presented in Table 4. It can be observed

61

M.R. Parsa et al. / Fuel 193 (2017) 54–64

Fig. 9. Macro, meso and micropore volume of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series.

Table 4 Equilibrium moisture content at relative vapour pressure 11.3%, 51.4% and 92.3% of raw, WW and AW coals and their densified products prepared with varying additions of NaOH; (a) LY series and (b) MW series. LY EMC (mmol/g db) RVP

Raw

WW

AW

Non-densified Densified with NaOH

Non-densified Densified with NaOH

Non-densified Densified with NaOH

MW EMC (mmol/g db) RVP

11.3%

11.3%

11.3%

11.3%

51.4%

92.3%

0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

2.9 ± 0.1 1.8 ± 0.1 2.1 ± 0.7 2.0 ± 0.5 2.7 ± 0.1

2.9 ± 0.1 1.8 ± 0.1 2.1 ± 0.7 2.0 ± 0.5 2.7 ± 0.1

26.2 ± 3.1 10.3 ± 0.9 11.6 ± 0.9 11.7 ± 0.5 15.4 ± 0.4

2.4 ± 0.2 2.0 ± 0.2 1.9 ± 0.4 2.0 ± 0.5 2.2 ± 0.4

8.9 ± 0.7 5.0 ± 0.0 5.0 ± 0.3 5.1 ± 0.4 5.4 ± 0.0

26.2 ± 3.1 10.3 ± 0.9 11.6 ± 0.9 11.7 ± 0.5 15.4 ± 0.4

0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

2.5 ± 0.1 1.9 ± 0.5 1.9 ± 0.6 1.7 ± 0.6 2.6 ± 0.1

2.5 ± 0.1 1.9 ± 0.5 1.9 ± 0.6 1.7 ± 0.6 2.6 ± 0.1

33.3 ± 3.1 9.3 ± 0.9 11.6 ± 0.5 11.5 ± 1.1 16.1 ± 1.0

0.0 mol/l 0.5 mol/l 1.0 mol/l 1.5 mol/l

2.7 ± 0.1 2.0 ± 0.5 1.9 ± 0.5 2.1 ± 0.7 2.3 ± 0.1

2.7 ± 0.1 2.0 ± 0.5 1.9 ± 0.5 2.1 ± 0.7 2.3 ± 0.1

22.5 ± 2.5 9.4 ± 2.0 10.7 ± 1.0 10.3 ± 0.5 14.7 ± 1.0

2.1 ± 0.4 1.7 ± 0.1 1.6 ± 0.2 1.5 ± 0.3 2.3 ± 0.1

6.9 ± 0.5 4.3 ± 0.1 4.7 ± 0.6 4.6 ± 0.4 5.2 ± 0.2

22.5 ± 2.5 9.4 ± 2.0 10.7 ± 1.0 10.3 ± 0.5 14.7 ± 1.0

The uncertainties are 90% confidence limits based on duplicate or triplicate determinations.

62

M.R. Parsa et al. / Fuel 193 (2017) 54–64

Fig. 10. SEM images of (a) nitrogen dried raw LY, and densified LY with (b) 0.5 M NaOH, (c) 1.0 M NaOH and (d) 1.5 M of NaOH.

M.R. Parsa et al. / Fuel 193 (2017) 54–64

that the effect of water washing was not significant on LY coal’s EMC value, whereas the acid washing led to a small decline of EMC value at all RVPs tested for both coals. It seems that although the inorganic contents of the coal samples were low, the removal of inorganic cations by acid washing slightly decreased the hydrophilicity of the coal. The results are in good agreement with previous studies [38,43,44]. The EMC values decreased significantly after densification of both coals without using the additive, compared to non-densified coals. The EMC values of densified products with 0.5 and 1.0 M NaOH were close to each other and to densified coal without additive in all series of samples, whereas raising the NaOH concentration to 1.5 M resulted in the increase of the EMC values. It is interesting that previous studies showed that the addition of cations to the acid-washed coals developed the hydrophilicity of the coal surface leading to higher equilibrium moisture content over a range of RVPs (9–75%) [45,46]. This is because the strong acid groups of coal are more hydrophilic in their salt form than in their acidic form [44]. Hulston et al. [47] have also reported that the reduction of surface area and pore volume leads to lower EMC values for dewatered brown coal products prepared by mechanical thermal expression. The EMC of densified products prepared without additive is controlled by the reduction of its surface area and pore volumes, causing a significant drop in EMC values in all RVPs. In the case of densified products prepared with NaOH, the results can be understood in terms of the two opposing factors mentioned in preceding paragraph. The reduction in surface area and pore volume is clearly very significant and has the dominating influence for all samples; yet the higher hydrophilicity of the sample with 1.5 M NaOH addition is apparent. These effects are most clearly discernible at high RVP, in agreement with previous studies [47,48]. Hulston et al. [47] suggested that, at high RVP the water molecules interact with other water molecules more willingly than with the brown coal, and behave more like bulk water, requiring available volume to occupy. Hence the pore volume, especially the macropore value, plays a more important role on the water adsorption of coal at higher RPV, so that the decrease in EMC at RVP 92.3% for densified products was more significant compared to two lower RVPs. 3.8. SEM results The SEM results revealed interesting features of morphology. As seen in Fig. 10, by increasing the concentration of NaOH, the surface roughness of densified products decreased as the surface evolved from spongy and porous for the nitrogen dried raw coal to very smooth and contiguous for the coal densified with 1.5 M NaOH. The layered shape of densified coals may be the result of the applied pressure on coal particles during the extrusion process. Although it is not possible to be quantitative, it was clear that there was an increase in the number of coal particles with smooth surfaces as the NaOH concentration was increased. SEM images of increasing smoothness compliment the results of CO2 surface area and mercury porosimetry; indicated that the surface area of coals densified with NaOH additive were less accessible to oxygen. 4. Conclusion The densification process is considered to be an efficient method to dewater brown coal, providing dense and relatively hard products with heavily reduced moisture content. The use of NaOH as an additive leads to reductions in the CO2 surface areas and pore volumes concomitant with increasing the hardness of

63

the products, this occurs as a result of modifying the molecular interactions within the coal structure, such that relatively weak hydrogen bonds are replaced by stronger electrostatic bonds where NaOH is added. Also a change of surface morphology from relatively porous structure for the raw coal products to relatively smooth structure for products densified with NaOH addition is evident by SEM imaging. These physico-chemical changes result in a ‘tightening’ of the coal structure, which suppresses the oxidation rate of coal particles by reducing the number and accessibility of active sites on the coal surface at which oxygen molecules can react. As a consequence the Tcr of densified brown coal with NaOH is postponed to significantly higher temperature up to a maximum difference of 20 °C in Tcr as determined by the wire basket test method.

Acknowledgements The financial support of this project by Brown Coal Innovation Australia (BCIA) is acknowledged. The authors also thank Dr. J. Taghavi Moghaddam for SEM imaging.

References [1] Li C-Z. Chapter 1 - introduction. In: Advances in the science of Victorian brown coal. Amsterdam: Elsevier Science; 2004. p. 1–10. [2] Fei Y, Aziz AA, Nasir S, Jackson WR, Marshall M, Hulston J, et al. The spontaneous combustion behavior of some low rank coals and a range of dried products. Fuel 2009;88:1650–5. [3] Mulcahy MFR, Morley WJ, Smith IW. Chapter 8 - combustion, gasification and oxidation. In: Durie RA, editor. The science of Victorian brown coal. Butterworth-Heinemann; 1991. p. 359–463. [4] Nugroho YS, McIntosh AC, Gibbs BM. Low-temperature oxidation of single and blended coals. Fuel 2000;79:1951–61. [5] Carras JN, Young BC. Self-heating of coal and related materials: models, application and test methods. Progr Energy Combust Sci 1994;20:1–15. [6] Wang D-M, Xu-yao Q, Xiao-xing Z, Jun-jie G. Test method for the propensity of coal to spontaneous combustion. Proc Earth Planet Sci 2009;1:20–6. [7] Kaymakci VDE. Relations between coal properties and spontaneous combustion parameters. Turk J Eng Environ Sci 2002;26:59–64. [8] Baris K, Kizgut S, Didari V. Low-temperature oxidation of some Turkish coals. Fuel 2012;93:423–32. [9] Sujanti W, Zhang D-K. A laboratory study of spontaneous combustion of coal: the influence of inorganic matter and reactor size. Fuel 1999;78:549–56. [10] Favas G, Jackson WR. Hydrothermal dewatering of lower rank coals. 1. Effects of process conditions on the properties of dried product. Fuel 2003;82:53–7. [11] Bergins C. Mechanical/thermal dewatering of lignite. Part 2: a rheological model for consolidation and creep process. Fuel 2004;83:267–76. [12] Johns RB, Chaffee AL, Harvey KF, Buchanan AS, Thiele GA. The conversion of brown coal to a dense, dry, hard material. Fuel Process Technol 1989;21:209–21. [13] Bergins C. Kinetics and mechanism during mechanical/thermal dewatering of lignite. Fuel 2003;82:355–64. [14] Johns RB, Cain DA, Chaffee AL, Buchanan AS. Upgrading solid fuels. Patents US 4,627,575, AU 52590/86; 1986. [15] Zhang DK, Sujanti W. Effect of exchangeable cations on low-temperature oxidation and self-heating of a Victorian brown coal. Fuel 1999;78:1217–24. [16] Sujanti W, Zhang DK. Investigation into the role of inherent inorganic matter and additives in low-temperature oxidation of a Victorian brown coal. Combust Sci Technol 2000;152:99–114. [17] Fei Y, Aziz A Abd, Nasir S, Jackson WR, Marshall M, Hulston J, et al. The spontaneous combustion behaviour of some low rank coals and a range of dried products. Fuel 2009;88:1650–5. [18] Greenspan L. Humidity fixed points of binary saturated aqueous solutions. J Res Natl Bur Stand - A Phys Chem 1977;81A:89–96. [19] Fei Y, Chaffee AL, Marshall M, Jackson W Roy. Lignite-water interactions studied by phase transition - differential scanning calorimetry. Fuel 2005;84:1557–62. [20] S.A. International. Methods for the analysis and testing of lower rank coal and its chars - lower rank coal - determination of ash. In: AS 2434.8-2002 (R2013). Australia: Standards Australia International Ltd; 2002. [21] Low F, Zhang L. Microwave digestion for the quantification of inorganic elements in coal and coal ash using ICP-OES. Talanta 2012;101:346–52. [22] S.A. International. Metallic materials - Vickers hardness test - test methods. In: AS 1817.1-2003. Australia: Standards Australia International Ltd; 2003. [23] Woskoboenko F, Stacy WO, Raisbeck D. Chapter 4 - Physical structure and properties of brown coal. In: Durie RA, editor. The Science of Victorian Brown Coal. Butterworth-Heinemann; 1991. p. 151–246.

64

M.R. Parsa et al. / Fuel 193 (2017) 54–64

[24] Bergins C, Hulston J, Strauss K, Chaffee AL. Mechanical/thermal dewatering of lignite. Part 3: Physical properties and pore structure of MTE product coals. Fuel 2007;86:3–16. [25] Pandolfo AG, Johns RB. Physical and chemical characteristics of densified lowrank coals. Fuel 1993;72:755–61. [26] Beamish BB, Hamilton GR. Effect of moisture content on the R70 self-heating rate of Callide coal. Int J Coal Geol 2005;64:133–8. [27] Chen XD, Stott JB. The effect of moisture content on the oxidation rate of coal during near-equilibrium drying and wetting at 50 °C. Fuel 1993;72:787–92. [28] Arisoy A, Akgün F. Modelling of spontaneous combustion of coal with moisture content included. Fuel 1994;73:281–6. [29] Bhat S, Agarwal PK. The effect of moisture condensation on the spontaneous combustibility of coal. Fuel 1996;75:1523–32. [30] Bhattacharyya KK. The role of desorption of moisture from coal in its spontaneous heating. Fuel 1972;51:214–20. [31] Clemens AH, Matheson TW. The role of moisture in the self-heating of lowrank coals. Fuel 1996;75:891–5. [32] Kadiog˘lu Y, Varamaz M. The effect of moisture content and air-drying on spontaneous combustion characteristics of two Turkish lignitesa. Fuel 2003;82:1685–93. [33] Morris R, Atkinson T. Seam factor and the spontaneous heating of coal. Min Sci Technol 1988;7:149–59. [34] Sondreal EA, Ellman RC, U.S.B.o. Mines. Laboratory determination of factors affecting storage of North Dakota lignite: computer simulation of spontaneous heating. U.S. Bureau of Mines; 1974. [35] Ll YH, Skinner JL. Deactivation of dried subbituminous coal. Chem Eng Commun 1986;49:81–98. [36] Vance WE, Chen XD, Scott SC. The rate of temperature rise of a subbituminous coal during spontaneous combustion in an adiabatic device: the effect of moisture content and drying methods. Combust Flame 1996;106:261–70. [37] Speight JG. The chemistry and technology of coal. 3rd ed. Taylor & Francis; 2012.

[38] Fei Y, Zhang C, Marshall M, Jackson WR, Chaffee AL, Allardice DJ. The effect of cation content of some raw and ion-exchanged Victorian lignites on their equilibrium moisture content and surface area. Fuel 2007;86:2890–7. [39] Wang H, Dlugogorski BZ, Kennedy EM. Coal oxidation at low temperatures: oxygen consumption, oxidation products, reaction mechanism and kinetic modelling. Progr Energy Combust Sci 2003;29:487–513. [40] Beamish BB, Barakat MA, George JD St. Spontaneous-combustion propensity of New Zealand coals under adiabatic conditions. Int J Coal Geol 2001;45:217–24. [41] Clemens AH, Matheson TW, Rogers DE. Low temperature oxidation studies of dried New Zealand coals. Fuel 1991;70:215–21. [42] Bouwman R, Freriks ILC. Low-temperature oxidation of a bituminous coal. Infrared spectroscopic study of samples from a coal pile. Fuel 1980;59:315–22. [43] Allardice DJ, Clemow LM, Favas G, Jackson WR, Marshall M, Sakurovs R. The characterisation of different forms of water in low rank coals and some hydrothermally dried productsq. Fuel 2003;82:661–7. [44] Clemow LM, Jackson WR, Chaffee AL, Sakurovs R, Allardice DJ. Understanding brown coal-water interactions to reduce carbon dioxide emissions. In: Maroto-Valer MM, Song C, Soong Y, editors. Environmental challenges and greenhouse gas control for fossil fuel utilization in the 21st century. Boston (MA, US): Springer; 2002. p. 203–16. [45] Quast KB, Hall SF, Readett DJ. Characterisation of the interaction of lignite with water - Part III: effect of ion exchange. In: The 4th Australian coal science conference, Brisbane. p. 323. [46] Allardice DJ, Chaffee AL, Jackson WR, Marshall M. Chapter 3 - water in brown coal and its removal. In: Li C-Z, editor. Advances in the science of Victorian brown coal. Amsterdam: Elsevier Science; 2004. p. 85–133. [47] Hulston J, Favas G, Chaffee AL. Physico-chemical properties of Loy Yang lignite dewatered by mechanical thermal expression. Fuel 2005;84:1940–8. [48] Fei Y, Artanto Y, Giroux L, Marshall M, Jackson WR, MacPhee JA, et al. Comparison of some physico–chemical properties of Victorian lignite dewatered under non-evaporative conditions. Fuel 2006;85:1987–91.