- Email: [email protected]
Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate
Energy Conversion and Management xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate P.U. Okoye, A.Z. Abdullah, B.H. Hameed ⇑ School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, 14300 Nibong Tebal, Penang, Malaysia
a r t i c l e
i n f o
Article history: Received 4 August 2016 Received in revised form 28 September 2016 Accepted 16 October 2016 Available online xxxx Keywords: Dimethyl carbonate Glycerol Glycerol carbonate Ladle furnace steel slag Transesterification
a b s t r a c t Abundant waste from ladle furnace (LF) steel slag industry was utilized to catalyze the glycerol transesterification reaction with dimethyl carbonate (DMC) to synthesize valuable glycerol carbonate (GC). LF slag was modified using various sodium hydroxide (NaOH) loadings (1–15 wt.%), and the catalyst samples were characterized using XRD, FTIR, TPD-CO2, and EDS techniques. Modifying the LF using different NaOH loadings enhanced the stability, basicity, and basic strength of LF. Favorable reaction conditions at the best glycerol conversion and GC yield were investigated via reaction influencing parameter screening. Thus, the 10 wt.% NaOH incorporated in LF under the reaction conditions of 75 °C, DMC-to-glycerol molar ratio of 2, and 3 wt.% catalyst dosage achieved 99% glycerol conversion and 97% GC yield. The catalyst exhibited sufficient heterogeneity and was utilized in five successive cycles of the experiment without serious deactivation. Ó 2016 Published by Elsevier Ltd.
1. Introduction Glycerol, a major biodiesel production by-product, constitutes 10% w/w of the total global biodiesel [1]. The oversupply and physicochemical properties of glycerol make it a valuable material that requires defunctionalization [2]. Pathways to synthesize fine chemicals from glycerol were discussed in detail in Refs. [3–8]. The base-catalyzed transesterification reaction of glycerol using dialkyl carbonate sources is a popular pathway to upgrade abundant glycerol into fine chemicals [9–11]. Dialkyl carbonate sources, such as dimethyl carbonate (DMC), is a popular choice because these sources are environment-friendly and nontoxic [12]. Under mild reaction conditions, basic catalysts were employed to modify the glycerol transesterification reaction with DMC to synthesize valuable glycerol carbonate (GC). The base catalytic materials are synthesized mainly from alkaline earth metals and alkali metal oxides. Their outstanding activity in transesterification reaction is attributed to their strong basic sites and high basic density [13]. Too high a basic strength above H_ > 18.4 can propagate GC decomposition or dehydration to glycidol [14]. Methanol is coproduced in the reversible equilibrium reaction, and it can be positively shifted to produce appreciably high GC, utilizing DMC in excess of glycerol. GC has various applications, such as cement composite aggregate, solvent, surfactant, plant bolster, carrier in ⇑ Corresponding author.
lithium ion battery, methanol replacement to avoid glycerol coproduction in biodiesel synthesis, and fuel additive [15]. Reported studies based on CaO, MgO, double-layered hydrotalcite of Ca or Mg, Na-zeolite, K-zeolite, and sodium hydroxide (NaOH)/cAl2O3 alkaline materials describe the considerable high activity performance toward a high GC yield [16–20]. Research in the recent decade focused on the beneficial utilization of abundant wastes generated from many processing industries. Waste utilization decreases handling problems and averts environmental hazards caused by improper disposal. Millions of tons of waste are generated in many steel industries around the world. Available information on steelmaking processes indicates that approximately 135 million tons of steel were produced in April 2016 from 66 countries [21]. The volume of generated slag increases with the consumption of steel products, which affects utilization methods. Slag waste is categorized as electric arc furnace (EAF) steel slag, ladle furnace (LF) basic slag, blast furnace (BF) iron slag, and basic oxygen furnace (BOF) slag based on the source furnace [22]. Iron and steel slag are mostly utilized in structural and civil construction. Slag is applied in various areas, such as soil stabilization in erosion-prone areas, soil acidity correction, fertilizers, and adsorbents in CO2 sequestration. The varying mineralogy in different proportions provides steel slag wastes with properties that could be further explored from the economic and technical standpoints. The slag mineral compositions for each slag category provide specific characteristics for specific applications [22]. For instance, BOF and EAF slag are often
E-mail address: [email protected] (B.H. Hameed). http://dx.doi.org/10.1016/j.enconman.2016.10.067 0196-8904/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
2
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
used as low-cost materials in road construction and asphalt aggregate mixes. BF slag from iron industry can be used in Portland cement manufacturing. LF slag application differs because of its specific mineral constituents and characteristics. Notably, LF slag is largely composed of lime (Ca(OH)2) with primary applications in soil acidity correction and is a recycle material to formulate Portland clinker [22]. The slag is generated during the desulfurization of steel at the final stage of the steelmaking process, known as secondary metallurgy process. LF slag is also known as basic slag because of its high calcium content. Thus, LF slag is a suitable material for base-catalyzed reactions. Calcium in LF can be extracted or used directly to synthesize hydroxyapatite, which is highly beneficial in bone treatment. Calcium-based materials have exhibited potential in modifying the glycerol transesterification reaction [17]. However, these materials are highly unstable because of their solubility in glycerol at glycerol/DMC equimolar ratio and high reactivity in the presence of moisture and CO2. Calcium leaches into the reaction mixture and forms an undesired homogeneous catalysis, which results in an evident decline in catalytic activity at less than three cycles of reuse. CaO formed active homogeneous species Ca(C3H7O3)(OCO2CH3) at a glycerol/DMC equimolar ratio but exhibited heterogeneity when DMC is in excess of glycerol [18]. However, leaching cannot be avoided because a sharp decline in GC yield was recorded after the third cycle of reuse. Magnesium and potassium sources were employed to improve the stability of calcium [17,23]. The incorporation of magnesium provided a strong bond interaction, resulting in the minimal leaching (7.4%) of metallic species after four cycles of reuse [24]. Incorporating potassium nitrate salt in CaO provided outstanding leaching resistance, which was attributed to surface masking of the catalyst by in situ generated amorphous calcium carbonate [23]. Thus, the rapid leaching of Ca species was minimized and the recovered catalyst can be reused for five cycles with negligible loss in CaO. NaOH exhibited good dispersion force similar to potassium hydroxide, and they can reduce the solubility of portlandite (Ca(OH)2) [25,26]. Moreover, NaOH is more cost-effective than potassium hydroxide. This study aims to synthesize GC using highly active and robust heterogeneous waste under mild reaction conditions. Abundant LF steel slag waste is reported for the first time to catalyze glycerol transesterification reaction with DMC. NaOH was loaded on LF slag via impregnation method. The synthesized NaOH-LF catalyst was characterized and utilized in the synthesis of GC. The stability and catalytic activities of NaOH-LF were investigated. 2. Materials and methods 2.1. Materials LF steel slag was obtained from a local company in Malaysia. NaOH pellets (approximately 99% assay), anhydrous DMC (>99%), HCl (37 wt.% fuming), and methanol (>99%) were supplied by R & M Chemicals. The standards (93–95% purity) used in gas chromatography–flame ionization detector (GC-FID) were obtained from Huntsman, USA. All the chemicals were used without further processing or treatment. 2.2. NaOH-LF synthesis LF slag was screened with a standard sieve of 100–125 lm mesh to remove large-sized particles and impurities (e.g., marble, stones, and metal debris). The obtained LF was dried for 4 h at 100 °C to remove absorbed water, followed by NaOH (typically 1–15 wt.% loading) impregnation under stirring (500 rpm) for 24 h. Thereafter, the solution was heated at 80 °C for 4 h to evapo-
rate water and dried in static air at 100 °C overnight. The samples were calcined at different temperature ranges of 100–900 °C for 4 h to investigate the effect of calcination. Fresh LF slag was calcined at 800 °C for 4 h and used directly in the transesterification reaction. NaOH-LF was also used in the glycerol transesterification reaction as a catalytic material to synthesize GC. 2.3. NaOH-LF characterization The mineral compositions of the calcined LF, NaOH impregnated LF and leaching of the constituent minerals for reused catalyst were characterized using energy-dispersive spectroscopy (EDS; FE-SEM LEO SUPRA 35VP). Nitrogen adsorption–desorption isotherms were obtained using Autosorb I (Quantachrome Corporation, USA) at 77 K. Multipoint Brunauer–Emmet–Teller (BET) method was utilized to calculate the surface area of the samples. Average pore diameter and pore size distribution were calculated using the Barrett–Joyner–Halen da method. The base amount of the samples was investigated using the acid–base titration method [23]. A total of 100 mg of the samples were dispersed in 10 mL of 0.5 M HCl solution, and the suspension was stirred for 24 h. Thereafter, the solution was filtered using filter paper, and the acid remaining in the liquid phase was titrated against 0.1 M NaOH solution using phenolphthalein as an indicator. The base amount was calculated using the same standard equation reported in Ref. [24]. Temperature-programmed desorption using CO2 (TPD-CO2) as the probe molecule was used to investigate the base amount and the basic sites strength of the catalyst samples. Impurities and trapped gasses were removed by heating 50 mg of the samples under helium flow (10 °C/min, 50 mL/min) for 2 h and from ambient temperature to 800 °C. The samples were cooled to 90 °C, and CO2 gas was allowed to adsorb (50 mL/min) for 1 h. The thermal desorption of adsorbed CO2 molecules was conducted from 90 to 900 °C under helium flow and the desorbed CO2 was monitored online using gas chromatography fitted with a thermal conductivity detector. Hammett indicators which include phenolphthalein (H_ = 9.3), 2,4-dinitroaniline (H_ = 15.0), 4-nitroaniline (H_ = 18.4), and aniline (H_ = 27.0) were used to investigate the basic strength distribution of LF, NaOH impregnated LF samples (5–15 wt.%) and five times reused catalyst [14,23,27]. Samples powder (50 mg) was transferred into 10 mL cyclohexane and sonicated for 2 h. Thereafter, indicator dissolved in benzene (0.75 wt.%) was added dropwise to the catalyst suspension in cyclohexane. Then, it was left to equilibrate for 3 h, until no further color changes were observed. The color changes on the catalyst were monitored virtually, and the relative base strength is described as being stronger than the weakest indicator that produces a color change but weaker than the strongest indicator that shows no color change. Fourier transform infrared spectrometer (FTIR; Perkin Elmer Spectrum GX) was utilized to analyze the surface functional groups over a range of 4000 cm 1 to 400 cm 1 and a resolution of 4 cm 1. The KBr method was employed by mixing 7 mg of the sample with 100 mg of KBr, followed by grinding and pressing to form a transparent pellet. X-ray diffraction (XRD; SIEMENS D5000) for the samples and five times reused NaOH-impregnated LF were conducted using Cu Ka radiation at 30 mA/40 kV with a scanning rate of 2°/min over a 2h range of 5–90°. 2.4. Glycerol transesterification reaction with DMC Transesterification reaction was conducted in a 50 mL glass reactor equipped with a K-type thermocouple for temperature
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
measurement. The reactor was placed in a heater system fixed with a magnetic stirrer at 500 rpm to overcome the mass transfer barrier [28]. A total of 9 mL (11.4 g, 124.2 mmol) of glycerol and 21 mL of DMC (22.34 g, 248.08 mmol) were reacted with 0.572 g of the NaOH-LF catalyst under specified reaction conditions. At the end of each reaction, samples were drawn for quantitative analysis using gas chromatography equipped with flame ionization detector (GC-FID; Shimadzu 2010 Plus, Japan). The samples were analyzed by extending 20 lm of the sample into 1 mL of pyridine solvent. The reaction products (i.e., glycerol conversion and GC yield) were calculated using methods reported in the Ref. [23]. The adopted GC-FID settings, column, and carrier gas are the same as reported by [20]. 2.5. Effect of glycerol water content on catalytic performance The applications of biodiesel-derived glycerol are highly limited because of in situ water and other impurities. Thus, water content test was conducted to determine catalyst robustness against water. Crude glycerol was represented by adding distilled water in the range of 0.75 to 5 wt.% relative to glycerol weight and its effect on glycerol conversion and GC yield was investigated. 2.6. Leaching and reusability studies The relative loss of mineral elements that are active sites for the transesterification reaction was determined using EDX technique. The recovered catalyst, 10 wt.% NaOH impregnated LF calcined at 800 °C, from the reaction was washed with methanol (50 mL per wash) and dried under static air at 100 °C for 5 h. NaOH has much lower solubility in methanol than water, although it can react to form sodium methoxide. However, the washing time (5 min per wash) and volume of methanol are limited to initiate sufficient or rapid leaching of incorporated Na. The recovered catalyst was reused in five successive cycles of transesterification reaction adopting the same pretreatment at the end of each cycle. 3. Results and discussion 3.1. Effect of NaOH loading on LF slag and calcination temperature Glycerol transesterification reaction with DMC was conducted using various NaOH loadings (1–15 wt.%) on the LF slag and calcined LF sample. The effect of calcination temperature on the conversion and yield of glycerol and GC was investigated. The obtained best NaOH loading on LF, which was based on the conversion and yield of glycerol and GC, was utilized for the reaction influencing parameter screening. Table 1 presents the glycerol conversion, GC yield, glycidol yield, and bulk basicity of the catalytic materials. The calcined LF (800 °C) slag can remarkably catalyze the glycerol transesterification reaction to appreciable glycerol conversion and GC yield of approximately 87%. The remarkable performance of calcined LF (800 °C) slag is ascribed to the high surface area and appreciable basicity (the material is predominantly calciumbased). The modification via NaOH incorporation increased the basicity and activity toward a high GC yield in contrast to the unmodified LF slag. NaOH loading increases the glycerol conversion and GC yield. Consequently, the surface and bulk basicity increase as the NaOH loading increases as shown in the Tables 1 and 4. The increased glycerol conversion and GC yield are ascribed to the increased amount of available strong basic sites as the NaOH loadings increase. However, GC decomposes to glycidol beyond a certain basicity value (H P 18), which possibly explain the observed 10.67% glycidol in the 15 wt.% NaOH impregnated LF
3
calcined at 800 °C [14]. The 1 wt.% and 5 wt.% NaOH impregnated LF exhibited a GC yield of 87.19% and 89.22%, respectively, whereas, 96.34% GC yield was attained using 10 wt.% NaOH impregnated LF. Based on these findings, 10 wt.% NaOH impregnated LF (denoted as 10 wt.% NaOH-LF) was selected as the appropriate NaOH loading on LF. The basic properties of mixed metal oxides depend on their elemental compositions and calcination temperature [23,31,36]. Calcination changes textural and topographical nature of the catalyst, imposing relevant catalytic properties that promote transesterification reaction. Calcination was conducted at a temperature range of 500–900 °C to investigate the influence of calcination temperature on the catalytic performance of 10 wt.% NaOH-LF catalyst. The influence of calcination temperature in Table 1 shows that the amount of basic sites on the material increases as calcination temperature increases from 500 to 900 °C. A corresponding increase in the GC yield from 50 to 96% was observed. The increase in basicity as the calcination temperature increases is attributed to the conversion of weak basic sites to strong basic sites and further opening of the pore structure [36]. Hence, the 10 wt.% NaOH-LF calcined at 800 °C was utilized as the best catalyst to further investigate the influence of the reaction parameters on glycerol transesterification with DMC. 3.2. Catalyst characterization 3.2.1. EDS of LF slag and 10 wt.%, NaOH-LF calcined at 800 °C The mineral composition of the understudied samples is presented in Table 2. The result showed that calcium and oxygen are the major mineral elements in the fresh LF slag without traces of any alkali metal hydroxides. The high calcium content is consistent with the results of previously conducted studies, which indicate that the estimated calcium content of LF slags from steel production industries is 30–60 wt.% [29–31]. Oxygen and calcium content slightly increased for NaOH-impregnated calcium, whereas the sulfur content may be burnt off during calcination. Aluminum may have leached out completely from the LF slag lattice structure because of the high hydroxide concentration, and the slight increase in Ca and Mg can be ascribed to the lattice framework distortion that resulted in enlarged unit cell crystalline structure [32]. The leaching of manganese is facilitated by decomposition temperature and high alkali hydroxide content [33]. The increase in oxygen indicates that NaOH may have transformed to Na2O during calcination. Thus, the observed sodium in the 10 wt.% NaOH-LF catalyst (4.38 wt.%) confirms that sodium was incorporated into the LF. In addition to the oxygen group, Ca and Na comprise approximately 80% of the material, which has sufficient basicity suitable for base-catalyzed glycerol transesterification reaction with DMC. 3.2.2. Catalyst textural properties Table 3 presents the BET surface area, pore volume, and average pore diameter of LF and 10 wt.% NaOH impregnated LF catalyst. The BET results indicate that calcined LF slag (800 °C) as a waste material has a remarkable surface area of 72.58 m2/g, which explains its appreciable catalytic performance. Impregnating LF with NaOH (10 wt.%) significantly decreased the surface area and pore volume of LF, which can be ascribed to the increased crystalline size of the catalyst. The evident increase in mesoporosity (average pore diameter from 5.25 to 11.13 nm) indicates that pore transformation occurred because of the molecular size of Na. Thus, the increased mesoporosity possibly promotes the adsorption– desorption of glycerol transesterification reactants and products because the geometrical properties of these molecules require a mesoporous catalytic surface [34]. The nitrogen adsorption–
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
4
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
Table 1 Catalytic activity of the calcined LF slag and NaOH impregnated LF and effect of calcination.
a b
Catalyst
Glycerol conv. (%)
GC yield (%)
Glycidol yield (%)
Base amount (mmol/g)
LF (800) 1 wt.% NaOH-LF (800) 5 wt.% NaOH-LF (800) 10 wt.% NaOH-LF (800) 15 wt.% NaOH-LF (800) 10 wt.% NaOH-LF(100)b 10 wt.% NaOH-LF (500) 10 wt.% NaOH-LF (600) 10 wt.% NaOH-LF (700) 10 wt.% NaOH-LF (900) 10 wt.% NaOH-LF(800)a
87.38 90.88 95.47 98.19 98.42 50.05 85.62 94.19 95.12 98.32 96.41
86.59 87.19 89.22 96.34 89.33 50.13 82.38 90.41 92.67 96.66 94.07
– – – – 10.67 – – – – – –
35.80 37.30 38.98 39.60 40.99 20.45 29.29 34.05 36.98 41.03 38.92
Regenerated five times reused 10 wt.% NaOH-LF after calcination at 800 °C. This sample was dried at 100 °C.
Table 2 EDS of the fresh LF slag, 10 wt.% NaOH-LF (800 °C) and five times reused 10 wt.% NaOH-LF (800 °C). Elements
C
O
Mg
Na
Al
Si
S
Ca
Mn
Fe
Fresh LF slag NaOH-LF (10 wt.%, 800) Reused NaOH-LF (10 wt.%, 800)
6.64 5.85 16.87
47.66 50.67 37.22
0.75 1.72 1.91
– 4.38 1.76
2.58 – –
2.00 1.09 1.13
0.28 – –
32.95 34.83 39.82
5.90 0.97 –
1.23 0.50 –
80
Adsorption Desorption
Pore volume (cm3/g)
Average pore diameter (nm)
LF (800 °C) 10 wt.% NaOH-LF (800 °C)
72.58 29.92
0.11 0.10
5.25 11.13
desorption presented in Fig. 1 for calcined LF (800 °C) and 10 wt.% NaOH-LF calcined at 800 °C exhibited a type IV isotherm with H2 hysteresis loop. The size, shape and pore structure of materials with an irregular distribution can be represented by the H2 hysteresis loop, which is typical of highly mesoporous materials [35]. An almost linear increase in adsorption is observed for the calcined LF slag. The 10 wt.% NaOH impregnated LF catalyst showed a step increase in adsorption at a relative pressure = 0.6 to 1. The step increase can be attributed to the capillary condensation occurring at the catalyst interconnected pores. The high quantity of adsorbed nitrogen in the calcined LF slag compared to 10 wt. % NaOH impregnated LF catalyst (y-axis) correlates with the observed larger pore volume of calcined LF slag compared with that of 10 wt.% NaOH-LF pore volume. 3.2.3. TPD-CO2 analysis The performance of catalytic materials used in the glycerol transesterification reaction mainly depends on the textural properties, catalyst surface basicity, and basic strength distribution of the material [31]. Thus, TPD-CO2 was conducted to investigate the LF slag and NaOH impregnated LF slag (5–15 wt.%) basic sites concentration strength which correlate with the temperature of CO2 desorption for each catalyst sample [34]. Fig. 2 shows the TPD-CO2 of the catalyst samples. The desorption peaks at <200 °C, 300 °C to 450 °C, and >450 °C correspond to weak, moderate, and strong basic sites, respectively [37]. The CO2 desorption for the calcined LF slag (800 °C) around 550–670 °C, displayed strong peak intensity with basic sites concentration of 0.16 mmol/g. LF slag modification with NaOH loadings from 5 to 15 wt.% required a higher temperature (590–700 °C) for CO2 desorption, indicating the presence of stronger basic sites. Also, a shoulder peak is observed for all NaOH impregnated LF and five times reused 10 wt.% NaOH-LF catalyst around 780– 900 °C. The intensity of the shoulder peaks consequently increased
3
SBET (m2/g)
(a)
70 60 50 40 30 20 10 0.0
0.2
0.4
0.6
0.8
1.0
P/P 0 (Relative pressure) 70
Quantity adsorbed (cm3 /g, STP)
Catalyst
Quantity adsorbed (cm /g, STP)
Table 3 Textural properties of LF slag and 10 wt.% NaOH-LF (800 °C) catalyst.
60
(b)
Adsorption Desorption
50 40 30 20 10 0 0.0
0.2
0.4
0.6
0.8
1.0
P/P0 (Relative pressure) Fig. 1. N2 physisorption isotherms of calcined LF (a) and NaOH-LF (10 wt.%, 800 °C) (b) catalysts.
with increasing NaOH loading (5–15 wt.%), with a corresponding increase in the total basic sites concentration as shown in Table 4. The five times reused 10 wt.% NaOH-LF catalyst has desorption
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
(e)
TCD Signal (a.u.)
(d) (c) (b)
(a)
100
200
300
400
500
600
700
800
900
Temperature (oC) Fig. 2. TPD-CO2 of calcined LF slag (800 °C) (a), 5 wt.% NaOH-LF (800 °C) (b), 10 wt.% NaOH-LF (800 °C) (c), 15 wt.% NaOH-LF (800 °C) (d), and five times reused 10 wt.% NaOH-LF (800 °C) (e).
Table 4 Basic sites concentration and basic strength distribution of catalyst samples. Catalyst
Basic amount (mmol/g)
Basic strength (H_)
LF (800 °C) 5 wt.% NaOH-LF (800 °C) 10 wt.% NaOH-LF (800 °C) 15 wt.% NaOH-LF (800 °C) Five times reused 10 wt.% NaOH-LF (800 °C)
0.16 0.08 0.14 0.15 0.11
9.3 < H_ < 15 15 < H_ < 18.4 15 < H_ < 18.4 18.4 < H_ < 27 15 < H_ < 18.4
temperature around 550–670 °C with a shoulder peak. The amount of the basic sites on the five times reused catalyst is lower than that of 10 wt.% NaOH-LF. The decreased basic sites concentration can be ascribed to leached Na elements, as evidenced in the EDX analysis. The distribution of the basic strength on the LF and NaOH impregnated LF was investigated using Hammett indicators. Evidently, the basic strength of fresh LF (9.3 < H_ < 15) is lower than all NaOH modified LF (5–15 wt.%) samples. This indicates that NaOH enhances the LF slag basic properties for appreciably higher catalytic performance. Pan et al. [14] and Hu et al. [23], reported that basic strength above H_ > 18.4, is likely to initiate GC decomposition to glycidol, which confirms the appearance of glycidol in the 15 wt.% NaOH-LF catalyst (see Table 1). Hence, moderate basic strength (15 < H_ < 18.4) and strong basic sites are crucial for higher glycerol conversion and GC yield without glycidol formation.
3.2.4. Fourier transform infrared, FTIR analysis The infrared spectra of the fresh LF slag, calcined LF slag (800 °C), 10 wt.% NaOH-LF calcined at 800 °C, and five times reused 10 wt.% NaOH-LF catalyst are presented in Fig. 3. The weak spectra band from 870 to 880 cm 1 and broad peaks at approximately 1400 cm 1 are attributed to the carbonate impurities usually generated from CaO carbonation in the air during FTIR spectrum analysis. The IR peaks at 999, 924, and 944 cm 1 are ascribed to the symmetric stretching of SiAO2 [38]. The IR spectra band at 2942.53 cm 1, which was observed in the reused calcined LF and 10 wt.% NaOH-LF catalysts is ascribed to the CAH stretching mode, which is likely from organic matter that are in contact with the catalysts [39]. The broad band at 3340 cm 1 is attributed to symmetric and asymmetric OAH stretching vibrations of water molecules.
5
The observed sharp peak at approximately 3640 cm 1 is assigned to the stretching hydroxyl functional group. Although several studies on Portland cement identified the IR band at 3640 cm 1 as AOH stretching of Ca(OH)2, assigning this IR band is contentious if NaOH is present [25,43,45,46] because Ca(OH)2 and NaOH have similar absorption bands, which can be attributed to their relatively close ionic radii. Ca and Na have strong basic sites and are regarded as strong bases. The possibility of mixed oxide formation may be negligible or the incorporated Na+ may have diffused and dispersed uniformly into the CaO structures, thus hindering the conspicuous identification of the Na2O crystalline phase. Thus, the initial interaction between two hydroxides can be attributed to the incorporation of alkali ions (Na+) into CaO during calcination and hydrogen bond formation [41]. Therefore, NaOH decomposed into the CaO matrix as Na2O during calcination, increasing the basicity of the LF. The reused fresh LF calcined at 800 °C exhibited a crystalline phase distortion because some of the IR spectra peaks patterns flatten out, which explain the poor catalytic activity observed at the fifth reuse of the unmodified LF. Incorporating NaOH into the LF resulted in a crystalline IR pattern similar to fresh LF sample. The sustained five cycles of reuse of 10 wt.% NaOH-LF catalyst exhibited a similar spectra pattern to that of the unused 10 wt.% NaOH-LF calcined at 800 °C. 3.2.5. XRD of catalyst samples The XRD pattern of the unmodified LF, 10 wt.% NaOH-LF calcined at 800 °C, five times reused 10 wt.% NaOH-LF and regenerated five times reused 10 wt.% NaOH-LF catalyst re-calcined at 800 °C are shown in Fig. 4. The XRD pattern showed the phases of CaO, portlandite (Ca(OH)2), calcium silicate, and calcium carbonate. The crystalline phase of the steel slag is mainly composed of portlandite, which is consistent with the results of previous studies. The characteristic peaks at 2h of 18.12°, 28.63°, 34.14°, and 50.82° were identified as portlandite (Ca(OH)2) [43,46,47]. The peaks at 2h of 32.14°, 37.47°, 54.57°, 62.58°, and 64.92° were ascribed to CaO. Calcium silicate (wollastonite) was observed for all the samples at 2h of 43.59° [29]. The calcium carbonate (CaCO3) peak observed at 2h of 39.81° is possibly from CaO carbonation during XRD analysis or storage under ambient conditions. The catalyst that was reused five times showed an irregular crystalline structure with a more intense peak of portlandite at 2h of 29.03° compared with that of the unused 10 wt.% NaOH-LF calcined at 800 °C catalyst. A reduction in the peak intensity of CaO occurred, indicating that organic materials from the reactants may have masked or stuck to the catalyst surface [42]. This observation corresponds to and explains the emergence of the CAH band on the FTIR spectra for the catalyst reused five times. Also, the conducted EDX confirms the presence of contacting organic matter evidenced by increased carbon content (see Table 2). However, when the reused catalyst was subjected to regeneration step by calcination at 800 °C, the crystalline structure was restored with a slight decrease in peak intensity. NaOH peaks were not observed in the XRD crystalline pattern of all the samples, thus confirming the transformation to Na2O that is uniformly dispersed in the slag lattice. This result is consistent with that of a reported study [41]. 3.3. Effect of reaction influencing parameters on glycerol transesterification with DMC-catalyzed by 10 wt.% NaOH-LF calcined at 800 °C The glycerol conversion and GC yield were probed at different reaction temperatures ranging from 30 to 80 °C under the DMCto-glycerol molar ratios of 1:1 and 2:1 (Fig. 5a). GC yield and glycerol conversion increase with temperature from 30 to 80 °C for the 1:1 and 2:1 DMC-to-glycerol molar ratios. This result explains the reaction rate dependence on temperature. Thus, the molecular
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
6
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
(d)
-OH
2942.53 -CO3 2-
Transmittance (%)
1427.39
(c) 941.30
877.65
(b) 1478.50
(a)
874.76 3642.73 1431.24
4000
3500
3000
2500
2000
1500
1000
500
-1
Wavenumber (cm ) Fig. 3. FTIR of calcined LF (800 °C) (a), five times reused LF slag (800 °C) (b), 10 wt.% NaOH-LF (800 °C) (c), and five times reused 10 wt.% NaOH-LF (800 °C) (d).
Ca(OH) 2 , CaCO3 ,
*
(d)
Intensity (a.u.)
*
CaO
CaSiO 3
*
(C)
(b)
(a)
10
20
30
*
40 50 60 2 ( o )
70
80
90
Fig. 4. XRD pattern of calcined LF (800 °C) (a), 10 wt.% NaOH-LF (800 °C) (b), five times reused 10 wt.% NaOH-LF (800 °C) (c), and regenerated five times reused 10 wt.% NaOH-LF (re-calcined at 800 °C) (d).
Glycerol conv. & GC yield (%)
100
80
Glycerol conv. (%) (@1:2) GC yield (%) (@1:2) Glycerol conv. (%) (@1:1) GC yield (@1:1)
60
40
20
0 30
45
60
70 75 Reaction temperature (oC)
80
Fig. 5a. Glycerol conversion and GC yield at the DMC-to-glycerol molar ratios of 1:1 and 2:1 and the temperature range of 30–80 °C. 3 wt.% catalyst dosage, 90 min, and 500 rpm.
speed of the reactants (glycerol and DMC) relative to one another increases with reaction temperature, leading to collisions and bond cleavages [43]. Meanwhile, the glycerol conversion and GC yield achieved equilibrium (because the product concentration does
not change as temperature increases) from 75 to 80 °C for the 2:1 DMC-to-glycerol molar ratio. The equimolar ratio of DMC-toglycerol increased steadily as temperature increases. However, the 1:1 DMC-to-glycerol molar ratio did not achieve full glycerol conversion (56% glycerol conversion and 51% GC yield) under the reaction conditions. A previous study on CaO-catalyzed glycerol transesterification revealed that the DMC-to-glycerol molar concentration of 1 results in the serious dissolution of Ca species in glycerol exhibiting homogeneous catalysis [18]. The solubility of Ca species in the polar glycerol decreased significantly when the DMC amount was increased. A consequent increase in glycerol conversion and yield was observed above the equimolar DMC-toglycerol molar ratio. Moreover, the hydrophilic glycerol and hydrophobic DMC creates a biphasic system that seriously limits miscibility. This could possibly result in preferential strong adsorption of glycerol on the 10 wt.% NaOH-LF catalyst active sites, because of glycerol polarity, leading to lower GC yield and glycerol conversion as shown in DMC/glycerol ratio of 1. Hence, increasing contact between DMC and glycerol and surmounting the biphasic system that drives the equilibrium reaction, excess DMC is utilized. The excess DMC above stoichiometric value acts as a solvent and reagent to promote the catalytic reaction. As a reagent and solvent (b.p. 90 °C), excess DMC positively shifts the equilibrium to promote higher glycerol conversion and can easily be separated from reaction products via distillation, respectively. Thus, increased DMC concentration limits the interaction of glycerol with portlandite (Ca(OH)2), thus preventing soluble Ca species formation. Under the reaction conditions of 2:1 DMC-to-glycerol molar ratio, 3 wt.% catalyst dosage, stirring speed of 500 rpm, and 90 min reaction time, the conversion and yield of glycerol and GC were approximately 99% and 97%, respectively, at 75 °C. Therefore, 75 °C and 2:1 DMC-to-glycerol molar ratio were selected as the best temperature and molar ratio for further reaction influencing parameter screening, respectively. Active materials containing functional sites speed up the chemical reaction for non-spontaneous chemical reactions by reducing the activation energy barrier. Thus, 10 wt.% NaOH-LF catalytic activity in glycerol transesterification reaction with DMC was investigated utilizing different amounts of catalyst dosage (1– 4 wt.%) relative to glycerol weight, as presented in Fig. 5b. The conversion and yield of glycerol and GC increased as the amount of catalyst increases from 1 to 4 wt.%. This observation can be attributed to the increase in available active basic sites necessary to
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
7
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
100
stable until 120 min. The reaction achieved equilibrium at 90 min because no changes in product concentration were observed.
GC yield (%)
Glycerol conv. & GC yield (%)
Glycerol conv. (%) 80
3.4. Water content test
60
40
20
0
1
2 3 Catalyst dosage (wt.%)
4
Fig. 5b. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 1–4 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
deprotonate glycerol and the subsequent reaction with DMC leading to GC formation. The 3 wt.% catalyst dose relative to glycerol weight achieved approximately 98% glycerol conversion and 97% GC yield at 90 min reaction time, 75 °C, and 2:1 DMC-to-glycerol molar ratio. Catalyst loading above a certain value may limit the mass transfer from the bulk to the catalyst active sites and induce particle agglomeration, thus hindering further glycerol conversion [36]. At >3 wt.% catalyst dosage, no significant changes (saturation of active basic sites) in glycerol conversion and GC yield were observed, indicating that NaOH-LF is stable against particle abrasion, which is normally experienced in slurry mixtures. The effect of reaction time on glycerol conversion and GC yield was investigated under reaction conditions of 75 °C, 500 rpm stirring speed, 3 wt.% catalyst dosage, 2:1 DMC-to-glycerol molar ratio, and 30–120 min. Fig. 5c shows that GC yield and glycerol conversion increases as the reaction time increases from 30 to 120 min. The 30 min reaction time resulted in glycerol conversion and GC yield of approximately 65%. The low glycerol conversion is possibly attributed to carbonate group proximity to the third glycerol OH group. The reactivity of the first OH group of glycerol proceeds faster than the third OH group and the intramolecular reaction regime for the second OH is faster than the intramolecular reaction of the third OH group with the next DMC molecule forming a thermodynamically stable cyclic carbonate [28]. Thus, more reaction time is required to achieve reaction equilibrium. Therefore, at 90 min reaction time, the glycerol conversion and GC yield were approximately 98% and 96%, respectively, and remained
10 wt.% NaOH-LF catalyst stability against leaching of active sites and reusability were investigated. Fig. 7 indicates that the catalyst could still sustain >90% glycerol conversion and 80% GC yield. However, the opposite result is the case for 800 °C calcined LF sample where the sample that was reused five times showed only 19% and 18% glycerol conversion and GC yield, respectively. The significant decrease in the catalytic performance of the LF slag is possibly attributed to the serious leaching of the active sites (CaO). The reduction in the performance of 10 wt.% NaOH-LF catalyst could be ascribed partly to the leached constituents of the materials and adhesion of the organic matters in the catalyst pores, as evidenced by the FTIR, EDX and XRD results. Although deactivation of the catalyst is not rapid as shown in the reusability studies, however, the slightly leached Na or CaO poses separation difficulties which increases overall energy input of the transesterification process. Nevertheless, the Na metals can be separated from the reaction mixture by neutralization step utilizing sulfuric acid, hydrochloric or phosphoric acid [44]. Utilizing phosphoric acid to remove leached NaOH via crystallization/precipitation is a considerable route because, depending on the concentration of dissolved Na, the pre-
GC yield (%) Glycerol conv. (%)
80
60
40
20
0
3.5. Leaching and reusability tests of calcined LF slag (800 °C) and 10 wt.% NaOH-LF calcined at 800 °C
Glycerol conv. & GC yield (%)
Glycerol conv. & GC yield (%)
100
Biodiesel-derived glycerol contains water. Thus, synthesizing water-resistant catalytic material is economical because refining glycerol is energy-intensive. Developing materials that exhibit similar catalytic activity with crude and refined glycerol is desirable for commercialization. Therefore, the influence of water content ranging from 0.75 wt.% to 5 wt.% relative to glycerol weight was investigated under reaction conditions of 75 °C, 2:1 DMC-toglycerol molar ratio, 3 wt.% catalyst dosage, 500 rpm stirring speed, and 90 min reaction time. Fig. 6 shows that GC yield and glycerol conversion decrease as water content increases, which is likely attributed to the selective strong adsorption of water molecules in the active sites of 10 wt.% NaOH-LF calcined at 800 °C catalyst or leaching of the active sites (CaO or Na2O) since they are highly unstable under moisture. Thus, the number of available sites for glycerol deprotonation is seriously reduced. However, the conversion and yield of glycerol and GC could still be higher than 50% at 5 wt.% water content, which indicates reasonable water resistance by the 10 wt.% NaOH-LF catalyst.
30
60 90 Reaction time (min)
120
Fig. 5c. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 30– 120 min; stirring speed = 500 rpm.
100 80 60 40 20 0
GC yield (%) Glycerol conv. (%) 0.75
1 3 Water content (wt.%)
5
Fig. 6. Effect of water content on glycerol conversion and GC yield. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
8
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
Glycerol conv. (%)-10wt% NaOH-LF
GC yield (%)-10wt% NaOH-LF
Glycerol conv. (%)-LF(800)
GC yield (%)-LF(800)
Glycerol conv. & GC yield (%)
100 80
the reaction conditions utilizing 10 wt.% NaOH-LF calcined at 800 °C achieved 99% glycerol conversion and 96.59% GC yield under 90 min reaction time. NaOH-LF exhibited a reasonable resistance to in situ water in glycerol (50% glycerol conversion at 5 wt.% glycerol water content), however, deactivation is inevitable due to leaching of active sites. Thus, 10 wt.% NaOH-LF was reused five times without showing serious deactivation.
60
Acknowledgment
40 20 0
0
1
2 3 No. of reuse cycles
4
5
Fig. 7. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
cipitated disodium or trisodium phosphates salts are almost insoluble in an organic solvent [45,46]. Moreover, the trisodium phosphate has been successfully applied as an efficient catalyst in base-catalyzed BD production [45]. EDX (see Table 2) of the five times reused catalyst sample, revealed about 11.02 wt.% increase in carbon element of the catalyst, suggesting possible masking of catalyst active sites by contacting organic matters, whereas, the Na element reduction about 2.62 wt.% confirms leaching of basic active site necessary for the transesterification reaction. Thus, plausible deactivation mechanisms can be proposed for 10 wt.% NaOH-LF calcined at 800 °C. Portlandite (calcium hydroxide), which is the bulk of LF, comes in contact with glycerol to form calcium diglyceroxide and a small amount of water. On the other hand, a small amount of glycerol is subjected to etherification to produce a small amount of water under a strong alkali hydroxide concentration. In the presence of in situ water in the glycerol transesterification process, GC and/or excess DMC likely undergo hydrolysis, producing CO2 and glycerol, as reported in [47]. Water may not be generated or produced in significant amounts to initiate serious or rapid deactivation of the catalytic material because of the mild reaction conditions used in this study. However, the interaction of the generated CO2 and portlandite impelled or promoted the formation of amorphous calcium carbonate layer on the catalyst surface. To a certain extent, this layer of CaCO3 inhibits the attrition of CaO (Na2O) and portlandite from the inner catalyst layer [23]. The potential mass transport of the reactant molecules into the loose CaCO3 outer layer structure is possible (or the formed CaCO3 outer layer is leached completely by the reactants during the reaction) into the inner catalyst layer, thus allowing sufficient glycerol contact with the catalyst active sites. After regeneration at 800 °C, the five times reused 10 wt.% NaOH-LF catalyst activity was restored with 96% glycerol conversion and 94% GC yield.
4. Conclusions LF slag can be utilized to synthesize GC in the transesterification reaction. However, the serious deactivation of active sites of the catalyst resulted in poor GC yield after five cycles of reuse. LF slag that impregnated with NaOH exhibited a remarkable GC yield because of the increased basicity and basic strength compared with that of calcined LF (800 °C). The calcination temperature modified the transesterification reaction to result in an evidently high GC yield because weak basic sites are converted to strong basic sites as calcination temperature increases. The unique severity of
The first author acknowledges the financial support provided by Universiti Sains Malaysia (USM), under the USM Fellowship Scheme. The authors acknowledge the financial support provided by Ministry of Education, Malaysia under the Transdisciplinary Research Grant Scheme (TRGS) Phase 2/2014 (203/ PJKIMIA/6762002). References [1] da Silva GP, Mack M, Contiero J. Glycerol: a promising and abundant carbon source for industrial microbiology. Biotechnol Adv 2009;27:30–9. [2] Yang F, Hanna Ma, Sun R. Value-added uses for crude glycerol–a byproduct of biodiesel production. Biotechnol Biofuels 2012;5:13. [3] Remon J, Gimenez JR, Valiente A, Garcia L, Arauzo J. Production of gaseous and liquid chemicals by aqueous phase reforming of crude glycerol: influence of operating conditions on the process. Energy Convers Manage 2016;110:90–112. [4] Teng WK, Ngoh GC, Yusoff R, Aroua MK. A review on the performance of glycerol carbonate production via catalytic transesterification: effects of influencing parameters. Energy Convers Manage 2014;88:484–97. [5] Dou B, Wang C, Song Y, Chen H, Xu Y. Activity of Ni-Cu-Al based catalyst for renewable hydrogen production from steam reforming of glycerol. Energy Convers Manage 2014;78:253–9. [6] Galera S, Gutiérrez Ortiz FJ. Life cycle assessment of hydrogen and power production by supercritical water reforming of glycerol. Energy Convers Manage 2015;96:637–45. [7] Zakaria ZY, Amin NAS, Linnekoski J. Optimization of catalytic glycerol steam reforming to light olefins using Cu/ZSM-5 catalyst. Energy Convers Manage 2014;86:735–44. [8] Veiga S, Bussi J. Steam reforming of crude glycerol over nickel supported on activated carbon. Energy Convers Manage 2016. http://dx.doi.org/10.1016/j. enconman.2016.04.103 (in press). [9] Ishak ZI, Sairi NA, Alias Y, Aroua MKT, Yusoff R. Production of glycerol carbonate from glycerol with aid of ionic liquid as catalyst. Chem Eng J 2016;297:128–38. [10] Liu Z, Wang J, Kang M, Yin N, Wang X, Tan Y, et al. Structure-activity correlations of LiNO3/Mg4AlO5.5 catalysts for glycerol carbonate synthesis from glycerol and dimethyl carbonate. J Ind Eng Chem 2015;21:394–9. [11] Esteban J, Fuente E, Blanco A, Ladero M, Garcia-Ochoa F. Phenomenological kinetic model of the synthesis of glycerol carbonate assisted by focused beam reflectance measurements. Chem Eng J 2015;260:434–43. [12] Ochoa-Gómez JR, Gómez-Jiménez-Aberasturi O, Ramírez-López C, Belsué M. A brief review on industrial alternatives for the manufacturing of glycerol carbonate, a green chemical. Org Process Res Dev 2012;16:389–99. [13] Malyaadri M, Jagadeeswaraiah K, Sai Prasad PS, Lingaiah N. Synthesis of glycerol carbonate by transesterification of glycerol with dimethyl carbonate over Mg/Al/Zr catalysts. Appl Catal A Gen 2011;401:153–7. [14] Pan S, Zheng L, Nie R, Xia S, Chen P, Hou Z. Transesterification of glycerol with dimethyl carbonate to glycerol carbonate over Na–based zeolites. Chinese J Catal 2012;33:1772–7. [15] Okoye PU, Hameed BH. Review on recent progress in catalytic carboxylation and acetylation of glycerol as a byproduct of biodiesel production. Renew Sustain Energy Rev 2016;53:558–74. [16] Zheng L, Xia S, Hou Z, Zhang M, Hou Z. Transesterification of glycerol with dimethyl carbonate over Mg-Al hydrotalcites. Chinese J Catal 2014;35:310–8. [17] Zheng L, Xia S, Lu X, Hou Z. Transesterification of glycerol with dimethyl carbonate over calcined Ca-Al hydrocalumite. Cuihua Xuebao/Chinese J Catal 2015;36:1759–65. [18] Simanjuntak FSH, Kim TK, Lee SD, Ahn BS, Kim HS, Lee H. CaO-catalyzed synthesis of glycerol carbonate from glycerol and dimethyl carbonate: isolation and characterization of an active Ca species. Appl Catal A Gen 2011;401:220–5. [19] Bai R, Wang S, Mei F, Li T, Li G. Synthesis of glycerol carbonate from glycerol and dimethyl carbonate catalyzed by KF modified hydroxyapatite. J Ind Eng Chem 2011;17:777–81. [20] Algoufi YT, Hameed BH. Synthesis of glycerol carbonate by transesterification of glycerol with dimethyl carbonate over K-zeolite derived from coal fly ash. Fuel Process Technol 2014;126:5–11. [21] Worldsteel Association. April 2016 crude steel production. Brussels, Belgium; 2016.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx [22] Lewis DW. Properties and uses of iron and steel slags. In: Nationa Slag Association, editor. Symp slag natl inst transp road res South Africa, South Africa. p. 1–11. [23] Hu K, Wang H, Liu Y, Yang C. KNO3/CaO as cost-effective heterogeneous catalyst for the synthesis of glycerol carbonate from glycerol and dimethyl carbonate. J Ind Eng Chem 2014;28:334–43. [24] Khayoon MS, Hameed BH. Mg1+xCa1 xO2 as reusable and efficient heterogeneous catalyst for the synthesis of glycerol carbonate via the transesterification of glycerol with dimethyl carbonate. Appl Catal A Gen 2013;466:272–81. [25] Duchesne J, Reardon EJ. Measurement and prediction of portlandite solubility in alkali solutions. Cem Concr Res 1995;25:1043–53. [26] Pallagi A, Tasi Á, Gácsi A, Csáti M, Pálinkó I, Peintler G, et al. The solubility of Ca (OH)2 in extremely concentrated NaOH solutions at 25 °C. Cent Eur J Chem 2012;10:332–7. [27] Simanjuntak FSH, Lim SR, Ahn BS, Kim HS, Lee H. Surfactant-assisted synthesis of MgO: characterization and catalytic activity on the transesterification of dimethyl carbonate with glycerol. Appl Catal A Gen 2014;484:33–8. [28] Yadav GD, Chandan PA. A green process for glycerol valorization to glycerol carbonate over heterogeneous hydrotalcite catalyst. Catal Today 2014;237:47–53. [29] Setién J, Hernández D, González JJ. Characterization of ladle furnace basic slag for use as a construction material. Constr Build Mater 2009;23:1788–94. [30] Shi C. Characteristics and cementitious properties of ladle slag fines from steel production. Cem Concr Res 2002;32:459–62. [31] Ochoa-Gómez JR, Gómez-Jiménez-Aberasturi O, Maestro-Madurga B, Pesquera-Rodríguez A, Ramírez-López C, Lorenzo-Ibarreta L, et al. Synthesis of glycerol carbonate from glycerol and dimethyl carbonate by transesterification: catalyst screening and reaction optimization. Appl Catal A Gen 2009;366:315–24. [32] Bkour Q, Faqir N, Shawabkeh R, Ul-Hamid A, Bart HJ. Synthesis of a Ca/Naaluminosilicate from kaolin and limestone and its use for adsorption of CO2. J Environ Chem Eng 2016;4:973–83. [33] Wang X, Zheng S, Xu H, Zhang Y. Leaching of niobium and tantalum from a low-grade ore using a KOH roast-water leach system. Hydrometallurgy 2009;98:219–23.
9
[34] Alvarez MG, Segarra AM, Contreras S, Sueiras JE, Medina F, Figueras F. Enhanced use of renewable resources: transesterification of glycerol catalyzed by hydrotalcite-like compounds. Chem Eng J 2010;161:340–5. [35] Benhouria A, Islam MA, Zaghouane-Boudiaf H, Boutahala M, Hameed BH. Calcium alginate–bentonite–activated carbon composite beads as highly effective adsorbent for methylene blue. Chem Eng J 2015;270:621–30. [36] Liu P, Derchi M, Hensen EJM. Promotional effect of transition metal doping on the basicity and activity of calcined hydrotalcite catalysts for glycerol carbonate synthesis. Appl Catal B Environ 2014;144:135–43. [37] Kumar P, With P, Srivastava VC, Gläser R, Mishra IM. Glycerol carbonate synthesis by hierarchically structured catalysts: catalytic activity and characterization. Ind Eng Chem Res 2015;54:12543–52. [38] Trezza MA. Hydration study of ordinary portland cement in the presence of zinc ions. Mater Res 2007;10:331–4. [39] Chen G, Zhang B, Zhao J. Dispersion process and effect of oleic acid on properties of cellulose sulfate-oleic acid composite film. Materials (Basel) 2015;8:2346–60. [41] Maiti GC, Baerns M. Dehydration of sodium hydroxide and lithium hydroxide dispersed over calcium oxide catalysts for the oxidative coupling of methane. Appl Catal A Gen 1995;127:219–32. [42] Syamsuddin Y, Hameed BH. Synthesis of glycerol free-fatty acid methyl esters from Jatropha oil over Ca-La mixed-oxide catalyst. J Taiwan Inst Chem Eng 2016;58:181–8. [43] Li J, Wang T. Chemical equilibrium of glycerol carbonate synthesis from glycerol. J Chem Thermodyn 2011;43:731–6. [44] Chiu CW, Dasari MA, Sutterlin WR, Suppes GJ. Removal of residual catalyst from simulated biodiesel’s crude glycerol for glycerol hydrogenolysis to propylene glycol. Ind Eng Chem Res 2006;45:791–5. [45] Thinnakorn K, Tscheikuna J. Biodiesel production via transesterification of palm olein using sodium phosphate as a heterogeneous catalyst. Appl Catal A Gen 2014;476:26–33. [46] Jiang ST, Zhang FJ, Pan LJ. Sodium phosphate as a solid catalyst for biodiesel preparation. Braz J Chem Eng 2010;27:137–44. [47] Li J, Wang T. On the deactivation of alkali solid catalysts for the synthesis of glycerol carbonate from glycerol and dimethyl carbonate. React Kinet Mech Catal 2011;102:113–26.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate P.U. Okoye, A.Z. Abdullah, B.H. Hameed ⇑ School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, 14300 Nibong Tebal, Penang, Malaysia
a r t i c l e
i n f o
Article history: Received 4 August 2016 Received in revised form 28 September 2016 Accepted 16 October 2016 Available online xxxx Keywords: Dimethyl carbonate Glycerol Glycerol carbonate Ladle furnace steel slag Transesterification
a b s t r a c t Abundant waste from ladle furnace (LF) steel slag industry was utilized to catalyze the glycerol transesterification reaction with dimethyl carbonate (DMC) to synthesize valuable glycerol carbonate (GC). LF slag was modified using various sodium hydroxide (NaOH) loadings (1–15 wt.%), and the catalyst samples were characterized using XRD, FTIR, TPD-CO2, and EDS techniques. Modifying the LF using different NaOH loadings enhanced the stability, basicity, and basic strength of LF. Favorable reaction conditions at the best glycerol conversion and GC yield were investigated via reaction influencing parameter screening. Thus, the 10 wt.% NaOH incorporated in LF under the reaction conditions of 75 °C, DMC-to-glycerol molar ratio of 2, and 3 wt.% catalyst dosage achieved 99% glycerol conversion and 97% GC yield. The catalyst exhibited sufficient heterogeneity and was utilized in five successive cycles of the experiment without serious deactivation. Ó 2016 Published by Elsevier Ltd.
1. Introduction Glycerol, a major biodiesel production by-product, constitutes 10% w/w of the total global biodiesel [1]. The oversupply and physicochemical properties of glycerol make it a valuable material that requires defunctionalization [2]. Pathways to synthesize fine chemicals from glycerol were discussed in detail in Refs. [3–8]. The base-catalyzed transesterification reaction of glycerol using dialkyl carbonate sources is a popular pathway to upgrade abundant glycerol into fine chemicals [9–11]. Dialkyl carbonate sources, such as dimethyl carbonate (DMC), is a popular choice because these sources are environment-friendly and nontoxic [12]. Under mild reaction conditions, basic catalysts were employed to modify the glycerol transesterification reaction with DMC to synthesize valuable glycerol carbonate (GC). The base catalytic materials are synthesized mainly from alkaline earth metals and alkali metal oxides. Their outstanding activity in transesterification reaction is attributed to their strong basic sites and high basic density [13]. Too high a basic strength above H_ > 18.4 can propagate GC decomposition or dehydration to glycidol [14]. Methanol is coproduced in the reversible equilibrium reaction, and it can be positively shifted to produce appreciably high GC, utilizing DMC in excess of glycerol. GC has various applications, such as cement composite aggregate, solvent, surfactant, plant bolster, carrier in ⇑ Corresponding author.
lithium ion battery, methanol replacement to avoid glycerol coproduction in biodiesel synthesis, and fuel additive [15]. Reported studies based on CaO, MgO, double-layered hydrotalcite of Ca or Mg, Na-zeolite, K-zeolite, and sodium hydroxide (NaOH)/cAl2O3 alkaline materials describe the considerable high activity performance toward a high GC yield [16–20]. Research in the recent decade focused on the beneficial utilization of abundant wastes generated from many processing industries. Waste utilization decreases handling problems and averts environmental hazards caused by improper disposal. Millions of tons of waste are generated in many steel industries around the world. Available information on steelmaking processes indicates that approximately 135 million tons of steel were produced in April 2016 from 66 countries [21]. The volume of generated slag increases with the consumption of steel products, which affects utilization methods. Slag waste is categorized as electric arc furnace (EAF) steel slag, ladle furnace (LF) basic slag, blast furnace (BF) iron slag, and basic oxygen furnace (BOF) slag based on the source furnace [22]. Iron and steel slag are mostly utilized in structural and civil construction. Slag is applied in various areas, such as soil stabilization in erosion-prone areas, soil acidity correction, fertilizers, and adsorbents in CO2 sequestration. The varying mineralogy in different proportions provides steel slag wastes with properties that could be further explored from the economic and technical standpoints. The slag mineral compositions for each slag category provide specific characteristics for specific applications [22]. For instance, BOF and EAF slag are often
E-mail address: [email protected] (B.H. Hameed). http://dx.doi.org/10.1016/j.enconman.2016.10.067 0196-8904/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
2
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
used as low-cost materials in road construction and asphalt aggregate mixes. BF slag from iron industry can be used in Portland cement manufacturing. LF slag application differs because of its specific mineral constituents and characteristics. Notably, LF slag is largely composed of lime (Ca(OH)2) with primary applications in soil acidity correction and is a recycle material to formulate Portland clinker [22]. The slag is generated during the desulfurization of steel at the final stage of the steelmaking process, known as secondary metallurgy process. LF slag is also known as basic slag because of its high calcium content. Thus, LF slag is a suitable material for base-catalyzed reactions. Calcium in LF can be extracted or used directly to synthesize hydroxyapatite, which is highly beneficial in bone treatment. Calcium-based materials have exhibited potential in modifying the glycerol transesterification reaction [17]. However, these materials are highly unstable because of their solubility in glycerol at glycerol/DMC equimolar ratio and high reactivity in the presence of moisture and CO2. Calcium leaches into the reaction mixture and forms an undesired homogeneous catalysis, which results in an evident decline in catalytic activity at less than three cycles of reuse. CaO formed active homogeneous species Ca(C3H7O3)(OCO2CH3) at a glycerol/DMC equimolar ratio but exhibited heterogeneity when DMC is in excess of glycerol [18]. However, leaching cannot be avoided because a sharp decline in GC yield was recorded after the third cycle of reuse. Magnesium and potassium sources were employed to improve the stability of calcium [17,23]. The incorporation of magnesium provided a strong bond interaction, resulting in the minimal leaching (7.4%) of metallic species after four cycles of reuse [24]. Incorporating potassium nitrate salt in CaO provided outstanding leaching resistance, which was attributed to surface masking of the catalyst by in situ generated amorphous calcium carbonate [23]. Thus, the rapid leaching of Ca species was minimized and the recovered catalyst can be reused for five cycles with negligible loss in CaO. NaOH exhibited good dispersion force similar to potassium hydroxide, and they can reduce the solubility of portlandite (Ca(OH)2) [25,26]. Moreover, NaOH is more cost-effective than potassium hydroxide. This study aims to synthesize GC using highly active and robust heterogeneous waste under mild reaction conditions. Abundant LF steel slag waste is reported for the first time to catalyze glycerol transesterification reaction with DMC. NaOH was loaded on LF slag via impregnation method. The synthesized NaOH-LF catalyst was characterized and utilized in the synthesis of GC. The stability and catalytic activities of NaOH-LF were investigated. 2. Materials and methods 2.1. Materials LF steel slag was obtained from a local company in Malaysia. NaOH pellets (approximately 99% assay), anhydrous DMC (>99%), HCl (37 wt.% fuming), and methanol (>99%) were supplied by R & M Chemicals. The standards (93–95% purity) used in gas chromatography–flame ionization detector (GC-FID) were obtained from Huntsman, USA. All the chemicals were used without further processing or treatment. 2.2. NaOH-LF synthesis LF slag was screened with a standard sieve of 100–125 lm mesh to remove large-sized particles and impurities (e.g., marble, stones, and metal debris). The obtained LF was dried for 4 h at 100 °C to remove absorbed water, followed by NaOH (typically 1–15 wt.% loading) impregnation under stirring (500 rpm) for 24 h. Thereafter, the solution was heated at 80 °C for 4 h to evapo-
rate water and dried in static air at 100 °C overnight. The samples were calcined at different temperature ranges of 100–900 °C for 4 h to investigate the effect of calcination. Fresh LF slag was calcined at 800 °C for 4 h and used directly in the transesterification reaction. NaOH-LF was also used in the glycerol transesterification reaction as a catalytic material to synthesize GC. 2.3. NaOH-LF characterization The mineral compositions of the calcined LF, NaOH impregnated LF and leaching of the constituent minerals for reused catalyst were characterized using energy-dispersive spectroscopy (EDS; FE-SEM LEO SUPRA 35VP). Nitrogen adsorption–desorption isotherms were obtained using Autosorb I (Quantachrome Corporation, USA) at 77 K. Multipoint Brunauer–Emmet–Teller (BET) method was utilized to calculate the surface area of the samples. Average pore diameter and pore size distribution were calculated using the Barrett–Joyner–Halen da method. The base amount of the samples was investigated using the acid–base titration method [23]. A total of 100 mg of the samples were dispersed in 10 mL of 0.5 M HCl solution, and the suspension was stirred for 24 h. Thereafter, the solution was filtered using filter paper, and the acid remaining in the liquid phase was titrated against 0.1 M NaOH solution using phenolphthalein as an indicator. The base amount was calculated using the same standard equation reported in Ref. [24]. Temperature-programmed desorption using CO2 (TPD-CO2) as the probe molecule was used to investigate the base amount and the basic sites strength of the catalyst samples. Impurities and trapped gasses were removed by heating 50 mg of the samples under helium flow (10 °C/min, 50 mL/min) for 2 h and from ambient temperature to 800 °C. The samples were cooled to 90 °C, and CO2 gas was allowed to adsorb (50 mL/min) for 1 h. The thermal desorption of adsorbed CO2 molecules was conducted from 90 to 900 °C under helium flow and the desorbed CO2 was monitored online using gas chromatography fitted with a thermal conductivity detector. Hammett indicators which include phenolphthalein (H_ = 9.3), 2,4-dinitroaniline (H_ = 15.0), 4-nitroaniline (H_ = 18.4), and aniline (H_ = 27.0) were used to investigate the basic strength distribution of LF, NaOH impregnated LF samples (5–15 wt.%) and five times reused catalyst [14,23,27]. Samples powder (50 mg) was transferred into 10 mL cyclohexane and sonicated for 2 h. Thereafter, indicator dissolved in benzene (0.75 wt.%) was added dropwise to the catalyst suspension in cyclohexane. Then, it was left to equilibrate for 3 h, until no further color changes were observed. The color changes on the catalyst were monitored virtually, and the relative base strength is described as being stronger than the weakest indicator that produces a color change but weaker than the strongest indicator that shows no color change. Fourier transform infrared spectrometer (FTIR; Perkin Elmer Spectrum GX) was utilized to analyze the surface functional groups over a range of 4000 cm 1 to 400 cm 1 and a resolution of 4 cm 1. The KBr method was employed by mixing 7 mg of the sample with 100 mg of KBr, followed by grinding and pressing to form a transparent pellet. X-ray diffraction (XRD; SIEMENS D5000) for the samples and five times reused NaOH-impregnated LF were conducted using Cu Ka radiation at 30 mA/40 kV with a scanning rate of 2°/min over a 2h range of 5–90°. 2.4. Glycerol transesterification reaction with DMC Transesterification reaction was conducted in a 50 mL glass reactor equipped with a K-type thermocouple for temperature
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
measurement. The reactor was placed in a heater system fixed with a magnetic stirrer at 500 rpm to overcome the mass transfer barrier [28]. A total of 9 mL (11.4 g, 124.2 mmol) of glycerol and 21 mL of DMC (22.34 g, 248.08 mmol) were reacted with 0.572 g of the NaOH-LF catalyst under specified reaction conditions. At the end of each reaction, samples were drawn for quantitative analysis using gas chromatography equipped with flame ionization detector (GC-FID; Shimadzu 2010 Plus, Japan). The samples were analyzed by extending 20 lm of the sample into 1 mL of pyridine solvent. The reaction products (i.e., glycerol conversion and GC yield) were calculated using methods reported in the Ref. [23]. The adopted GC-FID settings, column, and carrier gas are the same as reported by [20]. 2.5. Effect of glycerol water content on catalytic performance The applications of biodiesel-derived glycerol are highly limited because of in situ water and other impurities. Thus, water content test was conducted to determine catalyst robustness against water. Crude glycerol was represented by adding distilled water in the range of 0.75 to 5 wt.% relative to glycerol weight and its effect on glycerol conversion and GC yield was investigated. 2.6. Leaching and reusability studies The relative loss of mineral elements that are active sites for the transesterification reaction was determined using EDX technique. The recovered catalyst, 10 wt.% NaOH impregnated LF calcined at 800 °C, from the reaction was washed with methanol (50 mL per wash) and dried under static air at 100 °C for 5 h. NaOH has much lower solubility in methanol than water, although it can react to form sodium methoxide. However, the washing time (5 min per wash) and volume of methanol are limited to initiate sufficient or rapid leaching of incorporated Na. The recovered catalyst was reused in five successive cycles of transesterification reaction adopting the same pretreatment at the end of each cycle. 3. Results and discussion 3.1. Effect of NaOH loading on LF slag and calcination temperature Glycerol transesterification reaction with DMC was conducted using various NaOH loadings (1–15 wt.%) on the LF slag and calcined LF sample. The effect of calcination temperature on the conversion and yield of glycerol and GC was investigated. The obtained best NaOH loading on LF, which was based on the conversion and yield of glycerol and GC, was utilized for the reaction influencing parameter screening. Table 1 presents the glycerol conversion, GC yield, glycidol yield, and bulk basicity of the catalytic materials. The calcined LF (800 °C) slag can remarkably catalyze the glycerol transesterification reaction to appreciable glycerol conversion and GC yield of approximately 87%. The remarkable performance of calcined LF (800 °C) slag is ascribed to the high surface area and appreciable basicity (the material is predominantly calciumbased). The modification via NaOH incorporation increased the basicity and activity toward a high GC yield in contrast to the unmodified LF slag. NaOH loading increases the glycerol conversion and GC yield. Consequently, the surface and bulk basicity increase as the NaOH loading increases as shown in the Tables 1 and 4. The increased glycerol conversion and GC yield are ascribed to the increased amount of available strong basic sites as the NaOH loadings increase. However, GC decomposes to glycidol beyond a certain basicity value (H P 18), which possibly explain the observed 10.67% glycidol in the 15 wt.% NaOH impregnated LF
3
calcined at 800 °C [14]. The 1 wt.% and 5 wt.% NaOH impregnated LF exhibited a GC yield of 87.19% and 89.22%, respectively, whereas, 96.34% GC yield was attained using 10 wt.% NaOH impregnated LF. Based on these findings, 10 wt.% NaOH impregnated LF (denoted as 10 wt.% NaOH-LF) was selected as the appropriate NaOH loading on LF. The basic properties of mixed metal oxides depend on their elemental compositions and calcination temperature [23,31,36]. Calcination changes textural and topographical nature of the catalyst, imposing relevant catalytic properties that promote transesterification reaction. Calcination was conducted at a temperature range of 500–900 °C to investigate the influence of calcination temperature on the catalytic performance of 10 wt.% NaOH-LF catalyst. The influence of calcination temperature in Table 1 shows that the amount of basic sites on the material increases as calcination temperature increases from 500 to 900 °C. A corresponding increase in the GC yield from 50 to 96% was observed. The increase in basicity as the calcination temperature increases is attributed to the conversion of weak basic sites to strong basic sites and further opening of the pore structure [36]. Hence, the 10 wt.% NaOH-LF calcined at 800 °C was utilized as the best catalyst to further investigate the influence of the reaction parameters on glycerol transesterification with DMC. 3.2. Catalyst characterization 3.2.1. EDS of LF slag and 10 wt.%, NaOH-LF calcined at 800 °C The mineral composition of the understudied samples is presented in Table 2. The result showed that calcium and oxygen are the major mineral elements in the fresh LF slag without traces of any alkali metal hydroxides. The high calcium content is consistent with the results of previously conducted studies, which indicate that the estimated calcium content of LF slags from steel production industries is 30–60 wt.% [29–31]. Oxygen and calcium content slightly increased for NaOH-impregnated calcium, whereas the sulfur content may be burnt off during calcination. Aluminum may have leached out completely from the LF slag lattice structure because of the high hydroxide concentration, and the slight increase in Ca and Mg can be ascribed to the lattice framework distortion that resulted in enlarged unit cell crystalline structure [32]. The leaching of manganese is facilitated by decomposition temperature and high alkali hydroxide content [33]. The increase in oxygen indicates that NaOH may have transformed to Na2O during calcination. Thus, the observed sodium in the 10 wt.% NaOH-LF catalyst (4.38 wt.%) confirms that sodium was incorporated into the LF. In addition to the oxygen group, Ca and Na comprise approximately 80% of the material, which has sufficient basicity suitable for base-catalyzed glycerol transesterification reaction with DMC. 3.2.2. Catalyst textural properties Table 3 presents the BET surface area, pore volume, and average pore diameter of LF and 10 wt.% NaOH impregnated LF catalyst. The BET results indicate that calcined LF slag (800 °C) as a waste material has a remarkable surface area of 72.58 m2/g, which explains its appreciable catalytic performance. Impregnating LF with NaOH (10 wt.%) significantly decreased the surface area and pore volume of LF, which can be ascribed to the increased crystalline size of the catalyst. The evident increase in mesoporosity (average pore diameter from 5.25 to 11.13 nm) indicates that pore transformation occurred because of the molecular size of Na. Thus, the increased mesoporosity possibly promotes the adsorption– desorption of glycerol transesterification reactants and products because the geometrical properties of these molecules require a mesoporous catalytic surface [34]. The nitrogen adsorption–
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
4
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
Table 1 Catalytic activity of the calcined LF slag and NaOH impregnated LF and effect of calcination.
a b
Catalyst
Glycerol conv. (%)
GC yield (%)
Glycidol yield (%)
Base amount (mmol/g)
LF (800) 1 wt.% NaOH-LF (800) 5 wt.% NaOH-LF (800) 10 wt.% NaOH-LF (800) 15 wt.% NaOH-LF (800) 10 wt.% NaOH-LF(100)b 10 wt.% NaOH-LF (500) 10 wt.% NaOH-LF (600) 10 wt.% NaOH-LF (700) 10 wt.% NaOH-LF (900) 10 wt.% NaOH-LF(800)a
87.38 90.88 95.47 98.19 98.42 50.05 85.62 94.19 95.12 98.32 96.41
86.59 87.19 89.22 96.34 89.33 50.13 82.38 90.41 92.67 96.66 94.07
– – – – 10.67 – – – – – –
35.80 37.30 38.98 39.60 40.99 20.45 29.29 34.05 36.98 41.03 38.92
Regenerated five times reused 10 wt.% NaOH-LF after calcination at 800 °C. This sample was dried at 100 °C.
Table 2 EDS of the fresh LF slag, 10 wt.% NaOH-LF (800 °C) and five times reused 10 wt.% NaOH-LF (800 °C). Elements
C
O
Mg
Na
Al
Si
S
Ca
Mn
Fe
Fresh LF slag NaOH-LF (10 wt.%, 800) Reused NaOH-LF (10 wt.%, 800)
6.64 5.85 16.87
47.66 50.67 37.22
0.75 1.72 1.91
– 4.38 1.76
2.58 – –
2.00 1.09 1.13
0.28 – –
32.95 34.83 39.82
5.90 0.97 –
1.23 0.50 –
80
Adsorption Desorption
Pore volume (cm3/g)
Average pore diameter (nm)
LF (800 °C) 10 wt.% NaOH-LF (800 °C)
72.58 29.92
0.11 0.10
5.25 11.13
desorption presented in Fig. 1 for calcined LF (800 °C) and 10 wt.% NaOH-LF calcined at 800 °C exhibited a type IV isotherm with H2 hysteresis loop. The size, shape and pore structure of materials with an irregular distribution can be represented by the H2 hysteresis loop, which is typical of highly mesoporous materials [35]. An almost linear increase in adsorption is observed for the calcined LF slag. The 10 wt.% NaOH impregnated LF catalyst showed a step increase in adsorption at a relative pressure = 0.6 to 1. The step increase can be attributed to the capillary condensation occurring at the catalyst interconnected pores. The high quantity of adsorbed nitrogen in the calcined LF slag compared to 10 wt. % NaOH impregnated LF catalyst (y-axis) correlates with the observed larger pore volume of calcined LF slag compared with that of 10 wt.% NaOH-LF pore volume. 3.2.3. TPD-CO2 analysis The performance of catalytic materials used in the glycerol transesterification reaction mainly depends on the textural properties, catalyst surface basicity, and basic strength distribution of the material [31]. Thus, TPD-CO2 was conducted to investigate the LF slag and NaOH impregnated LF slag (5–15 wt.%) basic sites concentration strength which correlate with the temperature of CO2 desorption for each catalyst sample [34]. Fig. 2 shows the TPD-CO2 of the catalyst samples. The desorption peaks at <200 °C, 300 °C to 450 °C, and >450 °C correspond to weak, moderate, and strong basic sites, respectively [37]. The CO2 desorption for the calcined LF slag (800 °C) around 550–670 °C, displayed strong peak intensity with basic sites concentration of 0.16 mmol/g. LF slag modification with NaOH loadings from 5 to 15 wt.% required a higher temperature (590–700 °C) for CO2 desorption, indicating the presence of stronger basic sites. Also, a shoulder peak is observed for all NaOH impregnated LF and five times reused 10 wt.% NaOH-LF catalyst around 780– 900 °C. The intensity of the shoulder peaks consequently increased
3
SBET (m2/g)
(a)
70 60 50 40 30 20 10 0.0
0.2
0.4
0.6
0.8
1.0
P/P 0 (Relative pressure) 70
Quantity adsorbed (cm3 /g, STP)
Catalyst
Quantity adsorbed (cm /g, STP)
Table 3 Textural properties of LF slag and 10 wt.% NaOH-LF (800 °C) catalyst.
60
(b)
Adsorption Desorption
50 40 30 20 10 0 0.0
0.2
0.4
0.6
0.8
1.0
P/P0 (Relative pressure) Fig. 1. N2 physisorption isotherms of calcined LF (a) and NaOH-LF (10 wt.%, 800 °C) (b) catalysts.
with increasing NaOH loading (5–15 wt.%), with a corresponding increase in the total basic sites concentration as shown in Table 4. The five times reused 10 wt.% NaOH-LF catalyst has desorption
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
(e)
TCD Signal (a.u.)
(d) (c) (b)
(a)
100
200
300
400
500
600
700
800
900
Temperature (oC) Fig. 2. TPD-CO2 of calcined LF slag (800 °C) (a), 5 wt.% NaOH-LF (800 °C) (b), 10 wt.% NaOH-LF (800 °C) (c), 15 wt.% NaOH-LF (800 °C) (d), and five times reused 10 wt.% NaOH-LF (800 °C) (e).
Table 4 Basic sites concentration and basic strength distribution of catalyst samples. Catalyst
Basic amount (mmol/g)
Basic strength (H_)
LF (800 °C) 5 wt.% NaOH-LF (800 °C) 10 wt.% NaOH-LF (800 °C) 15 wt.% NaOH-LF (800 °C) Five times reused 10 wt.% NaOH-LF (800 °C)
0.16 0.08 0.14 0.15 0.11
9.3 < H_ < 15 15 < H_ < 18.4 15 < H_ < 18.4 18.4 < H_ < 27 15 < H_ < 18.4
temperature around 550–670 °C with a shoulder peak. The amount of the basic sites on the five times reused catalyst is lower than that of 10 wt.% NaOH-LF. The decreased basic sites concentration can be ascribed to leached Na elements, as evidenced in the EDX analysis. The distribution of the basic strength on the LF and NaOH impregnated LF was investigated using Hammett indicators. Evidently, the basic strength of fresh LF (9.3 < H_ < 15) is lower than all NaOH modified LF (5–15 wt.%) samples. This indicates that NaOH enhances the LF slag basic properties for appreciably higher catalytic performance. Pan et al. [14] and Hu et al. [23], reported that basic strength above H_ > 18.4, is likely to initiate GC decomposition to glycidol, which confirms the appearance of glycidol in the 15 wt.% NaOH-LF catalyst (see Table 1). Hence, moderate basic strength (15 < H_ < 18.4) and strong basic sites are crucial for higher glycerol conversion and GC yield without glycidol formation.
3.2.4. Fourier transform infrared, FTIR analysis The infrared spectra of the fresh LF slag, calcined LF slag (800 °C), 10 wt.% NaOH-LF calcined at 800 °C, and five times reused 10 wt.% NaOH-LF catalyst are presented in Fig. 3. The weak spectra band from 870 to 880 cm 1 and broad peaks at approximately 1400 cm 1 are attributed to the carbonate impurities usually generated from CaO carbonation in the air during FTIR spectrum analysis. The IR peaks at 999, 924, and 944 cm 1 are ascribed to the symmetric stretching of SiAO2 [38]. The IR spectra band at 2942.53 cm 1, which was observed in the reused calcined LF and 10 wt.% NaOH-LF catalysts is ascribed to the CAH stretching mode, which is likely from organic matter that are in contact with the catalysts [39]. The broad band at 3340 cm 1 is attributed to symmetric and asymmetric OAH stretching vibrations of water molecules.
5
The observed sharp peak at approximately 3640 cm 1 is assigned to the stretching hydroxyl functional group. Although several studies on Portland cement identified the IR band at 3640 cm 1 as AOH stretching of Ca(OH)2, assigning this IR band is contentious if NaOH is present [25,43,45,46] because Ca(OH)2 and NaOH have similar absorption bands, which can be attributed to their relatively close ionic radii. Ca and Na have strong basic sites and are regarded as strong bases. The possibility of mixed oxide formation may be negligible or the incorporated Na+ may have diffused and dispersed uniformly into the CaO structures, thus hindering the conspicuous identification of the Na2O crystalline phase. Thus, the initial interaction between two hydroxides can be attributed to the incorporation of alkali ions (Na+) into CaO during calcination and hydrogen bond formation [41]. Therefore, NaOH decomposed into the CaO matrix as Na2O during calcination, increasing the basicity of the LF. The reused fresh LF calcined at 800 °C exhibited a crystalline phase distortion because some of the IR spectra peaks patterns flatten out, which explain the poor catalytic activity observed at the fifth reuse of the unmodified LF. Incorporating NaOH into the LF resulted in a crystalline IR pattern similar to fresh LF sample. The sustained five cycles of reuse of 10 wt.% NaOH-LF catalyst exhibited a similar spectra pattern to that of the unused 10 wt.% NaOH-LF calcined at 800 °C. 3.2.5. XRD of catalyst samples The XRD pattern of the unmodified LF, 10 wt.% NaOH-LF calcined at 800 °C, five times reused 10 wt.% NaOH-LF and regenerated five times reused 10 wt.% NaOH-LF catalyst re-calcined at 800 °C are shown in Fig. 4. The XRD pattern showed the phases of CaO, portlandite (Ca(OH)2), calcium silicate, and calcium carbonate. The crystalline phase of the steel slag is mainly composed of portlandite, which is consistent with the results of previous studies. The characteristic peaks at 2h of 18.12°, 28.63°, 34.14°, and 50.82° were identified as portlandite (Ca(OH)2) [43,46,47]. The peaks at 2h of 32.14°, 37.47°, 54.57°, 62.58°, and 64.92° were ascribed to CaO. Calcium silicate (wollastonite) was observed for all the samples at 2h of 43.59° [29]. The calcium carbonate (CaCO3) peak observed at 2h of 39.81° is possibly from CaO carbonation during XRD analysis or storage under ambient conditions. The catalyst that was reused five times showed an irregular crystalline structure with a more intense peak of portlandite at 2h of 29.03° compared with that of the unused 10 wt.% NaOH-LF calcined at 800 °C catalyst. A reduction in the peak intensity of CaO occurred, indicating that organic materials from the reactants may have masked or stuck to the catalyst surface [42]. This observation corresponds to and explains the emergence of the CAH band on the FTIR spectra for the catalyst reused five times. Also, the conducted EDX confirms the presence of contacting organic matter evidenced by increased carbon content (see Table 2). However, when the reused catalyst was subjected to regeneration step by calcination at 800 °C, the crystalline structure was restored with a slight decrease in peak intensity. NaOH peaks were not observed in the XRD crystalline pattern of all the samples, thus confirming the transformation to Na2O that is uniformly dispersed in the slag lattice. This result is consistent with that of a reported study [41]. 3.3. Effect of reaction influencing parameters on glycerol transesterification with DMC-catalyzed by 10 wt.% NaOH-LF calcined at 800 °C The glycerol conversion and GC yield were probed at different reaction temperatures ranging from 30 to 80 °C under the DMCto-glycerol molar ratios of 1:1 and 2:1 (Fig. 5a). GC yield and glycerol conversion increase with temperature from 30 to 80 °C for the 1:1 and 2:1 DMC-to-glycerol molar ratios. This result explains the reaction rate dependence on temperature. Thus, the molecular
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
6
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
(d)
-OH
2942.53 -CO3 2-
Transmittance (%)
1427.39
(c) 941.30
877.65
(b) 1478.50
(a)
874.76 3642.73 1431.24
4000
3500
3000
2500
2000
1500
1000
500
-1
Wavenumber (cm ) Fig. 3. FTIR of calcined LF (800 °C) (a), five times reused LF slag (800 °C) (b), 10 wt.% NaOH-LF (800 °C) (c), and five times reused 10 wt.% NaOH-LF (800 °C) (d).
Ca(OH) 2 , CaCO3 ,
*
(d)
Intensity (a.u.)
*
CaO
CaSiO 3
*
(C)
(b)
(a)
10
20
30
*
40 50 60 2
70
80
90
Fig. 4. XRD pattern of calcined LF (800 °C) (a), 10 wt.% NaOH-LF (800 °C) (b), five times reused 10 wt.% NaOH-LF (800 °C) (c), and regenerated five times reused 10 wt.% NaOH-LF (re-calcined at 800 °C) (d).
Glycerol conv. & GC yield (%)
100
80
Glycerol conv. (%) (@1:2) GC yield (%) (@1:2) Glycerol conv. (%) (@1:1) GC yield (@1:1)
60
40
20
0 30
45
60
70 75 Reaction temperature (oC)
80
Fig. 5a. Glycerol conversion and GC yield at the DMC-to-glycerol molar ratios of 1:1 and 2:1 and the temperature range of 30–80 °C. 3 wt.% catalyst dosage, 90 min, and 500 rpm.
speed of the reactants (glycerol and DMC) relative to one another increases with reaction temperature, leading to collisions and bond cleavages [43]. Meanwhile, the glycerol conversion and GC yield achieved equilibrium (because the product concentration does
not change as temperature increases) from 75 to 80 °C for the 2:1 DMC-to-glycerol molar ratio. The equimolar ratio of DMC-toglycerol increased steadily as temperature increases. However, the 1:1 DMC-to-glycerol molar ratio did not achieve full glycerol conversion (56% glycerol conversion and 51% GC yield) under the reaction conditions. A previous study on CaO-catalyzed glycerol transesterification revealed that the DMC-to-glycerol molar concentration of 1 results in the serious dissolution of Ca species in glycerol exhibiting homogeneous catalysis [18]. The solubility of Ca species in the polar glycerol decreased significantly when the DMC amount was increased. A consequent increase in glycerol conversion and yield was observed above the equimolar DMC-toglycerol molar ratio. Moreover, the hydrophilic glycerol and hydrophobic DMC creates a biphasic system that seriously limits miscibility. This could possibly result in preferential strong adsorption of glycerol on the 10 wt.% NaOH-LF catalyst active sites, because of glycerol polarity, leading to lower GC yield and glycerol conversion as shown in DMC/glycerol ratio of 1. Hence, increasing contact between DMC and glycerol and surmounting the biphasic system that drives the equilibrium reaction, excess DMC is utilized. The excess DMC above stoichiometric value acts as a solvent and reagent to promote the catalytic reaction. As a reagent and solvent (b.p. 90 °C), excess DMC positively shifts the equilibrium to promote higher glycerol conversion and can easily be separated from reaction products via distillation, respectively. Thus, increased DMC concentration limits the interaction of glycerol with portlandite (Ca(OH)2), thus preventing soluble Ca species formation. Under the reaction conditions of 2:1 DMC-to-glycerol molar ratio, 3 wt.% catalyst dosage, stirring speed of 500 rpm, and 90 min reaction time, the conversion and yield of glycerol and GC were approximately 99% and 97%, respectively, at 75 °C. Therefore, 75 °C and 2:1 DMC-to-glycerol molar ratio were selected as the best temperature and molar ratio for further reaction influencing parameter screening, respectively. Active materials containing functional sites speed up the chemical reaction for non-spontaneous chemical reactions by reducing the activation energy barrier. Thus, 10 wt.% NaOH-LF catalytic activity in glycerol transesterification reaction with DMC was investigated utilizing different amounts of catalyst dosage (1– 4 wt.%) relative to glycerol weight, as presented in Fig. 5b. The conversion and yield of glycerol and GC increased as the amount of catalyst increases from 1 to 4 wt.%. This observation can be attributed to the increase in available active basic sites necessary to
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
7
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
100
stable until 120 min. The reaction achieved equilibrium at 90 min because no changes in product concentration were observed.
GC yield (%)
Glycerol conv. & GC yield (%)
Glycerol conv. (%) 80
3.4. Water content test
60
40
20
0
1
2 3 Catalyst dosage (wt.%)
4
Fig. 5b. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 1–4 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
deprotonate glycerol and the subsequent reaction with DMC leading to GC formation. The 3 wt.% catalyst dose relative to glycerol weight achieved approximately 98% glycerol conversion and 97% GC yield at 90 min reaction time, 75 °C, and 2:1 DMC-to-glycerol molar ratio. Catalyst loading above a certain value may limit the mass transfer from the bulk to the catalyst active sites and induce particle agglomeration, thus hindering further glycerol conversion [36]. At >3 wt.% catalyst dosage, no significant changes (saturation of active basic sites) in glycerol conversion and GC yield were observed, indicating that NaOH-LF is stable against particle abrasion, which is normally experienced in slurry mixtures. The effect of reaction time on glycerol conversion and GC yield was investigated under reaction conditions of 75 °C, 500 rpm stirring speed, 3 wt.% catalyst dosage, 2:1 DMC-to-glycerol molar ratio, and 30–120 min. Fig. 5c shows that GC yield and glycerol conversion increases as the reaction time increases from 30 to 120 min. The 30 min reaction time resulted in glycerol conversion and GC yield of approximately 65%. The low glycerol conversion is possibly attributed to carbonate group proximity to the third glycerol OH group. The reactivity of the first OH group of glycerol proceeds faster than the third OH group and the intramolecular reaction regime for the second OH is faster than the intramolecular reaction of the third OH group with the next DMC molecule forming a thermodynamically stable cyclic carbonate [28]. Thus, more reaction time is required to achieve reaction equilibrium. Therefore, at 90 min reaction time, the glycerol conversion and GC yield were approximately 98% and 96%, respectively, and remained
10 wt.% NaOH-LF catalyst stability against leaching of active sites and reusability were investigated. Fig. 7 indicates that the catalyst could still sustain >90% glycerol conversion and 80% GC yield. However, the opposite result is the case for 800 °C calcined LF sample where the sample that was reused five times showed only 19% and 18% glycerol conversion and GC yield, respectively. The significant decrease in the catalytic performance of the LF slag is possibly attributed to the serious leaching of the active sites (CaO). The reduction in the performance of 10 wt.% NaOH-LF catalyst could be ascribed partly to the leached constituents of the materials and adhesion of the organic matters in the catalyst pores, as evidenced by the FTIR, EDX and XRD results. Although deactivation of the catalyst is not rapid as shown in the reusability studies, however, the slightly leached Na or CaO poses separation difficulties which increases overall energy input of the transesterification process. Nevertheless, the Na metals can be separated from the reaction mixture by neutralization step utilizing sulfuric acid, hydrochloric or phosphoric acid [44]. Utilizing phosphoric acid to remove leached NaOH via crystallization/precipitation is a considerable route because, depending on the concentration of dissolved Na, the pre-
GC yield (%) Glycerol conv. (%)
80
60
40
20
0
3.5. Leaching and reusability tests of calcined LF slag (800 °C) and 10 wt.% NaOH-LF calcined at 800 °C
Glycerol conv. & GC yield (%)
Glycerol conv. & GC yield (%)
100
Biodiesel-derived glycerol contains water. Thus, synthesizing water-resistant catalytic material is economical because refining glycerol is energy-intensive. Developing materials that exhibit similar catalytic activity with crude and refined glycerol is desirable for commercialization. Therefore, the influence of water content ranging from 0.75 wt.% to 5 wt.% relative to glycerol weight was investigated under reaction conditions of 75 °C, 2:1 DMC-toglycerol molar ratio, 3 wt.% catalyst dosage, 500 rpm stirring speed, and 90 min reaction time. Fig. 6 shows that GC yield and glycerol conversion decrease as water content increases, which is likely attributed to the selective strong adsorption of water molecules in the active sites of 10 wt.% NaOH-LF calcined at 800 °C catalyst or leaching of the active sites (CaO or Na2O) since they are highly unstable under moisture. Thus, the number of available sites for glycerol deprotonation is seriously reduced. However, the conversion and yield of glycerol and GC could still be higher than 50% at 5 wt.% water content, which indicates reasonable water resistance by the 10 wt.% NaOH-LF catalyst.
30
60 90 Reaction time (min)
120
Fig. 5c. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 30– 120 min; stirring speed = 500 rpm.
100 80 60 40 20 0
GC yield (%) Glycerol conv. (%) 0.75
1 3 Water content (wt.%)
5
Fig. 6. Effect of water content on glycerol conversion and GC yield. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
8
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx
Glycerol conv. (%)-10wt% NaOH-LF
GC yield (%)-10wt% NaOH-LF
Glycerol conv. (%)-LF(800)
GC yield (%)-LF(800)
Glycerol conv. & GC yield (%)
100 80
the reaction conditions utilizing 10 wt.% NaOH-LF calcined at 800 °C achieved 99% glycerol conversion and 96.59% GC yield under 90 min reaction time. NaOH-LF exhibited a reasonable resistance to in situ water in glycerol (50% glycerol conversion at 5 wt.% glycerol water content), however, deactivation is inevitable due to leaching of active sites. Thus, 10 wt.% NaOH-LF was reused five times without showing serious deactivation.
60
Acknowledgment
40 20 0
0
1
2 3 No. of reuse cycles
4
5
Fig. 7. Glycerol conversion and GC yield trend. DMC-to-glycerol molar ratio = 2:1; reaction temperature = 75 °C; catalyst dosage = 3 wt.%; reaction time = 90 min; stirring speed = 500 rpm.
cipitated disodium or trisodium phosphates salts are almost insoluble in an organic solvent [45,46]. Moreover, the trisodium phosphate has been successfully applied as an efficient catalyst in base-catalyzed BD production [45]. EDX (see Table 2) of the five times reused catalyst sample, revealed about 11.02 wt.% increase in carbon element of the catalyst, suggesting possible masking of catalyst active sites by contacting organic matters, whereas, the Na element reduction about 2.62 wt.% confirms leaching of basic active site necessary for the transesterification reaction. Thus, plausible deactivation mechanisms can be proposed for 10 wt.% NaOH-LF calcined at 800 °C. Portlandite (calcium hydroxide), which is the bulk of LF, comes in contact with glycerol to form calcium diglyceroxide and a small amount of water. On the other hand, a small amount of glycerol is subjected to etherification to produce a small amount of water under a strong alkali hydroxide concentration. In the presence of in situ water in the glycerol transesterification process, GC and/or excess DMC likely undergo hydrolysis, producing CO2 and glycerol, as reported in [47]. Water may not be generated or produced in significant amounts to initiate serious or rapid deactivation of the catalytic material because of the mild reaction conditions used in this study. However, the interaction of the generated CO2 and portlandite impelled or promoted the formation of amorphous calcium carbonate layer on the catalyst surface. To a certain extent, this layer of CaCO3 inhibits the attrition of CaO (Na2O) and portlandite from the inner catalyst layer [23]. The potential mass transport of the reactant molecules into the loose CaCO3 outer layer structure is possible (or the formed CaCO3 outer layer is leached completely by the reactants during the reaction) into the inner catalyst layer, thus allowing sufficient glycerol contact with the catalyst active sites. After regeneration at 800 °C, the five times reused 10 wt.% NaOH-LF catalyst activity was restored with 96% glycerol conversion and 94% GC yield.
4. Conclusions LF slag can be utilized to synthesize GC in the transesterification reaction. However, the serious deactivation of active sites of the catalyst resulted in poor GC yield after five cycles of reuse. LF slag that impregnated with NaOH exhibited a remarkable GC yield because of the increased basicity and basic strength compared with that of calcined LF (800 °C). The calcination temperature modified the transesterification reaction to result in an evidently high GC yield because weak basic sites are converted to strong basic sites as calcination temperature increases. The unique severity of
The first author acknowledges the financial support provided by Universiti Sains Malaysia (USM), under the USM Fellowship Scheme. The authors acknowledge the financial support provided by Ministry of Education, Malaysia under the Transdisciplinary Research Grant Scheme (TRGS) Phase 2/2014 (203/ PJKIMIA/6762002). References [1] da Silva GP, Mack M, Contiero J. Glycerol: a promising and abundant carbon source for industrial microbiology. Biotechnol Adv 2009;27:30–9. [2] Yang F, Hanna Ma, Sun R. Value-added uses for crude glycerol–a byproduct of biodiesel production. Biotechnol Biofuels 2012;5:13. [3] Remon J, Gimenez JR, Valiente A, Garcia L, Arauzo J. Production of gaseous and liquid chemicals by aqueous phase reforming of crude glycerol: influence of operating conditions on the process. Energy Convers Manage 2016;110:90–112. [4] Teng WK, Ngoh GC, Yusoff R, Aroua MK. A review on the performance of glycerol carbonate production via catalytic transesterification: effects of influencing parameters. Energy Convers Manage 2014;88:484–97. [5] Dou B, Wang C, Song Y, Chen H, Xu Y. Activity of Ni-Cu-Al based catalyst for renewable hydrogen production from steam reforming of glycerol. Energy Convers Manage 2014;78:253–9. [6] Galera S, Gutiérrez Ortiz FJ. Life cycle assessment of hydrogen and power production by supercritical water reforming of glycerol. Energy Convers Manage 2015;96:637–45. [7] Zakaria ZY, Amin NAS, Linnekoski J. Optimization of catalytic glycerol steam reforming to light olefins using Cu/ZSM-5 catalyst. Energy Convers Manage 2014;86:735–44. [8] Veiga S, Bussi J. Steam reforming of crude glycerol over nickel supported on activated carbon. Energy Convers Manage 2016. http://dx.doi.org/10.1016/j. enconman.2016.04.103 (in press). [9] Ishak ZI, Sairi NA, Alias Y, Aroua MKT, Yusoff R. Production of glycerol carbonate from glycerol with aid of ionic liquid as catalyst. Chem Eng J 2016;297:128–38. [10] Liu Z, Wang J, Kang M, Yin N, Wang X, Tan Y, et al. Structure-activity correlations of LiNO3/Mg4AlO5.5 catalysts for glycerol carbonate synthesis from glycerol and dimethyl carbonate. J Ind Eng Chem 2015;21:394–9. [11] Esteban J, Fuente E, Blanco A, Ladero M, Garcia-Ochoa F. Phenomenological kinetic model of the synthesis of glycerol carbonate assisted by focused beam reflectance measurements. Chem Eng J 2015;260:434–43. [12] Ochoa-Gómez JR, Gómez-Jiménez-Aberasturi O, Ramírez-López C, Belsué M. A brief review on industrial alternatives for the manufacturing of glycerol carbonate, a green chemical. Org Process Res Dev 2012;16:389–99. [13] Malyaadri M, Jagadeeswaraiah K, Sai Prasad PS, Lingaiah N. Synthesis of glycerol carbonate by transesterification of glycerol with dimethyl carbonate over Mg/Al/Zr catalysts. Appl Catal A Gen 2011;401:153–7. [14] Pan S, Zheng L, Nie R, Xia S, Chen P, Hou Z. Transesterification of glycerol with dimethyl carbonate to glycerol carbonate over Na–based zeolites. Chinese J Catal 2012;33:1772–7. [15] Okoye PU, Hameed BH. Review on recent progress in catalytic carboxylation and acetylation of glycerol as a byproduct of biodiesel production. Renew Sustain Energy Rev 2016;53:558–74. [16] Zheng L, Xia S, Hou Z, Zhang M, Hou Z. Transesterification of glycerol with dimethyl carbonate over Mg-Al hydrotalcites. Chinese J Catal 2014;35:310–8. [17] Zheng L, Xia S, Lu X, Hou Z. Transesterification of glycerol with dimethyl carbonate over calcined Ca-Al hydrocalumite. Cuihua Xuebao/Chinese J Catal 2015;36:1759–65. [18] Simanjuntak FSH, Kim TK, Lee SD, Ahn BS, Kim HS, Lee H. CaO-catalyzed synthesis of glycerol carbonate from glycerol and dimethyl carbonate: isolation and characterization of an active Ca species. Appl Catal A Gen 2011;401:220–5. [19] Bai R, Wang S, Mei F, Li T, Li G. Synthesis of glycerol carbonate from glycerol and dimethyl carbonate catalyzed by KF modified hydroxyapatite. J Ind Eng Chem 2011;17:777–81. [20] Algoufi YT, Hameed BH. Synthesis of glycerol carbonate by transesterification of glycerol with dimethyl carbonate over K-zeolite derived from coal fly ash. Fuel Process Technol 2014;126:5–11. [21] Worldsteel Association. April 2016 crude steel production. Brussels, Belgium; 2016.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067
P.U. Okoye et al. / Energy Conversion and Management xxx (2016) xxx–xxx [22] Lewis DW. Properties and uses of iron and steel slags. In: Nationa Slag Association, editor. Symp slag natl inst transp road res South Africa, South Africa. p. 1–11. [23] Hu K, Wang H, Liu Y, Yang C. KNO3/CaO as cost-effective heterogeneous catalyst for the synthesis of glycerol carbonate from glycerol and dimethyl carbonate. J Ind Eng Chem 2014;28:334–43. [24] Khayoon MS, Hameed BH. Mg1+xCa1 xO2 as reusable and efficient heterogeneous catalyst for the synthesis of glycerol carbonate via the transesterification of glycerol with dimethyl carbonate. Appl Catal A Gen 2013;466:272–81. [25] Duchesne J, Reardon EJ. Measurement and prediction of portlandite solubility in alkali solutions. Cem Concr Res 1995;25:1043–53. [26] Pallagi A, Tasi Á, Gácsi A, Csáti M, Pálinkó I, Peintler G, et al. The solubility of Ca (OH)2 in extremely concentrated NaOH solutions at 25 °C. Cent Eur J Chem 2012;10:332–7. [27] Simanjuntak FSH, Lim SR, Ahn BS, Kim HS, Lee H. Surfactant-assisted synthesis of MgO: characterization and catalytic activity on the transesterification of dimethyl carbonate with glycerol. Appl Catal A Gen 2014;484:33–8. [28] Yadav GD, Chandan PA. A green process for glycerol valorization to glycerol carbonate over heterogeneous hydrotalcite catalyst. Catal Today 2014;237:47–53. [29] Setién J, Hernández D, González JJ. Characterization of ladle furnace basic slag for use as a construction material. Constr Build Mater 2009;23:1788–94. [30] Shi C. Characteristics and cementitious properties of ladle slag fines from steel production. Cem Concr Res 2002;32:459–62. [31] Ochoa-Gómez JR, Gómez-Jiménez-Aberasturi O, Maestro-Madurga B, Pesquera-Rodríguez A, Ramírez-López C, Lorenzo-Ibarreta L, et al. Synthesis of glycerol carbonate from glycerol and dimethyl carbonate by transesterification: catalyst screening and reaction optimization. Appl Catal A Gen 2009;366:315–24. [32] Bkour Q, Faqir N, Shawabkeh R, Ul-Hamid A, Bart HJ. Synthesis of a Ca/Naaluminosilicate from kaolin and limestone and its use for adsorption of CO2. J Environ Chem Eng 2016;4:973–83. [33] Wang X, Zheng S, Xu H, Zhang Y. Leaching of niobium and tantalum from a low-grade ore using a KOH roast-water leach system. Hydrometallurgy 2009;98:219–23.
9
[34] Alvarez MG, Segarra AM, Contreras S, Sueiras JE, Medina F, Figueras F. Enhanced use of renewable resources: transesterification of glycerol catalyzed by hydrotalcite-like compounds. Chem Eng J 2010;161:340–5. [35] Benhouria A, Islam MA, Zaghouane-Boudiaf H, Boutahala M, Hameed BH. Calcium alginate–bentonite–activated carbon composite beads as highly effective adsorbent for methylene blue. Chem Eng J 2015;270:621–30. [36] Liu P, Derchi M, Hensen EJM. Promotional effect of transition metal doping on the basicity and activity of calcined hydrotalcite catalysts for glycerol carbonate synthesis. Appl Catal B Environ 2014;144:135–43. [37] Kumar P, With P, Srivastava VC, Gläser R, Mishra IM. Glycerol carbonate synthesis by hierarchically structured catalysts: catalytic activity and characterization. Ind Eng Chem Res 2015;54:12543–52. [38] Trezza MA. Hydration study of ordinary portland cement in the presence of zinc ions. Mater Res 2007;10:331–4. [39] Chen G, Zhang B, Zhao J. Dispersion process and effect of oleic acid on properties of cellulose sulfate-oleic acid composite film. Materials (Basel) 2015;8:2346–60. [41] Maiti GC, Baerns M. Dehydration of sodium hydroxide and lithium hydroxide dispersed over calcium oxide catalysts for the oxidative coupling of methane. Appl Catal A Gen 1995;127:219–32. [42] Syamsuddin Y, Hameed BH. Synthesis of glycerol free-fatty acid methyl esters from Jatropha oil over Ca-La mixed-oxide catalyst. J Taiwan Inst Chem Eng 2016;58:181–8. [43] Li J, Wang T. Chemical equilibrium of glycerol carbonate synthesis from glycerol. J Chem Thermodyn 2011;43:731–6. [44] Chiu CW, Dasari MA, Sutterlin WR, Suppes GJ. Removal of residual catalyst from simulated biodiesel’s crude glycerol for glycerol hydrogenolysis to propylene glycol. Ind Eng Chem Res 2006;45:791–5. [45] Thinnakorn K, Tscheikuna J. Biodiesel production via transesterification of palm olein using sodium phosphate as a heterogeneous catalyst. Appl Catal A Gen 2014;476:26–33. [46] Jiang ST, Zhang FJ, Pan LJ. Sodium phosphate as a solid catalyst for biodiesel preparation. Braz J Chem Eng 2010;27:137–44. [47] Li J, Wang T. On the deactivation of alkali solid catalysts for the synthesis of glycerol carbonate from glycerol and dimethyl carbonate. React Kinet Mech Catal 2011;102:113–26.
Please cite this article in press as: Okoye PU et al. Stabilized ladle furnace steel slag for glycerol carbonate synthesis via glycerol transesterification reaction with dimethyl carbonate. Energy Convers Manage (2016), http://dx.doi.org/10.1016/j.enconman.2016.10.067