- Email: [email protected]
Etherification of glycerol with C4 and C5 reactive olefins
FUPROC-04612; No of Pages 8 Fuel Processing Technology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Etherification of glycerol with C4 and C5 reactive olefins Burcin Ikizer a, Nuray Oktar b,⁎, Timur Dogu a a b
Department of Chemical Engineering, Middle East Technical University, 06800 Ankara, Turkey Department of Chemical Engineering, Gazi University, 06430 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 29 April 2015 Received in revised form 17 June 2015 Accepted 24 June 2015 Available online xxxx Keywords: Glycerol Etherification i-Amylene i-Butene Amberlyst-36 Dowex DR-2030
a b s t r a c t Etherification of glycerol with C4 (i-butene) and C5 (i-amylenes: 2-Methyl-2-Butene (2M2B) and 2-Methyl-1Butene (2M1B) mixture) i-olefins was investigated for the production of transportation fuel oxygenates. Results proved formation of mono-, di- and tri-ethers of glycerol, as a result of its etherification with i-amylene at 120–140 °C over Amberlyst-36. Glycerol fractional conversion values and di-ether selectivity values approaching to 1.0 and 0.7 respectively were highly promising. Glycerol etherification results obtained by using i-butene as the reactant, proved the importance of Brønsted acidity of the catalyst, as well as the pore diffusion resistance, on the catalytic performance. Silicotungstic acid (STA), with very high acidity, showed very good performance at T ≤ 80 °C. However, at higher temperatures it also facilitated the oligomerization of i-butene, which caused a negative effect on glycerol etherification. Both Amberlyst-36 and Dowex DR-2030 showed excellent performance in catalyzing glycerol with i-butene, at 90 °C. Results proved that etherification of the biodiesel by-product glycerol could be successfully achieved by using i-amylenes, as well as i-butene, to improve the economics of biodiesel production. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Rapid depletion of petroleum reserves opened new avenues of research for the production of alternative transportation fuels, from non-petroleum feedstock [1,2]. Biodiesel, which is a renewable and environmentally friendly alternative to petroleum-based diesel fuel, is produced by trans-esterification of vegetable oils and animal fats with alcohols (methanol/ethanol), in the presence of a basic catalyst [3,4]. Due to its biodegradable nature, quite high cetane number, higher flash point and lower emission of pollutants than petroleum-based diesel fuel, biodiesel attracted significant attention of fuel producers in recent decades [5,6]. However, large amount of glycerol is also produced as a side product during trans-esterification of oils with alcohols. Glycerol (1, 2, 3-propanetriol) is the main side product of biodiesel production process and one mole of glycerol is produced for every three moles of methyl esters. This corresponds to about 10 wt.% of the biodiesel product. Production rate of biodiesel is expected to increase significantly in the coming years. Hence, a remarkable surplus of glycerol will be generated as a result of increase of biodiesel production. Economics of biodiesel production strongly depends upon the efficient utilization of glycerol, which is produced as a side product during transesterification [7–9]. Some of the possible alternative ways for glycerol utilization are; acrolein production by its dehydration, hydrogen and synthesis gas
⁎ Corresponding author. E-mail address: [email protected] (N. Oktar).
production by reforming, selective oxidation for dihydroxyacetone, production of 1,3-propanediol, and its esterification/etherification to produce fuel additives and/or valuable chemicals [8,9]. Among these alternatives, its etherification with i-butene or tert-butyl alcohol has been considered as attractive routes to produce fuel oxygenates [3, 5–9,11–16]. Etherification of glycerol with i-butene or tert-butyl alcohol would produce two mono-tert-butyl glycerol ether isomers (3-tertbutoxy-1,2-propanediol (MTBG1) and 2-tert-butoxy-1,3-propanediol (MTBG2)), two di-tert-butyl glycerol ether isomers (2,3-di-tertbutoxy-1-propanol (DTBG1) and 1,3-di-tert-butoxy-2-propanol (DTBG2)) and a tri-tert-butyl-glycerol ether (1,2,3-tri-tert-butoxypropane (TTBG)) [5,8,15,16]. These ethers were reported to have high octane numbers and can be used as gasoline additives for octane enhancement. They can also be mixed with biodiesel to increase its amount and to improve the economics of biodiesel production. Solubility of monoethers (MTBG) in diesel fuel is quite low. On the other hand, di-ethers (DTBG) and tri-ethers (TTBG) are highly soluble in diesel and gasoline. Hence, especially di-ethers and tri-ethers were considered as excellent fuel components which can be added to gasoline, instead of current commercial oxygenate additives, such as methyl tert-butyl ether (MTBE), ethyl tert-butyl ether (ETBE) and tert-amyl methyl ether (TAME). In the case of blending of these ethers to diesel fuel, important positive effects were reported in terms of reduction of particulate matter, carbon monoxide and hydrocarbon emissions of the fuels. Furthermore, these ethers can lower the cloud point and viscosity of the fuels [5,6,16]. Etherification of glycerol with i-butene has been investigated during the last decade by some researchers, using different solid acid
http://dx.doi.org/10.1016/j.fuproc.2015.06.039 0378-3820/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
2
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
catalysts [12–16]. Klepacova et al. [16] reported some results for the etherification of glycerol using Amberlyst-15, Amberlyst-35, H-Y, HBeta zeolites and homogeneous p-toluenesulfonic acid as the catalysts. Quite high di-and tri-ether selectivity values were achieved over ion exchange resin catalyst Amberlyst-35. Karinen and Krause [15] also investigated the etherification of glycerol with i-butene in the presence of Amberlyst-35, in a batch reactor. Optimal selectivity towards di-ethers and nearly complete glycerol conversion were obtained with optimization of reaction conditions. Etherification of glycerol with i-butene over two sulfonic acid modified mesostructured silicas and commercial HY zeolite were also reported in the literature [5,6]. Compared to commercial HY, nitric acid and citric acid treated catalysts gave an increased glycerol conversion and higher selectivities to DTBG and TTBG [6]. All of these studies were performed in batch reactors at quite long reaction times. Solubility of glycerol in i-butene is quite low, and hence mass transfer resistances play a significant role on the observed rate of this etherification process. Use of a solvent has been recommended to enhance the rate of etherification of glycerol with i-butene. Klepacova et al. [16] tested different solvents (dioxane, dimethyl sulfoxide and sulfolane) to eliminate or reduce mass transfer effects. Maximum solubility of i-butene was obtained in dioxane. More recently, etherification of glycerol was investigated in our research group, using tert-butyl alcohol, instead of i-butene [7–9]. Glycerol is highly soluble in tert-butyl alcohol and hence higher etherification rates were achieved. In these studies, catalytic performances of different solid acid catalysts were tested in batch and flow reactors. Results proved the advantages of flow reactors in terms of achieving very high conversions at quite low residence times. Amberlyst-15 was shown to give the highest activity in the temperature range of 90–100 °C. However, its stability was low at higher temperatures. At higher temperatures (110–130 °C), best catalytic performance was achieved by using Amberlyst-36. Results reported in these studies also indicated the negative effect of water on the catalyst performance. Water adsorbed on the active acid sites reversibly deactivated the catalyst [7–9]. Importance of Brønsted acidity of the catalyst used for this reaction is also discussed in a recent work of Gonzales et al. [13]. All of these reported studies on the etherification of glycerol were performed either with i-butene or tert-butyl alcohol. Another possible route may be etherification of glycerol with C5 i-olefins (i-amylenes: 2-Methyl-1-Butene (2M1B) or 2-Methyl-2-Butene (2M2B)). Nearly 10% of C5 reactive olefins are present in FCC gasoline. There are number of studies in the literature for the etherification of these olefins with methanol or ethanol to produce some octane enhancing oxygenates, such as tert-amyl methyl ether (TAME) and tert-amyl ethyl ether (TAEE) [17–20]. Comparison of etherification rates 2M2B and 2M1B with ethanol was illustrated in an earlier publication of our research group, indicating that 2M2B, with higher alkyl groups attached to the carbon atoms of double bond, is more stable and somewhat less reactive than 2M1B [17]. Occurrence of isomerization reaction between 2M2B and 2M1B is also expected to take place in the presence of solid acid catalysts. Etherification of glycerol with C5 i-olefins has the potential to contribute positively to the economics of biodiesel production. Such a process would also help to decrease the harmful effects of this olefin on environment and human health. There are only a few publications in the literature for the etherification of glycerol with i-amylenes. Izquierdo et al. [21,22] reported some results for etherification of glycerol with i-amylenes in the presence of various kinds of organic solvents and also using several macromolecular resin type catalysts. Their results proved the importance of pore structure of the resin, as well as the solubility of glycerol in the solvent used during the etherification process, on its conversion. Miscibility of both reactants in the solvent was found to be the main factor affecting the progress of this reaction. The highest glycerol conversion was obtained in the presence of tertiary amyl alcohol as the solvent, due to its good solvent polarity. In the present research, etherification of glycerol was investigated with both i-amylenes and i-butene. Catalytic performances of some
resin type solid acid catalysts (Amberlyst-36, Dowex DR-2030), as well as silicotungstic acid were tested and the results were compared.
2. Experimental 2.1. Catalysts and chemicals Chemicals used in this study were i-butene (≥ 98% Air Products), Glycerol (≥99.5% Merck), 1,4 Dioxane (≥99%, Merck), i-amylene mixture [(2-Methyl-2-Butene (~ 85%) and 2-Methyl-1-Butene (~ 15%), Merck)] and Ethanol (≥99.5%, Merck). Two ion exchange resin catalysts, namely Amberlyst-36 (Sigma Aldrich) and Dowex DR-2030 (The Dow Chemical Company) and silicotungstic acid (STA) (H4SiW12O40) were used as the solid acid catalysts. Ion exchange resin catalysts are copolymers in macroreticular structure and functionalized with sulfonic acid groups. Silicotungstic acid (STA) is a Keggin type heteropolyacid with very high proton mobility. One of the reasons of selection of these catalysts for the etherification of glycerol was their quite high Brønsted acidity. Another important factor of this selection was their maximum operating temperatures. Our initial experiments performed with various type ion-exchange resin catalysts (Amberlyst-15, Amberlyst-16 etc.) showed that temperatures higher than 110 °C were needed to achieve reasonable conversion rates in etherification of glycerol with i-amylenes. However, many of the resin type solid acid catalysts are not stable at high temperatures. On the other hand, the maximum operating temperatures of Amberlyst-36, Dowex DR-2030 and silicotungstic acid are 150, 130 and 400 °C, respectively (Table 1). Considering the inhibition effect of adsorbed water on the rate of etherification reactions performed using acidic resins, the catalysts used in our work were dried at 110 °C. As it was discussed by Soto et al. [19] the amount of adsorbed water was less than 3.5 wt.% over the resin type catalysts pretreated at this temperature and inhibition of etherification rate with this much adsorbed water was negligible. Some of the physical properties of these catalysts are given in Table 1.
2.2. Characterization of the catalysts Diffuse Reflectance FT-IR Spectroscopy (DRIFTS) analysis of pyridine adsorbed catalyst samples was performed by using a PerkinElmer Spectrum One instrument. The procedure of Ozbay et al. [9] was followed for this analysis. Before these tests, samples were pretreated at 110 °C to remove water adsorbed on the acid sites. After the adsorption of pyridine, samples were dried at 40 °C, before the DRIFTS analysis. A background spectrum was recorded with KBr. FT-IR spectrum of the sample which was not treated by pyridine was subtracted from the spectrum of pyridine-adsorbed sample for the detection of adsorbed surface species. Relative strengths of Lewis and Brønsted acid sites were then determined from the differences of these spectra.
Table 1 Physical properties of the catalysts used in this study. Property Exchange capacity (meq H+/g) Surface area (m2/g) Porosity (cc/g) Average pore diameter (nm) Particle diameter (cm) Max. operating temperature (°C) a b c
a
Dowex DR-2030 b
Silicotungstic acid (STA) c
5.4 33 0.2 24 0.06–0.085 150
4.7 30 0.35 20 0.043–0.0525 130
– 7.8 0.03 – Powder 400 [23]
Amberlyst-36
Data obtained from Sigma-Aldrich Chemistry. Data obtained from The Dow Chemical Company. Measured by Quantachrome Corp., Autosorb-6 device in Central Laboratory in METU.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
3
2.3. Reaction experiments Etherification reactions of glycerol with i-butene and i-amylene mixture were performed in a stainless steel autoclave batch reactor (75 mL), equipped with a PID temperature controller, a magnetic stirrer and a pressure gauge. Etherification reactions with i-butene were performed in a temperature range of 70–120 °C using Amberlyst-36, Dowex DR-2030 and silicotungstic acid as the solid acid catalysts. In each experiment, 0.055 mol of glycerol, 0.3 g of catalyst, 0.446 mol i-butene and 30 mL 1, 4 dioxane (solvent) were introduced into the reactor. During the reaction runs performed with i-butene, the autogenous pressure in the reactor was recorded as 8–10 bars. Etherification of glycerol with i-amylene mixture was performed using Amberlyst-36 and Dowex DR-2030 under different reaction conditions. In the case of using Amberlyst-36, experiments were also repeated with different amounts of catalyst, namely 0.3 and 1 g. Also, another set of experiments were performed at different reaction times (1, 3, 6, 12, 24 h) with this catalyst. Initial experiments performed with i-amylene mixture indicated that the reaction temperature should be higher than the experiments performed with i-butene to achieve sufficiently high glycerol conversions. Hence, reaction temperatures were selected as 120 and 140 °C. Due to the instability of the resin type catalysts at higher temperatures, maximum temperature was selected as 140 °C. In each test, 0.0326 mol of glycerol and 0.261 mol i-amylene mixture were used. During these reaction runs performed with the i-amylene mixture, the autogenous pressure in the reactor was recorded as 6–8 bars. After the completion of a reaction run, the reactor was cooled to room temperature, until the system pressure reached to atmospheric. In the case of using i-amylene mixture as the reactant, 50 mL of ethanol was added to the reactor immediately after opening it. As a result of ethanol addition, a single phase of glycerol, i-amylene and the reaction products mixture was obtained. This mixture was stirred until it became homogenous. At the end of each run, chemical composition of the product mixture was analyzed by a gas chromatograph (Agilent 6890N), which was equipped with a capillary HP INNOWax column and a flame ionization detector (FID). Flow rate of the carrier gas (He) was 1.4 mL/min and flow rates of dry air and hydrogen gas were 300 mL/min and 30 mL/min, respectively. Detector and the injector temperatures were adjusted as 270 and 295 °C, respectively. In order to be able to observe all of the peaks corresponding to different species, a temperature program was adjusted, starting at 50 °C, followed by a 4 °C/min ramp up to 115 °C and then followed with a 15 °C ramp up to 250 °C, where temperature was kept constant for about 10 min. Chromatographic peaks of 2M2B and 2M1B were very close to each other. Hence, we used summation of these peaks to evaluate the amount of i-amylenes in the mixture. 3. Results and discussion 3.1. DRIFTS analysis of catalysts Diffuse Reflectance Fourier Transform IR Spectroscopy (DRIFTS) analysis of pyridine adsorbed samples was performed to determine the relative strengths of the Brønsted and Lewis acid sites on the surface of the solid acid catalysts used in this study (Fig. 1). Distinctive bands, which appear at 1540 and 1640 cm−1 correspond to the Brønsted acid sites. However, for the Lewis acid sites, bands are expected at 1450 and 1598 cm−1 [9]. According to DRIFTS given in Fig. 1, for Amberlyst36 and Dowex DR-2030 acidic resins, the Lewis bands at 1450 and 1598 cm−1 were quite weak. However, the intensities of the bands at 1540 and 1640 cm−1 were quite strong, indicating a high number of Brønsted acid sites. Sulfonic acid functionality on the catalyst surface for the acidic resins is the cause of Brønsted acidity [9]. Due to the contributions of both Brønsted and Lewis acid sites another band was observed at 1490 cm−1 for Amberlyst-36 and Dowex DR-2030. Because
Fig. 1. DRIFT spectra of pyridine adsorbed solid acid catalysts (Amberlyst-36, Dowex DR2030, silicotungstic acid (STA)).
of low Lewis acidity of these resins, it can be considered that the band at 1490 cm−1 was also mainly due to Brønsted acid sites. Although the hydrogen exchange capacity of Amberlyst-36 was higher than that of Dowex DR-2030 (Table 1), DRIFTS analysis indicated somewhat a higher number of Brønsted acid sites for Dowex DR-2030. This was considered to be due to higher porosity of Dowex DR-2030 than Amberlyst-36, which allowed better penetration of pyridine into the catalyst to be adsorbed on the active sites. The DRIFT spectrum of pyridine adsorbed silicotungstic acid showed the presence of both Lewis and Brønsted acid sites (Fig. 1). However, comparison of the intensities of the bands showed that the Brønsted acidity of silicotungstic acid was much higher than its Lewis acidity. Strong Brønsted acidity is mainly due to high proton mobility of silicotungstic acid. Heteropolyacids are known to have acidities even higher than sulfuric acid and they are considered as super acids [23]. 3.2. Etherification of glycerol with i-butene Etherification of glycerol with i-butene is expected to take place following three consecutive reversible reaction steps over the solid acid catalysts. Reaction steps giving the products, namely mono-tert-butyl glycerol ether isomers (3-tert-butoxy-1,2-propanediol; MTBG1 and 2tert-butoxy-1,3-propanediol; MTBG2), two di-tert-butyl glycerol ether isomers (2,3-di-tert-butoxy-1-propanol; DTBG1 and 1,3-di-tertbutoxy-2-propanol; DTBG2), and a tri-tert-butyl glycerol ether (1,2,3tri-tert-butoxy-propane; TTBG) are shown in Fig. 2. Besides these reactions, oligomerization and hydration of i-butene are two of the possible side reactions, which may take place in the reactor [16]. As a result of oligomerization of i-butene, two isomers of di-i-butene, namely TMP1 (2,4,4-trimethyl-1-pentene) and TMP-2 (2,4,4-trimethyl-2-pentene) are expected to form. Oligomerization of i-butene is not desired and has a negative effect on ether yields. However, as a result of hydration of i-butene tert-butyl alcohol may form, which may also react with glycerol to produce the ethers. Glycerol conversion values obtained as a result of its etherification with i-butene are reported in Fig. 3. Results obtained at different temperatures by using Amberlyst-36, Dowex DR-2030 and silicotungstic acid catalysts, in the presence of 1,4 Dioxan as solvent, are reported in this figure. At 70 and 80 °C, maximum glycerol conversion values were obtained by using silicotungstic acid as the catalyst. Considering that the Brønsted acidity of silicotungstic acid was higher than the other resin type solid acid catalysts, this was an expected result. Heteropolyacids are sometimes considered as pseudo liquids, indicating easy penetration of reactants to the acid sites. In this temperature range, catalytic performance of Dowex DR-2030 was also higher than the performance of Amberlyst-36. Comparison of the activities of the two resin type catalysts was also in concordance with the Brønsted acidity results (Fig. 1). Dowex DR-2030 was more porous than Amberlyst-36, which
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
4
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Fig. 2. Etherification of glycerol with C4 i-olefin (i-butene).
facilitated transport of reactants to the active sites within the catalyst. Macroreticular resins are composed of interconnected gel-like micro grains and macropores between these micrograins. As it was discussed in an earlier publication of our research group, both macropore diffusion and penetration resistances of reactants into the micro-grains were expected to be significant in this reaction system [24]. Such transport resistances cause reduction in the observed activities of the catalysts. Macropore diameters of Amberlyst-36 and Dowex DR-2030 are quite high (24 and 20 nm, respectively) and much higher than the diameters
Fig. 3. Etherification of glycerol with C4 i-olefin (i-butene); glycerol conversion vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
of diffusing species. However, porosity of Dowex DR-2030 is 1.75 times higher than the porosity of Amberlyst-36. Since effective diffusivity within the macropores is expected to be proportional to the square of porosity (Wakao–Smith model), macropore diffusion resistance of reactants is expected to be more significant in Amberlyst-36 than in Dowex DR-2030. Hence, although the hydrogen exchange capacity of Amberlyst-36 was somewhat higher than Dowex DR-2030 (Table 1), transport of the reactants to the active sites within Amberlyst-36 was more difficult than Dowex DR-2030. Similar conclusions were reported in our earlier publication for the etherification of glycerol with tertbutyl alcohol over different solid acid catalysts [9]. These acidic resins include –SO3H groups which are directly related to hydrogen exchange capacity. A dense network of –SO3H groups were formed within the micro-grains of the resins. Penetration of pyridine into the resin catalysts during DRIFTS analyses might also have been hindered due to diffusional effects. Hence, DRIFTS results indicated a higher number of Brønsted acid sites for Dowex DR-2030, which correspond to the sites near the external surface of the catalysts, up to a depth that pyridine had penetrated into catalyst pores. Ozbay et al. [9] had also reported in a recent publication on etherification of glycerol with tert-butyl alcohol that, mainly the acid sites near the outer surface of the acidic resins contributed to the Brønsted acidity measured by DRIFTS analysis of pyridine adsorbed samples and also to the etherification reaction at low temperatures. As reaction temperature was increased to 90 °C, glycerol conversion increased and reached to about 100% over both Amberlyst-36 and Dowex DR2030. With both of these resin type catalysts, nearly the same glycerol conversion was obtained at 90 °C. This is mainly due to the increase of the etherification rate with an increase in temperature.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
5
Fig. 6. Etherification of glycerol with C4 i-olefin (i-butene); diether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g). Fig. 4. Etherification of glycerol with C4 i-olefin (i-butene); di-i-butene (DIB) selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
When reaction temperature was increased to 120 °C, glycerol conversion decreased with all of the catalysts used in this work. All three of these catalysts were expected to be stable at 120 °C. Increase in temperature is also expected to increase the etherification reaction rate. However, the rate of i-butene oligomerization (mainly to di-i-butene, (DIB)) was also expected to increase with an increase in temperature. As it was discussed in the literature the increase of the rate of i-butene oligomerization was more than the glycerol etherification reaction rate with an increase of temperature [15,16]. Especially over 90 °C, this caused a negative effect on glycerol conversion. To prove this conclusion, di-i-butene mole fractions in the product mixtures obtained at different temperatures were also experimentally determined. Influence of temperature on selectivity of di-i-butene (DIB), obtained with the three catalysts used in this study is given in Fig. 4. As seen in this figure, at 70 °C, almost no di-i-butene (DIB) was formed. However, as the temperature was increased to 80 and 90 °C, selectivity of di-i-butene (DIB) increased sharply especially in the case of using silicotungstic acid as the catalyst. At 90 °C, selectivity of di-i-butene (DIB) approached to 0.26, with silicotungstic acid. As temperature was increased to 120 °C, selectivity of di-i-butene (DIB) was further increased with all of the catalysts used in this study. Especially in the case of silicotungstic acid, di-i-butene (DIB) selectivity was recorded as nearly 0.5. Apparently, higher acidity of silicotungstic acid also facilitated the oligomerization of i-butene. Due to quite high chemical attraction of i-butene to Keggin anion of silicotungstic acid, protonation of i-butene and formation of the alkoxy species on the surface was expected. This may couple with a next i-butene molecule producing di-i-butene (DIB) [12]. Decrease of glycerol conversion at 120 °C is a direct consequence of increase of di-i-butene (DIB) formation. Mono, di and tri ether selectivities obtained as a result of glycerol etherification with i-butene in the presence of Amberlyst-36, Dowex
Fig. 5. Etherification of glycerol with C4 i-olefin (i-butene); monoether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
DR-2030 and silicotungstic acid are shown in Figs. 5, 6 and 7, respectively. Monoethers of glycerol (MTBG) are not preferred as oxygenate additives, because of their low solubility in diesel and gasoline. Therefore, di and tri ethers of glycerol (DTBG and TTBG) are preferred products. Product selectivity was towards monoethers at 70 °C. Mono-ether selectivity values were about 0.9 with all three of the acidic catalyst used in this study. As temperature was increased to 80 °C, a drastic decrease in monoether selectivity was observed, especially with silicotungstic acid catalyst. Di- and tri-ether selectivity values reached to a maximum at this temperature with silicotungstic acid. At 90 °C, di- and tri-ether selectivities obtained in the presence of Dowex DR-2030 and silicotungstic acid were quite similar (of about 60%). However, for Amberlyst-36, mono-ether selectivity was still higher than di- and triether selectivity values. This was considered to be due to the lower porosity of Amberlyst-36 than Dowex DR-2030, which created higher transport resistance for the penetration of reactants to the active sites. In the case of silicotungstic acid, the pseudo-liquid character of these catalysts allows easier penetration of the reactants to the acid sites. Silicotungstic acid is much more acidic than the resin catalysts used in this work. In fact, it was reported to be even more acidic than sulfuric acid and it was considered as a super acid [23]. Due to very high Brønsted acidity of silicotungstic acid, it also catalyzes cracking and oligomerization reactions. Decrease of di- and tri-ether selectivity values observed with silicotungstic acid and Dowex DR-2030 with a further increase of temperature to 120 °C may be related to the decrease of i-butene concentration in the reactor due to formation of di-i-butene (DIB) and other side products [15,16]. 3.3. Etherification of glycerol with i-amylene mixture Etherification of glycerol with i-amylene mixture (2M2B and 2M1B) is also an acid catalyzed reaction. Formations of five ethers were expected in this case, as a result of consecutive etherification reactions. Two
Fig. 7. Etherification of glycerol with C4 i-olefin (i-butene); triether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
6
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Fig. 8. Etherification of glycerol with C5 i-olefin (i-amylene: 2M2B).
mono-ethers, two di-ethers and a tri-ether were expected to form as a result of these reactions [10]. We denote these ethers as ME1, ME2, DE1, DE2 and TE, respectively (Fig. 8). In the present work etherification of glycerol with the i-amylene mixture was investigated in a batch reactor using Amberlyst 36 and Dowex DR-2030 as the solid acid catalysts. Initial experiments performed with different acidic resin catalysts indicated that the reaction temperature should be higher than 110 °C, to
obtain meaningful conversion values. Both of these acidic resins are stable at such high temperatures. Since initial experiments performed with STA indicated large amount of side products during etherification of glycerol with i-amylenes, further analysis was not performed with this catalyst. As it was discussed above in Section 3.2, silicotungstic acid is a super acid with very high Brønsted acidity and it also catalyzes cracking reactions of i-amylenes.
Fig. 9. Etherification of glycerol with C5 i-olefin (i-amylene mixture); glycerol conversion vs. catalyst amount (g) with different catalysts (reaction time = 6 h, reaction temperature = 120 °C).
Fig. 10. Etherification of glycerol with C5 i-olefin (i-amylene mixture); glycerol conversion values obtained at different; temperatures (120 and 140 °C), reaction times (1, 3, 6, 12, 24 h) and catalyst amount (mcat = 0.3 or 1 g, Amberlyst-36).
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx Table 2 Initial reaction rate of etherification of glycerol with C5 i-olefin (i-amylene mixture); over Amberlyst-36. Temperature (°C)
120
140
mcat (g) 0.3 1
–RA (mol L−1 min−1) 0.0012 0.0016
0.0015 0.0127
3.3.1. Influence of catalyst type and catalyst amount on glycerol conversion A set of experiments were performed for etherification of glycerol with i-amylene mixture, using different amounts of Amberlyst-36 and Dowex DR-2030. Considering that the maximum allowable operating temperatures of these catalysts were 150 and 130 °C, respectively, reaction temperature was selected as 120 °C. Glycerol conversion results obtained at the end of a 6 h reaction period indicated that the activities of these catalysts were quite comparable at this temperature (Fig. 9). A significant increase was observed in glycerol conversion values, with an increase in catalyst amount. This is due to the increased number of active sites per unit volume of the reaction mixture. Glycerol conversion values obtained as a result of its reaction with i-amylenes were less than the corresponding values obtained with i-butene as the reactant. This was considered to be mainly due to the larger molecular size of i-amylenes than i-butene, which caused higher pore diffusion resistance, as well as more stearic effects. Considering that the catalytic activities of Amberlyst-36 and Dowex DR-2030 were quite similar, detailed investigation of effects of reaction conditions on the catalytic performance and product distributions were continued with Amberlyst-36, which was more stable at high temperatures.
3.3.2. Influence of reaction temperature, reaction time and catalyst amount using Amberlyst-36 Etherification of glycerol with the i-amylene mixture was investigated at both 120 and 140 °C, over Amberlyst-36 in the batch reactor. Effects of reaction time on glycerol conversion and product distributions were evaluated. In order to determine the effect of reaction time and catalyst amount, experiments were repeated at 1, 3, 6, 12, 24 h reaction periods and using 0.3 and 1 g catalyst loadings. Results are illustrated in Fig. 10. At 140 °C with 1 g of catalyst, the reaction was very fast in the first hour and glycerol conversion value reached to 70%. As the reaction time increased, glycerol conversion increased until the 6th hour of reaction period, approaching to almost complete conversion. After this point no significant change was observed in terms of glycerol conversion. As, it is seen in Fig. 10, glycerol conversion rate is slower in the case of using less amount of catalyst, at 140 °C. In the first hour of reaction period, glycerol conversion was only 10%, in the case of using 0.3 g of
7
catalyst. As the reaction time increased, glycerol conversion increased sharply until the 6th hour of reaction time, and approached to almost complete conversion after 24 h of reaction time. Glycerol conversion values also varied drastically with the amount of catalyst loading in the reactions conducted at 120 °C, especially at reaction times higher than 1 h. As it was expected, lower reaction rates were observed at 120 °C than the corresponding values obtained at 140 °C. As shown in Table 2, initial rate of conversion of glycerol to the products (evaluated from the conversion values found at 1 h reaction time) significantly increased when the catalyst amount was increased from 0.3 to 1.0 g. However, the increase of initial rate observed at 120 °C, as a result of increase of catalyst amount, was quite low. This is a surprising result, indicating that a large fraction of active surface of the acidic resin is initially covered by adsorbed glycerol molecules, especially at low temperatures. Apparently, when temperature was increased to 140 °C some fraction of adsorbed glycerol molecules were desorbed, allowing empty sites for the adsorption of i-amylene molecules. As it was discussed in one of our earlier publications for TAME and TAEE syntheses, such etherification reactions follow a Langmuir-Hinshelwood type mechanism and ethanol is more strongly adsorbed on the –SO3H sites then i-olefins [25]. Hence, at high alcohol concentrations and at low temperatures, most of the active surface was covered by the alcohol molecules, which caused a negative effect on etherification rate. It is quite probable to have a similar mechanism for the etherification reaction of glycerol with i-amylenes. As reaction proceeds, glycerol concentration in the reaction mixture decreases, emptying more sites for the adsorption of i-amylene molecules. Hence, effect of increase of catalyst amount to glycerol conversion became more apparent at reaction times higher than 1 h. Effect of reaction period on product distributions obtained at 120 °C was illustrated in Fig. 11(A) and (B), for different amounts of catalyst (Amberlyst-36) charged to the reactor. Both of these figures indicated that mono-ethers were the main products at initial reaction times. Decrease of mono-ether selectivity with reaction time was accompanied with an increase in di-ether and tri-ether selectivity values. This is due to the production of these ethers through a consecutive reaction path. As it was shown in Fig. 11(A) and (B), increase of catalyst amount from 0.3 g to 1.0 g, caused significant enhancement of the rate of etherification reactions, causing much sharper decrease of mono-ether selectivity and much higher di- and tri-ether selectivity values at longer reaction times. Di-ether selectivity values of about 0.7 and tri-ether selectivities approaching to 0.1, obtained with 1.0 g catalyst at reaction times longer than 12 h, were highly promising. Effect of reaction temperature on ether selectivity values is illustrated in Fig. 12. Comparison of the results reported in this figure with the corresponding results given in Fig. 11(A) clearly showed the positive effect of increase of temperature from 120 °C to 140 °C on the di-ether and tri-ether selectivity values obtained by using 0.3 g of catalyst. Much sharper decrease of
Fig. 11. Etherification of glycerol with C5 i-olefin (i-amylene mixture); product selectivity vs. reaction time with Amberlyst-36 catalyst (A) reaction temperature = 120 °C, mcat = 0.3 g; (B) reaction temperature = 120 °C, mcat = 1 g.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
8
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Research Fund Projects (BAP-03-04-2013-008 and BAP-03-04-2014002) are gratefully acknowledged.
References
Fig. 12. Etherification of glycerol with C5 i-olefin (i-amylene mixture); Product selectivity vs. reaction time with Amberlyst-36 catalyst, Reaction temperature = 140 °C, mcat = 0.3 g.
mono-ether selectivity and a much sharper increase of di- and tri-ether selectivity values were obtained as a result of increase of temperature to 140 °C. Considering that di- and tri-ethers of glycerol were more soluble in diesel fuel and gasoline, this has been considered as an attractive result. 4. Conclusions As a result of this study it was concluded that, fuel oxygenates can be successfully produced through etherification of glycerol with iamylenes, as well as with i-butene. Results proved that, higher reaction temperatures (120–140 °C) are needed for the etherification reaction with i-amylene than the best reaction temperature of glycerol etherification with i-butene (about 90 °C). Results obtained at 140 °C with Amberlyst-36 showed very high glycerol conversion values, approaching to 100% and quite high di- and tri-ether selectivity values during the reaction of glycerol with i-amylene mixture. In the case of etherification of glycerol with i-butene, increase of reaction temperature over 90 °C enhanced its oligomerization, which caused reduction in ether formation. Rapid increase in oligomerization caused an inhibiting effect on glycerol etherification at higher temperatures. Silicotungstic acid, with a higher Brønsted acidity than the resin type catalysts (Amberlyst-36 and Dowex DR-2030), showed better catalytic performance in glycerol etherification with i-butene, at temperatures lower than 90 °C. However, it facilitated the undesired side reactions at higher temperatures. It was also shown that the catalytic performance of Dowex DR-2030 was better than Amberlyst-36 in etherification of glycerol with i-butene, at temperatures lower than 90 °C. This result indicated the importance of macropore diffusion resistance on the etherification rate, since Dowex DR-2030 had a higher porosity than Amberlyst-36. Results also proved that increase of catalyst amount and increase of reaction temperature to 140 °C caused significant increase in etherification rates of glycerol with i-amylenes, giving quite high di- and tri-ether selectivity values. Di-ether and tri-ether selectivity values, approaching 0.7 and 0.1 respectively, were highly promising.
[1] K.C. Tokay, T. Dogu, G. Dogu, Dimethyl ether synthesis over alumina based catalysts, Chem. Eng. J. 184 (2012) 278–285. [2] D. Varisli, T. Dogu, G. Dogu, Petrochemicals from ethanol over a W–Si-based nanocomposite bidisperse solid acid catalyst, Chem. Eng. Sci. 65 (2010) 153–159. [3] F. Frusteri, F. Arena, G. Bonura, C. Cannilla, L. Spadaro, O. Di Blasi, Catalytic etherification of glycerol by tert-butyl alcohol to produce oxygenated additives for diesel fuel, Appl. Catal. A Gen. 367 (2009) 77–83. [4] J.A. Melero, G. Vicente, G. Morales, M. Paniagua, J. Bustamante, Oxygenated compounds derived from glycerol for biodiesel formulation: influence on EN 14214 quality parameters, Fuel 89 (2010) 2011–2018. [5] J.A. Melero, G. Vicente, G. Morales, M. Paniagua, J.M. Moreno, R. Roldan, A. Ezquerro, C. Perez, Acid-catalyzed etherification of bio-glycerol and isobutylene over sulfonic mesostructured silicas, Appl. Catal. A Gen. 346 (2008) 44–51. [6] L. Xiao, J. Mao, J. Zhou, X. Guo, S. Zhang, Enhanced performance of HY zeolites by acid wash for glycerol etherification with isobutene, Appl. Catal. A Gen. 393 (2011) 88–95. [7] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Conversion of biodiesel by-product glycerol to fuel ethers over different solid acid catalysts, Int. J. Chem. React. Eng. 8 (2010) A18. [8] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Effects of sorption enhancement and isobutene formation on etherification of glycerol with tert-butyl alcohol in a flow reactor, Ind. Eng. Chem. Res. 51 (2012) 8788–8895. [9] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Activity comparison of different solid acid catalysts in etherification of glycerol with tert-butyl alcohol in flow and batch reactors, Top. Catal. 56 (2013) 1790–1803. [10] J.F. Izquierdo, M. Montiel, I. Pales, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, M. Izquierdo, M.P. Hermo, X. Ariza, Fuel additives from glycerol etherification with light olefins: State of the art, Renew. Sust. Energ. Rev. 16 (2012) 6717–6724. [11] W. Kiatkittipong, P. Intaracharoen, N. Laosiripojana, C. Chaisuk, P. Praserthdam, S. Assabumrungrat, Glycerol ethers synthesis from glycerol etherification with tertbutyl alcohol in reactive distillation, Comput. Chem. Eng. 35 (2011) 2034–2043. [12] H. Jong Lee, D. Seung, K.S. Jung, H. Kim, I.N. Filimonov, Etherification of glycerol by isobutylene: tuning the product composition, Appl. Catal. A Gen. 390 (2010) 235–244. [13] M.D. Gonzalez, Y. Cesteros, P. Salagre, Establishing the role of Brønsted acidity and porosity for the catalytic etherification of glycerol with tert-butanol by modifying zeolites, Appl. Catal. A Gen. 450 (2013) 178–188. [14] K. Klepacova, D. Mravec, M. Bajus, tert-Butylation of glycerol catalysed by ionexchange resins, Appl. Catal. A Gen. 294 (2005) 141–147. [15] R.S. Karinen, A.O.I. Krause, New bio components from glycerol, Appl. Catal. A Gen. 306 (2006) 128–133. [16] K. Klepacova, D. Mravec, A. Kaszonyi, M. Bajus, Etherification of glycerol and ethylene glycol by isobutylene, Appl. Catal. A Gen. 328 (2007) 1–13. [17] N. Oktar, K. Murtezaoglu, G. Dogu, l. Gonderten, T. Dogu, Etherification rates of 2methyl-2-butene and 2-methyl-1-butene with ethanol for environmentally clean gasoline production, J. Chem. Technol. Biotechnol. 74 (1999) 155–161. [18] Z. Obali, T. Dogu, Activated carbon–tungstophosphoric acid catalysts for the synthesis of tert-amyl ethyl ether (TAEE), Chem. Eng. J. 138 (2008) 548–555. [19] R. Soto, C. Fité, E. Ramírez, R. Bringué, F. Cunill, Equilibrium of the simultaneous etherification of isobutene and isoamylenes with ethanol in liquid-phase, Chem. Eng. Res. Des. 92 (2014) 644–656. [20] R. Soto, C. Fité, E. Ramírez, R. Bringué, M. Iborra, Green metrics analysis applied to the simultaneous liquid-phase etherification of isobutene and isoamylenes with ethanol over Amberlyst™ 35, Green Processes Synth. 3 (2014) 321–333. [21] J.F. Izquierdo, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, X. Ariza, New biodiesel additives from glycerol and isoamylenes, Biofuels Bioprod. Biorefin. 8 (2014) 658–669. [22] J.F. Izquierdo, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, M.P. Hermo, X. Ariza, I. Fernandez, Ethers of glycerol and isoamylenes as biodiesel additives: synthesis and characterization, Chem. Eng. Trans. 32 (2013) 877–882. [23] D. Varisli, T. Dogu, G. Dogu, Novel mesoporous nanocomposite WOx-silicate acidic catalysts: ethylene and diethylether from ethanol, Ind. Eng. Chem. Res. 48 (2009) 9394–9401. [24] N. Oktar, K. Murtezaoglu, T. Dogu, G. Dogu, Dynamic analysis of adsorption equilibrium and rate parameters of reactants and products in MTBE, ETBE and TAME production, Can. J. Chem. Eng. 77 (1999) 406–412. [25] N. Boz, T. Dogu, K. Murtezaoglu, G. Dogu, Mechanism of TAME and TAEE synthesis from diffuse–reflectance FTIR analysis, Catal. Today 100 (2005) 419–424.
Acknowledgments The financial support of Scientific and Technical Research Council of Turkey (TUBITAK-112M234) and Middle East Technical University
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Etherification of glycerol with C4 and C5 reactive olefins Burcin Ikizer a, Nuray Oktar b,⁎, Timur Dogu a a b
Department of Chemical Engineering, Middle East Technical University, 06800 Ankara, Turkey Department of Chemical Engineering, Gazi University, 06430 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 29 April 2015 Received in revised form 17 June 2015 Accepted 24 June 2015 Available online xxxx Keywords: Glycerol Etherification i-Amylene i-Butene Amberlyst-36 Dowex DR-2030
a b s t r a c t Etherification of glycerol with C4 (i-butene) and C5 (i-amylenes: 2-Methyl-2-Butene (2M2B) and 2-Methyl-1Butene (2M1B) mixture) i-olefins was investigated for the production of transportation fuel oxygenates. Results proved formation of mono-, di- and tri-ethers of glycerol, as a result of its etherification with i-amylene at 120–140 °C over Amberlyst-36. Glycerol fractional conversion values and di-ether selectivity values approaching to 1.0 and 0.7 respectively were highly promising. Glycerol etherification results obtained by using i-butene as the reactant, proved the importance of Brønsted acidity of the catalyst, as well as the pore diffusion resistance, on the catalytic performance. Silicotungstic acid (STA), with very high acidity, showed very good performance at T ≤ 80 °C. However, at higher temperatures it also facilitated the oligomerization of i-butene, which caused a negative effect on glycerol etherification. Both Amberlyst-36 and Dowex DR-2030 showed excellent performance in catalyzing glycerol with i-butene, at 90 °C. Results proved that etherification of the biodiesel by-product glycerol could be successfully achieved by using i-amylenes, as well as i-butene, to improve the economics of biodiesel production. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Rapid depletion of petroleum reserves opened new avenues of research for the production of alternative transportation fuels, from non-petroleum feedstock [1,2]. Biodiesel, which is a renewable and environmentally friendly alternative to petroleum-based diesel fuel, is produced by trans-esterification of vegetable oils and animal fats with alcohols (methanol/ethanol), in the presence of a basic catalyst [3,4]. Due to its biodegradable nature, quite high cetane number, higher flash point and lower emission of pollutants than petroleum-based diesel fuel, biodiesel attracted significant attention of fuel producers in recent decades [5,6]. However, large amount of glycerol is also produced as a side product during trans-esterification of oils with alcohols. Glycerol (1, 2, 3-propanetriol) is the main side product of biodiesel production process and one mole of glycerol is produced for every three moles of methyl esters. This corresponds to about 10 wt.% of the biodiesel product. Production rate of biodiesel is expected to increase significantly in the coming years. Hence, a remarkable surplus of glycerol will be generated as a result of increase of biodiesel production. Economics of biodiesel production strongly depends upon the efficient utilization of glycerol, which is produced as a side product during transesterification [7–9]. Some of the possible alternative ways for glycerol utilization are; acrolein production by its dehydration, hydrogen and synthesis gas
⁎ Corresponding author. E-mail address: [email protected] (N. Oktar).
production by reforming, selective oxidation for dihydroxyacetone, production of 1,3-propanediol, and its esterification/etherification to produce fuel additives and/or valuable chemicals [8,9]. Among these alternatives, its etherification with i-butene or tert-butyl alcohol has been considered as attractive routes to produce fuel oxygenates [3, 5–9,11–16]. Etherification of glycerol with i-butene or tert-butyl alcohol would produce two mono-tert-butyl glycerol ether isomers (3-tertbutoxy-1,2-propanediol (MTBG1) and 2-tert-butoxy-1,3-propanediol (MTBG2)), two di-tert-butyl glycerol ether isomers (2,3-di-tertbutoxy-1-propanol (DTBG1) and 1,3-di-tert-butoxy-2-propanol (DTBG2)) and a tri-tert-butyl-glycerol ether (1,2,3-tri-tert-butoxypropane (TTBG)) [5,8,15,16]. These ethers were reported to have high octane numbers and can be used as gasoline additives for octane enhancement. They can also be mixed with biodiesel to increase its amount and to improve the economics of biodiesel production. Solubility of monoethers (MTBG) in diesel fuel is quite low. On the other hand, di-ethers (DTBG) and tri-ethers (TTBG) are highly soluble in diesel and gasoline. Hence, especially di-ethers and tri-ethers were considered as excellent fuel components which can be added to gasoline, instead of current commercial oxygenate additives, such as methyl tert-butyl ether (MTBE), ethyl tert-butyl ether (ETBE) and tert-amyl methyl ether (TAME). In the case of blending of these ethers to diesel fuel, important positive effects were reported in terms of reduction of particulate matter, carbon monoxide and hydrocarbon emissions of the fuels. Furthermore, these ethers can lower the cloud point and viscosity of the fuels [5,6,16]. Etherification of glycerol with i-butene has been investigated during the last decade by some researchers, using different solid acid
http://dx.doi.org/10.1016/j.fuproc.2015.06.039 0378-3820/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
2
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
catalysts [12–16]. Klepacova et al. [16] reported some results for the etherification of glycerol using Amberlyst-15, Amberlyst-35, H-Y, HBeta zeolites and homogeneous p-toluenesulfonic acid as the catalysts. Quite high di-and tri-ether selectivity values were achieved over ion exchange resin catalyst Amberlyst-35. Karinen and Krause [15] also investigated the etherification of glycerol with i-butene in the presence of Amberlyst-35, in a batch reactor. Optimal selectivity towards di-ethers and nearly complete glycerol conversion were obtained with optimization of reaction conditions. Etherification of glycerol with i-butene over two sulfonic acid modified mesostructured silicas and commercial HY zeolite were also reported in the literature [5,6]. Compared to commercial HY, nitric acid and citric acid treated catalysts gave an increased glycerol conversion and higher selectivities to DTBG and TTBG [6]. All of these studies were performed in batch reactors at quite long reaction times. Solubility of glycerol in i-butene is quite low, and hence mass transfer resistances play a significant role on the observed rate of this etherification process. Use of a solvent has been recommended to enhance the rate of etherification of glycerol with i-butene. Klepacova et al. [16] tested different solvents (dioxane, dimethyl sulfoxide and sulfolane) to eliminate or reduce mass transfer effects. Maximum solubility of i-butene was obtained in dioxane. More recently, etherification of glycerol was investigated in our research group, using tert-butyl alcohol, instead of i-butene [7–9]. Glycerol is highly soluble in tert-butyl alcohol and hence higher etherification rates were achieved. In these studies, catalytic performances of different solid acid catalysts were tested in batch and flow reactors. Results proved the advantages of flow reactors in terms of achieving very high conversions at quite low residence times. Amberlyst-15 was shown to give the highest activity in the temperature range of 90–100 °C. However, its stability was low at higher temperatures. At higher temperatures (110–130 °C), best catalytic performance was achieved by using Amberlyst-36. Results reported in these studies also indicated the negative effect of water on the catalyst performance. Water adsorbed on the active acid sites reversibly deactivated the catalyst [7–9]. Importance of Brønsted acidity of the catalyst used for this reaction is also discussed in a recent work of Gonzales et al. [13]. All of these reported studies on the etherification of glycerol were performed either with i-butene or tert-butyl alcohol. Another possible route may be etherification of glycerol with C5 i-olefins (i-amylenes: 2-Methyl-1-Butene (2M1B) or 2-Methyl-2-Butene (2M2B)). Nearly 10% of C5 reactive olefins are present in FCC gasoline. There are number of studies in the literature for the etherification of these olefins with methanol or ethanol to produce some octane enhancing oxygenates, such as tert-amyl methyl ether (TAME) and tert-amyl ethyl ether (TAEE) [17–20]. Comparison of etherification rates 2M2B and 2M1B with ethanol was illustrated in an earlier publication of our research group, indicating that 2M2B, with higher alkyl groups attached to the carbon atoms of double bond, is more stable and somewhat less reactive than 2M1B [17]. Occurrence of isomerization reaction between 2M2B and 2M1B is also expected to take place in the presence of solid acid catalysts. Etherification of glycerol with C5 i-olefins has the potential to contribute positively to the economics of biodiesel production. Such a process would also help to decrease the harmful effects of this olefin on environment and human health. There are only a few publications in the literature for the etherification of glycerol with i-amylenes. Izquierdo et al. [21,22] reported some results for etherification of glycerol with i-amylenes in the presence of various kinds of organic solvents and also using several macromolecular resin type catalysts. Their results proved the importance of pore structure of the resin, as well as the solubility of glycerol in the solvent used during the etherification process, on its conversion. Miscibility of both reactants in the solvent was found to be the main factor affecting the progress of this reaction. The highest glycerol conversion was obtained in the presence of tertiary amyl alcohol as the solvent, due to its good solvent polarity. In the present research, etherification of glycerol was investigated with both i-amylenes and i-butene. Catalytic performances of some
resin type solid acid catalysts (Amberlyst-36, Dowex DR-2030), as well as silicotungstic acid were tested and the results were compared.
2. Experimental 2.1. Catalysts and chemicals Chemicals used in this study were i-butene (≥ 98% Air Products), Glycerol (≥99.5% Merck), 1,4 Dioxane (≥99%, Merck), i-amylene mixture [(2-Methyl-2-Butene (~ 85%) and 2-Methyl-1-Butene (~ 15%), Merck)] and Ethanol (≥99.5%, Merck). Two ion exchange resin catalysts, namely Amberlyst-36 (Sigma Aldrich) and Dowex DR-2030 (The Dow Chemical Company) and silicotungstic acid (STA) (H4SiW12O40) were used as the solid acid catalysts. Ion exchange resin catalysts are copolymers in macroreticular structure and functionalized with sulfonic acid groups. Silicotungstic acid (STA) is a Keggin type heteropolyacid with very high proton mobility. One of the reasons of selection of these catalysts for the etherification of glycerol was their quite high Brønsted acidity. Another important factor of this selection was their maximum operating temperatures. Our initial experiments performed with various type ion-exchange resin catalysts (Amberlyst-15, Amberlyst-16 etc.) showed that temperatures higher than 110 °C were needed to achieve reasonable conversion rates in etherification of glycerol with i-amylenes. However, many of the resin type solid acid catalysts are not stable at high temperatures. On the other hand, the maximum operating temperatures of Amberlyst-36, Dowex DR-2030 and silicotungstic acid are 150, 130 and 400 °C, respectively (Table 1). Considering the inhibition effect of adsorbed water on the rate of etherification reactions performed using acidic resins, the catalysts used in our work were dried at 110 °C. As it was discussed by Soto et al. [19] the amount of adsorbed water was less than 3.5 wt.% over the resin type catalysts pretreated at this temperature and inhibition of etherification rate with this much adsorbed water was negligible. Some of the physical properties of these catalysts are given in Table 1.
2.2. Characterization of the catalysts Diffuse Reflectance FT-IR Spectroscopy (DRIFTS) analysis of pyridine adsorbed catalyst samples was performed by using a PerkinElmer Spectrum One instrument. The procedure of Ozbay et al. [9] was followed for this analysis. Before these tests, samples were pretreated at 110 °C to remove water adsorbed on the acid sites. After the adsorption of pyridine, samples were dried at 40 °C, before the DRIFTS analysis. A background spectrum was recorded with KBr. FT-IR spectrum of the sample which was not treated by pyridine was subtracted from the spectrum of pyridine-adsorbed sample for the detection of adsorbed surface species. Relative strengths of Lewis and Brønsted acid sites were then determined from the differences of these spectra.
Table 1 Physical properties of the catalysts used in this study. Property Exchange capacity (meq H+/g) Surface area (m2/g) Porosity (cc/g) Average pore diameter (nm) Particle diameter (cm) Max. operating temperature (°C) a b c
a
Dowex DR-2030 b
Silicotungstic acid (STA) c
5.4 33 0.2 24 0.06–0.085 150
4.7 30 0.35 20 0.043–0.0525 130
– 7.8 0.03 – Powder 400 [23]
Amberlyst-36
Data obtained from Sigma-Aldrich Chemistry. Data obtained from The Dow Chemical Company. Measured by Quantachrome Corp., Autosorb-6 device in Central Laboratory in METU.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
3
2.3. Reaction experiments Etherification reactions of glycerol with i-butene and i-amylene mixture were performed in a stainless steel autoclave batch reactor (75 mL), equipped with a PID temperature controller, a magnetic stirrer and a pressure gauge. Etherification reactions with i-butene were performed in a temperature range of 70–120 °C using Amberlyst-36, Dowex DR-2030 and silicotungstic acid as the solid acid catalysts. In each experiment, 0.055 mol of glycerol, 0.3 g of catalyst, 0.446 mol i-butene and 30 mL 1, 4 dioxane (solvent) were introduced into the reactor. During the reaction runs performed with i-butene, the autogenous pressure in the reactor was recorded as 8–10 bars. Etherification of glycerol with i-amylene mixture was performed using Amberlyst-36 and Dowex DR-2030 under different reaction conditions. In the case of using Amberlyst-36, experiments were also repeated with different amounts of catalyst, namely 0.3 and 1 g. Also, another set of experiments were performed at different reaction times (1, 3, 6, 12, 24 h) with this catalyst. Initial experiments performed with i-amylene mixture indicated that the reaction temperature should be higher than the experiments performed with i-butene to achieve sufficiently high glycerol conversions. Hence, reaction temperatures were selected as 120 and 140 °C. Due to the instability of the resin type catalysts at higher temperatures, maximum temperature was selected as 140 °C. In each test, 0.0326 mol of glycerol and 0.261 mol i-amylene mixture were used. During these reaction runs performed with the i-amylene mixture, the autogenous pressure in the reactor was recorded as 6–8 bars. After the completion of a reaction run, the reactor was cooled to room temperature, until the system pressure reached to atmospheric. In the case of using i-amylene mixture as the reactant, 50 mL of ethanol was added to the reactor immediately after opening it. As a result of ethanol addition, a single phase of glycerol, i-amylene and the reaction products mixture was obtained. This mixture was stirred until it became homogenous. At the end of each run, chemical composition of the product mixture was analyzed by a gas chromatograph (Agilent 6890N), which was equipped with a capillary HP INNOWax column and a flame ionization detector (FID). Flow rate of the carrier gas (He) was 1.4 mL/min and flow rates of dry air and hydrogen gas were 300 mL/min and 30 mL/min, respectively. Detector and the injector temperatures were adjusted as 270 and 295 °C, respectively. In order to be able to observe all of the peaks corresponding to different species, a temperature program was adjusted, starting at 50 °C, followed by a 4 °C/min ramp up to 115 °C and then followed with a 15 °C ramp up to 250 °C, where temperature was kept constant for about 10 min. Chromatographic peaks of 2M2B and 2M1B were very close to each other. Hence, we used summation of these peaks to evaluate the amount of i-amylenes in the mixture. 3. Results and discussion 3.1. DRIFTS analysis of catalysts Diffuse Reflectance Fourier Transform IR Spectroscopy (DRIFTS) analysis of pyridine adsorbed samples was performed to determine the relative strengths of the Brønsted and Lewis acid sites on the surface of the solid acid catalysts used in this study (Fig. 1). Distinctive bands, which appear at 1540 and 1640 cm−1 correspond to the Brønsted acid sites. However, for the Lewis acid sites, bands are expected at 1450 and 1598 cm−1 [9]. According to DRIFTS given in Fig. 1, for Amberlyst36 and Dowex DR-2030 acidic resins, the Lewis bands at 1450 and 1598 cm−1 were quite weak. However, the intensities of the bands at 1540 and 1640 cm−1 were quite strong, indicating a high number of Brønsted acid sites. Sulfonic acid functionality on the catalyst surface for the acidic resins is the cause of Brønsted acidity [9]. Due to the contributions of both Brønsted and Lewis acid sites another band was observed at 1490 cm−1 for Amberlyst-36 and Dowex DR-2030. Because
Fig. 1. DRIFT spectra of pyridine adsorbed solid acid catalysts (Amberlyst-36, Dowex DR2030, silicotungstic acid (STA)).
of low Lewis acidity of these resins, it can be considered that the band at 1490 cm−1 was also mainly due to Brønsted acid sites. Although the hydrogen exchange capacity of Amberlyst-36 was higher than that of Dowex DR-2030 (Table 1), DRIFTS analysis indicated somewhat a higher number of Brønsted acid sites for Dowex DR-2030. This was considered to be due to higher porosity of Dowex DR-2030 than Amberlyst-36, which allowed better penetration of pyridine into the catalyst to be adsorbed on the active sites. The DRIFT spectrum of pyridine adsorbed silicotungstic acid showed the presence of both Lewis and Brønsted acid sites (Fig. 1). However, comparison of the intensities of the bands showed that the Brønsted acidity of silicotungstic acid was much higher than its Lewis acidity. Strong Brønsted acidity is mainly due to high proton mobility of silicotungstic acid. Heteropolyacids are known to have acidities even higher than sulfuric acid and they are considered as super acids [23]. 3.2. Etherification of glycerol with i-butene Etherification of glycerol with i-butene is expected to take place following three consecutive reversible reaction steps over the solid acid catalysts. Reaction steps giving the products, namely mono-tert-butyl glycerol ether isomers (3-tert-butoxy-1,2-propanediol; MTBG1 and 2tert-butoxy-1,3-propanediol; MTBG2), two di-tert-butyl glycerol ether isomers (2,3-di-tert-butoxy-1-propanol; DTBG1 and 1,3-di-tertbutoxy-2-propanol; DTBG2), and a tri-tert-butyl glycerol ether (1,2,3tri-tert-butoxy-propane; TTBG) are shown in Fig. 2. Besides these reactions, oligomerization and hydration of i-butene are two of the possible side reactions, which may take place in the reactor [16]. As a result of oligomerization of i-butene, two isomers of di-i-butene, namely TMP1 (2,4,4-trimethyl-1-pentene) and TMP-2 (2,4,4-trimethyl-2-pentene) are expected to form. Oligomerization of i-butene is not desired and has a negative effect on ether yields. However, as a result of hydration of i-butene tert-butyl alcohol may form, which may also react with glycerol to produce the ethers. Glycerol conversion values obtained as a result of its etherification with i-butene are reported in Fig. 3. Results obtained at different temperatures by using Amberlyst-36, Dowex DR-2030 and silicotungstic acid catalysts, in the presence of 1,4 Dioxan as solvent, are reported in this figure. At 70 and 80 °C, maximum glycerol conversion values were obtained by using silicotungstic acid as the catalyst. Considering that the Brønsted acidity of silicotungstic acid was higher than the other resin type solid acid catalysts, this was an expected result. Heteropolyacids are sometimes considered as pseudo liquids, indicating easy penetration of reactants to the acid sites. In this temperature range, catalytic performance of Dowex DR-2030 was also higher than the performance of Amberlyst-36. Comparison of the activities of the two resin type catalysts was also in concordance with the Brønsted acidity results (Fig. 1). Dowex DR-2030 was more porous than Amberlyst-36, which
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
4
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Fig. 2. Etherification of glycerol with C4 i-olefin (i-butene).
facilitated transport of reactants to the active sites within the catalyst. Macroreticular resins are composed of interconnected gel-like micro grains and macropores between these micrograins. As it was discussed in an earlier publication of our research group, both macropore diffusion and penetration resistances of reactants into the micro-grains were expected to be significant in this reaction system [24]. Such transport resistances cause reduction in the observed activities of the catalysts. Macropore diameters of Amberlyst-36 and Dowex DR-2030 are quite high (24 and 20 nm, respectively) and much higher than the diameters
Fig. 3. Etherification of glycerol with C4 i-olefin (i-butene); glycerol conversion vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
of diffusing species. However, porosity of Dowex DR-2030 is 1.75 times higher than the porosity of Amberlyst-36. Since effective diffusivity within the macropores is expected to be proportional to the square of porosity (Wakao–Smith model), macropore diffusion resistance of reactants is expected to be more significant in Amberlyst-36 than in Dowex DR-2030. Hence, although the hydrogen exchange capacity of Amberlyst-36 was somewhat higher than Dowex DR-2030 (Table 1), transport of the reactants to the active sites within Amberlyst-36 was more difficult than Dowex DR-2030. Similar conclusions were reported in our earlier publication for the etherification of glycerol with tertbutyl alcohol over different solid acid catalysts [9]. These acidic resins include –SO3H groups which are directly related to hydrogen exchange capacity. A dense network of –SO3H groups were formed within the micro-grains of the resins. Penetration of pyridine into the resin catalysts during DRIFTS analyses might also have been hindered due to diffusional effects. Hence, DRIFTS results indicated a higher number of Brønsted acid sites for Dowex DR-2030, which correspond to the sites near the external surface of the catalysts, up to a depth that pyridine had penetrated into catalyst pores. Ozbay et al. [9] had also reported in a recent publication on etherification of glycerol with tert-butyl alcohol that, mainly the acid sites near the outer surface of the acidic resins contributed to the Brønsted acidity measured by DRIFTS analysis of pyridine adsorbed samples and also to the etherification reaction at low temperatures. As reaction temperature was increased to 90 °C, glycerol conversion increased and reached to about 100% over both Amberlyst-36 and Dowex DR2030. With both of these resin type catalysts, nearly the same glycerol conversion was obtained at 90 °C. This is mainly due to the increase of the etherification rate with an increase in temperature.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
5
Fig. 6. Etherification of glycerol with C4 i-olefin (i-butene); diether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g). Fig. 4. Etherification of glycerol with C4 i-olefin (i-butene); di-i-butene (DIB) selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
When reaction temperature was increased to 120 °C, glycerol conversion decreased with all of the catalysts used in this work. All three of these catalysts were expected to be stable at 120 °C. Increase in temperature is also expected to increase the etherification reaction rate. However, the rate of i-butene oligomerization (mainly to di-i-butene, (DIB)) was also expected to increase with an increase in temperature. As it was discussed in the literature the increase of the rate of i-butene oligomerization was more than the glycerol etherification reaction rate with an increase of temperature [15,16]. Especially over 90 °C, this caused a negative effect on glycerol conversion. To prove this conclusion, di-i-butene mole fractions in the product mixtures obtained at different temperatures were also experimentally determined. Influence of temperature on selectivity of di-i-butene (DIB), obtained with the three catalysts used in this study is given in Fig. 4. As seen in this figure, at 70 °C, almost no di-i-butene (DIB) was formed. However, as the temperature was increased to 80 and 90 °C, selectivity of di-i-butene (DIB) increased sharply especially in the case of using silicotungstic acid as the catalyst. At 90 °C, selectivity of di-i-butene (DIB) approached to 0.26, with silicotungstic acid. As temperature was increased to 120 °C, selectivity of di-i-butene (DIB) was further increased with all of the catalysts used in this study. Especially in the case of silicotungstic acid, di-i-butene (DIB) selectivity was recorded as nearly 0.5. Apparently, higher acidity of silicotungstic acid also facilitated the oligomerization of i-butene. Due to quite high chemical attraction of i-butene to Keggin anion of silicotungstic acid, protonation of i-butene and formation of the alkoxy species on the surface was expected. This may couple with a next i-butene molecule producing di-i-butene (DIB) [12]. Decrease of glycerol conversion at 120 °C is a direct consequence of increase of di-i-butene (DIB) formation. Mono, di and tri ether selectivities obtained as a result of glycerol etherification with i-butene in the presence of Amberlyst-36, Dowex
Fig. 5. Etherification of glycerol with C4 i-olefin (i-butene); monoether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
DR-2030 and silicotungstic acid are shown in Figs. 5, 6 and 7, respectively. Monoethers of glycerol (MTBG) are not preferred as oxygenate additives, because of their low solubility in diesel and gasoline. Therefore, di and tri ethers of glycerol (DTBG and TTBG) are preferred products. Product selectivity was towards monoethers at 70 °C. Mono-ether selectivity values were about 0.9 with all three of the acidic catalyst used in this study. As temperature was increased to 80 °C, a drastic decrease in monoether selectivity was observed, especially with silicotungstic acid catalyst. Di- and tri-ether selectivity values reached to a maximum at this temperature with silicotungstic acid. At 90 °C, di- and tri-ether selectivities obtained in the presence of Dowex DR-2030 and silicotungstic acid were quite similar (of about 60%). However, for Amberlyst-36, mono-ether selectivity was still higher than di- and triether selectivity values. This was considered to be due to the lower porosity of Amberlyst-36 than Dowex DR-2030, which created higher transport resistance for the penetration of reactants to the active sites. In the case of silicotungstic acid, the pseudo-liquid character of these catalysts allows easier penetration of the reactants to the acid sites. Silicotungstic acid is much more acidic than the resin catalysts used in this work. In fact, it was reported to be even more acidic than sulfuric acid and it was considered as a super acid [23]. Due to very high Brønsted acidity of silicotungstic acid, it also catalyzes cracking and oligomerization reactions. Decrease of di- and tri-ether selectivity values observed with silicotungstic acid and Dowex DR-2030 with a further increase of temperature to 120 °C may be related to the decrease of i-butene concentration in the reactor due to formation of di-i-butene (DIB) and other side products [15,16]. 3.3. Etherification of glycerol with i-amylene mixture Etherification of glycerol with i-amylene mixture (2M2B and 2M1B) is also an acid catalyzed reaction. Formations of five ethers were expected in this case, as a result of consecutive etherification reactions. Two
Fig. 7. Etherification of glycerol with C4 i-olefin (i-butene); triether selectivity vs. reaction temperature with different catalysts (reaction time = 6 h, mcat = 0.3 g).
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
6
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Fig. 8. Etherification of glycerol with C5 i-olefin (i-amylene: 2M2B).
mono-ethers, two di-ethers and a tri-ether were expected to form as a result of these reactions [10]. We denote these ethers as ME1, ME2, DE1, DE2 and TE, respectively (Fig. 8). In the present work etherification of glycerol with the i-amylene mixture was investigated in a batch reactor using Amberlyst 36 and Dowex DR-2030 as the solid acid catalysts. Initial experiments performed with different acidic resin catalysts indicated that the reaction temperature should be higher than 110 °C, to
obtain meaningful conversion values. Both of these acidic resins are stable at such high temperatures. Since initial experiments performed with STA indicated large amount of side products during etherification of glycerol with i-amylenes, further analysis was not performed with this catalyst. As it was discussed above in Section 3.2, silicotungstic acid is a super acid with very high Brønsted acidity and it also catalyzes cracking reactions of i-amylenes.
Fig. 9. Etherification of glycerol with C5 i-olefin (i-amylene mixture); glycerol conversion vs. catalyst amount (g) with different catalysts (reaction time = 6 h, reaction temperature = 120 °C).
Fig. 10. Etherification of glycerol with C5 i-olefin (i-amylene mixture); glycerol conversion values obtained at different; temperatures (120 and 140 °C), reaction times (1, 3, 6, 12, 24 h) and catalyst amount (mcat = 0.3 or 1 g, Amberlyst-36).
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx Table 2 Initial reaction rate of etherification of glycerol with C5 i-olefin (i-amylene mixture); over Amberlyst-36. Temperature (°C)
120
140
mcat (g) 0.3 1
–RA (mol L−1 min−1) 0.0012 0.0016
0.0015 0.0127
3.3.1. Influence of catalyst type and catalyst amount on glycerol conversion A set of experiments were performed for etherification of glycerol with i-amylene mixture, using different amounts of Amberlyst-36 and Dowex DR-2030. Considering that the maximum allowable operating temperatures of these catalysts were 150 and 130 °C, respectively, reaction temperature was selected as 120 °C. Glycerol conversion results obtained at the end of a 6 h reaction period indicated that the activities of these catalysts were quite comparable at this temperature (Fig. 9). A significant increase was observed in glycerol conversion values, with an increase in catalyst amount. This is due to the increased number of active sites per unit volume of the reaction mixture. Glycerol conversion values obtained as a result of its reaction with i-amylenes were less than the corresponding values obtained with i-butene as the reactant. This was considered to be mainly due to the larger molecular size of i-amylenes than i-butene, which caused higher pore diffusion resistance, as well as more stearic effects. Considering that the catalytic activities of Amberlyst-36 and Dowex DR-2030 were quite similar, detailed investigation of effects of reaction conditions on the catalytic performance and product distributions were continued with Amberlyst-36, which was more stable at high temperatures.
3.3.2. Influence of reaction temperature, reaction time and catalyst amount using Amberlyst-36 Etherification of glycerol with the i-amylene mixture was investigated at both 120 and 140 °C, over Amberlyst-36 in the batch reactor. Effects of reaction time on glycerol conversion and product distributions were evaluated. In order to determine the effect of reaction time and catalyst amount, experiments were repeated at 1, 3, 6, 12, 24 h reaction periods and using 0.3 and 1 g catalyst loadings. Results are illustrated in Fig. 10. At 140 °C with 1 g of catalyst, the reaction was very fast in the first hour and glycerol conversion value reached to 70%. As the reaction time increased, glycerol conversion increased until the 6th hour of reaction period, approaching to almost complete conversion. After this point no significant change was observed in terms of glycerol conversion. As, it is seen in Fig. 10, glycerol conversion rate is slower in the case of using less amount of catalyst, at 140 °C. In the first hour of reaction period, glycerol conversion was only 10%, in the case of using 0.3 g of
7
catalyst. As the reaction time increased, glycerol conversion increased sharply until the 6th hour of reaction time, and approached to almost complete conversion after 24 h of reaction time. Glycerol conversion values also varied drastically with the amount of catalyst loading in the reactions conducted at 120 °C, especially at reaction times higher than 1 h. As it was expected, lower reaction rates were observed at 120 °C than the corresponding values obtained at 140 °C. As shown in Table 2, initial rate of conversion of glycerol to the products (evaluated from the conversion values found at 1 h reaction time) significantly increased when the catalyst amount was increased from 0.3 to 1.0 g. However, the increase of initial rate observed at 120 °C, as a result of increase of catalyst amount, was quite low. This is a surprising result, indicating that a large fraction of active surface of the acidic resin is initially covered by adsorbed glycerol molecules, especially at low temperatures. Apparently, when temperature was increased to 140 °C some fraction of adsorbed glycerol molecules were desorbed, allowing empty sites for the adsorption of i-amylene molecules. As it was discussed in one of our earlier publications for TAME and TAEE syntheses, such etherification reactions follow a Langmuir-Hinshelwood type mechanism and ethanol is more strongly adsorbed on the –SO3H sites then i-olefins [25]. Hence, at high alcohol concentrations and at low temperatures, most of the active surface was covered by the alcohol molecules, which caused a negative effect on etherification rate. It is quite probable to have a similar mechanism for the etherification reaction of glycerol with i-amylenes. As reaction proceeds, glycerol concentration in the reaction mixture decreases, emptying more sites for the adsorption of i-amylene molecules. Hence, effect of increase of catalyst amount to glycerol conversion became more apparent at reaction times higher than 1 h. Effect of reaction period on product distributions obtained at 120 °C was illustrated in Fig. 11(A) and (B), for different amounts of catalyst (Amberlyst-36) charged to the reactor. Both of these figures indicated that mono-ethers were the main products at initial reaction times. Decrease of mono-ether selectivity with reaction time was accompanied with an increase in di-ether and tri-ether selectivity values. This is due to the production of these ethers through a consecutive reaction path. As it was shown in Fig. 11(A) and (B), increase of catalyst amount from 0.3 g to 1.0 g, caused significant enhancement of the rate of etherification reactions, causing much sharper decrease of mono-ether selectivity and much higher di- and tri-ether selectivity values at longer reaction times. Di-ether selectivity values of about 0.7 and tri-ether selectivities approaching to 0.1, obtained with 1.0 g catalyst at reaction times longer than 12 h, were highly promising. Effect of reaction temperature on ether selectivity values is illustrated in Fig. 12. Comparison of the results reported in this figure with the corresponding results given in Fig. 11(A) clearly showed the positive effect of increase of temperature from 120 °C to 140 °C on the di-ether and tri-ether selectivity values obtained by using 0.3 g of catalyst. Much sharper decrease of
Fig. 11. Etherification of glycerol with C5 i-olefin (i-amylene mixture); product selectivity vs. reaction time with Amberlyst-36 catalyst (A) reaction temperature = 120 °C, mcat = 0.3 g; (B) reaction temperature = 120 °C, mcat = 1 g.
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039
8
B. Ikizer et al. / Fuel Processing Technology xxx (2015) xxx–xxx
Research Fund Projects (BAP-03-04-2013-008 and BAP-03-04-2014002) are gratefully acknowledged.
References
Fig. 12. Etherification of glycerol with C5 i-olefin (i-amylene mixture); Product selectivity vs. reaction time with Amberlyst-36 catalyst, Reaction temperature = 140 °C, mcat = 0.3 g.
mono-ether selectivity and a much sharper increase of di- and tri-ether selectivity values were obtained as a result of increase of temperature to 140 °C. Considering that di- and tri-ethers of glycerol were more soluble in diesel fuel and gasoline, this has been considered as an attractive result. 4. Conclusions As a result of this study it was concluded that, fuel oxygenates can be successfully produced through etherification of glycerol with iamylenes, as well as with i-butene. Results proved that, higher reaction temperatures (120–140 °C) are needed for the etherification reaction with i-amylene than the best reaction temperature of glycerol etherification with i-butene (about 90 °C). Results obtained at 140 °C with Amberlyst-36 showed very high glycerol conversion values, approaching to 100% and quite high di- and tri-ether selectivity values during the reaction of glycerol with i-amylene mixture. In the case of etherification of glycerol with i-butene, increase of reaction temperature over 90 °C enhanced its oligomerization, which caused reduction in ether formation. Rapid increase in oligomerization caused an inhibiting effect on glycerol etherification at higher temperatures. Silicotungstic acid, with a higher Brønsted acidity than the resin type catalysts (Amberlyst-36 and Dowex DR-2030), showed better catalytic performance in glycerol etherification with i-butene, at temperatures lower than 90 °C. However, it facilitated the undesired side reactions at higher temperatures. It was also shown that the catalytic performance of Dowex DR-2030 was better than Amberlyst-36 in etherification of glycerol with i-butene, at temperatures lower than 90 °C. This result indicated the importance of macropore diffusion resistance on the etherification rate, since Dowex DR-2030 had a higher porosity than Amberlyst-36. Results also proved that increase of catalyst amount and increase of reaction temperature to 140 °C caused significant increase in etherification rates of glycerol with i-amylenes, giving quite high di- and tri-ether selectivity values. Di-ether and tri-ether selectivity values, approaching 0.7 and 0.1 respectively, were highly promising.
[1] K.C. Tokay, T. Dogu, G. Dogu, Dimethyl ether synthesis over alumina based catalysts, Chem. Eng. J. 184 (2012) 278–285. [2] D. Varisli, T. Dogu, G. Dogu, Petrochemicals from ethanol over a W–Si-based nanocomposite bidisperse solid acid catalyst, Chem. Eng. Sci. 65 (2010) 153–159. [3] F. Frusteri, F. Arena, G. Bonura, C. Cannilla, L. Spadaro, O. Di Blasi, Catalytic etherification of glycerol by tert-butyl alcohol to produce oxygenated additives for diesel fuel, Appl. Catal. A Gen. 367 (2009) 77–83. [4] J.A. Melero, G. Vicente, G. Morales, M. Paniagua, J. Bustamante, Oxygenated compounds derived from glycerol for biodiesel formulation: influence on EN 14214 quality parameters, Fuel 89 (2010) 2011–2018. [5] J.A. Melero, G. Vicente, G. Morales, M. Paniagua, J.M. Moreno, R. Roldan, A. Ezquerro, C. Perez, Acid-catalyzed etherification of bio-glycerol and isobutylene over sulfonic mesostructured silicas, Appl. Catal. A Gen. 346 (2008) 44–51. [6] L. Xiao, J. Mao, J. Zhou, X. Guo, S. Zhang, Enhanced performance of HY zeolites by acid wash for glycerol etherification with isobutene, Appl. Catal. A Gen. 393 (2011) 88–95. [7] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Conversion of biodiesel by-product glycerol to fuel ethers over different solid acid catalysts, Int. J. Chem. React. Eng. 8 (2010) A18. [8] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Effects of sorption enhancement and isobutene formation on etherification of glycerol with tert-butyl alcohol in a flow reactor, Ind. Eng. Chem. Res. 51 (2012) 8788–8895. [9] N. Ozbay, N. Oktar, G. Dogu, T. Dogu, Activity comparison of different solid acid catalysts in etherification of glycerol with tert-butyl alcohol in flow and batch reactors, Top. Catal. 56 (2013) 1790–1803. [10] J.F. Izquierdo, M. Montiel, I. Pales, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, M. Izquierdo, M.P. Hermo, X. Ariza, Fuel additives from glycerol etherification with light olefins: State of the art, Renew. Sust. Energ. Rev. 16 (2012) 6717–6724. [11] W. Kiatkittipong, P. Intaracharoen, N. Laosiripojana, C. Chaisuk, P. Praserthdam, S. Assabumrungrat, Glycerol ethers synthesis from glycerol etherification with tertbutyl alcohol in reactive distillation, Comput. Chem. Eng. 35 (2011) 2034–2043. [12] H. Jong Lee, D. Seung, K.S. Jung, H. Kim, I.N. Filimonov, Etherification of glycerol by isobutylene: tuning the product composition, Appl. Catal. A Gen. 390 (2010) 235–244. [13] M.D. Gonzalez, Y. Cesteros, P. Salagre, Establishing the role of Brønsted acidity and porosity for the catalytic etherification of glycerol with tert-butanol by modifying zeolites, Appl. Catal. A Gen. 450 (2013) 178–188. [14] K. Klepacova, D. Mravec, M. Bajus, tert-Butylation of glycerol catalysed by ionexchange resins, Appl. Catal. A Gen. 294 (2005) 141–147. [15] R.S. Karinen, A.O.I. Krause, New bio components from glycerol, Appl. Catal. A Gen. 306 (2006) 128–133. [16] K. Klepacova, D. Mravec, A. Kaszonyi, M. Bajus, Etherification of glycerol and ethylene glycol by isobutylene, Appl. Catal. A Gen. 328 (2007) 1–13. [17] N. Oktar, K. Murtezaoglu, G. Dogu, l. Gonderten, T. Dogu, Etherification rates of 2methyl-2-butene and 2-methyl-1-butene with ethanol for environmentally clean gasoline production, J. Chem. Technol. Biotechnol. 74 (1999) 155–161. [18] Z. Obali, T. Dogu, Activated carbon–tungstophosphoric acid catalysts for the synthesis of tert-amyl ethyl ether (TAEE), Chem. Eng. J. 138 (2008) 548–555. [19] R. Soto, C. Fité, E. Ramírez, R. Bringué, F. Cunill, Equilibrium of the simultaneous etherification of isobutene and isoamylenes with ethanol in liquid-phase, Chem. Eng. Res. Des. 92 (2014) 644–656. [20] R. Soto, C. Fité, E. Ramírez, R. Bringué, M. Iborra, Green metrics analysis applied to the simultaneous liquid-phase etherification of isobutene and isoamylenes with ethanol over Amberlyst™ 35, Green Processes Synth. 3 (2014) 321–333. [21] J.F. Izquierdo, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, X. Ariza, New biodiesel additives from glycerol and isoamylenes, Biofuels Bioprod. Biorefin. 8 (2014) 658–669. [22] J.F. Izquierdo, P.R. Outon, M. Galan, L. Jutglar, M. Villarrubia, M.P. Hermo, X. Ariza, I. Fernandez, Ethers of glycerol and isoamylenes as biodiesel additives: synthesis and characterization, Chem. Eng. Trans. 32 (2013) 877–882. [23] D. Varisli, T. Dogu, G. Dogu, Novel mesoporous nanocomposite WOx-silicate acidic catalysts: ethylene and diethylether from ethanol, Ind. Eng. Chem. Res. 48 (2009) 9394–9401. [24] N. Oktar, K. Murtezaoglu, T. Dogu, G. Dogu, Dynamic analysis of adsorption equilibrium and rate parameters of reactants and products in MTBE, ETBE and TAME production, Can. J. Chem. Eng. 77 (1999) 406–412. [25] N. Boz, T. Dogu, K. Murtezaoglu, G. Dogu, Mechanism of TAME and TAEE synthesis from diffuse–reflectance FTIR analysis, Catal. Today 100 (2005) 419–424.
Acknowledgments The financial support of Scientific and Technical Research Council of Turkey (TUBITAK-112M234) and Middle East Technical University
Please cite this article as: B. Ikizer, et al., Etherification of glycerol with C4 and C5 reactive olefins, Fuel Processing Technology (2015), http:// dx.doi.org/10.1016/j.fuproc.2015.06.039