A R T I C L E
Efficient manganese decorated cobalt based catalysts
for hydrogenation of 5-hydroxymethylfurfural (HMF)
to 2,5-dimethylfuran (DMF) biofuel
Solmaz Akmaz | Merve Esen | Esra Sezgin | Serkan Naci Koc
Department of Chemical Engineering, Istanbul University-Cerrahpas¸a, Istanbul, Turkey
Correspondence
Solmaz Akmaz, Department of Chemical Engineering, Istanbul University-Cerrahpas¸a, 34320 Istanbul, Turkey. Email: solmaz@istanbul.edu.tr Funding information
Scientific Research Projects Coordination Unit of Istanbul University, Grant/Award Number: FYL-2017-24265; The Scientific and Technological Research Council of Turkey (TUBITAK), Grant/Award Number: 214M149
Abstract
2,5-dimethylfuran (DMF) is a promising compound in the production of biofuel with high-quality properties. In this study, it is aimed to develop new efficient cata-lysts to synthesize DMF from 5-hydroxymethylfurfural (HMF). Co, Mn/Co, and Ru/Co catalysts were prepared using the NaBH4reduction method. The catalysts were subjected to activity tests for the hydrogenation of HMF to DMF by changing the reaction parameters, such as temperature and time. Mn/Co catalysts prepared from metal precursors at various molar ratios of Mn/Co were found to be effective in hydrogenation reactions of HMF to DMF. A 91.8% DMF yield was achieved in the presence of a Mn/Co (50/50) catalyst without noble metal at 180C for 4 hours. The Brunauer-Emmet-Teller (BET) method, x-ray diffraction (XRD), x-ray photo-electron spectroscopy (XPS), and induction coupled plasma mass spectroscopy (ICP-MS) techniques were used to characterize the efficient Mn/Co catalyst.
K E Y W O R D S
biofuel, catalytic hydrogenation, DMF, HMF, hydrogenolysis
1
| I N T R O D U C T I O N
The world’s increasing need for vehicle fuels and limited petroleum resources have created the need for new alternative biofuels. Bioethanol and biodiesel are among the most known renewable biofuels.[1] These fuels are mainly produced from sources such as sugar, starch, and vegetable oil, which can be also used as foods. Cellulosic materials, the most renewable resource on Earth, have the potential to produce alternative chemicals and fuels. 5-hydroxymethylfurfural (HMF), which can be produced from cellulose and other carbohydrates, is an important intermediate product for the synthesis of alternative fuels.[2–4] 2,5-dimethylfuran (DMF), a non-toxic fuel or fuel additive with a high energy density and the high octane num-ber (RON 119), is suitable for modern automobile engines and can be produced by the catalytic hydrogenation of HMF.[5,6]
The volumetric energy density (31.5 MJ/L) of DMF is much closer to that of gasoline when compared to bioethanol, which has an energy density of 23 MJ/L. The water-insoluble prop-erty of DMF is also advantageous in terms of preventing mois-ture absorption from the air. Another advantage of DMF is that it can be stored for a long time due to its chemical stabil-ity.[7,8] Moreover, DMF has the potential to be an alternative fuel, and the use of DMF in vehicles with gasoline does not require any changes in the vehicle engine.[9,10] Besides the possible use of DMF as an alternative fuel or fuel additive, DMF can also be converted to valuable chemicals such as p-xylene through Diels-Alder reactions.[11–13]
DMF is synthesized by the catalytic hydrogenation of HMF. As seen in Scheme 1, according to the possible mech-anism, the synthesis of DMF takes place in two steps. HMF is catalytically hydrogenated to bis (hydroxymethyl) furan
DOI: 10.1002/cjce.23613
(BHMF), which is converted into 2-hydroxymethyl-5-methylfuran (MFA), and then to the desired DMF by hydrogenolysis. During the first hydrogenation step, hydrogenolysis can also take place to form 5-methylfurfural (MF), which can be converted into MFA, and then to DMF with the advanced hydrogenation/hydrogenolysis route.[14,15] Product selectivity of the hydrogenation reaction depends on factors such as the solvent, partial pressure of hydrogen, temperature and the structure of the catalyst.
The first study on DMF production from biomass-derived carbohydrate sources was published in 2007, and HMF, which was produced from fructose by Román-Leshkov et al[16]was converted to DMF in 1-butanol solvent at 220C for 10 hours with CuRu/C catalyst under 0.68 MPa hydrogen pressure. In their study, 100% HMF conversion was achieved with 71% DMF selectivity. Then, Binder and Raines[17]obtained a 32.5% DMF yield from HMF, which is synthesized from fructose by using LiCl and H2SO4, with 0.62 MPa of hydrogen at 220C for 10 hours in 1-butanol using a Cu-Ru/C catalyst. When stud-ies on the DMF synthesis are examined, it can be seen that mainly Ru-based catalysts are used and high DMF yields are obtained with Ru-containing catalysts. Zu et al[14] used Ru/Al2O3, Ru/ZSM-5, and Ru/Co3O4 catalysts for HMF hydrogenation reactions. They obtained 7% and 5.4% DMF yields at 130C under 0.7 MPa H2pressure for 24 hours with Ru/ZSM-5 and Ru/Al2O3, respectively. However, they reported that a 93.4% DMF yield was obtained with Ru/Co3O4at the same reaction conditions.[14] Hu et al[18] obtained a 94.9% DMF yield with 100% HMF conversion using Ru/C catalyst at 200C and under 2 MPa hydrogen pressure. Nishimura et al[19] achieved a very high efficiency with 99% HMF conversion and 99% DMF selectivity at 60C with Pd/C and PdAu/C catalysts in the presence of HCl. Thananatthanachon and Rauchfuss[20] transformed fructose to DMF with a 51% yield and they syn-thesized 95% DMF yield from HMF in 15 hours with Pd/C cat-alyst in the presence of HCOOH, H2SO4, and THF.
Noble metals (Pt, Pd, Ru, etc) have been generally used for HMF hydrogenation because of their high activity on the hydrogenation reactions. Metal-based bifunctional catalysts were also prepared using metal precursors such as Ni, Co, Cu, Zr, Fe salts, and noble metals such as Ru, Rh, Pd, and Pt and tested on DMF production. Iriondo et al[21]prepared Pt, Ru, Ni, and Cu on acidic supports such as HY zeolite/Al2O3 and Al2O3and on basic supports such as ZrO2and TiO2to
convert the 5-hydroxymethylfurfural (HMF) to DMF. The maximum 60% DMF selectivity was obtained in the fixed bed reactor at 200C under 1.5 MPa H2 atmosphere after 2 hours with Cu/ZrO2catalyst.[21]However, due to the high cost of these noble based catalysts, many researchers have aimed to produce high yield DMF using non-noble metal cat-alysts and different support materials.[22] Metals such as Co, Fe, Cu, and Ni have been tested for DMF production.[15,23,24] The results obtained from the studies show that bifunctional catalysts are suitable for DMF production. Yang et al[15] stud-ied hydrogenation reactions of HMF to DMF using Ni/Co3O4 catalyst and they achieved 76% DMF yield at 130C under 1.0 MPa pressure after 24 hours. Chen et al[24]used a carbon coated Cu-Co bimetallic nanoparticle catalyst for the hydroge-nation of HMF to DMF. Researchers reported that the bime-tallic structure of the Co-Cu catalysts showed higher activity due to the synergistic effect of the two metals. A DMF yield up to 99.4% was reached in the presence of ethanol solvent at 180C for 8 hours under 5 MPa hydrogen pressure.[24] In 2017, Srivastava et al[25] investigated the effect of Cu-Co bimetallic catalysts supported on CeO2, ZrO2, and Al2O3for DMF synthesis from HMF. 99.9% HMF conversion and 68.4% DMF selectivity were obtained using Cu-Co/Al2O3 catalyst at 200C under hydrogen pressure of 3 MPa after 8 hours. New types of catalysts are also investigated to improve the reaction conditions such as mild hydrogen pres-sure and lower reaction temperatures.
On the basis of the mentioned studies, it is observed that Co-based catalysts are effective for the hydrogenation of HMF and the use of bimetallic catalysts is more efficient in the synthesis of DMF from HMF. When the literature is examined, it is noticed that manganese, a transition-metal, is not tested in DMF production studies. This study aims to achieve a high yield of DMF using a Mn/Co efficient cata-lyst as an alternative to noble metal-based catacata-lysts in rela-tively mild reaction conditions. The effects of temperature and reaction time on the DMF yield were also investigated.
2
| M A T E R I A L S A N D M E T H O D S
2.1
| Materials
HMF (≥99%) and tetrahydrofuran (THF, HPLC grade) were purchased from Sigma-Aldrich. Cobalt (II) acetate tetra
S C H E M E 1 Reaction route of HMF hydrogenation to DMF[14,15]
hydrate (98%) and manganese (II) acetate (98%), provided by Sigma-Aldrich, was used for catalyst preparation. 2,5-dimethylfuran (DMF, Sigma-Aldrich,≥99%) as standard and naphthalene (Merck, 99%) was used as the external stan-dard for the gas chromatography-mass spectrometry (GC-MS) analysis.
2.2
| Preparation of catalysts
Catalysts were prepared using the NaBH4reduction method. The preparation of the Co catalyst was as follows. First of all, cobalt (II) acetate tetra hydrate (C4H8CoO44H2O) was dissolved in deionized water to prepare solution A. Solution B was prepared by dissolving of sodium borohydride (NaBH4) in 10 mL of ethanol. Solution B was added dropwise to solution A under stirring. The colour change in the solution confirmed the reduction to metallic Co, which is metallic gray. After reduction with sodium borohydride, the solution was stirred at ~ 23C for 4 hours, filtrated by centrifuging, and the precipitate was thoroughly washed sev-eral times with distilled water. Subsequently, the catalyst was dried at 110C in nitrogen atmosphere. To prepare the Mn/Co catalyst with ratios of 10/100 (mol/mol), 0.03 mol of cobalt (II) acetate tetra hydrate (C4H8CoO44H2O), and 0.003 mol of manganese (II) acetate (Mn[CH3COO]2) were dissolved in 100 mL of ultrapure water, and the solution was stirred for 30 minutes. 0.7 g of NaBH4 was dissolved in 10 mL of ethanol, and then a sodium borohydride solution was added dropwise to the first solution. The mixture was stirred for 4 hours at ~ 23C. After filtration and washing, drying was carried out at 110C in nitrogen atmosphere.
Mn/Co catalysts (Mn/Co) were prepared from manga-nese (II) acetate and cobalt (II) acetate tetra hydrate solu-tions with different molar ratios of 15/100, 50/50, and 70/30 (mol/mol) using same procedure and labelled with these molar ratios.
For the preparation of the Ru/Co (4 g/g Ru) catalyst, solution A was prepared by dissolving 0.005 mol of cobalt (II) acetate tetra hydrate (C4H8CoO44H2O) and 0.09 g of RuCl3 in 65 mL of water. 0.57 g of NaOH and 0.74 g of Na2CO3in deionized water were added dropwise by stirring to the solution A with a pH value of 10.7-11.2.[14]After the solution was stirred at 80C for 24 hours, it was cooled to ~ 23C and centrifuged. The filtrate was washed with ultra-pure water until the pH reached 7. NaBH4, which is 10 times the amount of molar Ru, was added with ethanol dropwise to the catalyst, and a colour change was observed. Stirring was continued for 2 hours. The catalyst was centrifuged, washed with deionized water, and finally dried at 110C under nitrogen atmosphere.
2.3
| Catalyst characterization
The structural properties of the catalysts were determined by inductive coupling plasma-mass spectrometry (ICP-MS), Brunauer-Emmet-Teller (BET) analysis, x-ray diffraction (XRD) method, and x-ray photoelectron spectroscopy (XPS). The surface area measurement of the catalysts was carried out at −196C in a Quantachrome Nova 3200e BET unit after being pre-treated under high vacuum at 300C. The XRD pat-terns were recorded with Rigaku D / Max-2200 / PC XRD. Samples were measured with a Cu Kα radiation (1.5404 × 10−10m) between 2θ 3-90. The surface structures of the catalysts were investigated by thermo scientific k-alpha XPS. The binding energies are referenced to the C1S line. The Mn, Co, and B contents of the catalyst were measured with a Thermo Elemental X Series 2 ICP-MS.
2.4
| Catalytic activity
Hydrogenation reactions of the HMF were carried out in 100 mL stainless steel Parr micro reactor system. The reactor was charged with 0.5 g of HMF, 0.2 g of catalyst, 23 mL of tetrahydrofuran (THF), and 0.013 g of naphthalene. Naph-thalene was used as external standard to identify the prod-ucts quantitatively. Prior to the reaction, the reactor was purged five times with H2to avoid air in the environment. Subsequently, reactions were carried out at 130C-200C under H2 pressure with stirring at 450-470 rpm. After the desired reaction time was completed, the reactor was cooled to ~23C and the solution was centrifuged to separate the catalyst. The products were analyzed by an Agilent Technol-ogies 7890A series gas chromatography and Agilent Tech-nologies 5975C series mass spectroscopy (GC-MS) system. A capillary column HP-5MS (30 m × 0.25 mm × 0.25μm) was used with a 116.81 mL/min He (carrier gas) flow rate. The GC oven was held at a controlled temperature of 303 K for 1 minute and then increased to 358 K with a temperature gradient of 10 K/min. This temperature was held for 2 minutes and then increased to 423 K with a temperature gradient of 10 K/min and hold for 10 minutes. Calibration curves were created using HMF and DMF standard solutions with different concentrations to identify products quantita-tively. The HMF conversion and DMF yield were calculated using following Equations (1) and (2), respectively:
HMF conversionð Þ =% consumed HMF moleð Þ
initial HMF moleð Þ × 100 ð1Þ
DMF yieldð Þ =% DMF moleð Þ
3
| R E S U L T S A N D D I S C U S S I O N
3.1
| Catalyst characterization
The BET surface area of the Mn/Co (50/50) catalyst with high activity in the hydrogenation reactions to obtain DMF from HMF is 5.48 m2/g. The XRD patterns of the Co, Mn/Co (10/100), Mn/Co (15/100), Mn/Co (50/50), Mn/Co (70/30), and Ru/Co catalysts are shown in Figures 1–3.
In the XRD pattern of the Co catalyst seen in Figure 1, the peaks at 2θ angle of 25 (JCPDS card no: 250241) and the peak at 2θ angle of 45 (JCPDS card no: 30959) can be assigned to Co2B and CoB, which can be formed by co-alloying cobalt and boron. Since the catalysts were prepared using the NaBH4reduction method, boron can participate in the catalyst structure by alloying with cobalt as reported in various studies.[26–29] The peaks at 2θ angles of 45, 51.7,
and 75.8 represent Co (0) (JCPDS card no: 150806). The peaks at 2θ angles of 20.2 and 30.5 may belong to Co(OH)2(JCPDS card no: 30-0443). In the XRD pattern of the Mn/Co (15/100), Mn/Co (50/50), and Mn/Co (70/30) cat-alysts shown in Figure 2, the peaks at 2θ angles of 44, 51.7, and 75.8 2θ correspond to Co (0) (JCPDS card no: 150806). In all of the catalysts, the peak at 2θ angle of ~42 value can indicate Co2B or Co3B structures (JCPDS card no: 250241, JCPDS card no: 120443). This peak may also be attributed to CoO (JCPDS card no: 780431), which is observed in the XPS analysis. Additionally, the peak at 2θ angle of 47 can be assigned to Co3B (JCPDS card no: 120443). The peak at 2θ angle of the 28 peak can represent the CoB alloy (JCPDS card no: 30959) for the XRD pattern of the Mn/Co (15/100). No peak representing crystalline manganese structures was observed. The XRD pattern of Ru/Co catalyst is shown in Figure 3. The peaks at 2θ angles of 31.4, 36.5, 59.4, and 65.2indicated Co3O4(JCPDS card no: 090418). The peak at 2θ angle of 38.6 and the peak at 2θ angle of 44 can be assigned to Ru (0) (JCPDS card no: 894903) and Co (0) (JCPDS card no: 150806), respectively. It is worth noting that the main peak of cobalt in the Ru/Co catalyst is Co3O4; how-ever, no peak corresponding to Co3O4phase was observed in the XRD patterns of the Co and Co/Mn catalysts.
The structures, metal ratio, and relationships of the Mn/Co (50/50) catalyst with highest activity were determined by XPS. The Co2p3/2spectrums of the XPS analysis of the fresh Mn/Co (50/50) catalyst and Mn/Co (50/50) catalyst after the fourth use for cobalt are shown in Figure 4A and B.
The peak of Co2p3/2 at 781.0 eV (1.249 × 10−16 J) belongs to Co [2+], which is provided by the satellite peak at 786 eV (1.257 × 10−16 J). It can be thought that Co (2+) peak represents CoO, which can be formed due to air oxida-tion during preparaoxida-tion.[29–32] The peak of Co2p3/2 at 782.9 eV (1.252 × 10−16 J) can be attributed to Co (2+) in
F I G U R E 1 XRD pattern of the Co catalyst
F I G U R E 2 XRD patterns of the Mn/Co (10/100), Mn/Co (15/100), Mn (50/50), and Mn/Co (70/30) catalysts
Co(OH)2.[33,34] As reported in the literature,[34,35] Co (3+) peak appears at 779.2 eV −779.4 eV (1.246 × 10−16 J– 1.247 × 10−16 J). There is a small peak at 778.6 eV (1.245 × 10−16 J) in the fresh Co/Mn catalyst that may be attributed to Co (3+), and it was disappeared after the fourth use. The peak of Co2p3/2 at 777.8 eV (1.244 × 10−16J) can be attributed to Co (0). It is seen that the Co (2+) peak becomes more considerable for the XPS spectrum of Mn/Co (50/50) catalyst after the fourth use in Figure 4B. The surface ratio of Co (2+)/Co (0) for the fresh Co/Mn (50/50) catalyst and the catalyst after the fourth use was calculated as 1.61 and 5.81, respectively, as determined by the XPS spectrum.
Although no crystalline manganese structures were observed with XRD, Mn (2+) and Mn (3+) were determined by the XPS. The peak of Mn 2p3/2at ~638.6 eV (1.021 × 10−16J) belongs to Manganese metal (Mn (0)) and was not observed for the XPS spectrum of the Mn/Co (50/50) catalyst in Figure 5A.[32,36,37]It is thought that the satellite peak at 648 eV (1.036 × 10−16J) may be due to Mn (2+). Although the Mn (2 +
) and Mn (3+) peaks are difficult to differentiate, it is thought
that the Mn (2+) satellite peak and Mn (+3) satellite peak may be present in the catalyst after the fourth use of the catalyst according to Figure 5B. Therefore, it is concluded that 640.3 eV (1.024 × 10−16J) is the Mn (2+) peak and 641.8 eV (1.026 × 10−16J) is the Mn (3+) peaks.[36–38]
According to the results of the ICP-MS analysis of the Mn/Co (50/50) catalyst, which is the most effective catalyst in the DMF synthesis, the Mn, Co, and B contents of catalyst were 0.65, 94.06, and 2.27 g/g, respectively. It is understood that boron participates with the catalyst structures during the reduction with NaBH4 in the catalyst structure. It is also understood that there is a small amount of manganese in the catalyst structure after the catalyst preparation process.
3.2
| Catalytic activity
3.2.1
| DMF synthesis from HMF with the Co
catalyst
The amount of the reducing agent was determined by using different amounts of NaBH4 in the preparation of the Co
(A) (B)
F I G U R E 4 A, Cobalt (Co 2p) XPS result of the fresh Mn/Co (50/50) catalyst; and B, Cobalt (Co 2p3/2) XPS result of the Mn/Co (50/50)
catalyst after the fourth use
(A) (B)
F I G U R E 5 A, Manganese (Mn 2p) XPS result of the fresh Mn/Co (50/50) catalyst; and B, manganese (Mn 2p) XPS result of the Mn/Co (50/50) catalyst after the fourth use
catalysts. The catalysts were prepared in 0.25:1, 0.5:1, and 1:1 M ratios of the metal precursor cobalt (II) acetate ([CH3COO]2Co.4H2O) to NaBH4, respectively. The cata-lysts were tested at 150C for 6 hours under 1 MPa H2initial reaction pressure conditions. DMF yield (%) and HMF con-version (%) of the reactions are given in Table 1. The highest 49.7% DMF yield was achieved by the catalyst with a molar ratio of 1:1. As a result, the effect of the temperature on the DMF yield using the cobalt catalyst prepared with a ([CH3COO]2Co.4H2O)/NaBH4molar ratio of 1:1 was inves-tigated. Hydrogenation reactions were carried out with the Co catalyst at temperatures of 130, 150, 180, and 200C for 6 hours using THF solvent under 1 MPa of H2initial pres-sure. The yields of the DMF and HMF conversions are shown in Figure 6.
While the yield of the DMF obtained at 130C was 32%, the yield increased to 49.7% at 150C and the HMF conver-sion also increased. The maximum DMF yield was reached at 150C, then a significant decrease in the yield of DMF was observed, and the yield of DMF at 180C was 14.2%. DMF was not formed at 200C and a significant decrease in HMF conversion was also observed (46.1%). It can be said that high reaction temperatures, such as 200C, negatively affect the hydrogenation reaction mechanism in the presence of the Co catalyst. Zu et al[14] obtained a 3.2% DMF yield and 54.9% HMF conversion with the Co3O4 catalyst at 130C, under 0.7 MPa H2pressure in 10 mL of THF for 24 hours.
3.2.2
| DMF synthesis from HMF with the
Mn/Co catalyst
Mn/Co catalysts were prepared in different molar ratios of manganese salt to cobalt salt (10/100, 15/100, 50/50, 30/70) to improve the DMF yields obtained as a result of hydroge-nation reactions using an effective Co catalyst. The activity of the catalysts was tested under 1 MPa of H2initial pressure at 150C for 6 hours and the percentages of DMF yields and HMF conversions are presented in Table 2.
Hydrogenation reactions with Mn/Co catalysts give higher percentages of DMF yields and HMF conversions compared to the reactions using the Co catalyst. It was
observed that the increase in the amount of Mn in the Mn/Co catalysts also has a positive effect on the DMF yield. The highest DMF yield of 74.3% was achieved in the presence of the Mn/Co (50/50) catalyst. It was observed that the DMF yield decreased using the Mn/Co (70/30) catalyst.
Accordingly, the reactions were carried out at 140, 160, 180, and 200C for 6 hours under 1.5 MPa H2initial reac-tion pressure to investigate the effect of temperature on the hydrogenation reaction of HMF using the Mn/Co (50/50) catalyst. The DMF yields and HMF conversions are shown in Figure 7. The DMF yield increased as the reaction tem-perature increased according to Figure 7. The DMF yield reached to the maximum yield of 89.1% at 180C and then decreased to 74.4% at 200C. The percentage of HMF con-version reached 100% for each temperature condition. Although the percentage of HMF conversion did not decrease, a decrease in the percentage of DMF yield indi-cates the formation of a by-product at 200C.
The reactions were carried out for holding times ranging from 1 to 6 hours at 180C, which was the most effective reaction temperature using the Mn/Co (50/50) catalyst, to investigate the effect of reaction time on DMF yield. The percentages of DMF yields and HMF conversions are given in Figure 8.
As seen in Figure 8, both the DMF yield and the HMF conversion are positively affected by the increased reaction times. Although the HMF conversion in the first hour reached 90%, the DMF yield remained at 50%, indicating that HMF was converted into by-products other than DMF. After 2 hours, the yield and selectivity of DMF increased and the reaction accelerated. The highest DMF yield was
T A B L E 1 The effect of the amount of NaBH4in the preparation
of the Co catalysts
(CH3COO)2Co4H2O/NaBH4molar ratio
1 0.5 0.25 DMF yield (%) 49.7 35.9
HMF conversion (%) 78.4 90.6 22.3
Reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g; naphthalene: 0.013 g; 150C; 6 h; and 1 MPa H2.
F I G U R E 6 The effect of temperature on the hydrogenation reactions of HMF to DMF using the Co catalyst (reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g; naphthalene: 0.013 g; 6 h; and 1 MPa H2)
obtained at a reaction time of 4 hours with a value of 91.8%. A small drop in the DMF yield occurred in the reaction time of 6 hours. Chen et al[24]used a carbon-coated Cu-Co bime-tallic nanoparticle catalyst for the hydrogenation of HMF to 2,5-Dimethylfuran (DMF) and obtained DMF yield up to 99.4% at 180C for 8 hours under 5 MPa hydrogen pressure. However, in this study, a 91.8% DMF yield was obtained using Mn/Co (50/50) under relatively moderate reaction con-ditions, especially in terms of hydrogen pressure.
The reusable properties of the Mn/Co (50/50) catalyst, which is the most effective in the conversion of HMF to DMF, were tested four times under 1.5 MPa H2 initial pressure at 180C for 2 hours. After the reactions, the catalyst was removed by filtration from the reaction medium and washed several times with THF and ethanol. After washing, the catalyst was dried in a vacuum oven at 110C.
As shown in Figure 9, the yield of 54.9% DMF obtained after the first use was reduced after each repeat-ability test and the final yield was obtained with a DMF yield of 12.7%. According to the XPS analysis results of the Co structures in Mn/Co (50/50) catalyst, the Co (2+)
peak in the catalyst is more pronounced than Co (0) after the fourth use of the catalysts. The surface ratio of Co (2+)/Co (0) in the catalyst structure increased from 1.61 to 5.81 after the fourth use of the catalyst in the reactions. It is thought that the decrease in activity of the catalyst after re-use may be due to the oxidation of the Co metal in the catalyst structure.[24]
In summary, as revealed by the XRD analysis, the metal-lic manganese and also manganese oxides and hydroxides crystal phases were not found in the Mn/Co catalysts, How-ever, Mn (2+) and Mn (3+) were observed by the XPS spec-trum. The manganese ratio was calculated as 0.65 g/g by ICP-MS in the Mn/Co (50/50) catalyst. However, the surface Mn/Co ratio was calculated as 6.47 by the XPS spectrum. It is clearly seen that manganese decorated the cobalt catalyst surface during the catalyst preparation. Since the lower DMF yield was obtained with the cobalt catalyst without manganese, it can be thought that the surface manganese structures have a synergetic effect with cobalt in this reac-tion. It has been proposed that the Co metal is responsible for the hydrogen adsorption, and then hydrogen spillover to oxygen vacancy of CoOx sites (due to Co [2+]) occurs.
T A B L E 2 DMF yields and HMF conversions using the Mn/Co catalysts prepared in different ratios Catalyst
Co Mn/Co (10/100) Mn/Co (15/100) Mn/Co (50/50) Mn/Co (70/30) DMF yield (%) 49.7 58.4 61.4 74.3 40.4
HMF conversion (%) 78.4 100 100 93.0 89.0
Reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g; naphthalene: 0.013 g; 150C; 6 h; and 1 MPa H2.
F I G U R E 7 The effect of different reaction temperatures on the DMF yield and HMF conversion using the Mn/Co (50/50) catalyst (reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g; naphthalene: 0.013 g; 6 hours; and 1.5 MPa H2)
F I G U R E 8 Variation in the DMF yields and HMF conversions with different reaction times at 180C using the Mn/Co (50/50) catalyst (reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g;
These oxygen vacancies adsorb the C = O bonds of HMF, and the deoxygenation of HMF occurs by releasing H2O on these oxygen vacant sites.[25] Due to the complex reaction nature of the NaBH4reduction method, it is very difficult to identify surface manganese structures in the Co/Mn catalyst. Manganese would be in the form of amorphous oxides, hydroxides, and oxyhydroxides. On the other hand, the DMF yield was highly enhanced with the incorporation of manga-nese on the surface of Co/Mn catalyst. Thus, it is thought that the presence of Mn (2+), Co (2+), and Mn (3+) on the Co/Mn catalyst surface may lead to electron transfer between cobalt and manganese domains and more oxygen vacancy forma-tion, which is responsible for the hydrodeoxygenation step on the surface. As reported in the study on XANES,[39] substitut-ing manganese increases the Co(+2)/Co(3+)ratio. This is con-sistent with our XRD and XPS data. Although a crystalline Co3O4phase was not observed in XRD spectrum, the small XPS peak at 778.6 eV (1.245 × 10−16J) in the fresh Co/Mn catalyst that can be attributed to Co (3+) disappeared after the fourth catalyst use. Moreover, the surface Mn (3+)/Mn (2+) ratio calculated from the deconvoluted XPS peaks increased from 1.67 to 2.66 after the fourth catalyst use.
3.2.3
| DMF synthesis from HMF with Ru/Co
catalyst
The activity of a noble metal with a Co catalyst in the hydroge-nation reaction of HMF to DMF was investigated to compare to the activity of the Mn/Co catalyst. For this purpose, ruthenium (Ru), one of the most effective hydrogenation catalysts, was cho-sen as the noble metal, and the Ru/Co catalyst was synthesized. Hydrogenation reactions were carried out under 1.5 MPa H2 pressure for 3 hours at 180C in THF solvent. The percentages of DMF yields and HMF conversions are given in Table 3.
When the Co, Mn/Co (50/50), and Ru/Co catalysts were compared, 42.4, 87.8, and 75.5% DMF yields were observed
in 3 hours, respectively. This means that the cheaper Mn/Co catalyst showed a better DMF yield than Ru/Co in this study. Zu et al[14] reported a 93.4% DMF yield at 130C under 0.7 MPa H2for 24 hours with Ru/Co3O4, which was reduced with hydrogen. Although a 93.4% DMF yield for 24 hours of reaction time has been reported with noble metal containing Ru/Co3O4in their study, 87.8% and 91.8% DMF yields were achieved with a non-noble metal containing Mn/Co catalyst in a shorter reaction time, such as 3 and 4 hours, respectively. Ru metal is effective for hydrogena-tion, and the CoOxstructures are effective for the adsorption of hydrogenation products and the subsequent breakdown of the C-O bonds during the hydrogenation reactions of HMF.[14,40,41]
In summary, although it is known that Ru is an effective noble catalyst for hydrogenation reactions, higher DMF yields were achieved with the cost effective Mn/Co (50/50) catalyst than that in the presence of Ru/Co in this study. This means that this reaction does not need a noble metal catalyst for hydrogenation. Although no peak representing crystal-line manganese phases was observed in the XRD patterns of the catalysts, Mn2+ and Mn3+ forms were detected by the XPS spectrum and the manganese ratio was calculated as 0.65 g/g by ICP-MS in the Mn/Co (50/50) catalyst. A possi-ble reason of the increase in the catalytic activity of the Mn/Co (50/50) catalyst is that Mn2+ and Mn3+forms deco-rate the Co catalyst surface and the surface manganese forms with cobalt synergistically promote DMF formation reactions.
4
| C O N C L U S I O N S
High DMF yields from HMF were obtained with the non-noble Co catalyst without the use of expensive non-noble metals such as Ru. In order to improve the DMF yield obtained in the presence of the Co catalyst, Mn/Co catalysts were pre-pared in different ratios of manganese salt to cobalt salt. The effect of the increase in the Mn content of the catalyst on the hydrogenation reaction was investigated. The highest DMF yield of 74% with 89% HMF conversion was obtained using Mn/Co (50/50) at 150C for 6 hours under 1 MPa H2initial pressure. To improve the DMF yield obtained in the
F I G U R E 9 Reusability of the Mn/Co (50/50) catalyst
T A B L E 3 DMF yields and HMF conversions using the Co, Mn/Co (50/50), and Ru/Co catalysts
Catalyst
Co Mn/Co (50/50) Ru/Co DMF yield (%) 42.4 87.8 75.5 HMF conversion (%) 91.1 100 100
Reaction conditions: catalyst: 0.2 g; HMF: 0.5 g; THF: 23 g; naphthalene: 0.013 g; 180C; 3 hours; and 1.5 MPa H2.
presence of Mn/Co catalyst, experiments were carried out at different reaction temperatures (140C, 160C, 180C, and 200C) and different reaction times (1, 2, 3, 4, and 6 hours). A 91.8% DMF yield was achieved at 180C after 4 hours. In order to compare the high DMF yields obtained with the Mn/Co catalysts, a Ru/Co catalyst was also prepared to investigate the effect of a noble metal with Co. Ru metal also had a positive effect with Co on the DMF yield percentage. Although Ru is an effective noble catalyst for hydrogenation reactions, higher DMF yields were achieved in the presence of the cost effective Mn/Co (50/50) catalyst. As a result of this study, an effective Mn/Co catalyst, which does not con-tain expensive noble metal, was designed for the hydrogena-tion reachydrogena-tion of HMF to DMF. Further studies will be done with different preparation methods to increase the reusable properties of the Mn/Co catalysts, since reusability is impor-tant for the industrial application for this reaction.
A C K N O W L E D G E M E N T S
This work was supported by the Scientific Research Projects Coordination Unit of Istanbul University (project number: FYL-2017-24265) and the Scientific and Technological Research Council of Turkey (TUBITAK) (project no.: 214 M149).
R E F E R E N C E S
[1] K. A. Agarwal, Prog. Energ. Combust. 2007, 33, 233. [2] X. Tong, Y. Ma, Y. Li, Appl. Catal. A-Gen. 2010, 385, 1. [3] J. N. Chheda, Y. Roma´n-Leshkov, J. A. Dumesic, Green Chem.
2007, 342, 9.
[4] R.-J. van Putten, J. C. Van der Waal, E. de Jong, C. B. Rasrendra, H. J. Heeres, J. G. de Vries, Chem. Rev. 2013, 113, 1499. [5] S. Dutta, S. Pal, Biomass Bioenerg. 2014, 62, 182. [6] B. Saha, M. M. Abu-Omar, ChemSusChem 2015, 8, 1133. [7] Y. Qian, L. Zu, Y. Wang, X. Lu, Renew. Sust. Energ. Rev. 2015,
41, 633.
[8] S. Zhong, R. Daniel, H. Xu, J. Zhang, D. Turner, M. L. Wyszynski, P. Richards, Energ. Fuel. 2010, 24, 2891. [9] R. Daniel, G. Tian, H. M. Xu, M. L. Wyszynski, X. S. Wu, Fuel
2011, 90, 449.
[10] A. Malinowski, D. Wardzınska, Chemik 2012, 66, 982.
[11] J. J. Pacheco, J. A. Labinger, A. L. Sessions, M. E. Davis, ACS Catal. 2015, 5, 5904.
[12] L. Ni, J. Xin, H. Dong, X. Lu, X. Liu, S. Zhang, ChemSusChem 2017, 10, 2394.
[13] R. Zhao, Z. Zhao, S. Li, A.-N. Parvulescu, U. Meller, W. Zhang, ChemSusChem 2018, 11, 3803.
[14] Y. Zu, P. Yang, J. Wang, X. Liu, J. Ren, G. Lu, Y. Wang, Appl. Catal. B-Environ. 2014, 146, 244.
[15] P. Yang, Q. Cui, Y. Zu, X. Liu, G. Lu, Y. Wang, Catal. Commun. 2015, 66, 55.
[16] Y. Román-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic, Nature 2007, 447, 982.
[17] J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979. [18] L. Hu, X. Tang, J. Xu, Z. Wu, L. Lin, S. Liu, Ind. Eng. Chem.
Res. 2014, 53, 3056.
[19] S. Nishimura, N. Ikeda, K. Ebitani, Catal. Today 2014, 232, 89. [20] T. Thananatthanachon, T. B. Rauchfuss, Angew. Chem. Int. Edit.
2010, 49, 6616.
[21] A. Iriondo, A. Mendiguren, M. B. Güemez, J. Requies, J. F. Cambra, Catal. Today 2017, 279, 286.
[22] X. Kong, Y. Zhu, H. Zheng, Y. Zhu, Z. Fang, ACS Sustain. Chem. Eng. 2017, 5, 11280.
[23] P. Yang, Q. Xia, X. Liu, Y. Wang, Fuel 2017, 187, 159. [24] B. Chen, F. Li, Z. Huang, G. Yuan, Appl. Catal. B-Environ. 2017,
200, 192.
[25] S. Srivastava, G. C. Jadeja, J. Parikh, Chinese J. Catal. 2017, 38, 699. [26] G. N. Glavee, K. J. Klabunde, C. M. Sorensen,
G. C. Hadjipanayis, Langmuir 1993, 9, 162.
[27] O. V. Komova, V. I. Simagina, O. V. Netskina, D. G. Kellerman, A. V. Ishchenko, N. A. Rudina, Catal. Today 2008, 138, 260. [28] U. B. Demirci, P. Miele, Phys. Chem. Chem. Phys. 2010, 12,
14651.
[29] U. B. Demirci, P. Miele, Phys. Chem. Chem. Phys. 2014, 16, 6872.
[30] O. V. Netskina, O. V. Komovo, V. I. Simagina, G. V. Odegova, I. P. Prosvirin, O. A. Bulavchenko, Renew. Energ. 2016, 99, 1073. [31] A. Rodrigues, P. Costa da, C. Methivier, S. Dzwigaj, Catal.
Today 2011, 176, 72.
[32] B. J. Tan, K. J. Klabunde, P. M. A. Sherwood, J. Am. Chem. Soc. 1991, 113, 855.
[33] Y. Meng, Nanomaterials 2015, 5, 554.
[34] L. Lukashuk, K. Föttinger, E. Kolar, C. Rameshan, D. Teschner, M. Hävecker, A. Knop-Gericke, N. Yigit, H. Li, E. McDermott, M. Stöger-Pollach, G. Rupprechter, J. Catal. 2016, 344, 1.
[35] S. Todorova, H. Kolev, J. P. Holgado, G. Kadinov, C. Bonev, R. Peren~ı´guez, A. Caballero, Appl. Catal. B-Environ. 2010, 94, 46.
[36] F. Zheng, D. Zhu, X. Shi, Q. Chen, J. Mater. Chem. A 2015, 3, 2815. [37] Z. Huang, W. Zhou, C. Ouyang, J. Wu, F. Zhang, J. Huang,
Y. Gao, J. Chu, Sci. Rep.-UK 2015, 5, 10899.
[38] J. H. Park, J. Man, M. Jin, J.-K. Keon, S.-S. Kim, S. H. Park, S. C. Kim, Y.-K. Park, Nanoscale Res. Lett. 2012, 7, 14.
[39] Y. Liang, H. Wang, J. Zhou, Y. Li, J. Wang, T. Regier, H. Dai, J. Am. Chem. Soc. 2012, 134, 3517.
[40] W. J. Xu, H. F. Wang, X. H. Liu, J. W. Ren, Y. Q. Wang, G. Z. Lu, Chem. Commun. 2011, 47, 3924.
[41] Z. Gao, G. Fan, M. Liu, L. Yang, F. Li, Appl. Catal. B-Environ. 2018, 237, 649.
How to cite this article: Akmaz S, Esen M, Sezgin E, Koc SN. Efficient manganese decorated cobalt based catalysts for hydrogenation of 5-hydroxymethylfurfural (HMF) to 2,5-dimethylfuran (DMF) biofuel. Can J Chem Eng. 2020;98:138–146.
