Abstract
We developed a series of new organic-inorganic hybrid hydrotalcite functionalized Ru catalytic systems. All the developed materials have been studied by FTIR, N2 physisorption, ICP-OES, XPS, NMR (1H, 13C, 29Si) and TEM analysis were performed to know the physiochemical behavior and structural morphology of functionalized hydrotalcite materials. XPS results strongly suggest that it involves the formation of N-Ru coordination bonds. We applied these well analyzed materials for CO2 hydrogenation reaction as catalyst (with and without ionic liquid medium). We found that Ru metal containing functionalized hydrotalcite materials were highly active and stable (in terms of catalyst leaching and recycling). The heterogeneous catalyst can be easily recovered and reused 8 times without significant loss of catalytic activity and selectivity, which is a better green alternative for practical applications.
Graphical Abstract
1 Introduction
The escalating level of carbon dioxide is one of the important pressing environmental concerns of our age and CO2 is considered as a major greenhouse gas [1, 2, 3]. Carbon capture and storage gave an alternative to reduce CO2 concentration in the environment but surprisingly the carbon capture process alone increases the demand for energy by 25-40% [3]. In most of the reports, CO2 is utilized as a cheap C1 source in the production of value-added chemicals like formic acid, methanol, urea, and polycarbonates [1]. Formic acid is widely applicable in different industries such as leather, rubber, and fragrance [1,3].
Many reports have been published to utilize the application of homogenous as well as heterogeneous catalysts with and without additives to hydrogenate CO2 gas to formic acid [4,8]. The main obstacle to achieve selective and easy hydrogenation of CO2 is the positive standard free energy of the hydrogenation reaction [1,4-8]. A series of supported and unsupported transition metal complexes or nanocatalysts were utilized with or without a base to achieve a reasonable conversion of CO2 to formic acid [9,10]. Nevertheless, the easy recovery of formic acid and catalyst recycling, as well as the requirement of the base as an additive, are still a challenge for the scientific community. This problem is also considered as one of the main shortcomings for the industrialization of CO2 hydrogenation reaction [10,11].
A series of unsupported, homogeneous transition metal catalysts such as Ru, Rh, Ir and Pt were effectively used to catalyze the CO2 hydrogenation reaction [1,9]. Unfortunately, very few scientific reports are offered on heterogeneous catalysts for CO2 hydrogenation reaction. Organic-inorganic hybrid catalysts are considered as a promising type of tethered heterogenous catalysts and are designed to retain the selectivity of homogeneous catalysts while being immobilized on heterogeneous support to obtain easy separation [9]. In one of the recent reports, Baiker et al. reported that the synthesis of transition metals incorporated silica material as catalyst followed by the co-condensation method to catalyze the preparation of N, N-diethylformamide [12]. In another report, Hicks and Jones prepared a series of silica tethered Ir metal catalyst for the selective hydrogenation of CO2 gas [13]. Although, the mentioned catalytic systems were found to be effective for CO2 hydrogenation, these systems also suffered in terms of long reaction, high catalysts loading and catalysts leaching.
Hydrotalcite clay (HTc) is a naturally occurring material having layered double hydroxide (LDH) structure [14,15]. The general formula of hydrotalcite material is [M1-x2+Mx3+ (OH)2]x+ (An-)x/n.mH2O, where A- are anions like CO32-, OH-, Cl- or SO42-. The morphology of natural and synthetic HTc is like brucite clay but the basicity varies with the concentration of cations [13, 14, 15, 16]. Such variation makes the HTc more interesting in terms of the catalyst as well as catalyst support [16,18]. Therefore, HTc materials have been effectively applied in numerous areas such as adsorption materials for gases (CO2, H2S, and SO2), precursors for functional materials, photochemistry and electrochemistry [15,16]. In addition, the presence of hydroxyl groups makes the HTc as an active and recyclable solid base catalyst for a variety of organic reactions [18]. In recent years, the applications of HTc have been increased not only as heterogeneous catalyst or catalyst support but also as anion exchanger, CO2 absorbent and in water treatment [18,19]. As a catalyst support, HTs have been used to immobilize various transition metal nanoparticles and homogeneous catalysts [19, 20, 21]. Alkoxysilane linkage is one of the easiest ways to link homogeneous or heterogeneous catalysts over a hydrotalcite surface [21, 22, 23]. Surprisingly, anchoring the homogeneous or heterogeneous catalysts over silica and alumina is well studied but there are still very few reports available with respect to alkoxysilane interaction with hydrotalcite clay [21].
In this report, we are offering the synthesis of hydrotalcite tethered Ru metal catalyst (HRUC) for the hydrogenation of CO2 gas in the functionalized ionic liquid medium. The HRUC catalytic system was synthesized by a multistep grafting process using an iminophosphine ligand tethered to a hydrotalcite inorganic support.
2 Experimental
Reagent Plus® grade chemicals were purchased from Sigma Aldrich and other chemical suppliers. FTIR measurements were conducted with Bruker Tensor 27 in DRIFT mode (KBr powder) with a scanning range 4004000 cm-1. Perkin Elmer Optima 3300 XL ICP-OES was used to study the elemental analysis. Nuclear Magnetic Resonance (NMR) spectra were recorded on a standard Bruker 300WB spectrometer with an Avance console at 400 and 100 MHz for 1H NMR. The morphology of catalysts was investigated by transmission electron microscopy (TEM) using a Philips CM12 instrument. Kratos-Axis 165 with Mg Kα radiation 1254 eV was used to perform X-ray photoelectron spectroscopy (XPS). DTA-TGA thermal analyzer apparatus (Shimadzu DTG-60H) was used to study the thermal stability of an ionic liquid. BET surface area, pore size, and pore volume measurements of the catalysts were determined from a physical adsorption of N2 using liquid nitrogen by an ASAP2420 Micromeritics adsorption analyzer (Micromeritics Instruments Inc). All the samples were degaussed at 250°C for 2 h before taking the measurements. The surface area and pore size distribution (PSD) were measured from the BET and BJH equations, respectively, by the instrument software. All the hogenation reactions were carried out in a 100 mL stainless steel autoclave (Amar Equipment, India).
2.1 Synthesis of Hydrotalcite material
The hydrotalcite with Mg and Al species was synthesized using the coprecipitation method [16]. An aqueous solution containing Mg(NO3)2. 6 H2O (80.0 g) and Al(NO3)3. 9 H2O (37.74 g) in 225 ml of water (solution A) was prepared (while maintaining a Mg/Al molar ratio of 3). A separate solution B, containing NaOH (54 g) and Na2CO3 (21.6 g) in 675 ml of water was prepared separately. Solution A and B were simultaneously added dropwise into distilled water under vigorous mechanical stirring while maintaining the pH of the resulting solution in the range of 9.5-10 at 55°C temperature. The slurry was ripened for 30 min under forceful stirring at 55°C and was allowed to stand in its mother liquor for 3 h. The precipitate was washed several times until a pH of 7 was reached and was then dried at 110°C for 12 h. The calcined hydrotalcite (HT) was obtained by calcining the hydrotalcite (HTc) at 600°C for 4 h.
2.2 Synthesis of iminophosphine ligands -Moiety A and iminophosphine ligands with Ru metal- Ru-Moiety A
A 100 ml round bottom flask was charged with 2-(diphenylphosphine) benzaldehyde (2.5 mmol) and 1-propylamine (10 mL). All reactants were refluxed under nitrogen for 5 hours to obtain the Schiff- base. After cooling, the reaction mass to room temperature, monophosphate ligand was isolated as a red-brown oil after vacuum distillation and column chromatography (yield 91%).
1H NMR (400 MHz, CD2Cl2): δ= 9.02 (s, 1H), 8.04 (s, 1H), 7.44 - 7.31 (m, 12 H), 6.97 (s, 1 H), 3.42 (t, 2 H), 1.57 - 1.53 (q, 2 H), 0.82 – 0.77 (t, 3 H); 13C NMR (100 MHz, CD2Cl2): δ= 160.02, 134.25, 128.73, 63.34, 23.95, 11.79 ppm. 31P NMR: (300 MHz, CD2Cl2, ppm) δ= - 13.02 ppm.
Ru-Moiety-A was obtained by reacting moiety -A (2 g) with RuCl3.3H2O (2.1 g) in dry ethanol (25 mL) under an inert atmosphere. The combined reaction mass was refluxed for 12 hours. After cooling, the reaction mass was washed with dry ethanol (10 x 2 mL) under a nitrogen atmosphere. Dried material (Ru-Moiety-A) through lyophilizer was kept under nitrogen atmosphere.
2.3 Synthesis of Alkoxysilane -Moiety B
A 100 mL round bottom flask was charged with 2-(Diphenylphosphino) benzaldehyde (0.75 g), 3-(Trimethoxysilyl)propylamine (0.375 g) and 40 mL dry THF under a nitrogen atmosphere. The combined reaction mass was reacted at 100oC for 5 hours. After completion of the reaction, we obtained an alkoxysilane compound with a bidentate iminophosphine ligand as a yellow oily liquid (after the careful removal of solvent and volatile impurities followed by vacuum distillation and column chromatography) (yield 95%).
1H NMR (400 MHz, CD2Cl2): δ = 8.9 (s, 1H), 8.03 (s, 1H), 7.45-7.28 (m, 12 H), 6.93 (s, 1 H), 3.52 (s, 9 H), 3.47 (m, 2 H), 1.68 - 1.61 (m, 2 H), 0.58 – 0.51 (m, 2 H); 13C NMR (100 MHz, CD2Cl2): δ= 158.58, 134.15, 128.78, 65.15, 50.80, 24.61, 8.74 ppm.; 31P NMR : (300 MHz, CD2Cl2, ppm) δ= -13.05ppm.
2.4 Synthesis of phosphine functionalized calcined hydrotalcite (ABIL-HT-A to C)
A post-synthetic grafting method was applied to obtain phosphine functionalized calcined hydrotalcite material under an inert atmosphere. Phosphine ligand (2 mmol) was allowed to react with calcined hydrotalcite (1 g) in dry toluene under (20 mL) nitrogen atmosphere. The reaction mass was stirred for the next 24 hours at room temperature (30-35oC). The resulting solid mass was washed several times to dry the toluene (10 x 2 mL). The recovered solid was carefully dried in a lyophilizer. The pale-yellow solid (ABIL-HT-A to C) was carefully placed under a vacuum at room temperature (yield 98%).
2.5 Synthesis of amine functionalized calcined hydrotalcite (ABIL-HT-D)
Calcined hydrotalcite (1 g) was mixed with 3-aminopropyltrimethoxysilane (1.2 mmol) in dry toluene (20 mL). The combined reaction mass was heated and vigorously stirred at room temperature (30-35oC) for next 24 hours under a nitrogen atmosphere. The resulting solid material was washed and filtered with dry toluene. (10 x 2 mL) The solid was then dried in a lyophilizer. The light yellow solid (ABIL-HT-D) was sensibly stored under vacuum in a nitrogen atmosphere (Yield 98%).
2.6 Synthesis of Ru-Tethered Pre- catalyst (HRUC-A to D)
The ABIL-HT-A to D was added to RuCl3. 3H2O in anhydrous ethanol (25 mL) under nitrogen by keeping the ratio of Ru and phosphorus (1:1). The resulting reaction mixture was refluxed for 10 hours. After cooling the reaction mass, the resulting solid material was carefully washed several times with dry ethanol (10 x 2 mL) under a nitrogen atmosphere. The recovered black solid was dried using a lyophilizer and then carefully stored under inert atmosphere (yield 98%).
2.7 CO2 hydrogenation reaction and recycling experiment
After performing the catalysts pretreatment process at 45oC for 20 minutes under 10 MPa pressure of hydrogen gas, all the gases were completely replaced with nitrogen gas at room temperature. Then all the reactants (except H2 and CO2 gas) were added along with a known quantity of dioxane (internal standard for 1HNMR analysis) to the reaction vessel (as per Table 3 and 4) without opening the autoclave. Then nitrogen gas was completely replaced by CO2 gas (by 2-3 times flushing). Absorption of CO2 gas was carried out at 80oC with 20 bar pressures for 1 hour. Later, after reducing the temperature of the autoclave to 40oC, hydrogen gas was added into the reactor. The combined reaction mass was permitted to cool at room temperature before opening the reaction vessel. A small quantity of crude reaction mass was used for 1HNMR analysis and titration to quantify the amount of formic acid in the reaction sample. No signs of ionic liquid, as well as dioxane decomposition, were recorded during 1H NMR analysis. The results obtained from 1H NMR analysis were found to be in good agreement with the titration method. Recovery of formic acid was carried out easily under reduced pressure. At 50oC initially, all the water and other volatile impurities were removed at 75-80 oC, and under nitrogen flow the formic acid was isolated.
2.7.1 Catalyst recycling with ionic liquid
After careful isolation of formic acid, the ionic liquid immobilized catalytic system was washed several times with dry diethyl ether (5 x 2 mL) to remove all the organic impurities and then further dried in a lyophilizer. The ionic liquid immobilized catalytic system was again pretreated at 45oC for 20 minutes under 10 MPa pressure of hydrogen gas, before going to recycling for CO2 hydrogenation.
2.7.2 Catalyst recycling without ionic liquid
After the successful isolation of formic acid, the catalytic system was washed several times with ether to remove all the organic impurities from the catalysts and then the catalyst was dried in a lyophilizer. The fresh catalyst was added to the used catalyst, in case of any catalyst’s loss during the workup process. After adjusting the quantity of catalyst again the autoclave was charged with a pretreated catalyst and reactants as per the above-mentioned method.
Ethical approval: The conducted research is not related to either human or animal use.
3 Results and Discussion
Synthetic hydrotalcite was prepared via a well-reported coprecipitation method using Mg(NO3)2 and Al(NO3)2, under basic condition. The corresponding white precipitates were dried at 110oC and further calcined at 600oC for 4 hours. The alkoxysilane containing the bidentate iminophosphine ligand [o-Ph2PC6H4CH=N(CH2)3Si(OMe)3] A was synthesized by reacting 2-(Diphenylphosphino) benzaldehyde with 3-(Trimethoxysilyl)propylamine under a nitrogen atmosphere in dry THF. All the analytical data obtained from NMR analysis (1H/13C/31P) were found in agreement with the reported data for the same [9,21]. Again, iminophosphine ligand A was grafted on the surface of a calcined hydrotalcite material by refluxing the reaction mass in dry toluene for 24 hours. Unreacted or loosely coordinated iminophosphine ligand A was isolated by Soxhlet extraction method. The resulting solid material was dried under vacuum at 100oC to obtain ABIL-HT-A (Figure 1). At last, ABIL-HT-A went to the metalation step. The resulting HRUC-A material was obtained in the black color solid and properly stored in an argon atmosphere. In addition, we synthesized a series of hydrotalcite anchored monodentate phosphine-based materials such as HRUC-B, HRUC-C and HRUC-D using ABIL-HT-B, ABIL-HT-C and ABIL-HT-D respectively (Figure 1). We also created an iminophosphine ligand without out alkoxysilane moiety A to use them as a reference material to understand the physiochemical properties of our developed materials (Figure 2).
Structures of Hydrotalcite tethered phosphine ligands.
DRIFT analysis data for Moiety A and B.
We used various sophisticated techniques to understand the physiochemical properties of HRUC materials (Figure 2 and 3) like 13 C & 29Si NMR, DRIFT, FTIR, XPS, elemental analysis using AAS, BET surface analysis and inductively coupled plasma optical emission spectroscopy (ICP-OES).
13 C analysis of HRUC-A (after removal of the solvent peak).
A solid state 13C & 29Si NMR analyss was applied to understand the surface structure of functionalized hydrotalcite materials (Figure 3 and 4). This analysis helps to understand the physiochemical structure of hydrotalcite with alkoxysilane moiety. The 13C NMR spectrum of an organic moiety of HRUC-A confirmed that the bidentate iminophosphine ligand is properly intact with hydrotalcite support. This was also confirmed from the report published by Wang et al. [24]. No peak corresponding to the methoxy group of trimethoxysilane was found in 13C NMR analysis which also supports the idea that all the three methoxy groups have been lost during functionalization of hydrotalcite as HRUC-A. This data confirms the tripodal (T3) type arrangement of silane moiety over hydrotalcite. 29Si NMR analysis provided the insight nature of M-O-Si bonds on the hydrotalcite surface [24,25]. The two distinct peaks confirmed in 29Si NMR confirmed the anchoring of silane groups to Mg3-OH and AlMg2-OH environment respectively. Similar results were observed in other HRUC-B to D materials [24, 25, 26].
29Si analysis of HRUC-A.
We performed diffuse reflectance infrared fourier transform spectroscopy (DRIFTS) for iminophosphine ligand (without alkoxysilane) and iminophosphine ligand with alkoxysilane moiety. Collectively, we used this data to understand the structure of HRUC-A to D. We located the position of a Schiff base (CH=NR) double bond near to 1640 cm-1. The vibrations peaks of phenyl rings in HRUC-A and B materials located in the following regions 3061, 1580, 1430, 741 and 694 cm-1. Similar peaks for phenyl rings were observed in ABIL-HT-A and ABIL-HT-B. Such data confirms the presence of the Schiff base before and after the metalation of our developed materials. In FTIR analysis, we also recorded different types of absorption peaks for HRUC-A to D (Figure 5 and 6).
DRIFT data for HRUC-A and ABIL-HT-A with respect to Hydrotalcite.
DRIFT analysis data for HRUC-A to D with respect to Hydrotalcite.
Before going to the metalation step, we performed surface analysis using N2-physisorption analysis for hydrotalcite (before and after calcination) and after the functionalization calcined hydrotalcite clay. All the data were tabulated in Table 1. A noticeable increase in the BET surface area, pore size, and pore (single point pore volume measured at P/Po =0.97 on absorption) volume was recorded in the calcined hydrotalcite material with respect to normal hydrotalcite material. A visible drop in terms of the BET surface area, pore size and pore volume were found in functionalized calcined hydrotalcite materials (ABIL-HT-A to D). The lowest value of BET surface area, pore size, and pore volume was noted for ABIL-HT-B. After the completion of the metalation process for ABIL-HT-A to D materials, we also checked the BET surface area, pore size and pore volume along with the elemental analysis. This data helped us to understand the metallic composition of both ABIL-HT-A to D and HRUC-A to D materials (Table 1). Considering the matrix effect, we performed the elemental analysis using the ASS method to confirm the quantity of Mg, Al, Si and Ru content in our developed materials. The silica content was only recorded in functionalized hydrotalcite clay, in the same pattern Ru metal signal was only logged for HRUC-A to D materials. A slightly higher amount of Ru metal was found in HRUC-A material with respect to other HRUC-B to D materials.
XPS analysis was performed to understand the composition of different HRUC-A to D materials. The Ru 3d, C1, N 1S and P2p levels were calculated at a normal angle with respect to the plane of the surface during XPS spectra calculation. The binding energy was carefully measured with a precision of ±0.2 eV. Shirley background subtraction, as well as Gaussian and Lorentzian principal for peak shape, was considered while doing XPS analysis of Ru 3d levels. The peaks near to 280.2 eV (Ru 3d5/2) and 284.3 eV (Ru 3d3/2) inveterate the attendance of Ru (0) species. Unexpectedly, no signs of RuO2 were recorded while performing the XPS analysis of HRUC-A to D materials. This observation was also supported as no peak was found to represent Ru 3p with binding energy 464 eV which confirmed the absence of Ru species with +2 oxidation state. We also linked the XPS data of RuCl3/ calcined hydrotalcite material with XPS data of HRUC-A to D materials and intimated the occurrence of Ru (0) species.
Some resemblance with respect to Ru 3d5/2 and Ru 3d3/2 was found between HURC-B and RuCl3/Calcined hydrotalcite materials which established the low loading of phosphine ligands on support. In contrast, much higher phosphine loading was recorded in HRUC-C, which resulted in a higher binding energy for Ru 3d5/2 and Ru 3d3/2. The binding energies of HRUC-A for Ru 3d5/2 and Ru 3d3/2 were found almost like Ru-PNPr1. These results indicate that the location of Ru metal in HRUC-A is identical to Ru-PNPr1 species. TEM data also used to confirm the morphology of the Ru metal loaded functionalized hydrotalcite material (Figure 7). TEM image of all the HRUC-A to D materials along with hydrotalcite and calcined hydrotalcite materials were recorded. The image analysis of HRUC-A to D materials inveterate the presence of well dispersed Ru nanometal in the range of 8-10 nm (± 0.25) with a mean diameter of 4.5 nm. TEM image analysis also disclosed the presence of highly crystalline nature of Ru nanometals.
TEM images of HRUC catalysts
3.1 Hydrogenation of CO2 to formic acid without ionic liquid medium
Formic acid is one of the important chemicals in organic chemistry. It has been reported as a starting chemical for the synthesis of a large variety of beneficial chemical derivatives such as aldehydes, ketones, amine and carboxylic acid. It is also utilized in the manufacturing of perfume and fragrance [1]. Transition metal complex catalysts were widely applied to utilize CO2 gas as a C1 synthetic unit in formic acid synthesis. Although, this developed catalytic system offers the production of formic acid, these catalytic protocols also suffer from the selectivity and low reactivity of CO2 gas due to its high thermodynamic stability (+ΔGo298= 32.9 KJ/mol). Considering the above-mentioned facts, we tested our developed catalytic systems (HRUC-A to D) for the selective hydrogenation of CO2 to formic acid under high-pressure reaction condition in the presence of triethylamine and a small quantity of water at 80oC. We used cyclohexanone as an internal standard in the reaction mass.
All results obtained while using HRUC-A to D catalysts were tabulated in Table 3, entry 1-18, Scheme 1. The quantity of formic acid was calculated using 1H NMR and acid-base titration (Figure 8). Dioxane was added in the reaction mass before starting the reaction as an internal standard. After completion of the reaction, a small quantity of reaction mass was used for 1H NMR analysis, which confirms the formation of formic acid along with no formation of any side product during the reaction. The quantity of formic acid calculated via 1H NMR was found to be in good agreement with the data obtained from acid-base titration of reaction mass.
1H NMR data after the hydrogenation of CO2.
Hydrogenation of CO2 in triethyl amine system.
A small quantity of formic acid was reported in absence of triethylamine which confirmed that the basic nature of hydrotalcite clay also influences the hydrogenation of CO2 gas up to certain extent. RuCl3. 3H2O precursor, as well as RuCl3, supported calcined hydrotalcite material gave formic acid with a low TON and TOF values. A low quantity of formic acid was recorded with amine functionalized HRUC- D catalysts with respect to phosphine functionalized HRUC-A to C catalytic system. This observation confirmed the high activity of phosphine functionalized HRUC-A to C catalytic systems for the hydrogenation of CO2 gas. The bidentate phosphine functionalized HRUC-A catalyst gave formic acid with a high TON and TOF value with respect to other catalytic systems (HRUC-B to D). We performed all the optimization steps with HRUC-A catalyst to get a high quantity of formic acid. Changing the temperature, reaction time, quantity of reactant/catalyst quantity gave a noticeable change in formic acid production. Ru metal loaded moiety-A gave the lowest value of formic acid due to the Ru-Ru metal dimerization in the reaction solution.
A filtration test was performed to know the stability of HRUC-A to D catalytic systems. After the completion of the hydrogenation reaction under optimized reaction conditions (Table 3, entry 4), the solid part of the reaction was isolated through 0.45 micrometer polytetrafluoroethylene (PTFE) filter and the filtered was used to perform the hydrogenation reaction. We found a small quantity of formic acid with TOF value of 12 h-1. This test confirmed the catalytic leaching during the reaction, which was further supported by XPX and ICP-OES analysis of Ru metal in the filtrate. The same observation was observed while recycling the HRUC-A catalyst. We only recorded the recycling test up to 4 cycles mainly because of catalysts leaching, as well as Ru metal agglomeration (Figure 7). The size of Ru metal was increased from 8 to 78 nm.
Results of Recycling tests of the HRUC-A catalytic system.
3.2 Hydrogenation of CO2 to formic acid with ionic liquid medium
Many attractive properties of ionic liquids like extremely low volatility, high thermal stability, strong solvation nature for various substances and wide liquid-temperature range makes them promising alternative solvents over conventional solvent systems. Additionally, functionalized ionic liquid (by changing the anion or cation or grafting functional group on to ions) extended the application ionic liquids in catalysis, extraction, absorption of gases such as CO2, H2, SO2 etc.
In our previous studies, we synthesized a series of functionalized ionic liquid liquids such as 1-(N,N-dimethylaminoethyl) 2,3-dimethylimidazolium trifluoromethanesulfonate ([mammim][TfO]), and 1,3-di(N,N-dimethylaminoethyl)-2-methylimidazolium bis (trifluoromethylsulfonyl) imide ([DAMI][NTf2]) were synthesized as per the reported procedures [31, 32] while ionic liquids such as 1-(N,N-dimethylaminoethyl) 2,3-dimethylimidazolium bis (trifluoromethylsulfonyl) imide ([mammim][NTf2]), 1-(N,N-dimethylaminoethyl)-2,3-dimethylimidazolium nonafluorobutanesulfonate ([mammim] [CF3CF2CF2CF2SO3]), 1-(N,N-dimethylaminoethyl)- 2,3-dimethylimidazolium trifluoromethanesulfonate ([mammim] [BF4]), 1,3-di(N,N-dimethylaminoethyl)-2-methylimidazolium trifluoromethanesulfonate ([DAMI][TfO]), , 1,3-di(N,N-dimethylaminoethyl) -2-methylimidazolium nonafluorobutanesulfonate ([DAMI][CF3CF2CF2CF2SO3]) and 1,3-di(N,N-dimethylaminoethyl)-2-methylimidazolium tetrafluoroborate ([DAMI][BF4]) in order to obtain a high degree of chemo selectivity, easy catalyst recycling and informal product isolation [4, 5, 6, 7, 8]. Moreover, we also did a comprehensive study on the absorption of CO2 gas in above mentioned functionalized ionic liquid. In our study, we found 1,3-di(N,N-dimethylaminoethyl)-2-methylimidazolium nonafluorobutanesulfonate ([DAMI] [CF3CF2CF2CF2SO3]) ionic liquid as a promising reaction medium [4, 5, 6, 7, 8, 9,18, 19, 20]. The density of this ionic liquid was measured by using a capillary pycnometer and found 1.235 g/mL. We also calculated the glass transition temperature of this ionic liquid using differential scanning calorimetry and recorded it to be -41oC. This ionic liquid was also analyzed by means of thermogravimetric analysis which indicates that the ionic liquid is found to be stable up to 225oC (Figure 10).
![Figure 10 TGA analysis data of [DAMI][CF3CF2CF2CF2SO3] ionic liquid.](/document/doi/10.1515/chem-2018-0094/asset/graphic/j_chem-2018-0094_fig_011.jpg)
TGA analysis data of [DAMI][CF3CF2CF2CF2SO3] ionic liquid.
In this manuscript, we took the ionic liquid as a reaction medium to perform the CO2 hydrogenation using HRUC-A catalyst. All the experimental results of the CO2 hydrogenation reaction are represented in Table 4, entry 1-18, Scheme 2. The heterogeneous nature of HRUC-A catalyst in ([DAMI][CF3CF2CF2CF2SO3]) ionic liquid was investigated as per the reported procedure of catalyst poisoning experiments [6]. The above-mentioned protocol was found to be relatively simple and both ionic liquids as well as the catalyst were recycled straightforwardly. It is also expected that the competence of the HRUC-A catalyst was increased due to the presence of two -NH functional groups on both the ends of the ionic liquid. The synergic effect of ionic liquid with HRUC-A catalysts was also studied and it was found that while using [bmim][NTf2] ionic liquid (as nonfunctional ionic liquid) with HRUC-A catalyst, formic acid was obtained with low TON/TOF value (Table 4, entry 16).
Ionic liquid mediated CO2 hydrogenation reaction.
After completion of CO2 hydrogenation, the reaction mass was heated to isolate first water and then formic acid with the help of N2 flow. The presence of a base (ionic liquid and hydrotalcite material) and formic acid in the reaction mass developed an equilibrium between free formic acid, formic acid salt, ionic liquid and hydrotalcite material. The amount of free formic acid was increased by increasing the temperature as the acid-base neutralization to salt is an exothermic process. Considering that fact, separation of formic acid was carried out under nitrogen flow at optimized temperature.
All the CO2 hydrogenation results were tabulated in Table 4 and Scheme 2. Surprisingly no by-product was reported during hydrogenation reaction (confirmed by 1H NMR). We started our study by changing the molar ratio of formic acid/HRUC-A+ionic liquid at 80oC and 20 MPa as a function of time. This is represented in Figure 3. The almost linear increase was found between the formic acid/ HRUC-A+ionic liquid ratio near to 1:81 at 5 hours. After this time no drastic change in ratio was reported while increasing the temperature and after 6 hours the ration between formic acid/HRUC-A+ionic liquid was found to be equal to 2.1 at 6-10 h. This observation showed that free formic acid combined with the basic HRUC-A+ionic liquid system and formed the formic acid salt. In starting, formic acid neutralization was fast with respect to the formic acid formation. At a high ratio of formic acid/ ionic liquid, most of the ionic liquid got neutralized but HRUC-A+ionic liquid system and formic acid produced remained same in the system.
The effect of water was also evaluated (Table 4, entry 1-7). It was clearly observed in our study that the TON-TOF value of formic acid was increased by adding water. This increase can be explained in two different manners. First, the addition of water will reduce the viscosity of the ionic liquid and second, presence water may increase the chances of bicarbonate formation due to the reaction of the CO2 gas with water and the basic ionic liquid system. In some of the reports, these bicarbonate species are treated as the true substrate for hydrogenation reaction (Figure 11 a). The bicarbonate species was identified in the 13C NMR spectra and no other species were detected (Figure 11 b). The errors in the peak areas of the spectra were estimated to be at most 3% on the basis of the standard deviations of peak areas from the different carbons of the same amine species of ionic liquid. It should be noted that one peak was detected for the carbon of the bicarbonate species in 13C NMR because of the fast exchange of protons.
FTIR data of format and carbonate species formed during the reaction.
![Figure 11 b 13C NMR spectra of CO2-loaded [DAMI][CF3CF2CF2CF2SO3] ionic liquid with peak interpretations for bicarbonate species.](/document/doi/10.1515/chem-2018-0094/asset/graphic/j_chem-2018-0094_fig_014.jpg)
13C NMR spectra of CO2-loaded [DAMI][CF3CF2CF2CF2SO3] ionic liquid with peak interpretations for bicarbonate species.
The hydrogenation reaction was carried out at different temperature and reaction pressure and time (Table 4, entry 14). On increasing the temperature, the rate of reaction increased and the TON value of formic acid was increased up to 1581.62 at 100oC (Table 4, entry 7 & 8). As pressure increased from 20 MPa to 30 MPa at 80oC, TON and TOF values were also found to be higher and recorded 1886.94 and 317.4 respectively. The pressure effect in this reaction can be explained according to Henry’s law, as the solubility of two gases increases along with increasing the pressure, resulting in the concentration of reactants having greater importance on the rate of reaction [4, 5, 6, 7, 8].
We recovered the formic acid followed by the distillation process under nitrogen flow. No sign of ionic liquid decomposing and side product formation was recorded during the reaction. This observation was supported by 1HNMR analysis of crude reaction mass. We used a simple titration method as well as 1HNMR analysis to quantify the amount of formic acid. Data from both analyses were found to be in good agreement.
To understand the stability of [DAMI] [CF3CF2CF2CF2SO3]/HRUC-A catalytic system we performed the filtration experiment. In this experiment, we mixed our catalytic system with 10 mL of water and heated the resulting mixture at 80oC for the next 6 hours under 40 bar hydrogen pressure in an autoclave. After cooling the reaction mass and degassing the autoclave, the resulting mixture was filtered, and the filtrate was used to for the CO2 hydrogenation experiment. Surprisingly, no sign of formic acid was found. This result was also supported by ICP-OES and XPS method that no Ru metal was detected in the solution.
Taking advantage of above experimental results, we exploited our catalytic system for a catalyst recycling test. After completing the reaction, the reaction product was isolated, and the complete catalytic system was recycled to the next run (after the pretreatment process).
We easily recycled our catalytic system up to 8 runs (Figure 12). The drop-in catalyst activity was recorded after the 8th run mainly due to agglomeration of Ru metal as the size of metal was increased from 8 to 34 nm (TEM image, Figure 7).
![Figure 12 Recycling test results of [DAMI][CF3CF2CF2CF2SO3]/HRUC-A catalytic system.](/document/doi/10.1515/chem-2018-0094/asset/graphic/j_chem-2018-0094_fig_015.jpg)
Recycling test results of [DAMI][CF3CF2CF2CF2SO3]/HRUC-A catalytic system.
4 Conclusion
In summary, we synthesized a series of active hydrotalcite supported Ru metal complexes (HRUC-A to D) to catalyze hydrogenation of CO2 gas with and without ionic liquids. HRUC-A catalyst was found to be highly active and gave a good quantity of formic acid. The combination of basic ionic liquid [DAMI][CF3CF2CF2CF2SO3] with HRUC-A catalytic was found extremely efficient to get formic acid with a high TON and TOF value. The addition of water and increase in pressure, as well as temperature, resulted in a high reaction rate. The molar ratio of formic acid to ionic liquid was reached up to 2.1 (0.276:1 w/w) in one reaction cycle. The recovery of formic acid from the reaction mass was very easy and [DAMI][CF3CF2CF2CF2SO3]/HRUC-A catalytic system was found recyclable up to 8 runs. A drastic effect of ionic liquid was recorded over the stability (in terms of catalyst leaching) HRUC-A catalyst. Without ionic liquid, HRUC-A catalyst was recycled only up to 4 cycles. This simple, efficient, and green protocol to get formic acid is economical, energy efficient and has the potential to be applied to industry.
Conflict of interest: Authors declare no conflict of interest.
References
[1] Upadhyay P., Srivastava V., Carbon sequestration: Hydrogenation of CO2 to formic acid, Present Environment and Sustainable Development, 2016, 10(2), 13-34.10.1515/pesd-2016-0022Search in Google Scholar
[2] Keeling R.F., Triage in the greenhouse, Nat. Geosci., 2009, 2(12), 820-822.10.1038/ngeo701Search in Google Scholar
[3] Gunasekar G.H., Park K., Jung K.-D., Yoon S., Inorg. Chem. Front., 2016,3, 882-895.10.1039/C5QI00231ASearch in Google Scholar
[4] Upadhyay P.R., Srivastava V., Titanium dioxide supported ruthenium nanoparticles for carbon sequestration reaction, Nanosyst.: Phys., Chem., Math., 2016, 7, 513–517.10.17586/2220-8054-2016-7-3-513-517Search in Google Scholar
[5] Upadhyay P.R., Srivastava V., Selective hydrogenation of CO2 using ruthenium nanoparticles intercalated montmorillonite clay, Lett. Org. Chem., 2016, 13(6), 459-465.10.2174/1570178613666160815095733Search in Google Scholar
[6] Upadhyay P.R., Srivastava V., Selective hydrogenation of CO2 gas to formic acid over nanostructured Ru-TiO2 catalysts, RSC Adv., 2016, 6(48), 42297-42306.10.1039/C6RA03660KSearch in Google Scholar
[7] Srivastava V., Ru-exchanged MMT clay with functionalized ionic liquid for selective hydrogenation of CO2 to formic acid, Catal. Lett., 2014, 144(12), 2221-2226.10.1007/s10562-014-1392-4Search in Google Scholar
[8] Upadhyay P., Srivastava V., Synthesis of monometallic Ru/TiO2 catalysts and selective hydrogenation of CO2 to formic acid in ionic liquid, Catal. Lett., 2016, 146(1), 12-21.10.1007/s10562-015-1654-9Search in Google Scholar
[9] Upadhyay P. R., Srivastava V., Heterogeneous Silica Tethered Ruthenium Catalysts for Carbon Sequestration Reaction, Catalysis Letters, 2016, 146, 1478-1486.10.1007/s10562-016-1772-zSearch in Google Scholar
[10] Li W., Wang H., Jiang X., Zhu J., Liu Z., Guo X., Song C., A short review of recent advances in CO2 hydrogenation to hydrocarbons over heterogeneous catalysts, RSC Adv., 2018, 8, 7651.10.1039/C7RA13546GSearch in Google Scholar PubMed PubMed Central
[11] Álvarez A., Bansode A., Urakawa A., Bavykina A.V., Wezendonk T.A., Makkee M., Gascon J., Kapteijn F., Challenges in the greener production of formates/formic acid, methanol, and DME by heterogeneously catalyzed CO2 hydrogenation processes, Chem. Rev., 2017, 117(14), 9804-9838.10.1021/acs.chemrev.6b00816Search in Google Scholar PubMed PubMed Central
[12] Schmid L., Rohr M., Baiker A., A mesoporous ruthenium silica hybrid aerogel with outstanding catalytic properties in the synthesis of N,N-diethylformamide from CO2, H2 and diethylamine, Chem. Commun., 1999,0, 2303-2304.10.1039/a906956iSearch in Google Scholar
[13] Zhang Y, Jinhua JF, Yu FY, Zheng X, The Preparation and Catalytic Performance of Novel Amine-Modified Silica Supported Ruthenium Complexes for Supercritical Carbon Dioxide Hydrogenation to Formic Acid, Catalysis letter 2004, 93, 231234.10.1023/B:CATL.0000017081.65272.5fSearch in Google Scholar
[14] Bernskoetter W. H., Hazari N., Reversible Hydrogenation of Carbon Dioxide to Formic Acid and Methanol: Lewis Acid Enhancement of Base Metal Catalysts, Acc. Chem. Res., 2017, 50, 4, 1049-1058.10.1021/acs.accounts.7b00039Search in Google Scholar PubMed
[15] Sikander U., Sufian S., Salam M.A., Synthesis and structural analysis of double layered Ni-Mg-Al hydrotalcite like catalyst, Procedia Eng., 2016, 148, 261-267.10.1016/j.proeng.2016.06.559Search in Google Scholar
[16] OthmanM. R., Martunus Z. H., Fernando W. J. N., Synthetic hydrotalcites from different routes and their application as catalysts and gas adsorbents: a review, Applied Organometallic Chemistry, 2009, 23, 335-346.10.1002/aoc.1517Search in Google Scholar
[17] Saifullah B., Hussein M.Z.B., Inorganic nanolayers: structure, preparation, and biomedical applications, Int. J. Nanomed., 2015, 10, 5609-5633.10.2147/IJN.S72330Search in Google Scholar PubMed PubMed Central
[18] Srivastava V., Recyclable hydrotalcite clay catalysed Baylis– Hillman reaction, J. Chem. Sci., 2013, 125(5), 1207-1212.10.1007/s12039-013-0472-0Search in Google Scholar
[19] Upadhyay P.R., Srivastava V., Clays: An encouraging catalytic support, Curr. Catal., 2016, 5(3), 162-181.10.2174/2211544705666160624082343Search in Google Scholar
[20] Lakshmi Kantam M., Vijaya Kumar K., Sreedhar B., Asymmetric hydrogenation of ethyl pyruvate using layered double hydroxides–supported nano noble metal catalysts, Synth. Commun., 2007, 37(6), 959-964.10.1080/00397910601163893Search in Google Scholar
[21] Baskaran T., Christopher J., Sakthivel A., Progress on layered hydrotalcite (HT) materials as potential support and catalytic materials, RSC Adv., 2015, 5(120), 98853-98875.10.1039/C5RA19909CSearch in Google Scholar
[22] Wasserscheid P., Keim W., Ionic liquids—New “Solutions” for transition metal catalysis, Ang. Chem. Intern. Ed., 2000, 39, 3772-3789.10.1002/1521-3773(20001103)39:21<3772::AID-ANIE3772>3.0.CO;2-5Search in Google Scholar
[23] Finn M., Ana N., Voutchkova-Kostal A., Immobilization of imidazolium ionic liquids on hydrotalcites using silane linkers: retardation of memory effect, RSC Adv., 2015,5, 13016-13020.10.1039/C4RA13839BSearch in Google Scholar
[24] Yang H., Han X., Li G., Wang Y., N-Heterocyclic carbene palladium complex supported on ionic liquid-modified SBA-16: an efficient and highly recyclable catalyst for the Suzuki and Heck reactions, Green Chem., 2009, 11(8), 1184-1193.10.1039/b904136bSearch in Google Scholar
[25] Nishimura S., Takagaki A., Ebitani K., Characterization, synthesis and catalysis of hydrotalcite-related materials for highly efficient materials transformations, Green Chemistry, 2013, 15, 2026-2042.10.1039/c3gc40405fSearch in Google Scholar
[26] Sahoo M., Parida K. M., Pd(II) loaded on diamine functionalized LDH for oxidation of primary alcohol using water as solvent, Appl. Catal., A, 2013, 460-461.10.1016/j.apcata.2013.04.003Search in Google Scholar
© 2018 Vivek Srivastava, published by De Gruyter
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
