Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of α,β-unsaturated esters

Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of α,β-unsaturated esters

G Model ARTICLE IN PRESS CATTOD-10497; No. of Pages 12 Catalysis Today xxx (2016) xxx–xxx Contents lists available at ScienceDirect Catalysis Tod...

2MB Sizes 3 Downloads 63 Views

G Model

ARTICLE IN PRESS

CATTOD-10497; No. of Pages 12

Catalysis Today xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Catalysis Today journal homepage: www.elsevier.com/locate/cattod

Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters Natalia Candu, Diana Paul, Ioan-Cezar Marcu, Vasile I. Parvulescu, Simona M. Coman ∗ Department of Organic Chemistry, Biochemistry and Catalysis, Faculty of Chemistry, University of Bucharest, Regina Elisabeta Blvd., No. 4-12, Bucharest 030016, Romania

a r t i c l e

i n f o

Article history: Received 29 September 2016 Received in revised form 14 November 2016 Accepted 1 December 2016 Available online xxx Keywords: Levulinic acid LDH Intercalated LDH hybrid organocatalyst trans-methylcinammate epoxidation Phenyl glycidates

a b s t r a c t Green efficient organocatalytic heterogeneous epoxidation of trans-methylcinnamate was achieved on LDH-intercalated levulinate using H2 O2 /acetonitrile as oxidant. The layered double hydroxide (LDH) was prepared by co-precipitation starting from an aqueous solution of Mg(NO3 )2 ·6H2 O and Al(NO3 )3 ·9H2 O. The preparation of the LDH hybrid organocatalyst was carried out following four different routes: i) step-by-step ion exchange in air (LEV@LDH-air); ii) one charge ion-exchange in N2 atmosphere (LEV@LDH-N2 ); iii) direct synthesis by co-precipitation (LEV@LDH-pp); and the iv) reconstruction method (LEV@LDH-mem). The synthesized catalysts were characterized using different techniques such as adsorption-desorption isotherms of nitrogen at −196 ◦ C, powder X-ray diffraction (XRD), differential thermal analysis and thermogravimetric analysis (TG-DTA), and diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy techniques. The levulinate was successfully intercalated by ion exchange and co-precipitation. In the investigated series the LEV@LDH-pp catalyst showed the highest performances at 40 ◦ C and 24 h (Ccinnamate = 21.4%, Sglycidate = 100%). © 2016 Elsevier B.V. All rights reserved.

1. Introduction The broad utility of epoxides justifies the current interest of synthetic chemists for more efficient and practical methods for the epoxidation of a wide variety of olefins [1,2]. Recent reports on the oxidation with hydrogen peroxide in the presence of different catalysts like iron in combination with porphyrin [3] or pyridine-2,6-dicarboxylic acid and different amines [4], polyoxovanadometalate [5], and different other metal catalysts [6] already confirmed the advantages of this strategy. However, the leaching of the metals and their presence in the synthesis products are nowadays considered as major limitations for the industrial development of the epoxidation processes. As another alternative, the epoxidation of olefins using dioxiranes as oxidants has also been studied. In this respect, both the traditional and the current methods proposed working in chlorinated solvents using peroxy acids [7] or, very often, mchloroperbenzoic acid (m-CPBA) as epoxidation agents [8,9]. With m-CPBA the oxidation occurs via an attack by electron-rich substrates such simple alkenes or alkenes carrying a variety of

∗ Corresponding author. E-mail address: [email protected] (S.M. Coman).

functional groups (e.g., ethers, esters, ketones) to the weak O O bond. Following this, the oxygen is transferred from the intermediate dioxirane to the substrate [10]. However, for cinnamic esters presenting electron withdrawing groups, epoxidation with m-CPBA is extremely slow [11], while in the case of cinnamates having electron donating groups, the epoxides react with the liberated m-chlorobenzoic acid to give diol ester derivatives as the major product [11]. On the other hand the epoxidation of the (E)-cinnamic acid, which had failed to react with m-CPBA in CH2 Cl2 , was achieved ® by Curci et al. [12] with high yields (95%) using as oxidant Oxone (a triple salt with the formula 2KHSO5 x KHSO4 x K2 SO4 ) dispersed in acetone. On this basis, organocatalysts (ie, acetone in the previous example [12]) able to generate dioxirane, have been suggested as possible alternative to the homogeneous metal catalysts for the ® epoxidation reactions [13,14]. Among these, Oxone in conjunction with a D-fructose-derived ketone, also known as Shi’s ketone organocatalyst, already proven to be highly active in the epoxidation of a wide variety of (E)-di- and trisubstituted alkenes [15]. ® However, as an inconvenient, Oxone may gradually decompose under the basic reaction conditions required for this type of epoxidation (ie, epoxides are usually more stable under basic conditions). ® Therefore, a greener alternative to the epoxidation with Oxone , also imposing substantially less solvent and salts, might be the use

http://dx.doi.org/10.1016/j.cattod.2016.12.007 0920-5861/© 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 2

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

of hydrogen peroxide (H2 O2 ), as primary oxidant, in combination with acetonitrile [16]. In this system, the peroxyimidic acid is likely to be the active oxidant for the formation of the dioxirane. However, in spite of the recent improvements and irrespective to their advantages upon the metal-complexes (e.g., robust, inexpensive, readily available, non-toxic), the homogeneous organocatalysis still preserve the main “classical” homogeneous catalysis disadvantage, difficulties in the recovery and the re-use of the organocatalytic species. Although in many cases it has been claimed that the absence of the metal lowers the price of the catalyst, based on the green chemistry principles, the resulted wastes generate technological inconveniences. Therefore, making organocatalysts insoluble and consequently, easily recoverable and reusable may be the way for a sustainable development in this field. In the nowadays chemistry organocatalysts also have an important place in asymmetric epoxidations providing an efficient way for the synthesis of important building blocks in the production of pharmaceuticals [17]. In the context of this work, not long ago, some of us reported the synthesis of an efficient magnetic and recyclable green SBILC (Supported Basic Ionic Liquid Catalyst)-based heterogeneous Shi-ketone organocatalysts for the asymmetric epoxidation of the trans-methylcinnamate to chiral (2R, 3S)-phenyl glycidate (Y = 35%, Sglycidate = 100% and e.e. of 100%) [18]. This synthesis strategy combined a number of concepts leading to an efficient green catalyst for the asymmetric epoxidation of the trans-methylcinnamate, as: i) di-acetate chiral ketone Shi as organocatalyst – generates suitable conformations for the epoxidation of trans-methylcinnamate with enhanced stability towards the basic species present in the reaction mixture; ii) hydrogen peroxide as oxidizing agent − compared to Oxone significantly reduced the amount of the solvent and introduced salt, which is in line with the principles of the green chemistry; iii) base ionic liquids – ensure the benefits of using a base in the epoxidation reaction, but expel the large amounts of generated wastes, corrosion problems and issues related to the organocatalyst recovery; iv) magnetic nanoparticles – ensure an efficient and elegant solution for the issue of recycling and reuse of the catalyst; v) MWCNT support – ensures the chemical stabilization of the ultrafine magnetic nanoparticles and offers the advantage of the simple functionalization. Although catalytically highly efficient, the laborious preparation procedure of this system involving large quantities of raw materials and high number of steps (in some of which dangerous reagents to the environment were used) restrain its development on a broader scale. Therefore, we invested a considerable effort by applying the green chemistry rules for the development of a novel environmentally more acceptable synthesis of an efficient catalytic system for the epoxidation of cinnamates. For comparison, catalytic results were compared with those obtained in homogeneous conditions by using sodium levulinate (ie, [Na][LEV]) as organocatalyst, a readily biodegradable and low toxicity ionic liquid, as recently classified [19]. Here we explored novel potential renewable organocatalystsintercalated layered double hydroxide (LDH) materials as efficient solid catalysts for the epoxidation of trans-methyl cinnamate to methyl-3-phenyl-glycidate. Layered double hydroxides (LDHs) or hydrotalcite-like compounds are anionic clays with the general M3+ (OH)2 An− · mH2 O with 0.2 ≤ x ≤ 0.4 [20]. M2+ formula of M2+ 1−x x x/n and M3+ cations are hexa-coordinated to hydroxyl groups forming brucite-like sheets which stack to create a layered structure. A large variety of inorganic and organic An− anions can be intercalated in the interlayer space to compensate the positive charge introduced by the M3+ cations partially replacing M2+ ones in the layers. Up to now, a high number of nanohybrids obtained by intercalating catalytically active species in LDHs such as metal complexes, oxometalates and polyoxometalates were investigated as

heterogeneous catalysts in liquid-phase oxidation [21]. However, to the best of our knowledge, levulinate species were never intercalated in LDHs to obtain solid catalysts for liquid-phase oxidation. Several advantages like the simple catalyst preparation methodology from cheap and renewable raw materials, the heterogeneous character of the organocatalyst, the use of H2 O2 and the absence of any inorganic soluble base were envisaged. All these elements can successfully compensate any additional costs related to the separation of enantiomers from the racemic (±)-phenyl glycidate consisting in the kinetic resolution of racemic (±)-phenyl glycidate through a lipase-catalyzed asymmetric hydrolysis [22]. The catalytic results are discussed by comparison with those obtained with previously developed and reported SBILC (Supported Basic Ionic Liquid Catalyst)-based heterogeneous Shi-ketone organocatalysts [18].

2. Experimental section 2.1. Catalysts preparation 2.1.1. Preparation of the reference LDH (layered double hydroxide) The layered double hydroxide (LDH) was prepared by coprecipitation at a constant pH of 10, starting from an aqueous solution of Mg(NO3 )2 ·6H2 O and Al(NO3 )3 ·9H2 O. The quantities of the Mg and Al salts were calculated for a molar ratio of Mg/Al = 3/1 in the final material. The preparation procedure was as follows: an aqueous solution containing 44.8 g of Mg(NO3 )2 ·6H2 O and 21.75 g of Al(NO3 )3 ·9H2 O in 100 mL of deionized water was first prepared. Then, a solution containing 8.294 g of Na2 CO3 ·10H2 O in 100 mL of deionized water was mixed with 100 mL solution of NaOH 2 M to obtain the precipitating agent. The solution of salts and the precipitating agent were simultaneously added dropwise into a beaker containing 200 mL of deionized water at room temperature with controlled rate to maintain the pH close to 10. After complete precipitation, the slurry was aged at 80 ◦ C overnight under vigorous stirring. The suspension was then separated by centrifugation, washed with deionized water and finally dried at 80 ◦ C for 12 h. The obtained LDH material was denoted MgAl-LDH.

2.1.2. Preparation of the intercalated LDH hybrid organocatalyst [Na][LEV] (density = 1.345 g mL−1 , dynamic viscosity = 36.21 mPa s) was prepared as follow: 17.54 g of NaHCO3 were dissolved in 160 mL of distilled water and to this solution 9.44 mL levulinic acid (LA) were added (NaHCO3 /LA molar ratio of 2.7/1.2). The mixture was stirred for 1 h, at room temperature and then the solvent was removed at 80 ◦ C, for 2 h. The intercalation of [Na][LEV] in the LDH structure followed four different routes: i) Step-by-step ion exchange in air (LEV@LDH-air): 3 g of MgAl-LDH and 1 g (7.2 mmoles) of [Na][LEV] were suspended in 150 mL of deionized water. The mixture was heated to 80 ◦ C for 9 h under stirring and then stirred at room temperature, overnight. The resulted product was centrifuged, washed twice with deionized water, four times with ethanol, once with acetone, and finally dried. The recovered product was then suspended with another 1 g (7.2 mmoles) of [Na][LEV] in 150 mL of deionized water, under the same conditions as in the first treatment and this procedure was repeated four times. The final [Na][LEV] loading was 21.7 mmoles. In the last treatment cycle the mixture was simply stirred without any [Na][LEV] addition. After the separation and washing, the final product was dried in oven at 80 ◦ C, for 8 h.

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model

ARTICLE IN PRESS

CATTOD-10497; No. of Pages 12

N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

3

Table 1 Preparation methods for the levulinate-intercalated LDH. Sample

LEV@LDH-air

LEV@LDH-N2

LEV@LDH-pp

LEV@LDH-mem

Precursors nature

• MgAl-LDH • [Na][LEV]

• MgAl-LDH • [Na][LEV]

• Metal nitrates salts • Levulinic acid

• LDH-derived mixed oxide (calcined at 450 ◦ C) • [Na][LEV]

Intercalation method

Ion-exchange (four steps, air)

Ion-exchange (one-step, nitrogen)

Co-precipitation

Reconstruction

Table 2 Catalytic performances of [Na][LEV] in the epoxidation of trans-methylcinnamate to phenyl-glycidate. Entry

Ionic liquid

T, ◦ C

H2 O2 , mL

C, %

S, %

E.e. (%)

1 2 3 4 5 6 7a

[BMIM][OH]

24 24 0 24 40 40 40

0.8 0.8 0.8 0.8 0.8 1.5 1.5

25.2 28.0 0.2 6.4 11.6 12.4 22.4

100 100 0 0 93.9 92.7 100

100 (2R, 3S) 100 (2R, 3S) – – – – –

[Na][LEV]

Reaction conditions: Entry 1: 43.5 mg ketone 5, 1.5 mL ionic liquid, 0.162 g (1 mmol) TMC, 1.5 mL AcCN, 24 h; Entry 2: 87.0 mg ketone 5, 1.5 mL ionic liquid 0.162 g (1 mmol) TMC, 1.5 mL AcCN, 24 h [18]. This work: 0.1 mL (0.1345 g, 1 mmol) [Na][LEV] 0.162 g (1 mmol) TMC, 1.5 mL AcCN, 0.138 g K2 CO3 , 24 h. a 0.3 mL (0.403 g, 3 mmol) [Na][LEV].

• One charge ion-exchange in N2 atmosphere (LEV@LDH-N2 ): 2.5 g of LDH was mixed with 2.5 g (18.1 mmoles) of [Na][LEV] in 150 mL of deionized water. The slurry was stirred at RT, for 18 h in nitrogen atmosphere. After 18 h the product was filtered, washed with distilled water until a pH = 7 and dried in oven at 80 ◦ C, for 8 h. • Direct synthesis by co-precipitation (LEV@LDH-pp): LDH was prepared by co-precipitation in the presence of levulinic acid. A mixed solution of 30 mL, 3.84 g Mg(NO3 )2 ·6H2 O and 2.81 g Al(NO3 )3 ·9H2 O was added dropwise to 150 mL solution containing 7.5 mmoles levulinic acid (Mg2+ /Al3+ /LA molar ratio = 2/1/1), 50 mL, 2 g NaOH 1 M and Na2 CO3 ·10H2 O in deionized water, under nitrogen atmosphere and at 40 ◦ C. The resulted product was filtered after aging overnight, washed with distillated water and dried under vacuum at 60 ◦ C [23]. • Reconstruction method (LEV@LDH-mem): 5 g of the reference MgAl-LDH were calcined at 450 ◦ C to the corresponding mixed oxide. Then, the resulted oxide (2 g) was mixed with 40 mL of [Na][LEV] under vigorous stirring at room temperature, for 3 h. The precipitate was filtered, washed with distilled water until pH = 7 and dried at 80 ◦ C, for 12 h. Fig 1. XRD diffraction pattern of the reference MgAl-LDH sample.

A summary of the intercalation methods and prepared materials is listed in Table 1: 2.2. Catalysts characterization Both pristine LDH and organocatalyst-based LDH samples were characterized using various techniques such as adsorptiondesorption isotherms at −196 ◦ C, powder X-ray diffraction (XRD), differential thermal analysis and thermogravimetric analysis (TG-DTA), and diffuse reflectance infrared Fourier transform spectroscopy (DRIFT). Adsorption-desorption isotherms were measured with a Micromeritics ASAP 2020 apparatus. Before measurements the samples were degassed at 110 ◦ C, under vacuum, for 8 h. Specific surface areas were calculated by BET method on a p/p0 range of 0.07–0.20. The pore size distribution was determined from the desorption branch by using the BJH formalism. Powder X-ray diffraction (XRD) patterns were recorded on a Shimadzu XRD-7000 diffractometer using Cu K␣ radiation (␭ = 1.5418 Å, 40 kV, 40 mA) at a step of 0.02◦ and a scanning speed of 2◦ min−1 in the 5–90 ◦ 2 range. The differential thermal and thermogravimetric analysis (TG-DTA) were carried out using a TG-DTA analyzer (Shimadzu

DTG-60 Simultaneous DTA-TG Apparatus), under air atmosphere, with a rate flow of 10 mL min−1 at a heating rate of 10 ◦ C min−1 , from RT to 700 ◦ C and using alpha-alumina as reference. Diffuse reflectance infrared Fourier transform spectroscopy (DRIFT) spectra were collected with a Thermo 4700 spectrometer (200 scans with a resolution of 4 cm−1 ) in the range of 600–4000 cm−1 . 2.3. Catalytic tests 2.3.1. Homogeneous catalytic epoxidation Homogeneous catalytic epoxidation with H2 O2 /acetonitrile, as oxidizing agent, and [Na][LEV], as organocatalyst, were carried as follows: trans-methylcinnamate (TMC, 0.162 g), [Na][LEV] (0.10.3 mL, d = 1.345 g mL−1 ) (TMC/[Na][LEV] molar ratio of 1/1-1/3), K2 CO3 (0.138 g) and acetonitrile (1.5 mL) were introduced in a glass vial and H2 O2 30% (0.4/0.8 mL) was added to this mixture. The reaction mixture was stirred at 0 − 40 ◦ C, for 24 h. After completion, the reaction products were extracted in hexane (3 × 10 mL), washed with NaCl 31% (1 × 10 mL), dried with Na2 SO4 and filtrated. The

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 4

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

Fig. 2. Nitrogen adsorption-desorption isotherms and pore size distribution for the MgAl-LDH sample.

Fig. 3. The TG-DTA profile of reference MgAl-LDH.

filtrate was concentrated under vacuum and analyzed by GC-FID chromatography.

2.3.2. Heterogeneous catalytic epoxidation All LEV-intercalated LDH samples were tested in the epoxidation of TMC and compared to the behavior of homogeneous catalytic epoxidation. Part of the reactions was carried out without the addition of K2 CO3 in the reaction mixture. After completion, the solid catalyst was separated by filtration, the filtrate was concentrated

under vacuum and the resulted reaction products were analyzed by GC-FID chromatography. The reaction products were analyzed by GC-FID chromatography (Shimadzu GC-2014 apparatus). Retention times: trans-methyl cinnamate (TMC) = 13.9 min, methyl-3phenyl glycidate: 14.5 − 14.9 min. The identification of the products was made using a GC–MS apparatus (GC Trace 2000 DSQ system coupled with an MS detector, from Thermo Electron Corporation) equipped with a Factor Four VF-5HT column with the following characteristics:

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

5

Fig. 4. IR spectrum of the reference MgAl-LDH.

Fig. 5. The IR spectra of the reference MgAl-LDH, LEV@LDH-air and LEV@LDH-N2 . The IR spectra of levulinic acid (LA) and sodium levulinate ([Na][LEV]) are given in inset.

0.32 mm × 0.1 ␮m × 15 m working with a temperature program at a pressure of 0.38 Torr with He as carrier gas. 3. Results and discussion One of the green chemistry concerns is the use of volatile organic compounds as solvents. Benign alternatives are highly recommended, and among the possible solutions, ionic liquids are

often considered [24]. Thus, the low volatility of ionic liquids effectively eliminates a major pathway for environmental release and contamination. However, this property is distinct from toxicity. Unfortunately, in many cases, the aquatic toxicity of ionic liquids is at least as severe as those of many current solvents [25]. A truly green IL-based process should consider nonenvironmental harmful ILs. Thus, the use of structures derived from biomaterials often results in higher biodegradability. In

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 6

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

Fig. 6. The TG-DTA profile of LEV@LDH-N2 .

Fig. 7. Nitrogen adsorption-desorption isotherms and pore distribution of LEV@LDH-N2 .

this line, [Na][LEV] can be classified as a readily biodegradable and low toxicity ionic liquid, in agreement with [19]. Besides these attractive benign properties, the results in the dioxirane epoxidation of the electron-deficient trans-methylcinnamate recommends [Na][LEV] as an efficient organocatalyst (Table 2). Note that in this table, the difference to reach 100% in selectivity is given by diols which were produced as a result of the non-selective decomposition of hydrogen peroxide to water and molecular oxygen. Once formed, water molecules attack the strained epoxide cycle with the consecutive formation of diols.

As Table 2 shows, the best catalytic performances (conversion, C = 22.4% and selectivity, SMPG = 100%) of the [Na][LEV] are slightly inferior to those determined in the presence of the Shi – like ketone 5 and [BMIM][OH] as IL (C = 25–28%, SMPG = 100%). Moreover, to reach these performances the [Na][LEV] needs higher temperatures (40 ◦ C versus 24 ◦ C), higher amounts of hydrogen peroxide (1.5 versus 0.8 mL) and the presence of K2 CO3 in the reaction medium. The increased yield in the presence of the Shi − like ketone 5 is a consequence of the electron-withdrawing acetate substituents at the ␣-carbonyl position [18]. Obviously, in the case of [Na][LEV]

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

Fig. 8. XRD patterns of the LEV-intercalated MgAl-LDH through ion-exchange method.

the presence of the carboxylate group at the ␥-carbonyl position has a diminished effect upon the reactivity of the C O. On the other hand, unsubstituted methyl cinnamate is an electrondeficient alkene, thus, it was expected to be less reactive toward in situ generated electrophilic dioxirane. As it was mentioned above, from environmental hazard point of view these preliminary catalytic results suggest that [Na][LEV] has great potential as green organocatalyst for epoxidation due to its biodegradability and low toxicity. Contrarily, studies of toxicity and biodegradability of imidazolium based ionic liquids (e.g., [BMIM][OH]) clearly showed a low or even lack of biodegradability while the toxic effect increases with the chain length of the side chain from C1 to C8 [26]. Therefore, the presented concept deserves to be extended to the epoxidation of cinnamate esters with electron-donating substituents in their structure, able to increases the reactivity toward the electrophilic dioxirane.

7

Also, due to the mass transfer limitations in ILs, the chemical industry still prefers to use heterogeneous catalysts. Combinations of heterogeneous catalysts and ILs are also preferred because of the easy separation. Thus, further improvements using supported ILs may lead to an increased efficiency. Encouraged by this idea, in the next step of the present study heterogeneous catalytic assemblies were prepared by intercalating [Na][LEV] or levulinic acid (LA) in LDH materials by different methodologies. The XRD pattern (Fig. 1) of the reference MgAl-LDH sample shows typical pattern of hydrotalcite-like materials, with diffraction lines indexed on a hexagonal unit cell with the space group R–3 m [27]. The sharp and symmetric diffraction lines at 2␪ of 11.4, 22.8 and 34.6◦ , assigned to (003), (006) and (009) planes [28], indicate a well-formed crystalline layered structure. The basal spacing distance d003 , determined from the position of the (003) reflection is 7.77 Å and corresponds to carbonate as compensating anion. Taking into consideration that the thickness of the brucite-like layer is 4.8 Å, the height of the gallery is 2.97 Å [29]. The lattice c (c = 3x d003 ) and a (a = 2x d110 ) parameters have been valued at 23.32 and 3.06 Å, respectively. The MgAl-LDH sample had a BET surface area of 42 m2 /g and a bimodal pore size distribution with two maxima at 3.5 and 14.6 nm, respectively (Fig. 2). The hysteresis loop suggests mesopores with a regular geometry and relatively uniform size. No microporosity has been detected using the micropore program. The molecular diameter of the N2 molecule (3.65 Å [30]) did not allow it to enter into the space between the layers (height of 2.97 Å (XRD measurements)). The evidenced mesoporous cavities are the result of the interconnection of multiple LDH layers [31]. The TG profiles comprise three characteristic main steps matching the literature reports: i) the loss of physically adsorbed and interlayer water molecules at temperatures lower than 200 ◦ C; ii) the dehydroxylation of the double hydroxide layers, at 200–400 ◦ C, and iii) the complete dehydroxylation together with the decomposition of the interlayer carbonate anions with the formation of mixed oxides, at higher temperatures than 400 ◦ C (Fig. 3) [32,33]. During the thermal decomposition 46.7% of the LDH sample is lost. Fig. 4 shows the IR spectrum that highlights the presence of the OH− (1647 cm−1 ) and CO3 2− (1430 cm−1 ) anions between the

Fig. 9. XRD pattern of the LEV@LDH-pp prepared via the direct intercalation (co-precipitation) of the levulinic acid in the MgAl-LDH structure.

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 8

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

Fig. 10. Nitrogen adsorption-desorption isotherms and pore distribution of LEV@LDH-pp samples.

Fig. 11. The TG-DTA profile of LEV@LDH-pp.

LDH layers (Fig. 4). The broad band at 3588 cm−1 is assigned to the (-OH) hydroxyl stretching vibration of the isolated and hydrogen bonded to the octahedral layer and water molecules [34]. A characteristic absorption band is the broad shoulder present around 3000 cm−1 that is attributed to carbonate hydrogen-bonded to the LDH hydroxides [35]. The more carbonate present in the LDH, the stronger this shoulder appears. The levulinate anion exchange depended on the preparation route (step-by-step ion exchange in air (LEV@LDH-air), single-charge ion exchange in nitrogen (LEV@LDH-N2 ) or direct co-precipitation (LEV@LDH-pp)). It occurred in different concen-

trations and with different orientations. The immobilization of LEV via the reconstruction route (LEV@LDH-mem) failed. The IR spectra of the reference and [Na][LEV]-intercalated MgAlLDH catalysts via the ion exchange are presented in Fig. 5. As Fig. 5 shows, all IR spectra display the broad band at 3400–3600 cm−1 due to the (-OH) hydroxyl stretching vibration of the isolated and hydrogen bonded to the octahedral layer and water molecules [34]. The shoulder at 3053 cm−1 is attributed to the hydrogen bonding of H2 O to CO3 2− ions in the interlayer space while the carbonate peak is observed at 1424 cm−1 . The new band with maximum at 1567 cm−1 in the IR spectra of the LEV@LDH-N2 sample can be associated with the asymmetric vibration of the ionized carboxylic

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model

ARTICLE IN PRESS

CATTOD-10497; No. of Pages 12

N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

9

Table 3 XRD data of LDH intercalated samples via ion-exchange routes. Entry

Sample

Interlayer spacing (Å)

Gallery height(Å)

1 2 3

MgAl-LDH LEV@LDH-air LEV@LDH-N2

7.77 7.78 7.65

2.97 2.98 2.84

groups from the levulinate molecules, while the band located at 1647 cm−1 is attributed to the OH stretching mode of the interlayer water [34]. These results indicate that the [Na][LEV] was intercalated as [LEV]␦− species by ion-exchange methodology in nitrogen atmosphere. Unfortunately, the ion exchange in air did not succeed the IR spectra of the obtained LEV@LDH-air sample being similar to that of the reference MgAl-LDH sample (Fig. 5). A similar behavior was evidenced by Kanoh et al. [36] for the intercalation of short-chain acids (e.g., acetic butyric or hexanoic acids) in LDH structures. Indeed, the IR spectrum of sodium levulinate ([Na][LEV]) (Fig. 5) presents strong bands at 1529 and 1413 cm−1 , assigned to as (OCO) and s (OCO) of carboxylate groups, respectively. Although with a lower intensity, the absorption peak of the non-dissociated carboxylic acid group (1720 cm−1 ) is still present, indicating that not all LA was transformed into corresponding [Na][LEV]. The presence of the sharp band at 3394 cm−1 associated to the (-OH) groups confirms this assumption. However, in the ion exchange process [Na][LEV] participated as a whole entity. No absorption bands characteristic to the non-dissociated carboxylic acid group (1720 cm−1 ) were evidenced for the LEV@LDH-N2 sample (Fig. 5). The intercalation of the [LEV]␦− species is also confirmed by the TG-DTA analysis. The TG profile (Fig. 6) of the LEV@LDH-N2 sample shows an increased mass loss for the LEV-intercalated sample, accompanied by an endothermic effect centered at 358 ◦ C (Fig. 6). This peak is associated to the vaporization of carboxylic acid and its degradation product. Additional mass loss steps until 200 ◦ C in the case of LEV@LDH-N2 can be associated to the decomposition of the residual ethanol from the sample preparation. The measured loading of the intercalated levulinate species (mass loss step between 330 and 405 ◦ C) was of 14.5 wt%. The decrease of the BET surface area of LEV@LDH-N2 was negligible, from 42 to 39 m2 /g. However, the pore size distribution changed significantly displaying in this case a mono-modal distribution with a maximum at around 10.6 nm (Fig. 7) and the hysteresis loop suggested the presence of mesopores with different geometries. The disappearance of the narrow mesopores evidenced for the pristine MgAl-LDH (35 Å) and the enlargement of the pores may confirm that part of the levulinate anions were inserted, by grafting, in the mesoporous cavities. Neither the LA kinetic diameter (5.7 Å [37]) nor the X-ray diffraction (XRD) pattern of the LEV@LDH-N2 (Fig. 8) supports an interlayer insertion. Indeed, as the XRD patterns of LEV@LDH-air and LEV@LDH-N2 samples (Fig. 8) and associated XRD data (Table 3) showed the interlayer distance (d003 ) characteristic to the pristine MgAl-LDH material remained unchanged after the intercalation process irrespective of the ion-exchange methodology. Table 3 However, the IR spectra, adsoption-desorption isotherms and TG-DTA measurements confirm the presence of LEV in LEV@LDH-N2 material. Cumulating these information it is very sure that the LEV species have not penetrated in between the LDH layers but are only anchored at the corners inside the mesoporous cavities. In this way the LDH layers remained unaffected, and only the mesoporosity generated through the layers interconnection has been affected. The result is not surprising if, in addition, we take into consideration the strong electrostatic interaction of the carbonate anions with LDH caused by the high anion charge density [38]. Such

Fig. 12. The IR spectrum of the reference MgAl-LDH and LEV@LDH-pp samples prepared via the co-precipitation method.

an interaction makes more complicated the levulinate-carbonate ion exchange process. However, the co-precipitation route led to materials in which part of carbonate groups were replaced with levulinate anions. In this case, the XRD pattern (Fig. 9) clearly indicates the existence of two kinds of inserted anionic species in the LDH support. This is confirmed by the two different basal distances of both (003) and (006) X-ray diffraction planes, that are associated to two anionic species with different sizes. The intensity of the diffraction lines corresponding to the phase containing the levulinate species is high, indicating a high quantity of embedded levulinate species in the LDH material. In addition, the shift of the (003) reflection line to lower 2␪ value corresponds to an enlargement of the interlayer distance (d003 ) from 7.77 Å (characteristic to the MgAl-LDH) to ca. 7.95 Å (characteristic to the LEV@LDH-pp). The interlayer distance value of 7.95 Å corresponds to the space between the layers (gallery) plus the brucite-like layer (4.8 Å). Thus, the gallery height is ca. 3.15 Å. Obviously, this space is not enough large to accommodate levulinate ions in a perpendicular position to the MgAl-LDH layers but their intercalation in a parallel position to the LDH layers might be possible. Using this preparation methodology, as both textural characteristics and TGDTA measurements confirmed, a large content of levulinate species has been encapsulated. Indeed, after the levulinate incorporation, the BET surface area drastically decreased (11 m2 /g), with a monomodal distribution of the pore size (centered at 6.3 nm) (Fig. 10), and the mass loss (330–407 ◦ C) of carboxylic species measured from TG was of 22.4 wt% (Fig. 11). Interesting enough, in this case, the DTA profile shows the presence of two organic species which decompose very fast and at very close temperatures (373 and 394 ◦ C, respectively) (Fig. 11). The existence of these species is also evidenced in IR spectra (Fig. 12). Although with a low intensity, the presence of the bands at 1626 and 1586 cm−1 indicates the intercalation of the levulinic acid in the R-COOH form, where H+ is ionized in the interlayer as C(O)O␦− H␦+ . The presence of the band centered at 1720 cm−1 characteristic to the free levulinic acid also indicate that although not all levulinic acid was transformed in levulinate during the synthesis, levulinic acid was incorporated in LDH framework in both ionized and nonionized forms. To sum up, the incorporation of levulinate (i.e., the organocatalyst) into the LDH structure was successful as following: i) for

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 10

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

Fig. 13. Schematic representation of the LEV@LDH-N2 (A) and LEV@LDH-pp adducts (B).

LEV@LDH-N2 (prepared via one-step ion exchange, in protected atmosphere) the anionic levulinate species have not penetrated in-between the LDH layers but they were anchored at the framework corners (this route left unaffected the LDH layers (Fig. 8) with changes in the textural characteristic of the pristine MgAl-LDH (Fig. 7)), and ii) for LEV@LDH-pp (prepared via co-precipitation) leading to two kinds of species, i.e., ionized and non-ionized forms of levulinic acid; most likely, these correspond to a) intercalated species in a parallel position to the LDH layers and b) grafted species at the framework corners. This procedure affects both the structural and textural characteristics of the pristine LDH. The amount of incorporated levulinate species is thus higher for LEV@LDH-pp than for the LEV@LDH-N2 sample (22.4 wt% versus 14.5 wt%). Obviously, the anchored species are preferable to intercalated ones, because, the catalytic sites of the organocatalyst ( C O group) are more easily accessible for bulky reactants and products (ie transmethylcinnamate and phenyl-glycidate). A schematic picture of the resulted LEV@LDH-N2 and LEV@LDHpp adducts is given in Fig. 13.

Since the LDH-type materials have basic properties and, therefore, efficiently transform hydrogen peroxide into a perhydroxyl anion on their surfaces [20], it may be expected that such systems can be developed for the epoxidation reactions of olefins and even of electron deficient olefins like 2-cyclohexen-1-one. However, in the case of the trans-methylcinnamate, reference MgAl-LDH does not show catalytic activity (Table 4). The comparison of the results obtained with LEV-intercalated LDH with those obtained from the homogeneous systems indicates that the yields in epoxides were slightly lower (21.4% yield, entry 7, Table 4 versus 22.4% yield, entry 7, Table 2) with the heterogeneous catalyst. Obiouvsly, in the case of LEV-intercalated materials the active sites (ie, C O groups of levulinate species) are in a much smaller amount (14.5% for LEV@LDH-N2 , and 22.4% for LEV@LDHpp, see TG-DTA results). However, their high dispersion on the LDH surface seems to favor the reaction rate of the epoxidation. Moreover, these species are located in different cavities of the LDH structure (i.e., gallery, narrow and larger mesopores) and in different structural orientations. Obviuosly, some of these molecules are not able to act as catalytic active sites, if their environment is

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model

ARTICLE IN PRESS

CATTOD-10497; No. of Pages 12

N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx Table 4 Catalytic performances of the LEV-intercalated LDH materials. Entry

Catalyst

Reaction time, h

C, %

SMPG , %

1 2 3 4a 5 6b 7c 8d

MgAl-LDH LEV@LDH-N2

18 2 18 24 24 24 24 24

0 2.3 3.1 0.7 1.5 5.1 12.3 21.4

– 77.9 86.7 100 100 100 100 100

LEV@LDH-pp

Reaction conditions: 0.05 g catalyst, 0.162 g TMC, 3 mL AcCN, 0.8 mL H2 O2 , 40 ◦ C. a room temperature. b 0.1 g catalyst. c 1.5 mL H2 O2 . d 0.1 g catalyst and 1.5 mL H2 O2 .

11

starting material for a wide number of chemical compounds and potential biofuels [42]. This contribution, for the first time in our best knowledge, offers a new opportunity of using levulinic acid as building block for the synthesis of novel solid materials with catalytic applications. 4. Conclusions In conclusion, an efficient LDH-based epoxidation hybrid catalyst was developed for the epoxidation of cinnamic esters choosing trans-methylcinnamate as reference. The reactions were carried out via in situ generated dioxiranes using levulinate as active organocatalyst and H2 O2 /acetonitrile mixture as oxidant. The low toxicity of sodium levulinate and the oxidation conditions obey the green chemistry requirements. Besides this, its anchoring on the corners of the LDH layers led to solid catalysts able to convert trans-methylcinnamate into corresponding epoxides in an efficient manner (Ccinnamate = 21.4%, Sglycidate = 100%). Acknowledgment Prof. Simona M. Coman kindly acknowledges the Romanian National Authority for Scientific Research, CNCS – UEFISCDI, for the financial support (project PN-II-ID-PCE-2011-3-0041, Nr. 321/2011). References [1] [2] [3] [4]

Scheme 1. The mechanism of trans-methylcinnamate epoxidation with dioxirane.

not large enough to allow the diffusion of the reactant molecules (i.e., trans-methylcinnamate) and the formation of the spirotransition state, in which the plane of the oxirane is perpendicular to the plane of the C C ␲-system permeating the oxygen transfer [18,39] (Scheme 1). The kinetic diameter of trans-methylcinammate (6.7 Å) and the corresponding phenyl-glycinate product (7.1 Å) has been estimated from the molecular weight correlation, using the equation:  = 1.234(Mw )1/3 For aromatic hydrocarbons (Mw = molecular weight in g mol−1 ) [40]. This approximation method for the kinetic diameters of oxygenated molecules which do not have critical properties is reasonable, as Huber et al. [41] demonstrated not long ago. On this basis it is logical that such large molecules are not able to interact with the levulinic acid in between the LDH layers (LEV@LDH-pp sample) or very slowly with the levulinic acid embedded in the narrow mesopores of the LDH (LEV@LDH-pp and LEV@LDH-N2 ). This constraint in the narrow spaces can also explain the increase of the selectivity to phenyl-glycidate in time for a very low increase of the trans-methylcinnamate conversion (entries 2 and 3, Table 4). These experiments also demonstrated that the basicity of the LDH can obviate the need for the presence of an inorganic base (e.g., K2 CO3 ) in the system that is, indeed, an important achievement from the standpoint of the environmental concerns. Also, the total selectivity in epoxide (100%) and the use of a benign oxidation agent (hydrogen peroxide instead of Oxone) bring high advantages from the green chemistry point of view avoiding the generation of wastes. Levulinic acid, one of the Top 15 biomass-derived platform molecules, is used in different purposes: precursor for pharmaceuticals, plasticizers, and it is recognized as a building block or

[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

[22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

C. Laurt, S.M. Robert, Aldrichim. Acta 35 (2002) 47–51. M.J. Porter, J. Skidmore, Chem. Commun. (2000) 1215–1225. N.A. Stephenson, A.T. Bell, J. Mol. Catal. A: Chem. 275 (2007) 54–62. B. Bianca, A. Gopinathan, G. Feyissa Gadissa, S. Bjoern, G. Anne, J. Ralf, T. Man Kin, B. Matthias, J. Chem Asian, (2007) 521–529. N. Yoshinao, N. Keigo, K. Miyuki, Y. Kazuya, M. Noritaka, Angew. Chem. Int. Ed. 44 (2005) 5136–5141. B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2474. A. Murphy, G. Dubois, T.D.P. Stack, J. Am. Chem. Soc. 125 (2003) 5250–5251. I. Bombarda, L. Cezanne, E.M. Gayou, J. Flavour Frag. 19 (2004) 275–280. F. Fringuelli, R. Germani, F. Pizzo, G. Savell, Tetrahedron Lett. 30 (1989) 1427–1428. R. l. Danheiser, S.D. Burke, Handbook of Reagents for Organic Synthesis Oxidizing and Reducing Agents, John Willey & Sons, 1999. G. Moyna, H.J. Williams, A.I. Scott, Synth. Commun. 26 (1996) 2235–2239. R. Curci, M. Fiorentino, L. Troisi, J.O. Edwards, R.H. Pater, J. Org. Chem. 45 (1980) 4758–4760. P.I. Dalko, L. Moisan, Angew. Chem. Int. Ed. 43 (2004) 5138–5175. A. Dondoni, A. Massi, Angew. Chem. Int. Ed. 47 (2008) 4638–4660. Y. Tu, Z.X. Wang, Y. Shi, J. Am. Chem. Soc. 118 (1996) 9806–9807. Z.X. Wang, Y. Shi, J. Org. Chem. 63 (1998) 3099–3104. I. Ojima, Catalysis Asymmetric Synthesis, Wiley-VCH Inc, Weinheim, 2000. N. Candu, C. Rizescu, I. Podolean, M. Tudorache, V.I. Parvulescu, S.M. Coman, Catal. Sci. Technol. 5 (2015) 729–737. M. Markiewicza, J. Maszkowska, V. Nardello-Rataj, S. Stolte, RSC Adv. 6 (2016) 87325–87331. D. Tichit, B. Coq, CATTECH 7 (2003) 206–217. I.C. Marcu, A. Urd˘a, I. Popescu, V. Hulea, Layered double hydroxides −based materials as oxidation catalysts, in: M.V. Putz, M.C. Mirica (Eds.), Sustainable Nanosystems Development, Properties, and Applications, IGI Global, Hershey, 2016, pp. 59–121. A. Gentile, C. Giordano, C. Fuganti, L. Ghirotto, S. Servi, J. Org. Chem. 57 (1992) 6635–6637. S. Aisawa, S. Takahashi, W. Ogasawara, Y. Umetsu, E. Narita, J. Solid State Chem. 162 (2001) 52–62. T. Akiyama, A. Suzuki, K. Fuchibe, Synlett (2005) 1024–1026. C. Pretti, C. Chiappe, D. Pieraccini, M. Gregori, F. Abramo, G. Monni, L. Intorre, Green Chem. 8 (3) (2006) 238–240. A. Romero, A. Santos, J. Tojo, A. Rodriguez, J. Hazard. Mater. 151 (2008) 268–273. H.C.B. Hansen, C.B. Koch, Appl. Clay Sci. 10 (1–2) (1995) 5–19. M. Raciulete, G. Layrac, D. Tichit, I.C. Marcu, Appl. Catal. A 477 (2014) 195–204. S. Miyata, Clays Clay Miner. 23 (1975) 369–375. D. Zhao, Z. Bai, F. Zhao, Mater. Res. Bull. 47 (2012) 3670–3675. P.S. Braterman, Z.P. Xu, F. Yarberry, Handbook of Layered Materials, in: S.M. Auerbach, K.A. Carrado, P.K. Dutta (Eds.), Marcel Dekker, 2004, pp. 373–474. V. Vagvolgyi, S.J. Palmer, J. Kristof, R.L. Frost, E. Horvath, J. Colloid Interface Sci. 318 (2008) 302–308. E. Kanezaki, Solid State Ion. 106 (1998) 279–284.

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007

G Model CATTOD-10497; No. of Pages 12 12 [34] [35] [36] [37] [38] [39]

ARTICLE IN PRESS N. Candu et al. / Catalysis Today xxx (2016) xxx–xxx

F.M. Labajos, V. Rives, M.A. Ulibarri, J. Mater. Sci. 27 (1992) 1546–1552. J.T. Kloprogge, R.L. Frost, J. Solid State Chem. 146 (1999) 506–515. T. Kanoh, T. Shichi, K. Takagi, Chem. Lett. (1999) 117–118. A. Sultana, T. Fujitani, Catal. Commun. 88 (2017) 26–29. F. Leroux, J.P. Besse, Chem. Mater. 13 (2001) 3507–3515. K.N. Houk, J. Liu, N.C. DeMello, K.R. Condroski, J. Am. Chem. Soc. 119 (1997) 10147–10152.

[40] H. Wang, M. Frenklach, Combust. Flame 96 (1994) 163–170. [41] J. Jae, G.A. Tompsett, A.J. Foster, K.D. Hammonda, S.M. Auerbach, R.F. Lobo, G.W. Huber, J. Catal. 279 (2011) 257–268. [42] J.J. Bozell, G.R. Petersen, Green Chem. 12 (4) (2010) 539–554.

Please cite this article in press as: N. Candu, et al., Levulinate-intercalated LDH: A potential heterogeneous organocatalyst for the green epoxidation of ␣,␤-unsaturated esters, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.12.007