Concerted Functions of Surface Acid-Base Pairs and Supported Copper Catalysts for Dehydrogenative Synthesis of Esters from Primary Alcohols.

Dehydrogenative synthesis of esters from primary alcohols proceeded efficiently over a ZrO2-supported copper catalyst. A variety of esters were obtained from primary alcohols as well as diols in good to high yields. The key to the dehydrogenative synthesis of esters is the concerted effect of the acid-base pairs on ZrO2 and metallic copper.

Introduction

Ester is one of the most important functionalities, and is found in a wide range of valuable chemicals such as fibers, inks, solvents, surfactants, fragrances, lubricants, food additives, and pharmaceuticals.[1] Although the condensation of alcohols with activated carbonyl compounds, such as acid chlorides and acid anhydrides, is the most convenient method for the preparation of esters, a stoichiometric amount of salt is inevitably formed as waste (eq 1 in Scheme ). Also, in dehydrative condensation of carboxylic acids with alcohols promoted by acid catalysts, the addition of dehydrating agents is indispensable for the high-yield production of esters (eq 2). On the other hand, the catalytic transformation of alcohols to esters in a dehydrogenative manner is a promising method thanks to the coproduction of only molecular hydrogen (eq 3).[2,3] In 2005, Milstein and co-workers first demonstrated an efficient and acceptorless dehydrogenative synthesis of esters from alcohols through the use of ruthenium complexes bearing a pincer-type ligand as catalysts.[4] Following this breakthrough, a series of homogeneous catalysts based on transition-metal complexes, such as Ru,[5−10] Rh,[11] Ir,[12] Re,[13] Os,[14] and Mn,[15] have been developed. On the other hand, efficient organic synthesis with heterogeneous catalysts is of great significance from the perspective of green and sustainable chemistry.[16−20] Shimizu and co-workers reported that supported Pt catalysts were effective for the dehydrogenative synthesis of esters.[21,22] Recently, much attention has also been paid to the development of an efficient catalytic system with abundant first-row transition metals. Particularly, many examples have demonstrated that supported Cu catalysts are highly effective for the dehydrogenative transformation of alcohols to the corresponding carbonyl compounds.[23−25] Dehydrogenative synthesis of esters from alcohols has been achieved by the use of supported Cu catalysts,[26−30] and the availability of ZrO2 as a support for Cu-catalyzed dehydrogenative synthesis of esters has also been explored. However, the correlations of the surface property of ZrO2 with the activity of Cu catalysts have not been fully understood.

Scheme 1

Synthetic Routes to Esters from Alcohols

Herein, we describe the dehydrogenative synthesis of esters from primary alcohols in the presence of supported Cu catalysts. The supports for the catalysts remarkably dominated the reaction efficiency and selectivity, and ZrO2-supported catalysts gave the corresponding esters with a high ester yield and selectivity. NH3- and CO2-temperature-programmed desorption (NH3-TPD and CO2-TPD) profiles revealed that metallic Cu and acid–base pair sites on the surface of ZrO2 promoted the catalytic transformation of primary alcohols to the corresponding esters with high selectivity.

Results and Discussion

Supported Cu catalysts were prepared by a simple impregnation method. An oxide support was impregnated in an aqueous solution of Cu(NO3)2·3H2O at 80 °C for 2 h. After drying, the resulting powder was calcined at 500 °C for 3 h under an air flow. Before being used in catalytic reactions, the supported Cu catalysts was pretreated under a H2 flow (10 mL min–1) at 400 °C for 1 h to reduce the Cu cation to metallic Cu.[31]

Table summarizes the results of the reaction of 1-octanol (1a, 1.5 mmol) in the presence of supported Cu catalysts (1 mol % as metal) at 170 °C for 24 h in mesitylene. Among the supported Cu catalysts tested, ZrO2-supported catalysts showed the highest conversion of the alcohol and the highest yield of the corresponding ester, octyl octanoate (2a). In all of the cases, the formation of a small amount of octanal (3a) was confirmed. Despite the high conversion of the alcohol, Al2O3-supported Cu catalysts resulted in a very low yield of the ester because Lewis acidic surface property of Al2O3 dominantly promoted dehydrative coupling to give octyl ether as a main byproduct. On the other hand, TiO2- and CeO2-supported catalysts gave α,β-unsaturated ketone as a main byproduct via the self-condensation of octanal. The effects of the supported metallic species on the present reaction were also investigated, whereas the reaction with Co, Ni, Ru, Pd, Ag, Ir, or Pt catalysts supported on ZrO2 resulted in a lower conversion of the alcohol and a lower yield of the ester than that with Cu/ZrO2. When only ZrO2 was subjected to the catalytic reaction, no conversion of 1a was confirmed. Although several Cu salts, such as Cu(OAc)2, Cu(NO3)2·3H2O, CuCl, CuCl2·2H2O, and Cu(OH)2, were employed as soluble or insoluble catalyst, no ester was obtained. These results indicate that the combination of metallic Cu and a ZrO2 support is essential for the selective formation of esters from primary alcohols in a dehydrogenative manner. Note that no leaching of Cu species into the solvent after the reaction with Cu/ZrO2 catalysts was confirmed by the atomic emission spectroscopic analysis.

Table 1

Dehydrogenative Synthesis of Ester with Supported Catalystsa

			yield (%)
entry	catalyst	conv. of 1a (%)	2a	3a
1	Cu/ZrO2	99	68	3
2	Cu/CeO2	63	24	7
3	Cu/Al2O3	96	8	2
4	Cu/SiO2	28	1	8
5	Cu/TiO2	100	4	0
6	Co/ZrO2	15	1	0
7	Ni/ZrO2	61	11	1
8	Ru/ZrO2	27	9	1
9	Pd/ZrO2	12	0	0
10	Ag/ZrO2	35	19	2
11	Ir/ZrO2	14	2	1
12	Pt/ZrO2	33	3	1
13	ZrO2	11	1	0

Reaction conditions: 1a (1.5 mmol), supported catalyst (0.01 mmol as Cu), mesitylene (1.0 mL), at 170 °C, 24 h, under Ar.

Yields (2a and 3a) and conversion (1a) were determined by gas–liquid chromatography based on 1a.

Under the optimized reaction conditions, a series of alcohols were subjected to the dehydrogenative synthesis of esters over Cu(5 wt %)/ZrO2 catalysts (Table ). The reactions of aliphatic primary alcohols (1a–f) with a straight alkyl chain gave the corresponding esters (2a–f) in moderate to high yields. Cyclohexylmethanol (1g) and 3-phenyl-1-propanol (1h) could also participate in the present catalytic system to afford 2g and 2h in high yields. The reaction of benzyl alcohol (1i) resulted in 35% yield of benzyl benzoate (2i), whereas an increased yield of 52% was obtained in the reaction with Cu/CeO2 catalyst. This is due to the higher reactivity of benzyl alcohol than that of aliphatic alcohols. The rapid conversion of alcohol to aldehyde reduces the amount of alcohols as the coupling reagent to form hemiacetal, leading to a low yield of the corresponding ester. In this respect, Cu/CeO2 is a suitable catalyst for the dehydrogenative synthesis of esters from benzyl alcohol because the catalyst shows moderate catalytic activity for the dehydrogenation of alcohols. The reactions of diols were also examined, and the intramolecular cyclization of 1,4-butanediol (1j) and 1,5-pentanediol (1k) took place to give the corresponding lactones (2j and 2k) in high yields under diluted conditions.

Table 2

Scope of Substratesa

Gas chromatography (GC) yields.

Reaction for 48 h.

Isolated yield.

Cu/CeO2 was used as a catalyst.

Solvent 2 mL.

Solvent 3 mL.

Figure shows the time course of the reaction of 1a in the presence of Cu/ZrO2 at 170 °C. At the initial stage of the reaction, the rapid formation of octanal 3a was observed, and the yield of the aldehyde gradually decreased after 5 h. In contrast, the yield of ester 2a gradually increased with the passage of time. This strongly suggests that aldehyde 3a was an intermediate for the ester in the present reaction. Two reaction pathways from aldehyde to ester can be assumed (Scheme ); dimerization of aldehyde, the so-called Tishchenko reaction (path A),[32] and the formation of hemiacetal via the condensation of aldehyde with alcohol, followed by dehydrogenation (path B). The reaction of octanal (3a) in the presence of Cu/ZrO2 under the same reaction conditions as in Table resulted in a very low yield of 2a (Scheme ). This strongly suggests that the ester was furnished through path B. Furthermore, the formation of aldehyde was observed (Figure ), which indicates that the dehydrogenation of 1-octanol was faster than the formation of ester. From this result and the general considerations for very low equilibrium concentration of hemiacetals, we can assume that the dehydrogenation of hemiacetal formed by the condensation of aldehyde with alcohol is the rate-determining step.

Figure 1

Time course of the reaction by Cu/ZrO2 catalyst.

Scheme 2

Reaction Pathway from Alcohol to Ester

Scheme 3

Reaction of 3a with Cu/ZrO2

It has become widely accepted that the activity of supported Cu catalysts for alcohol dehydrogenation is enhanced by the combination of metallic nanoparticles and base sites on the support because the basic nature of surface hydroxyl groups facilitates the dissociation of the O–H bond of alcohol to form copper alkoxide species, which are key intermediates for dehydrogenation.[33−37] In sharp contrast, the formation of hemiacetals via the condensation of carbonyl compounds with alcohols can be promoted by Lewis or Brønsted acid catalysis. On the basis of these facts, the coexistence of acid and base sites on the surface of the catalyst must be crucial for the selective formation of ester from primary alcohols. Hence, the surface acid–base amounts for each supported Cu catalyst were estimated by NH3- or CO2-temperature-programmed desorption (NH3-TPD or CO2-TPD).[38]

Figure shows the relationship between the amounts of both acid and base sites for the supported Cu catalysts and the yields of ester 2a. Al2O3-, SiO2-, and TiO2-supported catalysts had acid sites and much fewer base sites than ZrO2- and CeO2-supported catalysts. In contrast, the large amounts of both acid and base sites were present on Cu/ZrO2 catalyst, implying that the acid–base pair sites and adjacent metallic Cu species worked in concert as efficient catalysts for the formation of esters. Figure shows the correlation between the yield of ester 2a and the amount of acid–base pair sites per unit surface area in each supported Cu catalyst. In this case, the amount of acid–base pair sites was defined as the lesser acid and base sites. For example, the amounts in the ZrO2-, Al2O3-, TiO2-, and SiO2-supported catalysts were equivalent to those of base sites, whereas the amount of acid–base pair sites in Cu/CeO2 was equivalent to that of the acid sites. As shown in Figure , the yield of ester is closely correlated with the amount of acid–base pair sites in each supported Cu catalyst, suggesting that the selective formation of esters from primary alcohols must be dominated by the synergetic function of the surface acid–base pair sites with adjacent metallic Cu species. Such amphoteric functions of ZrO2 have been widely applied in the transformation of organic molecules, such as Meerwein–Ponndorf–Verley reduction[39−41] and CO2 fixation into alcohols,[42−44] and it is generally accepted that a Zr center and surface hydroxyl groups serve as the acid and base sites, respectively.[39−41]

Figure 2

Relationship between the yield of 2a (•) in Table and the amount of acid (white squares with black background) and base (chessboard pattern) on the catalysts.

Figure 3

Relationship between the yield of 2a and the amount of acid–base pair sites on the catalysts.

On the basis of these considerations, a possible reaction mechanism for the transformation of primary alcohols to the corresponding esters is proposed, as follows (Scheme ). The base sites on the ZrO2 surface promote the dissociation of the O–H bond to form copper alkoxide. Subsequently, β-hydride elimination gives aldehyde together with copper hydride species, followed by the coupling of hydride with a proton on the surface hydroxyl group to generate molecular hydrogen. The Lewis acidic Zr center activates aldehyde, which promotes the condensation with alcohols to furnish hemiacetal. Finally, the thus-formed hemiacetal is dehydrogenated by the adjacent surface base and metallic Cu to give the final product ester. A detailed mechanistic investigation by spectroscopic techniques is currently underway in our laboratory.

Scheme 4

Proposed Reaction Mechanism

Finally, reusability of Cu/ZrO2 catalyst was investigated (Table ). Unfortunately, severe deactivation of the Cu/ZrO2 catalyst was observed after the catalytic reaction. The reaction of 1a by the recovered Cu/ZrO2 without any regenerative treatment resulted in very low yield of 2a. In contrast, before being subjected to the repeated use, the recovered catalyst was treated with air calcination and H2 reduction. As a result, the regenerated Cu/ZrO2 catalyst gave 2a in an improved yield, whereas the activity of the catalyst was not fully recovered. As described above, a number of acid–base pair sites are crucial for high activity of the catalyst and high selectivity of the reactions. Such pair sites might be diminished during the catalytic reactions in an organic solvent, and they cannot be recovered even though the used catalysts were calcined at high temperatures under the air-flow conditions. Further investigation on the catalyst regeneration is now in progress.

Table 3

Reuse of Cu/ZrO2 Catalysts

			yield (%)a
procedure for regeneration		conv. (%)	2a	3a
	fresh	99	68	3
none	2nd use	14	12	2
calcination and reduction	2nd use	51	30	2

GC yields.

Conclusions

In summary, the dehydrogenative synthesis of esters from primary alcohols under mild reaction conditions was achieved by the use of a ZrO2-supported Cu catalyst. A variety of esters were obtained from primary alcohols as well as diols in good to high yields. Both NH3-TPD and CO2-TPD profiles revealed that the concerted functions of the acid–base pair sites and metallic Cu on the surface of ZrO2 should be responsible for the selective formation of esters.

Experimental Section

Materials

Cu(NO3)2·3H2O, Cu(OAc)2, CuCl, CuCl2·2H2O, and Cu(OH)2 were purchased from Wako Chemicals. Al2O3 (Sumitomo Chemical Co., Ltd, AKP-G015; JRC-ALO-8 equivalent), TiO2 (JRC-TIO-4), ZrO2 (JRC-ZRO-3), CeO2 (JRC-CEO-2), and SiO2 (JRC-SIO-9A) were obtained from the Catalysis Society of Japan. Other organic substrates, alcohols and aldehydes, were of analytical grade and used as received from TCI without further purification.

Physical and Analytical Measurements

The products of the catalytic Cuns were analyzed by gas chromatography–mass spectrometry (GC–MS) (Shimadzu GCMS-QP2010, CBP-10 capillary column, i.d. 0.25 mm, length 30 m, 50–250 °C) and gas chromatography (Shimadzu GC-2014, CBP-10 capillary column, i.d. 0.25 mm, length 30 m, 50–250 °C). NMR spectra were recorded on a JMN-ECS400 (FT, 400 MHz (1H), 100 MHz (13C)) instCument. Chemical shifts (δ) of 1H and 13C{1H} NMR spectra are referenced to SiMe4.

The supported catalysts were analyzed by temperature-programmed desorption (TPD), N2 adsorption, and X-ray diffraction (XRD). NH3-TPD and CO2-TPD were carried out to estimate the amount of acid and base on the catalysts. The TPD was performed on a MicrotracBEL BELCAT-II in the following manners: 100 mg of powder was reduced with H2 (50 mL min–1) gas at 400 °C for 1 h. Then, the temperature was kept for 1 h in He (50 mL min–1) gas (CO2-TPD: the sample was heated at the rate of 10 °C min–1 up to 500 °C, and the temperature was kept for 1 h in He gas). The sample was cooled down to 100 °C in He gas. Gaseous NH3/He (5/45 mL min–1) was adsorbed for 1 h and then removed in He gas for 1 h. Consecutively, NH3-TPD was started at 100 °C, and the temperature was raised to 800 °C at a ramping rate of 10 °C min–1 under the He gas flow. The products desorbed were determined by using a BELMass and recorded on an online personal computer. CO2-TPD was also carried out in a similar manner using CO2 gas. X-ray powder diffraction analyses were performed using Cu Kα radiation and a one-dimensional X-ray detector (SmartLab, Rigaku). The samples were scanned from 2θ = 10–70° at a scanning rate of 10° s–1 and a resolution of 0.01°. The Brunauer–Emmett–Teller specific surface area was estimated from the N2 isotherms obtained using a BELSORP-mini II (BEL Japan, Osaka, Japan) at 77 K. The analyzed samples were evacuated at 573 K for 2 h prior to the measurement. The contents of copper species leached into reaction solvent were determined by the atomic emission spectroscopic analysis with a SHIMADZU AA-6200.

Experimental Procedure

Typical Preparation of a Support Cu Catalyst

Supported catalysts were prepared by the impregnation method. A 1.0 g of support was added to the aqueous solution of Cu(NO3)2·3H2O in air at 353 K. After impregnation, the resulting powder was calcined in air at 773 K for 3 h to afford a supported Cu catalyst. Supported Ni, Co, Cu, Pd, Ag, Ir, and Pt catalysts were prepared in a similar manner.

Representative Procedure for Dehydrogenative Synthesis of Ester from Primary Alcohol

Dehydrogenative coupling of 1-octanol was carried out in a batch-type reactor (20 mL Pyrex tube). Cu/ZrO2 catalyst (100 mg, 1 mol % as Cu) was reductive pretreated in a H2 flow (10 mL min–1) at each temperature for 1 h in the reactor (Cu, Pd, and Pt catalysts at 423 K, Co, Cu, and Ir catalysts at 673 K, Ni catalyst at 773 K). Then, 1-octanol (1.5 mmol) and mesitylene (1.0 mL) were added to the reactor. The reaction was carried out at 443 K under Ar atmosphere. The products were quantified by gas chromatography using an internal standard technique.

Recycling of the Cu/ZrO2 Catalyst

After the reaction, the solid was separated from the reaction mixture by centrifugation and washed with 10 mL of diethyl ether, methanol/H2O (1:1), and again by diethyl ether. The resulting solid was dried overnight at 80 °C and calcined in air at 400 °C for 30 min to recover the Cu/ZrO2 catalyst for reuse. Before catalytic reactions, the supported Cu catalyst was subjected to reductive pretreatment in a H2 flow at 400 °C for 1 h.