One-pot synthesis of lactones from ketoacids involving microwave heating and sodium borohydride: application in biomass conversion

Abstract

A metal-free one-pot cascade method involving NaBH₄ and microwave (MW)-heating efficiently synthesized lactones, including the industrially significant γ-valerolactone at 90% yield. The process directly converts methyl levulinate, succinic acid and a wide range of aliphatic and aromatic ketoacids, in H₂O or THF, proceeding via reduction, COOH proton exchange, cyclization, and dehydration, showing promise for biomass conversion.

---

Lignocellulosic biomass has gained attention as a renewable energy source, offering a sustainable alternative to petroleum-based products due to its abundance.¹ The biorefinery industry and researchers are actively working to transition the economy from petroleum to biomass.^2a,b Despite progress, complete success is still a challenge. Lignocellulosic biomass, consisting of cellulose, hemicellulose, and lignin, is complex and low in density, requiring initial conversion into valuable platform chemicals such as 5-hydroxymethylfurfural (5-HMF), levulinic acid (LA), and furfural.^2c,d These intermediates can then be further processed into biofuel monomers like γ-valerolactone (GVL), ethyl levulinate, and 2-methyltetrahydrofuran (MTHF), among others.^3,4 Lactones, particularly GVL, are notable due to their versatile applications in industries ranging from food to bio-based polymers. GVL is non-toxic, biodegradable, and stable, making it an attractive green solvent, gasoline additive, flavoring agent, and intermediate in chemical synthesis.^5–7

Typically, GVL and other lactones are synthesized from LA or ketoacids through hydrogenation using metal catalysts such as Ru, Pd, or Fe, though these catalysts present challenges like high costs and complex industrial scaling.^7–14 Recent research has explored metal-free catalysts like I₂/Et₃SiH for GVL synthesis, but these also face issues such as toxicity and limited substrate scope.¹⁵ Thus, developing a cost-effective, metal-free catalytic system with high efficiency and broader application remains a key priority for advancing sustainable chemical production.

Sodium borohydride (NaBH₄) is among the most valuable hydrides used for reduction reactions in organic transformations. It serves as a mild and cost-effective reducing agent for converting aldehydes, ketones, and imines into their corresponding alcohols and amines in protic solvents.¹⁶

Karnik et al. demonstrated that NaBH₄ can mediate the reduction of levulinic ester, followed by HCl-catalysed cyclization, to produce GVL in a one-pot process with a yield of 45%.¹⁷ Additionally, Kobayashi et al. reported that certain aromatic keto acids can be converted into 6-membered lactones with yields ranging from 49–56%.¹⁸ However, there has been no systematic investigation into the ability of NaBH₄ to mediate the formation of aliphatic and aromatic-substituted lactones from keto-acid, ketoester and dicarboxylic acid substrates, particularly in the context of GVL synthesis.

Typically, levulinic ester is synthesized via catalytic esterification of LA (a platform chemical derived from biomass via glucose) in the corresponding alcohol, raising the cost and necessitating an additional step for GVL synthesis. Therefore, simplifying the process to achieve the direct conversion of LA to GVL is a desirable goal for cost reduction and improved efficiency. Previous methods rely on HCl for cyclization; however, we hypothesized that the carboxylic acid (COOH) group in LA could function as a mild acid, removing the need for HCl to induce self-cyclization. This cyclization could be further enhanced by MW heating following the hydrogenation of LA using NaBH₄ in polar solvents. In conventional heating, heat transfers from the surface of the reaction vessel inward, leading to uneven temperature distribution within the reaction mixture. In contrast, microwave (MW) heating provides uniform heating throughout the mixture via dielectric heating.^19a This process enables polar solvents and reactant molecules to rotate rapidly, accelerating cyclization and promoting the reaction to be faster and more efficient.^19b As a result, the even heating from MW technology enhances energy efficiency, increases reaction rates, and improves yields, making it especially beneficial for industrial applications.

In this study, we present a one-pot cascade method for synthesizing aliphatic and aromatic substituted lactones from keto-acid or dicarboxylic acid substrates. This approach employs NaBH₄ to provide hydride ions for reduction, followed by cyclization in an H₂O or THF solvent system.

Initially, we chose the conversion of biomass-derived levulinic acid 1a to GVL 2a. The optimization experiments commenced with dissolving LA (0.176 mmol) in H₂O, followed by the addition of NaBH₄ (0.066 mmol, 0.38 equiv.) and heating at 130 °C for 90 min. This process yielded 37% GVL using conventional heating (oil bath) (Table 1, entry 1). When employing MW heating instead, the GVL yield increased to 55% (entry 2), demonstrating higher conversion efficiency. Consequently, MW heating was selected for subsequent investigations. The absence of conversion in a control reaction without NaBH₄ (entry 3) demonstrated that NaBH₄ catalyses the reaction. Similar to H₂O, THF also provided similar conversion and GVL yield (entry 4). The findings indicate that NaBH₄ serves as an effective reducing agent for hydrogenation and cyclization in either ethers or H₂O for GVL synthesis. We further examined the Brønsted acid-catalysed conversion of LA to GVL in an H₂O/FA solvent system with FA as the hydrogen source, both with and without NaBH₄ (entries 5–8). The results show that FA mediates the reaction in H₂O regardless of NaBH₄, but achieves peak performance only with NaBH₄ in the absence of H₂O (entry 8). Interestingly, in CH₃OH, although the conversion rates were high (80–85%), the GVL yields remained low (2–5%) due to competitive esterification, primarily forming methyl levulinate (entries 9 and 10). Additionally, we conducted optimizations using other solvents (1,4-dioxane, toluene, and CH₃CN), to shed light on which type of solvent plays a crucial role in catalytic reactions (Table 1, entries 11–13). The findings indicate that NaBH₄ serves as an effective reducing agent for hydrogenation and cyclization in either ethers or H₂O for GVL synthesis.

Table 1 Optimization of the reaction conditions for LA (1a) conversion^a

Entry	NaBH4	H-Source	Solvent	Heating oil-bath^b/MW^c	Conversion 1a^d (%)	GVL yield 2a^d (%)
a Reaction conditions: LA (1a) (0.176 mmol, 1 equiv.), NaBH4 (0.066 mmol, 0.38 equiv.), solvent (1 mL, for entries 4 & 5 FA/H2O 1 : 1), 90 min. b Thermal heating: 130 °C, reflux, air, round bottom flask (10 mL), 600 rpm. c Microwave (MW) heating: 130 °C, air, 600 rpm, 4 W, 12 bar. d Conversion and yield were determined by GC-MS. e FA: formic acid. f By-product is methyl levulinate.
1	Yes	—	H2O	Thermal	45	37
2	Yes	—	H2O	MW	68	55
3	No	—	H2O	MW	0	0
4	Yes	—	THF	MW	70	53
5	Yes	FA^e	H2O	MW	30	15
6	No	FA	H2O	MW	35	20
7	No	FA	FA	MW	25	13
8	Yes	FA	FA	MW	40	25
9	Yes	—	CH3OH	MW	85^f	2
10	No	—	CH3OH	MW	80^f	5
11	Yes	—	1,4-Dioxane	MW	50	30
12	Yes	—	Toluene	MW	0	0
13	Yes	—	CH3CN	MW	0	0

---

We subsequently optimized other key parameters influencing the conversion, including NaBH₄ loading, temperature, and reaction time (Fig. 1). Varying the NaBH₄ amount for LA hydrogenation showed that 0.079 mmol (0.45 equiv.) of NaBH₄ yielded 75% GVL (Fig. 1a). Next, we investigated the effects of temperature (90–180 °C) and reaction time (30–150 min) individually (Fig. 1b and c). Through optimization, we achieved a 90% GVL yield with 93% selectivity using MW-heating at 150 °C (∼9 W, ∼18 bar), with NaBH₄ as the hydrogen donor in H₂O, within 120 min (Fig. 1d). To highlight the advantages of MW-heating over conventional thermal methods, we conducted experiments under similar conditions using an oil bath. The results showed a significantly lower yield (48%) and selectivity (68.5%), confirming that MW heating is crucial for this conversion (Fig. 1d). Comparisons with previously reported homogeneous catalysts showed that this non-metal-based system achieves high efficiency in GVL synthesis from LA, without requiring an external hydrogen source, and is on par with metal-based catalysts (Table S1, ESI†).

Fig. 1 Optimization of the reaction conditions for LA to GVL conversion, (a) effect of NaBH₄ loading, reaction conditions: LA (0.176 mmol, 1 equiv.), H₂O (1 mL), 90 min, and 130 °C; (b) effect of temperature, reaction conditions: LA (0.176 mmol, 1 equiv.), NaBH₄ (0.079, 0.45 equiv.), H₂O (1 mL), and 90 min; and (c) effect of time, reaction conditions: LA (0.176 mmol, 1 equiv.), NaBH₄ (0.079, 0.45 equiv.), H₂O (1 mL), and 150 °C, and (d) scheme with optimized parameters under MW- and thermal-heating conditions. Conversion and yield were determined by GC-MS (for details, see Experimental section in the ESI†).

The optimized conditions were then applied to test the substrate scope with other aliphatic keto acids (Table 2). When n is increased from 1 to 2, acetyl butyric acid (1b) resulted in a poor yield (10%) of the six-membered lactone (2b) in H₂O. However, replacing H₂O with a 4 : 1 THF/H₂O mixture, increased the yield to 48%. Whereas, with a further increase to n = 3 (1c), the corresponding seven-membered lactone 2c was not observed, presumably due to thermodynamically unfavourable cyclization. Interestingly, when R = OH, succinic acid (1d) was easily converted to a valuable chemical, i.e., succinic anhydride (2d) with a good yield of 80% in H₂O. However, the tert-butyl ester of oxoheptanoic acid (1e) did not undergo conversion to 2d, likely due to the poor leaving group nature of tert-butoxide compared to the hydronium ion in the case of 1d. Further testing with methyl levulinate (1f) showed a yield of >99% for GVL (2a). We compared LA and methyl levulinate conversion rates and GVL yields at 150 °C (Fig. S10, ESI†). The results show efficient GVL formation from both substrates. Since methyl levulinate is derived from LA, our system streamlines the process, achieving 98% conversion and 90% yield of GVL directly from LA, bypassing esterification.

Table 2 Substrate scope for aliphatic ketoacids/diacids/ketoesters to lactone synthesis^a

Entry	Substrate 1(b–f)			Conversion 1(b–f)^c (%)	2(a–d): R1, n yield (%) by GC-MS^c (isolated)^d
	n	R1	R2
a Reaction conditions: 1(b–f) (0.176 mmol, 1 equiv.), NaBH4 (0.079 mmol, 0.45 equiv.), H2O (1 mL). b THF/H2O (4 : 1, 1 mL). c Conversion and yield were determined by GC-MS. d Yield was determined by column or preparative thin-layer chromatography. e The structure represents succinic anhydride.
1^b	2	–Me	–H	60	–Me, 2
	1b				2b: 48 (40)
2	3	–Me	–H	0	–Me, 3
	1c				2c: 0
3	1	–OH	–H	90	=O,^e 1
	1d				2d: 80 (78)
4	1	tBuO–	–H	40	=O,^e 1
	1e				2d: 0
5	1	–Me	–Me	100	–Me, 1
	1f				2a: >99 (96)

---

To investigate the synthesis of other five-membered lactones similar to GVL by replacing the methyl group with a phenyl group (Table 3), we initially applied the optimized method to 4-oxo-4-phenylbutanoic acid (an aromatic keto acid, 1g). However, this resulted in only a 15% yield of lactone 2g in H₂O (Table 3, entry 1). To enhance the yield, we optimized the solvent choice for aromatic keto acids, using 4-oxo-4-phenylbutanoic acid as a model compound while keeping other reaction parameters consistent with those for aliphatic keto acid conversion (entries 2–7). The results revealed that NaBH₄ in cyclic ethereal solvents such as THF and 1,4-dioxane remarkably improved conversion producing 41–98% yields of lactone 2g at 150 °C within 2 h. In contrast, while CH₃OH facilitated 84% conversion, only a 10% yield was achieved due to competition with esterification. Non-ethereal solvents like toluene and CH₃CN did not promote conversion (0%). These findings clearly demonstrate that THF plays a crucial role in lactone formation by stabilizing NaBH₄ through solvation and coordination with boron, enhancing hydride reactivity and improving reaction rates and yields.

Table 3 Optimization of 4-oxo-4-phenylbutanoic acid (1g) conversion^a

Entry	Reducing agent	Solvent	1g conversion^b (%)	2g yield^b (%)
a Reaction conditions: 4-oxo-4-phenylbutanoic acid (1g) (0.176 mmol, 1 equiv.), NaBH4 (0.079, 0.45 equiv.), solvent (1 mL), 2 h, 150 °C, MW. b Conversion was determined by GC-MS. c TBME (tert-butyl methyl ether).
1	NaBH4	H2O	20	15
2	NaBH4	THF	99	98
3	NaBH4	1,4-Dioxane	50	41
4	NaBH4	CH3OH	84	10
5	NaBH4	Toluene	20	0
6	NaBH4	CH3CN	0	0
7	NaBH4	TBME^c	25	8

---

To further expand the substrate scope by modifying the aryl group in both the keto and middle carbon positions in 4-oxo-4-phenylbutanoic acid 1g, we investigated various aryl-substituted keto acids 1(h–k) using THF as the solvent, as H₂O and other solvents had not yielded good conversion for 1g. The results, summarized in Table 4, show that nearly all keto acids achieved excellent conversion to the corresponding lactones, with yields ranging from 83 to 95%. The products were characterized using GC-MS, as well as ¹H, and ¹³C-NMR spectroscopy (see ESI† for details).

Table 4 Substrate scope for aromatic ketoacids to lactone synthesis^a

Entry	Substrate 1(h–k)	Conversion^b (%)	Product 2(h–k)	GC-MS yield^b (%)	Isolated yield^c (%)
a Reaction conditions: 1(h–k) (0.176 mmol, 1 equiv.), NaBH4 (0.079, 0.45 equiv.), THF (1 mL). b Conversion and yield were determined by GC-MS. c Yield was determined by column or preparative thin-layer chromatography.
1		98		89	83
2		100		95	90
3		95		85	81
4		94		83	80

---

After successfully establishing the substrate scope, we focused on investigating the reaction mechanism. GVL synthesis from LA generally follows two reported pathways:⁷ (i) via the 4-hydroxypentanoic acid (HPA) intermediate, which involves hydrogenation followed by dehydration and cyclization, and (ii) via the α-angelica lactone intermediate, which involves dehydration and cyclization followed by hydrogenation. To determine which pathway NaBH₄ follows in converting LA to GVL and to identify potential intermediates, we performed the reaction in deuterated solvent (D₂O) under optimized conditions (LA, 0.176 mmol, 1 equiv.; NaBH₄, 0.079 mmol, 0.45 equiv.; D₂O, 1 mL; 150 °C) for up to 60 min to achieve partial conversion. The reaction mixture was analysed using NMR techniques, including ¹H-, ¹³C-, correlation spectroscopy (COSY), and diffusion ordered spectroscopy (DOSY), and high-resolution mass spectrometry (HR-MS) to detect intermediates (ESI,† Section B4). Based on the NMR data, we identified the peaks corresponding to GVL and HPA. Furthermore, the HR-MS spectrum displayed a [M + H] peak corresponding to HPA (m/z = 118.06 daltons) and its isotope (m/z = 120.06 daltons), attributed to H/D exchange in D₂O (Fig. S9, ESI†). The absence of the molecular ion peak [M] for α-angelica lactone (m/z = 98.06 daltons) in both NMR and HR-MS data indicates that the reaction proceeds through pathway I, involving the HPA intermediate. Therefore, we proposed a plausible mechanism (Fig. 2) based on these findings.

Fig. 2 Plausible reaction mechanism for the LA to GVL conversion.

Green metrics serve as key indicators for evaluating how well a chemical process aligns with green chemistry principles, assessing both its environmental impact and industrial efficiency. A detailed analysis was conducted to assess these metrics in the hydrogenation of ketoacids for lactone synthesis, with a focus on the process's environmental footprint and sustainability. Critical parameters such as the environmental factor (E-factor), process mass intensity (PMI), and reaction mass efficiency (RME) were evaluated, as they are crucial to promoting a greener, more sustainable chemical process. For the reaction from LA to GVL, the green metrics were calculated and compared with ideal values (Table S2, ESI†). The E-factor, which measures waste generation, was found to be 0.3, close to the ideal of 0.16. In an ideal scenario, where all reactants are fully converted to the product without a catalyst and reactants are used in H₂O, the PMI would be 1. In this case, the calculated PMI is 1.48, primarily due to the use of NaBH₄ in catalytic amounts. NaBH₄ is a low-cost commercial chemical and more sustainable than other noble metal-based catalysts and hydrogen sources like H₂. The favourable E-factor and PMI values underscore the environmental and sustainable merits of this synthetic route.

In conclusion, we have developed a sustainable, metal-free one-pot cascade synthetic approach using NaBH₄ under MW heating to efficiently convert a wide range of biomass-derived precursors, including keto acids, keto esters, and dicarboxylic acids, into their corresponding lactones. This method demonstrated excellent performance, achieving yields ranging from 48–99%. Notably, it was highly successful in synthesizing industrially significant GVL with a 90% yield and 93% selectivity. Our study showed that MW heating accelerates self-cyclization via the COOH group in keto acids, eliminating the need for HCl^17,18 and improving both the yield and selectivity in polar solvents. Mechanistic studies confirmed that the reaction proceeds through the HPA intermediate. This environmentally friendly and scalable approach offers promising potential for enhancing industrial sustainability in lactone production and expanding the use of renewable biomass in chemical synthesis.

This research is supported by grant numbers BT/CIAB-Flagship/2018 from the Department of Biotechnology, and CRG/2022/007139 from DST-SERB. GJ thanks DST-SERB for the Ramanujan Fellowship (SB/S2/RJN-047/2015).

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

The authors declare no competing financial interest.

References

1. R. Mori, RSC Sustainability, 2023, 1, 179–212.
2. (a) L. Yang, X. C. Wang, M. Dai, B. Chen, Y. Qiao, H. Deng and Y. Wang, Energy, 2021, 228, 120533; (b) R. Singathi, R. Raghunathan, R. Krishnan, S. Kumar Rajendran, S. Baburaj, M. P. Sibi, M. P. D. C. Webster and J. Sivaguru, Angew. Chem., Int. Ed., 2022, 61, e202203353; (c) Q. Yan, X. Wu, H. Jiang, H. Wang, F. Xu, H. Li, H. Zhang and S. Yang, Coord. Chem. Rev., 2024, 502, 215622; (d) H. Tian, A. Wang, H. Pan, H. Zhang and S. Yang, Ind. Crops Prod., 2024, 222, 119738.
3. (a) G. Gómez Millán, S. Hellsten, J. Llorca, R. Luque, H. Sixta and A. M. Balu, ChemCatChem, 2019, 11, 2022–2042; (b) K. Yan, T. Lafleur, X. Wu, J. Chai, G. Wu and X. Xie, Chem. Commun., 2015, 51, 6984–6987; (c) A. Selim, R. Sharma, S. M. Arumugam, S. Elumalai and G. Jayamurugan, ChemistrySelect, 2022, 7, e202104448; (d) R. Sharma, A. Selim, B. Devi, S. M. Arumugam, S. Sartaliya, S. Elumalai and G. Jayamurugan, Biomass Convers. Biorefin., 2024, 14, 11445–11457.
4. F. A. Kucherov, L. V. Romashov, K. I. Galkin and V. P. Ananikov, ACS Sustainable Chem. Eng., 2018, 6, 8064–8092.
5. T. Raj, K. Chandrasekhar, R. Banu, J. J. Yoon, G. Kumar and S. H. Kim, Fuel, 2021, 303, 121333.
6. (a) O. P. Taran, V. V. Sychev and B. N. Kuznetsov, Catal. Ind., 2021, 13, 289–308; (b) F. Kerkel, M. Markiewicz, S. Stolte, E. Müller and W. Kunz, Green Chem., 2021, 23, 2962–2976.
7. Z. Yu, X. Lu, C. Liu, Y. Han and N. Ji, Renewable Sustainable Energy Rev., 2019, 112, 140–157.
8. S. Dutta, K. M. Iris, D. C. Tsang, Y. H. Ng, Y. S. Ok, J. Sherwood and J. H. Clark, Chem. Eng. J., 2019, 372, 992–1006.
9. X. Gong, X. Feng, J. Cao, Y. Wang, X. Zheng, W. Yu, X. Wang and S. Shi, Chem. Commun., 2023, 59, 14717–14720.
10. L. T. Mika and I. T. Horváth, Adv. Inorg. Chem., 2021, 77, 1–25.
11. G. Metzker and A. C. Burtoloso, Chem. Commun., 2015, 51, 14199–14202.
12. X. Li, X. L. Shi, J. Wang, K. Shi and Q. Wang, J. Ind. Eng. Chem., 2024, 138, 17–33.
13. F. Liguori, C. Moreno-Marrodan and P. Barbaro, ACS Catal., 2015, 5, 1882–1894.
14. W. Zhao, S. Meier, S. Yang and A. Riisager, Green Chem., 2020, 22, 5972–5977.
15. P. Dolui, A. Nair, P. Saini, A. Verma and A. J. Elias, Asian J. Org. Chem., 2022, 11, e202200650.
16. M. Periasamy and M. Thirumalaikumar, J. Organomet. Chem., 2000, 609, 137–151.
17. A. V. Karnik, S. T. Patil, S. S. Patnekar and A. Semwal, New J. Chem., 2004, 28, 1420–1422.
18. K. Kobayashi, T. Mannami, M. Kawakita, J. Tokimatsu and H. Konishi, Bull. Chem. Soc. Jpn., 1994, 67, 582–585.
19. (a) C. O. Kappe, Angew. Chem., Int. Ed., 2004, 43, 6250–6284; (b) E. Cini, C. B. Botta, M. Rodriquez and M. Taddei, Tetrahedron Lett., 2009, 50, 7159–7161.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental information, precursor synthesis, and spectral details (¹H, ¹³C, COSY, DOSY) of the lactone products. See DOI: https://doi.org/10.1039/d4cc05194g

---