ReviewReview: Sustainable production of hydroxymethylfurfural and levulinic acid: Challenges and opportunities
Introduction
The rapid depletion of fossil fuels has led to an increased international effort to augment the use of renewable energy. However, alongside being the predominant source of energy, fossil fuels are also the leading source of organic chemicals which are the backbone of modern life. The replacement for fossil fuels in this area can come from biomass, with biorefineries being presented as the future substitutes for the present day petroleum refineries. In the same way that petroleum refineries use certain chemicals as the building blocks for more complex molecules such as polymers, a biorefinery will use simple molecules that can be readily obtained from a variety of feedstock as a base for the synthesis of biopolymers and other large molecules [1], [2]. Among the most promising building blocks are hydroxymethylfurfural (HMF) and levulinic acid (LA), which are the subject of this literature review.
HMF has been referred to as a ‘sleeping giant’ owing to the vast potential of this compound in the emerging bio-based economy due to the key position it holds in the production of biomass-derived intermediates [3]. It is said to be one of the few petroleum-derived chemicals that can also be readily synthesised from renewable resources, and is held to be a bridge between carbohydrate chemistry and industrial mineral oil-based organic chemistry [4]. While a vast number of chemicals can be produced from HMF, some of the most important ones are listed in Table 1 [5], [6], [7], [8], [9], [10], along with their potential uses. Among the chemicals not listed in Table 1 is LA, the attributes of which are also of importance for discussion in this paper.
In addition to being an intermediate in the production of chemicals mentioned in Table 1, HMF also has some potential uses on its own, such as in fuel cells [11], the treatment of sickle cell disease [12], etc. Ultimately, however, the major potential of HMF is as a key platform chemical, and this depends on its availability and cost [13]. In 1993, it was estimated that the cost of producing HMF would be between Deutsche Mark (DM) 5000–10,000 per ton based on a fructose price of DM 2000 per ton or inulin price of DM 1000 per ton [14], which at 2002 prices would be comparable to 2500–5000 €/ton [15]. These prices are fairly similar to those obtained for the pilot plant operated by Südzucker AG in the Federal Republic of Germany, with the manufacturing price of HMF being 6000 DM/ton for a fructose price of 500 DM/ton [13], and also the price of 2000 €/ton predicted by Bicker et al. for a fructose price of 500 €/ton [16].
Some researchers have investigated whether HMF could be produced at reduced cost in order to be used as a substitute for certain target chemicals. Torres et al. estimated the cost of production of HMF using a semi-batch biphasic reactor and compared it with the price of p-xylene, for which HMF can act as a substitute in the production of polyethylene terephthalate (PET) [17]. They obtained a minimum HMF cost of 0.248–0.273 $/mol (1967–2165 $/ton) for a fructose cost of 550 $/ton, depending on the solvent used at the extraction stage. However, this was higher than the cost of p-xylene (0.109 $/mol or 1027 $/ton). Since the cost of fructose is the dominant factor in the HMF price, it was concluded that lower fructose costs are necessary alongside the development of more efficient processes for the HMF price to be competitive [17]. More recently, Kazi et al. estimated that a plant co-producing dimethylfuran (2,5-DMF) and HMF from 300 metric ton/day of fructose would yield HMF with a minimum selling price of $ 1.33/l, which translates into approximately $ 1100/ton. However, this is still too expensive for HMF to be used as a feedstock for the production of FDCA as a replacement for terephthalic acid, which sells for around $ 800/ton [18]. Recently, Liu et al. stated that HMF produced at $1210/ton would be cost competitive with the petroleum-derived paraxylene-terephthalic acid selling at $1440/ton, and state that this HMF price is achievable for a fructose price of 460 $/ton [19].
LA is a compound derived from HMF that is itself a promising chemical intermediate. It was listed among the top 12 most promising value added chemicals from biomass by the Biomass Program of the US Department of Energy in 2004 [1], and continues to rank highly in more recent reviews of most important biorefinery target products [20]. This recognition of the potential of LA is not a recent phenomenon. The compound was first identified in the 1870s, and in 1956 a detailed report outlining the many derivatives from LA and their potential applications were published [21]. Despite this, the commercial applications of LA have been slow to develop. Among the reasons cited for this are the expensive raw materials, low yields, high equipment cost, problematic recovery and handling, etc [21]. Nevertheless, it is currently used in several industries including personal care products, lubricants, adsorbents, electronics, photography, batteries and drug delivery systems [22]. The production of LA at lower costs, however, should greatly enhance its use as a chemical intermediate, given the range of chemicals having numerous market applications that can be produced from it. Table 2 summarises some of the important chemicals that can be synthesised from LA and the sectors they can be used in Refs. [8], [23], [24], [25], [26].
The price for LA was about 8.8–13.2 $/kg in 2000, which meant that it was mostly used as a speciality chemical [22]. The small market size of LA at this time, around 450,000 kg annually [27], meant that it was produced largely from maleic anhydride and other petrochemicals [28]. The price in 2010 was around 3.2 $/kg, which was substantially lower than the earlier figures, but still too high for chemicals produced from LA to compete with those derived directly from petroleum [29]. Despite LA prices remaining in the range of 5–8 $/kg in 2013, global LA consumption rose to around 2600 tons, and is expected to increase steadily in the near-future, reaching 3800 tons in 2020 [30]. The Biofine process, which is dealt with in Section 9, has been projected to lower LA costs to as low as 0.09–0.22 $/kg [22], which should make a whole range of LA-derived chemicals economically attractive.
The above discussion has shown that while both HMF and LA have tremendous market potential, the realisation of this potential is dependent on the production of large quantities of these chemicals sustainably and at low prices. Hence, it is not surprising that these problems have attracted many researchers' interest. The large amount of research conducted on the synthesis of these two chemicals and the numerous reaction systems devised makes it necessary to analyse the merits and demerits of the different approaches, particularly with respect to the feedstocks used, the yields obtained, and the accompanying environmental impacts. The requirements for taking the laboratory schemes towards industrial production also need to be identified and addressed. This review, therefore, provides an overview of the production of HMF and LA from different carbohydrates and lignocellulosic biomass using various solvent and catalyst systems, and identifies the gaps that need to be filled for the two building blocks to attain the level of commercialisation they deserve. The use of ‘green chemistry’ techniques to synthesise HMF and LA is of particular interest to ensure that these chemicals are truly ‘sustainable’. Microwave heating provides an energy-efficient method of increasing reaction rates and yields while using benign solvents and moderate temperatures. Therefore, a special focus of this review will be on systems that utilise microwave irradiation as the heating medium.
Section snippets
Chemistry of HMF synthesis
Hydroxymethylfurfural (HMF), also known as 5-(hydroxymethyl)-2-furancarboxaldehyde and 5-(hydroxymethyl)-2-furaldehyde [31], is a chemical that has been of interest to researchers for over a century [32]. The HMF molecule has various functionalities arising from the presence of hydroxyl and aldehyde groups as well as a furan ring [14]. HMF is a yellow solid with a melting point between 28 and 34 °C, a boiling point between 114 and 116 °C and a density of 1.206 g/cc [13], [31], [33], [34]. It is
Chemistry of LA synthesis
Levulinic acid (LA), also known as 4-oxopentanoic acid [31], is a colourless crystalline compound (commercial product being yellow) with a melting point ranging from 33 to 37 °C, a boiling point between 245 and 246 °C and a density of around 1.14 g/cc [31], [34], [42]. LA is highly soluble in hot water, as well as in ethanol, diethyl ether, chloroform, acetone, dilute acids, etc [31], [34], [42]. In water, LA has a pKa (25 °C) value of 4.59, which means that its acidity is comparable with most
Monosaccharides
Since HMF is basically a molecule formed by the dehydration of a hexose molecule, and HMF itself is used for LA production, the hexose sugars fructose and glucose have been the starting point for HMF and LA synthesis for many researchers. Of these, fructose is easier to convert into HMF, but the wider availability of glucose means that it might be a better candidate as HMF feedstock [46]. One method of utilising glucose as feedstock is to use a catalyst that isomerises glucose to fructose.
Reaction systems for HMF and LA synthesis
A large number of reaction systems have been investigated for the synthesis of both HMF and LA from a range of different feedstocks. One of the main differences between the various reaction systems is the catalyst, with some non-catalytic systems being used, while others employ either liquid or solid catalysts. These have been discussed in detail in Section 5.1. The other major difference is in the solvent used, with the use of various protic and aprotic solvents being discussed in Section 5.2.
Laboratory-scale production of HMF
A large number of researchers have studied a range of reaction conditions for the laboratory-scale production of HMF from a number of different feedstocks. Some of the major work done in this area has been summarised below. To enable a better understanding of the large number of factors at play, the experiments have been categorised based on the feedstock used, with runs based on fructose in Table 3 and on monosaccharides other than fructose in Table 4. The reason for studying fructose
Pilot and commercial-scale production of HMF
Despite the potential importance of HMF and the number of laboratory schemes devised, till date no full-scale commercial plants for HMF manufacture have been set up, and only a few pilot processes, which are generally of the kilogram scale, have been reported. One of the earliest was mentioned in a patent by Cope in 1959, which involved a batch non-catalytic conversion of sucrose or glucose to HMF using MIBK as solvent. A distilled HMF yield of up to 63% was reported from a 50 wt% sucrose
Laboratory-scale production of LA
As LA is most commonly produced by a reaction route where HMF is an intermediate, it is understandable that the reaction systems used for LA synthesis are similar to those employed for HMF production. In general, it can be stated that higher temperatures, shorter reaction times, lower water concentration and higher pH maximise HMF production, while the opposite is true for LA [71], [97], [230], [231].
From Table 8, it can be seen that the LA yields obtained are lower than the HMF yields that had
Pilot and commercial-scale production of LA
One major process for the conversion of lignocellulosic biomass to LA is the Biofine process. This process, which can process a range of heterogeneous lignocellulosic feedstocks, is a two-stage high temperature acid hydrolysis process using sulphuric acid that differs from most acid hydrolysis processes in that the final products are not monomeric sugars but LA and furfural [8], [269]. The first stage of the process is a plug flow reactor that hydrolyses the cellulose and other carbohydrate
Microwave heating systems
Traditionally, heating apparatus like oil bath and muffle furnace have been used to carry out chemical synthesis. The pioneering work carried out by Gedye et al. [276], and Giguere et al. [277] in 1986, however, demonstrated the feasibility of using microwave ovens for chemical synthesis. Subsequently, microwave heating has steadily gained in importance as a heating method for synthetic transformations. The most important advantage of microwave heating is the dramatic increase in reaction rates
Microwave heating systems for HMF and LA synthesis
All the advantages of microwave heating outlined in Section 10 have led researchers to investigate the possibility of using microwave heating for producing HMF, LA and other chemicals from cellulose and lignocellulosic biomass. Table 11 summarises some of the work on HMF and LA synthesis using microwave irradiation that has been carried out in recent years. Some of the items in the table have been explained in greater detail in the following paragraphs.
Hansen et al., used fructose as the raw
Conclusions
HMF and LA are promising biorefinery platform chemicals, but for these to be commercialised successfully, they are to be synthesised economically to produce large yields from a range of different biomass feedstocks with a minimal environmental footprint. While monosaccharides like fructose may offer the highest product yields, the use of polysaccharides, and in particular lignocellulosic materials, is required for commercial-scale production to be viable from both the sustainability and
Acknowledgements
The financial support of McGill University through the Start-Up Fund, the National Science and Engineering Research Council of Canada (NSERC) and the Fonds de recherche du Québec – Nature et technologies (FRQNT) is gratefully acknowledged.
References (310)
- et al.
Process integration for the conversion of glucose to 2,5-furandicarboxylic acid
Chem Eng Res Des
(2009) - et al.
Dehydration of d-fructose to hydroxymethylfurfural in sub- and supercritical fluids
J Supercrit Fluids
(2005) - et al.
Carbohydrates as green raw materials for the chemical industry
Comp Rend Chim
(2004) Hydroxymethylfurfural, a possible basic chemical for industrial intermediates
- et al.
Techno-economic analysis of dimethylfuran (DMF) and hydroxymethylfurfural (HMF) production from pure fructose in catalytic processes
Chem Eng J
(2011) - et al.
Production of levulinic acid and use as a platform chemical for derived products
Resour Conserv Recycl
(2000) - et al.
Techno-economic analysis of 5-nonanone production from levulinic acid
Chem Eng J
(2010) - et al.
Mechanism of formation of 5-(hydroxymethyl)-2-furaldehyde from d-fructose and sucrose
Carbohydr Res
(1990) - et al.
Microwave-assisted hydrothermal degradation of fructose and glucose in subcritical water
Biomass Bioenergy
(2012) - et al.
Mechanism of levulinic acid formation
Tetrahedron Lett
(1985)