Agricultural byproducts-based biosorbents for purification of bioalcohols: a review

Globally, energy scarcity is a major problem and it is increasing with increasing population and number of vehicles. The transportation sector relies on fossil fuels, so vehicular emissions and associated environment change due to excessive usage of petroleum are a major concern. This has led to the in-depth research for development of environment friendly and possibly cost-effective alternatives. Liquid biofuels such as bioalcohols are alternative energy sources to fossil fuel. However, ethanol, butanol, methanol, isopropanol and other alcohols, produced by microbial fermentation, need intensive purification to obtain fuel grade purity (Niu et al. 2014). The traditional processing techniques have limitations, such as operating cost, energy requirement, process efficiency and scale up (Prakash et al. 2016; Balat 2011).

There are several processes for the purification of alcohols and production of anhydrous alcohols, such as azeotropic and extractive distillation (Zhao et al. 2017), absorption (Fei and Hongzhang 2009), gas stripping (Schläfle et al. 2017); pervaporation (Bello et al. 2014; Xu et al. 2018); solvent extraction (Lemos et al. 2017), and adsorption (Farzaneh et al. 2017; Pal et al. 2017; Raganati et al. 2018). Combination of one or more of these methods can also be employed to enhance the efficiency of the process. Generally, to get anhydrous (> 99%) ethanol, at first, distillation is carried out to get a purity of 72–92% (the azeotrope concentration is 95.6 wt%), followed by adsorption to remove rest of the water. The energy consumption of this combined process is about 3.9 MJ/kg compared with that of the single distillation processes (in the range of 6–9 MJ/kg) (Sun et al. 2007).

Adsorbents are natural or synthetic materials of amorphous or microcrystalline structure. Commonly used are molecular sieves, activated carbon, activated alumina, and silica gel on industrial scale (Perry 1997). Adsorbents can be classified based on their structure—amorphous or structured. Structured adsorbents such as molecular sieves zeolites (3A, 4A, 5A, 13X), as their name suggests, possess well-defined ordered structure and homogenous pore size, while amorphous adsorbents including activate carbon, silica gel, activated alumina, and different biosorbents have heterogeneous surface (Perry 1997). Biosorbents have the highest structural heterogeneity and a wide range of pore sizes to cater to their biological function in nature (Isaac et al. 2015). Biosorbents are also commonly used for dye, heavy metals, or antibiotics removal (Malik 2003; Crini 2006; Yan and Niu 2017a) due to their abundant availability, negligible cost, and their excellent capability to adsorb above mentioned components.

The present mini-review describes the purification of bioalcohols (mainly bioethanol and biobutanol) after initial azeotropic distillation followed by adsorptive dehydration to fuel grade purity (> 99 wt%.). There are number of review articles for the use of biosorbents for adsorption of dyes or heavy metals (Gupta and Suhas 2009; Bharathi and Ramesh 2013); however, as per the authors’ knowledge, there is no review article on the use of agricultural byproducts-based materials as adsorbents for bioalcohol dehydration. The focus of this paper is thus limited to the description of agri-based materials as adsorbents.

The adsorption process is one of the most commonly employed process in industries for ethanol dehydration as it is efficient and cost-effective. In a typical adsorption process, the component(s) of choice gets separated from a fluid mixture using an adsorbent material. As indicated before, adsorption is most commonly applied after the initial distillation of alcohols for selective water (or alcohol) adsorption from the water–alcohol mixtures. The adsorption of water on adsorbent is generally of physical nature which is an equilibrium-dependent process expressed by constructing isotherms. By comparing isotherm data at different temperatures, valuable information about the adsorption capacities and system design can be obtained.

The adsorbent must have a high surface to volume ratio; as generally, only a few layers of molecules are adsorbed on the surface. Biosorbents contain polar groups which strongly adsorb polar water. The more the polar groups on adsorbent, the more will be the water adsorption (Kidnay et al. 2011). The ideal adsorbent has high adsorption rate and capacity, high selectivity for the desired component, and low cost. All these characteristics allow for effective separation of the desired component from all other components present in the adsorption feed stream. The ease of desorption is another important factor to be considered to develop an efficient process (Abdehagh et al. 2013, 2016, 2017).

In addition to the adsorption capacity of a given adsorbent, its kinetic performance is also of great significance for the industrial application. In a fixed-bed system, breakthrough curves can be constructed (Qiu et al. 2009). The plots of relative adsorbate concentration (C/C₀) vs. time are termed as breakthrough curve which is studied for each adsorbing species (water, ethanol, or other components present in system). The breakthrough time of an adsorbent is an important criterion in an industrial operation. Alcohol feed concentration affects breakthrough time. The primary reason for shorter breakthrough times with increasing water concentration is that the adsorbent is exposed to more adsorbate per unit time (Pruksathorn and Vitidsant 2009). Furthermore, the adsorbate adsorption rate is also a critical factor.

For water adsorption on to the adsorbent surface, two different routes are commonly employed—liquid phase and vapor phase. Liquid phase adsorption is generally employed to screen an adsorbent for water adsorption capacity but has a disadvantage of leaching of biomass components into the solution which may be undesirable for product purity, while the vapor phase adsorption is favored as continuous operation is possible with high product purity.

Pressure swing adsorption (PSA) is one of the most commonly used processes for dehydration of gas/vapor phase as it is energy and cost-effective (Jeong et al. 2012; Simo 2013). The other reason that PSA is more favored than a similar process such as temperature swing adsorption, as it is easier and quick to switch from high to low pressure and vice versa as compared to temperature, leading to shorter cycles and prolonged life of adsorbent. Originally, the PSA process was invented by Skarstrom (1960), adsorption and desorption/purge steps are carried out in two separate adsorbent beds operated sequentially, enabling a continuous processing. PSA is a cyclical process consisting typically of two or more vessels packed with the adsorbent. The high partial pressure of component to be removed translates to high adsorption capacity (Simo 2013).

A typical PSA cycle for ethanol dehydration is presented in Fig. 1a (Simo 2013). It consists of a pressurization step (production step) in which vapor flows into the column at a high pressure; water is adsorbed while alcohol vapor passes through and collected as product at the bottom of bed. It is followed by regeneration (desorption), as the bed must be regenerated and prepared for the next cycle. It is referred to as blowdown step and occurs at low pressure which is followed by purge to desorb water from the bed under a lower pressure or vacuum. Near the end of the purge step, a portion of product vapor (> 99% alcohol) is used to purge the vessel to remove the adsorbed water that had been adsorbed during the production step. Then, the column is re-pressurized with product ethanol vapor from the operating vessel. This completes one cycle and the system is ready to enter a new production step (Jeong et al. 2012; Simo 2013). Pressure profile during a two bed PSA is presented in Fig. 1b. The most common cycle time for an industrial two bed PSA is 6–8 min as seen in Fig. 1b.

During the adsorption, there are three zones in an adsorbent bed namely: (1) the equilibrium zone, where the adsorbate on the adsorbent is in equilibrium with the adsorbate in the inlet fluid phase and no additional adsorption occurs; (2) the mass transfer zone (MTZ), the volume where mass transfer and adsorption take place; (3) the active zone, where no adsorption has yet taken place (Ladisch 1997). It is desirous to reach equilibrium (saturation) in minimum possible time, so a shorter MTZ is desirable (without the mass transfer limitation, the thickness of MTZ would be zero).

To maximize bed capacity, the MTZ needs to be as small as possible because the zone holds only about 50% of the adsorbate held by a comparable length of adsorbent at equilibrium. To minimize the thickness of MTZ and to make effective use of adsorbent bed, tall and slender beds and smaller particles size of adsorbent bed are recommended. However, smaller particle size, deeper beds, and increased gas velocity will increase pressure drop (Ladisch 1997). Type 3A molecular sieve (MS) is more efficient for drying the ethanol–water azeotrope than type 4A MS due to that the former can be more efficient than the later in terms of excluding ethanol molecules while adsorbing water molecules. In addition, the regeneration energy requirements for type 3A MS are lower than those of type 4A MS (Al-Asheh et al. 2004). However, for both of the two types of adsorbents, the temperature required for regeneration of the water saturated bed is high (around 190 °C or higher), which is energy intensive. There is an incentive to develop cost-effective adsorbents and processes.

In literature, several biomaterials have been found to be effective adsorbents, such as—corn grits, potato starch, amylose, and maize starch (starch based) and wheat straw and woodchips (cellulose based). In the latter, xylans and hemicelluloses act as the major adsorbing components instead of amylopectin in case of former (Benson and George 2005). Based on the production of major crops in the world, rice straw, wheat straw, corn stover/straw, and sugarcane bagasse are the most generated agricultural residues. In USA alone, the total biomass potential would be about 680 million tons till 2030 which can be utilized for biofuel production and purification as biosorbents (Union of Concerned Scientists 2012). The typical composition and global availability of major agricultural residues have been presented in Table 4. In addition, Fig. 2 presents the typical structure of a lignocellulosic biomaterial comprising cellulose, hemicellulose, and lignin. Plant biomass is typically composed of three polymers: cellulose, hemicellulose, and lignin along with smaller amounts of protein, extractives, pectin, and ash. The composition of biomass can vary from one plant species to another (Bajpai 2016).

Agricultural residues	Non-wood fibers annual production (million tons)	Chemical composition of non-wood lignocellulosic biomass
		Cellulose (%)	Hemicellulose (%)	Lignin (%)
Barley straw	51.3	33.3–42	20.4–28.0	17.1
Corn stover/straw	376.8	35.0–42.6	17.0–35.0	7–21
Oat straw	10.4	37.6	23.34	12.85
Rice straw	657.5	32.0–47.0	18.0–28.0	5.5–24.0
Sorghum straw	12.0	32.4	27.0	3.9
Wheat straw	472.2	33.0–45.0	20.0–32.0	155.8–212.5
Sugarcane bagasse	1044.8	45.4	28.7	474.3
Oil palm biomass	63.9	35.8–56.0	21.9–44.0	22.9–35.8
Canola meal	46.7	26.4 (cellulose + lignin)	6.3

In general, biomaterials contain about 50–80% in the form of cellulose and hemicellulose which are important with respect to water adsorption on biomass. Lignin does not contribute much for water adsorption (Benson and George 2005). Common sources of biomass are: agricultural—food grain, bagasse (crushed sugarcane), corn stalks, straw, seed hulls, nutshells, poultry and hogs; forest—trees, wood waste, wood or bark, sawdust, timber slash and mill scrap; municipal—municipal solid waste, sewage sludge, refuse-derived fuel, food waste, waste paper and yard clippings; energy crops—poplars, willows, switch grass, corn, soybean, canola and other plant oils; and biological—animal wastes, aquatic species and biological wastes (Fantini 2017).

It was the pioneering efforts of Ladisch and Dyck in Purdue University that opened up a new era—the use of biological materials as adsorbents for drying of ethanol using cracked corn, starch, and cellulose (Ladisch and Dyck 1979). Since then, extensive research has been carried out using a wide range of biomaterials such as corn grits, corn meal, canola meal, corn cobs, and barley straw for removal of water from alcohol and gases (Anderson et al. 1996; Ladisch 1997; Chang et al. 2006; Ranjbar et al. 2013; Tajallipour et al. 2013; Jayaprakash et al. 2017; Dhabhai et al. 2018; Hong et al. 1982). The only biomaterial that was applied at an industrial scale is corn grits, which has been used since 1984 for drying methanol, ethanol, propanol, and tert-butanol vapors (Ladisch 1997; Beery et al. 1998). PSA has been used for this drying application (Beery and Ladisch 2001). Corn grits are used in industry to dry 2.8 billion liters of fuel-grade ethanol annually produced by fermentation, which is over half of the fuel-grade ethanol produced in the US (Beery et al. 1998). However, use of corn exerts pressure on food availability. Water adsorption with biomass-based adsorbents may consume less energy than adsorption using other adsorbents. Furthermore, biomass adsorbents require less energy for regeneration as well compared to synthetic adsorbents. Also, when regeneration is not possible, the spent biosorbent can be utilized for biofuel or bioenergy production (Rattanaphanee et al. 2013).

The generation, use, and disposal/reuse of biosorbents have been depicted in Fig. 3. The use of biomaterials as adsorbents is advantageous as compared to most commonly used chemical-based adsorbents—commercial molecular sieves. Biomaterials are derived from renewable biomass and are nontoxic and biodegradable (Baylak et al. 2012). In addition, the production of biosorbents requires minimal processing and no chemical and thermal reactions like chemical adsorbents which often require high temperature activation. Moreover, the temperature of regeneration for biosorbents is much lower (< 110 °C) compared to commercial molecular sieves (> 190 °C) (Ranjbar et al. 2013; Tajallipour et al. 2013; Jayaprakash et al. 2017). Even if regeneration is not possible with biosorbent, it can still be utilized as a raw material for the production of biofuels (Benson and George 2005). It is to be noted that activate carbon produced from renewable biomass has not been included in the present definition of biosorbents as activate carbon production requires elevated temperature/pressure conditions.

Key features of biosorbents for purification of bioalcohols

Many of the categories of biomass described above can be used as adsorbents for dehydration of alcohols. The most common is the agricultural and forest residues because of their large-scale production and availability, renewability, cost-effectiveness, and non-competitiveness with food. One of the most predominant features with respect to water adsorption is the available surface area and pore volume. The higher the surface area, the higher may be the adsorption capacity. However, for raw biomaterials, surface area is generally low (in the range of 2–50 m²/g). The enhancement of surface characteristic such as increasing polar groups and surface area can be done by the treatment of biomass with mineral acids (Dhabhai et al. 2018; Yan and Niu 2017a), alkali (Imadi and Kazi 2015), or surface-active solvents or ionic liquids (Lee et al. 2009; Hou et al. 2012) which selectively remove components like volatile compounds, or lignin from the biomass. However, removal of biomass components may also be counterproductive as structural integrity of material is compromised. Jayaprakash et al. (2017) demonstrated that canola meal biosorbent has higher water adsorption capacity than the single substance of its main components such as cellulose and protein (Jayaprakash 2016).

Another key feature is the affinity with water molecules. One of the proposed mechanisms for adsorption is polar–polar interactions between adsorbent surface and adsorbate. Water adsorption by biomaterials is reported to involve the polar attraction between water and the cellulosic hydroxyl components and the protein carboxyl and amine groups in the adsorbent (Benson and George 2005; Jayaprakash et al. 2017; Dhabhai et al. 2018). Figure 4a presents the structural features of common biomaterials depicted using FTIR (Fourier transform infrared) spectra. Biomaterials contain cellulose and hemicellulose, which have polar groups such as carboxyl, hydroxyl, and/or amino groups, and have the potential for selective water adsorption as the polarity of water is higher as compared to alcohols. Xylan from hemicellulose itself can separate water from ethanol, and cellulose also has sorptive properties (Ladisch 1997). These polar groups can bond with water through hydrogen bonding and van der wall interactions which are weaker and reversible in nature as compared to ionic interaction. On the biomass, available polar groups are the key. The higher the number of polar groups on the surface of biomass, the better the water retention by biomaterial. In addition, most biomaterials are stable up to 200 °C or more as depicted in Fig. 4b in the thermogravimetric curves of representative biomaterials, they are suitable for being used in vapor phase adsorption at a temperature no higher than that range.

Another key feature for biosorbents for water adsorption is their pore size. As the size of water molecule is smaller (0.28 nm), as compared to ethanol (0.44 nm) and butanol (0.51 nm), it can penetrate more easily into the pores of the adsorbent (Dhabhai et al. 2018; Jayaprakash et al. 2017; Sun et al. 2007). In addition, the biomass particle size also affects the adsorption capability. Sun et al. (2007) compared three different biomaterials as adsorbent and found that higher water adsorption and selectivity were obtained with barley straw compared with wheat straw and crab shells. The smaller the particle size, the higher the surface area which may affect the adsorption rate of water or ethanol. Other operation parameters such as vapor flow rate, temperature, and water concentration have a large impact on performance of ethanol dehydration by starch-based adsorbents and cellulose-based adsorbents (Baylak et al. 2012).

Bioethanol and biobutanol are superior fuel additives. Bioalcohols require extensive purification after their fermentative production as current fuel alcohol standards require > 99% purity. Adsorption is one of the most efficient and cost-effective technologies for dehydration of bioalcohols. Based on their low cost of production and widespread availability, biobased materials which contain cellulose, hemicellulose, proteins, and lignin as their constituents have an excellent potential to be used industrially for dehydration of alcohols in vapor phase. Pressure swing adsorption is the choice of method industrially because of the ease of operation and maintenance and cost-effectiveness. The cellulose and hemicellulose content of biosorbents can be in the range of 50–80% and they primarily possess polar groups such as hydroxyl, carboxyl, and amine which help in water adsorption by polar–polar interactions. The pore size, low crystallinity, and thermal properties are also helpful characteristics for higher water adsorption. The low cost and water adsorption efficiency validate their industrial candidacy for alcohols dehydration. Future work is required to improve the properties of biosorbents for adsorption application in liquid phase.

ABE: acetone–butanol–ethanol (ABE); BTU: British thermal unit; FTIR: Fourier transform infrared; MTZ: mass transfer zone; PSA: pressure swing adsorption.