(662e) Optimal Solvent Selection Using Computer-Aided Molecular Design (CAMD) to Obtain Octacosanol from Sugarcane Rind

India stands as the world's second-largest sugar producer and a significant consumer. With over 700 sugar factories contributing to an annual production of approximately 340 lakh metric tonnes of sugar [1]. This sector's milling and clarification processes generate substantial solid wastes, including bagasse, rind, press mud, and bagasse fly ash. Nearly one-third of each ton of crushed sugarcane is collected as sugarcane waste and leaves, comprising esters, aldehydes, carboxylic acids, and alcohols [2]. Press mud, a byproduct, contains valuable nutraceuticals such as policosanols, which are aliphatic alcohols with carbon atoms ranging from C20 to C32. Nutraceuticals, a fusion of "nutrition" and "pharmaceuticals," are natural health supplements. The United States accounted for the largest market share for Nutraceuticals in the North American region, owing to consumers' rising interest in preventive healthcare and self-medication and an increase in the aging population's demand for functional food and beverages to prevent or mitigate aging-related conditions. In addition, increasing healthcare spending and rising consumer disposable income, coupled with changing consumer perception towards nutraceuticals due to their proven health benefits, further propels the market's growth [3].

Policosanols, primarily composed of octacosanol (66% - 67%) [4], can be derived from sources like plant waxes (sugarcane wax, soybean wax, wheat extract), fruits, leaves, seeds, and animal wax like beeswax [5]. These policosanols exhibit various health benefits, including reducing platelet aggregation, preventing low-density lipoprotein (LDL) peroxidation, and addressing cardiovascular diseases [5], [6]. Additionally, Octacosanol (C28), a prominent component, has been noted for its potential anti-Parkinsonism effect [7]. Being hydrophobic, it is insoluble in water but partially soluble in edible oils [8]. Given its significance, there has been a growing interest recently in extracting octacosanol from sugarcane.

1-octacosanol is a long-chain alcohol available in plant waxes and is a member of the broader family of policosanols (20-36 carbon). It is used as a dietary supplement and is known to have the potential for reducing blood cholesterol, stress-related sleep disorders, and benefits for Parkinson's disease. It contains antioxidants. A detailed review of the health benefits of octacosanol is available [9]. Irmak et al. (2006) report the number of dietary supplements containing policosanols that are commercially available in the US market [5]. Most of these products are prepared from beeswax or sugar cane extracts. While octacosanol is found in many plant waxes, its separation from filter mud produced after sugarcane juice clarification using extraction with hot ethanol reflux has been reported [10]. A US patent by Chinen (2009) [11] also reports a method of extracting octacosanol from sugarcane skin and extracting it from an organic solvent such as benzene, isopropanol, and hexane. The extract is then eluted using benzene and further purified using thin-layer chromatography. Analysis of octacosanol has been reported using GC at 320°C and an HP-5 capillary column with helium as carrier gas. The extraction step needs to be followed by a purification step that will fractionate the desired compound from the vast number of phytochemicals available in the filter press mud.

The primary goal of current research is to develop an economically viable process to obtain octacosanol from sugarcane wax or filter press mud. This involves starting from optimizing solvents used in extraction and adsorbents used in adsorption steps to complete process design and operations. In this paper, we address the first step of optimizing solvents used for extraction of octacosanol.

GCM for Solvent Selection

In the first stage of the process, we propose a framework for computer-aided molecular design (CAMD) to generate optimal solvents and adsorbents for efficient and cost-effective purification of octacosanol from sugar cane filter mud and wax.

A group contribution method (GCM) is a technique to estimate and predict thermodynamic and other properties from molecular structures. CAMD is the reverse use of the group contribution method to generate molecules with desirable properties.

CAMD for solvent selection involves consideration of various properties like distribution coefficient, and solvent selectivity. These properties can be defined in terms of infinite dilute activity coefficients as given below.

Distribution coefficient (m) is a measure of solvent capacity that provides an equilibrium relation between extract and raffinate [12]. The relation between infinite dilute activity coefficient γ^∞ and m is defined as shown in Eq. 1, [13]

m ≈ γ^∞_B,A/γ^∞_B,S (1)

where A, B, S represent raffinate, solute, and solvent, respectively.

Solvent selectivity (β) is defined as the measure of effectiveness of solvent to separate solute from the rest [12]. The relation between infinite dilute activity coefficient γ^∞ and β is shown in Eq. 2, [13]

β = m_B/m_A ≈ γ^∞_A,S/γ^∞_B,S (2)

The group contribution method of UNIFAC can be used to obtain the infinite dilute activity coefficients.

CAMD for Solvent Selection

CAMD methods are successfully applied to separations such as distillation, extraction, and absorption. However, this is the first time such an approach has been taken to extract and purify a nutraceutical from biomass.

An optimization approach to CAMD involves solving the following problem, Eq.3, using a combinatorial optimization method, as the resulting problem involves the combinatorial explosion of alternatives.

Maximize m = (γ^∞_B,A/γ^∞_B,S)(MW_A/MW_S) : Maxime distribution coefficient for solvent design (3)

Subject to

β = (γ^∞_A,S/γ^∞_B,S)(MW_B/MW_A) ≥ S : Solvent selectivity constraint for solvents (4)

∑∑V_ij ≤ ε : Structural Constraint (5)

where MW is molecular weight

UNIFAC model for calculation of activity coefficients

This poses a challenge to optimization techniques. The possibility of obtaining local minima or maxima is very high in such problems. Some of the biomimetic optimization techniques, like the ant colony optimization, show a higher probability of obtaining global solutions. However, these techniques can be computationally intensive. In earlier work, we have increased the efficiency of ant colony optimization by using the Hammersley Sequence Sampling (HSS) to generate random numbers needed for the algorithm [14], [15]. We have used this Efficient Ant Colony Optimization for our work here.

Results and Discussions

The groups used for the generation of molecules are tabulated in Table 1.

Using the above resources, the solvents generated are tabulated in Table 2 along with distribution coefficient values and solvent selectivity values.

Theoretically, for such a system of compounds, solvents were generated. Also, using this method, we found m and β of some of the common solvents like n-hexane (2.336 & 0.488), toluene (5.601 & 0.915), and ethanol (0.060 & 1.479), respectively. It has been found experimentally that these three solvents do require little heating for wax to dissolve and form a clear solution. Hence, the model used here has a good validation for separating the desired component from the rest. It can be seen from Table 2 that the solvents generated are better solvents in terms of distribution coefficient and selectivity than the commonly used solvents.

In the next step, we will validate these solvents using experiments.

Conclusion

Octacosanol is one of the policosanols which has wide-range of health benefits. This can be obtained from sources like rice-bran, beeswax, sugarcane, seeds, leaves, etc. Achieving this may be done in two stages: extraction (or leaching) and adsorption. Our present focus would be on selecting the solvent for extraction. This has been achieved using a computational technique called CAMD incorporating the UNIFAC model. Various solvents were generated with respect to certain properties like distribution coefficient and solvent selectivity.