Surfactants produced from carbohydrate derivatives: Part 2. A review on the value chain, synthesis, and the potential role of artificial intelligence within the biorefinery concept
Abstract
This comprehensive and critical review explores the synthesis and applications of carbohydrate-based surfactants within the biorefinery concept, focusing on biobased sugar-head molecules suitable for use across several manufacturing sectors, including cosmetics, pharmaceuticals, household products, detergents, and foods. The main focus relies on sustainable alternatives to conventional surfactants, which could reduce the final manufacturing carbon footprint of several industrial feedstocks and products. A thorough analysis of raw materials, highlighting the significance of feedstock sources, and the current biobased surfactants and rhamnolipid biosurfactants production trends, is presented. Key organic reactions for the production of sorbitan esters, sucrose esters, alkyl polyglycosides, and fatty acid glucamines, such as glycosidation, acylation, and etherification, as well as the production of rhamnolipids through fermentation are described. Given the scarce literature on the characterization of these surfactant types within the hydrophilic–lipophilic deviation (HLD) framework, the surfactant contribution parameter (SCP) in the HLD equation for sugar-head surfactants is critically assessed. The economic landscape is also discussed, noting the significant growth in the biobased surfactants and biosurfactant market, driven by environmental awareness and regulatory changes, with projections indicating a substantial market increase in the forthcoming years. Finally, the promising potential of generative artificial intelligence (AI) in developing customized surfactant molecules, with optimized properties for targeted applications, is emphasized as a promising avenue for future research.
Graphical Abstract
Potential application of AI in surfactants structure prediction.
INTRODUCTION
Climate change and the shift towards biobased surfactants
Climate change has emerged as a major concern for society in recent years, mainly as a consequence of the utilization of non-renewable resources in industrial processes (Hairon Azhar et al., 2022; Vera, Suarez, et al., 2023; Zambrano et al., 2021). This has resulted from the increased emissions of greenhouse gases, pollution, and the depletion of natural resources, all of which severely impact the environment (Sun et al., 2018; Walker & Rothman, 2020). In response to these challenges, there is a growing need for a shift toward the adoption of sustainable and eco-friendly alternatives to traditional fossil-based and non-renewable materials (Dudefoi et al., 2018; Philippini et al., 2020; Vera et al., 2022). These alternatives, often biodegradable, non-toxic, and derived from renewable resources, offer numerous advantages over their non-renewable counterparts in different areas, including the synthesis of surfactants (Hairon Azhar et al., 2022; Tan & Lamers, 2021).
Surfactants are amphiphilic compounds mainly produced from petroleum-based feedstocks, traditionally associated with a significant carbon footprint (Salager et al., 2023; Salager, Marquez, Bullon, & Forgiarini, 2022). Considering the crucial role that surfactants play in the formulation of a wide range of industrial products (detergents, emulsifiers, cosmetics, dispersants, and wetting agents), there is an urgent need for transitioning towards sustainable production patterns involving biobased surfactants (Hayes, 2009; Hayes et al., 2019; Hubbe et al., 2022; Ortiz et al., 2022). Biobased surfactants can be derived from various sources, including lipids, carbohydrates, as well as other alternatives such as lignin (Bajwa et al., 2019; Le Guenic et al., 2019; Lebeuf et al., 2018; Ontiveros et al., 2021).
Sugar-based surfactants typically consist of a hydrophobic moiety, often composed of a lengthy alkyl chain, and a hydrophilic head group, which is formed by carbohydrate molecules like glucose and sucrose, among others. As for the hydrophilic moiety, while safety and toxicity considerations come into play, a significant proportion of synthetic nonionic and ionic surfactants are still fossil-based, with nearly half of these surfactants being derived from polyethoxylated units (Santos et al., 2023). Consequently, the search for alternatives to the hydrophilic moiety remains an important challenge (Chin et al., 2023).
The production of biobased surfactants poses a number of challenges, including those related to selectivity and yield, to ensure that the process is economically viable (Mika et al., 2018; Smith, 2019). To address these challenges, the biorefinery concept has emerged as a promising approach, as it enables the integrated production of a range of value-added products from different raw materials, including carbohydrates (Gaudin et al., 2019; Le Guenic et al., 2019; Lichtenthaler, 2010; Polat et al., 2001). The circular bioeconomy aims to use biomass as a sustainable and renewable resource for the production of a range of value-added products, including chemicals, materials, and energy (Chandel et al., 2020; Tan & Lamers, 2021; Ubando et al., 2020). The use of biobased surfactants can be integrated into the circular bioeconomy through the use of various feedstocks and production pathways (Becker & Wittmann, 2019; Gaudin et al., 2019; Ortiz et al., 2022). Raw materials for these surfactants can be obtained from crops, such as corn, sugarcane, or cassava, as well as from food waste and other sources (Hayes et al., 2019; Ortiz et al., 2022; Vera, Zambrano, et al., 2023). However, the production of certain biobased materials with feedstocks derived from crops can negatively impact ecosystems or use land that could otherwise be used for food production (Begum et al., 2023; Dudefoi et al., 2018; Hairon Azhar et al., 2022). The use of crop waste allows addressing this by using a circularity framework, thus reducing environmental impact concerns (Byrne, 2023; Vera, Suarez, et al., 2023).
Biobased surfactants market growth and technical developments
The global surfactant market has experienced substantial growth, reaching approximately USD 43 billion in 2021, and is projected to reach USD 58 billion by 2028, with a compound annual growth rate (CAGR) of 4.9% (Markets and Markets, 2022; Nagrale, 2023). This growth is primarily driven by increased hygiene awareness post- Covid-19 pandemic and the demand for eco-friendly products (Marquez et al., 2022; Morone et al., 2021). Within this market, four distinct segments have emerged: synthetic surfactants, biobased surfactants, chemically synthesized biobased surfactants, and biosurfactants. In general, the cost of producing a biobased surfactant or a biosurfactant is higher than for synthetic ones (Hayes et al., 2019). However, biobased and biosurfactants, accounting for around 40% of the total surfactant market share, have seen remarkable growth, valued at USD 18 billion in 2022, and is projected to grow from USD 19 billion in 2023 to USD 26 billion by 2032, with a CAGR of 3.9% (Nagrale, 2023). This surge is attributed to increasing environmental concerns and stricter regulations on the use of synthetic surfactants (Zargar & Srivastava, 2024), which is further supported by a notable increase in research and technology development, indicating a 46% increase from 2010 to 2020 (Sayyed et al., 2022). The global market of sugar-head surfactants is predominantly led by alkyl polyglycoside (APG) surfactants (biobased but synthetic surfactants), with almost 23% of the global sugar-based surfactants by 2021 (Jha, 2022). On the other hand, glycolipids constitute the largest market share at 38% in 2021 for emerging biosurfactants (biosynthesized surfactants) (Farias et al., 2021; Zion Market Research, 2020). These types of biosurfactants are further subdivided into trehalolipids, rhamnolipids, and sophorolipids. Among these, sophorolipids and rhamnolipids have the most predominant interest to be produced at large scale (Adu et al., 2023).
On the other hand, the global biosurfactants market is projected to grow from USD 4 billion in 2022 to USD 6 billion by 2029, at a CAGR of 5.4% (Fortune Business Insights, 2022a). Among the main manufacturers Evonik (Germany), BASF (Germany), Arkema (France), Nouryon (Netherlands), Croda International (UK), Stepan Company (US), Kao Corporation (Japan), Sasol (South Africa), and Clariant (Switzerland) can be mentioned. From a technical perspective, the surge in market demand is being met through the enzymatic functionalization and coupling of carbohydrate-based heads to fatty aliphatic chains, forming the structural foundation of biobased surfactants (Agger & Zeuner, 2022).
By 2021, household detergents claimed the largest biosurfactant market share, followed by personal care, industrial and institutional cleaners, among others (Markets and Markets, 2022). While biosurfactants are gaining popularity in personal care products, mainly for moisturizing creams, the demand for household cleaners is expected to grow more rapidly (Roelants & Soetaert, 2022). This acceleration is attributed to the advocacy for eco-friendly cleaning products by consumers, governments, and regulatory bodies, coupled with increased consumer spending on hygiene products (Fortune Business Insights, 2022b). This is also reflected in the increasing willingness of consumers to pay a premium price for biobased alternatives (Herrmann et al., 2022; Morone et al., 2021). One of the advantages of the use of biobased surfactants and biosurfactants when compared to fossil-based ones is in their use in cosmetics and personal care products. Fossil-based surfactants are being regulated by government agencies such as the FDA through the Federal Food, Drug, and Cosmetic Act and the Modernization of Cosmetics Regulation Act of 2022 and the Cosmetics Regulation (EC) No 1223/2009 in Europe, in personal care and cosmetics industry. This is the case of ethoxylated surfactants, which can generate 1,4-dioxane in their synthesis (Hayes et al., 2022; Ortiz et al., 2022).
Europe has the largest revenue share of biosurfactants, with more than 48% in 2021, followed by Asia Pacific and North America (Fortune Business Insights, 2022b). Stringent environmental laws in Europe and a robust research and development infrastructure contribute to the region's leadership in the biosurfactant market (Roelants & Soetaert, 2022). On the other hand, the South American agro-industrial sector has an abundance of feedstocks that can be used as a substrate for surfactant production, which is expected to drive expansion in the region (Markets and Markets, 2022).
Although biosurfactant production, especially in the case of glycolipids, is on the rise, it is still not competitive compared to synthetic alternatives due to high production costs and low production capacity (Emergen Research, 2021; Müller, 2021). Synthetic and conventional surfactants, such as sodium dodecyl sulfate and plant-based amino acid surfactants have an average cost of USD 1–4/kg. On the other hand, sophorolipids and rhamnolipids can have a cost range between USD 6 and 34/kg (Farias et al., 2021), depending on conversion yields, substrate availability and price, processing times, downstream processing costs, energy requirements for sterilization, and maintenance of biological cultures (Ahalliya et al., 2022). Various renewable sources, including agricultural waste, high-oleic soybeans, and microalgae, are being explored for their potential as economically viable feedstocks. In some cases, government subsidies play a critical role in making these natural feedstocks economically viable (Axelsson et al., 2012). These subsidies can help offset the higher costs associated with the production and purification processes, which are often more expensive compared to their synthetic counterparts due to the need for specialized equipment and technologies (Bardhan et al., 2015; Vu et al., 2020).
A significant challenge lies in optimizing the fermentation process (including temperature, pH, and aeration) to reduce both time and resource intensity, thereby improving yield (Zion Market Research, 2020). Additionally, extracting and purifying biosurfactants from the fermentation broth is complex and expensive but necessary for improving yield, quality, and sensory perception (odor and color), primarily for cosmetics and personal care products (Müller, 2021). It should be noted that most surfactant manufacturers are based on synthetic chemistry and do not possess bioprocessing facilities. This is one of the reasons prices are driven up, as a significant initial capital investment is required for specialized equipment to scale microbial fermentation into large-scale production (Roelants & Soetaert, 2022). However, recent advances in fermentation and bioprocessing engineering offer opportunities to develop cost-effective techniques for producing more affordable and competitive biosurfactants. The past decade has seen the granting of many patents related to biosurfactant development and production, which indicates ongoing research efforts (Sayyed et al., 2022). Some inferences on producing rhamnolipids and sophorolipids at industrial scale can be extracted from Table 1. Substrate type and quality are crucial for the economics of the production process. A significantly higher minimum selling price (MSP) when the substrate is food waste compared to glucose and sugarcane bagasse can be deduced from this analysis. On the other hand, the biorefinery concept plays a crucial role in the economics for biosurfactant manufacturing, particularly in pretreatments typically applied to biomass sources (e.g., delignification, bleaching, enzymatic hydrolysis, etc.) that can be potentially used as substrate for biosurfactant production. The lower MSP for biosurfactants in certain cases is not solely attributed to the presence of co-products from a biorefinery process, such as using sugarcane bagasse after juice extraction for surfactant production. Instead, the reduction in MSP might be more closely tied to the specific technologies employed in the surfactant production process for each substrate. For instance, glucose presents a readily available substrate, potentially simplifying the process and reducing costs. Conversely, the variable composition in food waste could introduce complexity and unpredictability, negatively impacting production efficiency and costs.
Product | Substrate | Annualized CAPEX (USD/kg) | Interest rate (%) | Annual OPEX (USD/kg) | MSPa (USD/kg) |
---|---|---|---|---|---|
Sophorolipidb | Food waste | 7.5 | 7 | 20.2 | 20.4 |
Sophorolipidc | Sugarcane bagasse | 5.5 | 11 | 10.5 | 18.2 |
Rhamnolipidd | Food waste | 18.7 | 7 | 72.1 | 36.0 |
Rhamnolipide | Glucose | —f | 7 | 13.6 | 6.2 |
- a Minimum selling price.
- b Scenario II – syrup production (78% purity); production of 1793 ton/year.
- c Scenario B – syrup production (90% purity); production of 594 ton/year.
- d Scenario I – syrup production (50% purity); production of 627 ton/year.
- e Syrup production (46% purity); production of 8100 MT/year.
- f Annualized CAPEX could not be calculated since project lifetime was not reported.
Despite these challenges, the production and consumption of biosurfactants, particularly glycolipids, are expected to increase. Evonik Industries, a leading specialty chemical company, has set up a new industrial-scale facility to manufacture rhamnolipid-based surfactants to meet the anticipated demand in collaboration with Unilever, one of the major global manufacturers of home and personal care products (Markets and Markets, 2022). Like many other companies, Unilever has expressed its ambition to eliminate fossil feedstocks in the production of cleaning products by 2032 (Unilever, 2020). Evonik and Unilever are partnering and making substantial investments in response to consumer demand for sustainable products, as demonstrated by the success of the rhamnolipid-based liquid detergent QUIX (Roelants & Soetaert, 2022). While synthetic and biobased surfactants currently have differing production capacities, the superior growth rate of biobased surfactants is evident from their CAGR comparisons over recent years—4.9% for synthetic surfactants versus 5.6% for biosurfactants. (Roelants & Soetaert, 2022). This trend underscores the significant potential and market opportunity for biosurfactants. Consequently, the biosurfactants market is on an upward trajectory and is anticipated to witness a continued rise in value in the near future, driven by increasing consumer demand for more sustainable products.
Bibliometric data analysis of biobased surfactants and rhamnolipids biosurfactants
A detailed review of literature centered on the synthesis of biobased surfactants and rhamnolipids was conducted, employing specific keywords within the titles and abstracts of articles found in the Dimensions and Scopus Databases. Figures 1 and 2 provide an overview of the bibliometric landscape maps, visualized through the VOSviewer text analytics module. Figure 1 illustrates the countries with the most publications on biobased surfactant research, offering a visual representation of publication patterns and collaborative networks. Table 2 lists the journals leading in biobased surfactant publications. Likewise, Figure 2 shows the distribution of publications on rhamnolipid synthesis by country and publication volume, using the keywords "rhamnolipids" + "synthesis" + "biosurfactant". Table 3 identifies journals with the highest number of publications in this area, highlighting key platforms for rhamnolipid research dissemination.
Journal | Number of publications |
---|---|
ACS Sustainable Chemistry & Engineering | 15 |
Journal of Surfactants and Detergents | 8 |
Journal of Colloid and Interface Science | 6 |
Biomacromolecules | 6 |
ACS Omega | 6 |
ACS Applied Materials & Interfaces | 5 |
ChemSusChem | 5 |
Polymers | 4 |
International Journal of Biological Macromolecules | 4 |
RSC Sustainability | 4 |
Macromolecules | 3 |
Journal | Number of publications |
---|---|
Applied Microbiology and Biotechnology | 15 |
Microbial Cell Factories | 7 |
Journal of Biotechnology | 5 |
Frontiers in Microbiology | 4 |
Frontiers in Bioengineering and Biotechnology | 4 |
Applied Biochemistry and Biotechnology | 4 |
World Journal of Microbiology and Biotechnology | 4 |
BMC Microbiology | 3 |
Journal of Bacteriology | 3 |
Process Biochemistry | 3 |
Biotechnology Letters | 3 |
Literature gap and structure of the manuscript
Utilizing agricultural waste as a feedstock source for surfactant production enables the creation of value-added products with a lower carbon footprint compared to their counterparts derived from food crops or petroleum-based alternatives (Ortiz et al., 2022; Smith, 2019). An array of biobased building blocks derived from carbohydrate sources, mainly biomass, algae, and food waste, can serve as feedstocks for producing multiple biobased surfactants (Aricò, 2020; Hayes et al., 2019; Le Guenic et al., 2019; Ortiz et al., 2022; Tripathy et al., 2021). Over the past three decades, a diverse array of chemical pathways has been developed, enabling the creation of surfactant molecules derived primarily from carbohydrates found in biomass and waste (Arias et al., 2013; Estrine et al., 2019; Lichtenthaler, 2010).
Previous reviews have covered different parts of the value chain of biobased surfactants and biosurfactants synthesis, focusing on topics such as the use of various raw materials, methods for environmentally friendly synthesis, and their applications in different industries (Agger & Zeuner, 2022; Estrine et al., 2019; Hayes et al., 2019; Le Guenic et al., 2019; Mohanty et al., 2021; Nagtode et al., 2023). However, these studies tend to concentrate on particular aspects rather than encompassing the entire biobased surfactant production process. In the first part of this series, the biobased building blocks and surfactant derivatives that facilitate the production of carbohydrate-derived surfactants were reviewed (Ortiz et al., 2022). The present review seeks to provide a comprehensive analysis, including all stages of the value-chain involved in the synthesis of biobased surfactants and rhamnolipids. Chemical pathways for the production of sugar-head surfactant that can be used to obtain biobased molecules from carbohydrate sources are described. Moreover, the linkage between molecular structure and physicochemical formulation parameters, specifically, the hydrophilic–lipophilic deviation (HLD) multivariable equation, are addressed (Aubry et al., 2020; Forgiarini et al., 2021; Salager et al., 2020; Salager, Marquez, Delgado-Linares, et al., 2022). The literature offers limited information on HLD parameters for biobased surfactants (Hayes et al., 2019; Ontiveros et al., 2021). Hence, considering the trends in surfactant parameters within the HLD framework and the available data, this review outlines the reported surfactant contribution parameter (SCP) of sugar-head surfactants, and how the sugar head group could influence the SCP values.
This work is divided into the following sections. In section Raw materials, and industry trends on the synthesis of biobased surfactants and biosurfactants, we present an in-depth examination of the raw materials integral to the development of biobased surfactants and rhamnolipids biosurfactants. It includes a summary of current industry trends on the synthesis of biobased surfactants and biosurfactants production and recommendations on analysis techniques for these types of surfactants. Following this, section Relevant organic reactions for the synthesis of carbohydrate-derived biobased surfactants details key reactions including: Glycosidation of Alcohol, Glycosidation of Fatty Alcohols with Pentoses, Regioselective Acylation of Sucrose, Acylation of Other Monosaccharides and Disaccharides, Etherification of Saccharides and Disaccharides, Synthesis of N-Alkyl-Aldosylamines and their derivatives. Section Rhamnolipids derived from fermentation explores Rhamnolipids Derived from Fermentation, and Section Insights into HLD parameters of sugar head surfactants shows trends on the SCP parameter of Sugar Head Surfactants. Sections Recent advances in the field of carbohydrate-derived surfactant synthesis and future perspectives Recent Advancements in the Field of Carbohydrate-Derived Surfactant Synthesis and Future Perspective, provide insights into the latest developments and future directions in this field. The review concludes with Section Artificial intelligence and generative AI to develop advanced biobased molecules, where we include a perspective on the potential use of deep learning and generative AI to design new surfactant molecules for specific applications.
RAW MATERIALS AND INDUSTRY TRENDS ON THE SYNTHESIS OF BIOBASED SURFACTANTS AND BIOSURFACTANTS
Throughout history, the utilization of biological resources for the synthesis of surfactants has been a common practice, primarily motivated by two key factors: (i) the innate hydrophobic and hydrophilic properties of specific biomolecules, such as lipids and carbohydrates, which offer the essential duality required (Jessop et al., 2015), and (ii) the transformation of low-value crops and agricultural residues into valuable commodities (Domínguez Rivera et al., 2019; Gaur et al., 2022; Santos et al., 2023). In comparison to the modification of fats and oils, the development of sugar-derived surfactants is a relatively recent advancement. Research findings demonstrate that surfactants falling within the APG category exhibit exceptional efficiency, particularly in solubilizing hydrocarbons and organochlorine compounds during processes such as soil oil removal, the formulation of pesticides and agrochemicals, as well as their utility as carriers for medicinal compounds (e.g., amphiphilic glucan derivatives) (Chin et al., 2023). They offer the potential to replace more hazardous and allergenic surfactants in protein synthesis (e.g., trehalose fatty acid esters) (Le Guenic et al., 2019). Among these, a few exhibit antibacterial properties, like sugar-based Gemini surfactants (Gonçalves et al., 2023), while others, such as chemically pure APGs, are recognized for their non-denaturing characteristics (Mansi et al., 2023).
Within the spectrum of sugar molecules, only a few align with the stipulated standards encompassing quality, cost-effectiveness, and widespread availability. In addition to the category of synthetic carbohydrate-based surfactants, there are naturally occurring surfactants, commonly referred to as glycolipids, in the field of biochemistry. These glycolipids are predominantly situated in the cell membranes of diverse living organisms.
Feedstock sources and production of biobased surfactants within the biorefinery concept
The fundamental constituents of biobased surfactants include fatty acyl groups derived from sources such as oilseeds (e.g., palm, palm kernel, coconut, and various other oils, including algae), animal fats, and related derivatives (such as fatty alcohols and amines) (Figure 3). Additionally, carbohydrates, proteins, extractives, and their derivatives, alongside various byproducts from biorefineries (such as glycerol and extractives), play a significant role in the composition of these surfactants. In contrast to surfactants sourced from fossil fuels, the manufacturing and application of biobased surfactants result in reduced carbon dioxide emissions (Hayes et al., 2019; Ortiz et al., 2022). The structure–property relationships of biobased surfactants are crucial in determining their suitability for various applications such as home care, personal care, agrochemicals, and cosmetics. The challenges in replicating the performance of conventional petroleum-based surfactants stem mainly from the inherent differences in molecular structures between biobased and synthetic surfactants (Foley et al., 2012; Salager, Marquez, Bullon, & Forgiarini, 2022). This section provides an overview of feedstock sources used in the production of biobased surfactants.
The biorefinery concept, at its core, is linked to the efficient utilization of biobased feedstocks, which may include lignocellulosic biomass, oilseed crops, and aquatic organisms. These resources serve as the basis for generating a range of products, including fuels, chemical intermediates, fine chemicals, and materials. This production process mirrors the systematic fractionation seen in petroleum refining at petrochemical facilities, where various fractions are derived through methods such as distillation. These fractions are then employed in distinct applications, such as short- and medium-chain alkanes for fuels, long-chain alkanes for lubricants, and aromatics for the synthesis of chemicals and polymeric materials (Cherubini, 2010).
The primary source of biobased surfactant feedstock for the hydrocarbon chain is the fatty acyl group. These fatty acyl groups are typically extracted from oilseeds as triacylglycerol (TAG). However, they can also be sourced from byproducts of oleochemical processes, such as free fatty acids (FFA) or phospholipids. For instance, soap stock, which is a byproduct formed during degumming, can yield these fatty acyl groups. When incorporated as the lipophilic building blocks for surfactants, these fatty acyl groups are usually utilized in the form of FFA or fatty acid (FA) esters. These forms are obtained through processes like hydrolysis or alcoholysis of TAG, respectively (Smith, 2019).
Many surfactants based on fatty acids incorporate an ester bond, which serves as a link between the hydrophilic and lipophilic components. Ester bonds provide valuable attributes like biodegradability and biocompatibility (Hayes et al., 2019; Smith, 2019), making them well-suited for surfactants employed in food, cosmetics, personal care items, and pharmaceuticals. However, the ester bonds are relatively susceptible to degradation, limiting their application in certain product sectors, such as laundry detergents. In contrast, more robust bonds, including ethers, amides, and carbonates, are better suited for these applications. To facilitate the formation of these stable bonds, it is possible to convert fatty acyl groups into fatty alcohols or fatty amines.
The primary sources of high-lauric acid oils frequently utilized to produce non-food biobased surfactants include palm kernel oil, palm stearin (a byproduct rich in palmitic acid obtained from palm oil production), and coconut oil (Martínez et al., 2022). The preferred fatty acyl feedstock for this purpose typically comprises 10–14 carbon atoms and lacks double bonds, which enhances its resistance to oxidation. In applications related to food, biobased surfactants often rely on high-oleic acid oils, including those derived from corn, olives, cottonseeds, palm, or soybeans, as the primary sources of lipophilic components (Sharma et al., 2023).
The hydrophilic segment of the surfactant can also be sourced from constituents found in oilseeds, including polysaccharides and proteins. An economical option for this hydrophilic component is glycerol, a byproduct readily obtained from biodiesel manufacturing. Glycerol can be applied directly, for instance, in the production of monoacylglycerols (MAG) or their acylated or ethoxylated derivatives. Alternatively, it can be transformed into other glycols, such as glyceric acid (Bhadani et al., 2020), 1,2-propanediol, 1,3-propanediol, and glycerol carbonate. It can even be polymerized into polyglycerol, yielding a diverse mixture of linear, branched, and cyclic oligomers with varying degrees of polymerization (Ciriminna et al., 2015).
On the other hand, sugars are a frequently employed resource in the biorefinery process, serving as a valuable feedstock for the creation of hydrophilic components in surfactants. This includes the development of related derivatives, such as sugar alcohols (for example, sorbitol and sorbitan) (Liang et al., 2023), as well as derivatives like furfuryl (Park et al., 2016) and levoglucosanyl compounds, and glucaric acid (Machado et al., 2023), which were reviewed in detail in part 1 of this series. While ethoxylate groups are typically sourced from petrochemicals, these crucial components found in nonionic surfactants have the potential to be obtained from biobased ethylene. This biobased ethylene can be generated from bioethanol, which in turn is derived from sugarcane (Jamil et al., 2022).
The actual landscape of feedstock sources for the production of biobased surfactants
Substrate raw materials have a significant impact on the final cost of bioproducts, accounting for up to 80% of total costs in some cases (Longati et al., 2023). Regarding the economics of the substrates used in the production of biobased surfactants (presented in Figure 3), there is a high demand for these resources in other areas, such as biofuels and renewable chemicals. This competitive demand creates limitations in the supply chain for large biosurfactant production volumes. It also increases the final biosurfactants price as the cost of substrates can make up to 50% of the total manufacturing cost (Farias et al., 2021). In the case of sugar-based surfactants, the presence of essential sugars such as glucose is crucial. In this context, the sugar source can originate from food waste, and starch sources (e.g., corn), to cellulosic substrates like agricultural residues (e.g., wheat straw) and textile wastes (Longati et al., 2023; Ortiz et al., 2022; Vera, Vivas, et al., 2023). An extensive compilation of sources for carbohydrates production was presented in detail in Part 1 of this series (Ortiz et al., 2022). The quality of the sugar (e.g., purity and chemical structure) plays a fundamental role in the production cost of the biosurfactants. Some techno-economic studies indicate that the cost per ton of glucose syrup can be around USD 600–800/ton via enzymatic hydrolysis of lignocellulosic materials. Moreover, commercial sugar can be around USD 500/ton (Table 4) (USDA ERS, 2023).
Sugar | Cost, USD/ton | Ref. |
---|---|---|
Glucose | 300–500 | (Vera et al., 2022) |
Fructose | 400–500 | (SelinaWamucii, 2023) |
Xylose | 300–500 | (Ou et al., 2021) |
Arabinose | 400 | (SelinaWamucii, 2023) |
Fatty acid | Source | Carbon chain length | Cost, USD/ton | Ref. |
---|---|---|---|---|
Caprylic | Palm oil, coconut oil | 8:0 | 6915 | (CHEMANALYST, 2023a) |
Capric | Palm oil, coconut oil | 10:0 | ||
Lauric | Coconut oil and palm kernel oil | 12:0 | 1624 | (CHEMANALYST, 2023b) |
Mystiric | Bovine milk, palm kernel oil, coconut oil | 14:0 | ||
Palmitic | Palm oil, soybean oil, bovine milk | 16:0 | 2900 | (CHEMANALYST, 2023b) |
Stearic | Fish, butter, beef tallow, soybean, olive oil, corn oil | 18:0 |
Researchers have studied agro-based industrial waste products as substrates in the fermentation process to address this issue. This waste has been found to reduce production costs, improve biosurfactant yield, and lower downstream processing costs, although there are challenges related to purification and purity of the products (Ahalliya et al., 2022). On the other hand, the surfactant molecular structure and performance will be driven by the length of the lipophilic tail, generally given by fatty acids obtained from vegetable oils or animal sources. In this context, it is important to mention that costs will be sensitive to the desired application of the surfactant. Table 4 shows how the cost per ton of some fatty acids is driven by the length of their carbon chain.
Recent trends on the production of biobased surfactants and biosurfactants
- Alkyl glycosides and alkyl polyglycosides, produced via Fischer glycosylation reactions, are at the forefront due to their efficiency and cost-effectiveness. Their application in personal care and pharmaceuticals is propelled by consumer demand for safer ingredients and regulatory push for non-toxic alternatives, as well as by preferred performance properties such as foaming. With a consistent growth rate in the personal care industry, these surfactants are expected to maintain a steady market share, potentially growing at a CAGR aligned with the overall green chemicals sector, which some reports suggest could be between 5% and 10% in the coming years (Estrine et al., 2019; Lokesh et al., 2017).
- Polypentosides demonstrate a significant potential for market expansion. Their high yield and less stringent synthesis conditions from xylanase not only reduce production costs but also present an opportunity for scalable manufacturing. The property of reducing surface tension makes them suitable for a variety of industrial applications, where they can act as efficient wetting agents (Jérôme et al., 2018).
- Sucrose monoesters and long-chain fatty acid esters of maltose are niche products with specialized applications. Although their production is resource-intensive, the specificity of their synthesis allows for targeting high-value markets where consumers are willing to pay a premium for specialized functionality. Given the trend towards customization in industries such as cosmetics and pharmaceuticals, these surfactants may see a rise in demand, albeit in a more segmented market (Queneau et al., 2004).
- Monoethers of D-Glucose, D-Galactose, and D-Mannose offer a promising outlook due to their water solubility and ability to reduce interfacial tension. They have the potential to capture a significant portion of the market in specialized surfactant applications, especially in industries focusing on bioavailability and biocompatibility (Gozlan et al., 2016).
- N-Alkyl-Aldosylamines and derivatives face challenges due to their instability in aqueous solutions, which limits their direct application. However, with advances in chemical modification and stabilization techniques, they may find increasing use in niche markets, particularly where emulsifying capacity is critical (Costes et al., 1995; Latge et al., 1991).
- Rhamnolipids, despite their production challenges, are poised for significant growth. The demand for rhamnolipids is expected to surge in various sectors, notably in oil & gas, agriculture, cosmetics, and consumer cleaning products, driven by their biodegradability and effectiveness as well as mildness in complex formulations. Innovations in fermentation technology and genetic engineering may lead to more cost-effective and scalable production methods. Further, the current constraints associated with the pathogenic nature of Pseudomonas aeruginosa may be mitigated through the use of engineered organisms that do not pose the same hazard (Bahia et al., 2018; Zhao et al., 2021). Strategies to enhance yield, avoid the use of solvents, and improve downstream processing in fermentation process for biosurfactants production include: (i) utilizing advanced bioreactor designs that can help manage the unique challenges of biosurfactant fermentation, such as foaming; (ii) solvent-free extraction techniques, such as mechanical methods like centrifugation and filtration; or (iii) co-production strategies that integrate biosurfactant production with other valuable biochemicals, like biofuels or enzymes, can make the process economically viable. Additionally, innovations in downstream processing, such as membrane filtration and chromatographic methods can be deployed for simplifying purification (Sundaram et al., 2024).
Surfactant type | Production process | Yield and techno-economics | Industry trends | Ref. |
---|---|---|---|---|
Alkyl-glycosides and Alkyl-polyglycosides | Fischer glycosylation reactions under extreme conditions with toxic catalysts | Efficient synthesis process, wide commercial use and relatively economical | Widely used in personal care, pharmaceutical preparations, and membrane protein research, replacing hazardous surfactants | (Estrine et al., 2019; Lokesh et al., 2017) |
Polypentosides | Transglycosylation of xylane with alcohols under ambient pressure and temperature | Notably high monopentoside yield of 95%; less stringent synthesis conditions from xylanes | Alkylpolypentosides exhibit enhanced efficiency in reducing surface tension, applicable in synthesizing surfactants with tailored properties | (Jérôme et al., 2018) |
Sucrose Monoesters | Selective synthesis of specific saccharide esters through hydroxyl group protection and deprotection steps or regioselectivity | Time-consuming and resource-intensive separation process; requires elaborate and costly strategies for synthesis | Sucrose monoesters are used in regioselective esterification, with potential applications in specialized fields | (Queneau et al., 2004) |
Long-Chain Fatty Acid Esters of Maltose | Gradual addition of long-chain acid chlorides to a solution of maltose in dry pyridine | Yields between 15% and 28%; reduction of interfacial tension to as low as 34 mN/m | Long-chain fatty acid esters of maltose used in food, cosmetics, personal care items, and pharmaceuticals due to biodegradability and biocompatibility | (Allen & Tao, 2002) |
Monoethers of D-Glucose, D-Galactose, and D-Mannose | Selective preparation of monosaccharide monoethers using protective groups under Williamson etherification conditions | Achieving yields exceeding 70% under specific conditions | Monoethers of D-Glucose, D-Galactose, and D-Mannose show potential in specialized surfactant applications due to water solubility and interfacial tension reduction | (Gozlan et al., 2016) |
N-Alkyl-Aldosylamines and Derivatives | Condensation of a monosaccharide with a primary amine; preparation also possible from reducing disaccharides | Yields ranging between 60% and 70%; formation of gel-type aggregates with emulsifying capacity | N-Alkyl-Aldosylamines and Derivatives have emulsifying capacity, but instability in aqueous solutions limits direct application | (Costes et al., 1995; Latge et al., 1991) |
Rhamnolipids | Biosynthesis involving specific genes and pathways in Pseudomonas aeruginosa, synthesizing 3-(hydroxyalkanoyloxy)alkanoic acids | Challenges in large-scale production due to the pathogenic nature of the primary producing bacterium | Increasing demand for rhamnolipids in oil & gas, agriculture, food & beverage, cosmetics, and pharmaceuticals, despite production challenges | (Abdel-Mawgoud et al., 2011, 2010; Guzmán et al., 2023) |
Analytical techniques for characterizing biobased surfactants and biosurfactants
An overview of the current state-of-the-art characterization techniques for biobased surfactants and biosurfactants is presented in Figure 4. A comparative analysis on the sensitivity, resolution, quantitative capability, cost-effectiveness, technical ease, and information depth of analytical techniques used to characterize biobased surfactants and biosurfactants is provided. A scale from 1 to 5 was assigned for each attribute according to the technique (TLC, HPTLC, HPLC, FTIR, GC-MS/LC-MS, ESI-MS, MALDI-TOF-MS, NMR), being 1 very low and 5 a very high value of the attribute.
Each technique presents its own advantages and disadvantages. Thin-layer chromatography (TLC) provides simplicity and cost-effectiveness for preliminary screening, but is counterbalanced by limitations in quantification, resolution, and sensitivity, alongside an inability to furnish detailed structural insights (Ibrahim, 2018; Medhi et al., 2023; Silva et al., 2014). High-performance thin-layer chromatography (HPTLC) possesses resolution and swift analysis rendering it advantageous over traditional TLC, although HPTLC lacks comprehensive structural elucidation (Al-Wahaibi et al., 2014; Geissler et al., 2017). High-performance liquid chromatography (HPLC) coupled with mass spectrometry (MS) affords high-resolution, accurate, and quantitative analyses. It has been utilized for elucidating biosurfactant structures, albeit at a high cost (Dalili et al., 2015; Pantazaki et al., 2011). Fourier transform-infrared spectroscopy (FTIR) identifies biobased and biosurfactant functional groups, aiding molecular structure determination. It reveals vibrational frequencies of chemical bonds. While rapid and non-destructive, FTIR's qualitative bias and sensitivity to water content limit its quantitative accuracy and isomer distinction (Chakraborty & Das, 2017; Ratna & Kumar, 2022).
Advanced techniques, such as MS in conjunction with gas (GC-MS) or liquid chromatography (LC-MS), provide definitive molecular weight and structural elucidation. This sensitive technique offers detailed insights into functional groups and modifications but faces challenges in data interpretation, necessitating substantial investment and expertise (Anaukwu et al., 2020; Ibrahim et al., 2013; Shekhar et al., 2015). Electrospray Ionization Mass Spectrometry (ESI-MS) elucidates charge state distribution and molecular weights, enabling complex mixture analysis. Although it minimizes fragmentation, ESI-MS is prone to matrix effects, impacting measurement accuracy (Jimoh & Lin, 2019; Monteiro et al., 2007). Matrix-assisted laser desorption/ionization-time of flight mass spectroscopy (MALDI-TOF MS) identifies biosurfactant constituents and structural features. Despite its sensitivity and ability to analyze a wide molecular weight range, matrix selection and ion suppression effects pose significant challenges (Chakraborty & Das, 2017; Guo et al., 2012). Nuclear magnetic resonance (NMR) offers structural detail, identifying functional groups and elucidating stereochemistry. While it provides high-resolution characterization and dynamic insights, sensitivity limitations and complex spectra interpretation challenge its application, particularly for high-molecular-weight biosurfactants (Li et al., 2016). As an example, the physicochemical characterization of rhamnolipids involves sophisticated analytical techniques such as HPLC–MS for identifying RL congeners and determining their compositions. This method reveals the diversity in rhamnolipid structures, which directly influences their surface activity (El-Housseiny et al., 2020).
The body of knowledge and know-how developed on these analytics techniques over the years, through a synergy of university and industry research, has facilitated the characterization and analysis of products obtained during the synthesis of biobased surfactants. This is crucial for ensuring a degree of purity that enables formulators in industry to work with consistent mixtures, owing to the complex nature of the feedstocks used, in contrast to their fossil-based counterparts. Figure 4 illustrates a clear trade-off between resolution and cost, with GC-MS, NMR, and HPLC being among the less cost-effective methods, albeit with higher resolution, sensitivity, and quantitative capability.
RELEVANT ORGANIC REACTIONS FOR THE SYNTHESIS OF CARBOHYDRATE-DERIVED BIOBASED SURFACTANTS
The industrial-scale production of carbohydrate-based surfactants is relatively recent (Abdellahi et al., 2021; Gutsche & Behler, 1999; Kipshagen et al., 2019; Lang, 2002; Wang & Queneau, 2019). Commercially produced carbohydrate-derived surfactants include a wide variety of molecules from different sources and chemicals (Hayes et al., 2019; Ortiz et al., 2022). Among them, surfactants derived from biobased building blocks (i.e., succinic, fumaric, itaconic, citric, ascorbic, and levulinic acids, as well as furfural and 5-hydroxymethylfurfural [5-HMF]), according to the National Renewable Energy Laboratory (NREL, USA) (Bozell & Petersen, 2010; Kumar & Verma, 2020; Ortiz et al., 2022; Werpy & Petersen, 2004), and sugar-head surfactants as sucrose ester, sorbitan esters, alkyl polyglycosides, isosorbide and xylitol derivatives (Estrine et al., 2019; Krawczyk, 2018; Le Guenic et al., 2019; Ortiz et al., 2022). In this section, chemical pathways to synthesize sugar-head surfactants, such as glycosidation of alcohols and pentoses (Demchenko, 2008; Levy & Fügedi, 2005) to produce alkyl polyglycosides, acylation (Queneau et al., 2004; Wang & Queneau, 2019), and etherification of monosaccharides and disaccharides to obtain sucrose esters (Arias et al., 2013; Gagnaire et al., 1999; Queneau et al., 2004) and condensation of monosaccharides with primary amines to produce sugar-derived cationic surfactants (Hayes et al., 2019; Oskarsson et al., 2007), are discussed. Examples of industrial production and utilization of carbohydrate-based surfactants are shown in Table 6.
Fields of utilization | Specified suppliers | Production (ton/year) |
---|---|---|
Sorbitan esters | Nouryon, Cognis, Kao, Dai-Ichi Kogyo Seiyaku, SEPPIC | 20,000 |
Sucrose esters | Jiangsu Weixi, Cognis, Evonik, Croda, Dai-Ichi Kogyo Seiyaku | <10,000 |
Alkyl polyglycosides | Nouryon, BASF, Dai-Ichi Kogyo Seiyaku, and SEPPIC | 85,000 |
Anionic alkyl polyglycoside derivatives | Cognis | <10,000 |
Glycosidation of alcohols
Alkyl polyglycosides are synthesized via the glycosidation of long-chain alcohols with glucose. In chemical terms, glycosidation is defined as the transformation of a saccharide, which is in its hemiacetal form, to a glycoside (acetal) through the mediation of an alcohol in an acidic medium. Extending the scope of this definition, glycosidation involves a nucleophilic substitution, wherein a leaving group (often abbreviated as LG) on the glycoside donor is displaced by a hydroxyl group present on the glycoside acceptor (see Figure 5 for the reaction mechanism). In the context of Fischer glycosidation, the leaving group is typically protonated hydroxyl. However, alternative methodologies exist where the hemiacetal hydroxyl group is pre-converted to, or replaced by, a more efficacious leaving group prior to the glycosidation step (Hill et al., 2008; Kinanti et al., 2021; Pantelic & Cuckovic, 2014).
Consequently, the term “glycoside donor” is employed to encompass all derivatives fulfilling this role. Another pivotal component in glycosidation is the activator, which may serve as either a promoter or a catalyst. A designated activator facilitates the displacement of the leaving group located at the anomeric carbon of a given glycoside donor. Notably, the repertoire of activators compatible with a singular glycoside donor can be as comprehensive, if not broader, than the array of glycoside donors explored. Elaborating on this extensive topic exceeds the boundaries of the present work. Henceforth, the remaining discussion in this section will be confined to selected examples illustrating the application of Fischer glycosidation in the synthesis of surfactants. This focus is attributable to Fischer glycosidation being the most direct and economically viable approach for synthesizing alkylglucosides. Other glycosidation methods are reviewed by Levy and Fügedi (2005) and Demchenko (2008).
Fischer glycosidation stands as one of the most archaic techniques in the realm of chemical glycosidation. In this method, a deprotected hemiacetal glycoside donor undergoes acetalization with alcohol, mediated by an acid catalyst. Although protic acids were initially stipulated in the original methodology, a plethora of variants now exist that utilize Lewis acids (e.g., ZnCl2, FeCl3, BF3, SnCl4), ionic liquids, heteropoly acids, and clays as catalytic agents (Ryan et al., 2008). To maximize glycoside yield and circumvent the undesired self-condensation of glycoside donors, the reaction is generally conducted with a stoichiometric excess of alcohol (Ryan et al., 2008). Nevertheless, the protocol requires elevated temperatures and protracted reaction durations. Such conditions culminate in a product mixture that comprises furanosides and pyranosides in dynamic equilibrium (Levy & Fügedi, 2005). While the reaction condition does little to offer anomeric selectivity (Wang, 2010), the proportion of α-pyranosides is usually amplified due to their thermodynamic stability (Demchenko, 2008).
While glycosidation of unprotected hemiacetals typically lacks stringent selectivity, Ferrières et al. (1998) demonstrated that it is feasible to primarily yield alkyl-D-glycofuranosides from D-glucose, D-mannose, and D-galactose—all pyranose sugars—by employing glycosidation with long-chain alcohols in the presence of ferric trichloride. This methodology was further extended to synthesize alkyl-β-D-fructopyranosides derived from D-fructose (Ferrières et al., 1999). Goodby et al. (1995) showed that certain alkylfuranosides—namely, glucofuranosides, galactofuranosides, and manofuranosides—synthesized with n-octanol and n-decanol, could attain surface tension values in the range of 25 to 30 mN/m at their critical micelle concentration (CMC).
Fischer glycosidation serves as an expeditious and efficacious technique for the synthesis of alkylglycosides, therefore, its commercial-scale application in the production of alkyl polyglycosides is key. Nonetheless, the requisite high temperatures and extended reaction durations pose a significant drawback, leading to the formation of chromogenic by-products that necessitate subsequent purification, thereby escalating the overall production costs. The use of organic acid catalysts such as p-toluenesulfonic acid, coupled with a substantial excess of alcohol, has mitigated this issue to some extent. However, the incorporation of sulfonic acids introduces a contentious element in the synergy with the oil industry, especially in a process predominantly reliant on biomass-derived feedstocks.
Intriguingly, furan dicarboxylic acid and its decyl monoester have been shown to enhance both the yield and the chromatic purity of the product in the glycosidation of decanol with glucose via the Fischer method (van Es et al., 2013). This enhancement is ascribed to the in-situ formation of furane dicarboxylic acid decyl monoester, which functions as a phase-transfer agent in a system where the carbohydrate and alcohol initially constitute a heterogeneous medium. Moreover, the utilization of furane dicarboxylic acid confers a sustainability advantage, as it is derived from the oxidation of 5-hydroxymethylfurfural (HMF) (Bozell & Petersen, 2010).
Interfacial properties of alkyl polyglycosides
In addition to structural characteristics, it is imperative to examine the interfacial properties of alkyl polyglycoside surfactants. These properties facilitate the inference and projection of the colloidal applications of these amphiphilic compounds compared to their counterparts derived from petrochemicals (Haese et al., 2022; Rather & Mishra, 2013; Thenchartanan et al., 2020; Wu et al., 2021).
The comprehensive reviews conducted by von Rybinski (1996), von Rybinski and Hill (1998), Söderman and Johansson (1999), and Stubenrauch (2001) provide a detailed description and in-depth discussion of the interfacial properties of this class of surfactants. This section addresses the most relevant systematic surface effects resulting from their chemical structural differences, as well as the influence of temperature and electrolytes. In Table 7, fundamental interfacial properties of different alkylglycosides surfactants in pure water are presented.
Alkyl glycoside | CMC | Kp | DH | γCMC | Ref. |
---|---|---|---|---|---|
octyl-β-D-glucoside | 23.4 | — | 3 | — | (Vanaken et al., 1986) |
octyl-β-D-glucoside | 25 | 25 | — | ~30 | (Shinoda et al., 1961) |
octyl-β-D-maltoside | 23.4 | — | 3.1 | — | (Vanaken et al., 1986) |
n-octyl-β-D-glucopyranoside | 25.8 | 25 | — | — | (Pastor et al., 1998) |
C8/10-APGa | 0.05 | 25 | — | — | (Platz et al., 1995) |
decyl-β-D-glucoside | 2.2 | 25 | — | ~28 | (Shinoda et al., 1961) |
dodecyl-β-D-glucoside | 0.19 | 25 | — | ~39 | (Shinoda et al., 1961) |
dodecyl-glucoside | 0.17 | 60 | — | — | (Förster et al., 1995) |
lauryl-β-D-maltoside | 0.165 | — | 5.8 | — | (Vanaken et al., 1986) |
lauryl-α-D-maltoside | 0.156 | — | 5.5 | — | (Vanaken et al., 1986) |
C12/14-APGa | 0.003 | 25 | — | — | (Platz et al., 1995) |
oleoyl-β-D-maltotrioside | <0.005 | — | 8.4 | — | (Vanaken et al., 1986) |
- Note: The units of CMC, T, DH and γCMC are mM, °C, nm, and mN/m.
- a Industrial alkyl polyglycosides (APG) products with mixed alkyl chain lengths and head groups. The CMC unit for these surfactants is %m/m.
Glycosidic surfactants present critical micelle concentrations similar to those found in surfactants derived from the petrochemical industry, encompassing both ionic, zwitterionic, and non-ionic types (Agudelo et al., 2020; Barrios et al., 2022; Patiño-Agudelo et al., 2022; Patiño-Agudelo & Quina, 2022). The variation of 10 carbons in their hydrophobic chain (from octyl-β-D-glucoside to oleoyl-β-D-maltotrioside) promotes a three-order-of-magnitude reduction in CMC, indicating that the entropic factor primarily contributes to the thermodynamic stability of the formed aggregates. While the water-1H transverse relaxation effectively illustrates how the self-aggregation of APG into micelles impacts interactions with water and OH headgroups, either through steric hindrance of the OH groups or by modifying the network structure of water molecules at the micelle surface (Cardoso & Sabadini, 2013), it is crucial to recognize that the alkyl chain exerts a significantly greater influence on CMC compared to the number of glucose type groups in alkyl polyglycosides. Another relevant aspect related to CMC is its apparent invariability with temperature variations, as observed in Table 7 for dodecyl-glucoside, maintaining practically constant values at both 25 and 60°C. A more detailed analysis of this phenomenon can be found in the study by Aoudia and Zana (1998), which adopts a more comprehensive approach, integrating graphical representations and more concise results. Similar to conventional surfactants, the CMC of APGs decreases in the presence of strongly hydrated ions, while the aggregation number increases (Pastor et al., 1998).
Alkyl polyglycoside surfactants are stable in solution at room temperature, as evidenced by the Krafft point. They form nanometric-sized aggregates, with those containing ≤12 carbons forming spherical micelles and those with ≥12 forming rod-like micelles (Nilsson et al., 1998; Zhang et al., 1999). This results in a similar reduction in water surface tension compared to values reported by traditional surfactants. Moreover, a synergistic effect was observed when combined with an alcohol co-solvent, leading to the generation of ultra-low interfacial tension of 10 N/m (Iglauer et al., 2009; Li et al., 2019). This has the potential to significantly enhance oil recovery.
In addition to the surfactants presented thus far, in recent years, due to the significant utility of biobased surfactants, the family of alkyl polyglycosides has been explored as bioinspired anionics, known as anionic alkyl glycosides (Zan et al., 2023). They reduce surface tension similarly to non-ionics. However, the CMC increases by approximately 20% in a range of 30°C, and the size of the aggregates is around 10 times larger compared to non-ionics. Further studies are still needed to solidify the presence of these surfactants in the market.
Glycosidation of fatty alcohols with pentoses
Several methods have been for the synthesis of alkylglycosides from xylanes. A two-stage procedure (Bertho et al., 1997) involves the initial formation of n-butyl glycosides through the treatment of xylanes with n-butanol and an acid catalyst at 80°C. These n-butyl glycosides are then subjected to transglycosylation with the respective alcohols—namely, n-octyl-, n-decyl, and n-dodecyl—to yield polypentosides. Bouxin et al. (2010) demonstrated that higher alkylpentosides could be directly obtained via transglycosylation of xylane under ambient pressure and at 90°C, using a mixture of xylane and alcohol (butanol, octanol, or decanol), and introducing sulfuric acid at a concentration of 10%. The reaction is carried out by dispersing the xylane in the alcohol (butanol, octanol, or decanol) and adding sulfuric acid in a concentration of 10%. More recently it has been showed that it is not necessary to isolate xylane since alkylglycosides can be obtained by direct treatment of wheat stubble with decanol and acid at high temperature (90–110°C), with a monopentoside yield of 95% (Marinkovic et al., 2012). Notably, this method also generates a high-value residue enriched with glucose and lignin. Such residues can be subjected to saccharification using concentrated acid or enzymatic treatment, thereby recovering considerable quantities of glucose and lignin with yields surpassing those attainable from untreated hemicellulosic material (Ludot et al., 2014). The resultant glucose can further undergo glycosidation to augment the surfactant yield. Figure 6 schematically delineates the integration of lignocellulosic material for surfactant production within the context of a biorefinery concept (Marinkovic et al., 2012).
The synthesis of alkyl polypentosides can be alternatively realized through the telomerization of butadiene with saccharides such as D-xylose and L-arabinose. Telomerization entails the oligomerization of 1,3-dienes, also known as taxogens, concomitant with the addition of a nucleophilic compound referred to as the telogen. This mechanistic route is depicted in Figure 7a. Transition metals frequently act as the catalysts in this reaction scheme, with palladium standing as the most commonly employed (Behr et al., 2009). This telomerization approach has been effectively utilized to generate an array of compounds including monooctadienyl polyxylosides, polyarabinosides, and dioctadienylxilosides, as schematically detailed in Figure 7b (Estrine et al., 2004). Significantly, these synthesized compounds have manifested the capability to attenuate surface tension to levels as low as 35 mN/m (Hadad et al., 2006).
Alkyl polypentosides present several advantages over APGs, starting with their synthesis. The direct preparation of alkyl polypentosides from xylan requires less stringent conditions than does the synthesis of APGs from starch. This is because xylans possess less stable β-(1,4) glycosidic bonds compared to the α-(1,4) linkages in starch. Moreover, xylans are branched heteropolymers and thus exhibit fewer tendencies towards crystallinity, unlike starch, which consists of linear polymeric chains (Bouxin et al., 2010; Martel et al., 2010). Strategically, the production of alkyl polypentosides could offer significant advantages since xylans are derived from hemicellulose—a biomass fraction that is not only indigestible by humans but can also be sourced from agricultural residues. Conversely, APGs necessitate starch as a precursor. From a functional perspective, alkyl polypentosides are comparable to alkyl polyglycosides in terms of their ability to reduce surface tension but display slightly enhanced efficiency due to lower CMC values (Bouxin et al., 2010; Marinkovic et al., 2012; Martel et al., 2010). Notably, alkyl polypentosides with chain lengths ranging from 8 to 12 are as effective in reducing surface tension as conventional surfactants such as sodium dodecylbenzene sulfonate and lauryl ether sulfate (Bertho et al., 1997).
Furthermore, alkyl polypentosides serve as versatile platforms for synthesizing surfactants with tailored properties. For instance, Renault et al. introduced succinoyl substituents to alkyl polypentosides via esterification with biomass-derived succinic anhydride at 110°C (Renault et al., 2012). The modified surfactants displayed Krafft points below 0°C, along with enhanced water solubility.
Regioselective acylation of sucrose
Obtaining pure sucrose monoesters poses significant challenges on an industrial scale due to the polyfunctionality of the saccharide. The separation of different monoesters can be both time-consuming and resource-intensive. Laboratory-scale efforts, however, permit the application of elaborate and costly strategies to selectively synthesize specific saccharide esters. Such approaches often rely on a sequential series of hydroxyl group protection and deprotection steps. Alternatively, a more straightforward strategy leverages the preferential reactivity of specific hydroxyl groups over others, known as regioselectivity, which can be modulated by reaction conditions (Queneau et al., 2004).
Novel acylating agents such as 3-acyl-5-methyl-1,3,4-thiadiazol-2(3H)-thione have been deployed for the selective synthesis of monoesters from unprotected sucrose (Figure 8a). Other agents like 3-acyl-thiazoledin-2-thione (Figure 8) enable reactions under less stringent conditions, such as strong basic catalysts or elevated temperatures (Polat et al., 2001; Queneau et al., 2004). When such acylating agents are employed in the presence of sodium hydride or triethylamine and in a solvent such as dimethylformamide, 2-O-acylsucrose is predominantly formed. Upon subsequent addition of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU, Figure 8), intramolecular isomerization occurs, leading to 6-O-acylsucrose. Interestingly, DBU-catalyzed acylation in dimethylformamide predominantly yields 6-O-acylsucrose. In contrast, using 1,4-diazabicyclo[2.2.2]octane (DABCO, Figure 8) as the catalyst tips the balance towards the formation of 6′-O-acylsucrose (Chauvin & Plusquellec, 1991).
The primary hydroxyl groups, which are less prone to steric hindrance, are not always the ones that react more quickly (Baczko et al., 1995). Under specific conditions, the hydroxyl group at C-2 can exhibit higher reactivity. This increased reactivity is likely due to the acidity of the alcohol function at C-2. When sucrose is dissolved in aprotic polar solvents, it predominantly adopts two conformations. In both conformations, the oxygen atom linked to C-2 engages in intramolecular hydrogen bonding either with the hydroxyl group at C-1′ or C-3′, as delineated in Figure 9a (Benvegnu et al., 2008). This intramolecular hydrogen bonding endows the hydroxyl at C-2 with enhanced nucleophilicity. Consequently, when a catalytic amount of sodium hydride is added to the reaction mixture, it results in the formation of a highly nucleophilic and stabilized 2-oxyanion, as illustrated in Figure 9b. This oxyanion reacts swiftly with the acylating agent (Chauvin et al., 1993). Chauvin et al. (1993) elucidated that when the ratio of NaH to substrate exceeds the minimum necessary for the reaction, regioselectivity diminishes. This is attributed to the simultaneous ionization of multiple hydroxyl groups, which consequently dampens the preference for C-2 acylation.
Another advanced strategy to achieve regioselective monoesterification of unprotected sucrose involves the synthesis of an intermediary acetal from sucrose and dibutyltin oxide, as outlined in Figure 10. Upon isolation, this intermediate acetal undergoes reaction with fatty acid anhydride in the presence of triethylamine, exclusively yielding 6-O-acyl-sucrose (Vlahov et al., 1997). As an alternative pathway, the isolated acetal can react with fatty acid chloride in a pyridine medium. Under such conditions, the primary products are 6-O-acyl-sucrose and 6,3′-di-O-acyl-sucrose (Wang et al., 2007). Biocatalysis offers another avenue for the regioselective esterification of unprotected sucrose. Specifically, both proteinases and lipases have been employed to this end. Figure 11 presents the distinct regioselectivity profiles of various lipases and proteinases utilized in the synthesis of sucrose esters.
Acylation of other monosaccharides and disaccharides
While sucrose is abundant, it is worth noting that long-chain fatty acid esters can also be synthesized from other disaccharides and certain monosaccharides, exhibiting surfactant properties. Maltose, a disaccharide derived from enzymatic hydrolysis of starch, has been employed for the synthesis of monoesters through the gradual addition of long-chain acid chlorides (such as stearic, palmitic, myristic, and oleic acids) to a solution of maltose in dry pyridine (Allen & Tao, 2002). This procedure results in monoesters primarily at the 6 and 6′ positions of maltose, yielding between 15% and 28%. These compounds are capable of reducing interfacial tension to as low as 34 mN/m at their CMC.
Lactose, or β-D-galactopyranosyl-(1 → 4)-D-glucopyranose, has also been employed to produce long-chain lactose esters. Monoesters have been derived from lactose and maltose through enzymatically catalyzed acylation of the acetals of these disaccharides, as illustrated in Figure 12 (Sarney et al., 1994). Acetalization ensures the selective formation of 6′-O-acyl-lactose and 6′-O-myristoyl-maltose with yields of 61%–74% and 48%, respectively. Garofalakis et al. (2000) have investigated various properties of these amphiphiles, and their characteristic values can be found in Table 8, alongside properties of esters derived from other saccharides, disaccharides, and trisaccharides.
Compound | CMC (mM) | ΓCMC (mN/m) | T (°C) | Ref. |
---|---|---|---|---|
Sucrose monolaurate | 0.45 | 34.5 | 25 | (Zhang et al., 2014) |
Sucrose monocaprate | 0.60 | 33.8 | ||
Sucrose monocaprylate | 0.78 | 32.4 | ||
Maltose monolaurate | 0.32 | 36.0 | ||
Maltose monocaprate | 0.56 | 32.3 | ||
Maltose monocaprylate | 0.66 | 31.2 | ||
Lactose monolaurate | 0.31 | 33.1 | ||
Lactose monocaprate | 0.56 | 31.6 | ||
Lactose monocaprylate | 0.76 | 29.7 | ||
6-O-monocaprynoyl-α,α-trehalose | 4.32 | 29 | 25 | (Schiefelbein et al., 2010) |
6-O-monolauroyl-α,α-trehalose | 0.40 | 39 | 25 | |
6-O-monopalmitoyl-α,α-trehalose | 0.01 | n.d. | 45 | |
Fructose monocaprate | 0.22 | 26 | 20 | (Soultani et al., 2003) |
Fructose monomyristate | 0.04 | 30 | ||
Fructose monoestearate | 0.16 | 31.8 | ||
6-O-lauroyl sucrose | 0.250 | 31.5 | 25 | (Ferrer et al., 2002) |
6-O-palmitoyl sucrose | 0.028 | 35.3 | ||
6′-O-lauroyl sucrose | 0.240 | 34.7 | ||
6′-O-myristoyl maltose | 0.037 | 35.0 | ||
6′-O-palmitoyl maltose | 0.006 | 32.5 | ||
6′-O-stearoyl maltose | 0.032 | 32.5 | ||
6″-O-lauroyl maltotriose | 0.052 | 24.5 | ||
6″-O-myristoyl maltotriose | 0.028 | 36.5 | ||
6″-O-palmitoyl maltotriose | 0.013 | 35.0 | ||
6″-O-stearoyl maltotriose | 0.002 | 35.5 | ||
6-O-lauroyl lucrose | 0.135 | 30.0 | ||
6′-O-lauroyl sucrose | 0.21 | 35.9 | 32 | (Garofalakis et al., 2000) |
6′-O-myristoyl sucrose | 0.021 | 36.0 | ||
6′-O-palmitoyl sucrose | 0.0041 | 35.3 | ||
Sucrose myristic ester | 0.017 | 34.8 | ||
5-O-lauroyl xylose | 0.041 | 28.9 | ||
6′-O-myristoyl actose a | 0.043 | 38.6 | ||
6′-O-palmitoyl lactose a | 0.011 | 39.5 | ||
5-O-myristoyl xylose a | 0.015 | 36.0 | ||
5-O-palmitoyl xylose b | 0.022 | 41.0 | ||
6-O-myristoyl galactose a | 0.020 | 35.5 | ||
6-O-palmitoyl galactose b | 0.15 | 43.0 | ||
6-O-oleyl galactose 18: 1 | 0.023 | 31.0 |
Trehalose, or α-D-glucopyranosyl-(1 → 1)-α-D-glucopyranose, is a naturally occurring disaccharide found in both animals and plants. It consists of two D-glucose units linked through their anomeric carbons by a glycosidic bond. Recent advancements in technology have led to the synthetic production of trehalose from starch through enzymatic processes, making it economically viable for various applications in the food and pharmaceutical industries (Ohtake & Wang, 2011).
Schiefelbein et al. (2010) undertook the preparation of trehalose monoesters, specifically 6-O-monocaprinoyl-α,α-trehalose, 6-O-monolauroyl-α,α-trehalose, and 6-O-monopalmitoyl-α,α-trehalose. These compounds were investigated for their potential as substitutes for polysorbate 20 and 80 in pharmaceutical formulations containing proteins in aqueous solutions. Technical-grade polysorbates may contain trace amounts of residual peroxides from the production process, which can compromise the chemical stability of pharmaceutical formulations. In contrast, 6-O-monocaprinoyl-α,α-trehalose and 6-O-monolauroyl-α,α-trehalose exhibited similar physicochemical properties to polysorbates, albeit with slightly higher hemolytic activity. CMC and surface tension values of these trehalose monoesters are detailed in Table 8.
The synthesis of trehalose monoesters involved an initial protection step, where all hydroxyl groups of trehalose were protected by reacting it with trimethylsilyl chloride and bis(trimethylsilyl)amine in dry pyridine. Subsequently, selective deprotection at positions 6 and 6′ (the only primary hydroxyl groups) was achieved by treating 2,3,4,6,2′,3′,4′,6′-octa-O-(trimethylsilyl)-α,α-trehalose with potassium carbonate in methanol. This selective deprotection is feasible because the hydrolysis rate of secondary silyl ethers is significantly slower than that of primary silyl ethers, as depicted in Figure 13 (Schiefelbein et al., 2010; Smith, 2017).
After carefully neutralizing the partially unprotected trehalose the Steglich procedure was used to achieve hydroxyl acylation at C-6 (Schiefelbein et al., 2010). This process, described in Figure 14, initiates with the formation of an intermediate O-acylisourea, achieved by reacting a carboxylic acid with dicyclohexylcarbodiimide (DCC). Subsequently, the activated carboxylic acid, in the form of O-acylisourea, undergoes substitution with the hydroxyl group, leading to the ester and the by-product dicyclohexylurea (DHU). The final step of this alcohol acylation method is most efficient when 4-dimethylaminopyridine (DMAP) is employed as a catalyst. The selective production of trehalose monoesters hinges on the controlled protection and deprotection of specific hydroxyl groups and involves multiple stages (Schiefelbein et al., 2010). Some attempts to obtain trehalose monoesters enzymatically have been relatively successful (Chen et al., 2005).
Sugar acylation has also been used to synthesize dimeric and trimeric surfactants. Amphiphilic dimers have been crafted through the condensation of acetalized 5-O-lauroyl-xylose (compound I in Figure 15) or tetraacetalized 6′-O-acyl-lactose (compound II in Figure 15) with decanedioyl dichloride or hexanedioyl dichloride. Furthermore, as illustrated in Figure 15, asymmetric dimeric surfactants have been synthesized by the sequential condensation of compounds (I) and (IIb) with diacyl dichloride (Fregapane et al., 1991; Gao et al., 1999).
Another route used by Gao et al. (1999) to synthesize carbohydrate-based dimeric surfactants involves condensing 1,2:3,4-di-O-isopropylidene-α-D-galactopyranose with α-hydroxytetradecanoic acid. Subsequently, the α hydroxyl groups of two ester molecules are esterified with decanedioyl dichloride, resulting in a dimeric amphiphile (Figure 16) after deprotection. In addition to dimeric surfactants, they also synthesized trimeric surfactants employing phloroglucinol as the linker unit.
Etherification of monosaccharides and disaccharides
In terms of their reactivity under Williamson etherification conditions, all the hydroxyl groups of monosaccharides, with the exception of those associated with the anomeric center, can be considered standard secondary and primary alcohols (Levy & Fügedi, 2005). As a result, achieving the selective preparation of monosaccharide monoethers requires the use of protective groups to block certain positions. The isopropylidene ketal protecting group has been employed for the synthesis of D-glucose, D-galactose, and D-mannose monoethers (Figure 17) (Cabaret et al., 1986; Urata et al., 1995). This is achieved by reacting a cis vicinal hydroxyl pair with acetone under standard acetalization conditions. The heightened stability of the isopropylidene ketal group in an alkaline environment facilitates straightforward etherification of the free hydroxyl group. Urata et al. (1995) carried out the etherification of protected aldohexoses with alkyl bromide (8, 12, 16, 18) in aqueous sodium hydroxide and hexane, assisted by a phase transfer agent, achieving yields exceeding 70%. Subsequently, the thermotropic mesomorphic properties of a homologous series of 6-O-alkyl-D-galactopyranose were investigated (Bault et al., 1998).
Scorzza, Godé, Martin, et al., 2002 carried out the etherification of the hydroxyl group at position 3 of glucose using a polypropoxylated n-dodecanol mesylate, resulting in extended surfactants featuring a glucose polar head (Forgiarini et al., 2021; Salager et al., 2019). This multi-step process initiates with the selective protection of the primary hydroxyl group of a polypropylene glycol chain (with n~6) using the triphenylmethyl ether group (Figure 18). Subsequently, Williamson etherification links an n-dodecyl group to the secondary hydroxyl of the polypropylene glycol chain. After the removal of the triphenylmethyl ether group, the primary hydroxyl is transformed into a mesylate group to enable its displacement by the secondary hydroxyl group at C-3 of 1,2:5,6-di-O-isopropylidene-α-D-glucofuranose. In the final step, the compound is deacetalized in an acid medium to yield the end product. In comparison to 3-O-dodecyl-D-glucopyranose, the extended 3-dodecyl ether of D-glucopyranose is significantly more soluble in water, boasting an HLB of 12.6 (4 units higher than simple glucose 3-dodecyl ether). It also exhibits a slightly lower CMC and is equally effective in reducing interfacial tension (Salager et al., 2002; Scorzza, Godé, Goethals, et al., 2002).
Scorzza, Godé, Martin, et al. (2002) followed a similar procedure to synthesize a series of extended 3-O-alkyl ethers of D-glucopyranose extended surfactants, featuring an extension composed of polypropylene glycol and two ethylene glycol units. In this case, two additional steps are introduced: the preparation of 1,5-di-O-mesyl-3-oxapentane from diethylene glycol and its subsequent utilization for etherifying the C-3 hydroxyl group of 1,2:5,6-di-O-isopropylidene-α-D-glucofuranose. The resulting compound is then employed to etherify the terminal hydroxyl group of an alkyl polypropylene glycol chain, ultimately yielding the final product.
The etherification of unprotected sucrose tends to favor the formation of 2-O-alkyl-sucrose (Queneau et al., 2004). The relative reactivity of 2-OH has been attributed to the formation of an intramolecular hydrogen bond between O-2 and 1′-OH, or between O-2 and 3′-OH, even in polar aprotic solvents (Figure 19), which would provide some stability to the 2-oxyanion (Bock & Lemieux, 1982; Davies & Christofides, 1987; Lichtenthaler et al., 1991). Lichtenthaler et al. (1995) selectively obtained 2-O-benzyl sucrose (yield >80%) by reacting non-protected sucrose with benzyl bromide in DMF, in the presence of sodium hydride. In contrast, the use of bulky substrates, such as t-butyl-dimethylsilyl chloride or t-butyl-diphenylsilyl chloride, tends to favor etherification of the primary hydroxyls 6-OH, 6′-OH, and 1′-OH (Karl et al., 1982; Lichtenthaler et al., 1995). Gagnaire et al. (2000, 1999) produced sucrose monoethers through the etherification of sucrose with 1,2-epoxydodecane as a substrate, in an aqueous medium, with the addition of a tertiary amine (N-methylmorpholine, N,N-dimethylbutanamine, 1,4-diazabicyclo[2.2.2]octane) and phase transfer agents.
Sucrose ethers have also been obtained by palladium catalyzed telomerization of butadiene (Figure 19). However, controlling the degree of substitution in this case is more challenging. Desvergnes-Breuil et al. (2001) examined the influence of various reaction parameters on the degree of etherification of sucrose, particularly temperature, the butadiene/sucrose ratio, and the nature of the catalyst. The type of salt used as a catalyst precursor and the metal/ligand ratio were identified as critical parameters.
Synthesis of N-alkyl-aldosylamines and their derivatives
The condensation of a monosaccharide with a primary amine initially yields an N-alkyl-aldosylamine, which can be rearranged under acidic conditions into a more stable 1-amino-1-deoxy-2-ketosse, commonly known as Amadori's product (Figure 20). By avoiding reaction under low pH and high-temperature conditions, it is possible to obtain N-alkyl-aldosylamine free from rearranged products (van Doren et al., 2000). These nitrogenous analogs of alkylglycosides can also be prepared from reducing disaccharides. N-alkyl-lactosylamines and N-alkyl-maltosylamines have been synthesized with yields ranging between 60% and 70%. This synthesis involves stirring a mixture of the disaccharide and primary amine in isopropanol and water at a moderate temperature (~60°C) for 24 h (Bhattacharya & Acharya, 1999a; Costes et al., 1995; Latge et al., 1991). The formation of gel-type aggregates by N-hexadecyl-maltosylamine and N-hexadecyl-lactosylamine, as well as their emulsifying capacity, has been thoroughly evaluated (Bhattacharya & Acharya, 1999b, 1999a; Garg et al., 2010).
N-alkylaldosylamines exhibit instability in aqueous solutions, which poses challenges for their direct application as surfactants. However, acetylation of N-alkylaldosylamines enhances their stability in aqueous solutions while preserving their surfactant properties (El Ghoul et al., 1996). Notably, the acetylation of N-alkyl-lactosylamines preferentially generates the β anomer of N-acetyl, N-alkyl-lactosylamine (Figure 21). This preference arises because the α anomer experiences steric repulsion between the acetyl group and the hydroxyl group at C-2. As a result, this product has the potential to induce the formation of chiral aggregates (Costes et al., 1995).
Research has been conducted on the micellization properties of a series of N-propanoyl, N-alkyl-glucosylamines and N-propanoyl, N-alkyl-lactosylamines with various alkyl chain lengths (8, 10, and 12). Furthermore, CMC values have been determined for some N-acyl derivatives. These CMC values are found to be within the same order of magnitude as typical values for other nonionic surfactants, ranging from 10−2 to 10−4 M (Costes et al., 1995; Pestman et al., 1999). It is worth noting that CMC decreases by a factor of 10 with the addition of every two methyl groups in the carbon chain, a behavior consistent with other nonionic surfactants (Pestman et al., 1999).
The reduction of N-alkylaldosylamines can be achieved by employing sodium borohydride in water or hydrogen in the presence of Pd/C, resulting in the formation of N-alkylamino-1-deoxyalditols (Figure 22). Latge et al. (1991) successfully synthesized a series of N-alkylamino-1-deoxyalditols with varying alkyl chain lengths (7, 8, 9, 10, and 12) and assessed their surface activity. Their findings indicated that these compounds exhibit effective surfactant properties, with CMC values ranging from 32 to 44 mN/m, corresponding to concentrations of 10−2 to 10−4 M. Similar to N-alkylaldosylamines, the acetylation of derivatives of N-alkylamino-1-deoxyalditols is advantageous for enhancing their stability in aqueous solutions (Rico-Lattes & Lattes, 1997). Pestman et al. (1999) synthesized N-propanoyl, N-alkylamino-1-deoxy-D-lactitols and N-propanoyl, N-alkylamino-1-deoxy-D-glucitols with carbon chains of 8, 10, and 12, along with their acetylated counterparts. The CMC values of these N-acyl, N-alkylamino-1-deoxyalditols closely resembled those of the corresponding acetylated/propionylated N-alkylactosylamines and N-alkylglucosylamines.
N-alkylamino-1-deoxy-D-alditols can be further transformed into surfactants with two lipophilic tails by attaching nitrogen to a long-chain acyl group. This transformation is achieved by treating N-alkylamino-1-deoxy-D-alditol with N-acylthiazolidine-2-thione, a nitrogen-selective acylating agent synthesized through the condensation of 2-mercaptothiazoline with acid chloride. This method was successfully employed to produce a series of N-acyl, N-alkylamino-1-deoxy-lactitols and N-acyl, N-alkylamino-1-deoxy-glucitols, yielding between 20% and 60% (Figure 23). These surfactants possess sufficient hydrophobicity to be soluble in nonpolar organic solvents (Rico-Lattes & Lattes, 1997).
The N-alkylamino-1-deoxyalditol structure serves as a foundational framework for the synthesis of gemini surfactants due to the nitrogen atom's coordination properties. An extensive library of gemini surfactants derived from various saccharides has been created by incorporating lipophilic spacers and tails of varying lengths and compositions (Johnsson & Engberts, 2004). The synthesis methodology involves the initial preparation of a bis(1-amino-1-deoxy-D-aldityl)alkane bolaform structure, followed by the acylation or alkylation of its nitrogen atoms (Figure 24). The aggregation behavior of these gemini surfactants in aqueous solutions has been extensively investigated (Johnsson & Engberts, 2004).
The synthesis of the gemini surfactant series, bis(N-tetradecanoyl-1-amino-1-deoxy-D-glucityl)alkane, entails several steps. Initially, an α, ω-diaminoalkane (6, 8, or 10) is condensed with glucose in a reducing medium (H2 in the presence of Pd/C) (Figure 21). Subsequently, the nitrogen atoms of the resulting product are acylated with tetradecanoic anhydride in ethanol (Johnsson & Engberts, 2004). Pestman et al. (1997) conducted an evaluation of the lyotropic behavior of these compounds, confirming their tendency to form gel-like aggregates. Furthermore, compounds incorporating 12 and 16 alkanoyl groups, along with various spacers, have demonstrated remarkable capabilities for solubilizing organic liquids (van Doren et al., 2000).
Subsequently, Fielden et al. (2001) synthesized a series of bis(N-alkyl-1-amino-1-deoxy-D-glucityl)alkanes. In this context, the foundational bolaamphiphile structure, bis(1-amino-1-deoxy-D-glucityl)alkane, underwent reductive alkylation with saturated aldehydes (12, 14, 16, 18) in glacial acetic acid and hydrogen, catalyzed by Pd/C (Figure 25). Notably, alkylation with cis-9-octadecenal, while retaining the carbon–carbon double bond, was achieved using sodium cyanoborohydride in methanol as the reducing medium. Furthermore, the introduction of 1,8-diamino-3,6-dioxaoctane spacers and mannose as part of the synthesis led to the creation of diverse structures (Johnsson et al., 2003).
Following the same synthesis protocols as outlined above, a range of gemini surfactants were obtained, including bis(N-octadecen-9-oyl-1-amino-1-deoxy-D-glucityl)-3,6-dioxaoctane, bis(N-octadecen-9-yl-1-amino-1-deoxy-D-glucityl)-3,6-dioxaoctane, bis(N-octadecen-9-yl-1-amino-1-deoxy-D-mannityl)-3,6-dioxaoctane, and bis(N-octadecen-9-yl-1-amino-1-deoxy-D-mannityl)hexane (Figure 26) (Johnsson et al., 2003) further expanded this collection of gemini surfactants, including in their structures other sugars (galactose, lactose, arabinose, talose, melibiose, erythrose) into their structures, along with a variety of spacers.
Schuur et al. (2004) successfully synthesized asymmetric bolaamphiphiles through a two-stage process involving reductive amination of sugars (Figure 27). In the initial stage, the aldose/1,6-diaminohexane ratio was meticulously adjusted to 1:6.6, yielding a 1-(1-deoxy-D-alditol-1-ylamino)-6-aminohexane intermediate. Subsequently, this intermediate underwent reductive alkylation with a different aldose in the subsequent step. These surfactants possess significant potential as starting materials for the creation of novel sugar-based gemini surfactants.
Warwel et al. (2004) synthesized gemini surfactants from N-alkylamino-1-deoxy-glucitols using a different strategy based on linking two of these molecules through their nitrogen atoms with a spacer provided by di-epoxides of α,ω-olefins (8, 9, 10, and 14) and glycidic ethers of diols. This occurs by opening the oxirane ring in dry methanol, at 70°C. Among all the compounds prepared by Warwel et al. (2004), those with hydrophilic spacers (Figure 28) showed greater surfactant activity.
RHAMNOLIPIDS DERIVED FROM FERMENTATION
Rhamnolipids constitute a class of nonionic glycolipid biosurfactants predominantly synthesized from sugars, starches and/or oil sources (mainly vegetal). The sugars are fermented by utilizing the bacterium Pseudomonas aeruginosa, alongside other organisms. The molecular structure of rhamnolipids includes a rhamnose moiety serving as the glycosyl head group and a 3-(hydroxyalkanoyloxy)alkanoic acid (HAA) fatty acid tail, such as 3-hydroxydecanoic acid. The biosynthesis of rhamnolipids is governed by specific genes and metabolic pathways, primarily directed by P. aeruginosa, the most proficient producer of rhamnolipids (Anaukwu et al., 2020; Soberón-chávez et al., 2021). However, the pathogenic nature of this bacterium poses challenges for large-scale production. The biosynthetic pathway of rhamnolipids involves the synthesis of 3-(3-hydroxyalkanoyloxy) alkanoic acids by the enzyme RhlA, achieved through esterification of two 3-hydroxyacyl molecules bound to an acyl carrier protein from the fatty acid synthesis (Abdel-Mawgoud et al., 2011; Guzmán et al., 2023).
In recent years, the growing emphasis on sustainable processes and environmentally friendly products has generated interest in biosurfactant production. While biosurfactants offer a broad array of applications and numerous advantages, the principal technological hurdle to their industrial utilization remains the low yields and elevated production costs, particularly in relation to downstream processes. Glycolipid microbial biosurfactants derived from bacteria or yeast, such as rhamnolipids, sophorolipids, and lipopeptides, are major categories within the field of biosurfactants. Biosurfactant production can be achieved using either soluble carbohydrates or hydrophobic, insoluble substrates. To enhance sustainability, researchers have explored the use of waste materials to reduce costs and disposal requirements (Begum et al., 2023). These methods encompass both submerged and solid media and consider a variety of microorganisms, including bacteria and fungi (Esposito et al., 2023; Guzmán et al., 2023).
Rhamnolipids are a type of secondary metabolite synthesized by microorganisms. They consist of rhamnose components (forming the glycone part) and lipid portions (constituting the aglycone part) linked via an O-glycosidic bond (Abdel-Mawgoud et al., 2010). These compounds are particularly popular among biosurfactants due to their remarkable physicochemical attributes, making them excellent natural emulsifiers and wetting agents (Varjani et al., 2021). The primary producer of rhamnolipids is the pathogenic microorganism P. aeruginosa. However, for specific applications, nonpathogenic microorganisms, like Burkholderia spp., and even genetically modified microbes (Chong & Li, 2017), have been employed in rhamnolipid production. Nonetheless, the yields achieved with these alternative approaches were notably lower compared to those obtained with the well-established P. aeruginosa strains.
Over the past two decades, there have been documented instances in the literature of producing rhamnolipids using food and agro-industrial waste originating from biobased sources (Table 9). Some examples include olive oil remnants, unrefined glycerol, winery byproducts, residual cooking oil, cassava waste, lignocellulose residues, and even oil remnants sourced from petroleum.
Feedstock | Microorganism | Fermentation conditions | Biosurfactant concentration | Productivity | Ref. |
---|---|---|---|---|---|
Distillery wastewater (20% v/v) | Pseudomonas aeruginosa SRRBL1 | Batch, 37°C, 120 rpm, 120 h | 2.9 g/L | 0.040 g/L h | (Ratna & Kumar, 2022) |
Crude glycerol (60 g L − 1) | Pseudomonas aeruginosa PrhlAB | Batch, 37°C, 200 rpm, pH 6.8 (initial), 168 h | 2.9 g/L | 0.024 g/L h | (Zhao et al., 2019) |
Glycerol | Pseudomonas putida KT24C1 | 13 h at 30°C | 1.5 g/L | 0.16 g/L h | (Tiso et al., 2020) |
Glycerol | Pseudomonas putida KT2440 | 18 h at 30°C | 1.5 g/L | 0.35 g/L | (Wittgens et al., 2011) |
Waste glycerol | Pseudomonas aeruginosa RS6 | 0.2 M sodium nitrate when incubated at 35°C and pH 6.5 | — | 2.7 g/L | (Baskaran et al., 2021) |
Hydrolyzed pineapple skin, corncob, and glycerol (5%) | Planomicrobium okeanokoites IITR52 | 72 h batch at 30°C and 150 rpm | — | 1500 mg/L (glycerol), 568 mg/L (corncob), 304 mg/L (pineapple) | (Gaur et al., 2022) |
Petroleum oil waste, (2% oil) | Pseudomonas aeruginosa | Batch (bioreactor), 10 L (working volume: 5 L), 37°C, pH 7 (initial), 0.5 vvm, 360 h | 2.7 g/L | 0.019 g/L h | (Mostafa et al., 2019) |
Cooking oil fume condensates | Pseudomonas aeruginosa AB93066 |
Batch, 7-L bioreactor, pH 7.0 (initial), 110 h | 12.3 g/L, 45.0 mg/L | 0.16 g/L h | (Wu, Zhang, et al., 2019) |
Frying oil | Pseudomonas aeruginosa | 96 h at 30°C and 150 rpm | — | 6.2 g/L | (Pathania & Jana, 2020) |
Waste cooking oil | Pseudomonas aeruginosa M4 | 216 h of incubation at 35°C and 180 rpm | — | 1119.9 mg/L | (Shi et al., 2021) |
Industrial production of rhamnolipids is gaining momentum, with companies like NatSurFact, AGAE Technologies Ltd., Rhamnolipid, Inc., GlycoSurf, TensioGreen, and Jeneil Biosurfactant, and Evonik Industries actively involved in the production (Evonik, 2023; Evonik Corporation, 2022). This includes a new rhamnolipids plant in Slovakia, constituting a significant step towards commercial-scale production (Jiang et al., 2020). Despite these developments, a significant challenge hinders broader market penetration: high production cost, partly due to low yields and small-scale operations. However, the market's growth is driven by the increasing demand for rhamnolipids in various sectors, including oil & gas, agriculture, food and beverage, cosmetics, home care, and pharmaceuticals (Jiang et al., 2020; Soberón-chávez et al., 2021), (Tiso et al., 2020).
Biosynthesis, chemical structure, and production of rhamnolipids
Within microbial surfactants, five primary categories are recognized: glycolipids, lipopeptides, phospholipids, neutral lipids, and polymeric microbial biosurfactants. Among these, glycolipids, including sophorolipids (SLs), rhamnolipids (RLs), and mannosylerythritol lipids (MELs), are the most commonly commercialized entities (Baccile et al., 2021). Rhamnolipids are glycolipid biosurfactants produced by various microorganisms. Biosynthesis is controlled by specific genes and metabolic pathways, primarily regulated by P. aeruginosa, the most efficient source, despite the pathogenic challenges in large-scale production. The molecular structure of rhamnolipids consist of a rhamnose moiety as the glycosyl head group and a 3-(hydroxyalkanoyloxy)alkanoic acid (HAA) fatty acid tail, such as 3-hydroxydecanoic acid (Guzmán et al., 2023; Jahan et al., 2020).
Figure 29 shows the metabolic pathway for rhamnolipids biosynthesis in P. aeruginosa and highlights the enzymatic steps involved in converting the central metabolite D-glucose-1-phosphate into the rhamnose and fatty acid components of RLs, and eventually into the final RL molecules. The complexity of this pathway reflects the challenges in scaling up RL production due to the tight regulation of the involved genes, which are controlled by quorum sensing mechanisms within the bacterium (Bahia et al., 2018).
- RmlA (Glucose-1-phosphate thymidylyltransferase): This enzyme initiates the pathway by converting D-glucose-1-phosphate to dTDP-D-glucose.
- RmlB (dTDP-glucose 4,6-dehydratase): The second step, performed by RmlB, involves the dehydration of dTDP-D-glucose to dTDP-4-dehydro-6-deoxy-D-glucose.
- RmlC (dTDP-4-dehydrorhamnose 3,5-epimerase): RmlC then acts on the dehydrated sugar to form dTDP-4-dehydro-rhamnose via an epimerization reaction.
- RmlD (dTDP-4-dehydrorhamnose reductase): Finally, RmlD reduces the product to produce dTDP-L-rhamnose.
Parallel to the rhamnose biosynthesis, fatty acid synthesis occurs through a separate pathway where fatty acids are constructed “de novo” from acyl carrier protein (ACP) linked intermediates, which are eventually converted into β-hydroxyacyl-ACP intermediates. The enzyme RhlA takes over to convert β-hydroxyacyl-ACP intermediates into β-hydroxyalkanoyl-β-hydroxyalkanoic acids (HAAs). This is a crucial step as HAAs are the fatty acid components of rhamnolipids. Following the synthesis of HAAs and dTDP-L-rhamnose, two rhamnosyltransferases, RhlB and RhlC, are responsible for the coupling of these molecules to produce mono-rhamnolipids and di-rhamnolipids, respectively. The enzyme RhlB catalyzes the transfer of rhamnose to HAAs to form mono-rhamnolipids. RhlC further adds another rhamnose unit to mono-rhamnolipids to produce di-rhamnolipids (Bahia et al., 2018).
Rhamnolipids, an eco-friendly alternative to petrochemical-derived compounds, are being researched for increased production from renewable sources like sugar to support industrial use (Maťátková et al., 2022). The synthesis of rhamnolipids from glucose (Table 10) involves converting it into dTDP-l-rhamnose, a sugar moiety in the biosurfactant structure. Another key component, Acyl-CoA, can also be produced from glucose and used in the synthesis process (Tan & Li, 2018). Utilizing glucose or carbohydrates as feedstocks for rhamnolipid production aligns with the trend of using renewable resources in biosurfactant synthesis (Chong & Li, 2017; Wittgens et al., 2011).
Stage | Specific challenge | Strategy | Impact on value chain |
---|---|---|---|
Fermentation | High substrate cost | Use of low-cost, renewable substrates like agro-industrial waste | Reduces raw material costs significantly, making production more sustainable and cost-effective |
Fermentation | Foam formation causing operational issues | Implementation of mechanical foam breakers or addition of antifoam agents | Prevents loss of yield and contamination, improving operational efficiency and product consistency |
Fermentation | Oxygen transfer limitation in high-density cultures | Adoption of fed-batch or continuous fermentation strategies with optimal aeration | Enhances cell growth and product yield, reducing per-unit production costs |
Strain Selection | Pathogenicity of Pseudomonas aeruginosa | Genetic modification to attenuate pathogenicity or use of non-pathogenic strains | Ensures safety and regulatory compliance, broadening market acceptance and application range |
Strain Selection | Low productivity of non-pathogenic strains | Metabolic engineering to enhance biosurfactant synthesis pathways | Increases yield from non-pathogenic strains, making them viable alternatives to P. aeruginosa |
Product Purification | Complexity and cost of downstream processing | Development of simplified purification processes, like foam fractionation or in situ extraction | Lowers purification costs and minimizes product loss, significantly reducing overall production expenses |
Product Purification | Degradation or modification of product during purification | Optimization of purification conditions and use of mild extraction techniques | Preserves the functional integrity and quality of rhamnolipids, ensuring high product value |
Theoretical rhamnolipids yield from glucose has been assessed in various studies, though it may vary based on the specific bacterial strain and fermentations conditions. In one study involving engineered Pseudomonas putida KT2440, the yield from glucose was notably lower than the theoretical limit of 0.72 mol/mol (rhamnolipid/glucose) (Wittgens et al., 2011). The actual yield can be influenced by various factors, including the type and concentration of carbon source. For instance, in one study, a combination of glucose and fatty acids resulted in a yield of 2.1 g/L of rhamnolipids (Tan & Li, 2018). In another study, the maximum yield of 0.784 g/L was achieved when glucose served as the sole carbon source (Dobler et al., 2016). Another study reported a yield of 0.1 kg of product per kg of glucose in a 150-L fermentation system (Tiso et al., 2020). By adjusting the C/N ratio, carbon source type, and concentration, the yield improved from 2.2 to 13.5 g/L with productivity ranging from 11.6 to 45.3 mg/L/h.
In Figure 30, the process diagram for rhamnolipid production is shown (Evonik, 2020). This process primarily involves fermentation and purification, largely due to the complexity of the substances formed and the required purity of the products. The purity level notably depends on the feedstock's characteristics; agricultural residues are more complex than glucose but less costly. Table 10 provides an in-depth analysis of the challenges at each stage and strategies for overcoming them, including how these strategies impact the value chain (Chong & Li, 2017).
Interfacial properties of rhamnolipids
Pioneering work on the self-assembly of biosurfactants started in the late 1980s and early 1990s, with a particular focus on RLs, surfactins, MELs, SLs, and CLs. This timeline contrasts with the development of synthetic surfactants, which had reached maturity in the 1970s (Yu et al., 2008). This delay can be largely attributed to several factors, primarily the structural complexity and multi-functionality of biosurfactants in comparison to traditional surfactants.
Biosurfactants often exhibit bolaform structures with multiple ionizable groups, primarily carboxylic acids. This structural complexity imparts unique properties but also makes understanding their self-ssembly mechanisms more challenging. External factors, including pH, ionic strength, and temperature, exert a significant influence on their phase behavior (Jahan et al., 2020). The multifunctionality of biosurfactants leads to weak interactions, such as hydrogen bonding and π–π stacking, coexisting with stronger interactions like ionic, van der Waals, and entropic forces during self-assembly. As a result, conventional theories of surfactant self-assembly, such as the packing parameter theory, often prove inadequate for biosurfactants (Guzman et al., 2023). These challenges render the prediction of their aggregation morphology and phase behavior in aqueous environments exceedingly difficult.
Rhamnolipids are generally stable under neutral or mildly basic pH conditions (Baccile et al., 2021). In neutral pH conditions, rhamnolipids (RLs) exhibit similar micellar structures, predominantly forming ellipsoidal micelles. However, upon ionization, RLs undergo a morphological transition, resulting in more spherical micelle shapes characterized by thinner equatorial shells compared to their axial shells. The CMC of RL exhibits notable variability based on the substrates available to the producing microorganism. The molecular weight (Mw) values range from 277 to 647 g/mol highlighting significant structural differences contributing to CMC variability. Interestingly, RL predominantly display CMC values in the μM–mM range (Esposito et al., 2023).
Rhamnolipids exhibit excellent surfactant and emulsion properties, significantly reducing the surface tension of water to a range between 25 and 40 mN/m at the CMC (Dabaghi et al., 2023; Esposito et al., 2023; Leitermann et al., 2010) and decrease the interfacial tension in water/oil systems from 43 to below 1 mN/m (Kong et al., 2021). However, the available sources do not explicitly provide information about their emulsion and foaming capacity. Therefore, additional literature research or experimental data may be necessary to establish these characteristics.
As may be expected with a variance of one rhamnose unit, mono-RLs generally exhibit higher surface activity compared to di-RLs. Comparing mono- and di-RLs with the same hydrophilic groups will show a lower CMC for the mono-RL, along with a greater reduction in surface tension. The more efficient packing at the interface this variance demonstrates could be expected from the more compact head group. On the other hand, the mono-RL also demonstrates a lower solubility (Wu, Lai, et al., 2019), resulting in a higher effective diffusion coefficient. In one study, the CMC of a mono-RL was found to be around 26 mg/L, while the corresponding di-RL had a CMC of around 30 mg/L (Guzmán et al., 2023). This property is crucial for detergency and emulsification applications, such as cleaning agents and oil spill remediation. There are mixed reports on the emulsifying performance of mono and di-RLs.
Clearly, the solubility of RLs in water is influenced by the number of rhamnose units and the length of the fatty acid chains. On the other hand, it is the form in which sophorolipids are present (acid or lactone) that significantly influences their application and effectiveness. There are some trends, presented in Table 10, although they are not general and require further research. It should be noted that rhamnolipid and sophorolipid biosurfactants produced by microorganisms growing on different substrates can have different molecular structures and compositions (Nguyen & Sabatini, 2011).
As mentioned, the unusual structure of biosurfactants can also result in unusual aggregation properties. Under conditions of higher temperature (>45°C) or higher pH (~10) large micellar structures (350–450 nm) are formed with indications that these structures may actually be vesicles. Apparently, these conditions adjust the repulsive interactions among the charged hydrophilic groups, resulting in a rapid reduction in micellar curvature (Wu, Lai, et al., 2019). Similarly, lactonic SLs (closed ring formation) will produce microvesicles at low concentrations (0.2–3.0 mM) and move to disordered aggregates at higher concentrations. Acid SLs (linear formation) are more consistent in producing typical globular micelles. The most typical composition of SLs are a mix of lactone and acid structures, and it is in this case that small non-associating micelles are present at lower concentrations (<30 mM), but take on more elongated structures with higher concentrations (Penfold et al., 2011).
The greatest practical challenge for formulators working with biosurfactants is understanding how the properties of these materials can be best harnessed to provide effective industrial or consumer products. Formulators typically work within frameworks established by properties of their tools. In the case of surfactants, those properties are tied to somewhat familiar structures and a skilled formulator recognizes the benefits and drawbacks of changes in the structure. Biosurfactants are rather different in their properties and deviate significantly from common structural motifs in industry.
As an example, commonly used surfactants often have a single hydrophobic and a single hydrophilic structure and the hydrophilic structure usually represents a single type of functional group. SLs and RLs, on the other hand, have two types of hydrophiles in the nonionic sugar group(s) and the potentially charged carboxylate. As a result, there is now substantial interest in presenting successful formulations for particular applications. One theme that has arisen from much of this work has been the determination that biosurfactants tend to work best when blended, either as types or as variations upon a type (Benhur et al., 2020; Freitas et al., 2016; Nguyen et al., 2010).
INSIGHTS INTO HLD PARAMETERS OF SUGAR HEAD SURFACTANTS
The SCP in the HLDN equation is related to the system's phase behavior, as a term in the difference in affinity of the surfactant with oil and water (Salager, 2021; Salager et al., 2020). However, it's important to note that the SCP, along with other equivalent parameters like Nmin, PACN, σ, β, and CC, is not an inherent property of the surfactant alone. Instead, it depends on other variables such as temperature and salinity (Salager et al., 2020; Salager, Graciaa, & Marquez, 2022). Therefore, the use of a so-called “characteristic curvature” (Acosta et al., 2008; Leng & Acosta, 2023) term for this parameter could lead to confusion (Salager et al., 2020; Salager, Graciaa, & Marquez, 2022). The SCP of biobased surfactants is intrinsically linked to its molecular structure, particularly the sugar head and the hydrocarbon chain length. To date there are only a handful of studies where the SCP (or an equivalent parameter) of sugar head surfactants has been assessed. In Figure 31 the SCP of a series of sugar head surfactants studied at the University of Lille are presented (Lemahieu et al., 2020). Surfactants with higher SCP values, such as those derived from hexahydrofarnesol (HHF) and sugars HHFA (A: arabinose) and HHFX (X: xylose), exhibit a strong lipophilic nature due to their affinity for longer hydrocarbon chains. This characteristic makes them highly suitable for oil-rich environments, where their molecular structure allows for effective interaction with long-chain hydrocarbons. Conversely, a surfactant with a maltose sugar moiety such as HHFMalt (Malt: Maltose), with a negative SCP value, present more hydrophilicity. This unique behavior is attributed to the specific microstructural attributes of its sugar head, which facilitate stronger hydrogen bonding and polar interactions, making it effective in environments where typical surfactants may exhibit limited performance.
The sugar head of a surfactant significantly influences its hydrophilicity. Sugar molecules, with their multiple hydroxyl groups, can form extensive hydrogen bonding networks with water molecules, enhancing the surfactant's solubility in aqueous phases (Lemahieu et al., 2020). This attribute is particularly important in surfactants with more complex sugar heads, such as disaccharides, which offer additional sites for hydrogen bonding. The hydrocarbon tail, on the other hand, dictates the surfactant's lipophilicity. Longer hydrocarbon chains, as seen in the nCm-β-Malt series, increase the surfactant's affinity for oil phases. This trend of increasing lipophilicity with elongated hydrocarbon tails is a direct consequence of the enhanced van der Waals interactions between the surfactant and hydrocarbon oils. Surfactants with moderate SCP values, such as nC10 and nC12-β-Glu, HHFG and HHFMan, exhibit balanced amphiphilic behavior. Their molecular structures are such that they can interact favorably with both water and oil phases, making them versatile for a range of applications, from emulsification to detergency (Lemahieu et al., 2020). The shaded region in Figure 31 map presents the range of SCP values that can be attained with different mixtures of these surfactants from negative to positive values.
RECENT ADVANCES IN THE FIELD OF CARBOHYDRATE-DERIVED SURFACTANT SYNTHESIS AND FUTURE PERSPECTIVES
Sugar-based surfactants, derived from simple sugars such as monosaccharides and disaccharides, are gaining popularity due to their renewable nature and functional attributes (Estrine et al., 2019; Hayes et al., 2019; Ortiz et al., 2022). To enhance the efficacy of these surfactants, future research could explore specific chemical processes like the acylation of sucrose and the etherification of monosaccharides. These surfactants are not only environmentally friendly but also at a lower cost when compared to other types of biobased alternatives, particularly when agricultural waste is utilized as the raw material (Vučurović et al., 2024). Furthermore, the incorporation of waste materials would be beneficial to the circular economy not only because of the reduction of waste, but also through the development of value from previously valueless (or negative value via disposal) materials. In Figure 32, a tree diagram representing the current perspective of carbohydrate-derived biobased surfactants and the main chemical companies producing biobased surfactants and biosurfactants are presented.
Alkyl polyglycosides
The most industrially mature biobased surfactants are the APG (Estrine et al., 2019). Rohm & Haas began supplying APG in commercial quantities in the late 1970's. Since that time a sufficient number of other chemical manufacturers have begun to produce and supply APG that such surfactants could reasonably be considered a commodity products (Hill et al., 2008). Mono- and polypentoses, are relatively new to the market with French chemical manufacturer, Wheatoleo, being the most visible. Dow also produces at least one product based on this technology as well, and the focus of these surfactants is cleaning, personal care, and the agricultural market (DOW, 2023). In addition to the pentose hydrophile, the fatty acid hydrophobe is primarily based on coconut oil.
Microbial biosurfactants
There are many options for the industrial utilization of biobased and biosurfactants, as detailed above. Options for sustainably incorporating raw materials into surfactant molecules include the use of agricultural products and biproducts. Although this primarily can be done to generate the hydrophilic moiety, it decreases the carbon footprint and in some cases performance (Stubbs et al., 2022). The significant strides in microbial biosurfactant production, including rhamnolipids (RLs), sophorolipids (SLs), and mannosylerythritol lipids (MELs), across various industrial sectors like detergents and cosmetics and innovations, such as metabolic engineering, have enhanced the production efficiency of biosurfactants, making them more viable for industrial applications. Despite their eco-friendly profile and diverse applications, challenges remain, particularly in achieving cost-effective production and formulation consistency due to their complex mixtures compared to synthetic surfactants. Strategies to overcome these limitations include exploring cheaper raw materials and advancing fermentation processes. The focus on RLs in research is due to their unique structural properties influencing their functionality, although there's a call for balanced attention to SLs and MELs, considering their potential in various applications. The effort to chemically modify glycolipid biosurfactants aims to tailor their properties for specific industrial needs, marking a promising direction for future development (Guzmán et al., 2023; Rahman et al., 2024).
Polyglycosides have been rapidly adopted in the market and rhamnolipids, notable for their surface properties and clear path to industrial-level production may follow the same path (Chin et al., 2023; Guzmán et al., 2023; Lokesh et al., 2017). Adjusting the mix between the lactose and acid forms in sophorolipids, or ratio of mono- and di-rhamnolipids, can modify their formulation and surface properties, such as foamability and emulsification capability, suggesting a focused potential for rhamnolipids and possibly other biosurfactants in various applications (Table 11) (Zhao et al., 2019).
Property | Sophorolipids (acid form) | Sophorolipids (lactone form) | Mono-rhamnolipids | Di-rhamnolipids |
---|---|---|---|---|
Chemical Structure | Open chain with carboxylic acid group | Closed ring lactone structure | Single rhamnose sugar unit attached | Two rhamnose sugar units attached |
Solubility | Higher due to the presence of the carboxylic acid group, enhancing water solubility | Low due to the closed ring minimizing hydrophilic interactions | Generally low due to the single sugar moiety | Higher than monorhamnolipids due to increased hydrophilicity from the additional rhamnose unit |
Surface Activity | High, effective in reducing surface and interfacial tension | Strong adsorption to the surface driven by low solubility; offers unique interactions due to structural conformation | High, effective at forming micelles and even and reducing surface tension | High but may require higher concentrations to achieve similar surface tension reduction as monorhamnolipids |
Potential Applications | Detergents, cosmetics, bioremediation due to high solubility and surface activity | May offer specific interactions in pharmaceutical applications due to stable structure | Wide range including bioremediation, pharmaceuticals, and cosmetics | Similar to monorhamnolipids but with potential for enhanced activity in certain applications due to structural differences |
Recent industry trends in raw materials and products
The renewable options for the hydrophobic portion of surfactants (mainly a simple hydrocarbon chain) are generally limited to fats and oils. Life cycle assessment studies comparing hydrophiles derived from agriculturally sourced materials to those derived from petrochemical sources indicate a greater carbon footprint results from the production, refining, and functionalization of agricultural materials (Shah et al., 2016). However, efforts are ongoing to lower the carbon footprint for sources such as palm and palm kernel oil. Non-profit organizations such as the Roundtable on Sustainable Palm Oil (RSPO) are allowing the implementation of certifications to prove the credibility of such efforts, which aim to introduce non-petroleum hydrophobes into industrial production (Stubbs et al., 2022).
Regardless of source, increased use and research on biobased surfactants and biosurfactants in commercial products reflect the influence of current sustainability consumer trends and initiatives to minimize the carbon footprint. This megatrend encourages major producers of consumer goods, such as cosmetics, household items, and cleaning detergents, to incorporate surfactants from natural sources, aiming to declare their product contents as predominantly natural or biobased (Pätäri et al., 2016). Based on investment and expected consumption, biosurfactants seem to be recognized as the most viable option for carbohydrate-based surfactants. The fermentation process is carried out by a genetically modified Pseudomonas putida bacteria to avoid the potential issues of the pathogenic organism, P. aeruginosa and is expected to produce double-digit metric kilotons per year (Bettenhausen, 2024).
Industry, in general, shows confidence in the future of these biosurfactants, as evidenced by the new plant from Evonik in Slovakia, which indirectly shows a trend on the part of consumer companies to bring biosurfactants to the market. As it was presented in the previous section, a partnership with Unilever motivated this significant scale-up project and the resolve to launch a hand dishwashing product in Chile utilizing Evonik's rhamnolipids (Unilever, 2019). The effort in Chile was followed by bringing the same technology to Vietnam (Unilever, 2022).
Finally, the potential of artificial intelligence (AI), especially generative AI, in discovering new molecules and identifying novel chemical or biochemical pathways for surfactant synthesis opens new opportunities. The synergy between AI and surfactant research enables to identify new surfactant molecules and refine production methods, thus advancing the field and overcoming the current challenges of biobased surfactants and biosurfactants, as will be discussed in the following section.
ARTIFICIAL INTELLIGENCE AND GENERATIVE AI TO DEVELOP ADVANCED BIOBASED MOLECULES
Recent developments and massification of AI across academia and industry have opened an opportunity of funding research supported by AI in the chemical industry. The integration of machine learning, deep learning, and generative artificial intelligence technologies represents a step forward in the development of more efficient, sustainable, and performance-enhancing surfactant development. These tools have the potential to advance the way surfactant structures are discovered, designed, and optimized, offering opportunities to tailor surfactant molecules for specific applications, including pharmaceutical and biomedical applications (Bannigan et al., 2021; Vamathevan et al., 2019), personal care products, food additives, detergents and cleaners, among others (Boiko et al., 2023; Lei et al., 2023; Yang et al., 2023).
Machine learning
The utilization of the Simplified Molecular Input Line Entry System (SMILES) notation (Weininger, 1988) has been utilized in the development of new molecules using machine learning models, providing a concise and text-based representation of chemical structures. This enables the efficient encoding of molecules for computational analysis, facilitating the rapid and accurate input of molecular data into machine learning algorithms. However, the SMILES notation may not adequately capture stereochemistry or dynamic aspects of molecules, such as tautomeric forms, which can lead to ambiguities in molecule representation, requiring complementary approaches to fully understand and design novel molecules with desired properties (Thacker et al., 2023).
Machine learning (ML) has shown great promise in predicting the physicochemical properties of surfactants. ML combined with molecular simulations can predict different physicochemical properties such as the dispersion and the viscosity of the self-assembled surfactants solutions, CMC, surface tension, interfacial tension, among others (Brozos et al., 2024; Patel et al., 2020; Rashidi-Khaniabadi et al., 2023; Thacker et al., 2023). This predictive capability is crucial and plays an important role in the discovery process of surfactants with desired behavior in different applications, enabling the fine-tuning of some properties, such as solubility and micelle formation, without extensive trial-and-error experimentation.
Machine learning offers a pathway to overcome challenges such as complexity of surfactant systems and the need for advanced computational models to predict properties. This could provide insights into the non-linear relationship between molecular structure and macroscopic properties (Thacker et al., 2023). Moreover, ML algorithms could predict, within some boundaries, the physical properties of surfactant solutions, such as dispersion and viscosity, from their molecular structures, facilitating the design of surfactants with tailored physical characteristics for specific applications (Inokuchi et al., 2018). By understanding these relationships, ML might facilitate the design of surfactants that exhibit optimal performance in different environments, such as high salinity or extreme temperatures. Deep learning models (Wong et al., 2024) have also been proposed as a tool for discovering novel biochemical molecules.
Machine learning models implemented to determine surfactant properties
Artificial Intelligence, including Machine Learning and Neural Networks, has been used recently as an approach for understanding and predicting surfactant properties. The properties that have been studied include the EACN, Surface Tension, Interfacial Tension (IFT), and predictions related to Critical Micelle Concentration (CMC) and Surface Excess Concentration (Γm) (Table 12).
Property measured | Method used | Precision | Advantages | Limitations | Ref. |
---|---|---|---|---|---|
Equivalent Alkane Carbon Number (EACN) | Graph Machines (GM) & Neural Networks (NN) based on σ-moment descriptors | GM: RMSE = 0.5, NN: RMSE = 0.7 | Quick prediction from molecular structure; GM utilizes SMILES codes, NN uses σ-moments | GM restricted to molecules with C, H, or O; NN needs pre-calculation of σ-moments with COSMO-RS | (Delforce et al., 2022) |
Surface Tension | Machine Learning hybrid method, fitting data to Szyszkowski equation | R2 = 0.69–0.87 for different models | Good correlation with experimental data; provision of open-source code; integrates data-based and knowledge-based methods | Limited predictive capacity due to complex molecular interactions and dataset size | (Seddon et al., 2022) |
Interfacial Tension | Tree-based Machine Learning algorithms (DT, ET, GBRT) | R2 = 0.9939 | High accuracy; accommodates various input parameters; GBRT shows optimal performance | Depends on extensive experimental data for model training; sensitive to data quality | (Rashidi-Khaniabadi et al., 2023) |
Phase Behavior of Surfactants | Various Machine Learning classifiers | Variable; ≈50% Recall for de novo prediction | Effective for “gap filling” missing data; some classifiers are suitable for de novo predictions | Limitations due to data bias and insufficient chemical space information | (Thacker et al., 2023) |
CMC and Surface Excess Concentration (Γm) | Graph Neural Networks (GNNs) | High accuracy for CMC; improved Γm prediction with multi-task GNNs | Predicts multiple properties simultaneously; enhanced accuracy with ensemble learning | Limited applicability to surfactants similar to those in the training set; stereochemistry not considered | (Brozos et al., 2024) |
The AI methods that have been used include graph machines (GM) and neural networks (NNs) based on σ-moment descriptors to tree-based Machine Learning algorithms such as decision tree, extra trees, gradient boosted regression trees, and graph neural networks (GNNs). These methods are applied to a variety of tasks, including rapid prediction from molecular structure, fitting experimental data to established equations like the Szyszkowski equation for surface tension, and developing comprehensive predictive models for IFT and phase behavior.
Precision metrics are provided in Table 12 to indicate the accuracy of the AI models. Methods include an efficient prediction capability, the ability to correlate closely with experimental data, and the capacity to handle complex molecular interactions and a broad range of input parameters. Although the main restriction is on the types of molecules that can be analyzed, there is also a need for extensive experimental data for model training, and high-sensitivity models related to the quality and size of the dataset. This means that extensive experimental testing would be needed, and economically, it is not interesting to make such a large dataset. This is why generative AI is opening an opportunity in the next few years to predict properties and identify structures, which will be discussed in Section Generative AI.
Deep learning models implemented to discover new biochemical molecules
Recently, a deep learning approach was utilized to screen millions of chemical compounds to evaluate their potential applications as antibiotics (Wong et al., 2024). The AI-driven methodology employed identified several compounds that showed efficacy against drug-resistant pathogens, such as the methicillin-resistant Staphylococcus aureus (MRSA) and the vancomycin-resistant enterococci. Distinguishing itself from conventional AI models, this particular algorithm was designed to be transparent, enabling scientists to comprehend the model's decision-making processes and the underlying biochemistry involved (Table 13).
Aspect | Description | Techniques/models used | Key findings/improvements |
---|---|---|---|
Data Preparation | The study began with the collection of a large dataset encompassing antibiotic activities and human cell cytotoxicity profiles of over 39,000 compounds. | Screening of chemical libraries; Utilization of the MCULE purchasable database for virtual screening. | Enabled the identification of compounds with potential antibiotic activity and low cytotoxicity. |
Model Architecture & Training | Ensembles of graph neural networks were trained using the collected dataset, focusing on predicting antibiotic activity and cytotoxicity. | Graph neural networks augmented with RDKit features for enhanced prediction accuracy. | Achieved robust model performance, enabling the prediction of antibiotic activity for over 12 million compounds. |
Validation & Feedback Mechanism | The models' predictions were empirically validated by testing a subset of predicted compounds against Staphylococcus aureus, with a focus on those exhibiting high antibiotic activity and low cytotoxicity. | Monte Carlo tree search method for identifying chemical substructures associated with antibiotic activity. | Empirical testing confirmed the models' predictive accuracy, with several compounds showing effective antibiotic properties. |
Discoveries & Future Directions | The study uncovered novel structural classes of antibiotics, including a class selective against MRSA and vancomycin-resistant enterococci, which demonstrated efficacy in mouse models. | Substructure-based approach for exploring chemical spaces; In-depth analysis of chemical scaffolds using graph-based rationales. | Opened new avenues for antibiotic discovery by demonstrating the potential of deep learning-guided exploration of chemical spaces. The approach also highlighted the importance of model explainability in identifying chemical substructures critical for antibiotic activity. |
Generative AI
Deep learning (DL) and its subset, generative artificial intelligence, are extending the boundaries of molecule design by generating novel molecular structures with potentially superior properties (Cesaro et al., 2023; Wong et al., 2024; Yang et al., 2023). These models could potentially learn from vast datasets of chemical compounds and predict new surfactants that meet criteria such as biodegradability, low toxicity, and high efficacy in targeted applications. The ability to generate and screen thousands of potential surfactants in silico drastically reduces the time and cost associated with bio-based surfactant development.
Furthermore, the application of ML and DL might extend beyond molecular design to the prediction and optimization of surfactant properties and their interactions in biological systems. Studies such as those by Bannigan et al. (2021), Vamathevan et al. (2019) and Cesaro et al. (2023) highlight the role of ML and DL in enhancing the efficiency and efficacy of drug discovery processes, which can be adapted to surfactant research for improved drug delivery mechanisms. The integration of chemical and biological data through AI enables the design of surfactant-like molecules with enhanced properties, addressing challenges such as drug solubility and stability (Vora et al., 2023).
The potential of artificial intelligence in surfactant development is not limited to the prediction of physical properties and molecular design. It also extends to the optimization of surfactant formulations for specific applications, as demonstrated by the emerging research on surfactant-assisted drug delivery systems (Colombo, 2020; Hassanzadeh et al., 2019). Here, deep learning models can potentially predict how surfactants interact with other components, such as drugs and polymers, to improve the stability, solubility, and bioavailability of pharmaceutical formulations.
- Advanced Data Collection and Integration: Beyond traditional datasets, this involves collecting a wide array of chemical data, including experimental results, literature findings, and proprietary databases. Integrating diverse data types, such as quantum mechanical calculations, molecular dynamics simulations, and experimental physicochemical properties, can provide a more holistic dataset for AI models.
- Enhanced Data Preprocessing with AI: Use of AI to identify and fill data gaps, predict missing properties, and generate synthetic data points to improve model training. Techniques such as data augmentation and generative adversarial networks (GANs) can be particularly advantageous here.
- Multi-Objective Optimization: Instead of focusing on a single property, AI models can optimize multiple objectives simultaneously, such as biodegradability, efficacy, and low toxicity. This approach ensures the development of surfactants that are not only effective but also environmentally friendly and safe.
- Integrative Model Training with Transfer Learning: Utilization of transfer learning and multi-task learning to leverage knowledge from related domains, such as pharmaceuticals and materials science, to improve the predictiveness for surfactant molecules. This can significantly reduce the need for extensive surfactant-specific data.
- Iterative Design and Synthesis Loop: Implementation of an iterative loop where AI not only predicts the structure of new surfactants but also suggests modifications to existing molecules based on desired properties. Coupled with automated synthesis platforms, this can accelerate the cycle of hypothesis, testing, and refinement.
- Virtual Screening and Property Prediction: Use of deep learning models for virtual screening of vast chemical spaces to identify promising surfactant candidates. Simultaneously, prediction of critical properties such as solubility, critical micelle concentration, and interfacial tension using specialized AI models.
- Experimental Validation and Feedback Loop: After synthesizing the AI-predicted surfactants, conduction of rigorous experimental testing to validate their properties and performance. The results should feed back into the AI models to refine predictions and guide future synthesis, creating a dynamic learning process.
Additionally, generative AI presents a significant opportunity to improve the production of biobased products. Generative AI has the potential to enhance the design and operational efficiency of chemical processes, particularly in generating and refining process flowsheets and assisting in error correction and hazard studies. This could lead to more efficient data utilization, innovative problem-solving, and the creation of models that combine traditional knowledge with AI insights (Schweidtmann, 2024).
Recently large language models (LLM) (Boiko et al., 2023) and generative pretrained transformer (GPT) models (Yang et al., 2023) have been proposed as a tool to optimize chemical synthesis, design new polymer molecules. The body of knowledge that is being created in different disciplines, mainly the pharmaceutical field, will allow the use of deep learning and generative AI in other applications, such as novel biobased surfactants synthesis.
Large language models in scientific research
Large language models (LLMs) can be used for a wide array of scientific tasks, including chemical synthesis. These models, based on transformer architectures, are adept at navigating and interpreting extensive scientific literature and databases. By training on diverse datasets from scientific publications, patents, and experimental records, LLMs develop a comprehensive understanding that can be applied to complex scientific queries. The application of LLMs extends to the autonomous design, planning, and optimization of experiments, significantly enhancing the efficiency of scientific inquiry. However, a notable challenge lies in maintaining a balance between the innovative output of generative models and the stringent demands for scientific accuracy and relevance. This necessitates sophisticated tuning of the models to ensure the generation of hypotheses and experimental designs that are both novel and scientifically sound.
Use of generative models for the discovery of new polymer electrolyte molecules to substitute polyethylene oxide polymers
It has been demonstrated that molecular structures enabling the enhancement of specific properties, such as conductivity, can be achieved using Generative AI (Yang et al., 2023), For surfactants, depending on the application, enhancing a property—such as detergent effectiveness, surface tension reduction, or emulsification capacity—is desirable. This would involve compiling a database with the structures of a certain surfactant family to identify candidate molecules that could maximize these properties. The selected molecule would likely, if synthesized, offer superior characteristics. This approach not only facilitates molecule design but also increases the likelihood that the synthesized molecules in the laboratory will exhibit the desired properties. The next example, which is related to the discovery of novel polymer electrolyte structures, allows us to envision the development in this field, coupled with MD simulations and laboratory experiments.
The discovery of high-performance polymer electrolytes, analogous to PEO polymers with superior ionic conductivity relies on the application of conditional generative models, such as those based on the minGPT framework (Figure 34 & Table 14). These models are instrumental in suggesting potential polymer structures that exhibit high ionic conductivity, a crucial attribute for the performance and safety of energy storage devices. The primary source of data for training these models is derived from molecular dynamics simulations, offering a rich foundation that encapsulates the complex interplay between molecular structures and their ionic conductivities. Through an iterative learning process enhanced by feedback from additional molecular dynamics simulations, the models continually refine their polymer suggestions (Yang et al., 2023). This approach not only speeds up the identification of novel polymer electrolytes but also ensures that the proposals progressively align with the target property of enhanced ionic conductivity. The main challenge in this research area is the significant computational resources required for detailed simulations, which are essential for accurate property predictions. Moreover, the reliability of force fields used in these simulations is critical for generating dependable data. Looking ahead, the field aims to extend to automated synthesis and experimental testing, creating a comprehensive loop from computational design to practical validation. Refining models with experimental feedback will improve the accuracy of predictions, potentially transforming the development of future energy storage technologies.
Aspect | Description | Techniques/models used | Key findings/improvements |
---|---|---|---|
Data Preparation | The initial step involves curating a dataset of polymers, including polymer electrolytes, with their ionic conductivities computed from MD simulations. This dataset serves as the seed data for training the generative model. | High-Throughput Polymer Design – Molecular Dynamics (HTP-MD) database, consisting of 6024 linear chain homopolymers. | Oversampling of high-conductivity polymers to balance the dataset and ensure a focus on generating high-performance candidates. |
Model Architecture & Training | A conditioned generative model based on the minGPT architecture is employed. The model learns to generate polymer SMILES strings conditioned on their ionic conductivity classes, aiming to produce polymers with targeted properties. | minGPT architecture, with modifications to include ionic conductivity classes in the input. | The model successfully generates novel polymer candidates with shifted distributions towards higher ionic conductivity compared to the training set. |
Validation & Feedback Mechanism | Each batch of generated polymers undergoes validation through MD simulations to evaluate their ionic conductivities. High-performing polymers are added to the training dataset, and the model is retrained, enhancing its ability to target desirable properties in subsequent iterations. | Molecular dynamics simulations; active learning and feedback loop for continuous model improvement. | Increased average ionic conductivity of polymers generated in subsequent iterations, showcasing the model's evolving efficiency. |
Discoveries & Future Directions | The platform identified 19 novel polymer repeating units with ionic conductivities surpassing that of Polyethylene Oxide (PEO), a benchmark material. These discoveries open new avenues for the development of high-performance polymer electrolytes. | Analysis of polymer repeating units discovered; comparison of ionic conductivity, ion diffusivity, and concentrations of free ions between generated polymers and PEO. | The discovered polymers exhibit enhanced ion transport properties, suggesting potential for significant advancements in energy storage technologies. Future work will focus on experimental validation and exploration of more complex polymer systems. |
CONCLUSIONS AND PERSPECTIVES
Surfactant molecules play a critical role in various applications across various industrial sectors, requiring tailored functionalities and environmental sustainability. The development of biobased surfactants and biosurfactants from initial research to their current market presence illustrates a significant transition within the industry, motivated by global sustainability trends. Since the early production of alkyl polyglycosides in the 1980s by Henkel and BASF in Germany, to the start of production of biosurfactants by Evonik nowadays, there has been considerable progress, highlighted by substantial investments in production facilities globally. This transition reflects an industry-wide shift from traditional fossil-based surfactants towards sustainable alternatives with a lower carbon footprint.
The eco-friendly attributes of biobased surfactants and biosurfactants, including biodegradability, reduced toxicity, and utilization of renewable resources, are central to their consumer perception and potential for broader adoption. However, economic considerations in production and competitive demand for raw materials, coupled with the aim of achieving performance comparable or superior to conventional surfactants pose significant challenges. Addressing these challenges requires employing agricultural waste or other types of biomasses with a low carbon footprint. The use of lower cost raw materials will require advancements in biotechnological production methods, which could benefit from advancements in reactor design and process separation. Furthermore, artificial intelligence emerges as a promising area that eventually will assist the transition towards sustainable materials and processes, especially in the complex domain of biobased products.
Another challenge lies in the inherent complexity of biobased surfactants and biosurfactants, often resulting from their heterogeneous mixtures. Ensuring consistency in these mixtures requires careful selection of raw materials, analytical techniques, and synthesis routes. There is a crucial need for surfactants with improved characteristics and purity to facilitate easier formulation. This scenario offers an opportunity to employ AI and generative techniques for discovering novel molecules in the domain of biobased surfactants and biosurfactants, potentially leading to surfactants with superior properties and environmental benefits. Generative models hold the potential to design new surfactant molecules that align with specific criteria such as biodegradability, low toxicity, and high efficacy. Among existing challenges are the necessity for large and diverse datasets for model training, the complexity of accurately simulating surfactant behavior, and ensuring that AI-generated molecules are synthesizable and economically viable.
AUTHOR CONTRIBUTIONS
Ronald Marquez: Conceptualization, methodology, writing – original draft, writing – review and editing. Maria S. Ortiz: Conceptualization, methodology, writing – original draft. Nelson Barrios: Conceptualization, writing – original draft, writing – review and editing. Ramon Vera: Conceptualization, writing – original draft, writing – review and editing. Álvaro Javier Patiño-Agudelo: Writing – original draft, Writing – review and editing. Keren A. Vivas: Writing – review and editing. Mariangeles Salas: Writing – review and editing. Franklin Zambrano: Writing – review and editing. Eric Theiner: Writing – original draft, writing – review and editing.
ACKNOWLEDGMENTS
The authors wish to express their gratitude to Emeritus Professor Jean-Louis Salager, Emeritus Editor in Chief of the Journal of Surfactants and Detergents, to whom this special issue is dedicated. We would also like to express our acknowledgments to our colleagues and former co-authors on the subject of formulation with surfactants: Johnny Bullón and Ana Forgiarini from Venezuela, Jesus Ontiveros, Valérie Molinier, Dominique Langevin, Véronique Nardello-Rataj, and Jean-Marie Aubry from France. The authors also wish to thank all the Ph.D. students and researchers from FIRP Laboratory (University of Los Andes), PHD Laboratory (University of Carabobo), and CISCO-UCCS (University of Lille), who have significantly contributed to the body of knowledge on this subject. The authors also acknowledge the continuous support of Dr. Ronalds Gonzalez and Dr. Lokendra Pal from North Carolina State University. Álvaro Javier Patiño-Agudelo acknowledges post-doctoral fellowships provided by the CNPq (151115/2023-0) and FAPESP (2023/11091-9). Professor Douglas Hayes is thanked for his recommendations on the structure and contents of the review.
FUNDING INFORMATION
This research received no external funding.
CONFLICT OF INTEREST STATEMENT
The authors declare that they have no conflict of interest.
ETHICS STATEMENT
Human or animal studies were not part of this research.