In recent years, topics such as renewable, green, eco-friendly, or recyclable are widespread because of increasing attention to environmental matters. In spite of they are sometimes only being used to make products more appealing to consumers, as evidenced by the exponential growth in eco-marketing and green labeling initiatives, their use should always come under the broad heading of sustainable development, which requires making every decision with the future in mind. In chemistry, sustainable mentality is guided by the (12 Principles of Green Chemistry) introduced in 1998 by Anastas and Warner.1 One of the principles of green chemistry is to prioritize the use of renewable materials including the use of agricultural waste or biomass. Other principles focus on prevention of waste, less hazardous chemical synthesis, and design safer chemicals including safer solvents. Others focus on the design of chemical products to safely degrade in the environment and the efficiency and simplicity in chemical processes.
When these principles are evaluated by a polymer chemist, the one that encourages the exploitation of available renewable resources can be easily identified as the most important, because the life cycle of a polymer or material starts from the raw material. Today, it is entirely impossible to envisage a polymer chemistry business that is not based predominantly on carbon. Thus, thinking about our future generations, it is obvious that in ages of depleting fossil reserves and an increasing emission of greenhouse gases, this carbon must come in the future from either biomass or from CO2, with no alternative.
Nowadays, some of the most widely applied renewable raw materials in the chemical industry for nonfuel applications include plant oils, polysaccharides (mainly cellulose and starch), sugars, wood, and others.2 Indeed, plant oils and especially fatty acids derived therefrom are among the most important raw materials for polymer chemistry.3–5 They have already been established as a well-known pool of renewable raw materials for polymer synthesis, including polyesters, polyamides, polyanhydrides, epoxy resins, polyolefin analogs, and to greater extent polyurethanes.6–8 In general, three general approaches have been considered to convert plant oils to both crosslinked and linear polymers: (a) the direct polymerization of triglyceride double bonds, (b) their chemical modification and later polymerization, and (c) the synthesis and subsequent polymerization of monomers from plant oil-derived chemicals such as fatty acids, which can be easily obtained by either simple hydrolysis or alcoholysis of triglycerides.
Guided by this green philosophy, undoubtedly there is still a need to develop environmentally friendly monomer and polymer synthesis strategies to keep moving toward more sustainable polymer chemistry. To achieve this end, the family of reactions collectively termed (click chemistry) can definitely lend a hand. Click chemistry reactions are often cited as a style of chemical synthesis that is consistent with the goals of green chemistry. This concept, introduced by Sharpless and coworkers in 2001, encompasses a wide range of reactions characterized by selectivity, facile experimental set-up, applicability in aqueous and aerobic systems, tolerance to a variety of functional groups, quantitative yields, and minimal synthetic work-up.9 Initially, the Cu(I)-catalyzed azide–alkyne cycloaddition attracted most of the attention in the field; however, this concept can be extended to many other highly efficient reactions, such as nucleophilic substitutions, radical additions, Michael additions, as well as Diels–Alder and retro Diels–Alder reactions.
Among the multiple reactions that have been accepted into the click chemistry realm, the radical addition of thiols to CC bonds, which is currently called thiol-ene addition, is absolutely the champion in the field of oleochemistry. Thiol-ene addition is century-old chemistry10 that offers high yields, regiospecificity and stereospecificity, and outstanding functional group tolerance under simple reaction conditions.11 Moreover, one remarkable feature inherent to this reaction is that virtually any alkene functional group can participate. The reaction proceeds via a free radical chain mechanism and can be initiated by UV light or radical initiators to mainly yield the anti-Markovnikov products.12 However, it is important to point out that the addition of thiyl radicals to olefins is reversible and can thus lead to cis/trans isomerization of the starting material.13 Interestingly, the extent of reversibility of C—S bond formation with terminal olefins is much less pronounced than that of the addition to olefins with nonterminal double bonds. Thus, terminal double bonds are more reactive than internal ones.14 In a context of sustainable polymer chemistry, there is no doubt that plant-derived oils bearing intrinsic double-bond functionality together with hydrothiolation reaction can be considered a promising marriage.
In this highlight article, we will provide a complete account of the recent contribution of click chemistry reactions in the transformation of oleochemicals into monomers and polymers. The focus will be on the thiol-ene addition reaction although other click reactions such as thiol-yne addition, and Cu(I)-catalyzed azide–alkyne and Diels–Alder cycloadditions, will also be discussed (Scheme 1). The expectation is to help readers in inspiring new challenges and ideas for plant oil-based products and processes.
THIOL-ENE ADDITION IN OLEOCHEMISTRY
Thiol-Ene Addition of Fatty Acids and Derivatives
Nowadays, fatty acids are absolutely in the spotlight of polymer chemistry community, and have already become valuable renewable building blocks for the synthesis of designed monomers and polymers. Among the variety of fatty acids available from plant oils, oleic acid, a C18 fatty acid containing a CC bond at the ninth position that can be isolated by hydrolysis of high-oleic sunflower oil, and ricinoleic acid, in the form of its 10-undecenoic acid C11 derivative containing a terminal CC bond, have been by far the most used compounds. Although hydrothiolation reaction was intermittently applied to fatty acids since more than 50 years ago,15–26 oleochemistry undoubtedly has made the most of this reaction in the last years.27 In particular, attention has been paid to the preparation of AA and AB polycondensable monomers from isolated fatty acids for polyurethane, polyester, polyanhydride, as well as polyamide synthesis. Monomer chemical structures are systematically outlined in Table 1. However, it is worth nothing at this point that hydrothiolation reaction has also been used as a polymerization tool in the preparation of linear polymers from fatty acid-based α,ω-dienes, as well as a postpolymerization modification tool of well-defined reactive fatty acid-based polymers. Within the following paragraphs, all the contributions will be presented following this structure.
Table 1. Chemical Structures of AA and AB-Type Fatty Acid-Derived Monomers Prepared Using Thiol-Ene Addition as a Key Reaction Step
Synthesis of AA and AB Fatty Acid-Based Monomers
Given the large number of commercially available functional thiols, several research groups envisioned thiol-ene addition as a very attractive tool to introduce different functional groups to fatty acids(esters) leading to a variety of well-defined polycondensable monomers. As mentioned above, all the attention has almost focused on monounsaturated fatty acids such as oleic and 10-undecenoic acids as well as their derivatives because of their potential for the preparation of strictly bifunctional polymer precursors. As can be observed in Table 1, intensive efforts have been paid to design and prepare diols, suitable for polyurethane technology.40
To this end, initial findings were reported in 2011 by our group who described the preparation of two diols (M1 and M2, Table 1) containing primary hydroxyl groups using 10-undecenoic and oleic methyl esters as starting materials.28 After optimization of the reaction conditions, both fatty ester CC bonds could be easily and quantitatively functionalized with a primary hydroxyl group taking advantage of 2-mercaptoethanol (T1, Scheme 2) radical addition. The aforementioned outstanding high reaction rates under thiol-ene conditions of terminal CC bonds allowed carrying out methyl 10-undecenoate functionalization in the absence of 2,2-dimethoxy-2-phenylaceto-phenone (DMPA) photoinitiator. Although thiol-ene addition to 10-undecenoic acid and derivatives is known to proceed completely at stoichiometric ratio,29 in this case a slight excess of thiol (thiol to CC molar ratios of 1.8:1) was used to reduce the reaction time scale from several hours to minutes. The excess of thiol reagent could be eliminated simply by a liquid/liquid extraction with water, following thus one of the most important click chemistry issues. Subsequently, methyl ester group was reduced in both cases, using LiAlH4, introducing the second primary hydroxyl functionality in the hydrocarbon chain.
Using a similar strategy but taking advantage of the carboxylic acid group reactivity to tune diol structure, other authors reported the preparation of different diols containing ester, amide, or ester/amide linkages. In this case, a two-step methodology consisting of an esterification or amidation step followed by a thiol-ene addition with T1 was used. For example, Cramail and coworkers38 demonstrated that the ester-containing diol M16 can be obtained in two steps by heating methyl 10-undecenoate with a large excess of 1,3-propanediol at 120 °C in the presence of 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) to obtain an ω-unsaturated alcohol, which can be easily functionalized with a second hydroxyl group via hydrothiolation with T1 under UV irradiation without photoinitiator. Similarly, replacing 1,3-propanediol with the stoichiometric amount of 1,3-propanolamine, the amide-containing diol M17 was also obtained.
Diester (M21 and M22)-, ester-amide (M23)-, and diamide (M24)-containing diols were also prepared using a similar general procedure using 1,3-propanediol, isosorbide, 1,3-aminopropanol, or 1,4-diaminobutane as spacers. Surprisingly, higher amounts of T1 were necessary to perform efficient thiol-ene additions on amide-containing precursors. Moreover, because of solubility problems, final thiol-ene step to prepare diamide-containing diol M24 was carried out at 80 °C in N-methylpyrrolidone using thermal initiation in the presence of 2,2′-azobis(2-methylpropionitrile) (AIBN). David and coworkers extended this synthetic pathway to an oleic acid-rich fatty methyl esters mixture prepared by methanolysis of soybean oil.37 Taking into account the use of natural oils, in this case, pseudo-telechelic polyols having an average functionality of 2 (or slightly higher than 2) were obtained. The authors used ethyleneglycol, 1,4-butanediol, 1,6-hexanediol, 2-aminoethanol, 5-aminopentanol, and 1,8-octanediamine as central spacers, and turned to AIBN-thermally initiated thiol-ene additions to prepare asymmetric (M14, M15, M18, M19, and M20) and symmetric (M25, M26, M28, and M31) monomers. To avoid solubility problems during thiol-ene coupling reactions with amide-containing monomers, amidation step was carried out at the end. Although a moderate excess of T1 was used (thiol to CC molar ratios of 5:1), long reaction times (8 days at 60 °C) quite far from click standards were required to achieve high conversions under these conditions. Later, rapeseed oil was also transformed into polyols applying the same approach.41 A dramatic reduction of the thiol-ene addition reaction time was reported by Cramail and coworkers when carried out the functionalization of similar oleic acid-based structures with 1,5-pentanediol (M27) and polyethyleneglycols of different length (13 and 45 repeating units, M29 and M30) spacers.39 Using photochemical initiation (225 nm) and a thiol to CC molar ratio of 6:1, 90% conversion determined by 1H NMR spectroscopy was achieved at 0 °C after 2 h demonstrating a temperature effect in the dissociation rate of the radical generated from addition of thiyl radicals to CC bonds.42 Finally, esterification of oleic or 10-undecenoic acids with allyl alcohol followed by the double hydrothiolation with T1 was proved as an alternative way to transform both fatty acids into diols (M32 and M33).28
As expected, all the aforementioned diols proved to be suitable for polyurethane synthesis by combination with either aliphatic or aromatic conventional diisocyanates. Generally, all the synthesized polyurethanes showed medium to high molecular weight, good solubility in conventional organic solvents, moderate thermal stability, and a broad range of thermal and mechanical properties being those dominated by the chemical structure of the parent diol. Although little has been studied about potential applications of such biobased polyurethanes, our group has been investigating the application of some of those materials as biomedical implants, showing that combined with bioactive molecules such as gelatin can be used as valid substrates to prepare biomimetic materials and interfaces.40–43
Beyond the preparation of diols, several research groups have recently been evaluating the thiol-ene reaction as a facile and convenient tool for the preparation of other well-defined polycondensation monomers. For example, Türünç and Meier29 reported the preparation of several polyester precursors (M3, M6,32M7, M8, M34, and M35) via hydrothiolation of methyl 10-undecenoate and 10-undecenol with T1, 1-thioglycerol (T2), methyl thioglycolate (T3), and 1,4-butanedithiol (T4). Pushing the limits of thiol-ene addition, reactions were performed at 1:1 ratio of reactive groups without photochemical or thermal initiator. Unfortunately, this clearly punished the reaction times. For example, the T1 addition to methyl 10-undecenoate (4 g product scale) at 35 °C resulted in a satisfactory conversion after 68 h. The resulting monomers were then polymerized to linear as well as hyperbranched polyesters with molecular weights between 4 and 9 kDa. Most suitable conditions for polymerization were 120 °C and 5 mol % TBD catalyst with continuous vacuum. Recently, Li and coworkers also reported T2 addition to methyl 10-undecenoate benefiting from click chemistry feature of DMPA-photoinitiated thiol-ene reaction.33 In this case, large-scale reaction with yields up to 95% in less than 10 min without any stirring were reported compared with the low monomer preparation efficiency reported by Türünç and Meier (66% yield at 35 °C and 1500 rpm for 6 days at 50 mmol scale). Indeed, the authors reported the self-polycondensation of M7 under various reaction conditions producing high-molecular-weight hyperbranched polyesters with unusual crystalline properties. Meier and coworkers also demonstrated that α,ω-aminomethylester M9 based on methyl 10-undecenoate is also easily accessible using cysteamine (T5) hydrochloride.34 This product was used to prepare polyamides of varying thermal properties by TBD-catalyzed copolymerization with adipic acid and 1,6-hexamethylene diamine. Extending this approach to 3-mercaptopropionic acid (T6), our group described the preparation of dicarboxylic (M11) and tricarboxylic acid monomers based on 10-undecenoic acid.35 These monomers were examined with respect to their ability to form polyanhydrides by melt condensation under vacuum. For instance, M11-derived linear polyanhydride of molecular weight of about 22.3 kDa showed a melting process at around 65 °C, and therefore was investigated as drug delivery vehicle. Thus, Rhodamine B as model drug could be incorporated by a melting/crystallization process into polyanhydride devices, and could be delivered via a surface erosion degradation mechanism (Fig. 3). On the other hand, thiol-ene addition of T3 to methyl 10-undecenoate has also been investigated as an initial step in the preparation of fatty acid-derived isocyanates (M13) via Curtius rearrangement.36 Moreover, it is worth nothing at this point the isocyanate-free approach to polyurethanes based on hydroxy-acyl azide monomer M4, in which hydroxyl group was also introduced via hydrothiolation with T1.30 Finally, Gandini and coworkers also investigated the functionalization of methyl 10-undecenoate with 2-furfuryl thiol (T7) to prepare Diels–Alder reactive monomers (vide infra).44, 45
Hydrothiolation has also been applied to the functionalization of oleic acid with several commercial thiols including T5 hydrochloride, T6, and T1 giving a variety of difunctional monomers (M10, M12, and M5). Typically, moderate excess of thiol, the presence of radical initiator, and reaction time scale of several hours are the parameter reactions required to ensure quantitative conversion of the internal CC bond. However, Auvergne and coworkers demonstrated that using the appropriate irradiation, T1 addition to oleic acid can be performed under mild conditions requiring neither solvent nor photoinitiator in 1 h.46 The prepared oleic acid-derived polycondensation monomers were used to obtain polyanhydrides, polyamides, as well as isocyanate-free polyurethanes that usually behave as amorphous materials because of the presence of dangling chains.30–35 On the other hand, Lapinte and coworkers recently used this strategy to prepare an oleic acid-based initiator for 2-methyl-2-oxazoline cationic ring-opening polymerization.47 Thus, methyl oleate was coupled with T1 under UV irradiation to introduce hydroxyl groups, which were further transformed in p-toluenesulfonyl, trifluoromethanesulfonyl, or chloroacetyl esters. This initiator was used to successfully polymerize 2-methyl-2-oxazolines with good control of the molecular weight and polydispersities. The same approach was applied to grapeseed oil triglyceride to obtain macroinitiators with a more complex architecture. In both cases, the obtained lipopolymers were able to self-organize into nanostructures in aqueous medium without formation of any supra-aggregates or uncontrolled self-assembly.
Synthesis of Fatty Acid-Based Polymers
The outstanding efficiency and high reaction rate demonstrated by thiol-ene addition performed on terminal CC bonds encouraged several researchers to investigate its performance as oligomerization/polymerization tool of 10-undecenoic acid-derived α,ω-dienes (M36–43, Scheme 4) using bifunctional thiols. It should be noted that such a polymerization reaction follows typical step-growth polymerization rules; therefore, the diene/dithiol ratio is an important parameter to achieve high-molecular-weight polymers. Thiol-ene addition was shown to be useful when low-molecular-weight oligomers are targeted. For example, reaction of a slight excess of allyl 10-undecenoate (M36) with 3,6-dioxa-1,8-octanedithiol (T8) under DMPA/UV initiation conditions resulted in the rapid and quantitative formation of oligomers that NMR analysis proved to contain mainly allyl ester end groups.48
Model studies were used to demonstrate that allyl ester CC bond is 1.8 times less reactive toward hydrothiolation than the vinylic one. Oligomers with molecular weight up to 3 kDa, which were further modified at the chain ends demonstrating the high efficiency of thiol-ene coupling during chain-ends functionalization (vide infra), could be readily obtained in a controlled manner by varying the ene/thiol ratio. Meier and coworkers also reported the successful preparation of medium molar mass polymers containing different functionalities in the main chain from fully renewable dienes based on 10-undecenoic acid using a dithiol as coupling agent.49–51 Thus, an ether-containing diene M37, obtained from 11-bromo-1-undecene and 10-undecenol, an ester-containing diene M38, prepared from methyl 10-undecenoate and 1,3-propanediol, and a diene M39 synthesized via a Ugi four-component reaction, bearing amide groups both in the main as well as side chains, were successfully polymerized generally applying the AIBN thermally initiated version of thiol-ene addition using dithiols. For example, polyester from M38 and bis(2-mercaptoethyl) ether (T9) with molecular weight of about 12 kDa and PDI of 2.00 was obtained at 80 °C for 2 h in the presence of only 2.5 mol % AIBN. On the other hand, a polyether with molecular weight of about 18 kDa and PDI of 2.63 was obtained from diene M37 and the same dithiol. Unfortunately, hydrothiolation failed in the attempt to obtain high-molecular-weight polyanhydride from the anhydride-containing diene M40. This was attributed to the high reactivity of the anhydride functionalities toward nucleophiles, in this case thiol groups, that caused the scission of either monomer or polymer backbone via thioester formation. Recently, the same group also examined the polymerization of other α,ω-dienic monomers based on methyl oleate and methyl erucate (M41), and vanillin alcohol and 10-undecenoic acid (M42 and M43) obtaining satisfactory results, although the molar masses of the reported polymers were relatively limited.32, 52 However, in a complimentary approach, Du Prez and coworkers demonstrated that high-molecular-weight polythioethers (up to 40 kDa) are also accessible by UV- or thermal-initiated thiol-ene polyaddition polymerization of 10-undecenethiol monomer M44 as an AB-type reactive monomer.53 To obtain high-molecular-weight polymer, the concept of using an α-olefinic ω-thiol monomer has the advantage, in contrast to the polymerization of two difunctional monomers, of having intrinsic stoichiometry.
Modification of Fatty Acid-Based Polymers
Beyond the synthesis and polymerization of fatty acid-based monomers, the radical-mediated thiol-ene addition has also been used by several research groups as a facile and convenient tool for the postpolymerization modification at the main chain [Scheme 5(A)], side chain [Scheme 5(B)], and end groups [Scheme 5(C)] of some well-defined plant oil-derived polymers.
Heise and coworkers described the thiol-ene functionalization of polyester synthesized by enzymatic ring-opening polymerization of globalide, an unsaturated macrolactone synthesized from hydroxyl fatty acid.54 The aliphatic polyester (Mn = 16 kDa and PDI = 2.5) was prepared and modified on the main chain with a range of thiols including N-acetyl T5, butyl ester of T6, and 6-mercapto-1-hexanol (T10). After optimization of the reaction conditions, thiol-ene additions were performed with 6.6–11.0 equiv of thiol in the presence of AIBN at 80 °C for 24 h. 1H NMR analysis was used to demonstrate that the degree of modification was >95% for T10 and N-acetyl T5 modifications; however, lower efficiency (75%) was achieved when butyl ester of T6 was used.
Hoogenboom and coworkers also investigated the modification of well-defined ene side-chain functional polymers prepared from 2-(dec-9-enyl)–2-oxazoline, based on 10-undecenoic acid.55 Using 1H NMR spectroscopy and MALDI-TOF MS investigations, the authors demonstrated that CC bond of the monomer stayed unaffected after the polymerization initiated by methyl tosylate under both conventional heating (100 °C) and microwave irradiation, allowing their use for subsequent thiol-ene modifications with dodecanethiol (T11) and 2,3,4,6-tetra-O-acetyl-1-thio-β-d-glycopyranose (T14).
Both modifications were performed using slight excess 1.3–3.0 of thiol under irradiation with UV light for 24 h in 2-methyl-THF (mTHF)56 and methyl laureate as (green) solvents (mTHF is obtained from waste biomass exhibiting similar properties as THF, whereas methyl laureate is the ester of the lauric acid derived from coconut and palm kernel oil). A quantitative addition of both thiols onto the pendant double bonds could be confirmed by NMR spectroscopy. In addition, size-exclusion chromatography (SEC) investigations of the polymers showed a shift of the SEC trace of the starting compound to lower elution REFVIDumes after both thiol-ene modifications. High degrees of modification (>99%) were also reported by Kolb and Meier during the thiol-ene modifications of side chains of poly(malonate) (10.2 kDa) bearing C9 unsaturated pendant moieties with several commercially available thiols.57 This polyester was prepared by condensation of 9-nonenyl malonate, a methyl 10-undecenoate derivative, with 1,6-hexanediol catalyzed by 1.0 mol % titanium (IV) isopropoxide. For all the investigated thiols, bearing different functionalities such as hydroxyl, carboxylic acid, and ester, equimolecular amounts of the thiol with 5.0 mol % DMPA and UV irradiation were reported to be convenient for a quantitative conversion of the CC bond after 1 h with very low polymer–polymer coupling (<5%) and no other side reactions such as radical-initiated ring closure of the side chains.
Finally, our group also investigated the end-groups functionalization of the aforementioned low-molecular-weight linear oligomers (<3 kDa) prepared from M36.48 It was demonstrated that photoinitiated hydrothiolation reaction at the chain ends of these fatty acid-derived oligomers with slight excess of thiol allowed the preparation of perfect telechelics with hydroxyl (T1), carboxyl (T6), or trimethoxysilyl (3-mercaptopropyl-trimethoxysilane T15) groups in essentially quantitative yield as determined by 1H NMR spectroscopy and MALDI-TOF MS end-group analysis (Fig. 2). Hydroxyl-terminated oligomers were further used as the soft, flexible block in segmented polyurethanes.58
Thiol-Ene Addition of Natural and Modified Triglycerides
It is well known that CC bonds in the triglyceride structure are not sufficiently reactive for any viable polymerization process, except for cationic polymerization.59 However, thiol addition to CC bonds of naturally occurring unsaturated triglyceride is a priori a straightforward and attractive way to functionalize these products with reactive groups with higher aptitude to be polymerized. Aside from the highly efficiency demonstrated by the thiol-ene addition in polymer/materials chemistry as well as in traditional (bio)organic synthesis, functional group tolerance is probably its most outstanding characteristic. This valuable feature has also been exploited in the field of oleochemistry to introduce several functional groups into unsaturated triglycerides. This is exemplified by the recent work of Lapinte and coworkers60 who modified grapeseed oil triglycerides with T5 hydrochloride (thiol to CC molar ratios of 3:1) in the presence of DMPA as initiator at room temperature for 8 h under UV (365 nm) irradiation. In this case, reaction was performed in the presence of solvent (1,4-dioxane/ethanol 70/30 v/v) because of the insolubility of T5 hydrochloride in the oil. In this way, starting from grapeseed oil containing 4.75 double bonds per triglyceride, an oil derivative with an amine equivalent weight of 290 g equiv–1 for 87% conversion was obtained. This vegetable-oil-based polyamine was used as hardener for epoxidized linseed oil. Using the same approach, Auvergne and coworkers functionalized rapeseed oil triglycerides with primary hydroxyl groups.46 The authors highlighted the fact that photoreaction with T1 required neither solvent nor photoinitiator to obtain complete CC conversion using a reasonable excess of thiol to CC (3:1 molar ratio). Additionally, model studies were systematically performed for the first time to determine the byproducts during thiol addition to fatty compounds, being sulfide formation and intermolecular recombination the most significant secondary processes. It is important to note that in both cases, the excess of thiol reagents could be easily removed by nonchromatographic methods (crystallization at low temperature for T5 hydrochloride and liquid/liquid extraction with water for T1).
Given the facile and efficient performance of thiol-ene coupling on the functionalization of natural triglycerides, it is not therefore surprising that such chemistry has also been used as a means of oligomerization process using high thiol to ene ratios. The synthesis of thiol-functional soybean oil oligomers achieved by AIBN thermally initiated thiol-ene addition using commercially available multifunctional thiols such as ethyleneglycol di-3-mercaptopropionate (T16), trimethylolpropane tri-3-mercaptopropionate (T17), and pentaerythritol tetra-3-mercaptopropionate (T18) was investigated by Webster and coworkers.61 As expected, it was found that only a small amount of thiol-terminated oligomers was formed at low thiol to CC molar ratios (2:1 and 4:1). However, successful oligomerization was confirmed using SEC at higher molar ratios (9.4:1 and 18.8:1) (Fig. 3). For example, oligomers of 6 kDa could be obtained after 20 h at 90 °C using T18.
Using an analogous strategy, the same group examined the oligomerization of sucrose soya ester with eight unsaturated fatty acid chains affording high-molecular-weight biobased thiols.62 Generally, it was found that high —SH and thermal radical catalyst concentrations, long reaction time, and nitrogen atmosphere favored faster consumption of the oil CC bonds and yielded higher molecular weight oligomers. This variety of soybean oil-based oligomeric thiols was used as a base of thiourethane coatings with either hexamethylene or isophorone isocyanate trimers. It should be noted that, although not as widely examined as the thiol-ene addition, thiol-isocyanate has also been demonstrated to proceed with some of the click chemistry characteristics.63 To the same end, Upshaw reported a different approach to prepare di-, tri-, and tetrathiols with high renewable content.64 In this case, terminal CC bonds of 9-decenoic and 10-undecenoic ester derivatives of cyclohexane diol, glycerol, trimethylol propane, pentaerythritol, or combinations were functionalized with thiol groups by thiol-ene addition of hydrogen sulfide (H2S) at 40 °C under UV light. Surprisingly, as is evidenced within this literature revision, the potential of H2S has not been exploited in the field of oleochemistry yet.
Importantly, aside from offering a facile way to oligomerize natural triglycerides, thiol-ene coupling is not, a priory, a good choice as crosslinking reaction of natural triglycerides. This is because, as discussed above, it is not highly efficient at low or stoichiometric ratios of thiol to CC because of the low reaction rates with internal alkenes. For example, Bantchev et al. when conducted 1-butanethiol (T13) addition to corn oil using low ratios (1.5–3) of thiol to ene determined degrees of hydrothiolation in the range of 18–55% by 1H NMR spectroscopy after 8 h of reaction.65 Generally, to overcome this problem, some authors proposed the introduction of readily reactive functional groups toward thiol-ene addition into triglyceride structure. As described by Hoyle et al., reactivity in radical thiol-ene reaction can vary considerably depending on the chemical structure of ene as well as thiol components.66 As such, vinyl ether, allyl ether, and acrylate functionalities were introduced to castor oil by Black and Rawlins67 through urethane linkages. Such macromonomers were cured with T17 via UV-initiated radical thiol-ene addition. Notably, the authors determined that vinyl ether and acrylate radical homopolymerizations are competitive processes under thiol-ene conditions. Cured films exhibited high solvent resistance and hardness as well as excellent adhesion and flexibility, regardless of the different macromonomer functionality evidencing a successful crosslinking. Acrylated castor oil also showed noteworthy reactivity in the presence of oligomeric silsesquioxane-containing thiol derivative (T19).68
Alternatively, our group recently demonstrated that maleated soybean oil triglycerides, prepared in two steps through glycerolysis of soybean oil followed by reaction with maleic anhydride, are reactive enough to afford elastomeric materials via UV (365 nm)/DMPA-initiated thiol-ene crosslinking.69 Although curing reactions were performed using stoichiometric thiol to CC ratios using multifunctional thiols, insoluble fractions as high as 89% after 12 h were reached. On the other hand, Webster and coworkers70 modified epoxidized soybean oil allyl alcohol in the presence of Lewis acid. Subsequently, thiol-ene-cured coatings were prepared using commercially available petroleum-based multifunctional thiols as well as soybean oil-based thiols prepared from such commercial thiols and epoxidized soybean oil by ring opening. Interestingly, the reaction between a thiol and an epoxy group has also been recently included into the realm of click chemistry.71 As such, in the field of oleochemistry, Erhan and coworkers72 already demonstrated that thiol addition to epoxidized soybean oil proceeds in an effective way under mild acid catalysis conditions. Thus, for instance, ring opening of epoxidized soybean oil with T13 was highly efficient (95% as determined by 1H NMR spectroscopic analysis), although competitive epoxy hydrolysis was also identified. Additionally, the distinguishing feature of this process is the formation of a reactive hydroxyl group upon coupling reaction. Because of the readily availability of epoxidized vegetable oils, there is no doubt that this reaction will be exploited in the near future.
THIOL-YNE ADDITION AND AZIDE–ALKYNE CYCLOADDITION IN OLEOCHEMISTRY
Alkynes are one of the most popular functional groups within the click chemistry paradigm. Terminal alkynes are, together with azides, the protagonists of the Cu(I)-catalyzed cycloaddition reaction to yield 1,4-triazoles (Scheme 1). This reaction, used by Sharpless and coworkers to introduce the click chemistry concept, is high yielding, selective, and devoid of side reactions but unfortunately, most alkynes have sluggish reaction rates and only proceed at temperatures higher than 100 °C in the absence of Cu(I) catalysts. On the other hand, alkynes are also well known for the recent rediscovery of their radical-mediated reaction with thiols, defined as thiol-yne addition, where two thiols are coupled to one alkyne using either a chemical radical source, UV irradiation, or sunlight at ambient temperature (Scheme 1).73–4 The beauty of this reaction is that it combines the readily available building blocks of the azide–alkyne and the thiol-ene reactions. Interestingly, unlike azide–alkyne cycloaddition, it does not need any potentially toxic metal catalyst. Moreover, it is normally much more efficient when carried out with equimolar amounts of reagents than thiol-ene addition, because intermediate vinyl radicals are formed in a virtually irreversible manner and moreover they abstract a hydrogen atom from the thiol reagent more rapidly than their alkyl counterparts.
Although both chemistries are well documented in the fields of chemistry, biology, and macromolecular science,75–78 they have been practically overlooked in oleochemistry for polymer/materials synthesis purposes because acetylenic fatty acids rarely occur in living organisms. Nevertheless, internal and terminal monoacetylenic fatty acids can be synthesized in acceptable yields, using a well-established bromation/dehydrobromination procedures, from oleic and 10-undecenoic acids, respectively.79, 80 Unfortunately, this procedure does not accomplish some of the green chemistry ideals, because it has an extremely low atom economy and uses toxic reagents. Alternatively, Shah and coworkers reported that alkynated plant oils are also accessible from the corresponding epoxidized derivatives taking advantage of the propensity of the epoxy rings to undergo ring-opening nucleophilic addition.81, 82 A real challenge for the near future is definitively to develop new methods (e.g., catalytic dehydrogenation of alkenes) to directly transform fatty acids into alkyne derivatives under sustainability criteria.
Thiol-Yne Addition of Alkyne-Derivatized Fatty Acids
In a polymer chemistry context, the application of thiol-yne coupling to oleochemicals is limited to only two contributions focused to transform alkyne-derivatized fatty acids into (a) polyurethane building blocks and (b) functional comb-like polymers. In both examples, the authors took advantage of the aforementioned ability of acetylene groups to accept rapidly two thiols under radical conditions.
In the search of biobased functional diols for polyurethane technology, our group applied thiol-yne addition to 10-undecynoic and stearolic acid methyl esters.83 Unlike thiol-ene addition to fatty acids, thiol-yne addition to their alkyne derivatives allows double primary hydroxyl functionalization without compromising carboxylic acid group. Reactions were investigated at room temperature using an excess of T1 (3 equiv relative to CC). Double thiol addition in the presence of DMPA was faster for the terminal CC reaching complete conversion in 5 min, whereas the absence of initiator extended the reaction to ∼60 min. For the internal yne, it required 60 min to be complete in the presence of DMPA, whereas reached a plateau at around 40% conversion without photoinitiator. Both synthesized diols, obtained in high yields as viscous oils, were combined with methylene diphenyl diisocyanate in DMF at 50 °C to yield thermoplastic polyurethanes containing pendant methyl ester groups [Fig. 4(A)]. Alternatively, those products were also used to prepare two fatty acid-derived triols suitable for the preparation of polyurethane networks by subsequent reduction of the ester groups. In another work,84 pendant methyl ester groups of the thermoplastic polyurethanes films were reacted with 1,6-hexamethylenediamine to functionalize film surface with amine groups, which were further used to ionically immobilize chondroitin sulfate as a bioactive molecule [Fig. 4(B)]. Contact angle measurements and cell culture experiments with osteoblastic cells demonstrated that chondroitin sulfate immobilization significantly increases surface wettability as well as osteoblast cytocompatibility in comparison with base polyurethanes [Fig. 4(C)].
Türünç and Meier85 investigated otherwise the direct polymerization of monoalkynes via thiol-yne addition. First, optimal polymerization conditions were determined using 1-octyne and octanethiol (T12), as model compounds, in the absence of initiator or using either thermally (AIBN) or photochemically (DMPA) radical initiation. Using an alkyne:thiol ratio of 1:2 UV-initiated reaction gave the best results. GC measurements determined that self-initiated addition at 80 °C yielded ∼70% of the bisaddition product after 2 h, whereas reached 90 and 99% in the presence of AIBN and DMPA, respectively. Expanding on these initial findings, the authors examined the ability to prepare a range of functional polymers using the radical thiol-yne addition using dithiols. The polymerization of several functional monoalkynes such as 10-undecynoic acid and propargylic acid was investigated showing that highly functional comb-like linear polymers are accessible via this approach. For example, polymerization of 10-undecynoic acid with T4 in the presence of DMPA at room temperature under UV irradiation yielded a 11.2 kDa polymer with PDI of 2.04.
Azide–Alkyne Cycloaddition of Azidated and Alkynated Triglycerides
Biswas et al. reported that azidated triglycerides are easily accessible via the reaction of the intermediate epoxide with sodium azide in water with 1-methyl imidazolium tetrafluoroborate ionic liquid as catalyst.86 This environmentally friendly methodology was successfully applied to epoxidized fatty acids and epoxidized soybean oil. For example, quantitative modification of epoxidized soybean oil was achieved in 2 days at 65 °C. Interestingly, this is a straightforward approach to convert triglycerides into clickable building blocks that can be used to be either modified or polymerized via the cycloaddition reaction between an azide and a terminal alkyne. To this end, Shah and coworkers reported that azidated natural oils such as castor, canola, corn, soybean, and linseed oils readily undergo Cu(I)-catalyzed polymerization at room temperature with diynes as well as alkynated soybean oil.81, 82, 87 Similarly, polymerization of alkynated soybean oil with diazide linkers resulted in a highly crosslinked polymers in moderate yields (65–85%). The major drawback of this approach is that long reaction times were necessary (48–72 h). Importantly, the authors also reported that these biobased monomers undergo more efficient and rapid polymerization in a catalyst-free and solvent-free environment at 100 °C, which represents an interesting green procedure to make polymers. Figure 5 shows how polymers obtained in the presence of Cu(I) catalyst were green in color even after extraction with THF several times, indicating the entrapment of copper salts in the polymer network, whereas polymers obtained thermally retained the color of the respective soybean oil-derived monomers. Generally, under these conditions, polymerizations were faster (12–24 h) and high yielding (>90%). These polymers exhibited behaviors ranging from soft rubbers to hard plastics.
DIELS–ALDER CYCLOADDITION IN OLEOCHEMISTRY
The Diels–Alder reaction is a [4+2] cycloaddition between an electron-rich diene and an electron-poor dienophile that has been established as one of the organic chemistry's classic reactions. This well-understood reaction allows an efficient synthesis of cyclic compounds with predictable regiocontrol and stereocontrol in the absence of metal catalysts. Moreover, the reaction is largely immune to solvent effects, and the reactants are typically nonreactive toward alcohols, amines, acids, carboxyl, and many other functional groups eliminating the need for protection/deprotection steps. Indeed, there is no doubt that Diels–Alder reaction fulfills many of the requirements of click chemistry philosophy.88 From the classical dimerization of linoleic acid to the more recent preparation of thermosetting resins from dehydrated castor oil,89 oleochemistry has exploited Diels–Alder reaction for different purposes although usually the click chemistry label cannot be used. For example, the extremely unreactive CC bonds of soybean oil were used during the norbornylization of the oil with cyclopentadiene. This process required harsh reaction conditions (240 °C and ∼200 psi).90 In addition to unfavorable reaction conditions, the aforementioned crosslinking of dehydrated castor oil with bismaleimides takes place via both Diels–Alder and ene reactions in addition to radical polymerization. A similar situation was observed during the modification of epoxidized soybean oil with pyridine derivatives and subsequent polymerization.91 It was not until recently that Gandini and coworkers44, 45 gave a push to an efficient application of furan/maleimide92 Diels–Alder reaction to oleochemistry. These particular studies clearly highlighted the fact that the furan/maleimide adduct formation dominates at temperatures up to 60 °C, whereas above 100 °C the uncoupling predominates, thus opening the way to promising properties in terms of thermoreversibility, mendability, and recyclability. The authors described the use of undecenyl compounds as suitable substrates for appending two furan heterocycles or a combination of a furan and a protected maleimide end group (Scheme 5).44, 45 Additional interest stems for exploiting furan derivatives that descend from renewable resources.
Synthesis was achieved in high yields using thiol-ene addition with T7 in conjunction with more classical chemical condensations. The synthesized AA (M45 and M46) and protected AB (M47) monomers were then polymerized at 65 °C via the Diels–Alder reaction, producing in both cases after several days low-molecular-weight materials (<10 kDa). Even more interesting are the results concerning the thermoreversible character of these polymeric systems. 1H NMR spectroscopy was used to demonstrate that retro-Diels–Alder reaction takes place heating up the polymers to 110 °C, regenerating the monomers that could in turn be polymerized again by heating at 65 °C.
The use of raw materials based on plant oils and fatty acids historically played an important role in the polymer chemistry. Although many different synthetic strategies have been proposed to transform oleochemicals into monomers and polymers, a more sustainable production of polymers from these renewable resources is still on the agenda. Polymer chemistry based on oleochemicals has recently realized of the click chemistry sound. In this line, the results presented within this contribution demonstrate that the impact of the click chemistry concept and, especially the renewed interest in efficient and simple chemical transformations, headed up by thiol-ene addition, has made an important breakthrough in the field. Even if some of the above presented strategies do not fulfill all the click chemistry requirements, they still can be classified as highly efficient transformations. Thus, irrespective of whether it is click or not, the marriage between efficient transformations and these natural resources will undoubtedly help to their exploitation in a sustainable fashion, and will bring us closer to the dreamed sustainable development.
Financial support from CICYT (Comisión Interministerial de Ciencia y Tecnología) (MAT2011–24823) is kindly acknowledged.
Gerard Lligadas was born in 1980 in Sitges (Spain) and studied chemistry at the University Rovira i Virgili (URV, Spain). In 2006, he obtained his Ph.D. under the supervision of Prof. Marina Galià and Prof. Juan C. Ronda. After postdoctoral research with Prof. Virgil Percec at the University of Pennsylvania (USA), he obtained a lecturer position at URV in 2008. His research interests lie in the field of polymer design and synthesis.
studied Chemistry at the University of Barcelona. In 1993, he obtained the PhD at the University Rovira i Virgili (URV). After two postdoctoral positions (1994 and 1996) at the Case Western Reserve University in the group of Prof. Virgil Percec, he got his Habilitation at the URV. Since 2011 he holds a Professorship in Organic Chemistry at the URV. His current research interest focuses in the synthesis of polymers starting from renewable resources using efficient synthetic methodologies.
Marina Galià received her undergraduate (BS 1987) and graduate education (PhD 1992) at the University of Barcelona. After a period of postdoctoral research with Prof. J.M.J. Fréchet at Cornell University, she began the academic career as Assistant Professor at the University Rovira i Virgili in 1995 and was promoted to Full Professor in 2010. Her current research focuses in sustainable polymer chemistry and involves the synthesis of functional polymers from renewable resources.
Virginia Cádiz received her BSc and PhD in Chemistry from the Complutense University of Madrid in 1975. This year she incorporated to the University of Barcelona (since 1992 University Rovira i Virgili). In 1992, she became Full Professor of Organic Chemistry and in 2004 was made Distinguished Professor. She is the founding head of the Polymer Group where she initiated, built up, and supervised significant research projects. Recently, her interest is in sustainable polymer chemistry: halogen-free flame-retardant thermosets, and polymers from vegetable oils, using efficient synthetic procedures.