Abstract
This review article is dedicated to special polysaccharide esters – the polysaccharide toluenesulfonic acid esters (tosylates) and polysaccharide carbonate esters. After describing the specifics of the synthesis, particular emphasis is placed on the use of polysaccharide tosylates and polysaccharide phenyl carbonates for subsequent modification by nucleophilic substitution (SN) reactions. For this purpose, the advantages and limitations of the respective derivatives are discussed with regard to their application in chemical modification with nucleophiles containing functional groups. A few functional polysaccharide derivatives and their properties are presented. Finally, reactive derivatives for click chemistry approaches are featured. These can be prepared starting from the reactive intermediate of either polysaccharide tosylate or polysaccharide phenyl carbonate.
Download PDF
Full Article
Clickable Polymers Accessible through Nucleophilic Substitution on Polysaccharides: A Sophisticated Route to Functional Polymers
Agnes Kemmer,a Haisong Qi,b,* and Thomas Heinze b,*
This review article is dedicated to special polysaccharide esters – the polysaccharide toluenesulfonic acid esters (tosylates) and polysaccharide carbonate esters. After describing the specifics of the synthesis, particular emphasis is placed on the use of polysaccharide tosylates and polysaccharide phenyl carbonates for subsequent modification by nucleophilic substitution (SN) reactions. For this purpose, the advantages and limitations of the respective derivatives are discussed with regard to their application in chemical modification with nucleophiles containing functional groups. A few functional polysaccharide derivatives and their properties are presented. Finally, reactive derivatives for click chemistry approaches are featured. These can be prepared starting from the reactive intermediate of either polysaccharide tosylate or polysaccharide phenyl carbonate.
DOI: 10.15376/biores.17.4.Kemmer
Keywords: Polysaccharide tosylate; Polysaccharide phenyl carbonate; Nucleophilic substitution; Reactive polysaccharide; Click chemistry
Contact information: a: Center of Excellence for Polysaccharide Research, Institute for Organic Chemistry and Macromolecular Chemistry, Friedrich-Schiller-University Jena, Humboldtstr. 10, D-07743 Jena, Germany; b: State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, 510641, China; *Corresponding authors: qihs@scut.edu.cn; thomas.heinze@uni-jena.de
INTRODUCTION
The chemical modification of polysaccharides remains the most important tool to obtain products with application-specific properties. Cellulose, as the most frequently used polysaccharide, offers access to broad variety commercial derivatives that are produced on a large scale. Cellulose ethers are produced by the Williamson ether synthesis with alkyl halides such as methyl- and ethyl chloride as well as by ring-opening reaction of ethylene- and propylene oxide as reagents. Prior to etherification, cellulose must be activated, which is done exclusively with aqueous sodium hydroxide in a concentration range from 15 to 50%. The reaction occurs under heterogeneous conditions. Both the aqueous alkaline and heterogeneous conditions are reasons for the limited number of commercially available cellulose ether products. Most of the cellulose ethers are water-soluble and applied in aqueous systems, as discussed in detail by Wuestenberg (2013).
Another important group of commercial cellulose derivatives is esters of carboxylic acids. Regarding the carboxylic acid ester of cellulose, the number of different products is also limited. The acetate, acetate propionate, acetate butyrate, and the acetate phthalate of cellulose are industrially manufactured.
Due to the conditions of esterification on a technical scale, carboxylic acids with more than 4 C atoms do not react efficiently with the biopolymer. These cellulose esters are soluble in organic solvents and find widespread applications (Edgar et al. 2001). Despite these commercial products from cellulose, it is important to develop new methods for chemical modification. With the increasing complexity of applications, especially in the biomedical context, the requirements for the synthesis of polysaccharide derivatives are also becoming more demanding. The robust synthesis of products should be feasible, selective, and easily adaptable by changing parameters such as solvent and functionality.
Although the physico-chemical properties of the polysaccharides are different, the modification can be adapted to other backbones of neutral polysaccharides carrying hydroxyl groups (Cumpstey 2013). This is particularly important to ensure that the requirements of applicability are achieved and that the range of diversity in structure and source of the polysaccharides is exploited. Bacterial dextran, which is popular in the biomedical field, but also other neutral polysaccharides, such as pullulan from fungi or pure xylan from wood, can be attractive starting materials. Chemical modification of polysaccharides based on click chemistry can be such a strategy as highlighted in the review of Meng and Edgar (2016). Although some derivatives already exist, it is still important to better implement the diversity of these reactions in the field of polysaccharides.
Special cases of previous mentioned esters of polysaccharides are sulfonic acid esters and carbonate esters of cellulose, which are not applied directly as commercial products, but are used as a platform for the synthesis of subsequent products (Trivedi and Fardim 2019).
The main part of the feature article will deal with the use of, in particular, p-toluene sulfonate (tosylate) and phenyl carbonate, as a reactive platform and the different opportunities of nucleophilic substitution (SN) reactions with nucleophiles (inverse reactivity than hydroxyl group), the products accessible, their properties, and applications. The advantages (achievable regioselectivity, complete conversion or derivatives with mixed reactive moieties) and disadvantages (e.g., limited reactivity) of each reactive platform are discussed. Some examples of “clickable” derivatives, starting from cellulose and extended to other polysaccharides, obtained by the SN reactions will also be highlighted. It must be mentioned that many issues discussed resulted from the authors’ own studies.
POLYSACCHARIDE p-TOLUENE SULFONIC ACID ESTER (POLYSACCHARIDE TOSYLATE)
Synthesis of Polysaccharide tosylates
Although the synthesis of cellulose tosylates (Fig. 1) was first realized heterogeneously by reaction of cellulose with p-toluene sulfonic acid chlorides (tosyl chloride) in aqueous alkaline media (NaOH, Schotten-Baumann reaction), homogeneous procedures are preferred to obtain pure products with controlled degree of substitution (DS) and distribution of substituents (Heinze et al. 2006a).
Fig. 1. Reaction scheme for homogeneous tosylation of cellulose
Thus, the synthesis of cellulose tosylate is carried out by reacting the biopolymer with tosyl chloride, applying the efficient cellulose solvent N,N-dimethylacetamide (DMAc) in combination with LiCl. As early as 1986, McCormick and Callais prepared pure cellulose tosylate (McCormick and Callais 1987). As summarized in Table 1, products with a DS as high as 2.4 could be obtained. In other systematic studies, it turned out that the reaction efficiency decreases with increasing degree of polymerization (DP) of the cellulose sample (cellulose powder, DP 280; spruce sulfite pulp, DP 330 and 650; cotton linters, DP 850; beech sulfite pulp, DP 1020, and bacterial cellulose, DP 5100). Moreover, the DS depends on the molar ratio cellulose to tosyl chloride. The highest DS value reached was 2.3, by applying pulp with DP of 330 and a molar excess of 4.5 moles tosyl chloride (Rahn et al. 1996). A complete tosylation of the cellulose was not possible even at higher molar excess of reagent. The reason for this upper limit DS of about 2.4 is still unknown. Unlike other sulfonates like mesylate, brosylate, and triflate, the polysaccharide tosylate is a very stable derivative (Siegmund and Klemm 2002) and the limitation of the DS could be attributed to the size of the sulfonate and the associated steric hindrance at the polysaccharide coil. In addition, a base must be applied in the tosylation reaction to bind the hydrogen chloride formed during the reaction. Pyridine was used in previous studies, and it leads to side reactions, mainly to the formation of dexoychloro and acetate moieties. The most appropriate base is trimethylamine to exclude this side reaction and to apply reaction temperature up to room temperature.
Recently, other solvents were studied as reaction media for the homogeneous tosylation of cellulose. Aqueous sodium hydroxide (Elchinger et al. 2012) and NaOH-urea aqueous system (Schmidt et al. 2014) were applied as “green” solvents. The tosyl cellulose was capable of being obtained. Compared to the reaction in the water-free DMAc/LiCl solvent, a significantly higher molar excess of reagent to cellulose is needed to achieve comparable DS values because of the partial hydrolysis of the tosyl chloride in the aqueous systems.
Ionic liquids (ILs) are powerful solvents for cellulose (El Seoud et al. 2007). The tosylation of cellulose in 1-allyl-3-methylimidazolium chloride (AMIM Cl) in the presence of pyridine as a base at 10 °C yields no reproducible results (Granström et al. 2008). The disadvantage of this solvent is its high viscosity, which leads to insufficient stirring and mixing of the reaction mixture, so that the products tend to be similar to those of heterogeneous reactions. Applying AMIM Cl or 1-ethyl-3-methylimidazolium diethyl phosphate (EMIM DEP) in combination with pyridine or organic solvents, such as N,N-dimethylformamide (DMF) or 1,3-dimethyl-2-imidazolidinoneem, leads to homogeneous conditions and consequently to tosylation predominantly at the primary hydroxyl group of C6 (Gericke et al. 2012). The organic co-solvent decreases the viscosity of the reaction systems significantly and guarantees a mixing of the reaction components. These synthesis paths are also not preferred due to the high price of the ILs and the lack of sufficient recycling concept for ILs. It was unexpected that the conversion of cellulose dissolved in 1-ethyl-3-methylimidazolium acetate (EMIM Ac) with tosyl chloride resulted in the formation of pure cellulose acetate. This result is explained by a side reaction of the IL solvent EMIM Ac with tosyl chloride, forming the mixed acetic acid-toluene sulfonic anhydride, which was demonstrated by nuclear magnetic resonance (NMR) spectroscopic studies (Köhler et al. 2007). Subsequent reaction of cellulose with the mixed anhydride occurs on the acetyl moiety. Therefore, ILs with less nucleophilic anions must generally be used in the reaction of acid chlorides.
Table 1. Overview of Solvent/Base Systems for Tosylation of Cellulose
The homogeneous synthesis of tosylate in DMAc/LiCl can be transferred to other polysaccharides in addition to the most commonly used cellulose. The preparation of tosylates was investigated for polysaccharides as α-1,4-linked starch (Dicke et al. 2001), pullulan (α-1,4; α-1,4; α-1,6), dextran (α-1,6), lichenan (β-1,4 and β-1,3) (Koschella et al. 2006), and other 1,3-linked glucans (Koschella et al. 2020a), as well as agarose (Gericke and Heinze 2015). Although the efficiency of the reaction is sufficient for these polysaccharides and the efficiency is often comparable to the conversion with cellulose, the regioselectivity observed differs for the polysaccharides. It was found that the regioselectivity is connected to the type of glycosidic linkage of the polysaccharide. For cellulose the reactivity of hydroxyl group 6>>2>3 occurs. For agarose, the primary hydroxyl groups possess also the highest reactivity with tosyl chloride. In contrast, starch reacts preferentially at position 2. The α-1,3-linked glucan also shows this reactivity, while the preference is not as pronounced as with starch. Moreover, dextran, which consists predominantly of a main chain of α-1,6-linked glucose, tosylation could be realized. However, the product is not well appropriate for SN reactions (see limitations).
The synthesis of polysaccharide tosylates is well explored and there is not much scope for optimization. A more sustainable aqueous medium leads to lower efficiency and therefore does not hold promise. The conditions have been extensively studied, and mild conditions, such as low temperatures, have a positive effect on the efficiency of the reaction and reduce side reactions. By upscaling the process, the amount of solvent DMAc/LiCl can be reduced without significantly decreasing the DS values, while keeping the processing manageable. The solvent for the isolation of the product by precipitation can be recycled to a large extent.
NUCLEOPHILIC DISPLACEMENT REACTIONS WITH POLYSACCHARIDE TOSYLATE
Advantages
An advantage of the introduction of tosylate groups into the polysaccharide backbone is the improvement of solubility in organic solvents, especially for cellulose, which is insoluble in conventional solvents in its unmodified form (Koschella et al. 2006). The stability of the tosyl group is considered as a further advantage; it is stable for further modification of the remaining hydroxyl groups before the reactive group is reacted with nucleophiles such as azide or amines (Fig. 2). To modify the properties of cellulose-p-toluene sulfonic acid ester and hence of the final product obtained by SN reaction, the remaining hydroxyl groups of cellulose tosylate may be modified prior to the SN reaction. The tosylate groups are stable under the usual conditions of esterification. Thus, cellulose tosylates were converted into mixed cellulose tosylate/carboxylates (Heinze et al. 1996), tosylate/urethanes (Tiller et al. 2001), and tosylate/sulfate (Heinze and Rahn 1996). Even a complete functionalization of all remaining hydroxyl groups was achieved with acetic- and propionic acid anhydride. The products are readily soluble in dipolar aprotic solvents, such as acetone, dimethyl sulfoxide (DMSO), N,N-dimethylacetaminde (DMAc), and tetrahydrofuran (THF), and can be cast as films from solution. Mixed cellulose esters that carry the tosylate and monoester of a dicarboxylic acid (e.g., tosylate/phthalate or succinate) are soluble in aqueous NaOH and even in water, depending on the ratio of the DS values. Cellulose tosylate sulfate shows self-aggregation in water due to the amphiphilic character of the derivative (Rahn et al. 1996). The hydrophobic character of the tosylate in combination with the repulsive electrostatic force of the sulfate leads to aggregation in aqueous media.
Fig. 2. Chemical modification of cellulose tosylate and preparation of different aminocelluloses
The main advantage of polysaccharide tosylates is the high reactivity with nucleophiles. Nucleophiles are bound to the polymer backbone by nucleophilic substitution. For example, aminolysis works well for cellulose, whereby aminocellulose with terminal amino groups can be obtained without protection and side reactions (Fig. 2). As the structurally simplest compound (Fig. 2, without linker), hydrazine hydrate can be taken as nucleophile (Nazir et al. 2021). Formally, there is also a terminal NH2 group in the product, although its reactivity is not the same as that of the diamines. Even tertiary diamines react sufficient with the tosylate of the primary hydroxyl group of cellulose and also asymmetric diamines can be applied for synthesis. The terminal tertiary amino moiety of the obtained derivative could be further converted in a ring-opening reaction to yield sulfobetainic polyampholytes based on cellulose (Pfeifer et al. 2022). For cellulose, a regioselective functionalization at the primary hydroxyl group at C6 is achievable up to a DS of 0.8. Consequently, it is possible to substitute all the tosylate groups by nucleophiles such as sodium azide or functional amines, forming a product with a defined structure and substitution pattern. Even pure 6-deoxyazido- or 6-deoxyamino agarose up to a DS of 0.6 could be synthesized by this route using sodium azide or ethylenediamine as nucleophile (Gericke and Heinze 2015). The conversion of polysaccharide tosylates with sodium azide was completed not only for cellulose (see section clickable polysaccharide derivatives), but also for dextran-, pullulan-, and lichenan tosylates, yielding well soluble samples of mixed tosylate-deoxyazido derivatives that are appropriate for the click chemistry approach.
Limitations
In the synthesis of polysaccharide tosylate, halogenation of the hydroxyl group is a possible side reaction. However, the reaction conditions of the homogeneous reaction in DMAc/LiCl can be controlled to avoid the deoxychloro group as a by-product. Thus, it is possible to obtain a pure polysaccharide tosylate as a reactive starting derivative for nucleophilic substitution reactions. The tosylation of various polysaccharides such as dextran, pullulan, lichenin, starch, and other glucans is well possible in DMAc/LiCl (see section synthesis of polysaccharide tosylates). However, they can only be used to a limited extent for subsequent chemical modification by nucleophilic displacement reactions. The highest rate of nucleophilic substitution reaction is observed for the tosylate of the primary hydroxyl group. For the conversion of dextran tosylate, which has no primary hydroxyl group in the main chain, a low DS of deoxyazido of 0.6 with a remaining DS of 0.7 could be synthesized. Up to now, the selective cleavage of remaining tosylate groups of a deoxyamino- or deoxyazido derivative is not reported.
CELLULOSE CARBONATES
Synthesis of Polysaccharide Carbonates
Heterogeneous conversion of cellulose is again less appropriate for the cellulose carbonate synthesis; for instance, the conversion of cellulose with ethyl chloroformate in DMSO (a non-solvent for cellulose) leads to non-uniform products containing both the trans-2,3-cyclic carbonate structure and the acyclic O-ethoxycarbonyl moiety (Barker et al. 1997).
In contrast, homogeneous reaction in DMA/LiCl yields the corresponding acyclic cellulose carbonates as uniform and soluble products with defined DS values. For the esterification of cellulose in DMA/LiCl, chloroformates and pyridine as base are most suitable (Elschner et al. 2013a). Applying an equimolar amount of repeating unit to phenylchloroformate, a DS of 0.84 was achieved. Increasing the amount of phenyl chloroformate to 10 equivalents resulted in an increase in DS to 1.98. Although electronic aspects (smaller +I effect) should favor a reaction of the primary hydroxyl groups, this regioselective reactivity cannot be observed. In the solvent DMAc/LiCl highly functionalized cellulose phenyl carbonates are difficult to synthesize and require high molar ratios because chloroformate reacts with DMAc and is thereby consumed.
The reaction of cellulose with p-NO2-phenyl chloroformate (molar ratio of 1:3) yields a product with a DS value of 1.34, which shows gel formation due to the crosslinking by intermolecular carbonate moieties. A comparable crosslinking was observed for cellulose phenyl carbonate with low DS (molar ratio 1:0.5). The crosslinking is based on primary hydroxyl groups due to the smaller +I effect and hence the increased reactivity of the carbonyl group, compared to carbonates derived from secondary hydroxyl groups. Thus, the remaining hydroxyl groups of cellulose may react despite their low nucleophilicity. The crosslinking of cellulose phenyl carbonates with low DS is a result of the high number of non-derivatized hydroxyl groups that are accessible for intra- and intermolecular carbonate formation. This hypothesis was proven by conversion of 6-O-trityl cellulose (Ganske and Heinze 2018). In this context, it is interesting to note that xylan, a hemicellulose consisting of β-1,4-glycosidically linked xylose, does not show any crosslinking during the formation of the corresponding phenyl carbonates independent of the DS. Both secondary hydroxyl groups present in the biopolymer could be converted with phenyl chloroformate either in dipolar aprotic solvents (DMAc, NMP) with LiCl or in an ionic liquid to products with DS of up to 2 (Fig. 3 structure characterization; Fig. 8 reaction scheme). Further, these xylan phenyl carbonates with high DS could be processed into stable reactive nanoparticles by precipitation approaches (Fig. 3). Unexpectedly, the hydroxyl group at position 3 is the most reactive, while substitution in position 2 only occurred if the neighboring position 3 is already modified (Gericke et al. 2018). The reason for this special reactivity is still unknown. Additionally, for the polysaccharides (starch, pullulan, and dextran), high DS values up 2.0 and 2.9 can be obtained (Elschner et al. 2013a).
Ionic liquids are efficient reaction media for the carbonate synthesis (Fig. 8, reaction of cellulose) because they are inert, i.e., they do not react with rather reactive chloroformates. Cellulose, dissolved in 1-butyl-3-methylimidazolium chloride (BMIM Cl) and diluted with pyridine to decrease the viscosity of the system, was allowed to react with phenyl chloroformate at different molar ratios, yielding cellulose phenyl carbonates with controlled DS and even complete functionalization (DS = 3). The homopolymer gave well-resolved NMR spectra. Under mild reaction conditions no biopolymer degradation occurred. A DS of 2.83 at a molar ratio of 3:1 could be obtained using the IL as reaction medium, while the DS was only 1.49 when applying DMA/LiCl as solvent. The efficiency of the conversion increased from 50 to 94%, i.e., the conversion occured almost quantitatively. The molar ratio of 5:1 yields a sample with DS of 3 whereas the DS is 1.75 in case of DMA/LiCl. Even at high molar ratios (10:1), the DS of 1.98 only could be realized in the tertiary amide solvent (Elschner et al. 2013b). Thus, exchanging the polar aprotic solvent by the ionic liquid BMIMCl is a good choice as a solvent to optimise polysaccharide-phenyl carbonate synthesis. The efficiency of the process (especially for high DS values/ complete modification) can be increased and the reagents and by-products reduced.
Fig. 3. 13C NMR spectra (DMSO-d6) of xylan (a) and xylan phenyl carbonate (XPhC) with rising degree of substitution of 0.46 (b), 1.17 (c), and 1.98 (d); scanning electron microscopy image of reactive nanoparticles from XPhC, DS = 1.54 (e); TFA = trifluoroacetic acid
CHEMICAL MODIFICATION OF CELLULOSE PHENYL CARBONATES
Advantages
Polysaccharide carbonates in general and polysaccharide phenyl carbonates in particular are promising reactive biopolymer derivatives for the design of advanced products and materials based on the most important renewable resource cellulose by conversions with nucleophilic compounds. Thus, anionic, cationic, and ampholytic polymers are accessible by advanced organic chemistry, as shown in Fig. 5.
Research revealed that the chemical modification of cellulose phenyl carbonates can be made with different nucleophiles in one-step reactions or stepwise. For nucleophiles with more than one reactive site with comparable reactivity, only one reactive moiety should be available; the other ones must be protected to exclude crosslinking and hence insoluble products. For amino-functions, typical protecting groups such as alkyloxycarbonyl moieties are well appropriate. Moreover, p-aminobezylamin yields organo-soluble p-aminobenzylcellulose carbamate (Ganske et al. 2016). Additionally, ethylene diamine moieties were introduced into the biopolymer via the protected intermediate as discussed above. The ω-aminoethylcellulose carbamate itself possesses a strong activity against Candida albicans and Staphylococcus aureus that is improved by p-aminobenzylamine substituent. Moreover, the mixed cellulose carbamate exhibits a high biocompatibility and forms films on cotton and PES that exhibit a strong activity against Staphylococcus aureus and Klebsiella pneumonia.
Even regioselective functionalization is possible, which is nicely documented by novel zwitterionic biopolymer derivatives of cellulose, i.e., the positively charged ammonium groups present at position 6 while the negative charge of carboxylate is located at positions 2 and 3 of the repeating unit (Fig. 5). This uniform structure of the novel polymeric zwitterion could be realized by reaction of cellulose with 1.3 eq. phenyl chloroformate at 0 °C in DMAc/LiCl to induce selective modification of the primary OH at position 6. The subsequent aminolysis of the carbonate takes place at the activated moiety only, yielding 6-O-(N-Boc-2-aminoethyl) cellulose carbamate. Subsequently, phenyl chloroformate was allowed to react with the 6-O-(N-Boc-2-aminoethyl) cellulose carbamate; the product was treated with β-alanine ethyl ester to introduce the latent anionic moiety. Finally, deprotection leads to the zwitterionic product (Elschner et al. 2016). The antimicrobial activity of a coating based on this zwitterionic biopolymer derivative was demonstrated by confocal laser scanning microscopy and live/dead staining. As support material, glass, titanium, or tissue culture poly(styrene) were used. The antimicrobial activity was influenced by the support material. Additionally, for the polysaccharide dextran a regioselective functionalization is possible without protecting any hydroxyl group. Using the p-nitrophenyl carbonate of dextran, a regioselective substitution at C2 hydroxyl group with a DS of 1.0 is observed. Also for dextran a complete modification with maleimido group could be achieved starting from the dextran furfuryl carbamate. Processing the derivative into reactive films by spin-coating allowed the immobilization of protein (Fig. 4) (Elschner et al. 2017).
Unlike the polysaccharide tosylates, the subsequent modification of polysaccharide phenyl carbonate by aminolysis is not limited to a distinct group of polysaccharides. The reaction with functional amines is well transferrable. Further, not only the synthesis of cellulose phenyl carbonates works successfully for other polysaccharides than cellulose, but also the subsequent aminolysis is efficient (see section clickable polysaccharide derivatives).
Fig. 4. Dextran furfuryl carbamate with a DS = 3.0, subsequent “click” reaction to obtain a maleimido-modified derivative (polysaccharide backbone and linker structure in blue), and processing of the maleimido derivative into reactive films for immobilization of proteins.
Limitations
Other than the undesired crosslinking of the carbonate during synthesis (see section synthesis of polysaccharide carbonates), there is another path of crosslinking of cellulose carbonate of low DS (0.3) by aminolysis. It became obvious that the primary carbonate is more reactive compared to the secondary one; even remaining hydroxyl groups may undergo SN reaction and hence crosslinking. Nevertheless, there are strategies to overcome this undesired side effect. The crosslinking reaction can be avoided using protected 6-O-triphenylmethyl cellulose for the carbonate synthesis, on the one hand. On the other hand, the reactivity and stability of the cellulose carbonates could be controlled by substituents in the p-position of the phenyl ring, such as methoxy-, nitro-, or chloro group. Thus, cellulose phenyl carbonates with low DS of 0.3 were also accessible (Ganske and Heinze 2018). While the p-NO2-substituent leads to regioselective substitution for dextran, and can be used successfully for other polysaccharides that only have secondary hydroxyl groups, such as xylan, it can only be used to a limited extent with a protective group for polysaccharides with a primary OH group. An example of the necessary use of the p-nitrophenyl carbonate of dextran is the synthesis of furfuryl modified dextran. Here the more reactive p-nitrophenyl carbonate reacted with furfuryl amine to form the corresponding carbamate (Elschner et al. 2017).
Fig. 5. Reaction scheme for the synthesis of cellulose carbamates with anionic, cationic, and ampholytic moieties (adapted from Elschner and Heinze 2014)
As mentioned, nucleophiles with two or more aliphatic amino moieties lead to crosslinking, while aromatic amino moieties do not possess sufficient nucleophilicity to be able to react with cellulose phenyl carbonate. Thus, amine protecting groups are necessary (Fig. 5) in contrast to the use of the polysaccharide tosylate derivatives, which can be reacted with unprotected diamines without crosslinking (Fig. 2).
Clickable Polysaccharide Derivatives Obtained from Tosylates and Carbonates
Further, the nucleophilic substitution reaction of polysaccharide tosylates and –phenyl carbonates can be used for the synthesis of derivatives that can be exploited for click chemistry approaches (Fig. 6). The use of clickable groups has a great influence on the sustainability of the process. The reactions mentioned, such as the Diels-Alder reaction with maleimide, the azide-alkyne reaction of the corresponding modified polysaccharide derivatives, and the polysaccharide allyl/ norbornene derivative for the thiol-ene reaction, all have a high atom economy. The reactants (with the exception of the catalytic amounts of initiator) are completely incorporated into the product. The reactions can also be carried out in aqueous media under mild conditions and with appropriate previous functionalization of the polysaccharides.
Fig. 6. General scheme for cellulose tosylate and phenyl carbonate as reactive intermediates towards reactive “clickable” derivatives
An important type of click chemistry is the Huisgen reaction azide–alkyne cycloaddition. The reaction can be exploited for the preparation of various cellulose derivatives (Abdellatif et al. 2015). Because of the catalysis by a Cu(I) species (commonly copper(II) sulfate and sodium ascorbate) the conversion can be completed efficiently under mild conditions and even in aqueous media. The two reactive groups needed for the Huisgen reaction are an azide- and an alkyne moiety that can be easily introduced by SN reaction with cellulose tosylate. The azide anion is one of the strongest nucleophiles and reacts well with cellulose tosylate to regioselectively yield 6-deoxy-6-azide cellulose, i.e., the primary tosylates are substituted only (Heinze et al. 2006b). The alkyne containing cellulose derivative may be obtained from cellulose tosylate by SN reaction with propargylamine (Koschella et al. 2011). Both polymers may react with each other yielding hydrogels. The activated cellulose derivatives can be functionalized to a myriad of novel biopolymers with organic compounds containing azide- and alkyne moieties. Dendron-like moieties could be attached to cellulose (Montanez et al. 2011). Either the azide (Heinze et al. 2008) or the alkyne moiety was attached to the complementarily modified polymer backbone (Pohl and Heinze 2008).
Oligosaccharides such as O-linked β-maltoside and β-lactoside containing terminal alkyne could be linked to 6-azido-6-deoxycellulose (Negishi et al. 2011). This approach was successfully applied for the preparation of novel all sugar-based cellulose derivatives (Fig. 7). The clickable groups attached to the cellulose (azide- or alkyne moiety) were allowed to react with modified sugar compounds containing the corresponding reactive group. With the modified sugars, up to 100% of the reactive moieties on the polymer backbone could be converted, whereby the conversion rate is dependent on the steric demand of the components (Koschella et al. 2020b).
One of the first papers to discuss clickable chemistry with cellulose was published in 2006 (Liebert et al. 2006). 6-Azido-6-deoxy cellulose obtained by SN reaction of cellulose tosylate with sodium azide was allowed to react with methyl propiolate, 2-ethynylaniline, and 3-ethynylthiophene, yielding the corresponding cellulose derivatives. The substituents are attached via the 1,4-disubstituted 1,2,3-triazol linker. The 1,3-dipolar cycloaddition catalyzed with copper(I) (copper(II) sulfate pentahydrate/sodium ascorbate) occurs regioselectively regarding the 1,2,3-triazole and under mild reaction conditions. Moreover, the distribution of the substituents is also limited to the primary hydroxyl group at the C6-atom because of the regioselective introduction of the azide moieties.
Not only organo-soluble 6-azido-6-deoxy cellulose but also water-soluble derivatives are known. The biopolymer derivative can be transferred to a water-soluble one by modification of the remaining OH group at the secondary positons 2 and 3 of the AGU with ionic moieties. A cationic moiety 4-(N,N,N-trimethylammonium)butyrate chloride can be attached to the 6-azido-6-deoxy cellulose to result a water-soluble product (DS of cationic group up to 0.24). Alternatively, etherification with 4-bromobutyltrimethyl-ammonium bromide yielded 6-azido-6-deoxy-2,3-O-(4-trimethylammonium)butyl cellulose bromide with a similar DS. These polymers are reactive and can be modified via azide–alkyne click chemistry in aqueous media. These polymers are reactive and can be modified via azide–alkyne click chemistry (Bretschneider et al. 2015).
Fig. 7. Reaction scheme for the preparation of sugar modified cellulose derivatives by Huisgen click reaction (Cu(I) catalyzed azide–alkyne 1,3-dipolar cycloaddition); Me = methyl
The simple heterogeneous carboxymethylation of 6-azido-6-deoxy cellulose applying 2-propanol/aqueous NaOH as slurry medium represents another path to obtain water-soluble precursors for the azide–alkyne click chemistry (Pohl et al. 2009). It is important to note that under the drastic alkaline conditions needed for the etherification, the azido moieties are stable. Chemoselective dendronization could be achieved by copper-catalyzed azide-alkyne reaction, yielding water-soluble carboxymethyl-6-deoxy-(1-N-[1,2,3-triazolo]-4-polyamidoamine) cellulose (first generation with DS = 0.51, second generation with DS = 0.44, and third generation with DS = 0.39). The dendronized cellulose possesses a persistence length of 2.8 to 4.0 nm without any change in flexibility, as could be concluded from conformation zoning. A comparable approach was used to get the water-soluble carboxymethyl-6-deoxy-6-azdio cellulose for crosslinking with 6-deoxy-6-aminopropargyl cellulose in water (Koschella et al. 2011).
Another approach, which deals with the preparation of a clickable derivative using nucleophilic substitution on the polysaccharide, was reported for the synthesis of dextran furfuryl carbamate. After synthesis of the derivative, the Diels-Alder reaction was completed successfully with a bismaleimide linker to obtain reactive films (Fig. 4). These films could be applied for the surface immobilization of the protein bovine serum albumin (Elschner et al. 2017).
To make thiol-ene reaction better accessible for highly modified polysaccharides, the activated starting material cellulose phenyl carbonates with DS from 2 to 3 were allowed to react with allylamine resulting in the clickable cellulose allyl carbamate with high DS of up to 2.2 (Fig. 8). Pure products without remaining phenyl carbonate substituents were accessible, but also mixed products with remaining phenyl carbonate moieties (Lindemann and Heinze 2022). The phenyl carbonate groups can be used for modification with further functional moieties, for example to adjust the physico-chemical properties. In addition to other propene-containing cellulose derivatives in the form of allyl ether or acrylate units, this path is an important alternative that despite the two-step synthesis, is highly efficient because a higher content of reactive C=C bonds attached to the polymer backbone is achieved. The high reactivity of these C=C double bonds with thiols could be demonstrated in a model reaction with the cationic thiol N,N-diethylamine ethanethiol hydrochloride. This makes the cellulose allyl carbamate into an appealing carrier polymer. A comparable chemistry can be carried out with propargylamine and a functional amine with terminal azido moiety as well (Fig. 8). The polysaccharide phenyl carbonate reacts readily with propargylamine under formation of the corresponding carbamate containing the reactive triple bond. In a similar way, the corresponding azide functionality could be attached to the polysaccharide (Skodda 2022).
Fig. 8. Reaction scheme for the synthesis of clickable cellulose/xylan carbamates by the conversion of cellulose phenyl carbonate with amines; BMIM Cl = 1-butyl-3-methylimidazolium chloride
CONCLUSIONS
It became obvious that the synthesis paths discussed by conversion of the polysaccharide to the polysaccharide tosylate or polysaccharide phenyl carbonate as reactive intermediates and the functionalization by nucleophilic substitution is a promising approach useful for chemical modification of polysaccharides that contain hydroxyl groups, such as neutral polysaccharides. The advantages of this chemistry clearly outweigh the limitations. It is obvious that the structural diversity of the chemically modified derivatives is enhanced. To make optimal use of the advantages of the nucleophilic substitution of the respective derivative, recommendations for the synthesis can be given. For the synthesis of non-natural polysaccharides with amino groups, the reactive polysaccharide tosylate provides the starting compound of choice, even though compounds with terminal amino groups also could be produced using polysaccharide carbonates. In contrast, polysaccharide phenyl carbonates can be used for screening the optimal composition of the polymer, as in a modular synthesis approach different amines can be used in a one-pot reaction, e.g., using a functional group and a solubility-mediating group. Functional polymers with different properties can be obtained by varying the molar ratio of both substituents alone. It may be expected that nucleophilic displacement reaction with polysaccharide will be increasingly applied for the design of advanced polysaccharide derivatives.
Although the synthesis method presented here is convincing, it is not suitable for optimizing already established technical processes. The multi-step synthesis with the reactive intermediate tosylate or phenyl carbonate formally adds a step, but the efficiencies are very high, and the conversion rates on average are higher than with a direct modification of the polysaccharide. The reactive intermediates can be used primarily to bind various reactive groups to the polymer backbone in a targeted and selective manner. This can be particularly advantageous for applications in the biomedical/ pharmaceutical field, where different functionalities (drug, dye, targeting group, solubility mediator, crosslinker) need to be introduced. The modular approach is also advantageous in this field, as many nucleophiles are suitable for the incorporation of the clickable groups, which can be easily adapted. The click reaction can be carried out with a variety of structures for the final functional moiety. The combination possibilities can contribute to creating a polymer library from the same reactive starting material. This step of the process can be easily adapted without the need for extensive optimization of reaction conditions.
REFERENCES CITED
Abdellatif, F. H. H., Babin, J., Arnal-Herault, C., and Jonquieres, A. (2015). “Grafting of cellulose and cellulose derivatives by CuAAC click chemistry,” in Cellulose-Based Graft Copolymers, V. K. Thakur (Ed.), CRC Press, Boca Raton, FL, USA, pp. 518-539. DOI: 10.1201/b18390-26
Barker, S. A., Tun, S. C., Doss, S. H., Gray, C. J., and Kennedy, J. F. (1997). “Preparation of cellulose carbonate,” Carbohydrate Research 17(2), 471-474. DOI: 10.1016/S0008-6215(00)82559-8
Bretschneider, L., Koschella, A., and Heinze, T. (2015). “Cationically modified 6-deoxy-6-azido cellulose as a water-soluble and reactive biopolymer derivative,” Polymer Bulletin 72, 473-485. DOI: 10.1007/s00289-014-1281-4
Cumpstey, I. (2013). “Chemical modification of polysaccharides,” ISRN Organic Chemistry 2013, Article ID 417672. DOI: 10.1155/2013/417672
Dicke, R., Rahn, K., Haack, V., and Heinze, T. (2001). “α-1,4-, Starch derivatives of high degree of functionalization. Part 2. Determination of the functionalization pattern of p-toluenesulfonyl starch by peracylation and NMR spectroscopy,” Carbohydrate Polymers 45(1), 43-51. DOI: 10.1016/S0144-8617(00)00219-8
Edgar, K. J., Buchanan, C. M., Debenham, J. S., Rundquist, P. A., Seiler, B. D., Shelton, M. C., and Tindall, D. (2001). “Advances in cellulose ester performance and application,” Progress in Polymer Science 26(9), 1605-1688. DOI: 10.1016/S0079-6700(01)00027-2
Elchinger, P.-H., Faugeras, P.-A., Zerrouki, C., Montplasir, D., Brouillette, F., and Zerrouki, R. (2012). “Tosylcellulose synthesized in aqueous medium,” Green Chemistry 14(11), 3126-3131. DOI: 10.1039/c2gc35592b
Elschner, T., Ganske, K., and Heinze, T. (2013a). “Synthesis and aminolysis of polysaccharide carbonates,” Cellulose 20(1), 339-353. DOI: 10.1007/s10570-012-9819-9
Elschner, T., Kötteritzsch, M., and Heinze, T. (2013b). “Synthesis of cellulose tricarbonates in 1-butyl-3-methylimidazolium chloride/pyridine,” Macromolecular Bioscience 14(2), 161-165. DOI: 10.1002/mabi.201300345
Elschner, T., and Heinze, T. (2014). “A promising cellulose-based polyzwitterion with pH-sensitive charges,” Journal of Organic Chemistry 10(1), 1549-1556. DOI: 10.3762/bjoc.10.159
Elschner, T., and Heinze, T. (2015). “Cellulose carbonates: A platform for promising biopolymer derivatives with multifunctional capabilities,” Macromolecular Bioscience 15(6), 735-746. DOI: 10.1002/mabi.201400521
Elschner, T., Lüdecke, C., Kalden, D., Roth, M., Löffler, B., Jandt, K. D., and Heinze, T. (2016). “Zwitterionic cellulose carbamate with regioselective substitution pattern – A coating material possessing antimicrobial activity,” Macromolecular Bioscience 16(4), 522-534. DOI: 10.1002/mabi.201500349
Elschner, T., Obst, F., Stana-Kleinschek, K., Kargl, R., and Heinze, T. (2017). “Synthesis and film formation of furfuryl- and maleimido carbonic acid derivatives of dextran,” Carbohydrate Polymers 161, 1-9. DOI: 10.1016/j.carbpol.2016.12.038
El Seoud, O., Koschella, A., Fidale, L. C., Dorn, S., and Heinze, T. (2007). “Applications of ionic liquids in carbohydrate chemistry: A window of opportunities,” Biomacromolecules 8(9), 2629-2647. DOI: 10.1021/bm070062i
Ganske, K., Wiegand, C., Hipler, U.-C., and Heinze, T. (2016). “Synthesis of novel cellulose carbamates possessing terminal amino groups and their bioactivity,” Macromolecular Bioscience 16(3), 451-461. DOI: 10.1002/mabi.201500324
Ganske, K., and Heinze, T. (2018). “Evaluation of the synthesis of soluble aromatic cellulose carbonates of low degree of substitution,” Macromolecular Chemistry Physics 219(15), article ID 1800152. DOI: 10.1002/macp.201800152
Gericke, M., Schaller, J., Liebert, T., Fardim, P., Meister, F., and Heinze, T. (2012). “Studies on the tosylation of cellulose in mixtures of ionic liquids and a co-solvent,” Carbohydrate Polymers 89(2), 526-536. DOI: 10.1016/j.carbpol.2012.03.040
Gericke, M., and Heinze, T. (2015). “Homogeneous tosylation of agarose as an approach toward novel functional polysaccharide materials,” Carbohydrate Polymers 127, 236-245. DOI: 10.1016/j.carbpol.2015.03.025
Gericke, M., Gabriel, L., Geitel, K., Benndorf, S., Trivedi, P., Fardim, P., and Heinze, T. (2018). “Synthesis of xylan carbonates – An approach towards reactive polysaccharide derivatives showing self-assembling into nanoparticles,” Carbohydrate Polymers 193, 45-53. DOI: 10.1016/j.carbpol.2018.03.083
Granström, M., Kavakka, J., King, A., Majoinen, J., Mäkelä, V., Helaja, J., Hietala, S., Virtanen, T., Maunu, S.-K., Argyropoulos, D. S., et al. (2008). “Tosylation and acylation of cellulose in 1-allyl-3-methylimidazolium chloride,” Cellulose 15(3), 481-488. DOI 10.1007/s10570-008-9197-5
Heinze, T., Rahn, K., Jaspers, M., and Berghmans, H. (1996). “p-Toluenesulfonyl esters in cellulose modifications: Acylation of remaining hydroxyl groups,” Macromolecular Chemistry and Physics 197(12), 4207-4224. DOI: 10.1002/macp.1996.021971218
Heinze, T., and Rahn, K. (1997). “The first report on a convenient synthesis of novel reactive amphiphilic polysaccharides,” Macromolecular Rapid Communications 17(10), 675-681. DOI: 10.1002/marc.1996.030171002
Heinze, T., Liebert, T., and Koschella, A. (2006a). Esterification of Polysaccharides, Springer Science & Business Media, Heidelberg, Germany, pp. 117-128.
Heinze, T., Koschella, A., Brackhagen, M., Engelhardt, J., and Nachtkamp, K. (2006b). “Studies on non-natural deoxyammonium cellulose,” Macromolecular Symposia 244(1), 74-82. DOI: 10.1002/masy.200651206
Heinze, T., Schöbitz, M., Pohl, M., and Meister, F. (2008). “Interactions of ionic liquids with polysaccharides. IV. Dendronization of 6-azido-6-deoxy cellulose,” Journal of Polymer Science Part A: Polymer Chemistry 46(11), 3853-3859. DOI: 10.1002/pola.2269
Köhler, S., Liebert, T., Schöbitz, M., Schaller, J., Meister, F., Günther, W., and Heinze, T. (2007). “Interactions of ionic liquids with polysaccharides 1. Unexpected acetylation of cellulose with 1-ethyl-3-methylimidazolium acetate,” Macromolecular Rapid Communications 28(24), 2311-2317. DOI: 10.1002/marc.200700529
Koschella, A., Heinze, T., Tied, A., Geitel, K., Chien, C.-Y., and Iwata, T. (2020a). “Comparative studies on regioselectivity of α-and β-Linked glucan tosylation,” Molecules 25(22), article no. 5382. DOI: 10.3390/molecules25225382
Koschella, A., Chien, C.-Y., Iwata, T., Thonhofer, M. S., Wrodnigg, T. M., and Heinze, T. (2020b). “All sugar based cellulose derivatives synthesized by azide–alkyne click chemistry,” Macromolecular Chemistry Physics 221(1), article ID 1900343. DOI: 10.1002/macp.201900343
Koschella, A., Leermann, T., Brackhagen, M., and Heinze, T. (2006). “Study of sulfonic acid esters from 1→ 4‐, 1→ 3‐, and 1→ 6‐linked polysaccharides,” Journal of Applied Polymer Science 100(3), 2142-2150. DOI: 10.1002/app.23600
Koschella, A., Hartlieb, M., and Heinze, T. (2011). “A “click-chemistry” approach to cellulose-based hydrogels,” Carbohydrate Polymers 86(1), 154-161. DOI: 10.1016/j.carbpol.2011.04.031
Liebert, T., Hänsch, C., and Heinze, T. (2006). “Click chemistry with polysaccharides,” Macromolecular Rapid Communications 27(3), 208-213. DOI: 10.1002/marc.200500686
Lindemann, H., and Heinze, T. (2022). “Cellulose allylcarbamate with high content of reactive double bond for thiol-ene reaction,” Reactive and Functional Polymers 176, article ID 105306. DOI: 10.1016/j.reactfunctpolym.2022.105306
McCormick, C. L., and Callais, P. A. (1987). “Derivatization of cellulose in lithium chloride and N,N-dimethylacetamide solutions,” Polymer 28(13), 2317-2323. DOI: 10.1016/0032-3861(87)90393-4
Meng, X., and Edgar, K. J. (2016). “ ‘Click’ reactions in polysaccharide modification,” Progress in Polymer Science 53, 52-85. DOI: 10.1016/j.progpolymsci.2015.07.006
Montanez, M. I., Hed, Y., Utsel, S., Ropponen, J., Malmstrom, E., Wagberg, L., Hult, A., and Malkoch, M. (2011). “Bifunctional dendronized cellulose surfaces as biosensors,” Biomacromolecules 12(6), 2114-2125. DOI: 0.1021/bm200201y
Nazir, F., Ashraf, I., Iqbal, M., Ahmad, T., and Anjum, S. (2021). “6-Deoxy-aminocellulose derivatives embedded soft gelatin methacryloyl (GelMA) hydrogels for improved wound healing applications: In vitro and in vivo studies,” International Journal of Biological Macromolecules 185, 419-433. DOI: 10.1016/j.ijbiomac.2021.06.112
Negishi, K., Mashiko, Y., Yamashita, E., Otsuka, A., and Hasegawa, T. (2011). “Cellulose chemistry meets click chemistry: Syntheses and properties of cellulose-based glycoclusters with high structural homogeneity,” Polymers 3(1), 489-508. DOI: 10.3390/polym3010489
Pfeifer, A., Kemmer, A., and Heinze, T. (2022). “Synthesis and structure characterization of novel polyampholytes based on cellulose,” Advanced Industrial and Engineering Polymer Research 5(1), 26-32. DOI: 10.1016/j.aiepr.2021.06.001
Pohl, M., and Heinze, T. (2008). “Novel biopolymer structures synthesized by dendronization of 6‐deoxy‐6‐aminopropargyl cellulose,” Macromolecular Rapid Communications 29(21), 1739-1745. DOI: 10.1002/marc.200800452
Pohl, M., Morris, G. A., Harding, S. E., and Heinze, T. (2009). “Studies of the molecular flexibility of novel dendronized carboxymethyl cellulose derivatives,” European Polymer Journal 45(4), 1098-1110. DOI: 10.1016/j.eurpolymj.2009.01.009
Rahn, K., Diamantoglou, M., Klemm, D., Berghmans, H., and Heinze, T. (1996). “Homogeneous synthesis of cellulose p-toluenesulfonates in N,N-dimethylacetamide/LiCl solvent system,” Die Angewandte Makromolekulare Chemie: Applied Macromolecular Chemistry and Physics 238(1), 143-163. DOI: 10.1002/apmc.1996.052380113
Schmidt, S., Liebert, T., and Heinze, T. (2014). “Synthesis of soluble cellulose tosylates in an eco-friendly medium,” Green Chemistry 16(4), 1941-1946. DOI: 10.1039/c3gc41994k
Siegmund, G., and Klemm, D. (2002). “Cellulose sulfonates: Preparation, properties, subsequent reactions,” Polymer News 27, 84-90.
Skodda, L. (2022). Modulare Synthese reaktiver Xylanderivate zur Funktionalisierung und Gelbildung mittels 1,3-dipolarer Cycloaddition [Modular Synthesis of Reactive Xylan Derivatives for Functionalization and Gelation by 1,3-Dipolar Cycloaddition], M.Sc. Thesis, Friedrich Schiller University Jena, Jena, Germany.
Tiller, J., Klemm, D., and Berlin, P. (2001). “Designed aliphatic aminocellulose derivatives as transparent and functionalized coatings for enzyme immobilization,” Designed Monomers and Polymers 4(4), 315-328. DOI: 10.1163/156855501753210808
Trivedi, P., and Fardim, P. (2019). “Recent advances in cellulose chemistry and potential applications,” in: Production of Materials from Sustainable Biomass Resources, Z. Fang, R. Smith, Jr., X. F. Tian (eds.), Springer, Singapore, pp. 99-115. DOI: 10.1007/978-981-13-3768-0_4
Wuestenberg, T. (2013). Cellulose und Cellulose derivate. Behr’s Verlag GmbH & Co. KG, Hamburg, Germany.
Article submitted: July 7, 2022; Peer review completed: September 11, 2022; Revisions accepted: October 23, 2022; Published: October 28, 2022.
DOI: 10.15376/biores.17.4.Kemmer