by Susanne Pleus and Carl H. Hamann

Electrons can be transferred directly to sugar molecules in non-aqueous solvents using the negative electrode of an electrolysis cell. This is utilised for the introductory step of a new way of obtaining modified sugars, e.g. in the form of alkyl and acyl glycosides. The latter are important fine or speciality chemicals that can be used as precursors in medicine and pharmaceuticals as well as in petrochemistry, colour chemistry and the plastics industry. The work presented here was funded by the chemical industry and the Federal Ministry of Education and Research.

The utilisation of renewable raw materials has a long tradition in chemistry. Until 1930, for example, plastics (at that time rubber, vulcanised fibre, celluloid, galalith) were produced 100 percent from renewable raw materials. Today, renewable raw materials (in the form of oils, fats, sugar, starch and cellulose) account for 1.8 million tonnes of the chemical industry's raw material base. This corresponds to 10 per cent of the weight and 20 per cent of the value of all chemical raw materials used. Products include Pharmaceuticals (antibiotics, vitamins), cosmetics (skin and hair care sector), detergents (surfactants for the production of detergents, rinsing agents and cleaning agents, emulsifiers and auxiliaries in dyeing), crop protection agents, packaging materials (paper, films, foils), textiles (finishes), adhesives (wallpaper paste, glues for wood panels), building materials (plasterboard) and chemicals (synthetic resins, polyurethanes, polyethers, phenolic resins, plasticisers, organic acids).

For medical research, glycoconjugates of carbohydrates and proteins (glycoproteins) or lipids (glycolipids) as the main components of cell membranes are of decisive importance, especially since the role of the carbohydrate part in the pharmacological and cytotoxic sense has become known. The carbohydrate part fulfils important functions in cell-cell recognition and interaction, in cell growth control and carcinogenesis.

From an industrial point of view, the substances listed represent fine or speciality chemicals that can be sold at good prices. Technical and economic criteria determine whether such products are based on renewable raw materials or petrochemicals. A competitive advantage - and thus an intrinsic (even in the absence of subsidies) economic efficiency - is all the greater for the renewable raw material, the better the desired molecular target structure for the product is already given in the raw material.

Access to basic chemical products (such as ethylene, propylene, benzene, toluene) from renewable raw materials, on the other hand, is uneconomical. The complex molecular structures formed by nature with considerable energy input (solar radiation) - in addition to the energy and economic input required for cultivation - are transformed into low-molecular, low-priced substances with further energy input. Here, the petrochemical basis - in the future century the coal basis - will remain decisive.

The production of motor fuels from renewable raw materials is also problematic. In the case of so-called sugar alcohol - the fermentation of sugar into ethanol - the considerations outlined above apply. In the case of biodiesel, the large molecules of rapeseed oil created by nature are not destroyed, but only modified (to form rapeseed oil methyl ester). However, the cultivation of the rapeseed plant is very energy-intensive and hardly any other plant requires more fertiliser. The question of improving the carbon footprint also remains open: calculated with a sharp pencil, the CO₂ emissions from the cultivation, rearing and processing of the renewable raw material would have to be taken into account, as would the (proportionate) production of the machinery. The charm of biodiesel, however, remains the double subsidy (in the form of the cultivation premium and the waiver of mineral oil tax) - and blossoming rapeseed fields are aesthetically pleasing.

In the following, a new process for the simple extraction of valuable products from renewable raw materials is presented. Natural carbohydrates and easily obtainable secondary products are modified electrochemically (i.e. by means of electrolysis). This produces important fine chemicals which are particularly important for medical research - as building blocks in the synthesis of glycoconjugates.

History of the electrochemistry of carbohydrates

The earliest work on the electrochemistry of carbohydrates dates back to 1872, when H. T. Brown described the electrolysis of sucrose and glucose in a beaker cell with platinum electrodes. It led to the development of CO₂ and an unspecified product mixture. And reference is made in this work to a publication by the Dutchman Brewster from 1866. This would make the electrochemistry of carbohydrates 125 years old - assuming that there are no older works.

However, the systematic electrochemistry of carbohydrates only began at the turn of the century. C. In 1908, Neuberg converted glucose on a lead anode in aqueous sulphuric acid to a mixture of gluconic acid, arabinonic acid, formaldehyde and other products. Gluconic acid as the desired sole product could only be realised by H. S. Isbell and H. L. Frush in 1931. They produced hypobromide at the anode from bromide, which converted the glucose into gluconic acid. Calcium hydroxide added to the latter immediately forms insoluble calcium gluconate. This process was used industrially in the 1930s. Today, gluconic acid is obtained enzymatically from glucose.

Inspired by the work of Emil Fischer in 1880, who succeeded in converting aldoses to alcohols on sodium amalgam, W. Loeb attempted the first direct electrolysis of glucose on lead as a cathode in a sulphuric acid solution in 1910. It did not yet lead to the desired goal, the conversion of glucose to sorbitol. It was not until Creighton's work in 1926 that the desired reaction was achieved. It was made possible by the use of a weak alkaline solution. In 1937, the Atlas Powder Company began the commercial electrochemical production of sorbitol from glucose on this basis. The process was subsequently expanded. Today, the process is no longer competitive with chemical catalytic hydrogenation on Raney nickel. It is reported that one small-capacity plant is still in operation in India alone.

Own work

Some years ago, our own work in the field of cathodic conversion of simple aliphatic alcohols such as methanol and ethanol showed that these molecules can be split into anions (alcoholate ions) and hydrogen by direct transfer of electrons at high conversion rates (Figures 1 and 2). If chloride ions are discharged at the anode and a membrane that is selectively permeable to the alkali cations is placed between the electrodes, the coveted industrial intermediate alkali alcoholate CH₃ONa is produced in the cathode chamber at an electrolysis voltage of around 3 V in an electrochemical direct synthesis (the current industrial process is based on previously produced sodium amalgam).

However, the direct transfer of electrons to alcohols quickly loses speed when switching to long-chain or branched aliphatic alcohols. Technical utilisation then fails. In the case of carbohydrates - which are also alcohols ("polyols") from a chemical point of view - the reaction speed then increases again to such an extent that preparative utilisation is possible.

We have recently studied this reaction in detail. Since an electrolysis voltage of more than 3 V is required, we use dimethylformamide (DMF) as a solvent - water as a solvent would immediately decompose to hydrogen and oxygen at this electrolysis voltage. Figure 3 shows an example of the substrate molecule methyl a-D-glucopyranoside in the top left-hand section (simplified illustration. Carbon atoms are positioned at the points labelled 1 - 6) and its transition into the anion and hydrogen after electron transfer. The reaction at the counter electrode is the formation of elemental bromine from lithium bromide Li+Br-. The remaining lithium ions form the alkali alcoholate (sugar alcoholate) as in the previous case in Fig. 2.

However, this is not yet the desired product. Instead, after the end of the electrolysis - when the sugar dissolved in the DMF is largely exhausted - a capture reagent is added. In the illustration, benzyl bromide was chosen as the reagent, the sugar is thus benzylated (formation of the benzyl ether).

As can be seen in Fig. 3, different anions and therefore different products can be formed due to multiple OH groups ("polyfunctionality") - identical according to the gross chemical formula, but with different molecular configurations ("isomeric products"). In the chemical system, depending on the substituted position, one speaks of 2-0-benzyl methyl-a-D-glucopyranoside (example shown in the drawing) or of 6-0-, 3-0- and 4-0-benzyl methyl-a-D-glucopyranoside.

After completion of the reaction with the capture reagent, the products are separated from the solvent and salt component and separated from each other (isolated from each other using column chromatography). The exact identification (according to substituted C-positions) is carried out using nuclear magnetic resonance (NMR).

Naturally, the product distribution - the proportion of different substitution positions in the product spectrum - will depend on the feedstock and intercepting reagent.

In the case of the reaction shown in Fig. 3, the main product is the 2-0 benzyl ether at 55 per cent and the 6-0 benzyl ether in second place at 19 per cent. The remainder is distributed between the 3 and 4 positions (5 and 7 per cent respectively) and various multiple substitutions in the percentage range. In other cases, 100 per cent substitutions were achieved in only one position, which is usually highly desirable in production processes.

In our work to date, we have analysed a total of 18 different sugars, including naturally occurring species such as sucrose (cane sugar), as feedstocks. We have used 12 different species as scavenging reagents - alkyl halides (e.g. the benzyl bromide shown in Fig. 3), silyl chlorides (e.g. tert-butyl diphenyl silyl chloride), acid halides and anhydrides. Since, as explained above, different isomers can be formed in the individual experiment by regioselective substitution, the total number of species synthesised is considerable. It amounts to more than 150, a considerable proportion of which were not previously described in the literature, i.e. synthesised for the first time. Figure 4 shows an example of a corresponding table with NMR data to document the analysis from a specialist publication.

The synthesised species are chemically sugars in which a specific OH group has been replaced by a 0-alkyl or 0-acyl group. Such groups are considerably less reactive than OH groups themselves. The molecule can therefore now be reacted with other substances without attacking the position in question. The alkyl or acyl group is therefore chemically labelled as a protective group, which makes further regioselective reactions (e.g. leading to the fine chemicals mentioned in the introduction) possible in the first place.

In organic preparative chemistry, the formation of sugar anions is achieved with the aid of common bases such as NaH, LiH, KOH and Ag₂0. In order to be able to carry out targeted, regioselective alkylations and acylations, which are irreplaceable in the protective group chemistry of carbohydrates, auxiliary reagents such as organometallic compounds must also be used. If electrons are used as a reagent, as described here, a considerable amount of corresponding waste materials can be saved.

Conclusion

New synthesis routes for known substances or the synthesis of new substances are important results of basic chemical research. This still applies even if the results are not - as is often demanded today - implemented in industrial practice in zero time. The famous Maxwell is said to have once replied to an English finance minister's question about the purpose of his research (to paraphrase): "One of your successors will raise taxes on it".

The authors

Dr Susanne Pleus (31) studied chemistry in Münster and Oldenburg. She obtained her doctorate in 1994 in the local Applied Physical Chemistry working group. She is now working there as a research assistant on electrodes designed to enable enantioselective electrolysis.

Prof Dr Carl H. Hamann (58) studied mathematics, physics, biology and economics in Hamburg and Bonn. He was appointed to the newly established Oldenburg Chair of Applied Physical Chemistry in 1975. His main interest is in electrolytic production processes.