This invention relates to the derivatisation of carbon, i.e. to the covalent attachment of molecules to the surface of elemental carbon.
Derivatisation of carbon surfaces has attracted considerable interest in recent years, in particular in connection with attaching molecules to carbon nanotubes.
The derivatisation of carbon may be carried out for a range of purposes which include modification of the surface properties of a carbon substrate, preparation of carbon-epoxy composites and attaching a molecule to a carbon electrode so that it can take part in an electrochemical reaction in an electrochemical sensor or an electrochemical catalyst.
Traditionally, carbon surfaces were modified by vigorous oxidation on the surface leading to the formation of carboxylic, quinonic, ketonic or hydroxylic groups, which were then reacted further with the target molecule. This aggressive process was difficult to control.
A number of procedures have been described for electrochemical induced derivatisation leading to the formation of a single covalent bond between a carbon electrode and a moiety which becomes attached. Examples have included: electro-oxidation of alcohols, —see Gouet al Chem. Pharm. Bull., vol 44, page 860 (1996);
electro-oxidation of primary and secondary amines—see Barbier et al J. Electrochem. Soc., vol 137, page 135 (1990), Antoniadou et al J. Appl. Electrochem., 1992, vol 22, page 1060 (1992) and Deinhammer et al Langmuir vol 10 page 1306 1994);
electro-oxidation of carboxylates—see C. P. Andrieux et al, J. Am. Chem. Soc., vol 119, page 4292 (1997);
One approach to derivatisation of carbon without electrochemistry is the homogeneous reduction of diazonium compounds in reducing media—see Pandurangappa et al Analyst, vol 127, page 1568 (2002) and Leventis et al, Talanta vol 63, page 1039 (2004). Also in this category is WO2005/066618 (Schlumberger) which describes the diazocoupling of anthraquinonyl and nitrophenyl groups onto carbon nanotubes by means of the reduction of diazonium salts.
U.S. Pat. No. 6,645,455 (William Rice University) discloses derivitisation of carbon nanotubes by exposure to fluorine gas, possibly followed by further reaction to replace fluorine with some other group. Derivatisation of carbon nanotubes has also been carried out using amines or thiols to attach at defect sites of the carbon nanotubes—see Basiuk et al J Phys Chem B vol 106 age 1588 (2002) and Zanella et al J Phys Chem B vol 109 page 16290 (2005). A number of rather esoteric species have been reacted with carbon nanotubes: amongst these are reaction with nitrenes and the attachment of dipyridyl imidazolidene which is an unusual molecule in which a carbene forms part of a large delocalized system and in consequence does not undergo normal carbene reactions, see Holzinger et al Angewandte Chemie. Int Ed vol 40 page 4002 (2001).
We have now found a new and versatile method of attaching molecules to elemental carbon. In a first aspect this invention provides a process which comprises exposing a carbon substrate to an aprotic reaction mixture containing a strong base and a compound (herein referred to as the derivitising compound) with a general formula
or
in which X denotes a covalently attached leaving group capable of forming an X− anion and each Z is a covalently attached moiety other than X comprising a plurality of atoms, which remains attached and does not have a hydrogen atom directly attached at the alpha position, i.e. does not have a hydrogen atom directly attached to the atom which itself is attached to —CH2X or —CHX.
Without wishing to be bound by theory, we believe that the strong base detaches a hydrogen atom from the terminal —CH2X or —CHX. The atom or group X then leaves as the anion X− so that the compound is converted to a reactive carbene species having the general formula:
which then reacts with the carbon substrate. This reaction scheme is illustrated in
The base needs to be strong enough to remove a covalently bonded hydrogen atom from the derivitising compound. The derivitising compound must not have hydrogen attached to an atom which is in the alpha position, directly adjacent to the terminal —CH2X or —CHX, so that the base cannot take hydrogen from this atom, hence leading to formation of a double bond rather than a carbene.
The strong base may be an uncharged, strongly basic compound and preferably it is a nitrogenous base. Two strong nitrogen-containing organic bases which are commercially available and which can be used in the process of this invention are 1,8-diazabicyclo[5.4.0]undec-7-ene (hereafter referred to as DBU) which has the formula
and its homologue 1,5-diazabicyclo[4.3.0]non-5-ene (hereafter DBN) which has the formula
The strong base used in the process of this invention may therefore be an uncharged compound with a pKa which is at least as high as that of DBN in an aprotic solvent. The pKa values of numerous bases in acetonitrile as solvent have been reported by Kaljurand and coauthors, see Kaljurand et al J. Org. Chem vol 65 pp 6202-6208 (2000), Kaljurand et al J Org Chem vol 70 pp 1019-1028 (2005) and Kaljurand et al J Phys Chem B vol 111 pp 1245-50 (2007).
DBU has a reported pKa value in acetonitrile of 24.34 which is well above the reported pKa of 18.82 for triethylamine in the same solvent, see Kaljurand et al J Org Chem vol 70 pp 1019-1028 (2005). DBN has been reported to have a pKa slightly below that of DBU: the values reported by Leffek et al in Can J. Chem vol 67 pp 590-595 (1989) were 23.9 and 23.4 for DBU and DBN respectively.
The base used in the process of this invention may therefore be defined as a base having a pKa in acetonitrile of at least 22 or alternatively may be defined as having a pKa in acetonitrile which is at least equal to that of DBN in acetonitrile.
A number of nitrogen-containing and phosphorus-containing bases are known to have higher pKa values than DBU and could be used in the process of this invention. The papers by Kaljurand et al mentioned above provide references to many of these very strong bases.
Provided the strong base is a liquid, the process may be carried out using the strong base itself as reaction solvent. Alternatively the process may be carried out in an aprotic solvent such as toluene. The reaction may be carried out at a temperature of 0° C. or above, such as in the range from 0 to 40° C., preferably from 10 to 30° C.
The amount of strong base will generally be at least the stoichiometric amount needed to remove HX from the derivitising compound which is to be attached to carbon. An excess of base may be used.
This process is versatile in that a very great range of compounds may be used as the derivitising compound which is attached to a carbon substrate. The requirement for a strong base can be met using DBU or DBN which are commercially available reagents and the process can be carried out under ambient temperature conditions.
The moiety Z (or both moieties Z) in the derivitising compound must remain covalently attached to the terminal —CH: or >C: of the carbene species and so the formation of the leaving anion should be favoured over any formation of a Z anion. Preferably each moiety Z provides a carbon atom at the position alpha to the terminal —CH: or >C: i.e. provides a carbon atom which is directly covalently attached to the carbon of the terminal —CH: or >C: entity.
The derivatising compound used in the process of this invention may have any of the following five general formulae
wherein R1 to R6 each independently denote attached groups, or two or more of the groups R1 to R6 may together form a ring.
These groups R1 to R6 may each independently be saturated or unsaturated aliphatic or cycloaliphatic (i.e. alicyclic) or may be aromatic. They may be substituted, and so an aromatic group may have one or more aliphatic, cycloaliphatic or aromatic substituents on it and an aliphatic or cycloaliphatic group may have one or more aromatic, aliphatic or cycloaliphatic substituents on it. Aliphatic groups may be straight or branched. These groups R1 to R6 may include hetero atoms such as nitrogen or silicon and they may be complexed with one or more metal atoms.
If two of the groups R1 to R6 form a ring, this ring may be aromatic. Thus a compound of the above general formula
may have an aromatic ring formed by the groups R1 and R2. Examples of such compounds are
The part of the derivatising compound which becomes attached to the carbon substrate should generally be stable in the presence of the strong base, although it is possible that it would itself undergo another chemical reaction in the presence of that base. For example the strong base might remove a protecting group at the same time as carrying out the process of this invention.
The leaving group X is preferably a bromine, chlorine or iodine atom, but it may be some other leaving group. It may be a single atom or a group which contains a plurality of atoms. Preferably it is at least as good a leaving group as a chlorine atom. This may be defined by reference to the pKa of the acid HX (in water or other solvent) which should be no greater than the pKa of HCl in the same solvent. Leaving groups which are as good as or better than a chlorine atom include bromine, iodine, para-toluene sulphonate and trifluoroacetate.
The relative ratio of the carbon substrate and the compound which is to be attached to it may vary considerably. Generally the amount of the compound will be insufficient to attach to all possible binding sites on the surface of the carbon substrate.
The carbon substrate may be any form of carbon including glassy carbon, carbon fibres, carbon black and graphite or carbon paste, boron doped diamond and carbon epoxy. One further form of carbon, to which the invention may be applied, is the carbon nanotube (CNT) which was discovered in 1991. The structure of carbon nanotubes approximates to rolled-up sheets of graphite and can be formed as either single or multi-walled tubes. Single-walled carbon nanotubes (SWCNTs) constitute a single, hollow graphite tube. Multi-walled carbon nanotubes (MWCNTs) on the other hand consist of several concentric tubes fitted one inside the other.
The carbon substrate may be conductive and may be for use in an electrode. Forms of conducting carbon used in electrode manufacture are glassy carbon, carbon fibres, carbon black, various forms of graphite, carbon paste, boron doped diamond and carbon epoxy. Carbon nanotubes may also be used as part of an electrode and may be immobilized on the surface of another form of conducting carbon.
As indicated at the beginning, carbon which has been derivatised by the attachment of a molecular species may be used for a variety of purposes. Carbon which has been derivatised by the process of this invention may be utilised for any such purposes. In particular, the Z—CH or Z2C moiety which becomes attached to carbon may be such that it can undergo electrochemical reduction or oxidation, so that the derivatised carbon can be used in an electrochemical sensor, for example in the manner described in WO 2005/66618, the disclosure of which is incorporated herein by reference. A reversible oxidation and reduction which can be monitored by voltammetry is particularly useful. This may be preceded by an initial irreversible alteration of the covalently attached moiety to form a species which is still covalently attached to the carbon substrate and can undergo reversible electrochemical reduction and oxidation.
This is exemplified below by covalent attachment of a nitrophenyl group whose nitro group can be reduced electrochemically to a nitroso group and thereafter reversibly reduced to a hydroxylamino group.
Embodiments of the invention will now be described with reference to the accompanying drawings and examples.
As shown in
Several experiments were carried out in order to provide evidence for this reaction scheme.
1 ml DBU and 2 ml CDCl3 were quickly mixed in a 20 ml vial, which was then sealed with a butyl rubber septum. Heat was liberated, indicating that an exothermic reaction was taking place. The mixture turned yellow in 16 hours and became dark coloured and viscous in 2 to 3 days, after which time the adduct DBU.HCl was detected by NMR spectroscopy. It was confirmed by single crystal X-ray diffraction after isolating it
1 ml DBU and 2 ml CHBr3 were quickly mixed in a 20 ml vial, which was then sealed with a butyl rubber septum. The solution turned yellow immediately the reactants were mixed and reaction released a considerable amount of heat. The solution turned dark brown within 4 hours. DBU.HBr was detected by NMR spectroscopy.
The above experiments 1 and 2 showed that DBU was able to break carbon-hydrogen (and likewise carbon-deuterium) bonds in deuterochloroform and bromoform. The following three experiments demonstrate formation of carbene species.
1 ml DBU, 2 ml CDCl3 and 0.4 ml cyclohexene were quickly mixed in a 20 ml vial and then sealed with a butyl rubber septum. Several days later, after the solution had turned a dark brown colour, the mixture was analysed by headspace GC-MS.
7,7-dichloro-bicyclo[4, 1, 0]heptane was detected. The latter is the product obtained by reaction of dichlorocarbene at the double bond of cyclohexene. It appeared as a clear peak in the gas chromatogram and the mass spectrum for that peak included a molecular ion at mass 164.
1 ml DBU, 2 ml CDCl3 and 0.4 ml tetramethylethylene were quickly mixed in a 20 ml vial and then sealed with a butyl rubber septum. The solution turned a dark brown colour over several days, after which the mixture was analysed by headspace GC-MS and 3,3-dichloro-1,1,2,2-tetramethyl-cyclopropane was detected. The product 3,3-dichloro-1,1,2,2-tetramethylcyclopropane was observed as a sharp peak in the gas chromatogram which could be identified by the associated mass spectrum. This experiment once again confirmed the production of dichlorocarbene in the reaction of DBU with CDCl3.
1 ml DBU, 2 ml CHBr3 and 0.4 ml cyclohexene were quickly mixed in a 20 ml vial, which was then sealed with a butyl rubber septum. Upon mixing the mixture turned yellow immediately and after several days it had turned black. The mixture was examined by headspace GC-MS and 7,7-dibromo-bicyclo[4,1,0]heptane was detected. It was observed as a sharp peak in the gas chromatogram which was identified by the associated mass spectrum with a molecular ion at mass 256. This experiment confirmed that dibromocarbene was produced in the reaction of DBU with CHBr3.
The following examples demonstrate the process of the invention being used to attach reactive moieties to carbon substrates. All chemicals were supplied by Aldrich and used without further purification.
This reaction is illustrated as
4-nitrobenzylbromide (1.4421 g, 6.68 mmol) supplied as a crystalline pale yellow powder was mixed with graphite (1.0182 g) in a 20 ml vial. DBU (1 ml) was added to the vial which was then sealed with a butyl rubber septum. The DBU acted as the required strong base and also provided the reaction solvent. The mixture was stirred at first, but after one day the mixture had solidified and could no longer to be stirred. Acetone was added to the mixture, then the solvent was filtered off and the graphite was washed repeatedly with acetone by filtration. Initially the acetone filtrate was dark brown but after repeated washing the acetone filtrate was colourless. The graphite which had been filtered off was allowed to dry in air.
The binding of nitrophenyl groups to graphite was demonstrated by mechanically immobilising the dried graphite on an electrode and examining its electrochemistry.
The dried graphite which had been derivatised with nitrophenyl groups was placed on a fine filter paper. A basal plane pyrollytic graphite (bppg) electrode was polished smooth on glass polishing paper and then on silicon carbide paper, after which the polished surface was gently rubbed on the dried graphite on the filter paper, so that this graphite became immobilised on the electrode. Cyclic voltammetry was then carried out using this electrode.
The reduction process at −0.5 V was attributed to the nitrophenyl group undergoing a 4 electron, 4 proton reduction to form the corresponding hydroxylamine species as shown in the reaction scheme which is
Plots of oxidation and reduction peak current, measured at +0.39 V and +0.29 V respectively, against scan rate were found to be linear from 25 to 400 mV s−1. These results showed that the 4-nitrophenyl group was strongly attached to the carbon surface. Further evidence to show that the nitrophenyl group had been immobilised onto the carbon was obtained by exchanging the buffer solution with a fresh solution and recording the voltammetry again. The voltammetric waveshapes (not shown) for each response were similar to those obtained previously, thereby confirming that the nitrophenyl group was attached to the graphite particles.
The stability of the electrode was further tested by placing it in fresh buffer solution and repetitively cycling over the voltage scan range. It was observed that although there was a 50% loss in peak current over the first 50 cycles, the peak current and voltammetric response were almost constant for the next 100 cycles, indicating strong attachment to the graphite.
As a control experiment the derivatisation procedure was carried out without DBU. Acetone was substituted as the solvent. The reaction mixture was stirred at room temperature for 2 days. The solvent was filtered off, and the graphite was washed with acetone by filtration several times until the acetone filtrate was colourless. The resulting graphite was dried in air. Testing as above did not show the redox behaviour described above, indicating that attachment of nitrophenyl groups to graphite had not occurred in this control experiment.
As with the previous example, the scheme of the reaction is shown in
The binding of nitrophenyl groups to carbon nanotubes was demonstrated by immobilising the dried carbon nanotubes on a glassy carbon electrode and examining its electrochemistry. Before immobilizing on an electrode, the derivatised carbon nanotubes (1 mg) were washed again using DMF (N,N-dimethylformamide) for many times by centrifugation until the DMF supernatant was colorless. An immobilising suspension was prepared by introducing 1 mg of carbon nanotubes into 1 mL of DMF.
The liquid was then sonicated for 20 minutes to aid in the dispersion of the carbon. A 20 μl_aliquot of the sonicated suspension was then placed directly onto the glassy carbon surface and left to dry. The electrode was then ready for use.
Cyclic voltammetry was then carried out.
Subsequently the potential scan was narrowed to examine only the voltammetric response of the hydroxylamine/nitroso redox couple, and various scan rates were used. As the scan rate was increased from 0.025 V to 0.4 V, an increase in the oxidative and reductive peak current was observed. A plot of peak current against scan rate was found to be linear over the entire scan rate studied indicating that the nitrophenyl species is attached to the surface of the carbon nanotubes.
In a further experiment, the square wave voltammetric response of these nitrobenzyl-derivatised carbon nanotubes was examined in buffer solutions with various pH values. As the pH was varied from 4 to 9 the oxidative peak potential for the oxidation of hydroxylamine to the nitroso species increased. A plot of peak potential versus pH was found to be linear with a gradient of 52.2 mV/pH unit consistent with a 2 electron, 2 proton transfer as illustrated for the hydroxylamine/nitroso redox couple in
Again the fact that the electrode can be successfully removed, washed and placed into various buffer solutions shows that the species undergoing electrochemical oxidation and reductions is attached to the carbon nanotubes.
In a control experiment 4-nitrobenzylbromide (1.3826 g) was mixed with carbon nanotubes (123 mg) in ca. 3 ml acetone but without the presence of DBU. The reaction mixture was stirred at ambient temperature for 2 days. The acetone was filtered off and was pale yellow in colour. The remaining solid was washed with acetone several times more by filtration until the acetone filtrate was colourless. The acetone filtrate was evaporated and almost all of the 4-nitrobenzylbromide was recovered. The CNTs which had been used in this experiment were immobilised on a glassy carbon electrode in the same manner as described above. The voltammetric response of these carbon nanotubes did not show the redox couple at 0.0 V, indicating that there is no hydroxylamine or nitroso species present on the surface and hence that derivatisation of the CNTs had not taken place.
The scheme of this reaction is illustrated as
The derivatised graphite, graphite from the control experiment, and untreated graphite were all immobilised on glassy carbon electrodes using the procedure of the previous example (except that acetone was used in place of DMF). The derivatised graphite showed a large reductive wave at −0.8 V followed by emergence of a new redox wave at 0.0 V. By contrast, these features were not seen in the voltammetric response of graphite from the control experiment which was the same as that of untreated graphite.
The procedures described in the above examples could also be used for attaching a variety of other species to various forms of carbon, including carbon nanotubes. The attached species could be redox active. One possibility is to attach ferrocene as illustrated in
Number | Date | Country | Kind |
---|---|---|---|
0904813.3 | Mar 2009 | GB | national |
This patent application claims priority, pursuant to 35 U.S.C. §119(e), to U.S. Non-Provisional application Ser. No. 13/142,515 filed Nov. 23, 2011, entitled “Derivatisation of Carbon”, which claims priority to International Application No. PCT/11310/00293 filed Feb. 12, 2010, which claims priority to United Kingdom Application No. 0904813.3 filed Mar. 20, 2009, and the disclosures of which are fully incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
Parent | 13142515 | Nov 2011 | US |
Child | 14665832 | US |