Patent Application
20090078581
The field of the invention is synthesis of antioxidants, and especially electrochemical synthesis of dihydrolipoic acid.
R-lipoic acid (α-lipoic acid, thioctic acid, or 5-(1,2-dithiolan-3-yl)pentanoic acid; RLA) and the corresponding S-enantiomer and mixtures thereof are often used in dietary supplements, topical formulations, and pharmaceutical formulations to provide antioxidant activity and/or to supply various enzymes in the energy metabolism (e.g., pyruvate dehydrogenase, 2-ketoglutarate dehydrogenase, etc.) with their respective cofactors. RLA is widely distributed in animal, plant and microbial cells, and is found to be essential in many biochemical reactions involving hydrogen transfer and acyl transfer. Other uses include metal chelation (e.g., mercury chelation), induction of glutathione synthesis, modulation of insulin and PKB/Akt signaling, and treatment or prevention of various neuropathies. Regardless of the particular use, the reduced form of lipoic acid, dihydrolipoic acid, is often considered the more active form. The reduction of lipoic acid to dihydrolipoic acid is schematically illustrated in Equation I below:
While lipoic acid can be produced in numerous manners (e.g., fully synthetic as shown in U.S. Pat. No. 7,208,609 or 6,140,512, or in vivo in many organisms as, for example, described in U.S. Pat. App. No. 2004/0235123), preparation of dihydrolipoic acid from lipoic acid is often a technically challenging process as relatively severe reducing conditions are required. Among other difficulties, where reducing agents are employed, the reduction is often incomplete and residual reducing agent must be removed from the preparation. Moreover, in most reductions, the product is often a racemic mixture. For example, current industrial dihydrolipoic acid production typically employs NaBH4 as reducing agent. Unfortunately, such process only affords yields of about 80% and requires removal of NaBH4 form the reaction mixture, which is a relatively expensive process. Alternatively, lipoic acid can be enzymatically reduced in an in vivo or in vitro system (e.g., U.S. Pat. App. No. 2006/0234359, 2005/0112111, or Biochem Pharmacol. 1996 Feb. 9; 51(3):233-8). However, such processes are typically not amenable to large scale reactions (e.g., for quantities of more than 100 g) and/or fail to provide a high degree of dihydrolipoic acid.
To improve yield, dihydrolipoic acid can be synthesized from precursors in a route that does not require reduction as described in U.S. Pat. Nos. 5,489,694, 6,906,210, and 7,109,362. While such systems often provide significantly improved process control and yield, non-natural precursors are often used that typically necessitate de novo synthesis. In other known methods, lipoic acid is electrochemically reduced while being bound to a graphite electrode functionalized with Fe(II)-phthalocyanine. Such reduction is employed to provide for co-enzyme recycling for NAD(P)(+)-specific dehydrogenases in a sensor assembly (Biosens Bioelectron. 2001 June; 16(4-5):245-52). Here the authors report a 95% yield at −1200 mV in phosphate buffer pH 7.0. While the yield of dihydrolipoic acid in such catalytic process is quite high, such reduction is entirely unsuitable for large scale preparation of dihydrolipoic acid as the lipoic acid must be bound to the surface of the catalytically activated electrode surface. Still further, the Fe(II)-phthalocyanine used in such processes is often cost prohibitive for industrial-scale production.
Therefore, while numerous methods of reduction of lipoic acid are known in the art, all or almost all of them suffer from one or more disadvantages. Consequently, there is still a need to provide improved configurations and methods for reduction of lipoic acid.
The present invention is directed to devices and methods of electrochemical reduction of lipoic acid to dihydrolipoic acid in which dihydrolipoic acid product is preferably separated from the lipoic acid using differential lipophilicity of the dihydrolipoic acid and the lipoic acid. In an especially preferred aspect, reduction is performed in highly alkaline electrolyte on a carbon felt cathode, and separation is preferably in a continuous manner and performed using a membrane that is preferentially permeable to the dihydrolipoic acid. In another preferred aspect, RLA is solubilized in strong alkaline medium (e.g., 1M NaOH at 50-80° C. at a high concentration [e.g., 300 g/L]) and then electrochemically reduced at 25° C. in a flow cell. The DHLA and RLA are then separated by neutralizing the solution (e.g., with 1M HCl), which results in the formation of an oily layer that is substantially completely pure (chemically and enantiomerically) DHLA.
In one aspect of the inventive subject matter, a method of reducing lipoic acid includes a step of electrochemically reducing lipoic acid to dihydrolipoic acid in an alkaline electrolyte on a cathode that comprises a carbon containing material. In a further step, the dihydrolipoic acid is separated from the lipoic acid using differential lipophilicity of the dihydrolipoic acid and the lipoic acid. In particularly preferred aspects, the alkaline electrolyte comprises at least 0.1M of an alkaline metal hydroxide or alkaline earth metal hydroxide, the carbon containing material comprises a carbon felt, and the anode comprises Ir2O3. Furthermore, it is generally preferred that the step of separating is performed in a phase separation step in which the dihydrolipoic acid is separated from an aqueous phase in which the lipoic acid is disposed. For example, the phase separation may be performed using a lipophilic solvent, preferably concurrently with the step of electrochemically reducing the lipoic acid. Alternatively, or additionally, separation may be performed using a membrane that is preferentially permeable to dihydrolipoic acid, also preferably concurrently with the step of electrochemically reducing the lipoic acid. Where a membrane is used for separation, it is generally preferred that the membrane is located in a separation compartment that is fluidly coupled to a cathode compartment in which the lipoic acid is reduced to dihydrolipoic acid, and that the separation compartment comprises an outlet for the separated dihydrolipoic acid.
Consequently, in another aspect of the inventive subject matter, one exemplary system for reduction of lipoic acid will include an electrolytic cell having an anode compartment with an anode, a cathode compartment with a cathode, and a separator, wherein the cathode compartment comprises an alkaline electrolyte in which lipoic acid is dissolved, and wherein the cathode comprises a carbon containing material. Most preferably, both anode and cathode compartments are configured as flow cells. In such systems, a power source is coupled to the anode and cathode to provide current at a voltage sufficient to reduce lipoic acid to dihydrolipoic acid.
Particularly preferred systems include a separation compartment fluidly coupled to the cathode compartment, wherein the separation compartment has a membrane (e.g., fluorinated polymer) that is preferentially permeable to dihydrolipoic acid, and wherein the membrane is coupled to the separation compartment such that dihydrolipoic acid is preferentially removable from the separation compartment via the membrane. It is further preferred that the separation compartment and the cathode compartment are configured to allow continuous flow of the electrolyte between the separation compartment and the cathode compartment, and/or that the cathode is configured as a flow-through cathode. Particularly preferred cathodes include carbon felt while particularly preferred anodes comprise Ir2O3. In such configurations, electrochemical reduction is typically carried out using a constant current electrolysis (e.g., 10 A, 5 V), and lipoic acid is reduced in significant quantities (e.g., 100-350 g/L).
Various objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of preferred embodiments of the invention along with the accompanying drawing.
The inventor has discovered efficient and simple devices and methods of reduction of lipoic acid to form dihydrolipoic acid in which lipoic acid is electrochemically reduced on a cathode in alkaline medium (preferably at a pH of at least 9), wherein the cathode allows a two-electron transfer reaction and in which the so formed dihydrolipoic acid is separated from the electrolyte using differential lipophilicity (e.g., via phase separation of a neutralized solution and/or membrane separation across a hydrophobic membrane).
Most advantageously, such two-electron reductions will reduce the RLA to DHLA in a single step without production of intermediates (e.g., semiquinone structures) and/or polymerization products. In especially preferred aspects, the cathode material has the capacity to store hydrogen within the cathode material while the RLA passes through the cathode. Typically, the cathode material is therefore porous and comprises carbon felt or grapheme-type material (e.g., graphene, pyrolytic or exfoliated graphite, etc.), and is configured as a flow-through electrode.
In contrast to heretofore known methods of reduction of lipoic acid, reduction according to the inventive subject matter does not require removal of any reduction agent or precursors from which the dihydrolipoic acid can be formed. Moreover, contemplated systems and methods allow reduction in a single process step while maintaining enantiomeric purity. Still further, it should be noted that contemplated systems and methods allow production of highly pure redox compositions with predetermined relative quantities of lipoic acid and dihydrolipoic acid.
One exemplary system for electrochemical reduction of RLA with concurrent removal of dihydrolipoic acid is illustrated in
Similarly, catholyte is provided to the cathode compartment from catholyte tank 120 via feed line 125. The reduced catholyte is then routed as stream 126 via pump 164 to tank 120 as stream 126A. Where online separation is desired, reduced catholyte 126 is fed to online separator 150 that employs a lipophilic membrane (not shown) to separate DHLA product 152 from recycle stream 154 that predominantly comprises RLA. Recycle stream 154 is then mixed in mixing chamber 140 with fresh RLA via stream 121, and base via stream 122. Where desired, nitrogen purge may be provided via stream 123. Mixed stream 124 is then fed to catholyte tank 120 from which hydrogen is vented via stream 127.
In preferred separation compartments, separation is driven by the substantial difference in lipophilicity between the RLA educt and DHLA product. Here, an online separator includes a polytetrafluoroethylene (PTFE) membrane against which slight pressure is applied on the inflow side. Due to the hydrophobic nature of the membrane, only DHLA will pass across the membrane while the less hydrophobic RLA will be retained, and thus provides a simple and effective online way of separating RLA and DHLA. Additionally, or alternatively, separation may be assisted by mixing a lipophilic solvent (preferably immiscible with electrolyte) or an acid to promote phase separation in to a lipophilic DHLA and an aqueous phase. In such configuration, dihydrolipoic acid will partition into the lipophilic solvent or organic phase which is then separated via the lipophilic membrane, typically using a relatively moderate pressure difference (e.g., 100 mbar). Alternatively, the lipophilic solvent may be omitted and separation may be carried out exclusively on the basis of different lipophilicity of the lipoic acid and dihydrolipoic acid as the electrochemically synthesized dihydrolipoic acid is less soluble in the reaction system than the starting material. Lipophilic solutions are able to pass through the PTFE membrane, while hydrophilic liquids will not, and the cross membrane pressure applied will further aid the separation of the reduced and oxidized products. Alternatively, the reduced catholyte may also be acidified to force the DHLA into the organic phase and the RLA in the aqueous phase. Separation can then be carried out in a simple phase separator. It should be appreciated that this general concept is applicable to all organic electrochemical processes and can be used for any process involving electrochemical oxidation/reduction followed by hydrophobic/hydrophilic separation.
With respect to the electrolyte, it should be appreciated that numerous electrolytes other than NaOH at a concentration of 1M are also deemed suitable, and it is generally preferred that the electrolyte is an alkaline electrolyte because of improved solubility and stability of RLA and DHLA. In addition, previous attempts to reduce RLA at a neutral or acidic electrolytes (2<pH<9) with glassy carbon, mercury, and platinum electrodes were not successful. Among other problems, such electrolytes often lead to undesired free radicals and polymers. In contrast, highly alkaline conditions and the cathode materials were found to improve various process parameters. The highly alkaline system was also found to be an ideal solution for the anode compartment for the rate limiting oxygen evolution reaction, which allowed using same solution for the anode and cathode compartments, which in turn helped avoiding concentrations gradient between the chambers. While the pH of the electrolyte may vary to some degree, it is generally preferred that the electrolyte has a pH of between 8-10, more preferably between 10-14, and most preferably between 12-14. Therefore, among other choices, alternative electrolytes include various alkaline metal hydroxide or alkaline earth metal hydroxide containing electrolytes at a concentration of between 0.01M to 0.1M, more preferably between 0.1M and 0.5M, and most preferably between 0.5M and 2M. Furthermore, it is noted that the electrolytes may be buffered or unbuffered, and may include additional ingredients (e.g., detergents, catalysts, etc.). In particularly contemplated aspects, the electrolyte is an aqueous electrolyte that may also include a non-aqueous solvent. Such non-aqueous solvents will typically be immiscible with water (less than 1% solubility of solvent in water) and form a separate phase. In particularly preferred aspects, the solvent is chosen such that the dihydrolipoic acid is preferentially dissolved in the solvent, and that the solvent together with the dihydrolipoic acid can pass across the membrane in the separation compartment while the remaining aqueous phase with the lipoic acid will not pass across the membrane. Alternatively, and especially where the difference in lipophilicity between the oxidized and reduced reagents is relatively large, additional solvent may be omitted. In such case, the more lipophilic reduced product will have a lipophilicity that allows the reduced product to selectively pass across the membrane in the separation compartment.
Numerous cathode materials are deemed suitable for use herein and include all materials that allow reduction of lipoic acid in alkaline electrolyte, including stainless steel, metal oxides, etc. However, particularly preferred cathode materials include high-surface materials, and especially carbon-containing high-surface materials. Additionally, it should be appreciated that particularly preferred cathode materials will allow a two-electron transfer reduction such that semiquinone intermediate products and polymer formation are substantially avoided (less than 5% of total products). The term “two-electron transfer reaction” refers to reductions in which the educt requires two electrons to form the product, and in which the product is formed in a single reaction (i.e., without measurable formation of an intermediate that has received only a single electron). Therefore, preferred cathodes include carbon felt, carbon fibers, graphite, vitreous carbon, and pyrolytic graphite, all of which may be part of a porous carrier structure or may be at least partially embedded in a conductive material. In preferred aspects, the cathode is configured as a flow-through cathode such that at least 50%, more typically at least 70%, and most typically at least 90% of the electrolyte flows through the body of the cathode. Still further, it is preferred that the cathode material has a configuration that allows storage of hydrogen within the cathode. While not wishing to be bound by any theory or hypothesis, it is contemplated that such stored hydrogen promotes two-electron transfer reductions as further elaborated below.
Similarly, the anode material may be made from numerous materials, however, it is generally preferred that the anode material will withstand the considerably harsh conditions without substantial degradation or loss of performance of the course of the electrochemical process. Therefore, suitable materials will generally be non-organic materials, and most preferably noble metals, noble metal-coated conductive materials, and metal oxides. For example, especially suitable anode materials include iridium oxide and nickel. Suitable separators include all known separators for electrochemical reactors. However, it is especially preferred that the separator is a cation exchange membrane to allow for proton flux into the anode compartment and cation (typically Na) flux into the cathode compartment.
In still further preferred aspects, the cathode compartment is optionally coupled to a fluid conduit from which electrolyte can be withdrawn in an on-line manner (i.e., while the reduction is in progress). At least part of the so withdrawn fluid is ultimately returned to the cathode compartment after at least some of the reduced product has been removed from the electrolyte in a separation device that is fluidly coupled to the cathode compartment via the fluid conduit(s). In such separation devices, the electrolyte may be combined with a solvent (typically immiscible with the electrolyte) for extraction of the reduced product and phase separation to thus remove the reduced product from the electrolyte. Where desirable, a mixing device may be included to promote intimate contact between the solvent and the electrolyte. Alternatively, separation may be performed in the separation compartment by a membrane that is preferentially (permeation ratio of reduced to non-reduced at least 5:1) or selectively (permeation ratio of reduced to non-reduced at least 100:1) permeable to the reduced product. Most typically, separation across the membrane is assisted by a pressure differential of at least 1 mbar, more typically at least 10 mbar, and most typically at least 100 mbar. It should be appreciated that there are numerous membranes suitable for use in conjunction with the teachings presented herein, however, especially preferred membranes include fluorinated polymeric membranes, and particularly PTFE (polytetrafluoroethylene). Consequently, it should be appreciated that separation of the dihydrolipoic acid from the lipoic acid can be performed in batch-wise manner or continuously.
In a typical experiment, 100-300 g of lipoic acid was dissolved in 1 liter of 1 M NaOH solution at 50-90° C. The solution was then brought to room temperature and electrochemically reduced at −4.7 V (cell voltage) using a carbon felt electrode of geometric area 100 cm in a continuous flow electrochemical cell. The limiting current was set to 10 A. Iridium oxide or Nickel was used as the counter electrode and 1M NaOH was used as the electrolyte in the anode compartment. After the reaction, the cathode electrolyte solution was transferred into a separating funnel and neutralized using 1 M HCl. The organic layer was analyzed using HPLC on a column that contained L1 packing (C18) and detection was via UV detection at 215 nm. RLA was 50% reduced after 24 hrs and almost 100% reduced after 48 hrs. Dissolved air was removed from the solution by bubbling argon through the cell for 30 minutes and later by passing argon over the solution during electrolysis.
The major thermodynamically feasible reactions in these electrochemical conditions are described below.
4OH—→O2+2H2O+4e− Eo=−0.401 (1)
O2+2H2O+4e−→4OH— Eo=0.401 (2)
2H2O+2e−→H2+2OH— Eo=−0.828 (3)
RLA+2e−+2H+→DHLA Eo=−0.29 (4)
It is therefore contemplated that the mechanism involved transfer of adsorbed hydrogen on the carbon felt to RLA. Carbon felt electrodes have recently been recognized as potential candidates for hydrogen storage. Especially in the alkaline electrolyte the carbon felt electrode can store a higher amount of hydrogen. Due to the high overvoltage value in alkaline medium (η=0.55) the recombination steps of Had leading to the molecular hydrogen evolution through the chemical (Tafel) or electrochemical (Heyrovsky) reactions are less favored than in an acid medium. Hence a meaningful sorption of hydrogen is observed in the alkaline electrolyte, which shows a reversible capacity of 350 mA h/g (1.3 wt %) of hydrogen storage with a good electrical efficiency. The carbon felt electrode is a unique cathode for this process. It has several advantages as compared to conventional metal electrodes: (1) No chemisorption, the thiol group in RLA can very easily absorbed on metal electrode surface and dissociate (2) High surface area (0.33 m2/g) due to its porous nature (3) high overpotential for hydrogen evolution reaction (4) The micro-porous nature allows transport of fluids and effective hydrogen transfer, and (5) The absorption process on the carbon electrode surface inhibits the polymer formation on carbon electrode surface unlike the metal electrodes.
The reduction reaction presented herein follows an interesting mechanism in which the first thermodynamically possible reaction on the cathode is the oxygen reduction. This can be avoided by using a nitrogen flow during electrolysis or by removing the oxygen from the electrolyte prior to the electrolysis using sodium dithionite. Both of these treatments were found to enhance the overall efficiency of the process. Although the electrochemical reduction of RLA is a thermodynamically favorable process compared to HER, a direct reduction does not occur on the electrode surface. The reduction of RLA is starting only after the pores of the C-felt electrode are saturated with hydrogen. This is further supported by the fact that, the kinetics of the process can be further enhanced by using a pre-treatment. The pretreatment phase involves the electrolysis in KOH without RLA. This helps in cleaning up the pores of the carbon felt electrode and also in increasing the amount of hydrogen stored prior to the electrolysis of RLA, which accelerates the process be avoiding the time lag due the unavailability of adsorbed hydrogen. This mechanism is supported by the increase in the NaOH concentration observed after the electrolysis in the cathode compartment. Since OH− ion concentration is increasing due to reactions 2 and 3 in the anode compartment, Na+ ions are migrating from the anode compartment and H+ ions are migrating to the anode compartment to maintain the charge balance. This is further supported by the decrease in the pH at the anode compartment. The rate of the process is improved by maintaining the pH of the anode compartment above 13 using an automated system. This allows keeping a constant current throughout the process and a continuous transfer of hydrogen to RLA.
In the anode compartment oxygen evolution reaction (1) is the most plausible reaction. The rate of the reaction is controlled by the pH ion concentration. Due to the H+ ion migration from the anode compartment, the pH is decreasing with time and it is therefore of significance to keep a constant pH where steady kinetics for the process are desired. Ir2O3 is found to be the best electrode for this process, although Ni or other similar electrodes can be used. The overall kinetics of the process is controlled by the oxygen evolution reaction, at least during the initial stages. The reaction was found to stop completely once the pH of the anode compartment came to neutral as a result of the oxygen evolution reaction shown above. Interestingly the reaction could be restored by replacing the solution with fresh prepared 1M NaOH solution. Remarkably, the reaction started only after the pores of the electrodes were saturated with hydrogen. Thus, it should be noted that the carbon felt provides a unique electrode surface for the process. Further, it was found that the electrode was stable during the entire process. No mechanical or chemical damage was observed on the electrode even after several electrochemical processes as can be seen in
Since RLA reduction is a two electron process, 96485 Coulombs (26.8 Ahr) will reduce 103 g of RLA (mwt 206). 300 g of RLA theoretically needs 78 Ahr.
Thus, specific devices and methods of electrochemical synthesis of dihydrolipoic acid have been disclosed. It should be apparent, however, to those skilled in the art that many more modifications besides those already described are possible without departing from the inventive concepts herein. The inventive subject matter, therefore, is not to be restricted except in the spirit of the appended claims. Moreover, in interpreting both the specification and the claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms “comprises” and “comprising” should be interpreted as referring to elements, components, or steps in a non-exclusive manner, indicating that the referenced elements, components, or steps may be present, or utilized, or combined with other elements, components, or steps that are not expressly referenced. Furthermore, where a definition or use of a term in a reference, which is incorporated by reference herein is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies and the definition of that term in the reference does not apply.
