The main reason most CO₂ reduction catalysts fail to produce carbon-carbon bonds is that the catalyst cannot produce a high local concentration of the reduced form of CO₂. The majority of the CO₂ reduction catalysts reported in the literature are able to produce the radical anion of CO₂; however, they are not able to produce two reduced molecules in close enough proximity to react with each other before they react with a normal CO₂ molecule. The key to overcoming this problem would be to manipulate the chemical environment to produce a high local concentration of reduced CO₂^-o radicals.

One research group, Kubiak et al., has successfully performed the above manipulation.¹¹ The unique "trick" Kubiak employed to beat the disproportionation reaction was to form a catalytic dimer by chemically tethering two catalytic centers in close proximity to each other (see Scheme 2). The tether he used has a seven-carbon spacer, which puts the two catalytic centers as close as 10Å from each other. In this way he was able to generate locally "high" concentrations of CO₂^-o. Considering the fact that the two catalytic centers were located in distinct solvation spheres, there was the concern that when the two CO₂^-o radicals left the local proximity of the catalytic center, they might still encounter a CO₂ (normal) before they found each other, and the catalytic dimer idea would fail. However, after several hours of catalysis, Kubiak was able to isolate several milligrams of oxalate. ¹¹

Scheme 2

Using the principles employed by Kubiak, et al., we designed a novel catalytic system for the reduction of carbon dioxide. Our catalyst, hereafter referred to as Cat-A,^D is distinct from Kubiak's in that it places two active sites on the same metal center. By doing this, Cat-A has the potential to reduce two CO₂ molecules in the same solvation sphere. Cat-A has successfully electrolytically reduced CO₂ and produced locally high concentrations of reduced carbon dioxide molecules. As a result, these CO₂^-o have dimerized to form oxalate, as will be seen in data to follow.

To electrolytically reduce a chemical species, an electrical voltage or current must be applied. In experimentation with Cat-A and its analogs, a negative voltage is applied to an electrode in contact with the solution containing the catalyst. This solution contains a set amount of both the catalyst and either potassium hexafluorophosphate (KPF₆) or tetrabutylammonium hexafluorophosphate (TBAP), which are used as the supporting electrolyte. The solvent is distilled acetonitrile which has been dried over molecular sieves. A standard three-electrode configuration is used in all electrochemistry experimentation. In micro-scale experiments, there is a platinum working electrode, a tungsten wire counter electrode, and a leakproof silver/silver chloride counter electrode. For large-scale experiments, a platinum mesh electrode is used as the working electrode, a platinum wire is the counter, and the same reference electrode is used as above. Schematics of the two experimental set-ups can be seen in Figures 1 and 2, respectively.

Figure 1. Micro-scale Cell	Figure 2. Bulk Cell (BAS)

Using the micro-scale cell as described above and TBAP as the supporting electrolyte, cyclic voltammagrams (CVs) of Cat-A, Cat-B, and Cat-C were performed. In CVs, a given electric potential is applied to the solution, the potential then moves out to a set point, and then it returns to the original potential. The change in current as a result of the applied voltage is measured. A CV is first run with the given catalyst without the presence of CO₂. In Figures 3, 4, and 5 below, this CV appears as the blue line. Next, the same solution is purged with CO₂ gas and another CV is taken; in the figures below, the green line represents the CV taken in the presence of CO₂.

Figure 3. CVs with Cat-A	Figure 4. CVs with Cat-B

Figure 5. CVs with Cat-C

The important feature to note in these comparisons of CVs with and without the presence of CO₂ is the increase in current in runs with CO₂ present (green). This spike in the current is indicative of catalytic regeneration of the oxidized form of the catalyst; in other words, as Cat-A, B, or C becomes reduced at the electrode surface, the catalytic formation of CO₂^-o regenerates the catalyst in its oxidized form and allows it to take another electron from the electrode. The increase of current on the addition of CO₂ indicates that we have successfully electrolytically reduced CO₂.

The products of this catalytic reduction of CO₂ were isolated from the CV solutions and analyzed using Gas Chromatography/Mass Spectrometry (GC/MS). If oxalate had indeed been formed during the process, then it would be present in the form of potassium oxalate. This is a white crystalline solid; in order to confirm its presence, we had to convert it into a form that had a vapor pressure and could be taken into phase. This conversion process is given below in Scheme 3.

Scheme 3. Conversion of K₂C₂O₄ to Diethyloxalate

Once the potassium oxalate was converted to diethyloxalate, it could be analyzed by (GC/MS). A GC/MS basically takes a complex mixture and separates each component into a pure form, which is then fragmented in the MS. The MS portion of the machine then evaluates the mass of the total component and its fragments. By comparing the results of the GC/MS data obtained from our experiment to data obtained from a commercially available sample of diethyloxalate, we were able to conclude that Cat-A had indeed reduced CO₂ and formed oxalate. This comparison is given below in Figure 6.

Figure 6. GC/MS of commercially obtained diethyloxalate (bottom);
GC/MS of diethyloxalate obtained from catalytic reduction of CO₂ (Top)

Using the bulk cell as described in Figure 2 and KPF₆ as the supporting electrolyte, potentiostatic experiments were conducted using Cat-A, Cat-C, and Cat-D. In potentiostatic experiments, the potential is set at a fixed voltage. In our case, this potential was between the range of -1.3 V and -2.0 V, depending on the run and the catalyst used. The potential was held constant, and the cell had a constant purge of CO₂ for the course of the experiment, which ranged from two to twelve hours. The products from these experiments were converted using the same process used for the micro-scale experiments. A representative GC trace of the reaction products of these experiments is given below in Figure 7. The molecules shown pointing to the different peaks indicate molecules that matched a library database with a confidence limit of at least 95%. The first two large molecules on the right are similar to phthalic acid derivatives. The cluster of peaks between 28 and 29 minutes is a group of phenol derivatives that have several different side chains, indicated by the R. The last molecule is a long chain alcohol, 2-ethyl-1-hexanol.

Figure 7. GC data for one of the bulk runs, Cat-D used

It appears that instead of producing oxalate, which is a simple one carbon-carbon bond molecule, we have succeeded in making numerous carbon-carbon bonds with our catalyst. These complex molecules, shown in Figure 7 above, are very similar in structure to those commonly found in plants. Although research is in progress to determine the exact reaction products and the process by which they are formed, there is a plausible mechanism for this reaction. We speculate that the more complex molecules are formed from the resulting oxalate anchoring itself onto the metal center of the catalyst, followed by free radical chain polymerization of additional carbon dioxide radical anions into the carbon-oxygen double bond of the oxalate. In the formation of bonds between the anchored molecule and the free reduced molecules of CO₂ in this reaction scheme, two oxygen molecules would have to be given off. In several of the experimental runs, bubbles were observed coming off of the electrodes, which could have been oxygen gas, O₂. Further investigation is underway.

With this research, it appears that we have made great strides in the direction of understanding the complex processes behind photosynthesis. It appears from the data that we have in fact produced an artificial photosynthesis catalyst. With continuing research into Cat-A, B, C, and D, we hope to produce a catalytic system that is both economically and practically feasible for use in closed environments such as space and underwater stations.

End Notes:
^s The term reduction is used whenever an electron is added to a molecule i.e. CO₂ + e- ›CO₂^-o.
^¥ Dimer means two, and the term dimerization is the process of connecting two discreet units to form a larger molecule.
^b The term disproportionation is used when one part of a molecule is involved in an electrochemical reaction with another part of the same molecule. The result is a splitting of the parent molecule and two new pieces, one oxidized and the other reduced.
^§ The name and chemical formula of our catalyst is proprietary information. Therefore, for the purpose of this paper, the catalyst will be referred to as Cat-A. Forms of the catalyst with different metal centers will be referred to in the same manner using subsequent letters.

References:
1) Musashi, Y.; Sakaki, S. J. Am. Chem. Soc. 2000, 122, 3867-77.
2) Grodkowski, J.; Neta, P. J. Phys. Chem. A 2000, 104, 1848-53.
3) Abbott, A.P.; Eardley, C. A. J. Phys. Chem. B 2000, 104, 775-9.
4) Kelly, C. A.; Blinn, E. L.; Camaioni, N. Inorg. Chem. 1999, 38.
5) Hirota, K.; Tryk, D. A.; Yamamoto, T. J. Phys. Chem. B 1998, 102, 9834-43.
6) Paul, P.; Tyagi, B.; Bilakhiya, A. K. Inorg. Chem. 1998, 37, 5733-42.
7) Behar, D.; Dhanasekaran, T.; Neta, P. J. Phys. Chem. A 1998, 102, 2870-7.
8) Chardon-Noblat, S.; Deronzier, A.; Ziessel, R. Inorg. Chem. 1997, 36, 5384-9.
9) Cheng, S. C.; Blaine, C. A.; Hill, M. G. Inorg. Chem. 1996, 35, 7704-8.
10) Mochizuki, K.; Manaka, S.; Takeda, I. Inorg. Chem. 1996, 35, 5132-6.
11) Morgenstern, D. A, Wittrig, R. E, Fanwick, P. E., Kubiak, C. P., J. Am. Chem. Soc., 1993, 115, 6470-6471.
12) Oral presentation at Purdue University by Al Hepp. Electrical Engineer and Director of Photovoltaic studies at NASA Lewis Research Station. Cleveland, OH. 1991.
13) (a) Hepp, A. F.; Landis, G. A.; Kubiak, C. P. Proc. Intersoc. Energy Convers. Eng. Conf., 1991, 26, 68-73.
(b) Hepp, A. F.; Landis, G. A.; Kubiak, C. P. NASA Tech. Memo. 1991. 22pp.
14) Oxidation and Reduction of CO. Eisenberg, R.; Kubiak, C. P. Inorganic Reactions and Methods, VCH Publishers, A. D. Norman, ed., 1993.
15) Pearce, D. J.; Pletcher, D. J. Electroanal. Chem. 1986, 197, 317.
16) A Chemical Approach to Carbon Dioxide Utilization on Mars. Hepp, A. F.; Landis, G. A.; Kubiak, C. P. Resources of Near Earth Space, University of Arizona Press, Tuscon, Ariz., J. Lewis, M. S. Matthews, and M. L. Guerrieri, eds., 1993, 799-818.
17) Eisenbert, R.; Hendriksen, D. E. Adv. Catal. 1979, 28, 79.
18) Palmer, D. A.; van Eldik, R. Chem Rev. 1983, 83, 651.
19) Darensbourg, D. J.; Kudaroski, R. A. Adv. Organomet. Chem. 1983, 22, 129.
20) Inoue, S.; Koinuma, H.; Rev. Inorg. Chem. 1984, 6, 291.
21) Lundquist, E. G.; Huffman, J. C.; Caulton, K. G.; J. Am. Chem. Soc., 1986, 108, 8309.
22) Gressin, J. C.; Michelet, D.; Nadjo, L.; Saveant, J. M. Nouv. J. Chem. 1979, 3, 545.
23) Lamy, E.; Nadjo, L.; Saveant, J. M. J. Electroanal. Chem. 1977, 78, 403.
24) Saveant, J. M.; Amatore, C. J. Am. Chem. Soc. 1981, 103, 5021-5023.
25) Froehlich, H.O.; Schreer, H. Ger (East) Patent 1985, DD 219, 488.

return to table of contents

This page updated March 1, 2004