Reversible carboxylation of N-heterocyclic carbenes

Hung A. Duong , Thomas N. Tekavec , Atta M. Arif and Janis Louie *
Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City, UT, USA 84112. E-mail: louie@chem.utah.edu; Fax: (801) 581-8433; Tel: (801)581-7309

Received (in Cambridge, UK) 17th September 2003 , Accepted 28th October 2003

First published on 10th November 2003


Abstract

Spectroscopic analysis, thermogravimetric analysis, and crossover experiments performed on a series of imidazolium carboxylates revealed carboxylation was reversible with N-aryl substituted adducts.


Since their isolation by Arduengo in 1991,1 N-heterocyclic carbenes (NHCs) have been widely used compounds in several areas in chemistry. For example, when used as ligands for transition metals, they often enhance performance and selectivity.2 Independently, they have been shown to function as nucleophilic organic catalysts for transesterifications,3,4 acylations,5 and ring-opening polymerizations.6 Although NHCs are typically air and moisture sensitive, Waymouth demonstrated how a biphasic system that generates catalytic amounts of NHCs from imidazolylidene-based ionic liquids can be used to circumvent the need to use Schlenk techniques.7 Similar biphasic systems have been recently integrated with supercritical CO2 to create environmentally friendly separation systems.8–10 While studying the use of NHCs as ligands for metal-mediated cycloadditions,11 we discovered that NHCs freely add CO2 to form the corresponding imidazolium carboxylates. Herein, we report the synthesis of a series of imidazolium carboxylates and present evidence which indicates that their formation is reversible. These results should be of prime importance to the aforementioned programs.

There are very few examples in the chemical literature of synthesizing and studying imidazolium carboxylates. Although Kuhn demonstrated that the direct addition of CO2 to 1,3-diisopropyl-4,5-dimethylimidazol-2-ylidene (iPrIm) cleanly afforded the corresponding carboxylate,12 mechanistic details were not provided. The only other known example was reported by Holbrey in their “unexpected synthesis” of 1,3-dimethylimidazolium-2-carboxylate from the reaction between 1-methylimidazole and dimethyl carbonate.13,14

We discovered that the addition of CO2 (1 atm) to iPrIm, 1,3-dimesitylimidazol-2-ylidene (IMes), or 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene (IPr) afforded the corresponding zwitterionic imidazolium carboxylate quantitatively (Scheme 1, Pathway A). However, carboxylates 2 (IMes CO2) and 3 (IPrCO2) were also prepared using a more convenient protocol. Simply deprotonating the corresponding salts with potassium tert-butoxide under an atmosphere of CO2 (Pathway B) afforded the products in high yields (75–95%). All carboxylates were sparingly soluble in THF, aromatic solvents, and water. However, carboxylates 2 and 3 displayed enhanced solubilities in methanol, DMSO, and chlorinated solvents. Although air-stable as solids, solutions of 2 and 3 decomposed readily in aerated solutions.


Synthesis of imidazolium carboxylates 1, 2 and 3.
Scheme 1 Synthesis of imidazolium carboxylates 1, 2 and 3.

The X-ray crystal structure of carboxylate 3 was determined15 and is shown in Fig. 1.§ The near equivalent C–O bond distances (1.221(4) and 1.225(4) Å) indicated that the negative charge was equally distributed between the central carbon (C6) and the two oxygens (O1 and O2). In contrast, the two C–N bond distances are substantially different (1.334(4) Å vs. 1.376(4) Å) which suggested some delocalization of the positive charge between a single N atom and the aryl ring. There appeared to be no delocalization of the carboxylate charge into the imidazole ring as its plane was perpendicular to the plane of the CO2 group. The crystal structure also indicated that the methine protons on the isopropyl groups were twisted inward toward the imidazolylidene plane. This was also observed in solution as two doublets for the isopropyl methyls were observed in the 1H NMR spectrum of compound 3. Combined, this suggested that their rotation was restricted.


Molecular structure of IPrCO2
					(3). Selected bond lengths (Å) and angles (°): C(6)–O(1) 1.225(4), C(6)–O(2) 1.221(4), C(2)–N(1) 1.334(4), C(2)–N(3) 1.332(4), O(1)–C(6)–O(2) 129.9(3), N(1)–C(2)–N(3) 107.1(2).
Fig. 1 Molecular structure of IPrCO2 (3). Selected bond lengths (Å) and angles (°): C(6)–O(1) 1.225(4), C(6)–O(2) 1.221(4), C(2)–N(1) 1.334(4), C(2)–N(3) 1.332(4), O(1)–C(6)–O(2) 129.9(3), N(1)–C(2)–N(3) 107.1(2).

Using 13C NMR analysis, signals attributable to the carboxylate carbon were observed in all complexes. In addition, no free IPr or CO2 was observed in solution, even after prolonged periods of time, which indicated a high coordination constant between the two species.16 Furthermore, the addition of 13CO2 (1 atm) to a solution of IPrCO2 led to an enhancement in the carbonyl peak of the IPrCO2 and suggested that the bound CO2 was equilibrating with free 13CO2. As expected, reintroduction of CO2 (1 atm) attenuated the signal.

Thermogravimetric analysis provided further evidence for decarboxylation in 2 and 3. Clean loss of 10.6% of the weight of IPrCO23 (corresponding to the mass of CO2) was observed between 136 and 164 °C (Fig. 2). Decomposition occurred at 250 °C. IMesCO22 appeared to have a higher stability toward decarboxylation as weight loss began at 187 °C and continued until decomposition ensued (203 °C). No decarboxylation was observed for iPrImCO2 (1) which decomposed at 160 °C.


TGA of IPrCO2
					(3) under N2 atmosphere. Ramp rate = 5 °C min−1.
Fig. 2 TGA of IPrCO2 (3) under N2 atmosphere. Ramp rate = 5 °C min−1.

Crossover experiments between free imidazolylidenes and their carboxylates revealed a trend in their relative stabilities (Equations 1–5). The addition of iPrIm to either IMesCO2 (2) or IPrCO2 (3) led to quantitative formation of iPrImCO2 (1) and IMes or IPr, respectively. The reverse addition of IMes or IPr to iPrImCO2 (1) gave no reaction. Similarly, the addition of IMes to IPrCO2 (3) led to the quantitative formation of free IPr and IMesCO2 (2), but the reverse reaction was not observed. Thus, the relative order of adduct stability appeared to be: iPrImCO2 (1) > IMesCO2 (2) > IPrCO2 (3).17 The same trend in relative stabilities was extracted from our TGA data.

 
IMesCO2 (2) + iPrIm → IMes + iPrImCO2 (1)(1)
 
IPrCO2 (3) + iPrIm → IPr + iPrImCO2 (1)(2)
 
ugraphic, filename = b311350g-t1.gif(3)
 
IPrCO2 (3) + IMes → IMesCO2 (2) + IPr(4)
 
IMesCO2 (2) + IPr → no reaction(5)

In conclusion, free N-heterocyclic carbenes were converted to their respective imidazolylidene carboxylates by introducing them to CO2. The adducts were found to readily exchange their carboxylate groups with free CO2 in solution and with relatively less stable imidazolium carboxylates. Likewise, adduct stability also affected the onset temperature of a thermally induced decarboxylation. Collectively, our data indicate that the formation of imidazolium carboxylates may be reversible under certain conditions. These results should be considered when working with imidazolylidene or NHC-based ligands or ionic liquids in the presence of CO2.

We gratefully acknowledge the University of Utah and the Petroleum Research Fund (Type G) for support of this research.

Notes and references

  1. A. J. Arduengo, R. L. Harlow and M. J. Kline, J. Am. Chem. Soc., 1991, 113, 361 CrossRef.
  2. W. A. Herrmann, T. Weskamp and V. P. W. Bohm, Adv. Organomet. Chem., 2001, 48, 1 CAS.
  3. G. W. Nyce, J. A. Lamboy, E. F. Connor, R. M. Waymouth and J. L. Hendrick, Org. Lett., 2002, 4, 3587 CrossRef.
  4. G. A. Grasa, R. M. Kissling and S. P. Nolan, Org. Lett., 2002, 4, 3583 CrossRef.
  5. G. A. Grasa, T. Guveli, R. Singh and S. P. Nolan, J. Org. Chem., 2003, 68, 2812 CrossRef CAS.
  6. E. F. Connor, G. W. Nyce, A. Möck and J. L. Hendrick, J. Am. Chem. Soc., 2002, 124, 914 CrossRef CAS.
  7. G. W. Nyce, T. Glauser, E. F. Connor, A. Möck, R. M. Waymouth and J. L. Hedrick, J. Am. Chem. Soc., 2003, 125, 3046 CrossRef CAS.
  8. T. Welton, Chem. Rev., 1999, 99, 2071 CrossRef CAS.
  9. A. M. Scurto, S. N. V. K. Aki and J. F. Brennecke, J. Am. Chem. Soc., 2002, 124, 10276 CrossRef CAS.
  10. F. Liu, M. B. Abrams, R. T. Baker and W. Tumas, Chem. Commun., 2001, 433 RSC.
  11. J. Louie, J. E. Gibby, M. V. Farnworth and T. N. Tekavec, J. Am. Chem. Soc., 2002, 124, 15188 CrossRef CAS.
  12. N. Kuhn, E. Niquet, M. Steimann and I. Walker, Z. Naturforsch., Teil B, 1999, 54, 427 CAS.
  13. J. D. Holbrey, W. M. Reichert, I. Tkatchenko, E. Bouajila, O. Walter, I. Tommasi and R. D. Rogers, Chem. Commun., 2003, 28 RSC.
  14. In addition, a dihydroimidazolylidenecarboxylate is known, see: W. Schössler and M. Regitz, Chem. Ber., 1974, 107, 1931 Search PubMed.
  15. Carboxylate 3 has a similar structure to 1,3-dimethylimidazolium-2-carboxylate and 1,3-diisopropyl-4,5-dimethylimidazolium-2-carboxylate (1). A notable difference is that the CO2 groups on the latter complexes are not perpendicular to the imidazole plane.
  16. K. A. Connors, Binding Constants, John Wiley & Sons, New York, 1987 Search PubMed.
  17. Semi-empirical (PM3) calculations performed using the Titan Software Program (Wavefunction, Inc., Irvine, CA 92612, USA) support this trend.
  18. N. Kuhn and T. Kratz, Synthesis, 1993, 561 CrossRef CAS.

Footnotes

Electronic supplementary information (ESI) available: crystal structure data and ORTEP for 1,3-bis(2,6-diisopropylphenyl)imidazolylium-2-carboxylate (3). See http://www.rsc.org/suppdata/cc/b3/b311350g/
iPrIm, IMes, and IPr were prepared using literature procedures.1,18General procedure for the synthesis of 1, 2 and 3 through direct carboxylation of the carbenes: In a Schlenk flask, a solution of N-heterocyclic carbene in ether was cooled to −78 °C and CO2 (1 atm) was added via condensation. The solution was warmed to room temperature and a white precipitate formed. The precipitate was collected by filtration and washed with ether to afford the imidazolylidene carboxylate in 85, 95, and 90% yields for 1, 2 and 3, respectively.2: δH(500 MHz, DMSO-d6) 2.09 (s, 12H, ArCH3), 2.31 (s, 6H, ArCH3), 7.08 (s, 4H, ArH), 7.87 (s, 2H, NCHCHN). δC(126 MHz, DMSO-d6): 17.0, 20.8, 125.0, 129.5, 131.1, 134.5, 138.6, 140.8, 149.2. IR (KBr): 1675 cm−1 (vs). mp: 315 °C (dec). Anal. Calc. for C22H24N2O2: C, 75.83; H, 6.94; N, 8.04. Found: C, 75.57; H, 6.98; N, 7.89%.3: δH(300 MHz, CD2Cl2) 1.22 (d, 12H, 3JH–H = 6.9 Hz, CH(CH3)2), 1.28 (d, 12H, 3JH–H = 6.6 Hz, CH(CH3)2), 2.53 (sept., 2H, CH), 7.19 (s, 2H, NCHCHN), 7.35 (d, 4H, 3JH–H = 7.8 Hz, ArH), 7.55 (t, 2H, 3JH–H = 7.8 Hz, ArH). δC(75.6 MHz, CD2Cl2): 23.4, 24.3, 29.4, 122.9, 124.3, 130.8, 132.8, 145.1, 147.6, 152.3. IR (KBr): 1678 cm−1 (vs). mp: 216 °C (dec). Anal. Calc. for C28H36N2O2: C, 77.74; H, 8.39; N, 6.48. Found: C, 77.71; H, 8.34; N, 6.42%.General procedure for the synthesis of 2 and 3 through carboxylation of the imidazolium salts: An oven-dried two-neck round-bottomed flask equipped with a magnetic stir bar, septum, gas adapter and balloon was evacuated and filled with CO2. A solution of the imidazolylidene salt and potassium tert-butoxide (1.2 equiv.) was then added and the reaction was stirred at room temperature for 2–6 hours. For the synthesis of 2, the solvent was removed and the residue was extracted with CH2Cl2. The solution was filtered through Celite and concentrated in vacuo. Subsequent washing with Et2O afforded the imidazolium carboxylate as a white solid in 95% yield. For the synthesis of 3, hexane was added to induce precipitation. The precipitate was collected via filtration and dissolved in CH2Cl2. The solution was then filtered through Celite and concentrated in vacuo. Subsequent washing with Et2O afforded the imidazolium carboxylate as a white solid in 75% yield.
§ Crystal data for C28H36N2O23: M = 432.59, monoclinic, a = 12.6385(7), b = 12.5779(8), c = 16.6868(12) Å, β = 92.993(2)°, U = 2649.0(3) Å3, T = 150 K, space group P21/n, Z = 4, µ(Mo-K) = 0.71073 Å, 9106 reflections collected, 4822 unique (Rint = 0.0721) which were used in all calculations. The R1 was 0.1638 and the final wR(F2) was 0.2314 (all data). Single crystals of 3 were grown from slow diffusion of CO2 into a solution of IPr in THF at room temperature. The crystals were mounted in inert oil and transferred to a cold gas stream in the diffractometer. The structure was solved by a combination of direct methods and heavy atoms using SIR 97 and refined by full-matrix least-squares on F2. CCDC 219086. See http://www.rsc.org/suppdata/cc/b3/b311350g/ for crystallographic data in .cif or other electronic format.

This journal is © The Royal Society of Chemistry 2004
Click here to see how this site uses Cookies. View our privacy policy here.