Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 1.
Published in final edited form as: Angew Chem Int Ed Engl. 2015 Oct 16;54(49):14974–14977. doi: 10.1002/anie.201508777

O2 Activation by Non-adiabatic Oxidative Addition to a Single Metal Center

Eser S Akturk, Glenn P A Yap, Klaus H Theopold a,✉
PMCID: PMC4894523  NIHMSID: NIHMS753168  PMID: 26473394

Abstract

A chromium(I) dinitrogen complex reacts rapidly with O2 to form the mononuclear dioxo complex TptBu,MeCrV(O)2, whereas the analogous reaction with sulfur stops at the persulfido complex TptBu,MeCrIII(S2). The transformation of the putative peroxo intermediate TptBu,MeCrIII(O2)(S = 3/2) into TptBu,MeCrV(O)2 (S = ½) is spin forbidden. The minimum energy crossing point on the two potential energy surfaces has been identified. While the dinuclear complex (TptBu,MeCr)2(μ-O)2 exists, mechanistic experiments suggest that the O2 activation occurs on a single metal center, by an oxidative addition on the quartet surface followed by crossover to the doublet surface.

Keywords: Chromium, O-O activation, spin forbidden reaction, sulfur, ab initio calculations

TOC image

The 4-electron oxidative addition of O2 to TpCr(I) is spin forbidden. And yet, it proceeds via a non-adiabatic crossing from a quartet to a doublet potential energy surface at the ‘minimum energy crossing point’ with the structure shown here.

graphic file with name nihms753168u1.jpg

In the context of aerobic oxidations of organic compounds there arises the issue of spin conservation, due to the triplet ground state of dioxygen.[1] This restriction may be circumvented by the ‘activation’ of O2 with the aid of metal catalysts.[2] Thus the formation of dioxygen complexes, and their subsequent transformation into highly reactive metal oxo species has been the focus of intense research interest for many years.[3] However, both the formation and the transfer of oxo moieties may also involve elementary reactions that are ‘spin forbidden’.[4] We have been interested in the mechanistic consequences of such occurrences.[5] Herein we describe the activation of O2 (and S8) by low-valent chromium complexes supported by tris(pyrazolyl)borate ligands. Specifically, we suggest that the oxidative addition of O2 to mononuclear Cr(I) proceeds, despite facing a spin forbidden elementary step.[6]

We have recently described the synthesis and reactivity of several complexes of chromium in its formal oxidation state +I, including the dinuclear dinitrogen complex [TptBu,MeCr]2(μ-η1:η1-N2) (1) and the even more labile alkyne complex TptBu,MeCr(η2-C2(SiMe3)2) (2).[7] Upon exposure to O2 gas, cold (−78°C) THF solutions of 1 gradually turned from green to brown, yielding the chromium oxo complex TptBu,MeCr(O)2 (3), i.e., the apparent product of a 4-electron oxidative addition of O2 to the metal. The structure of 3 is shown in Figure 1a;[8] it features approximately square pyramidal coordination of chromium, with N5 occupying the apical position. The average Cr–O distance of 1.603(2) Å is in the middle of the range established by chromium oxo complexes.[9] Befitting its d1 electronic configuration, the effective magnetic moment of 3 at room temperature was μeff(RT) = 1.9(1) μB. By contrast, the room temperature reaction of 1 with sulfur produced a deep red solution, from which TptBu,MeCr(S2) (4) was isolated in 43% yield. The structure of 4 (see Figure 1b) approximates trigonal bipyramidal coordination of chromium with S1 and N1 occupying the axial positions.[10] The molecule retains a bond between the two sulfur atoms; at 2.0565(18) Å the S–S distance is consistent with a single bond such as would be expected in a chromium persulfido (S22−) ligand.[11] A band at 541 cm−1 in the IR spectrum (KBr) of 4 has been tentatively identified as the S–S stretching frequency (νS–S) of the ligand. Chromium’s formal oxidation state in 4 is thus +III and the magnetic moment of 4 − μeff(RT) = 3.7(1) μB – is consistent with the quartet ground state of a d3 ion.

Figure 1.

Figure 1

Open in a new tab

Molecular structures of a) 3 and b) 4 (30 % probability level). Selected interatomic distances (Å) and angles(°): a) Cr–O1, 1.6007(19); Cr–O2, 1.6050(19); Cr–N1, 2. 216(2); Cr–N3, 2.265(2); Cr–N5(1.979(2); O1–Cr–O2, 108.14(10); b) Cr–S1, 2.3220(15); Cr–S2, 2. 2398(16); S1–S2, 2.0565(18), Cr–Navg, 2.109; S1–Cr–S2, 53.55(6).

The juxtaposition of congeners 3 and 4 is striking, but makes perfect chemical sense. Whereas oxygen is oxidizing enough to stabilize the +V oxidation state of chromium, reaction with the less electronegative sulfur halts at the lesser +III oxidation state of the metal. This explanation is also supported by specific computations. Thus, we have carried out DFT calculations on TpCr(E2) and TpCr(E)2 (Tp = HB(C3N2H3)3, E = O, S).[12] We find that the conversion of putative TpCrIII(O2) to TpCrV(O)2 (i.e., 3’) is exoergic by ΔG = −23.8 kcal/mol, while the analogous transformation of TpCrIII(S2) (i.e., 4’) to TpCrV(S)2 is thermodynamically unfavorable to the tune of ΔG = +70.8 kcal/mol. Furthermore, we note that the peroxo complex TpCrIII(O2) (A’, tbp geometry, dO-O = 1.464 Å, νO-O = 908 cm−1, S = 3/2) is structurally closely related to 4’. Its physical analog, TptBu,MeCr(O2) (A), is a prime candidate for an early intermediate in the formation of 3, and persulfido complex 4 serves as a near perfect model for it.

This is where the spin conservation issue may impact the mechanism of the dioxygen activation! A simple unimolecular O-O bond cleavage suffices to transform A into 3. And yet, the former has a quartet ground state (4A) and the latter is a doublet (23), rendering this elementary reaction step spin forbidden. The question then is whether this surface crossing is the path of lowest energy or if the reaction proceeds by an alternate – more complicated mechanism. The estimation of the activation barrier and rate of a non-adiabatic reaction transitioning from one potential energy surface (PES) to another involves the identification of ‘minimum energy crossing points’ (MECPs); i.e., the points of lowest energy on the seam forming the intersection of two multidimensional PESs (here the quartet and the doublet state of TpCrO2). Methods for finding MECPs have been described, and we have used the algorithm developed by Harvey et al.[13] Starting with a structure intermediate between 4A’ and 23’, we have located a MECP (X’) – see Figure 2. The structure of X’ features an O⃛O distance of 2.363 Å and Cr–O distances of 1.625 and 1.836 Å, and it has an energy only 5.3 kcal/mol above that of 43’. Thus X’ is rather product-like. We have also calculated the transition state (T’) for the conversion of 4A’ to 43’ on the quartet PES. This reaction has ΔG‡ = 19.5 kcal/mol, and the activated complex (O⃛O = 1.900 Å, Cr–Oavg = 1.765 Å) resembles the reactant more than does X’.

Figure 2.

Figure 2

Open in a new tab

Reaction coordinate diagram for the ‘spin forbidden’ transformation of A into 3. Structures shown are those calculated with DFT, showing the evolution of the O⃛O distance.

In keeping with Harvey’s ‘third rule’, which posits that “for reactions involving a change in atom connectivity as well as a spin state change, the preferred mechanism will usually occur in more than one step, with a spin crossover step preceding or following the bond connectivity change step”,[4f] we postulate that the oxidative addition of the peroxo ligand of A proceeds on the quartet surface, followed by a spin crossover converting 43 into 23.

A well-precedented mechanistic alternative for the cleavage of O2 would involve binuclear intermediates. Thus, trapping of A with a second equivalent of TptBu,MeCrI might generate (TptBu,MeCr)2(μ-O2), followed by cleavage of the O-O bond to yield (TptBu,MeCr)2(μ-O)2 and subsequent reaction with additional O2 to produce 3. It was thus of interest to explore the existence of such a binuclear species. Figure 3 shows the product of relevant experiments – e.g., the reaction of 3 with 1, or the treatment of 1 with only 1.0 equivalent of O2 or with an excess of N2O.[14] (TptBu,MeCr)2(μ-O)2 (5) is a binuclear chromium complex featuring a Cr2(μ-O)2 diamond core, which exhibits significant bond length alternation. Of particular note is the fact that one of the Tp ligands (on Cr2) is coordinated via two pyrazolyl groups and one agostic Cr–H–B interaction – the latter also manifests in a red-shifted νB–H at 2035 cm−1. The likely origin of this rearrangement is severe steric congestion caused by six tert-butyl groups brought into close proximity by the close approach of the two metal atoms (dCr–Cr = 2.739(2) Å). If nothing else, this might suggest that the formation of 5 faces a significant activation barrier, leaving some room for the unimolecular – albeit spin forbidden alternative. However, 5 reacted with O2 to yield 3 in good yield; thus a binuclear pathway must be considered viable.

Figure 3.

Figure 3

Open in a new tab

The molecular structure of binuclear 5 (30% probability level). Selected interatomic distances (Å) and angle (o): Cr1–O1, 1.9203(15); Cr1–O2, 1.7871(15); Cr2–O1, 1.7608(15); Cr2–O2, 1.8924(15); Cr1–NTp,avg, 2.158; Cr2–NTp,avg, 2.154; Cr2–H2B, 1.78(2); O1–Cr1–O2, 82.38(7); O1–Cr2–O2, 83.88(7); Cr1–O1–Cr2, 96.07(7); Cr1–O2–Cr2, 96.18(7).

Faced with a similar mechanistic question in the oxygenation of (i-Pr2Ph)2nacnacCr(η2-C2(SiMe3)2), we had turned to 16O/18O isotope labeling studies to provide strong evidence that the reaction proceeded via a binuclear intermediate.[5d] Unfortunately this proved impossible here, as a control experiment with an equimolar mixture of 3-16O2 and 3-18O2 resulted in rapid isotope scrambling, presumably via a dinuclear intermediate of the type TpCr(O)(μ-O)2(O)CrTp. Searching for other incisive mechanistic experiments, we reasoned that the binuclear pathway depends critically on the availability of substitutionally labile TptBu,MeCrI fragments to trap the initially formed A. Therefore, if a reaction could be devised to generate A in the absence of the former, a distinction might be made. To this end we have used the previously prepared chromium superoxo complex TptBu,MeCr(O2)Cl.[15] KC8 reduction of this complex in THF at −35°C effected a rapid color change from red to brown, and work-up of the solution yielded crystalline 3 in 77 % isolated yield. We propose that reduction of TptBu,MeCr(O2)Cl caused loss of chloride with concomitant formation of TptBu,MeCr(O2) (i.e., A, in which the erstwhile superoxo ligand has been reduced to a peroxo moiety), but in the absence of coordinatively unsaturated chromium and thus unable to form a binuclear compound. The ready production of 3 in high yield under these conditions then argues strongly that the unimolecular rearrangement of 4A to 23 is indeed facile.

In summary, we have investigated – experimentally and computationally – the 4-electron oxidative addition of O2 to a labile TpCr(I) precursor. The preponderance of the evidence suggests that in this system the unimolecular transformation of a Cr(III) peroxo intermediate (A) to the isomeric Cr(V) dioxo product (3) – while spin forbidden – is nevertheless the prevailing mechanistic pathway. We conclude that the effects of spin conservation on the mechanism of inorganic reactions are variable and require case-by-case evaluation. In other words, ‘spin forbidden’ reaction may indeed face particular constraints, but they cannot be ruled out categorically.

Supplementary Material

Supporting Information

Acknowledgments

This research was supported by DOE (DE-FG02-92ER14273). Shared instrumentation for NMR, LIFDI-MS, and X-ray diffraction was supported by grants from NIGMS (1 P30 GM110758-01), NSF (CHE-1229234), and NSF (CRIF 1048367), respectively.

References

  • 1.Ingraham LL, Meyer DL. Biochemistry of Dioxygen. Plenum Press; New York: 1985. [Google Scholar]
  • 2.Sheldon RA, Kochi JK. Metal-Catalyzed Oxidations of Organic Compounds. Academic Press; New York: 1981. [Google Scholar]
  • 3.a) Gubelmann MH, Williams AF. Struct Bonding (Berlin) 1983;55:1–65. [Google Scholar]; b) Martell AE, Sawyer DT. Plenum Press, New York. 1988. p. 341. [Google Scholar]; c) Klotz IM, Kurtz DM. Chem Rev. 1994;94:567–568. [Google Scholar]; d) Dickmann MH, Pope MT. Chem Rev. 1994;94:569–584. [Google Scholar]; e) Mirica LM, TDP Stack OX. Chem Rev. 2004;104:1013–1045. doi: 10.1021/cr020632z. [DOI] [PubMed] [Google Scholar]; f) Costas M, Mehn MP, Jensen MP, Que L. Chem Rev. 2004;104:939–986. doi: 10.1021/cr020628n. [DOI] [PubMed] [Google Scholar]; g) Lewis EA, Tolman WB. Chem Rev. 2004;104:1047–1076. doi: 10.1021/cr020633r. [DOI] [PubMed] [Google Scholar]; h) Kieber-Emmons MT, Riordan CG. Acc Chem Res. 2007;40:618–625. doi: 10.1021/ar700043n. [DOI] [PubMed] [Google Scholar]; i) Gong Y, Zhou MF, Andrews L. Chem Rev. 2009;109:6765–6808. doi: 10.1021/cr900185x. [DOI] [PubMed] [Google Scholar]; j) Cramer C, Tolman W, Theopold K, Rheingold A. Proc Natl Acad Sci U S A. 2003;100:3635–3640. doi: 10.1073/pnas.0535926100. [DOI] [PMC free article] [PubMed] [Google Scholar]; k) Gunay A, Theopold KH. Chem Rev. 2010;110:1060–1081. doi: 10.1021/cr900269x. [DOI] [PubMed] [Google Scholar]
  • 4.a) Schröder D, Shaik S, Schwarz H. Acc Chem Res. 2000;33:139–145. doi: 10.1021/ar990028j. [DOI] [PubMed] [Google Scholar]; b) Poli R, Harvey JN. Chem Soc Rev. 2003;32:1–8. doi: 10.1039/b200675h. [DOI] [PubMed] [Google Scholar]; c) Harvey JN, Poli R, Smith KM. Coord Chem Rev. 2003;238:347–361. [Google Scholar]; d) Harvey JN. Phys Chem Chem Phys. 2007;9:331–343. doi: 10.1039/b614390c. [DOI] [PubMed] [Google Scholar]; e) Besora M, Carreon-Macedo JL, Cimas A, Harvey JN. Adv Inorg Chem. 2009;61:573–623. [Google Scholar]; f) Harvey JN. Wiley Interdisciplinary Reviews-Computational Molecular Science. 2014;4:1–14. [Google Scholar]
  • 5.a) Detrich JL, Reinaud OM, Rheingold AL, Theopold KH. J Am Chem Soc. 1995;117:11745–11748. [Google Scholar]; b) Hess J, Leelasubcharoen S, Rheingold A, Doren D, Theopold K. J Am Chem Soc. 2002;124:2454–2455. doi: 10.1021/ja0167553. [DOI] [PubMed] [Google Scholar]; c) Monillas WH, Yap GPA, Theopold KH. Angew Chem, Int Ed. 2007;46:6692–6694. doi: 10.1002/anie.200701933. [DOI] [PubMed] [Google Scholar]; d) Dai F, Yap GPA, Theopold KH. J Am Chem Soc. 2013;135:16774–16776. doi: 10.1021/ja408357x. [DOI] [PubMed] [Google Scholar]
  • 6.Fiedler A, Kretzschmar I, Schröder D, Schwarz H. J Am Chem Soc. 1996;118:9941–9952. [Google Scholar]
  • 7.Akturk ES, Yap GP, Theopold KH. Chem Commun (Camb) 2015 doi: 10.1039/c5cc05993c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.C48H80B2Cr2N12O2, Mr = 982.86, crystal dimensions = 0.339 × 0.241 × 0.237 mm, monoclinic P21/c, a = 20.7125(7), b = 14.7706(5), c = 22.5952(8) Å, β = 103.7730(10) °, V = 6713.9(4) Å3, Z = 4, ρcalcd = 0.972 g cm−3, μ = 0.362 mm-1, Mo-kα λ = 0.71073 Å, T = 200(2) K, 2θmax = 55.0, 86171 reflections measured, 15408 independent reflections, Rint = 0.0536, R1 = 0.0558, wR2 = 0.1435, residual electron density = 0.802 e Å−3. The t-butyl group on the noncoordinated pyrazole was located disordered in two positions with a refined site occupancy of 58/42. Two disordered pentane molecules of solvation per compound molecule were treated as diffused contributions. Borohydride H-atoms located from difference map. CCDC number: 1414458.
  • 9.A search of CSD revealed 288 complexes with terminal oxo ligands coordinated to Cr. 4 of these were eliminated as outliers, leaving 781 Cr=O bonds Their mean Cr-O distance was 1.604(31) Å.
  • 10.C32H56BCrN6O2S2, Mr = 683.75, crystal dimensions = 0.438 × 0.388 × 0.364 mm, orthorhombic Pnma, a = 18.487(14), b = 21.314(16), c = 9.395(7) Å, V = 3702(5) Å3, Z = 4, ρcalcd = 1.227 g cm−3, μ = 0.458 mm−1, Mo-kα λ = 0.71073 Å, T = 200(2) K, 2θmax = 56.8, 48217 reflections measured, 4723 independent reflections, Rint = 0.0418, R1 = 0.0405, wR2 = 0.1201, residual electron density = 0.640 e Å−3. Compound molecule resides on a mirror plane. THF solvent molecule treated with rigid bond restraints. CCDC number: 1414459.
  • 11.a) Brunner H, Wachter J, Guggolz E, Ziegler ML. J Am Chem Soc. 1982;104:1765–1766. [Google Scholar]; b) Goh LY, Hambley TW, Robertson GB. J Chem Soc, Chem Commun. 1983:1458–1460. [Google Scholar]; c) Herrmann WA, Rohrmann J, Noth H, Nanila CK, Bernal I, Draux M. J Organomet Chem. 1985;284:189–211. [Google Scholar]; d) Goh LY, Hambley TW, Robertson GB. Organometallics. 1987;6:1051–1057. [Google Scholar]; e) Brunner H, Pfauntsch J, Wachter J, Nuber B, Ziegler ML. J Organomet Chem. 1989;359:179–188. [Google Scholar]; f) Toupadakis A, Kubas GJ, Burns CJ. Inorg Chem. 1992;31:3810–3817. [Google Scholar]; g) Herberhold M, Jin GX, Rheingold AL. Zeitschrift für Naturforschung Section B - A Journal of Chemical Sciences. 1993;48:1488–1498. [Google Scholar]; h) Tamne ES, Noor A, Qayyum S, Bauer T, Kempe R. Inorg Chem. 2013;52:329–336. doi: 10.1021/ic3020805. [DOI] [PubMed] [Google Scholar]
  • 12.DFT calculations using the B3LYP functional and TZVP basis set were carried out using the program package Gaussian 09. Details are in the ESI. For computational economy the alkyl substituents of the Tp ligands were replaced with hydrogen atoms; the resulting compounds are labeled with a prime (ʹ).
  • 13.Harvey JN, Aschi M, Schwarz H, Koch W. Theor Chem Acc. 1998;99:95–99. [Google Scholar]
  • 14.C24H40BCrN6O2, Mr = 507.43, crystal dimensions = 0.274 × 0.271 × 0.150 mm, monoclinic P21/n, a = 9.309(3), b = 30.574(10), c = 9.374(3) Å, β = 101.167(6) °, V = 2617.4(14) Å3, Z = 4, ρcalcd = 1.288 g cm−3, μ = 0.467 mm−1, Mo-kα λ = 0.71073 Å, T = 100(2) K, 2θmax = 55.0, 24396 reflections measured, 6001 independent reflections, Rint = 0.0773, R1 = 0.0514, wR2 = 0.1296, residual electron density = 0.333 e Å-3. CCDC number: 1414460.
  • 15.Qin K, Incarvito C, Rheingold A, Theopold K. Angew Chem, Int Ed. 2002;41:2333–2335. doi: 10.1002/1521-3773(20020703)41:13<2333::AID-ANIE2333>3.0.CO;2-F. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
NIHMS753168-supplement-Supporting_Information.pdf (230.5KB, pdf)

RESOURCES