High-Valent Pyrazolate-Bridged Platinum Complexes: A Joint Experimental and Theoretical Study.

Complexes [{Pt(C^C*)(μ-pz)}2] (HC^C*A = 1-(4-(ethoxycarbonyl)phenyl)-3-methyl-1H-imidazol-2-ylidene 1a, HC^C*B = 1-phenyl-3-methyl-1H-imidazol-2-ylidene 1b) react with methyl iodide (MeI) at room temperature in the dark to give compounds [{PtIV(C^C*)Me(μ-pz)}2(μ-I)]I (C^C*A 2a, C^C*B 2b). The reaction of 1a with benzyl bromide (BnBr) in the same conditions afforded [Br(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Bn] (5a), which by heating in BnBr(l) became [{PtIV(C^C*A)Bn(μ-pz)}2(μ-Br)]Br (6a). Experimental investigations and density functional theory (DFT) calculations on the mechanisms of these reactions from 1a revealed that they follow a SN2 pathway in the two steps of the double oxidative addition (OA). Based on the DFT investigations, species such as [(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)R]X (RX = MeI Int-Me, BnBr Int-Bn) and [(C^C*A)PtII(μ-pz)2PtIV(C^C*A)(R)X] (RX = MeI Int'-Me, BnBr Int'-Bn) were proposed as intermediates for the first and the second OA reactions, respectively. In order to put the mechanisms on firmer grounds, Int-Me was prepared as [(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Me]BF4 (3a') and used to get [I(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Me](4a), [(C^C*A)PtII(μ-pz)2PtIV(C^C*A)(Me)I](Int'-Me), and [{PtIV(C^C*)Me(μ-pz)}2(μ-I)]BF4 (2a'). The single-crystal X-ray structures of 2a, 2b, 3a', and 5a along with the mono- and bi-dimensional 1H and 195Pt{1H} NMR spectra of all the named species allowed us to compare structural and spectroscopic data for high-valent complexes with the same core [{Pt(C^C*)(μ-pz)}2] but different oxidation states.

Introduction

Oxidative addition (OA)–reductive elimination processes on d8 transition metal complexes account for many organic transformations.[1−5] High-valent metal–metal-bonded binuclear species play very often a key role as intermediates.[6−9] Compared to those of Rh2(I,I) and Ir2(I,I), the mechanisms of OA reactions of haloalkanes (RX) to Pt2(II,II) complexes have been scarcely studied. In the case of Pt2(II,II) complexes with the metal atoms far away from each other, the OA reactions follow monometallic pathways.[10−12] However, if the metal centers are held in proximity by bridging ligands, different kinds of mono- or bimetallic mechanisms can operate. In lantern- or half-lantern-shaped complexes, exhibiting short intermetallic separation (<3.0 Å), it is well known that the [Pt2(POP)4]4– (POP = pyrophosphite, dPt–Pt = 2.925 Å) complex undergoes thermal two-electron two-center [2e, 2c] OA of RI (R = Me, Et, Pr, Pr, n-pentyl) following a radical mechanism, although contribution of a SN2-type one cannot be excluded for MeI.[13] Furthermore, the half-lantern compound [{Pt(bzq)(μ-N^S)}2] [bzq = benzo[h]quinoline, HN^S = 2-mercaptopyrimidine] also undergoes [2e, 2c] thermal OA of CH3I and CHX3 (X = Br, I) following a bimetallic SN2 or radical mechanism.[14] On the other hand, complexes [Pt2Me2(C^N)2(μ-P^P)] [C^N = 2-phenylpyridyl-H, benzo[h]quinoline; P^P = dppf (1,1′-bis(diphenylphosphino)ferrocene), dppa (1,1′-bis(diphenylphosphino)acetylene)],[11,12,15] and cis,cis-[Me2Pt(μ-NN)(μ-dppm)PtMe2] (NN = phthalazine, dppm = bis(diphenylphosphino)methane), with flexible bridging ligands, reacted with MeI in two steps, via a monometallic SN2 mechanism. As a result, the diplatinum(IV) derivatives [Pt2Me4I2(C^N)2(μ-P^P)] and [Me3Pt(μ-I)2(μ-dppm)PtMe3] were obtained.[16]

Pyrazolates are a kind of adaptative bridging ligands. Because they have a proven ability to hold two metal atoms in close proximity while enabling a wide range of intermetallic separations; they allow for different kinds of OA mechanisms. For instance, haloalkanes such as CH3I or CH2I2 add to [{IrI(μ-pz)}2] complexes following mostly a bimetallic SN2[17−19] or radical[20,21] mechanisms, leading to metal–metal-bonded Ir2(II,II) complexes. Monometallic SN2[22,23] pathways resulting in mixed valence Ir(I)–Ir(III) compounds have sometimes been proven. The Ir2(II,II) species, once rarely formed, undergo further OA; when this happens, no metal–metal-bonded Ir(III)–Ir(III) compounds were obtained.[24,25] In the field of pyrazolate-bridged Pt2(II,II) complexes, we observed that [{PtII(C^C*A)(μ-pz)}2] (HC^C*A = 1-(4-(ethoxycarbonyl)phenyl)-3-methyl-1H-imidazol-2-ylidene 1a) reacted with haloforms, CHX3 (X = Cl, Br, I), following a radical mechanism. Complexes [{Pt(C^C*A)(μ-pz)X}2], [XPt(C^C*A)(μ-pz)2Pt(C^C*A)CHX2] or mixtures of both were obtained, depending mostly on the environmental conditions (argon atmosphere, oxygen or light).[26,27] In depth mechanistic investigations evidenced that complex 1a exists in solution in two forms: the butterfly-wing-spreading form 1a–s characterized by long intermetallic distances and the wing-folding one 1a–f, with short ones. These two conformers interconvert one into the other, resembling a butterfly flapping process. Species 1a–f are those which trigger the reaction with haloforms in the ground state (S0) with CHBr3 and CHI3 or in the excited state S1 with CHCl3.[27] These results highlighted the relevance of metal–metal cooperativity to enable the oxidation of 1a. These findings, along with the importance of high-valent organometallic complexes in many organic synthesis, encouraged us to widen the scope of our earlier investigations, exploring the reactivity of 1a and the analogous complex [{PtII(C^C*B)(μ-pz)}2] (HC^C*B = 1-phenyl-3-methyl-1H-imidazol-2-ylidene (1b) toward other halogenated species, such as methyl iodide (MeI) and benzyl bromide (BnBr). Supported by density functional theory (DFT) studies on the OA mechanisms, we were able to prepare dinuclear complexes with different oxidation states: Pt2(III,III), Pt2(III,III) ↔ Pt2(II,IV) and Pt2(IV,IV). They allowed us to substantiate the modeled mechanisms and to compare their structural and spectroscopic data.

Experimental Section

General information about instrumentation, X-ray structure determinations (CCDC 2160386–2160389), DFT computational details with Figure S1, and NMR spectra for characterization are available in the Supporting Information.

Compounds [{Pt(C^C*A)(μ-pz)}2] (1a),[26] and [{Pt(C^C*B)(μ-Cl)}2] (B)[28] were prepared as described elsewhere. MeI, BnBr, Hpz, and AgClO4 were used as purchased from Acros Organics, Fluka, Merck, and Aldrich, respectively. NMR spectra were recorded at r.t., except if a different value is indicated. Data are given according to Figure .

Figure 1

Numerical scheme for NMR purposes.

Synthesis of [{Pt(C^C*B)(μ-pz)}2] (1b)

AgClO4 (55.1 mg, 0.263 mmol) was added to a stirred suspension of B (102.1 mg, 0.132 mmol) in acetone (40 mL) in the dark at room temperature. After 2.5 h, pzH (35.8 mg, 0.527 mmol) was added to the mixture and allowed to react for 18.5 h in the darkness. Then, the resulting suspension was filtered through Celite and the solution was concentrated to 50 mL. Afterward, NEt3 (0.5 mL, 3.62 mmol) was added to the solution at r.t. and allowed to react for 2 h. The suspension was concentrated to 15 mL, and the solid was filtered and washed with 2 mL of acetone to give 1b as a white solid. Yield: 74.0 mg, 0.088 mmol, 67%. Anal. calcd for C26H24N8Pt2: C, 37.23; H, 2.88; N, 13.36. Found: C, 37.22; H, 2.98; N, 12.97. 1H NMR (400 MHz, DMSO-d6): δ = 7.43 (d, 3J2,3 = 1.6, 2H, H2), 7.25 (s, 2H, Hpz), 7.16 (s, 2H, Hpz), 6.82 (d, JH,H = 7.4, 2H, HAr), 6.76 (d, 2H, H3), 6.63 (d, JH,H = 7.3, 2H, HAr), 6.50 (t, JH,H = 7.4, 2H, HAr), 6.35 (t, JH,H = 7.0, 2H, HAr), 5.88 (s br, 2H, Hpz), 2.72 (s, 6H, H4). 1H–195Pt HMQC NMR (85.6 MHz, DMSO-d6): δ = −3767.4 (s).

Synthesis of [{Pt(C^C*A)Me(μ-pz)}2(μ-I)]I (2a)

CH3I (26 μL, 0.403 mmol) was added to a solution of 1a (98.9 mg, 0.101 mmol) in anhydrous CH2Cl2 (5 mL) under argon atmosphere in the dark. After 14 h of reaction, the precipitate was filtered, washed with Et2O (4 × 10 mL), and dried to give 2a as a white solid. Yield: 117.8 mg; 0.093 mmol; 92%. Anal. calcd for C34H38I2N8O4Pt2: C, 32.24; H, 3.02; N, 8.85. Found: C, 31.84; H, 2.82; N, 8.45. 1H NMR (400 MHz, CD2Cl2, 248 K): δ = 7.97 (s, 3JH,Pt = 37.7, 2H, H7), 7.94 (d, 3J9,10 = 8.2, 2H, H9), 7.87–7.73 (m, 6H, Hpz and H2), 7.47 (d, 3J10, 9 = 8.2, 2H, H10), 7.34 (d, 3J3,2 = 1.8, 2H, H3), 6.50 (s br, 2H, Hpz), 4.27 (q, 3JH,H = 6.9, 4H, OCHCH3), 3.52 (s, 6H, H4), 1.78 (s, 2JH,Pt = 65.5, 6H, Pt-CH3), 1.31 (t, 3JH,H = 6.9, 6H, OCH2CH). 13C{1H} NMR plus HMBC and HSQC (101 MHz, CD2Cl2, 248 K): δ = 165.7 (s, 2C, C=O), 146.3 (s, 2C, C5), 142.1 (s, 2C, 1JC,Pt = 1126.2, C1), 140.1 and 139.2 (4C, Cpz), 133.3 (s, 2C, C7), 128.4 (s, 2C, C9), 125.5 (s, 1C, C3), 117.5 (s, 1C, C2), 113.6 (s, 2C, C10), 107.6 (s, 2C, Cpz), 61.5 (s, 2C, OCH2CH3), 38.1 (s, 2C, C4), 14.3 (s, 2C, OCH2CH3), 11.6 (s, 1JC,Pt = 502.2, 2C, Pt-CH3). 195Pt{1H} NMR (85.6 MHz, CD2Cl2, 248 K): δ = −2688.0 (s). MS (MALDI+): m/z = 1138.88 [{Pt(C^C*A)(CH3)(μ-pz)}2(μ-I)]+. IR (ATR, cm–1) υ = 1698 (m, C=O).

Synthesis of [{Pt(C^C*B)Me(μ-pz)}2(μ-I)]I (2b)

CH3I (18 μL, 0.2862 mmol) was added to a suspension of 1b (60 mg, 0.072 mmol) in anhydrous DMF (5 mL) under argon atmosphere in the dark. After 8 h of reaction, 100 mL of Et2O was added and the precipitate was filtered, washed with Et2O (5 × 10 mL), and dried to give 2b as a white solid. Yield: 70.7 mg; 0.063 mmol; 88%. Anal. calcd for C28H30I2N8Pt2: C, 29.96; H, 2.69; N, 9.98. Found: C, 29.76; H, 2.62; N, 9.65. 1H NMR (400 MHz, CD2Cl2): δ = 7.78 (m, 4H, Hpz), 7.70 (d, 3J2,3 = 2.0, 2H, H2), 7.41–7.24 (m, 8H, H3 and HAr), 7.13 (m, 2H, HAr), 6.47 (m, 2H, Hpz), 3.52 (s, 6H, H4), 1.80 (s, 2JH,Pt = 66.1, 6H, Pt-CH3). 1H–195Pt HMQC NMR (85.6 MHz, CD2Cl2): δ = −2664.4 (s). MS (MALDI+): m/z = 994.07 [{Pt(C^C*B)(CH3) (μ-pz)}2(μ-I)]+.

Synthesis of [(C^C*A)Pt(μ-pz)2Pt(C^C*A)Me]BF4 (3a′)

Me3OBF4 (49.3 mg, 0.320 mmol) was added to a solution of 1a (262.0 mg, 0.267 mmol) in anhydrous CH2Cl2 (15 mL) under argon atmosphere in the dark at −25 °C. After 2 h of reaction, the solution was dried in vacuo. The residue was treated with 20 mL of dried Et2O, and the resulting solid was filtered, washed with Et2O (2 × 20 mL), and dried to give 3a′ as a brown solid. Yield: 244.8 mg; 0.226 mmol; 85%. Anal. calcd for C33H35BF4N8O4Pt2·2CH2Cl2: C, 33.51; H, 3.13; N, 8.93. Found: C, 33.27; H, 3.10; N, 9.33. 1H NMR (400 MHz, CD2Cl2): δ = 7.92 (d, 3JH,H = 2.1, 1H, Hpz), 7.85–7.77 (m, 3H, 2Hpz, and H9 [Pt–Me]), 7.74 (d, 3JH,H = 2.1, 1H, Hpz), 7.66 (dd, 3J9,10 = 8.2, 4J9,7 = 1.6, 1H, H9 [Pt]), 7.52 (d, 4J7,9 = 1.6, 3JH,Pt = 41.2, 1H, H7[Pt]), 7.50 (d, 4J7,9 = 1.6, 3JH,Pt = 51.7, 1H, H7 [Pt–Me]), 7.43 (d, 3J2,3 = 2.0, 1H, H2 [Pt–Me]), 7.24–7.16 (m, 2H, H10 [Pt–Me] and H2 [Pt]), 7.03 (d, 3J10,9 = 8.1, 1H, H10 [Pt]), 6.60–6.52 (m, 3H, 2Hpz and H3), 6.37 (d, 3J3,2 = 2.1, 1H, H3 [Pt]), 4.38–4.23 (m, 4H, OCHCH3), 3.17 (s, 3H, H4 [Pt–Me]), 3.05 (s, 3H, H4 [Pt]), 2.42 (s, 2JH,Pt = 70.7, 3JH,Pt = 14.7, 3H, [Pt-CH3]), 1.40–1.30 (m, 6H, OCH2CH). 13C{1H} NMR plus HMBC and HSQC (101 MHz, CD2Cl2): δ = 166.5 (s, 1C, C=O), 166.0 (s, 1C, C=O), 153.8 (s, 1C, 1JC,Pt = 1391.4, C1 [Pt]), 150.4 (s, 1C, C5 [Pt]), 147.2 (s, 1C, C5 [Pt–Me]), 144.5 (s, 1C, 1JC,Pt = 1179.2, C1 [Pt–Me]), 140.2 (s, 1C, Cpz), 137.9 (s, 1C, Cpz), 135.9 and 135.5 (s, 4C, Cpz), 134.7 (s, 1C, C7 [Pt]), 133.2 (s, 1C, C7 [Pt–Me]), 129.8 (s, 1C, C9 [Pt–Me]), 129.6 (s, 1C, C9 [Pt]), 124.8 (s, 1C, C3 [Pt–Me]), 123.6 (s, 1C, C3 [Pt]), 117.4 (s, 1C, C2 [Pt–Me]), 116.8 (s, 1C, C2 [Pt]), 113.7 (s, 1C, C10 [Pt–Me]), 112.1 (s, 1C, C10 [Pt]), 107.8 and 107.7 (s, 2C, Cpz), 61.9 and 61.8 (s, 2C, OCH2CH3), 37.2 (s, 1C, C4 [Pt–Me]), 37.0 (s, 1C, C4 [Pt]), 14.6 and 14.5 (s, 2C, OCH2CH3), −1.9 (s, 1JC,Pt = 411.2, 1C, Pt-CH3).19F NMR (376 MHz, CD2Cl2): δ = −151.4 (m, 4F, BF4). 195Pt{1H} NMR (85.6 MHz, CD2Cl2): δ = −2589.2 (s, 1JPt,Pt = 1023.4, Pt–Me), −3064.2 (s, Pt). MS (MALDI+): m/z = 997.2 [(C^C*A)Pt(μ-pz)2Pt(C^C*A)CH3]+. IR (ATR, cm–1) υ = 1702 (m, C=O), 1043, 1012 and 519 (s, BF4).

Synthesis of [I(C^C*A)Pt(μ-pz)2Pt(C^C*A)Me] (4a)

KI (33.6 mg, 0.202 mmol) was added to a solution of 3a′ (109.8 mg, 0.101 mmol) in MeCN (3 mL) in the dark at −25 °C. After 3 h of reaction, the suspension was filtered and the solid was washed with water (7 × 10 mL) and dried to give 4a as a yellow solid. Yield: 44.0 mg; 0.039 mmol; 39%. Anal. calcd for C33H35IN8O4Pt2: C, 35.54; H, 3.14; N, 9.96. Found: C, 35.88; H, 2.88; N, 9.65. 1H NMR (400 MHz, CD2Cl2, 223 K): δ = 8.01 (d, 3JH,H = 2.0, 1H, Hpz), 7.89 (d, 3JH,H = 2.0, 1H, Hpz), 7.72 (d, 3JH,H = 2.0, 1H, Hpz), 7.64 (d, 3JH,H = 2.0, 1H, Hpz), 7.61 (dd, 3J9,10 = 8.2, 4J9,7 = 1.6, 1H, H9 [Pt–Me]), 7.53 (d, 4J7,9 = 1.6, 3JH,Pt = 58.7, 1H, H7 [Pt–I]), 7.44–7.33 (m, 2H, H9 [Pt–I] and H7[Pt–Me]), 7.04 (d, 3J2,3 = 2.1, 1H, H2 [Pt–Me]), 6.89 (d, 3J10,9 = 8.2, 1H, H10 [Pt–Me]), 6.84 (d, 3J2,3 = 2.1, 1H, H2 [Pt–I]), 6.93 (d, 3J10,9 = 8.2, 1H, H10 [Pt–I]), 6.38–6.30 (m, 3H, Hpz and H3 [Pt–Me]), 6.15 (d, 3J3,2 = 2.0, 1H, H3 [Pt–I]), 4.35–4.10 (m, 4H, OCHCH3), 3.04 (s, 3H, H4 [Pt–I]), 3.01 (s, 3H, H4 [Pt–Me]), 1.48 (s, 2JH,Pt = 61.0, 3JH,Pt = 14.5, 3H, Pt-CH3), 1.32–1.24 (m, 6H, OCH2CH). 13C{1H} NMR plus HMBC and HSQC (101 MHz, CD2Cl2, 223 K): δ = 166.3 (s, 1C, C=O), 165.8 (s, 1C, C=O), 153.0 (s, 1JC,Pt = 1358.7, 1C, C1 [Pt–I]), 147.8 (s, 1C, C5), 147.5 (s, 1JC,Pt = 1153.6, 1C, C1 [Pt–Me]), 145.7 (s, 1C, C5), 140.5, 139.1 and 134.3 (s, 3C, Cpz), 132.9 (s, 1C, C7 [Pt–I]), 132.0 (s, 1C, C7 [Pt–Me]), 131.7 (s, 1C, Cpz), 128.1, 127.1, 126.2 and 125.2 (C6, C6, C8 and C8), 126.4 (s, 1C, C9 [Pt–Me]), 125.3(s, 1C, C9 [Pt–I]), 123.0 (s, 1C, C3 [Pt–Me]), 121.7 (s, 1C, C3 [Pt–I]), 114.1 (s, 1C, C2 [Pt–Me]), 113.7 (s, 1C, C2 [Pt–I]), 111.1 (s, 1C, C10 [Pt–Me]), 110.0 (s, 1C, C10 [Pt–I]), 105.6 (m, 2C, Cpz), 61.1 and 60.9 (s, 2C, OCH2CH3), 36.6 and 36.5 (s, 2C, C4, [Pt–Me] and [Pt–I]), 14.1 (s br, 2C, OCH2CH3), −16.0 (s, 1JC,Pt = 467.1, 1C, Pt-CH3). 195Pt{1H} NMR (85.6 MHz, CD2Cl2, 223 K): δ = −2848.2 (s, 1JPt,Pt = 1239.8, Pt–Me), −3018.8 (s, Pt–I). MS (MALDI+): m/z = 997.2 [(C^C*A)Pt(μ-pz)2Pt(C^C*A)(CH3)] +. IR (ATR, cm–1) υ = 1699 (m, C=O).

Synthesis of Int′-Me

PPh4I (83.4 mg, 0.179 mmol) was added to a solution of 3a′ (97.0 mg, 0.089 mmol) in anisole (5 mL) in the dark at 30 °C, and the mixture was allowed to react for 2 h. Then, the solvent was removed under vacuum and the residue was treated with H2O. The resulting yellow solid was identified as [(C^C*A)Pt(μ-pz)2(C^C*A)PtI(CH3)] (Int′-Me) by 1H and 195Pt NMR, although anisole and PPh4BF4 were detected as impurities. 1H NMR (400 MHz, 223 K, CD2Cl2): δ = 8.10–6.00 (16 HAr of Int′-Me and HAr of PPh4), 4.33–4.07 (m, 4H, OCHCH3), 3.75 (s, OCH3, anisole), 3.49 and 3.19 (s, 6H, H4), 1.75 (s, 2JH,Pt = 65.9, 3H, [Pt-CH3]). 195Pt{1H} NMR (85.6 MHz, 223 K, CD2Cl2): δ = −2697.0 (s, PtIV), −3776.0 (s, PtII).

Synthesis of [Br(C^C*A)Pt(μ-pz)2Pt(C^C*A)Bn] (5a)

BnBr (33 μL, 0.277 mmol) was added to a solution of 1a (68.0 mg, 0.069 mmol) in MeCN (20 mL) in the dark. After 3.5 h of reaction, the solvent was removed under vacuum. The residue was treated with a mixture of Et2O and n-hexane (1:20) to give 5a as an orange solid. Yield: 74.0 mg; 0.064 mmol; 93%. Anal. calcd for C39H39BrN8O4Pt2: C, 40.60; H, 3.41; N, 9.71. Found: C, 40.32; H, 3.36; N, 9.68. 1H NMR (400 MHz, CD2Cl2, 248 K): δ = 7.97 (d, 3JH,H = 1.8, 1H, Hpz), 7.91 (d, 3JH,H = 1.8, 1H, Hpz), 7.78 (d, 3JH,H = 2.1, 1H, Hpz), 7.65 (dd, 3J9,10 = 8.2, 4J9,7 = 1.7, 1H, H9 [Pt–Bn]), 7.50 (d, 4J7,9 = 1.7, 3JH,Pt = 43.3, 1H, H7 [Pt–Bn]), 7.44 (d, 4J7,9 = 1.7, 3JH,Pt = 45.2, 1H, H7 [Pt–Br]), 7.39 (dd, 3J9,10 = 8.2, 4J9,7 = 1.7, 1H, H9 [Pt–Br]), 7.16 (d, 3JH,H = 2.1, 1H, Hpz), 7.06–6.98 (m, 1H, H), 6.89–6.80 (m, 5H, H2, H2, H10 [Pt–Bn] and H), 6.78–6.68 (m, 3H, H10 [Pt–Br] and H), 6.40 (pt, 3JH,H = 2.0, 1H, Hpz), 6.30 (pt, 3JH,H = 2.1, 1H, Hpz), 6.11 (d, 3J3,2 = 2.1, 1H, H3 [Pt–Bn]), 6.07 (d, 3J3,2 = 2.1, 1H, H3 [Pt–Br]), 4.36–4.10 (m, 5H, CH2 (Bn) and OCHCH3), 3.73 (d, 2JH,H = 7.9, 2JH,Pt = 69.0, 3JH,Pt = 28.9, 1H CH2 (Bn)), 3.07 (s, 3H, H4 [Pt–Br]), 2.71 (s, 3H, H4 [Pt–Bn]), 1.32 (t, 3JH,H = 7.3, 3H, OCH2CH), 1.25 (t, 3JH,H = 7.3, 3H, OCH2CH).13C{1H} NMR plus HMBC and HSQC (101 MHz, CD2Cl2, 248 K): δ = 166.5 (s, 1C, C=O), 166.0 (s, 1C, C=O), 153.2 (s, 1JC,Pt = 1323.4, 1C, C1 [Pt–Br]), 148.7 (s, 1JC,Pt = 1239.5, 1C, C1 [Pt–Bn]), 148.1 (s, 1C, C5 [Pt–Br]), 147.1 (s, 1C, C), 145.8 (s, 1C, C5 [Pt–Bn]), 138.9, 137.2 and 134.4 (s, 3C, Cpz), 133.3 (s, 1C, C7 [Pt–Br]), 132.1 (s, 1C, C7 [Pt–Bn]), 131.9 (s, 1C, Cpz), 128.9 (s, 2C, C), 127.0 and 126.9 (2C, C9 [Pt–Bn] and C), 125.7 (s, 1C, C9 [Pt–Br]), 124.5 (s, 1C, C), 122.7 (s, 1C, C3 [Pt–Bn]), 121.9 (s, 1C, C3 [Pt–Br]), 114.1 and 113.9 (2C, C2), 111.4 (s, 1C, C10 [Pt–Bn]), 110.1 (s, 1C, C10 [Pt–Br]), 105.8 and 105.7 (2C, Cpz), 61.3 and 61.0 ( 2C, OCH2CH3), 36.6 (s, 1C, C4 [Pt–Br]), 36.1 (s, 1C, C4 [Pt–Bn]), 15.2 (s, 1JC,Pt = 448.1, 2JC,Pt = 192.9, 1C, Pt-CH2Ph), 14.3 (s, 2C, OCH2CH3). 195Pt{1H} NMR (85.6 MHz, CD2Cl2, 248 K): δ = −2693.6 (s br, Pt–Bn), −2742.8 (s, 1JPt,Pt = 1028.9, Pt–Br,). MS (MALDI+): m/z = 1061.6 [Br(EtO2C-C^C*)Pt(μ-pz)2Pt(EtO2C-C^C*)]+, 1072.8 [(C^C*A)Pt(μ-pz)2Pt(C^C*A)Bn]+. IR (ATR, cm–1) υ = 1700 (m, C=O).

Synthesis of [{Pt(C^C*A)Bn(μ-pz)}2(μ-Br)]Br (6a)

A suspension of 5a (95 mg, 0.082 mmol) in BnBr (5 mL) was heated 70 °C for 5 h. Then, the suspension was cooled down to room temperature, and the resulting solid was filtered and dried to give 6a. Yield: 94.7 mg; 0.071 mmol; 87%. Anal. calcd for C46H46Br2N8O4Pt2: C, 41.70; H, 3.50; N, 8.46. Found: C, 41.38; H, 3.27; N, 8.44. 1H NMR (400 MHz, CD2Cl2, 248 K): δ = 8.46 (s br, 2H, Hpz), 8.35 (s br, 2H, Hpz), 8.10 (d, 4J7,9 = 1.1, 3JH,Pt = 37.5, 2H, H7), 7.89 (dd, 3J9,10 = 8.2, 4J9,7 = 1.1, 2H, H9), 7.18 (s, 2H, H2), 7.15–7.03 (m, 6H, H10, H and H3), 6.88 (pt, JH,H = 7.6, 4H, H), 6.79 (s br, 2H, Hpz), 6.64 (pd JH,H = 7.6, 4H, H), 4.45–4.20 (m, 6H, CH2 (Bn) and OCHCH3), 4.02 (d, 2JH,H = 8.6, 2JH,Pt = 90.6, 2H, CH2 (Bn)), 3.41 (s, 6H, H4), 1.35 (t, 3JH,H = 7.3, 6H, OCH2CH). 13C{1H} NMR plus HMBC and HSQC (101 MHz, CD2Cl2, 248 K): δ = 165.7 (s, 2C, C=O), 146.4 (s, 2C, C5), 142.6 (s, 2C, 1JC,Pt = 1202.2, C1), 141.1 (2C, Cpz), 141.0 (s, 2C, C), 138.4 (2C, Cpz), 132.2 (s, 2C, C7), 128.8 (s, 2C, C9), 128.7 (s, 2C, C), 128.3 (s, 4C, 3JC,Pt = 23.0, C), 127.1 (s, 2C, C), 124.5 (s, 2C, C3), 116.5 (s, 2C, C2), 113.6 (s, 4JC,Pt = 28.7, 2C, C10), 107.4 (s, 3JC,Pt = 18.0, 2C, C), 61.5 (s, 2C, OCH2CH3), 37.6 (s, 2C, C4), 32.7 (s, 2JC,Pt = 521.4, 2C, Pt-CH2Ph), 14.3 (s, 2C, OCH2CH3). 195Pt{1H} NMR (85.6 MHz, CD2Cl2, 248 K): δ = −2357.0 (s). MS (MALDI+): m/z = 1245.69 [{Pt(C^C*A)Bn(μ-pz)}2(μ-Br)]+. IR (ATR, cm–1) υ = 1715 (m, C=O).

Results and Discussion

Reactivity of [{PtII(C^C*)(μ-pz)}2](C^C*A1a, C^C*B1b) with MeI and BnBr: Experimental and Computational Investigations for the Mechanistic Studies

The reaction of [{PtII(C^C*A)(μ-pz)}2] (1a) with MeI in MeCN in the dark afforded the Pt2(IV,IV) compound [{PtIV(C^C*A)Me(μ-pz)}2(μ-I)]I (2a) as result of a double OA of MeI, regardless the reactant molar ratio (Scheme , path a).

Scheme 1

Reaction Pathway for OA Reactions of MeI to 1a and 1b

The use of other solvents such as acetone or dichloromethane does not change the nature of the final compound. Compound [I(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Me] (4a), resulting from the bimetallic OA of one MeI molecule was just detected by 1H NMR. To evaluate if the CO2Et fragment plays a role in the redox behavior of complex 1a, we prepared the new complex 1b (Experimental Section in the Supporting Information and Figure S2). Then, it was reacted with MeI in DMF due to its low solubility in other organic solvents, giving rise to the Pt2(IV,IV) complex [{PtIV(C^C*B)Me(μ-pz)}2(μ-I)]I (2b). This result indicates that the CO2Et fragment does not affect the reactivity of these Pt2(II,II) complexes toward MeI; however, it increases their solubility, allowing for a better study of it. Complexes 2a and 2b were isolated as white solids in very good yields (92%, 2a; 88%, 2b) and fully characterized (Figures S3 and S4). Just two complexes with the same bridging system have been reported to date, PPN[{PtIVMe3(μ-pz)}2(μ-I)][29] and (PPh4)[{PtIVMe2Br(μ-pz)}2(μ-Br)],[30] both of them prepared by assembly of mononuclear PtIV fragments. Therefore, 2a and 2b are the first Pt2(IV,IV) derivatives obtained by OA of MeI to {PtII(μ-pz)}2 fragments.

Keeping in mind that compound 1a is oxidized by CHX3 (X = Br, I) in the dark through a radical mechanism,[26,27] we checked this possibility for MeI. The reaction of 1a with MeI in MeCN-d3 in the dark was performed with and without galvinoxyl (Gal·) as radical (R·) trap, and they were followed by 1H NMR for 1 h. It resulted to be almost unaffected by the presence of Gal·(see Figure S5), which led us to dismiss a radical mechanism and to consider a SN2 one for the first and the second OA of MeI to 1a. For an in depth knowledge of these reaction mechanisms, we carried out a DFT study (see Computational Details in Supporting Information). The free energy profiles in MeCN have been represented in Scheme , the reference energy value being 0.0 kcal/mol for one of the Pt2(II,II) reactant, 1a.

Scheme 2

Computed (PCM(MeCN)-M06/6-31G(d), MWB60(Pt), and MWB46(I)) Free Energy Profile (ΔG, kcal/mol) for the Thermal Conversion of 1a into 4a (or 5a) and Int′-R (Step i) and Int′-R into 2a (or 6a) (Step ii) Following SN2 Mechanisms

In the modeled mechanism, the first OA is a SN2 reaction, Nu + MeI → NuMe+ + I–, with the dinuclear compound 1a acting as nucleophile (Nu) to give a cationic [Pt(II)–Pt(IV) −Me]+ intermediate Int-Me. The reaction would proceed through a transition state TS1, with one imaginary frequency (436i cm–1), which shows a hypervalent C atom with two long Pt···C and C···I distances. The energy barrier (EaTS1 = 15.02 kcal/mol) is low enough to allow the reaction to proceed at room temperature in the dark. Once the intermediate Int-Me was formed, the migration of the halide to the Pt(II) center will afford the I–Pt(III) −Pt(III) −Me derivative (4a) while, if the halide bonds to the Pt(IV) center, Pt(II,IV) species (Int′-Me) will be generated.

The small barrier (2.92 kcal/mol) for the conversion of Int-Me into Int′-Me through the transition state TS2 (43i cm–1) competes with the barrierless formation of 4a (Figure S6). This along with the low free energy difference between 4a and Int′-Me (ΔGInt′-Me-4a = 4.03 kcal/mol) support the formation of the two species, 4a and Int′-Me from Int-Me, which are believed to be in equilibrium in solution of MeCN at r.t.

The second OA reaction (Scheme ) would start with the nucleophilic attack of the dz2 orbital of the Pt(II) center in complex Int′-Me to a second MeI molecule to give 2a as the final product. This step could proceed through a transition state TS3 (423i cm–1), with the energy barrier (EaTS3 = 14.78 kcal/mol) being similar to that of the first OA and thus small enough to be surpassed at room temperature in the dark. Therefore, this calculated mechanism shows the feasible access to Int′-Me, which would explain the observed double OA of MeI to 1a to give 2a. Besides, it shows that 4a is thermodynamically more stable than 2a. Because of this, the scarce solubility of the latter in the reaction media is likely to be the driving force for 2a, which will be the final product of the reaction of 1a with MeI.

Species like Int-Me and Int′-Me have been proposed as intermediates in OA reactions of one or two RX molecules to M2(I,I) (M = Rh, Ir).[17,25] Besides, the mixed-valence species Int′-Me could also be available by a monometallic SN2 OA of MeI to 1a.[31] Aiming to test the proposed mechanism and to compare the structural and spectroscopic features of high-valent Pt2 complexes, with the same core “{Pt(C^C*A)(μ-pz)}2” but with different oxidation states, we addressed the synthesis and characterization of additional compounds such as 3a′, 4a, and Int′-Me (Scheme S1 and Figures S7–S10).

First, to achieve our challenging tasks, Int-Me was prepared as the BF4 salt, [(C^C*A)Pt(μ-pz)2Pt(C^C*A)Me]BF4 (3a′), in a very good yield (85%) (see Scheme , path b) by reacting 1a with Me3OBF4, at −25 °C in anhydrous CH2Cl2 in the dark, under argon atmosphere. Compound 3a′ resulted to be stable in the solid state and solution at room temperature and could be fully characterized (Figure S7). Then, 3a′ was reacted with KI in MeCN at low temperature (−25 °C) to favor the exothermic process (Scheme , path c). In these conditions, [I(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Me] (4a) precipitated in the reaction media and could be obtained as a pure species in a moderate yield (39%) and then characterized (Figure S8). A mixture of 4a and Int′-Me remains in the mother liquor, as it was detected by 1H NMR, which explains the low yield in the synthetic procedure. Further support for the simultaneous formation of both 4a and Int′-Me along with the equilibrium between them was obtained following this reaction by NMR, as can be seen in Figure S9. At −30 °C, this reactions leads to the simultaneous formation of 4a and Int′-Me, with the former being the major species, which becomes Int′-Me as the temperature raises, in such a way that after 24 h at room temperature, both species are present in the mixture in about a 1:1 molar ratio. To reach Int′-Me as pure species, we searched for solvents that give a smaller free energy difference between Int′-Me and 4a than the one obtained in MeCN, so as to ensure a larger amount of Int′-Me in the equilibrium. As can be seen in Scheme , the computed ΔGInt′-Me-4a in anisole (1.79 kcal/mol) is clearly smaller than that in MeCN (4.03 kcal/mol). Accordingly, [(C^C*A)PtII(μ-pz)2PtIV(C^C*A)(Me)I] (Int′-Me) was the single organometallic species detected by 1H NMR in the reaction of 3a′ with Ph4PI in anisole in the dark at 30 °C (Scheme , path d), although it was obtained from the reaction mixture unpurified with anisole and Ph4PBF4 (see Experimental section in the Supporting Information and Figure S10).

Scheme 3

Computed (PCM-M06/6-31G(d), MWB60(Pt), and MWB46(I)) Free Energy Profiles (ΔG, kcal/mol) in MeCN (ε = 35.688) and Anisole (ε = 4.2247) for the Thermal Conversion of 4a, Int-Me, and Int′-Me

Additionally, 3a′ was reacted with MeI in acetonitrile at r.t., rendering 2a′ as the final product (Scheme , path e). This result is consistent with Pt2(III,III) ↔ Pt2(II,IV) formulations for 3a′, the contribution of the Pt2(II,IV) one being significant. Therefore, since all the intermediate species in the double OA of MeI to 1a resulted to be experimentally available, the proposed mechanism, initiated with a bimetallic OA of MeI to the Pt2(II,II) complexes 1a and 1b, seems suitable.

To expand these studies, we focused on the OA of benzyl bromide (BnBr) to [{PtII(C^C*A)(μ-pz)}2] (1a). The reaction of 1a with BnBr in a 1:4 molar ratio in MeCN in the dark at r.t. rendered the Pt2(III,III) complex [Br(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Bn] (5a) (Scheme , path a), which was isolated as an orange solid in very good yield (93%).

Scheme 4

Reaction Pathway for OA Reactions of BnBr to 1a

A second OA to give compound [{PtIV(C^C*)Bn(μ-pz)}2(μ-Br)]Br (6a) was achieved by heating 5a at 70 °C in BnBr(l) in the dark for 5 h. In these hard conditions, 6a was obtained in good yield (87%) (Scheme , path b). Then, 5a and 6a were fully characterized (Figures S11 and S12). The selective formation of 5a in the presence of oxygen, an efficient radical trap, points to a SN2 mechanism, like in the case of MeI (Figure S13), which was modeled by DFT in MeCN. For comparison, the free energy profiles obtained are represented in Scheme and Figure S6.

The energy barrier for the first OA (EaTS1 = 15.92 kcal/mol, TS1: 294i cm–1) is low enough to enable the reaction go at r.t. in the dark, which is not much different from that for MeI. Once the Int-Bn was formed, species 5a or Int′-Bn becomes available. Thermodynamically, the formation of 5a from 1a is clearly favored (calculated ΔG5a–1a = −5.78 kcal/mol; ΔGInt′-Bn-1a = 2.05 kcal/mol). Although the energy barrier (6.24 kcal/mol) for conversion of Int-Bn into Int′-Bn through TS2 (44i cm–1) is in principle not large enough to prevent it to occur, experimentally, 5a is the only species formed at r.t. in the dark. Therefore, it seems that the free energy difference between 5a and Int′-Bn (calculated ΔGInt′Bn-5a = 7.83 kcal/mol) hinders significant formation of Int′-Bn, thus preventing the second OA to occur at r.t. Only by heating at 70 °C in BnBr(l) is the formation of Int′-Bn achieved, enabling it to convert into 6a through TS3 (260i cm–1).

Again, in view of the lower stability of 6a compared to 5a, the scarce solubility of the former in the reaction media is likely the driving force for its formation.

The characterization of all these Pt2 compounds has been addressed together for an overall perspective, as can be seen below.

Characterization of All New High-Valent {Pt(μ-pz)}2 Complexes

In addition to elemental analysis, the most valuable information for the full characterization of these new complexes came from their 1H and 195Pt{1H} NMR spectra in solution (Figures S2–S4 and S7–S12). All these Pt2 complexes, except 1a/b and 3a′, are not stable in solution at r.t. without excess of RX in the media. Thus, the characterization of all these has been carried out at low temperature. Besides, single-crystal X-ray diffraction studies on complexes 2a, 2b, 3a′, and 5a have been carried out. Their molecular structures are depicted in Figure , and selected bond distances and angles are given in Tables S2 and S3.

Figure 2

Molecular structure of the cationic complexes 2a, 2b, 3a′, and 5a. Ellipsoids are drawn at their 50% probability level; solvent molecules, I– (2a, 2b), BF4– (3a′), and hydrogen atoms have been omitted for clarity.

As it can be seen in Figure , in all of them, the Pt2N4 rings exhibit a boat-like shape (angle between the Pt–N–N–Pt fragments being 73.0° 2a, 71.7° 2b, 89.32° 3a′, and 89.64° 5a) with an anti-configuration of the C^C* groups (C1–Pt–Pt#–C#1 torsion angles: 96.7(4)° 2a, 96.1(2)° 2b; C1–Pt1–Pt2–C14 torsion angles: 79.29° 3a′, 72.92° 5a).

The molecular structures of the cationic complexes, [{PtIV(C^C*)Me(μ-pz)}2(μ-I)]+, in 2a and 2b consist of a Pt2(IV,IV) core bridged by two pyrazolates and one iodide ligand. The intermetallic distances (dPt–Pt: 3.5909(9) Å 2a, 3.6228(6) Å 2b) are in between the observed ones in the Pt2(IV,IV) compounds (PPN)[{PtIVMe3(μ-pz)}2(μ-I)] (dPt–Pt: 3.706(1) Å)[29] and (PPh4) [{PtIVMe2Br(μ-pz)}2(μ-Br)] (dPt–Pt: 3.593(1) Å).[30] The PtIV centers exhibit octahedral PtIN2C3 coordination environments with the axial positions occupied by one Me group and the iodine bridge, the Pt–I–Pt angle being close to 80° (81.55(2) 2a, 82.69(13) 2b). The Pt–I, Pt–N, and Pt–C distances seem to be not affected by the metal oxidation state since they are quite similar to those observed in Pt2(III,III) complexes containing the same kind of ligands.[26,27]

The cationic complex [Pt2(C^C*A)2(μ-pz)2Me]+ in 3a′ and 5a show short intermetallic distances (dPt–Pt: 2.6700(4) Å 3a′, 2.6545(5) Å 5a) indicative of the existence of a Pt–Pt bond in each of them. All bond distances and angles are very similar to those observed in analogous complexes with the [(C^C*)PtIII(μ-pz)]2 core and octahedral environment at each Pt center.[26,27] In 3a′, the platinum centers show different coordination environments: octahedral for Pt1 with the Pt2 and the methyl group (C33) in the apex positions and distorted square pyramidal for Pt2, with Pt1 in the apex position. The intermetallic distance in 3a′ is in the range reported for “Pt2III(μ-L)2R” species, no matter if they exhibit an octahedral environment of each Pt center [2.529(1)–2.7910(2) Å][32,33] or octahedral geometry at one and square pyramidal at the other center. The latter is exemplified by complex [R(H3N)2Pt(μ-L-N,O)2Pt(NH3)2]3+ (L-N,O- = amidate or pyridonate) [2.676(1)–2.7542(11) Å].[34−41]

Regarding pyrazolate-bridged complexes, the intermetallic distance in 3a′ is a little longer than those in [(CHX2)(C^C*)PtIII(μ-pz)PtIII(C^C*)X],[26,27] (dPt–Pt = 2.6302(4) Å X = Br; 2.6324(3) Å X = I) or in 5a, which can be attributed to the larger trans influence of CH3 compared to CHX2 and CH2Ph.

The NMR spectra of the Pt–Me derivatives, 2a, 2b, 3a′, 4a, and Int′-Me, as well as the Pt–Bn ones, 5a and 6a, were performed in CD2Cl2 solution (see Table and Figures and 4).

Table 1

Relevant NMR Data for the New High Oxidation State Pt2 Complexesa

compound	δ 195Pt–X	δ 195Pt–R	1JPt–Pt	δ1H (Pt–R)	2JPt–H	3JPt–H
2a		–2688.0 (R = Me)		1.78	65.5
2b		–2664.4b (R = Me)		1.80	66.1
3a′	–3064.2 X = vacant	–2589.2 (R = Me)	1023.4	2.42	70.7	14.7
4a	–3018.8 (X = I)	–2848.2 (R = Me)	1239.8	1.48	61.0	14.5
Int′-Me	–3776.0 (PtII)	–2697.0 (PtIV)		1.75	65.9
5a	–2742.8 (X = Br)	–2693.6 (R = Bn)	1028.9	3.73 (1H)c	69.0	28.9
6a		–2357.0 (R = Bn)		4.02 (2H)c	90.6

CD2Cl2, more details in experimental Section, δ (ppm); J (Hz).

Indirect detection by 1H–195Pt HMQC NMR.

An equal signal appears overlapped with OCHCH3; δ 195Pt = −3778.0 ppm (1a, acetone-d6), −3767.4 ppm (1b, DMSO-d6).

Figure 3

Expanded view of the 1H NMR spectra of 2a (left), 3a′ (middle), and 4a (right) in CD2Cl2.

Figure 4

Left: 195Pt{1H} NMR spectra in CD2Cl2 of 2a, 3a′, and 4a. Right: 195Pt–195Pt{1H} COSY spectrum of 4a in CD2Cl2.

Their 1H NMR and 195Pt{1H} NMR spectra showed that in all cases, the major isomer is that observed in the X-ray single-crystal structures, with the C^C* groups in an anti-conformation, which provided structurally relevant details. In agreement with the absence of metal–metal interactions and their symmetry, the Pt2(IV,IV) complexes 2a, 2b, and 6a show the coupling to just one 195Pt nucleus (see 2JPt-H in Table ) of their corresponding Pt–R (R = CH3, CH2Ph) 1H NMR signals. Besides, their 195Pt{1H} NMR spectra exhibit just one singlet in the typical spectral range for Pt(IV) compounds (see Table and Figure for 2a).[42]

By contrast, in compounds 3a′, 4a, and 5a, the NMR spectra denote the non-equivalence of two Pt fragments joined by a metal–metal bond. That is, each compound exhibits a signal due to the Pt-CH3 (singlet) or Pt-CH2Ph (doublet) flanked by two sets of 195Pt satellites in its 1H NMR spectrum and two singlets in the 195Pt{1H} NMR one, each one flanked by 195Pt satellites.

The existence of a Pt–Pt bond was confirmed by a 195Pt–195Pt{1H} COSY spectrum, which displays a crosspeak due to scalar coupling (Figure right for 4a and S6 for 3a′).

The assignment of these resonances was made from 1H–195Pt HMQC and 1H {selective195Pt} NMR experiments.

All the 195Pt signals appear clearly downfield-shifted with respect to those of the Pt2(II,II) complexes, 1a and 1b (Table ), according to the higher oxidation state of the metal centers. They appear more deshielded as the oxidation state is higher [see δ195Pt for PtIV-CH3 (2a) and PtIV-CH2Ph (6a) vs PtIII-CH3 (4a) and PtIII-CH2Ph (5a)], and the electronegativity of the axial ligand is greater [see δ195Pt for Pt–Br (5a) vs Pt–I (4a)]. Besides, a downfield shift of the 195Pt-CH2Ph resonances with respect to the 195Pt–Me one is observed (see δ195Pt for 6a vs 2a), which is attributed to the effect of the π system of the benzyl fragment.[43]

The proposed structure for complex Int′-Me was based on its NMR data. The presence of two singlets in the spectral range expected for PtII and PtIV and the absence of platinum satellites in its 195Pt{1H} NMR spectrum denote the mixed-valence nature of Int′-Me and the absence of a metal–metal bond between the platinum centers. This fact was confirmed by its 1H NMR spectrum, which shows only one singlet corresponding to the Pt–Me group flanked just by one set of platinum satellites (Table , Figure S10).

The complex cation in 3a′ deserves some additional attention. In this complex, the average oxidation number of the platinum centers is +3, but it can be regarded as a metal–metal bonded Pt2(III,III) complex with just one axial ligand, or as a mixed valence Pt2(II,IV) one[44] with the metals linked by a PtII → PtIV donor–acceptor bond. The short intermetallic distance (2.6700(4) Å) observed in the X-ray structure points to a Pt2(III,III) formulation, while the NMR data (Figure S7) point to a PtII → PtIV one. In this sense, the different coordination environments of the Pt centers cause a big separation between the two 195Pt resonances up to 480 ppm. The one corresponding to Pt–Me appears even more deshielded than that in the Pt2(IV,IV) compounds (2a and 2b), while the other is shielded 50 ppm with respect to the Pt–I resonance in the Pt2(III,III) complex 4a. To help determine the correct oxidation states of the Pt centers, we performed additional computational and electrochemical studies. The Mulliken population analysis in MeCN for 3a′ provided an estimated partial charge for the two platinum centers (0.49 Pt, 0.46 Pt–Me) not much different one to another, the difference (Δ = 0.03) being even lower than in the Pt2(III,III) complex, 4a (0.35 Pt–I, 0.40 Pt–Me, Δ = 0.05). The Pt–Pt MO bond order in 3a′ (0.38) is close to the calculated value for the Pt2(II,II) complex 1a (0.39) and smaller than that found for the Pt2(III,III) complex 4a (0.59). These calculations are consistent with a Pt2(III,III) ↔ Pt2(II,IV) formulations, the contribution of the Pt2(II,IV) one being significant, in line with earlier calculations on catalytic processes involving [{PdIII(C^N)(OAc)}2XY]. They showed that when a strong σ-donor group is “axially” coordinated to one of the metal centers in dinuclear complexes, the dz2 orbital from the other metal gets populated, increasing the M(II) character and favoring the Pt2(II,IV) formulation.[45] Therefore, in our case, the presence of Me as the electron-donating group in 3a′ would increase the Pt2(II,IV) contribution to this molecule.

In this sense, oxidative CV in MeCN showed for 3a′ an irreversible oxidation at 0.39 V (given vs the Fc+/Fc couple). The value is quite similar to that of 1a, 0.44 V, measured under the same conditions and to the related cyclometalated pyrazolate-bridged dinuclear platinum(II) complexes,[46] while being far from the value observed for the Pt2(III,III) complex [{Pt(C^C*A)(μ-pz)I}2] (Figure S14). In the Pt2(II,II) complexes, the irreversibility of this oxidation process has been attributed to the square-planar geometry of each Pt(II) unit with little or no metal–metal interaction. In them, the metal centers are highly susceptible to nucleophilic attack by coordinating solvents, such as MeCN, resulting in permanent oxidized products. Therefore, a mixed-valence Pt2(II,IV) formulation with the metals linked by a PtII → PtIV donor–acceptor bond seems to be the most likely for 3a′ in MeCN solution, which is also compatible with the observed metal–metal coupling.[47] According to that, we observed that compound 3a′ reacted with MeI in acetonitrile to give 2a′.

Conclusions

Compound [{PtII(C^C*A)(μ-pz)}2] (1a) reacted with MeI and BnBr at room temperature in the dark to give the high-valent dinuclear complexes [{PtIV(C^C*A)Me(μ-pz)}2(μ-I)]I (2a) and [Br(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)Bn] (5a), resulting from the double or single OA of RX, respectively. Also, [{PtII(C^C*B)(μ-pz)}2] (1b) reacted with MeI, affording [{PtIV(C^C*B)Me(μ-pz)}2(μ-I)]I (2b), indicating that the CO2Et substituent in C^C*A does not affect the redox behavior of these Pt2(II,II) compounds 1a and 1b. DFT modeling of the SN2 mechanisms for the OA of RX to 1a proposed species such as [(C^C*A)Pt(μ-pz)2Pt(C^C*A)R]X (RX = MeI Int-Me, BnBr Int-Bn) as intermediates for the first OA reaction. Once formed, two species are accessible, [X(C^C*A)PtIII(μ-pz)2PtIII(C^C*A)R](RX = MeI 4a, BnBr 5a) and [(C^C*A)PtII(μ-pz)2PtIV(C^C*A)(R)X] (RX = MeI Int′-Me, BnBr Int′-Bn), the latter being the intermediate for the second OA. Keeping in mind the small energy barrier for the transformation of Int-R into Int′-R, the free energy difference between the species Pt2(III,III) (4a or 5a) and Pt2(II,IV) (Int′-Me, Int′-Bn) seems to determine the nature of the compounds obtained at r.t. When it is small (ΔGInt′-Me-4a = 4.03 kcal/mol), the feasible formation of Int′-Me allows the second OA to occur, providing the Pt2(IV,IV) complex, 2a. When it is bigger (ΔGInt′-Bn-5a = 7.83 kcal/mol), the reaction leads to the selective formation of the Pt2(III,III) complex 5a. In this case, the second OA to get [{PtIV(C^C*A)Bn(μ-pz)}2(μ-Br)]Br (6a) is possible under harder conditions. Species Int-Me could be prepared and isolated as the BF4 salt, 3a′, and then used to get 4a, Int′-Me, and 2a′, which indicate this computed mechanism as the most likely one and allow us to compare structural and spectroscopic data for complexes with the same core [{Pt(C^C*)(μ-pz)}2] but different oxidation states.