P- and N-Coordination of the Ambidentate Ligand HN[P(i-Pr)2]2 with

May 31, 2017 - Five different coordination motifs were observed upon reaction of the simple group 13 Lewis acids MCl3 (M = In, Ga, Al, B) or BF3·Et2O...
0 downloads 9 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

P- and N‑Coordination of the Ambidentate Ligand HN[P(i‑Pr)2]2 with Group 13 Trihalides Diane A. Dickie,†,‡ Ujwal Chadha,† and Richard A. Kemp*,†,§ †

Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, New Mexico 87131, United States Advanced Materials Laboratory, Sandia National Laboratories, Albuquerque, New Mexico 87106, United States

§

S Supporting Information *

ABSTRACT: Five different coordination motifs were observed upon reaction of the simple group 13 Lewis acids MCl3 (M = In, Ga, Al, B) or BF 3 ·Et 2 O with the ambidentate bis(diisopropylphosphino)amine ligand HN[P(i-Pr)2]2. In a 1:1 reaction mixture, the softer Lewis acids InCl3, GaCl3 and BCl3 coordinate to one of the two P atoms of the ligand. In contrast, AlCl3 and BF3 prefer coordination to the harder N atom. In all cases, the acidic N−H proton is shifted to P upon complexation with a metal. By altering the reaction stoichiometry, 2:1 metal− ligand complexes could be isolated for three of the combinations. BCl3 gives a bis-adduct via the two P atoms. GaCl3 produces a salt consisting of a [GaCl4]− anion and a P,P-chelated [LGaCl2]+ cation. Most unexpectedly, the reaction with InCl3 in methanol resulted in solvent deprotonation by the ligand to give two symmetric [(i-Pr2PH)2N]+ cations in which all the basic P sites are coordinated to H rather than the group 13 Lewis acid. These cations are balanced by the unique complex dianion [(MeO)6In4Cl8·2MeOH]2−. All complexes were characterized with a combination of multinuclear NMR spectroscopy and singlecrystal X-ray diffraction.

■

INTRODUCTION When introducing the concept of Lewis acid−base theory to undergraduate students, textbook examples most often include a trivalent group 13 element for the Lewis acid and an amine or phosphine as the base. This is also the most common, although far from exclusive,1−4 combination used in the flourishing field of frustrated Lewis pairs (FLPs).5−7 Because FLPs are too sterically hindered to bond to one another, they are often very reactive toward small molecules such as H2 and CO2. This steric “frustration”, however, is not always needed to obtain comparable FLP-like reactivity.8−15 For example, (Mes)3P and the simple, relatively inexpensive Lewis acid AlCl3 form a crystalline adduct whose association is maintained in solution.16 This adduct facilitates the reduction of CO2 to MeOH, although the optimized conditions require a second equivalent of AlCl3. Given the myriad of commercially available or easily synthesized tertiary phosphines that are known, and their potential for FLP-like reactivity when paired with simple Lewis acids such as group 13 halides, it is surprising that only ∼60 examples of R3P-MCl3 (M = B, Al, Ga or In) adducts are currently listed in the Cambridge Structural Database.17 We have had a long-standing interest in using phosphine ligands in the form HN(PR2)2 (R = Ph, i-Pr) or HN(SiMe3) (PR2) (R = i-Pr) to create complexes of divalent main group elements and earth-abundant transition metals including Mg,18 Ca,18,19 Sr,18,19 Ge,20 Sn,15,20 Zn,21,22 and Ni.23 All of these species have been shown to react with CO2 in various ways. Recently, we have expanded our studies to include Lewis acidic trivalent © 2017 American Chemical Society

group 13 elements. Deprotonation of HN[P(i-Pr)2]2 (1) followed by addition of InCl3 in a 2:1 ratio gives a P,Pchelated complex that reversibly inserts two equivalents of CO2 into its In−P bonds,24 but the need to use the anionic form of the ligand limits the potential of this ligand system to further reduce CO2 into valuable products like methanol as it is incompatible with protic environments. Neutral 1 is far more tolerant of such conditions, and it reacts with bulky B(C6F5)3 to form a stable 1:1 adduct into whose P−B bond CO2 readily inserts.25 One drawback with this system is that B(C6F5)3 is far more expensive on a per mole basis than the simple group 13 MX3 Lewis acids. To determine if the less bulky group 13 halides could be substituted for B(C6F5)3 in this system, MX3 salts were combined with 1 under various conditions, giving a series of new molecules whose syntheses, characterization, and preliminary reactivity are described below.

■

RESULTS AND DISCUSSION The group 13 trichlorides MCl3 (M = In, Ga, Al, B) were each combined with HN[P(i-Pr)2]2 1 in a 1:1 ratio in concentrated CH2Cl2 or Et2O solutions (Scheme 1). After standard workup, white solids were obtained in high yields. The 1H NMR spectra of each of the complexes 2−5 (2 = In, 3 = Ga, 4 = Al, 5 = B) were quite similar to one another, as well as to the previously prepared adduct of 1 with B(C6F5)3.25 In each case, two sets of isopropyl peaks were observed, indicating that the symmetric Received: April 25, 2017 Published: May 31, 2017 7292

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry Scheme 1. Synthesis of Complexes 2−5

showed no signs of broadening or coupling that would be consistent with a P−Al bond. Likewise, no P−Al coupling was resolved in the very broad resonance at δ 52 ppm in the 27Al NMR spectrum, although a small, sharp doublet at δ 108 ppm (1JAl−P = 316 Hz) was also present. Given that 1 is an ambidentate molecule with both “hard” N and “soft” P donor atoms, the spectroscopic data of 4 could be explained if AlCl3, the hardest of the four MCl3 Lewis acids used, was bound to N instead of P with only trace amounts of the P-bound isomer being present in solution. Single-crystal X-ray diffraction studies were undertaken to look for evidence of this coordination mode. Suitable crystals of 2−5 were readily grown from concentrated CH2Cl2 or Et2O solutions, and the solid-state structures were consistent with those proposed based on the solution-state NMR data (Figures 1−4). In complexes 2, 3, and

nature of 1 was not maintained for 2−5. The CH3 groups appeared as four distinct doublets of doublets due to splitting by both a CH proton and a P atom. Of the two CH signals in each complex, the downfield signal was the expected doublet of a septet, showing coupling to two methyl groups and one P atom. This was also the case for the upfield signal in 5, but an additional splitting into a doublet of doublets of a septet was observed for the heavier congeners 2−4. This extra coupling corresponded to the proton that had migrated from its position on the nitrogen atom in free 1 to a phosphorus atom upon complexation with the MCl3 moieties. In compounds 2−5, proof of the N/P tautomerization is the 1 H NMR signal with a large P−H coupling of 438 Hz in 2, 448 Hz in 3, 470 Hz in 4, and 451 Hz in 5 with chemical shifts of δ 6.83, 6.82, 6.52, and 6.86 ppm for 2−5, respectively. Further confirmation for the formation of the P−H tautomer rather than the N−H can be found in the IR spectra. No signal corresponding to an N−H stretch was observed for any of the complexes. Instead, signals found between 2337 and 2403 cm−1 can be readily assigned to the P−H stretches of the complexes. Similar N/P tautomerization was first reported for HN(PR2)2 ligands by Bertrand 26 and has since been observed experimentally specifically for 1 in its reactions with tris(pentafluorophenylborane)25 as well as carbon disulfide,27 elemental selenium,28 and tellurium.29 Computational studies have been preformed for the latter two examples.30 No such tautomerization was observed, however, in the reaction of 1 with a variety of NiX2 salts.23 In contrast to the relative similarities of the 1H NMR and IR spectra of the four molecules 2−5, a difference was immediately apparent upon examination of the 31P{1H} NMR spectra. In 2, 3 and 5, but not in 4, a P,P-coupled doublet appeared at δ 39.4, 36.7, and 32.7 ppm, respectively. The coupling constants were in the range of 11−13 Hz. When proton decoupling was turned off, the strong 1JPH coupling noted in the 1H spectra was observed, confirming that these signals corresponded to the P− H fragment of each complex. In addition to this resonance, the 31 1 P{ H} spectrum of 5 contained a broad quartet of doublets centered at δ 33.9 ppm, precisely what would be expected if the BCl3 moiety was coordinated to that P atom. In the case of 2 and 3, no corresponding signal was resolved, consistent with pronounced broadening by a quadrupolar In or Ga nucleus bound to P. The 31P{1H} spectrum of 4 was quite distinct. A very small doublet at δ 35.5 ppm was observed, consistent with the signal for 2, 3, and 5, but the major signals were two sharp doublets at δ 86.6 and 56.3 ppm. These signals were well downfield of the resonances measured for 2, 3, and 5, and the coupling constant of 54 Hz was significantly larger as well. The proton-coupled 31 P NMR spectra allowed the assignment of the more upfield of the two signals to a P−H group, but the downfield doublet

Figure 1. Structure of 2. Thermal ellipsoids are shown at 50% probability, and all H atoms except P−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): In1−P1 = 2.5423(4), P1−N1 = 1.6054(14), P2−N1 = 1.5794(14), P1−N1−P2 = 134.91(9).

Figure 2. Structure of 3. Thermal ellipsoids are shown at 50% probability, and all H atoms except P−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Ga1−P1 = 2.3704(6), P1−N1 = 1.6033(18), P2−N1 = 1.5751(18), P1−N1−P2 = 138.50(12). 7293

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry

fragmentation and rearrangement similar to what we and others have previously reported for [N(PR2)2]− anions; however, no evidence of such reactivity was detected in this case. As noted above, the preference of AlCl3 for N-coordination rather than P-coordination is consistent with the predictions of hard−soft acid−base theory. To see if this pattern would hold, we next tested the prototypical hard acid BF3. The reaction of 1 with BF3·Et2O (Scheme 2) in either CH2Cl2 or Et2O produces 6 in high yield. The 1H NMR spectrum was similar to those of 2−5, showing two sets of chemically inequivalent isopropyl groups. The resonance assigned to the acidic H atom appeared as a broad doublet at δ 6.13 ppm. The P−H coupling constant of 465 Hz indicated once again that the product was a P−H tautomer rather than an N−H tautomer, and though the magnitude of the coupling is far closer to that of 4 than the others, that alone is not sufficient evidence to distinguish N−B vs P−B binding. For that, it was necessary to look at the spectra of the other NMR-active nuclei of 6. The 31P{1H}, 11B{1H}, and 19F{1H} spectra each displayed a pattern of signals consistent with N−B coordination. Two doublets at δ 101 and 76 ppm (2JPP = 110 Hz) were found in the 31P{1H} spectrum, whereas the 11B{1H} and 19F{1H} signals were each quartets with B−F coupling constants of 22 Hz due to one I = 3/2 boron nucleus and the three I = 1/2 fluorine nuclei. As was the case with 4, a second, much smaller set of peaks consistent with P−B coordination was apparent in the heteronuclear spectra, but no conditions were found that would allow us to isolate this isomer from the solution. Final confirmation for the identity of 6 as the N-bound isomer was provided by single-crystal X-ray diffraction (Figure 5). Two crystallographically independent molecules were found

Figure 3. Structure of 4. Thermal ellipsoids are shown at 50% probability, and all H atoms except P−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Al1−N1 = 1.8658(11), P1−N1 = 1.7623(11), P2−N1 = 1.6437(10), P1−N1−P2 = 114.33(6), P1−N1−Al1 = 130.12(6), P2−N1−Al1 = 115.42(6).

Figure 4. Structure of 5. Thermal ellipsoids are shown at 50% probability, and all H atoms except P−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): B1−P1 = 1.9803(15), P1−N1 = 1.6016(11), P2−N1 = 1.5734(11), P1−N1−P2 = 135.84(7).

5, the metal is bound to P1, and the acidic H atom is found on P2. The P2−N1 bonds are shorter than the P1−N1 bonds but not dramatically so, suggesting that there is some delocalized double bond character across the P−N−P moiety consistent with the resonance forms drawn in Scheme 1. The crystal structure of 4 shows the aluminum is bound to N rather than to one of the P atoms, and this solid-state structure is consistent with the solution-phase NMR data. To the best of our knowledge, this is the first time that monodentate Ncoordination has been observed for neutral (nondeprotonated) HNPR2 ligands among some 200 crystallographically characterized structures of this ligand class.17 The location of the AlCl3 moiety forces the P1−N1−P2 angle of 4 to be far more acute than for the P-bound adducts but not nearly as acute as the angles found in previously reported P,P-chelated complexes.15,20,21,23 As in 2, 3, and 5, the P−N bonds of 4 are unequal, but both are longer than any in the other three complexes. In fact, the P1−N1 bond is substantially longer than in free 115 and well into the upper quantile of the range of P−N bonds reported in the CSD.17 This unusually long, presumably weaker bond could be expected to encourage ligand

Figure 5. Structure of 6. Thermal ellipsoids are shown at 50% probability, and all H atoms except P−H have been omitted for clarity. Only one of two crystallographically independent molecules is shown. Selected bond lengths (Å) and angles (deg): B1−N1 = 1.5790(16), B2−N2 = 1.5740(16), P1−N1 = 1.7538(10), P2−N1 = 1.6461(10), P3−N2 = 1.7523(10), P4−N2 = 1.6382(9), P1−N1−P2 = 112.79(5), P3−N2−P4 = 114.35(5), P1−N1−B1 = 128.30(8), P2−N2−B1 = 118.91(7), P3−N2−B2 = 125.31(7), P4−N2−B2 = 119.48(8).

in the asymmetric unit with the greatest difference between them being the P−N−B angles. Otherwise, the structure of 6

Scheme 2. Synthesis of BF3 Complexes 6 and 7

7294

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry

[H3NP(i-Pr)2][BF4], but we have no evidence for anything but 7 at this time. The 11B{1H} NMR signal of 6 was a B−F coupled quartet, but in 7 it is a triplet due to the loss of one F atom. No P−H or N−H resonance was observed in the 1H NMR spectrum, but two doublets assigned to the geminal protons of the exocyclic CC double bond of the former acetonitrile fragment were detected at δ 5.43 and 3.75 ppm. Their very different 3JPH coupling constants of 38 and 14 Hz, respectively, are consistent with their distinct spatial relationships to the endocyclic P atom. The 1H NMR spectrum also shows the presence of three unique isopropyl environments. One belongs to the substituents of the endocyclic P atom, one to the other P atom of “intact” 6, and the final group is the diisopropylphosphino fragment that was formally cleaved and transferred to acetonitrile N. With 7 serving as evidence that the addition of multiple bonds across these Lewis pairs is possible, we next turned our attention to CO2. One of the potential advantages of incorporating 1 into systems designed to capture CO2 is that it contains multiple Lewis basic sites and could therefore interact with carbon dioxide’s electrophilic central atom in more than one way. In previously published work (vide supra), the mode of interaction had always been formal insertion into a P−M or N−M bond, but 7 suggested other modes of interaction might also be possible. Regrettably, no evidence of CO2 reactivity was detected when solutions of 2−6 in a variety of solvents were exposed to CO2 for up to 1 h. This was not entirely surprising given that the adduct of 1 with B(C6F5)3 reacts readily with CO2, but its P−B bond is 2.083(5) Å,25 whereas the corresponding bond in 5 is much shorter at 1.9803(15) Å. Likewise, the In−P bond of 2 measures 2.5423(4) Å, which is 0.1−0.2 Å shorter than those found in the CO2-reactive compound [(i-Pr2P)2N]InCl.24 This suggests that the M−P or M−N bonds of 2−6 are too strong to be disrupted under these relatively mild conditions. Nevertheless, the isolation of 7 is evidence that 6 can be reactive toward at least one small molecule (acetonitrile) under at least some conditions. Therefore, we decided to focus on the remaining Lewis basic sites. Amines and other nitrogen bases have long been known to capture CO2 on their own,31,32 but we are aware of only one report of phosphine being electron-rich enough to complex CO2 without a Lewis acidic counterpart to simultaneously bind carbon dioxide’s nucleophilic oxygen atoms.33 Because pairing one Lewis basic site of 1 with a Lewis acid did not give the desired reactivity, we set out to prepare 2:1 combinations of MX3 with 1 to bring a second acid/base pair into play. Immediately upon reaction with two equivalents of GaCl3 in CH2Cl2, a sharp singlet with a chemical shift of δ 113 ppm was

was quite similar to that of 4, particularly in the relatively long P1−N1/P3−N2 bonds. We noted above that the long, weak P1−N1 bond of 4 remained intact under all conditions that it was exposed to, but the same is not true of 6. Diethyl ether and CH2Cl2 were found to work equally well as solvents for the synthesis and crystallization of compounds 2−6, but they were not the only solvents tested. During one attempted preparation of 6, acetonitrile solvent was used, and a very different product 7 was obtained (Scheme 2). Despite numerous attempts, we have been unable to repeat our initial isolation of crystalline 7. Nevertheless, through a

Figure 6. Structure of 7. Thermal ellipsoids are shown at 50% probability, and all H atoms have been omitted for clarity. Selected bond lengths (Å) and angles (deg): B1−N1 = 1.574(3), B1−N2 = 1.536(3), N1−P1 = 1.6322(19), N1−P2 = 1.735(2), N2−P3 = 1.7338(19), N2−C19 = 1.395(3), C19−C20 = 1.328(3), P1−N1−P2 = 118.74(11), P1−N1−B1 = 110.19(15), P2−N1−B1 = 130.46(15), N1−B1−N2 = 105.69(18), B1−N2−P3 = 130.02(15).

combination of single-crystal X-ray diffraction (Figure 6) and 1 H, 11B{1H}, and 19F{1H} NMR spectral data obtained from the initial synthesis, we have been able to conclusively identify 7 as the diazaphosphaborole ring shown in Figure 6. One of the P atoms from the nearly intact fragment of 6, along with its N and B atoms, makes up three-fifths of the heterocyclic core of 7. The remainder of the ring consists of the terminal N and central C of an acetonitrile solvent molecule that have formed bonds with the B and P of 6. Without suggesting any possible mechanism of formation, it is useful to think of 7 as the cycloaddition product of one molecule of 6 with acetonitrile, appended by a P(i-Pr)2 fragment that has been cleaved from the long, weak N−P bond of a second molecule of 6. Formally lost from that process are fragments that add up to the salt Scheme 3. Synthesis of the 2:1 Adducts 8 and 9

7295

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry the only signal observed in the 31P{1H} NMR spectrum of the crude reaction mixture. It was assigned to the presumed 2:1 complex 8. Under similar conditions, the reactions of 1 with two equivalents of InCl3, AlCl3, and BF3·Et2O showed only the 1:1 products 2, 4, and 6, respectively, over several hours of monitoring by 31P{1H} NMR spectroscopy. The reaction with BCl3 also appeared to form 5 initially, but a broad quartet at δ 51.6 ppm slowly grew in over time, and it was assigned to a new product 9. The 1H NMR spectra of 8 and 9 were collected, and each showed a single isopropyl environment, suggesting that unlike 2−7, 8 and 9 were symmetric complexes. A resonance integrating to one hydrogen atom appeared at δ 5.16 ppm in 8 and δ 3.48 ppm in 9. As both N−H and P−H have rather broad ranges for chemical shifts, this information alone was not sufficient to distinguish between the two tautomers, but the absence of P−H coupling suggested that N−H was far more likely to be present. This assignment was confirmed by the IR spectra, which showed unambiguous N−H stretches for 8 and 9 at 3217 and 3238 cm−1, respectively. Assuming that the 2:1 stoichiometry is correct, two possible arrangements, shown in Scheme 3, could be consistent with the combination of NMR and IR data. Although gallium is not particularly amenable to NMR spectroscopy, boron is, and the 11B{1H} spectrum of 9 showed only one signal, a doublet at δ 3.6 ppm (1JPB = 155 Hz). This effectively ruled out the cation/anion product for 9, in which one would expect to see two 11B NMR signals including a P-coupled triplet and indicated instead the bis-adduct. The former seemed more likely for 8 because its 31P chemical signal was shifted so far downfield, as one would predict if the P atoms were bound to an electron-poor cationic metal center. Fortunately, both 8 and 9 were isolated as colorless, crystalline products, and thus we were able to determine their solid-state structures unambiguously by single-crystal X-ray diffraction. The structure of 8 (Figure 7) revealed the predicted

The structure of 9 (Figure 8) was also consistent with the predictions based on the NMR spectra. The P−B bonds of the

Figure 8. Structure of 9. Thermal ellipsoids are shown at 50% probability, and all H atoms except N−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): P1−B1 = 2.011(3), P2−B2 = 1.995(3), P1−N1 = 1.684(2), P2−N1 = 1.683(2), B1−P1−N1 = 117.24(11), B2−P2−N1 = 103.29(12), P1−N1−P2 = 141.87(14).

bis-adduct are the same as in the 1:1 complex 4, within experimental error, but the P−N bonds are longer. The P−N− P angle is also somewhat greater at 141.87(14)° in 9 compared to 135.84(7)° in 4. As with the cation/anion pair, there are a handful of related MX3 bis-adducts of bidentate P-ligands in the literature,35,40−48 although only one with boron.49 It is not immediately clear to us why one structure is favored over the other, although the difference in solvent polarity caused by the hexane solvent of the commercial BCl3 solution could potentially be a factor that favors the formation of the neutral adduct rather than the cation/anion pair. As noted above, in CH2Cl2 solution, no 2:1 complex of InCl3 could be isolated or identified spectroscopically. However, because InCl3 is known to be soluble in alcohol, an attempt was made to run the 2:1 reaction in methanol. The reaction was very rapid, and 10 was isolated as colorless crystals. The IR spectrum of 10 displayed a P−H stretch at 2348 cm−1, in contrast to the N−H stretch seen in the other 2:1 adducts 8 and 9. This was not a simple tautomerization as in the 1:1 compounds 2−6, however, as the 1H and 31P{1H} NMR spectra indicated a symmetric compound. Instead, single-crystal X-ray diffraction of 10 revealed that ligand 1 had been protonated to give a [(i-Pr2PH)2N]+ cation (Figure 9). We are aware of only one related cation that has been crystallographically characterized, namely, the unsymmetric (t-Bu)2PHN-PH(t-Bu)(Me)50 isolated by Cristóbal-Lecina et al. as a BF4− salt. The anion of 10 is far more complex than that of 8 or the above-mentioned BF4 salt. It is a unique (MeO)6In4Cl8· 2MeOH dianionic cluster. The indium atoms are each 6coordinate octahedral centers with two terminal, cis-oriented chloride ligands. The methoxy ligands act as either μ2 (O2 and O3) or μ3 (O4) bridges between the indium atoms. Four of these methoxy groups are bound to In2, whereas In1 accepts three and completes its coordination sphere with a neutral methanol. Four broad singlets of three protons each were observed in the 1H NMR spectra, consistent with the three unique methoxy and one methanol environments of the anion. Although the GaCl3, BCl3 and InCl3 complexes 8−10 each demonstrated new coordination modes for 1, initial reactivity studies with CO2 had similar results to those of the 1:1 complexes 2−6. This is not entirely surprising, as the Lewis acidic and basic sites of the metal and ligand, respectively, are fully satisfied in these complexes, and the M−P bonds are of

Figure 7. Structure of 8. Thermal ellipsoids are shown at 50% probability, and all H atoms except N−H have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Ga1−P1 = 2.4070(4), Ga1−P2 = 2.4038(4), P1−N1 = 1.6928(15), P2−N1 = 1.6857(14), N1···Cl3 = 3.3722(13), P1−Ga1−P2 = 72.430(15), P1− N1−P2 = 114.55(7), N1−H1···Cl3 = 178.(2).

salt consisting of a P,P-chelated GaCl2 cation and a GaCl4 anion. Although this type of complex is well-known for nitrogen-based ligands such as diimines, diamines, or bipyridines, it has limited precedent with P donor ligands.34−39 Each gallium atom is in a tetrahedral environment, although the angles around Ga1 are distorted due to the chelating ligand. A hydrogen bond runs from the ligand N−H to one of the GaCl4 chloride atoms in the counterion. 7296

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry

received. HN[P(i-Pr)2]2 (1) was prepared according to literature procedures.52 1H and 13C{1H} spectra were referenced to residual solvent downfield of TMS. 31P{1H}, 27Al, 11B{1H}, and 19F{1H} spectra were referenced to 85% H3PO4, Al(NO3)3, BF3·Et2O, and CFCl3, respectively. IR spectra were recorded as mulls or thin films on KBr windows. Elemental analyses were obtained from ALS Global (Tucson, AZ) or on a PerkinElmer 2400 Series II CHNS/O Analyzer. Single-crystal X-ray diffraction studies were performed on a Bruker Kappa APEX II CCD system equipped with a graphite monochromator and a Mo Kα fine-focus tube (λ = 0.71073 Å). Crystals were coated in Paratone-N oil and mounted on a MiTeGen MicroLoopTM. The Bruker Apex2 and Apex3 software suites were used for data collection, structure solution, and refinement. Relevant data and parameters are summarized in Tables S1 and S2. Synthesis of Cl3In[P(i-Pr)2NP(H)(i-Pr)2] (2). InCl3 (0.44 g, 2.0 mmol) was added to a solution of HN[P(i-Pr)2]2 (0.50 g, 2.0 mmol) in ∼5 mL of CH2Cl2. The colorless solution was allowed to stir at rt for 1.5 h, and then the solvent was removed under vacuum to give 2 as a white powder. Yield = 0.89 g (95%), mp 84−86 °C. 1H NMR (300 MHz, C6D6) δ 6.83 (br dt, 1JPH = 438 Hz, 3JHH = 2.1 Hz, 1H, PH), 1.87 (d sept, 3JHH = 7.2 Hz, 2JPH = 7 Hz, 2H, CH), 1.50 (br dd sept, 3 JHH = 7.2 Hz, 3JHH = 1.8 Hz, 2JPH = 7 Hz, 2H, CH), 1.09 (dd, 3JHH = 7.2 Hz, 3JPH = 18.5 Hz, 6H, CH3), 0.95 (dd, 3JHH = 7.2 Hz, 3JPH = 16.5 Hz, 6H, CH3), 0.70 (dd, 3JHH = 7.2 Hz, 3JPH = 16.7 Hz, 6H, CH3), 0.66 (dd, 3JHH = 7.2 Hz, 3JPH = 18.7 Hz, 6H, CH3) ppm. 13C{1H} NMR (75 MHz, C6D6) δ 29.6 (d, 1Jpc = 37 Hz, CH), 25.5 (d, 1JPC = 67 Hz, CH), 17.5 (s, CH3), 16.6 (s, CH3), 16.1 (s, CH3), 15.5 (s, CH3) ppm. 31 1 P{ H} NMR (121 MHz, C6D6) δ 39.4 (d, 2JPP = 12 Hz, PH) ppm (In-bound P not observed). IR (nujol mull) ν 2337 (P−H) cm−1. Anal. Calcd for C12H29Cl3InNP2: C, 30.63; H, 6.21; N, 2.98. Found: C, 30.84; H, 6.36; N, 2.75. X-ray quality single crystals were grown from a concentrated CH2Cl2 solution. Synthesis of Cl3Ga[P(i-Pr)2NP(H)(i-Pr)2] (3). Same as 2 using GaCl3 (0.35 g, 2.0 mmol). Yield = 0.81 g (95%), mp 85−86 °C. 1H NMR (300 MHz, C6D6) δ 6.82 (br dt, 1JPH = 448 Hz, 3JHH = 1.8 Hz, 1H, PH), 2.06 (d sept, 3JHH = 7.2 Hz, 2JPH = 7 Hz, 2H, CH), 1.56 (dd sept, 3 JHH = 7.2 Hz, 3JHH = 1.8 Hz, 2JPH = 7 Hz, 2H, CH), 1.16 (dd, 3JHH = 7.2 Hz, 3JPH = 17.5 Hz, 6H, CH3), 1.03 (dd, 3JHH = 7.2 Hz, 3JPH = 16 Hz, 6H, CH3), 0.74 (dd, 3JHH = 7.2 Hz, 3JPH = 19 Hz, 6H, CH3), 0.69 (dd, 3JHH = 7.2 Hz, 3JPH = 19 Hz, 6H, CH3) ppm. 13C{1H} NMR (75 MHz, C6D6) δ 27.1 (d, 1JPC = 40 Hz, CH), 25.5 (d, 1JPC = 67.8 Hz, CH), 17.2 (s, CH3), 16.5 (s, CH3), 16.1 (s, CH3), 15.5 (s, CH3) ppm. 31 1 P{ H} NMR (121 MHz, C6D6) δ 36.7 (d, 2JPP = 11 Hz, PH) ppm (Ga-bound P not observed). IR (nujol mull) ν 2349 (P−H) cm−1. Anal. Calcd for C12H29Cl3GaNP2: C, 33.88; H, 6.87; N, 3.29. Found: C, 34.28; H, 7.89; N, 3.21. X-ray quality single crystals were grown from a concentrated CH2Cl2 solution. Synthesis of (i-Pr)2PN(AlCl3)P(H)(i-Pr)2 (4). A solution of 1 (0.50 g, 2.0 mmol) in 1 mL of CH2Cl2 was added to a slurry of AlCl3 (0.27 g, 2.0 mmol) in 1 mL of CH2Cl2. After stirring for 30 min at rt, the colorless solution was placed in a −25 °C freezer. Colorless crystals of 4 were isolated after 24 h. Yield = 0.59 g (77%), mp 118−119 °C. 1H NMR (300 MHz, C6D6) δ 6.52 (ddt, 1JPH = 470 Hz, 3JPH = 24.6 Hz, 3 JHH = 4.2 Hz, 1H, PH), 2.70 (d sept, 2JPH = 7.2 Hz, 3JHH = 7.2 Hz, 2H, CH), 2.13 (dd sept, 2JPH = 14.5 Hz, 3JHH = 7.2 Hz, 3JHH = 4.2 Hz, 2H, CH), 1.24 (dd, 3JPH = 14 Hz, 3JHH = 7.2 Hz, CH3), 0.98 (dd, 3JPH = 17 Hz, 3JHH = 7.2 Hz, CH3), 0.88 (dd, 3JPH = 19.2 Hz, 3JHH = 7.2 Hz, CH3), 0.73 (dd, 3JPH = 18.3 Hz, 3JHH = 7.2 Hz, CH3) ppm. 13C{1H} NMR (75 MHz, C6D6) δ 29.4 (d, 1JPC = 21 Hz, CH), 26.4 (d, 1JPC = 56 Hz, CH), 21.5 (d, 2JPC = 14 Hz, CH3), 19.9 (d, 2JPC = 25 Hz, CH3), 18.4 (s, CH3), 16.7 (s, CH3) ppm. 31P{1H} NMR (121 MHz, C6D6) δ 86.6 (d, 2JPP = 54 Hz, P), 56.3 (d, 2JPP = 54 Hz, PH) ppm. 27Al NMR (78 MHz, C6D6) δ 52 (br) ppm. IR (nujol mull) ν 2403 (P−H) cm−1. Anal. Calcd for C12H29AlCl3NP2: C, 37.67; H, 7.64; N, 3.66. Found: C, 37.53; H, 7.65; N, 3.55. P-isomer of AlCl3 (not isolated): 31P{1H} NMR (121 MHz, C6D6) δ 35.5 (s, PH) ppm, (Al-bound P not observed). 27Al NMR (78 MHz, C6D6) δ 108 (d, 1JPAl = 316 Hz) ppm.

Figure 9. Structure of 10. Thermal ellipsoids are shown at 50% probability, and all C-H atoms have been omitted for clarity. Selected bond lengths (Å) and angles (deg): In1−O1 = 2.264(2), In1−O2 = 2.128(2), In1−O3 = 2.137(2), In1−O4 = 2.291(2), In2−O2 = 2.116(2), In2−O3 = 2.118(2), In2−O4 = 2.247(2), In2−O4′ = 2.270(2), P1−N1 = 1.583(3), P2−N1 = 1.580(3), P2···Cl1 = 2.76(3), P1−N1−P2 = 139.6(2).

comparable lengths and therefore likely comparable strength. In the cases of AlCl3 and BF3·Et2O, however, it is a positive sign that no 2:1 adducts were isolable. The first equivalent of these two Lewis acids binds to 1 through nitrogen, compressing the P−N−P angle and strongly suggesting that there will be FLPtype reactivity between the second LA equivalent and the unprotonated but sterically hindered pendant phosphorus site. The scope of this reactivity is now being explored.

■

CONCLUSIONS The reaction of AlCl3 or BF3·Et2O with HN[P(i-Pr)2]2 in a 1:1 ratio produced the first two crystallographically characterized examples of monodentate coordination at nitrogen for the class of ambidentate HN(PR2)2 ligands that have been in widespread use for several decades. For the softer Lewis acids InCl3, GaCl3, and BCl3, coordination occurs at phosphorus instead. In all cases, coordination induces tautomerization, shifting the acidic H from N to P. Each adduct maintains its structure in solution, as shown by the close correlation of the solution phase multinuclear NMR spectra with the structures identified by single-crystal X-ray diffraction. Because of this tight association maintained in solution, no further reactivity of these 1:1 adducts was noted with the exception of BF3·Et2O, which in one case was found to react with acetonitrile to form an unusual diazaphosphaborole. Increasing the metal:ligand ratio to 2:1 gives a bis-adduct for BCl3, a simple salt of the form [LMX2][MX4] for GaCl3 and a far more complex salt [(iPr2PH)2N][(MeO)6In4Cl8·2MeOH] for InCl3. For the latter, both the cation and anion are reported here for the first time. Neither AlCl3 nor BF3·Et2O form isolable 2:1 adducts under the conditions explored, leading us to speculate that they are the most promising for ongoing exploration of FLP-type activation of small molecules.

■

EXPERIMENTAL SECTION

General Experimental Considerations. All manipulations were carried out in an argon-filled glovebox or by using standard Schlenk techniques. Anhydrous solvents were stored in the glovebox over 4 Å molecular sieves prior to use. Prior to use, AlCl3 and GaCl3 were sublimed, and BF3·Et2O was distilled.51 All other reagents were used as 7297

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry

(thin film) ν 3217 (N−H) cm−1. Anal. Calcd for C12H29Cl6Ga2NP2: C, 23.96; H, 4.86; N, 2.33. Found: C, 23.94; H, 4.96; N, 1.90. Synthesis of Cl3BP(i-Pr)2N(H)P(i-Pr)2(BCl3) (9). BCl3 (4.0 mL, 1.0 M in hexanes) was added to a solution of 1 (0.5o g, 2.0 mmol) in 1.5 mL of CH2Cl2. The solution was allowed to stir at rt for 4 h and was then placed in a −25 °C freezer. Colorless crystals of 9 were isolated after 24 h. Yield = 0.74 g (77%), mp 94−96 °C. 1H NMR (300 MHz, C6D6) δ 3.48 (s, 1H, NH), 2.57 (d sept, 3JHH = 7.2 Hz, 2JPH = not resolved, 4H, CH), 1.17 (dd, 3JHH = 7.2 Hz, 3JPH = 15 Hz, 12H, CH3), 1.12 (dd, 3JHH = 7.2 Hz, 3JPH = 15 Hz, 12H, CH3) ppm. 13C{1H} NMR (75 MHz, C6D6) δ 27.4 (d, 1JCP = 32 Hz, CH), 18.0 (s, CH3), 17.6 (s, CH3) ppm. 31P{1H} NMR (121 MHz, C6D6) δ 51.6 (br q, 1JPB = 155 Hz) ppm. 11B NMR (96 MHz, C6D6) δ 3.6 (d, 1JPB = 155 Hz) ppm. IR (nujol mull) ν 3238 (N−H) cm−1. Anal. Calcd for C12H29B2Cl6NP2: C, 29.80; H, 6.04; N, 2.90. Found: C, 29.94; H, 5.94; N, 2.59. Synthesis of [(i-Pr2PH)2N][(MeO)6In4Cl8·2MeOH] (10). Solid 1 (0.25 g, 1.0 mmol) was added to a solution of InCl3 (0.44 g, 2.0 mmol) in 2 mL of MeOH. The colorless solution was allowed to stir at rt for 5 min and was then placed in a −25 °C freezer. After 18 h, the supernatant solution was decanted from colorless crystals of 10. Yield = 0.26 g (35%), mp 95−97 °C. 1H NMR (300 MHz, CDCl3) δ 6.76 (dt, 1JPP = 463 Hz, 3JHH = 2.7 Hz, 2H, PH), 3.78 (br s, 3H, OCH3), 3.75 (br s, 3H, OCH3), 3.72 (br s, 3H, OCH3), 3.63 (br s, 3H, HOCH3), 2.35 (dd sept, 2JPH = 7.2 Hz, 3JHH = 2.7 Hz, 3JHH = 7.2 Hz, 4H, CH), 1.33 (dd, 3JPH = 19.5 Hz, 3JHH = 7.2 Hz, 12H, CH3), 1.29 (dd, 3JPH = 18.3 Hz, 3JHH = 7.2 Hz, 12H, CH3) ppm. 13C{1H} NMR (75 MHz, CDCl3) δ 53.5 (br s, OCH3), 53.2 (br s, OCH3), 53.0 (br s, OCH3), 51.3 (br s, HOCH3), 25.1 (vm, CH), 16.7 (s, CH3), 15.5 (CH3) ppm. 31P{1H} NMR (121 MHz, CDCl3) δ 38.7 (s) ppm. IR (thin film) ν 3260 (O−H), 2348 (P−H) cm−1. Anal. Calcd for C32H86Cl8In4N2O8P4 (10 − 2 MeOH): C, 25.17; H, 5.63; N, 1.96. Found: C, 24.87; H, 5.19; N, 1.55.

Synthesis of Cl3B[P(i-Pr)2NP(H)(i-Pr)2] (5). BCl3 (2.0 mL, 1.0 M in hexanes) was added to a solution of 1 (0.5o g, 2.0 mmol) in 2 mL of Et2O. The solution was allowed to stir at rt for 1.5 h, and then the solvent was removed under a vacuum to give 5 as a white powder. Yield = 0.65 g (89%), mp 63−64 °C. 1H NMR (300 MHz, C6D6) δ 6.86 (d, 1JPH = 451 Hz, 1H, PH), 2.42 (d sept, 3JHH = 7.2 Hz, 2JPH = 7.0 Hz, 2H, CH), 1.64 (d sept, 3JHH = 7.2 Hz, 2JPH = 7.0 Hz, 2H, CH), 1.26 (dd, 3JHH = 7.2 Hz, 3JPH = 14.4 Hz, 6H, CH3), 1.17 (dd, 3JHH = 7.2 Hz, 3JPH = 14.4 Hz, 6H, CH3), 0.80 (dd, 3JHH = 7.2 Hz, 3JPH = 16.6 Hz, 6H, CH3), (dd, 3JHH = 7.2 Hz, 3JPH = 16.6 Hz, 6H, CH3) ppm. 13 C{1H} NMR (75 MHz, C6D6) δ 26.2 (d, 1JPC = 50.1 Hz, CH), 25.5 (d, 1JPC = 69.2 Hz, CH), 17.9 (s, CH3), 17.5 (s, CH3), 16.6 (s, CH3), 15.6 (s, CH3) ppm. 31P{1H} NMR (121 MHz, C6D6) δ 33.9 (br qd, 1 JBP = 173 Hz, 2JPP = 13 Hz, PB), 32.7 (d, 2JPP = 13 Hz, PH) ppm. 11B NMR (96 MHz, C6D6) δ 6.59 (d, 1JPB = 173 Hz) ppm. IR (nujol mull) ν 2382 cm−1. Anal. Calcd for C12H29BCl3NP2: C, 39.33; H, 7.98; N, 3.82. Found: C, 39.46; H, 8.81; N, 3.80. X-ray quality single crystals were grown from a concentrated Et2O solution. Synthesis of (i-Pr)2PN(BF3)P(H)(i-Pr)2 (6). Same as for 2 beginning with BF3·Et2O (0.28 g, 2.0 mmol). Yield = 0.52 g (82%), mp 59−60 °C. 1H NMR (300 MHz, C6D6) δ 6.13 (br d, 1JPH = 465 ppm, 1H, PH), 2.65 (d sept, 3JHH = 7.2 Hz, 2JPH = 7.2 Hz, 2H, CH), 2.17 (br, 2H, CH), 1.35 (dd, 3JHH = 7.2 Hz, 3JPH = 14 Hz, 6H, CH3), 1.14 (dd, 3 JHH = 7.2 Hz, 3JPH = 14 Hz, 6H, CH3), 1.00 (dd, 3JHH = 7.2 Hz, 3JPH = 16.5 Hz, 6H, CH3), 0.73 (dd, 3JHH = 7.2 Hz, 3JPH = 17.5 Hz, 6H, CH3) ppm. 13C{1H} NMR (75 MHz, C6D6) δ 28.2 (dd, 1JPC = 17 Hz, 3JPC = 1.5 Hz, CH), 25.8 (d, 1JPC = 52 Hz, CH), 21.3 (d, 2JPC = 16 Hz, CH3), 19.3 (d, 2JPC = 20 Hz, CH3), 17.7 (s, CH3), 17.0 (s, CH3) ppm. 31 1 P{ H} NMR (121 MHz, C6D6) δ 101 (d, 2JPP = 110 Hz, P), 76 (d, 2 JPP = 110 Hz, PH) ppm. 11B{1H} NMR (96 MHz, C6D6) 3.0 (q, 1JBF = 22 Hz) ppm. 19F{1H} NMR (282 MHz, C6D6) −133 (q, 1JBF = 22 Hz) ppm. IR (nujol mull) ν 2348 (P−H) cm−1. Anal. Calcd for C12H29BF3NP2: C, 45.45; H, 9.22; N, 4.42. Found: C, 44.85; H, 9.46; N, 4.27. X-ray quality single crystals were grown from a concentrated CH2Cl2 solution. P-isomer of BF3 (not isolated): 31P{1H} NMR (121 MHz, C6D6) δ 37 (s, PH) ppm, (B-bound P not observed). 11B{1H} NMR (96 MHz, C6D6) δ 4.9 (q, 1JBF = 61 Hz) ppm. 19F{1H} NMR (282 MHz, C6D6) δ −128 (dq, 1JBF = 61 Hz, 2JPF = 208 Hz) ppm. Synthesis of 1,3-Bis(diisopropylphosphino)-2,2-difluoro-4,4-diisopropyl-5-methylene-1,3,4,2-diazaphosphaborole (7). Solid 1 (0.25 g, 1.0 mmol) was added to a solution of BF3·Et2O (0.14 g, 1.0 mmol) in 1 mL of acetonitrile. The solution was allowed to stir at rt for 2 h, and then volatiles were removed under a vacuum. The residue was redissolved in a minimal amount of acetonitrile and placed in a −25 °C freezer. After 24 h, the supernatant was decanted from crystalline 7. Yield = 0.06 g. Note! Due to irreproducibility and low yield, characterization data is incomplete. 1H NMR (300 MHz, C6D6) δ 5.43 (br d, 3JPH = 38 Hz, 1H, CH2), 3.75 (br d, 3JPH = 14 Hz, 1H, CH2), 2.72 (d sept, 3JHH = 7 Hz, 2JPH not resolved, 2H, CH), 2.62 (d sept, 3JHH = 7 Hz, 2JPH not resolved, 2H, CH), 1.99 (d sept, 3JHH = 7 Hz, 2JPH not resolved, 2H, CH), 1.41 (dd, 3JHH = 7 Hz, 3JPH = 15 Hz, 6H, CH3), 1.29 (dd, 3JHH = 7 Hz, 3JPH = 15 Hz, 6H, CH3), 1.23 (dd, 3 JHH = 7 Hz, 3JPH = 15 Hz, 6H, CH3), 1.21 (dd, 3JHH = 7 Hz, 3JPH = 15 Hz, 6H, CH3), 1.10 (dd, 3JHH = 7 Hz, 3JPH = 17 Hz, 6H, CH3), 0.90 (dd, 3JHH = 7 Hz, 3JPH = 17 Hz, 6H, CH3) ppm. 11B{1H} NMR (96 MHz, C6D6) δ 7.3 (br t, 1JBF = 40 Hz) ppm. 19F{1H} NMR (282 MHz, C6D6) δ −131 (br s) ppm. Synthesis of [{HN[P(i-Pr)2]2}GaCl2][GaCl4] (8). A solution of 1 (0.25 g, 1.0 mmol) in 0.5 mL of CH2Cl2 was added to a solution of GaCl3 (0.35 g, 2.0 mmol) in 0.5 mL of CH2Cl2. The colorless solution was stirred at rt for 5 min and then placed in a −25 °C freezer. After 24 h, the supernatant was decanted from colorless crystals of 8. Yield = 0.52 g (87%), mp 110 °C (dec). 1H{31P} NMR (300 MHz, CD2Cl2) δ 5.16 (s, 1H, NH), 2.79 (sept, 3JHH = 7.2 Hz, 4H, CH), 1.48 (d, 3JHH = 7.2 Hz, 12H, CH3), 1.47 (d, 3JHH = 7.2 Hz, 12H, CH3) ppm. 13C{1H} NMR (75 MHz, CD2Cl2) δ 28.6 (vt, CH), 17.8 (s, CH3), 17.6 (s, CH3) ppm. 31P{1H} NMR (121 MHz, CD2Cl2) δ 113 (s) ppm. IR

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01051. CIF files for 2−10 and crystallographic data for compounds 2−6, 7, and 2:1 complexes of 8−10 (PDF) Accession Codes

CCDC 1543984−1543992 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Diane A. Dickie: 0000-0003-0939-3309 Richard A. Kemp: 0000-0002-2063-3812 Present Address ‡

D.A.D.: Department of Chemistry, Brandeis University, Waltham, MA 02453, United States

Notes

The authors declare no competing financial interest. 7298

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry

■

(16) Ménard, G.; Stephan, D. W. Room Temperature Reduction of CO2 to Methanol by Al-Based Frustrated Lewis Pairs and Ammonia Borane. J. Am. Chem. Soc. 2010, 132, 1796−1797. (17) Groom, C. R.; Allen, F. H. The Cambridge Structural Database in Retrospect and Prospect. Angew. Chem., Int. Ed. 2014, 53, 662−671. (18) Dickie, D. A.; Gislason, K. B.; Kemp, R. A. Formation of Phosphino-Substituted Isocyanate by Reaction of CO2 with Group 2 Complexes Based on the (Me3Si)(i-Pr2P)NH Ligand. Inorg. Chem. 2012, 51, 1162−1169. (19) Dickie, D. A.; Parkes, M. V.; Kemp, R. A. Insertion of Carbon Dioxide into Main-Group Complexes: Formation of the [N(CO2)3]3‑ Ligand. Angew. Chem., Int. Ed. 2008, 47, 9955−9957. (20) Miller, C. J.; Chadha, U.; Ulibarri-Sanchez, J. R.; Dickie, D. A.; Kemp, R. A. Structure and Lewis-base reactivity of bicyclic low-valent germanium and tin complexes bridged by bis(diisopropylphosphino)amine. Polyhedron 2016, 114, 351−359. (21) Dickie, D. A.; Kemp, R. A. Structures and CO2 Reactivity of Zinc Complexes of Bis(diisopropyl-) and Bis(diphenylphosphino)amines. Organometallics 2014, 33, 6511−6518. (22) Dickie, D. A.; Ulibarri-Sanchez, R. P., III; Jarman, P. J.; Kemp, R. A. Activation of CO2 and CS2 by (Me3Si)(i-Pr2P)NH and its zinc complex. Polyhedron 2013, 58, 92−98. (23) Dickie, D. A.; Chacon, B. E.; Issabekov, A.; Lam, K.; Kemp, R. A. Nickel(II) and nickel(0) complexes of bis(diisopropylphosphino)amine: Synthesis, structure and electrochemical activity. Inorg. Chim. Acta 2016, 453, 42−50. (24) Dickie, D. A.; Barker, M. T.; Land, M. A.; Hughes, K. E.; Clyburne, J. A. C.; Kemp, R. A. Rapid, Reversible, Solid-Gas and Solution Phase Insertion of CO2 into In-P Bonds. Inorg. Chem. 2015, 54, 11121−11126. (25) Barry, B. M.; Dickie, D. A.; Murphy, L. J.; Clyburne, J.; Kemp, R. A. NH/PH Isomerization and a Lewis Pair for Carbon Dioxide Capture. Inorg. Chem. 2013, 52, 8312−8314. (26) Caminade, A.-M.; Ocando, E.; Majoral, J.-P.; Cristante, M.; Bertrand, G. First Example of Prototropism in Iminobis(phosphines) Induced by Phosphorus Alkylation. Inorg. Chem. 1986, 25, 712−714. (27) Dickie, D. A.; Ulibarri-Sanchez, R. P.; Kemp, R. A. Zwitterionic CS2 Adducts of Bis(dialkylphosphino)amines: Syntheses, Spectroscopy, and Structures. Aust. J. Chem. 2015, 68, 351−356. (28) Robertson, S. D.; Chivers, T. Synthesis, NMR characterization and X-ray structures of mixed chalcogenido PNP ligands containing tellurium: crystal structures of SeiPr2PNP(H)iPr2 and [NaN(EPiPr2)2]∞ (E = Se, Te). Dalton Trans. 2008, 1765−1772. (29) Chivers, T.; Eisler, D. J.; Ritch, J. S.; Tuononen, H. M. An Unusual Ditelluride: Synthesis and Molecular and Electronic Structures of the Dimer of the Tellurium-Centered Radical [TePiPr2NiPr2PTe]•. Angew. Chem., Int. Ed. 2005, 44, 4953−4956. (30) Elder, P. J. W.; Chivers, T.; Thirumoorthi, R. Experimental and Computational Investigations of Tautomerism and Fluxionality in PCP- and PNP-Bridged Heavy Chalcogenides. Eur. J. Inorg. Chem. 2013, 2013, 2867−2876. (31) Kortunov, P. V.; Baugh, L. S.; Siskin, M.; Calabro, D. C. In Situ Nuclear Magnetic Resonance Mechanistic Studies of Carbon Dioxide Reactions with Liquid Amines in Mixed Base Systems: The Interplay of Lewis and Brønsted Basicities. Energy Fuels 2015, 29, 5967−5989. (32) Murphy, L. J.; Robertson, K. N.; Kemp, R. A.; Tuononen, H. M.; Clyburne, J. A. C. Structurally simple complexes of CO2. Chem. Commun. 2015, 51, 3942−3956. (33) Buβ, F.; Mehlmann, P.; Mück-Lichtenfeld, C.; Bergander, K.; Dielmann, F. Reversible Carbon Dioxide Binding by Simple Lewis Base Adducts with Electron-Rich Phosphines. J. Am. Chem. Soc. 2016, 138, 1840−1843. (34) Sava, X.; Melaimi, M.; Mézailles, N.; Ricard, L.; Mathey, F.; Le Floch, P. Cationic diphosphaferrocene gallium dichloride complexes. New J. Chem. 2002, 26, 1378−1383. (35) Cheng, F.; Hector, A. L.; Levason, W.; Reid, G.; Webster, M.; Zhang, W. Preparation, Characterization, and Structural Systematics of Diphosphane and Diarsane Complexes of Gallium(III) Halides. Inorg. Chem. 2007, 46, 7215−7223.

ACKNOWLEDGMENTS This work was financially supported by the National Science Foundation (Grant CHE12-13529). The Bruker X-ray diffractometer was purchased via a National Science Foundation CRIF:MU award to the University of New Mexico (CHE04-43580), and the NMR spectrometers were upgraded via grants from the NSF (CHE08-40523 and CHE09-46690). Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under Contract No. DE-AC0494AL85000.

■

REFERENCES

(1) Chapman, A. M.; Haddow, M. F.; Wass, D. F. Frustrated Lewis Pairs beyond the Main Group: Synthesis, Reactivity, and Small Molecule Activation with Cationic Zirconocene-Phosphinoaryloxide Complexes. J. Am. Chem. Soc. 2011, 133, 18463−18478. (2) Metters, O. J.; Forrest, S. J. K.; Sparkes, H. A.; Manners, I.; Wass, D. F. Small Molecule Activation by Intermolecular Zr(IV)-Phosphine Frustrated Lewis Pairs. J. Am. Chem. Soc. 2016, 138, 1994−2003. (3) Normand, A. T.; Richard, P.; Balan, C.; Daniliuc, C. G.; Kehr, G.; Erker, G.; Le Gendre, P. Synthetic Endeavors toward Titanium Based Frustrated Lewis Pairs with Controlled Electronic and Steric Properties. Organometallics 2015, 34, 2000−2011. (4) Barnett, B. R.; Moore, C. E.; Rheingold, A. L.; Figueroa, J. S. Cooperative Transition Metal/Lewis Acid Bond-Activation Reactions by a Bidentate (Boryl)iminomethane Complex: A Significant MetalBorane Interaction Promoted by a Small Bite-Angle LZ Chelate. J. Am. Chem. Soc. 2014, 136, 10262−10265. (5) Stephan, D. W. Frustrated Lewis Pairs: From Concept to Catalysis. Acc. Chem. Res. 2015, 48, 306−316. (6) Stephan, D. W.; Erker, G. Frustrated Lewis Pair Chemistry: Development and Perspectives. Angew. Chem., Int. Ed. 2015, 54, 6400−6441. (7) Cardenas, A. J. P.; Hasegawa, Y.; Kehr, G.; Warren, T. H.; Erker, G. Cooperative 1,1-addition reactions of vicinal phosphane/borane frustrated Lewis pairs. Coord. Chem. Rev. 2016, 306, 468−482. (8) Boudreau, J.; Courtemanche, M.-A.; Fontaine, F.-G. Reactivity of Lewis pairs (R2PCH2AlMe2)2 with carbon dioxide. Chem. Commun. 2011, 47, 11131−11133. (9) Courtemanche, M.-A.; Larouche, J.; Légaré, M.-A.; Bi, W.; Maron, L.; Fontaine, F.-G. A Tris(triphenylphosphine)aluminum Ambiphilic Precatalyst for the Reduction of Carbon Dioxide with Catecholborane. Organometallics 2013, 32, 6804−6811. (10) Bertini, F.; Hoffmann, F.; Appelt, C.; Uhl, W.; Ehlers, A. W.; Slootweg, J. C.; Lammertsma, K. Reactivity of Dimeric P/Al-Based Lewis Pairs toward Carbon Dioxide and tert-Butyl Isocyanate. Organometallics 2013, 32, 6764−6769. (11) Roters, S.; Appelt, C.; Westenberg, H.; Hepp, A.; Slootweg, J. C.; Lammertsma, K.; Uhl, W. Dimeric aluminum-phosphorus compounds as masked frustrated Lewis pairs for small molecule activation. Dalton Trans. 2012, 41, 9033−9045. (12) Xu, X.; Kehr, G.; Daniliuc, C. G.; Erker, G. Frustrated Lewis Pair Behavior of [Cp2ZrOCR2CH2PPh2]+ Cations. Organometallics 2015, 34, 2655−2661. (13) Wang, X.; Kehr, G.; Daniliuc, C. G.; Erker, G. Internal Adduct Formation of Active Intramolecular C4-bridged Frustrated Phosphane/ Borane Lewis Pairs. J. Am. Chem. Soc. 2014, 136, 3293−3303. (14) Holtrichter-Rößmann, T.; Rösener, C.; Hellmann, J.; Uhl, W.; Würthwein, E.-U.; Fröhlich, R.; Wibbeling, B. Generation of Weakly Bound Al−N Lewis Pairs by Hydroalumination of Ynamines and the Activation of Small Molecules: Phenylethyne and Dicyclohexylcarbodiimide. Organometallics 2012, 31, 3272−3283. (15) Dickie, D. A.; Coker, E. N.; Kemp, R. A. Formation of a Reversible, Intramolecular Main-Group Metal-CO2 Adduct. Inorg. Chem. 2011, 50, 11288−11290. 7299

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300

Article

Inorganic Chemistry (36) Sigl, M.; Schier, A.; Schmidbaur, H. Bis[(2-diphenylphosphino)phenyl]phenylphosphine as an Inflexible Tridentate Ligand for Indium Trichloride. Z. Naturforsch., B: J. Chem. Sci. 1999, 54, 1417−1419. (37) Cheng, F.; Friend, S. I.; Hector, A. L.; Levason, W.; Reid, G.; Webster, M.; Zhang, W. Preparation, Characterization, and Structural Systematics of Diphosphane and Diarsane Complexes of Indium(III) Halides. Inorg. Chem. 2008, 47, 9691−9700. (38) Yurkerwich, K.; Parkin, G. Gallium and indium compounds supported by a [PNP] pincer ligand. Inorg. Chim. Acta 2010, 364, 157−161. (39) Sigl, M.; Schier, A.; Schmidbaur, H. Contributions to the Coordination and Structural Chemistry of Gallium(III) and Indium(III) Halides: Complexes with Bi- and Tridentate Tertiary Phosphanes. Eur. J. Inorg. Chem. 1998, 1998, 203−210. (40) Gans-Eichler, T.; Jones, C.; Aldridge, S.; Stasch, A. Crystal Structure of 1,4-Bis(triiodogallium(III))-1,4-bis(2,4,6-tri-tert-butylphenyl)-1,4-diphosphabuta-1,3-diene. Anal. Sci.: X-Ray Struct. Anal. Online 2008, 24, x109−x110. (41) Sigl, M.; Schier, A.; Schmidbaur, H. Indium Triiodide Complexes of Bis(diphenylphosphino)ethane (dppe) and its Disulfide (dppeS2). Z. Naturforsch., B: J. Chem. Sci. 1999, 54, 21−25. (42) Weigand, J. J.; Burford, N.; Decken, A. The Binary Ph2PCl/ GaCl3 System: A Room-Temperature Molten Medium for P-P Bond Formation. Eur. J. Inorg. Chem. 2008, 2008, 4343−4347. (43) Brown, M. A.; Castro, J. A.; Tuck, D. G. Spectroscopic and crystallographic studies of phosphino adducts of gallium(III) iodide. Can. J. Chem. 1997, 75, 333−341. (44) Sigl, M.; Schier, A.; Schmidbaur, H. Lewis Acid Catalyzed Z to E Isomerization of 1,2-Bis(diphenylphosphino)ethene. Z. Naturforsch., B: J. Chem. Sci. 1998, 53, 1301−1306. (45) Burt, J.; Levason, W.; Light, M. E.; Reid, G. Phosphine complexes of aluminium(III) halides - preparation and structural and spectroscopic systematics. Dalton Trans. 2014, 43, 14600−14611. (46) Degnan, I. A.; Alcock, N. W.; Roe, S. M.; Wallbridge, M. G. H. Structures of Adducts formed by Indium(III) Halides and Phosphine Ligands: Tris(1,2-bis(diphenylphosphino)ethane]bis[triiodoindium(III)] and Trichlorobis(trimethylphosphine)indium(III). Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1992, 48, 995−999. (47) Cheng, F.; Codgbrook, H. L.; Hector, A. L.; Levason, W.; Reid, G.; Webster, M.; Zhang, W. Gallium(III) halide complexes with phosphines, arsines and phosphine oxides − a comparative study. Polyhedron 2007, 26, 4147−4155. (48) Cheng, F.; Hector, A. L.; Levason, W.; Reid, G.; Webster, M.; Zhang, W. μ-1,2-Bis(diphenylphosphino)ethane-κ 2 P:P-bis[trichloridogallium(III)]. Acta Crystallogr., Sect. E: Struct. Rep. Online 2007, 63, m1761−m1761. (49) Charmant, J. P. H.; Fan, C.; Norman, N. C.; Pringle, P. G. Synthesis and reactivity of dichloroboryl complexes of platinum(II). Dalton Trans. 2007, 114−123. (50) Cristóbal-Lecina, E.; Etayo, P.; Doran, S.; Revés, M.; MartínGago, P.; Grabulosa, A.; Costantino, A. R.; Vidal-Ferran, A.; Riera, A.; Verdaguer, X. MaxPHOS Ligand: PH/NH Tautomerism and Rhodium-Catalyzed Asymmetric Hydrogenations. Adv. Synth. Catal. 2014, 356, 795−804. (51) Armarego, W. L. F.; Chai, C. L. L. Purification of Laboratory Chemicals, 5th ed.; Elsevier: New York, NY, 2003. (52) Ritch, J. S.; Chivers, T.; Ahmad, K.; Afzaal, M.; O’Brien, P. Synthesis, Structures, and Multinuclear NMR Spectra of Tin(II) and Lead(II) Complexes of Tellurium-Containing Imidodiphosphinate Ligands: Preparation of Two Morphologies of Phase-Pure PbTe from a Single-Source Precursor. Inorg. Chem. 2010, 49, 1198−1205.

7300

DOI: 10.1021/acs.inorgchem.7b01051 Inorg. Chem. 2017, 56, 7292−7300