Transformation of Metal–Organic Framework Secondary Building

Aug 11, 2017 - Bonnie Charpentier is 2018 ACS president-elect. Bonnie Charpentier, senior vice president for regulatory, quality, and safety at Cytoki...
3 downloads 8 Views 1MB Size
Communication pubs.acs.org/JACS

Transformation of Metal−Organic Framework Secondary Building Units into Hexanuclear Zr-Alkyl Catalysts for Ethylene Polymerization Pengfei Ji,† Joseph B. Solomon,† Zekai Lin, Alison Johnson, Richard F. Jordan, and Wenbin Lin* Department of Chemistry, The University of Chicago, 929 E 57th Street, Chicago, Illinois 60637, United States S Supporting Information *

ABSTRACT: We report the stepwise and quantitative transformation of the Zr6(μ3-O)4(μ3-OH)4(HCO2)6 nodes in Zr-BTC (MOF-808) to the [Zr 6 (μ 3 -O) 4 (μ 3 OH)4Cl12]6− nodes in ZrCl2-BTC, and then to the organometallic [Zr6(μ3-O)4(μ3-OLi)4R12]6− nodes in ZrR2-BTC (R = CH2SiMe3 or Me). Activation of ZrCl2BTC with MMAO-12 generates ZrMe-BTC, which is an efficient catalyst for ethylene polymerization. ZrMe-BTC displays unusual electronic and steric properties compared to homogeneous Zr catalysts, possesses multimetallic active sites, and produces high-molecular-weight linear polyethylene. Metal−organic framework nodes can thus be directly transformed into novel single-site solid organometallic catalysts without homogeneous analogs for polymerization reactions. Figure 1. Zr coordination environments in (a) homogeneous mononuclear Zr olefin polymerization catalysts, (b) homogeneous dinuclear Zr catalysts, and (c) mononuclear vs hexanuclear view of the ZrMe-BTC catalyst.

G

roup IV metal complexes form an important class of single-site catalysts with broad applications in olefin polymerization,1,2 hydrogenation,3−6 hydroboration,7 and hydroamination.8−11 For many reactions, group IV metal catalysts are advantageous over their precious metal counterparts because of their unique reactivity, high natural abundance, and low toxicity. For example, several classic Zr-based catalyst systems have been developed over the past few decades, including zirconocenes,2 ansa-zirconocenes,12 and constrained geometry catalysts (CGCs),12 which all exhibit extremely high activities for the polymerization of olefins such as ethylene and propene (Figure 1a). All of these catalysts are supported by electron-rich cyclopentadienyl (Cp) and substituted Cp ligands. A large variety of other ancillary ligands, such as amido, amidinate, aryloxide, and imine ligands, have also been used to construct homogeneous group IV metal olefin polymerization catalysts.13,14 Here we report a simple strategy of directly transforming the Zr6 secondary building units (SBUs) in a readily available metal−organic framework (MOF) into alloxygen ligand supported hexanuclear Zr-alkyl complexes for olefin polymerization. MOFs constructed from metal cluster SBUs and organic linkers have provided a unique platform for developing singlesite solid catalysts,15−21 and in rare cases, multifunctional catalysts for cooperative and/or tandem catalysis. 22−25 Recently, electrophilic Zr-benzyl species were supported on the SBUs of Hf-NU-1000 to afford active catalysts for olefin polymerization.26 MOF-based olefin polymerization catalysts have also been generated by partial substitution of Zn centers in the Zn5Cl4 SBUs of MFU-4l with Ti and Cr centers followed by © 2017 American Chemical Society

MAO activation.27 To our knowledge, however, the multimetallic nature of metal cluster SBUs in MOFs has not been utilized to effect novel reactivities. In olefin polymerization, there has been significant interest in developing multinuclear catalysts to harness the cooperative effects of multiple metal centers to influence catalyst performance and polymer molecular weight (MW) and microstructure.28,29 For example, dinuclear zirconocene catalysts30,31 and CGC catalysts generate polyolefins with higher MW and branching content than their mononuclear analogs, which was ascribed to agostic interactions and bimetallic chain transfers (Figure 1b).32−34 We hypothesize that MOFs can provide a novel strategy to access multinuclear Zr catalysts with electronic and steric properties that cannot be achieved with homogeneous catalysts. We illustrate this approach by transforming the Zr6(μ3-O)4(μ3OH)4(HCO2)6 nodes in Zr-BTC (MOF-808) to the organometallic Zr 6 (μ 3 -O) 4 (μ 3 -OLi) 4 R 12 Li 6 (ZrR 2 -BTC, R = CH2SiMe3 or Me) and Zr6(μ3-O)4(μ3-OLi)4Me6 (ZrMeBTC) nodes within the MOFs and demonstrating that ZrMeBTC catalyzes the polymerization of ethylene to high-MW linear polyethylene (PE). Zr-BTC was synthesized in 70% yield using a literature procedure (Figure 2a).35 1H NMR analysis of digested Zr-BTC Received: June 12, 2017 Published: August 11, 2017 11325

DOI: 10.1021/jacs.7b05761 J. Am. Chem. Soc. 2017, 139, 11325−11328

Communication

Journal of the American Chemical Society

formate species followed by the formation of a Zr(OH2)2+ intermediate that is deprotonated to form the Zr(OH)(OH2) species in ZrOH-BTC (Figure S2, SI). The terminal ZrOH/OH2 bonds in ZrOH-BTC were cleaved with 10 equiv of Me3SiCl at 23 °C to generate ZrCl bonds in ZrCl2-BTC with the [Zr6(μ3-O)4(μ3-OH)4Cl12]H6 SBU along with 2 equiv of (SiMe3)2O. The identity of (SiMe3)2O was confirmed by 29Si and 1H NMR spectroscopy (Figure S3, SI). The amount of (SiMe3)2O was determined by 1 H NMR to be 1.97 ± 0.22 equiv. w.r.t. Zr. Me3SiCl was previously used to activate UO and CdOR bonds to form UCl and CdCl species in molecular systems.38,39 We propose that the ZrOH species is first silylated to generate Zr-OSiMe3, which further reacts with Me3SiCl to form a labile [Zr-O(SiMe3)2]Cl intermediate. Dissociation of (SiMe3)2O and coordination of Cl− form the ZrCl bond. The ZrOH2 bond can be analogously converted to the [ZrCl]H unit (Figure S5, SI). ZrCl2-BTC remains crystalline as indicated by PXRD studies (Figure 2c), and TEM images display clear lattice fringes with a distance of 1.72 nm, corresponding to the calculated d-spacing of 1.75 nm along the (200) planes (Figure 2d). N2 sorption isotherms of ZrCl2-BTC at 77K gave a Brunauer−Emmett−Teller (BET) surface area of 1693 m2/g, slightly lower than those of Zr-BTC (1843 m2/g) and ZrOH-BTC (1779 m2/g), due to the increased SBU weight of ZrCl2-BTC (Figure S6, SI). Thermogravimetric analysis (TGA) showed a 48.7% of weight loss in the 140−800 °C range, corresponding to the decomposition of ZrCl2-BTC to (ZrO2)6 (calculated 47.7%). Extended X-ray adsorption fine structure (EXAFS) fitting of the Zr coordination environment validated the proposed [Zr6(μ3-O)4(μ3-OH)4Cl12]H6 SBU with a ZrCl distance of 2.50 Å (Figure 3c). The ZrCl bonds in ZrCl2-BTC can be alkylated with alkyllithium reagents. Treatment of ZrCl2-BTC with 10 equiv of LiCH 2SiMe 3 converted the ZrCl2 species into the Zr(CH2SiMe3)2 species in the Zr6(μ3-O)4(μ3-OLi)4(CH2SiMe3)12Li6 SBU, while releasing 1.69 ± 0.17 equiv of SiMe4 (calculated 1.67 equiv). The Li to Zr ratio was determined to be 1.68 ± 0.05 by inductively coupled plasma-mass spectrometry (ICP-MS), consistent with the calculated value of 1.67. ZrMe2-BTC was similarly generated by treating ZrCl2BTC with 10 equiv of LiCH3. The formation of ZrR2 (R = CH2SiMe3 or Me) species was confirmed by solid-state 13C-cross-polarization magic angle spinning (13C-CPMAS) NMR spectroscopy. The methylene group of Zr(CH2SiMe3)2-BTC appears at δ 73.8, which is slightly more downfield than the corresponding resonances for homogeneous Zr-alkyl analogues,4 consistent with the electrondeficient nature of Zr centers in Zr(CH2SiMe3)2-BTC. The SiMe3 group appears as a sharp peak at 25.1 ppm. The methyl groups of ZrMe2-BTC appear as a broad peak at δ 71.1. In comparison, ZrCl2-BTC showed only 13C signals from the BTC ligand, with the carboxylate peak at δ 171.4 and arene peaks at δ 134 ppm. EXAFS fitting of Zr coordination environments supported the Zr-[CH2SiMe3)]2 structural model, with a Zr C bonding distance of 2.41 Å, close to those reported in the literature for molecular analogues.40 ZrR2-BTC is inactive for olefin polymerization, presumably due to the coordinative saturation at the anionic ZrMe2 centers. By analogy to homogeneous catalysts, in which LnMCl2 species are converted to active LnMR+ monoalkyl species by reaction with MAO, we hypothesized that ZrCl2-BTC could be activated by conversion to ZrMe-BTC, which contains neutral Zr-

Figure 2. (a) Synthesis of ZrCl2-BTC via formate removal from ZrBTC followed by deoxygenation with Me3SiCl. (b) Structural model of ZrCl2-BTC showing large open channels of 2.2 nm in the largest dimension (left) and chemical structure of octahedral Zr6Cl12 SBUs (right). (c) PXRD patterns of Zr-BTC (black), ZrOH-BTC (red), ZrCl2-BTC (blue), and Zr(CH2SiMe3)2-BTC (pink) indicate the retention of crystallinity through SBU transformations. (d) TEM image of ZrCl2-BTC shows the particle size of ∼200 nm. The measured distances between lattice fringes match well with the calculated d-spacings based on the crystal structure.

indicated the presence of six formic acids per Zr6 node. The capping formate groups were removed using a modified literature procedure to afford ZrOH-BTC,36,37 by stirring in 1 M aqueous solution of HCl at 90 °C for 12 h, which was confirmed by 1H NMR analysis of the digested ZrOH-BTC. The MOF crystallinity was maintained as indicated by the similarity between the PXRD pattern of Zr-BTC and that of ZrOH-BTC. The inorganic node in ZrOH-BTC is formulated as Zr6(μ3-O)4(μ3-OH)4(OH)6(OH2)6 by charge balance. Formate removal likely occurs via protonation of the Zr11326

DOI: 10.1021/jacs.7b05761 J. Am. Chem. Soc. 2017, 139, 11325−11328

Communication

Journal of the American Chemical Society

Table 1. ZrMe-BTC Catalyzed Ethylene Polymerization at Different Temperaturesa

Entry T (°C) 1 2 3 4b 5b 6b,c 7b,c

20 40 60 80 100 120 140

Yield (g)

Activity (kg × mol−1 × h−1)

Mn (×103)

Mw/Mn

Tm

0.289 0.499 1.781 1.190 1.431 1.386 1.921

2.41 4.16 14.84 19.83 23.85 23.10 32.02

158 189 191 114 55 109 47

2.69 2.62 3.55 2.44 4.91 7.79 5.78

137 138 134 138 134 132 131

a

Fisher-Porter tube, 30 mL toluene, 0.06 mmol Zr, 12.0 mmol MMAO-12 (200 equiv), 100 psi C2H4, 2 h. b0.03 mmol Zr, 6.0 mmol MMAO-12 (200 equiv) was used. c30 mL mesitylene.

Table 2. ZrMe-BTC Catalyzed Ethylene Polymerization under Different Pressuresa Entry P (psi) 1 2 3 4 5

Figure 3. (a) Alkylation of ZrCl2-BTC to form Zr(CH2SiMe3)2-BTC. (b) 13C-CPMAS NMR spectra of Zr(CH2SiMe3)2-BTC (red) and ZrCl2-BTC (black). (c, d) EXAFS fitting of ZrCl2-BTC (c) and Zr(CH2SiMe3)2-BTC.

100 200 400 600 800

Yield (g)

Activity (kg × mol−1 × h−1)

Mn (×103)

Mw/Mn

Tm

0.219 0.379 0.917 1.841 6.586

21.9 37.9 91.7 184.1 658.6

100 113 306 345 119

3.03 2.72 3.70 2.30 1.82

138 135 135 137 135

Conditions: stainless steel Parr autoclave, 30 mL toluene, 100 °C, 0.005 mmol Zr, 1.00 mmol MMAO-12 (200 equiv), 2h.

a

monoalkyl centers with open coordination sites within the Zr6(μ3-O)4(μ3-OLi)4Me6 nodes.41 Indeed, the reaction of ZrCl2-BTC with MMAO-12 generates ZrMe-BTC. The formation of the Zr-Me species was supported by the observation of the Zr-Me 13C-CPMAS signal at δ 73.3 (Figure S11, SI) and satisfactory EXAFS fitting of the Zr coordination with a ZrC bond distance of 2.24 Å in addition to the ZrO bonds (Figure S12, SI). The in situ generated ZrMe-BTC is active for ethylene polymerization, producing high-MW linear PE. The activity of ZrMe-BTC increases with temperature (20−140 °C) and ethylene pressure (100−800 psi), reaching a maximum value of 658.6 kg PE·mol−1 Zr·h−1 at 100 °C and 800 psi.42 This level of activity compares favorably to those of other MOF catalysts for olefin polymerization.26,27 The effect of temperature on activity of the catalyst and MW of the polymer is in agreement with well-established trends (Table 1).43 The effect of pressure on the catalytic performance of ZrMe-BTC varies significantly from homogeneous Zr catalysts (Table 2). Activity typically increases linearly with respect to monomer pressure for a homogeneous Zr catalyst. However, we observe a 30-fold increase in activity with an 8-fold increase in ethylene pressure. This effect may arise from polymer-induced fracture of the MOF particles, leading to exposure of additional active sites.42 ZrMe-BTC has high thermal stability and a long lifetime. Even at 100 °C, no decrease in activity was observed in 12 h. PXRD of the polymer product shows that the crystallinity of the MOF is still maintained after polymerization and work up, further demonstrating the robustness of the MOF catalyst. Site isolation of the MOF framework may stabilize the active species as seen in other catalytic MOFs.44−46

PE samples produced by ZrMe-BTC at 20 and 40 °C have monomodal MW distributions with PDIs of ca. 2.6, consistent with single-site catalysis. Broader MW distributions are observed at higher polymerization temperatures, which we ascribe to mass transport phenomena. At elevated temperatures, the reactor becomes filled with solvent-swollen PE and diffusion of ethylene to the active sites is limited. Control experiments show that ZrOH-BTC and Zr-BTC are not active for the polymerization of ethylene following treatment with MMAO-12. To further demonstrate that the MOF catalyst is the only active species, ZrCl2-BTC was treated with activator in the absence of ethylene. The supernatant collected was inactive for ethylene polymerization. In summary, we have demonstrated that the Zr6(μ3-O)4(μ3OH)4(HCO2)6 nodes in Zr-BTC can be converted to the [Zr6(μ3-O)4(μ3-OH)4Cl12]6− nodes and then to the organometallic [Zr6(μ3-O)4(μ3-OLi)4(Me)n]6− (n = 6 or 12) nodes. The stabilization of multimetallic active sites in ZrMe-BTC affords a robust catalyst with a long lifetime for the production of high-molecular-weight polyethylene. Notably, ZrMe-BTC is the first MOF polymerization catalyst that uses the metals of the SBUs as the active species, providing a cost-effective way of designing solid catalysts for olefin polymerization.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b05761. 11327

DOI: 10.1021/jacs.7b05761 J. Am. Chem. Soc. 2017, 139, 11325−11328

Communication

Journal of the American Chemical Society



(16) Wu, C.-D.; Hu, A.; Zhang, L.; Lin, W. J. Am. Chem. Soc. 2005, 127, 8940−8941. (17) Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen, S. T.; Hupp, J. T. Chem. Soc. Rev. 2009, 38, 1450−1459. (18) Zhao, M.; Ou, S.; Wu, C.-D. Acc. Chem. Res. 2014, 47, 1199− 1207. (19) Johnson, J. A.; Zhang, X.; Reeson, T. C.; Chen, Y.-S.; Zhang, J. J. Am. Chem. Soc. 2014, 136, 15881−15884. (20) Lykourinou, V.; Chen, Y.; Wang, X.-S.; Meng, L.; Hoang, T.; Ming, L.-J.; Musselman, R. L.; Ma, S. J. Am. Chem. Soc. 2011, 133, 10382−10385. (21) Ingleson, M. J.; Barrio, J. P.; Bacsa, J.; Dickinson, C.; Park, H.; Rosseinsky, M. J. Chem. Commun. 2008, 1287−1289. (22) Lin, Z.; Zhang, Z. M.; Chen, Y. S.; Lin, W. Angew. Chem. 2016, 128, 13943−13947. (23) Manna, K.; Zhang, T.; Greene, F. X.; Lin, W. J. Am. Chem. Soc. 2015, 137, 2665−2673. (24) Beyzavi, M. H.; Vermeulen, N. A.; Howarth, A. J.; Tussupbayev, S.; League, A. B.; Schweitzer, N. M.; Gallagher, J. R.; Platero-Prats, A. E.; Hafezi, N.; Sarjeant, A. A.; et al. J. Am. Chem. Soc. 2016, 138, 3251− 3251. (25) Song, F.; Wang, C.; Lin, W. Chem. Commun. 2011, 47, 8256− 8258. (26) Klet, R. C.; Tussupbayev, S.; Borycz, J.; Gallagher, J. R.; Stalzer, M. M.; Miller, J. T.; Gagliardi, L.; Hupp, J. T.; Marks, T. J.; Cramer, C. J.; et al. J. Am. Chem. Soc. 2015, 137, 15680−15683. (27) Comito, R. J.; Fritzsching, K. J.; Sundell, B. J.; Schmidt-Rohr, K.; Dincă, M. J. Am. Chem. Soc. 2016, 138, 10232−10237. (28) McInnis, J. P.; Delferro, M.; Marks, T. J. Acc. Chem. Res. 2014, 47, 2545−2557. (29) Delferro, M.; Marks, T. J. Chem. Rev. 2011, 111, 2450−2485. (30) Jüngling, S.; Müllhaupt, R.; Plenio, H. J. Organomet. Chem. 1993, 460, 191−195. (31) Cano Sierra, J.; Hüerländer, D.; Hill, M.; Kehr, G.; Erker, G.; Fröhlich, R. Chem. - Eur. J. 2003, 9, 3618−3622. (32) Li, L.; Metz, M. V.; Li, H.; Chen, M.-C.; Marks, T. J.; LiableSands, L.; Rheingold, A. L. J. Am. Chem. Soc. 2002, 124, 12725−12741. (33) Motta, A.; Fragalà, I. L.; Marks, T. J. J. Am. Chem. Soc. 2009, 131, 3974−3984. (34) Li, H.; Marks, T. J. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15295−15302. (35) Furukawa, H.; Gándara, F.; Zhang, Y.-B.; Jiang, J.; Queen, W. L.; Hudson, M. R.; Yaghi, O. M. J. Am. Chem. Soc. 2014, 136, 4369−4381. (36) Moon, S. Y.; Liu, Y.; Hupp, J. T.; Farha, O. K. Angew. Chem., Int. Ed. 2015, 54, 6795−6799. (37) Jiang, J.; Gándara, F.; Zhang, Y.-B.; Na, K.; Yaghi, O. M.; Klemperer, W. G. J. Am. Chem. Soc. 2014, 136, 12844−12847. (38) Kiernicki, J. J.; Zeller, M.; Bart, S. C. Angew. Chem. 2017, 129, 1117−1120. (39) Chakrabarti, N.; Ruccolo, S.; Parkin, G. Inorg. Chem. 2016, 55, 12105−12109. (40) Radlauer, M. R.; Agapie, T. Organometallics 2014, 33, 3247. (41) Chen, E. Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391−1434. (42) As in previous examples,26,27 the activity of ZrMe-BTC is grossly underestimated due to the predominant contributions of surface sites to olefin polymerization. (43) Pietikainen, P.; Seppala, J. V. Macromolecules 1994, 27, 1325− 1328. (44) Zhang, T.; Manna, K.; Lin, W. J. Am. Chem. Soc. 2016, 138, 3241−3249. (45) Ji, P.; Sawano, T.; Lin, Z.; Urban, A.; Boures, D.; Lin, W. J. Am. Chem. Soc. 2016, 138, 14860−14863. (46) Manna, K.; Ji, P.; Lin, Z.; Greene, F. X.; Urban, A.; Thacker, N. C.; Lin, W. Nat. Commun. 2016, 7, 12610.

Synthesis and characterization of Zr-BTC, ZrOH-BTC, ZrCl2‑BTC, Zr(CH2SiMe3)2-BTC, ZrMe2-BTC, and ZrMe-BTC; screening of conditions, general reaction procedures, and product characterizations for ZrMe-BTC catalyzed olefin polymerization (PDF)

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Pengfei Ji: 0000-0002-8109-7929 Alison Johnson: 0000-0002-7297-5400 Richard F. Jordan: 0000-0002-3158-4745 Wenbin Lin: 0000-0001-7035-7759 Author Contributions †

These authors contributed equally.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by NSF (CHE-1464941). We thank Dr. Kuangda Lu, Dr. Feng Zhai, Erik Reinhart, and Marek Piechowicz for experimental help and helpful discussions. XAS analysis was performed at Beamline 20BM-B, supported by the Materials Research Collaborative Access Team (MRCAT) Use of the Advanced Photon Source, an Office of Science User Facility operated for the U.S. DOE Office of Science by ANL, was supported by the U.S. DOE under Contract No. DE-AC0206CH11357.



REFERENCES

(1) Jordan, R. F.; Bajgur, C. S.; Willett, R.; Scott, B. J. Am. Chem. Soc. 1986, 108, 7410−7411. (2) Chien, J. C. W. Top. Catal. 1999, 7, 23−36. (3) Troutman, M. V.; Appella, D. H.; Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 4916−4917. (4) Williams, L. A.; Guo, N.; Motta, A.; Delferro, M.; Fragalà, I. L.; Miller, J. T.; Marks, T. J. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 413− 418. (5) Stalzer, M. M.; Nicholas, C. P.; Bhattacharyya, A.; Motta, A.; Delferro, M.; Marks, T. J. Angew. Chem. 2016, 128, 5349−5353. (6) Gu, W.; Stalzer, M. M.; Nicholas, C. P.; Bhattacharyya, A.; Motta, A.; Gallagher, J. R.; Zhang, G.; Miller, J. T.; Kobayashi, T.; Pruski, M.; et al. J. Am. Chem. Soc. 2015, 137, 6770−6780. (7) Eedugurala, N.; Wang, Z.; Chaudhary, U.; Nelson, N.; Kandel, K.; Kobayashi, T.; Slowing, I. I.; Pruski, M.; Sadow, A. D. ACS Catal. 2015, 5, 7399−7414. (8) Manna, K.; Everett, W. C.; Schoendorff, G.; Ellern, A.; Windus, T. L.; Sadow, A. D. J. Am. Chem. Soc. 2013, 135, 7235−7250. (9) Leitch, D. C.; Turner, C. S.; Schafer, L. L. Angew. Chem., Int. Ed. 2010, 49, 6382−6386. (10) Walsh, P. J.; Baranger, A. M.; Bergman, R. G. J. Am. Chem. Soc. 1992, 114, 1708−1719. (11) Sun, Q.; Wang, Y.; Yuan, D.; Yao, Y.; Shen, Q. Chem. Commun. 2015, 51, 7633−7636. (12) McKnight, A. L.; Waymouth, R. M. Chem. Rev. 1998, 98, 2587− 2598. (13) Gibson, V. C.; Spitzmesser, S. K. Chem. Rev. 2003, 103, 283− 315. (14) Makio, H.; Terao, H.; Iwashita, A.; Fujita, T. Chem. Rev. 2011, 111, 2363−2449. (15) Ma, L.; Abney, C.; Lin, W. Chem. Soc. Rev. 2009, 38, 1248− 1256. 11328

DOI: 10.1021/jacs.7b05761 J. Am. Chem. Soc. 2017, 139, 11325−11328