Syntheses and Studies of Group 6 Terminal Pnictides, Early-Metal Trimetaphosphate Complexes, and a New Bis-Enamide Ligand

Home , Adiponitrile, Phosphorus monoxide, Salt metathesis reaction

Syntheses and Studies of Group 6 Terminal Pnictides, Early-Metal Trimetaphosphate Complexes, and a New bis-Enamide Ligand

by MASSACHUSMS INSTITUTE

Christopher Robert Clough OF TECHNOLOGY B.S., Chemistry (2002) JUN 072011 M.S., Chemistry (2002) The University of Chicago Submitted to the Department of Chemistry ARCHIVES in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY June 2011

Author...... Department of Chemistry May 6,2011

Certified by...... Christopher C. Cummins Professor of Chemistry Thesis Supervisor

Accepted by ...... Robert W. Field Chairman, Department Committee on Graduate Studies 2 This Doctoral Thesis has been examined by a Committee of the Department of Chemistry as follows:

Professor Daniel G. Nocera ...... Henry Dreyfus Profe~ssofof Energy and Pr6fessor of Chemistry Chairman

Professor Christopher C. Cummins...... Professor of Chemistry Thesis Supervisor

Professor Richard R. Schrock ...... Frederick G. Keyes Professor of Chemistry Committee Member 4 For my Grandfather:

For always encouraging me to do my best and to think critically about the world around me. 6 Alright team, it's the fourth quarter. The Lord gave us the atoms and it's up to us to make 'em dance.

-Homer Simpson 8 Syntheses and Studies of Group 6 Terminal Pnictides, Early-Metal Trimetaphosphate Complexes, and a New bis-Enamide Ligand by Christopher Robert Clough

Submitted to the Department of Chemistry on May 6, 2011, in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Abstract Investigated herein is the reactivity of the terminal-nitrido, trisanilide tungsten complex, NW(N[i- Pr]Ar) 3 (Ar = 3,5-Me 2C6H3, 1). Nitride 1 has been shown to undergo an "N for (O)C1 metathesis with a variety of acid chlorides to form oxochloride (Ar[i-Pr]N) 3W(O)C1 (2) and the corresponding nitriles. The reaction of 1 with acid chlorides has been shown to proceed through an acylimido chloride intermediate. Furthermore, oxochloride 2 has been converted to a terminal phosphido trisanilide tungsten complex, PW(N[i-Pr]Ar) 3 (9) by treatment with the anionic niobium phosphide complex [Na(OEt2)][PNb(N[Np]Ar) 31. Nitride 1 and oxochloride 2 have been converted to pseudo-octahedral complexes through the use of electrophilic reagents such as oxalyl chloride and phosphorus pentachloride. (Ar[i- Pr]N)3W(OCN)(Cl) 2 (10) and (Ar[i-Pr]N)3W(N=PCl 3)(Cl) 2 (11) are synthesized by treating compound 1 with oxalyl chloride and PCl5, respectively. Similarly, (Ar[i-Pr]N) 3W(Cl) 3 (12) is formed by treatment of oxochloride 2 with PC15 with concomitant loss of oxyphosphorus trichloride. Reaction studies of trichloride 12 undertaken in the attempt to generate a low-coordinate tungsten species are also presented. Also reported presently is a new procedure for synthesis of the terminal phosphoryl complex (Ar[t-Bu]N)3MoPO (17) by treating phosphide (Ar[t-Bu]N) 3MoP (16) with the potent oxygen atom transfer (OAT) reagent mesitylnitrile oxide (MesCNO). In conjunction with collaborators, the thermodynamic and kinetic aspects of MesCNO as an OAT reagent with phosphide 16 and phosphines have been investigated. Density Functional Theory calculations of OAT reactions of MesCNO are also shown. In an effort to further develop the coordination chemistry of the trianionic, tridentate ligand trimetaphosphate, studies are described whereupon trimetaphosphate is metallated with molybdenum. (MeCN)3Mo(CO) 3 reacts with [PPN] 3[P30 9] -H20 ([PPN] = [Ph3P=N-PPh3]') to form the trimetaphosphate salt [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3[18]) in high yield. Efforts to generate trimetaphosphate vanadium oxo ((P30 9)V--O, 19) are also revealed. Finally, the synthesis of a new bis-enamide ligand class is described by the double addition of ketenimines to dilithium arylphosphanides. Formation of [Li(thf)]2 {PhP[C(CPh2)NPh] 21 ([Li(thf)] 2[20]) and [Li(thf)]2{MesP[C(CPh2)NPh] 2} ([Li(thf)]2[21]) are presented. The synthesis of tantalum species utilizing these new bis-enamide ligands is also demonstrated.

Thesis Supervisor: Christopher C. Cummins Title: Professor of Chemistry 10 Acknowledgements

It has been said that it takes a village to raise a child. In this case, it has taken a small metropolis to help an imbecile like me finish my Ph.D. In my opinion, the acknowledgements section is the most important part of this document. First off, this may be the only section that most people read. More importantly, none of the work in this dissertation would be possible without the love, support, encouragement, guidance, advice and ass-whipping that the people listed here have supplied. I am making every effort to thank each and every person who has made an impact on me, no matter how small, while I pursued my studies at MIT. In no way will I make any attempt at brevity in this section. My deepest apologies to anyone whom I forget to thank-in no way was the omission deliberate. Finishing this degree has truly been the hardest thing that I've ever done in my life and is one of the accomplishments of which I am most proud. To anyone reading this and especially to those listed in this section, I thank you from the bottom of my heart and am forever indebted to you. Science has always fascinated me-it was the one subject I always looked forward to in school for as long as I can remember. When most children wanted to be police officers, truck drivers or ballplayers, I dreamed of becoming a scientist. From the moment I could pronounce the word, I wanted to be a paleontologist. As most childhood dreams die, so did mine-I settled for chemistry. As I look back upon my career in chemistry, I feel it is best to proceed chronologically throughout my development as a scientist. I seriously doubt that I would have become a chemist had it not been for Mr. Quigel, my chemistry teacher sophomore year of high school. Mr. Quigel challenged us not just to learn the material presented, but to truly understand it-to understand why. I still recall the lab where we measured the freezing point of p-dichlorobenzene... The idea was simple enough, melt the material and then observe the temperature as a function of time while it cooled back down to room temperature. If people had simply read the material in the textbook, they would know that upon reaching the freezing point of the dichlorobenzene, the temperature would stabilize until the material had frozen fully. When Mr. Quigel was asked why the temperature had suddenly stopped decreasing at ca. 53 C, he simply gave the answer, "Maybe the thermometer is broken. Here, try this one." Not only did this save him from having to answer the question 20-some odd times (along with lecturing each student for not doing the required reading), it forced people to actually THINK about what's going on. I had the pleasure of taking Quigel's class for a full year my sophomore year and his AP chemistry class my senior year. It's because of him that things like HONClBrIF and the charge of every conceivable ion are burned into my head forever. Throughout all the pain and frustration of learning things like stoichiometry for the first time, Mr. Quigel helped (forced) me to think about why things are the way they are. It was then that I realized the beauty of chemistry. I can honestly say that by the end of my sophomore year of high school that I knew I wanted to be a chemist. After High School, I took a short trip down the Kennedy to attend the University of Chicago. I suppose at this point I should thank Prof. Viresh Rawal for putting me down the path to be an inorganic chemist. Here is my dark secret-I became an inorganic chemist not as an act of rebellion for hating Prof. Rawal's organic chemistry class, truly the opposite. I became an inorganic chemist entirely by accident. Early in my second year of college, I approached Prof. Rawal about joining his laboratory as an undergraduate researcher. He kindly explained to me that he already had four undergraduates working for him and he suggested four other faculty members in chemistry that were looking for undergraduates: Hillhouse, Hopkins, Jordan and Piccirilli. Alphabetically, "Hillhouse" comes first, so I went to talk to Greg. (Admittedly, three out of the four are inorganic chemists, so Prof. Rawal had stacked the deck...) Greg Hillhouse gave me several reprints of his articles and told me to go home, read them over and then decide if that's the sort of thing that I wanted to do. I tried. I really did. But never having taken an inorganic chemistry course, it was like trying to read Greek. I remember how bizarre it looked seeing carbons bound to nickel. After about a week, Greg called me up and asked what I thought. (Eleven years later, I'm paraphrasing the conversation that took place.) "Dude, wondering if you wanted to sign up. I had another girl come by after you asking if she could join the lab. I only want to take one undergrad this quarter and since you were here first, the choice is yours." Without hesitation, I said yes. And it was perhaps the best decision I have made in my life. Working in Greg's lab was a wonderful experience. I had figured that I would be treated as a glorified lab tech-washing glassware, allowed to run an occasional reaction... But from the very start, I was given my own project. And I loved it. If I wasn't in class or sleeping, I was in lab (this may be a slight exaggeration as I do recall having some fun in college). Greg provided a wonderful environment in which I was very eager to learn as much as I could. Although not fun at the time, I'm very thankful for being grilled in group meeting and being required to assign every single bump, no matter how small, in the NMR spectra that I presented. The result was that analyzing my compounds by NMR spectroscopy became second nature. Despite "accidentally" becoming an inorganic chemist, I thank Greg for making me realize how fun and exciting this field is. I'll always remember heading to Jimmy's or "Homo Mex" for a few liquid refreshments while being asked, "dude... any hot new compounds?" I would be remiss if I didn't mention some of the other denizens of the Hillhouse group. Beatrice Kendall nde Lin was my mentor upon joining the group. She taught me just about everything-from using the high-vac line to simply doing literature searches using Scifinder. I thank her for her patience as I learned the ropes of working in a chemistry lab. It was with Bea that I got my first publication. I thank her for everything she did to help me get started as a researcher. At the same time I joined the lab, so did Rory Waterman n6 Melenkivitz. Still not sure what the hell that name change was about, but I guess I'll let it slide. Back to the nice stuff. Rory was a constant source of entertainment in lab. He was also always willing to help with any sort of question I had about lab or chemistry in general. I knew better than to ask Rory questions about life, though. Rory, if you're reading this, I'm sorry I tried to gas you by leaving those stinky thiiranes in the rotovap. I guess it makes us even for the time you tried to blow me up with the waste bottle explosion. I still miss heading over to the Divinity School cafe with you for coffee and falafel. Andy Tennyson, I miss you man. Had some good times with you both in Chicago and at MIT. Good luck to you at Clemson! While in the Hillhouse Lab, I had the unique experience of not only working with, but also living with Professor Herr Doktor Daniel Jos6 Mindiola. I learned A LOT from Dan. When just about everything we did in the lab was on the high-vac line, Dan came in and asked us why we were still living in the Dark Ages. Here we had two gloveboxes that we used for little more than storing materials and loading/unloading swivel-frit reaction mixtures. He showed us the power that one could yield by doing reactions in that glovebox. Suddenly there was the possibility to do more than one to two reactions a day! And my God, I have to say that Mindiola probably possesses the most talented set of hands of any chemist that I have ever seen. He was kind enough to share some of his secrets with me. I think just about every trick I know about crystallization I learned from Dan. And despite the trouble I got myself into in the process, I really enjoyed the "Cadillac Dan" incident. I'll save Mindiola the embarrassment of telling any stories about us living together. I can say this though, that was an interesting year. In Autumn of 2002, I moved East to the greater Boston area to attend grad school at MIT. In my first year I joined Kit's lab. I think it's fairly obvious that Kit and I have occasionally had our differences. Despite that, I could not have asked for a better adviser. Kit's enthusiasm towards all things chemistry absolutely amazes me. I have never met anyone (and seriously doubt that I ever will meet anyone) that truly loves chemistry as much as Kit does. Kit has shown incredible patience with me as I turned my graduate career into an endurance event. Kit, you have always been incredibly helpful with advice (both chemical and other) over the years. I would not have accomplished any of the beautiful chemistry in the thesis without your help, guidance and suggestions. And the ass- whipping helped keep me on track, too. You have demanded nothing but the best from me over the years. Your high expectations have helped me to become the chemist that you always knew I was capable of being. Also, thank you for giving me the opportunity to finish my Ph.D. by letting me back into the group. I'm sorry I was such a royal pain in the ass. Since I've spent quite a bit of time at and around MIT, there are a lot of people that I need to thank. Even though it's been quite a collection of alcoholics and drug addicts, I need to thank each and every member of the Cummins Group with whom I have overlapped over the years. I spent my first couple years working in a the same glovebox as Paula and Fran. Since Paula was finishing up at the time I joined the group, I'm mostly thankful that she didn't harm me as I was learning "the rules" for her glovebox. Fran, you were a wonderful labmate. I had a great time working with you and an even better time going out for the occasional cocktail. I wish you and Bird Dog the best out in New Mexico. Josh Figueroa/jGoni/Blackstash the Pirate or whatever you're going by now, thanks. You were always incredibly helpful and insightful when it came to chemical advice. Thanks for showing me some great times late night at the "Britch". I also had quite a bit of fun with you, chasing the ladies at the Middlesex. Arjun, you were always fun to mess with. Oh how I enjoyed making "art" to put on your desk. You certainly got me back by defending yourself-and I had the black eye to prove it. Always had a great time with you, especially when we were imbibing or synthesizing our own imbibables. HanSen Soo and Erin Daida, you were two of the most talented undergraduate chemists that I've ever seen. You both certainly inspired me to step up my game. Erin, I still miss smoking P-Funks with you in the Atomic Courtyard. Tetsu, thank you for taking such great care of the THF still. I always enjoyed talking baseball or chemistry with you. And James Blackwell. Man, I have missed SO MUCH good music since you left. I treasured the nights were you rolled into my lab and dragged me along to a show at TT's or the Middle East. I'll never forget the late night walks back from Boston across the Harvard Bridge or the cab rides to Toad to catch last call and some good Folk. Death, I owe you so much. From all the advice you gave me over the years to how invaluable and instrumental you were in helping me get my new job. Thank you. That just about sums up everyone from Tour of Duty, I. Except for the foursome. John Curley, Nick Piro, Alex Fox and Glen Alliger. I congratulate you guys for disproving the n - 1 rule. You four joined up towards the end of my first tour and you poor SOB's were still in the lab when I rejoined in '08. John, keep reaching for that rainbow! I had the pleasure of working with you both in Greg's lab and in Kit's lab. And wow, you were an insufferable %@ &* for most of that time. In hindsight, though, you reap what you sow-I really had it coming. Towards the end of your tenure here, you really lightened up. I mean it when I say I miss having you around. Nick, wow, I'm still in awe of what you did to the Goddess Niobe. Great work man. Had some pretty insightful conversations with you that helped me a lot. ArFox, I hope that twitch is gone. We sure had some great times in the 327 while we worked together. Glen, thanks. After some time at QD Vision out in Watertown, MA, I returned to MIT to finish my Ph.D. For the folks I overlapped with in Tour II, here goes. Matt Rankin, you were an absolute wizard when it came to NMR. I thought I knew what I was doing when it came to NMR, your knowledge blew me out of the water. Wow. Thanks for the pointers. You did an incredible job with Tantalum. It was always fun to walk into group meeting and see what magic you'd accomplished in the lab that week. I had a whole lot of fun with you while you were here, even though Brandi banned us from going out for drinks unchaperoned. I'll certainly never forget the whirlwind that was your last couple of weeks here. Manuel Temprado, I miss you bro! Glad we're still working together of late. Great job on those calcs, by the way! Heather Spinney, thank you for your constant encouragement. You were always one of the folks "on my team"-and you still are. Thank you thank you thank you for all your kind words reminding me that I can actually do this. Brandi Coissart, I don't know if I've ever met a harder working chemist. You were the latest magician to show the beauty of niobium. Had some wonderful times hanging out with you and that loser you date. Hope all is going well in NYC. And I know I'll see you soon. Jared, thanks. Alexandra and Tofan, the "twins"-you two still have a long road ahead of you and I wish you the best. Alexandra, your enthusiasm is infectious. Thank you for making the lab a brighter place. Tofan, you got some good stuff going on, keep at it! Take it from me, don't screw this one up! Hieu, you too. Get at it! Get it done! You can do it! I also have to thank Anthony Cozzolino. Anthony, good times hanging out-especially when there's moose involved. And thank you for all your help with crystallography. Whenever there is some minor aspect that's making checkcif go bonkers, you always know how to fix it. In my last few months here, you saved me considerable amounts of time by always knowing how to fix that A-level alert. Nazario Lopez, wow. Just wow. You've been a blast to have in the lab. Always looking where to go next. I've had a good time with you since you joined up-it's always nice to have another "smoker-chemist", we're a dying breed. Alex Vai, it's been great working with you. You've done some great stuff. Good luck next year at Oxford. And thank you to the revolving door of Germans we've had here of late: Andreas, Jens, Mona, Albert, Manuel, and Christian. Allison Kelsey, you get your own paragraph. Through a weird turn of events, I've known you longer than anyone else here at MIT. What were you doing in my apartment in Chicago? Thank you for everything! I spend almost as much time in your office as John did in Nick's. Anyway I can talk you into moving to Haverhill? I'm really not sure what the hell I'll do without being able to talk to you on a daily basis. No matter what I need, you're always willing to help. Thank you for being yet another source of constant encouragement as I struggled to finish this degree. Without all the help you've given me over the years, getting to this point would have been infinitely harder. While here, I haven't been completely xenophobic and have occasionally interacted with people from other groups. From the Schrock Group, I have to thank my two favorite girls: Andrea Gabert and Tanya Pilyugina. Had a wonderful time with both of you. Thank you for all the help and encouragement you two have given me over the years. Tanya, congrats on the twins! I wish you and Matt the best. Andrea, I'll see you soon, I still owe you dinner! And Andrea, I'm still sorry about the "second year orals incident". Zach Tonzetich, you are an incredible chemist. Thank you for all the discussions we had over the years. Good luck down in Texas, buddy. Hock, Bailey and Wampler, take care guys. It's been fun! From the Nocera group, I have to thank the old guard. Bart, Manke, Shores, Niels, Streece, Arthur... Wow, the stories. I could go on for ages. Thank you all. Little Emily and Arturo from the new crew, you two are the best. Can't tell you how much fun I've had with you both. Dorothy, thanks. And to everyone I've ever played softball with (too many to list), whether on the Organomets or the Red NOX, thank you! Summer just won't be summer without playing ball... I would also like to thank my thesis committee: Joseph Sadighi, Dan Nocera and Dick Schrock. Joseph was my first thesis chair. Joseph, thank you for all your help and advice during my first tour. Good luck at Georgia Tech. Dan, thanks for stepping in and filling the gap I had in the "Thesis Chair" department when I came back to MIT. You've been incredibly helpful during our thesis chair meetings. Thank you for the invaluable advice. Dick, thank you for all your advice over the years. It's truly been a pleasure to be able to talk to one of the masters of inorganic chemistry. When I started out working on tungsten, you were a great help as I was trying to learn the ropes of my multistep starting material synthesis. Thanks to all the DCIF staff, past and present. Special thanks to Mark, Dave and Jeff. You guys were always incredibly helpful whenever I needed help. I wouldn't have been able to characterize all of the compounds that I did without your help. In the X-ray lab, past and present, I have to thank Bill and Michael. And of course Peter. Peter, thank you for all of your help with structures over the years. I don't know how you do the magic that you do, but it is impressive, to say the least. I have learned so much from you and I hope to collaborate with you in the future. Thank you to everyone at QD Vision. Especially the chemistry folks. Craig, I still miss sharing a hood with you. Those were some good times. Over the years, I've managed to meet a few people outside of chemistry. Here is that short list. Robot, you've been a great roommate. Always enjoyed the late night chat sessions at HuMo where I tried to teach you chemistry and you tried to teach me EE. Despite all the collateral damage, I especially had a great time the night we almost got arrested-twice. Bradford, good luck down in Texas buddy. Can't wait to hear that album. Thanks for all the late nights on Douglass St. Laura, good luck at NYU. Thanks for all the memories. To all the folks at Hungry Mother, especially John, Alon, Rachel, Ned and Heather, thanks for making us feel at home. Some of my best chemistry was sketched out at your bar next to an old fashioned. Also have to thank the Miracle of Science and all the employees past and present. The rest of my chemistry ideas were probably sketched out there. I have to thank the Alliger family as a whole. I had some good times with Andrew, Grog and David in their frequent visits out to Boston. Some of the barbecues at the 42 were especially memorable. Mr. and Mrs. Alliger, I also have to thank you for your hospitality in letting Jared and myself stay at your house for a weekend getaway in Albany. We both had a wonderful time. Not sure why you let Glen back in the house, though. And this is for you, Grog: Chump, Chumpette, Yours, Up, Pimpmobile, Bite, My, Shiny, Daffodil, Ass. Hope you're happy. I suppose I should also thank the roommates I've had here at the 42. Kevin and Kyle, I'll never forget our first year. Sitting around mid-August, waiting for classes to start. Roasting. Drinking ice-cold Coors Light and eating freeze pops. Kyle, I still have no freaking clue what you're doing in Singapore, but I wish you and Alli the best. Get back to the states soon. Kevin, I miss eating mass quantities of meat while watching sports with you. Not once did we find a movie on that we considered unwatchable. From "Kung Pow: Enter the Fist" to anything with subs or aliens, always a good time. Venda, thanks for putting up with me while we lived together. I had a great time discussing the mass of a hole. Kevin and Venda, I consider you two family. I've had a great time the past few years having Thanksgiving dinner with you. I can't wait until November this year. See you both soon. I love you both. OK, here's the real acknowledgements for Glen and Jared. Glen, you're my boy! I owe you so many beers. Thank you for your invaluable help in getting my thesis done. Glen took the time to edit my entire thesis. So, if you see any mistakes, blame him. I miss you bro, can't wait to see you this June at graduation. I'll miss our late night mistakes that seemed to happen far too often when we lived together. Whether we were talking music, sports, chemistry or life-I wouldn't trade that time for anything. Thanks for always encouraging me and helping me through my time at MIT. You know better than anyone that I didn't exactly have the easiest time doing it. It's Miller Time! As I write this, I'm toasting a High Life to you. Jared, as we finish up at the same time, I am pretty sure now that you will not, in fact, stab me. And that's a relief. It hasn't been easy, but I have actually had a good time with you during this mad dash towards the end. We seem to have both been affected by "the crazies" during the last month while we finish our theses. It's comforting to know that I'm not the only one losing my grip on reality as I write up the past eight-odd years of my life. Thank you for all your help in lab, as well. You've kept the 327 running as a well-oiled machine. Lastly, I need to thank you for encouraging me and keeping me focused on getting done. It's truly been a great help. Jared, Glen, I can't wait to get that little piece of paper with you both in June. Thank you both, you've been better roommates and labmates than a bum like me deserves. Matt Hill, you're still my fishing girlfriend. Thanks for always being willing to drive my ass all around New England to go rip some lips. Looking forward to our next trip. I can't wait to wet a line again. It's been too long. I'd also like to thank Ed Mitchell for sharing his second home with Matt and me. Ed, I've had a great time using your cabin as a base of operations for fishing trips. Thank you. I've got to thank my friends from afar, too. Fred, you've always been willing to field my late night calls. Thanks for all the advice over the years. You'll always be my brother from another mother. I love you, kid. Rob, thanks for making me feel at home every time I come back to Chicagoland. I love the carnage that ensues every time we get together. Jenny, it's been fun talking with you over the past year. You've been a great friend. Thanks for all the advice. Matt, David, Candy, Shawn, I love hanging out with you guys. Can't wait to see you all again. Miranda, thank you. Whether you realize it or not, you've been a great help for me in finishing up here at MIT. For the past twelve years, you've always been there for me. I hope to be there for you in the future. Anything you need, don't hesitate to ask-I'm only a phone call away. I wish you nothing but the best. Take care of Wyatt. I miss you both. I also need to thank the Girl in the Moon. You've always been good to me and helped me through some tough times. No matter where I end up or who I'm with, you'll always hold a special place in my heart. Finally, I'd like to thank my family. Your love and support has been invaluable over this long road that has been my Ph.D. I love you all. Ben and Jenny, I can't wait to be around a little more so that I can see my beautiful niece and nephew. Straight sister, I still don't see what you see in my brother. Ben, I'll miss the late nights that I pull in lab where I call you just as the sun is coming up here in Boston. I swear, each time I expected you to be driving to work, but it never fails. You're somewhere on the west coast and it's some unholy hour. Man, I really enjoyed that. Katie, thanks for Petey! Love you little sister. I miss seeing you and your smile. Uncle Dr. Holland and Aunt Lori, thank you for keeping me focused and making me realize what is important. You both helped me a great deal in realizing that I needed to go back and finish my Ph.D. Grandpa and Rose, thank you both very, very much. I couldn't have done it without your help and encouragement. Thank you for giving me a hand my last year here at MIT. Grandpa, I've really enjoyed talking to you. Thanks for helping me stay focused. And Rose, I'll remember your phrase. It's helped quite a bit. Lastly, ------......

Mom and Dad, thank you for everything. Thank you for all the help over the years-keeping me focused, talking me off ledges, etc. I love you both. There's no way I would've become the man I am today without your love, support and encouragement. I'll do my best to get back to the Midwest someday. Short of that, I'll visit a hell of a lot more often. I also have to thank Petey, 'cause he's my real tight homeboy. Petey's my cat. Since Little Emily included a picture of a cat in her acknowledgements, I figured I'd see her picture and raise her a better one. So, here's Peter: Table of Contents

1 Synthesis of Nitriles from a Tungsten Nitride Complex 29 1.1 Introduction ...... 30 1.2 Background ...... 31 1.3 Nitriles Synthesized from a Tungsten Terminal Nitride ...... 33 1.4 Probing the Mechanism of Nitrile Formation ...... 35 1.5 Treatment of Oxochloride with Triflic Anhydride ...... 40

1.6 Synthesis of a Terminal Phosphide, PW(N[i-Pr]Ar) 3 (9) . .. --- . -- . -- .. - 40 1.7 Conclusions ...... 41 1.8 Experimental Section ...... 42 1.9 DFT Calculations...... 47 1.10 Crystallographic Structure Determinations ...... 48 1.11 References ...... 53

2 Synthesis and Reactivity of Pseudo-Octahedral Tungsten Trisanilide Species 2.1 Introduction ...... 2.2 Synthesis of a Pseudo-OctahedralTungsten Trisanilide Species ...... 2.3 Syntheses of Pseudo-Octahedral Tungsten Trisanilide Species using PCl5 2.4 Reduction Studies of (Ar[i-Pr]N) 3W(Cl) 3 (12)......

2.5 Synthesis of OW(N[i-Pr]Ar) 3 (15)...... 2.6 Conclusions...... 2.7 Experimental Section ...... 2.8 Crystallographic Structure Determinations...... 2.9 References......

3 Oxygen Atom Transfer from MesCNO to (Ar[t-Bu]N)3MoP to form (Ar[t-Bu]N) 3MoPO. 83 3.1 Introduction ...... 83 3.2 Results and Discussion ...... 84 3.3 Conclusions ...... 92 3.4 Experimental Section...... 94 3.5 References ...... 95 4 Synthesis of Early Metal Trimetaphosphate Complexes 4.1 Introduction ......

4.2 Synthesis and Characterization of [PPN]3[(P30 9)Mo(CO) 3] ([PPN] 3[18]) ... .. 4.3 Synthesis and Characterization of (P30 9)V=O (19) ...... 102 4.4 Conclusions Sectio...... 105 4.5 Experimental Section ...... 106 4.6 DFT Calculat ions ...... 108 4.7 Crystallograp 108 hic.trutur.Deermnaton...... 4.8 References . 112

5 Chemistry of a new bis-Enamide Ligand 115 5.1 Introduction ...... 116 5.2 Synthesis of bis-Enamide Ligands ..... 118 5.3 Metallation of bis-Enamide Complexes with Tantalum 119 5.4 Discussion...... 123 5.5 Conclusions ...... 124 5.6 Experimental Section ...... 125 5.7 Crystallographic Structure Determinations . 129 5.8 References ...... 134

A Unpublished Crystal Structures 135 A. 1 Introduction...... 135 A.2 Discussion...... 135 A.3 Experimental Section ...... 140 A.4 Crystallographic Structure Determinations . 142 A.5 References...... 147 List of Figures

1.1 ORTEP drawing of NW(N[i-Pr]Ar) 3 (1) ...... 32

1.2 ORTEP drawing of (Ar[i-Pr]N) 3W(O)C1 (2) ...... 34

1.3 ORTEP drawing of (Ar[i-Pr]N) 3W(=NC[O]CF 3 )(OC[O]CF3 ) (4) ...... 37

1.4 ORTEP drawing of (Ar[i-Pr]N) 3W(=NC[O]CF 3 )C1 (5) ...... 38

1.5 ORTEP drawing of PW(N[i-Pr]Ar) 3 (9) ...... 42

1 2.1 H NMR spectrum of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10) -.-...... 57

2.2 FT-IR spectrum of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10) ...... 58

2.3 ORTEP drawing of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10) -.-.-...... 59

2.4 ORTEP drawing of (Ar[i-Pr]N) 3W(N=PC 3 )(Cl) 2 (11) . -- ...... 61

2.5 ORTEP drawing of (Ar[i-Pr]N) 3W(Cl) 3 (12) ...... 63

2.6 ORTEP drawing of (Ar[i-Pr]N) 3W(Cl) 2 (13) ...... 64 2.7 ORTEP drawing of [Na(OEt 2 )1.5 (thD0.5] 2 [W(H)(rf -Me 2C=NAr)2 -cyclo-(N[iPr]Ar)] ([Na(OEt 2)1.5(thf)0.5][14]) ...... 66

2.8 ORTEP drawing of OW(N[i-Pr]Ar) 3 (15) ...... 68

2.9 ORTEP drawing of OW(N[i-Pr]Ar) 3 (15) highlighting the tungsten oxo dimer core 69

3.1 UV/Vis spectrum of (Ar[t-Bu]N)3MoPO (17) ...... 89

3.2 Plot of observed rate constants for the reaction of (Ar[t-Bu]N)3MoP (16) + MesCNO at various temperatures ...... 90

3.3 Eyring plot for the reaction of MesCNO and (Ar[t-Bu]N) 3MoP (16) ...... 90

3.4 Computed frontier orbitals for PhCNO and PMe 3 ...... 92

3.5 Intrinsic reaction coordinates for PhCNO + PMe3 .... . - ...... 93

3.6 Computed structures of intermediates and transition states for PhCNO + PMe 3 . .. 93

3 4.1 ORTEP drawing of anion [(P3 0 9 )Mo(CO) 3 ] - (183-) ...... 100

4.2 ORTEP drawing of [PPN] 3 [(P30 9 )Mo(CO) 3] ([PPN] 3 [18]) ...... 101 3 1 4.3 P NMR spectrum of VOCl 3 + [PPN] 3[P 30 9] -H20 ...... 104 51 4.4 V NMR spectrum of VOCl 3 + [PPN] 3 P3 0 9] -H20 ...... 104

4.5 Calculated structure of (P30 9 )V=O (19) ...... 106 3 4.6 Solid-state structure of [(P30 9)Mo(CO) 3] - (183-) showing the ca. 95:5 positional disorder ...... 110

5.1 ORTEP drawing of [Li(OEt2)] 2{PhP[C(CPh2)NPh] 2} ([Li(OEt2)] 2 [20]) ...... 120 5.2 ORTEP drawing of {PhP[C(CPh2)NPh] 2}TaMe3 (22) ...... 121 5.3 ORTEP drawing of {MesP[C(CPh2)NPh] 2}TaMe3 (23) ...... 122 5.4 ORTEP drawing of {MesP[C(CPh2)NPh] 2}Ta(I)Me 2 (24) ...... 123

A.1 ORTEP drawing of NW(O-t-Bu)(N[i-Pr]Ar) 2 ...... 136 A.2 ORTEP drawing of OW(N[i-Pr]Ar) 2(Cl) 2 ...... 137

A.3 ORTEP drawing of the asymmetric unit of [Ar(i-Pr)NBH 2]2 . - ...... 138 A.4 ORTEP drawing of [Ar(i-Pr)NBH 2]2 . . . . - ...... -...... 139

A.5 ORTEP drawing of [(CH 3CN) 2V(O)(C)-p-Cl] 2 ...... 140 List of Schemes

1.1 Examples of nitrile synthesis from acid chlorides...... 31

1.2 Synthesis of NW(N[i-Pr]Ar) 3 (1) ...... 32 1.3 Synthesis of nitriles from NW(N[i-Pr]Ar) 3 (1) ...... 33 1.4 Synthesis of adiponitrile from NW(N[i-Pr]Ar) 3 (1)...... 34 1.5 Stack plot of a 1H NMR spectrum showing NW(N[i-Pr]Ar)3 (1) + R(O)C1 -

acylimido -+ (Ar[i-Pr]N) 3W(O)Cl (2) + RCN ...... 35 3 1.6 Treatment of NW(N[i-Pr]Ar) 3 (1) with 1 C-labeled 1-adamantoyl chloride . . . . . 36 1.7 Synthesis of acylimidos (Ar[i-Pr]N)3W(=NC[O]CF 3)(OC[O]CF 3) (4)

and (Ar[i-Pr]N) 3W(=NC[O]CF 3 )Cl (5) ...... 36 1.8 Proposed intermediate in the nitrile forming reaction...... 39

1.9 Synthesis of (Ar[i-Pr]N) 3W(O)(OTf) (8)...... 40 1.10 Synthesis of PW (N[i-Pr]Ar)3 (9) ...... 41

2.1 Synthesis of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10)...... 57 2.2 Synthesis of (Ar[i-Pr]N)3W(N=PCl 3)(Cl) 2 (11) .. - -...... 60 2.3 Synthesis of (Ar[i-Pr]N) 3W(Cl) 3 (12) ...... 62 2.4 Synthesis of (Ar[i-Pr]N) 3W(Cl) 2 (13)...... 64 2 2.5 Synthesis of [Na(OEt 2) t 5(thf)0.5][W(H)(i1 -Me 2C=NAr) 2-cyclo-(N[iPr]Ar)] ([Na(OEt 2) 1.5(thf)o.s][14]) .. - . . -. ------...... 65 2.6 Synthesis of OW(N[i-Pr]Ar) 3 (15) ...... 67 2.7 Reaction of OW(N[i-Pr]Ar) 3 (15) with triflic anhydride ...... 68

3.1 Synthesis of (Ar[t-Bu]N) 3MoPO (17)...... 84 3.2 Generic reaction of AnP with MesCNO ...... 85

3.3 Measured and derived enthalpies for (Ar[t-Bu]N) 3MoEO (E = P, N) ...... 87

4.1 Synthesis of [N(n-Bu)4]2[(P30 9)M(CO) 3] (M = Mn, Re) ...... 98

4.2 Syntheses of Na 3P30 9 -6 H20, [PPN] 3[P30 9]. H20 and [PPN] 3 [(P30 9)Mo(CO) 3] ([PPN] 3[18]) ...... 99 4.3 Proposed synthesis of (P309)V=O (19)...... 103

5.1 Ligand Syntheses utilizing arylphosphanides ...... 117 5.2 Reaction of dilithium phenylphosphanide with two equivalents of diephenylketene 118 5.3 Syntheses of bis-enamide starting materials ...... 118

5.4 Synthesis of [Li(thf)]2{ArP[C(CPh2)NPh]2} (Ar = Ph, Mes) ...... 119 5.5 Synthesis of {ArP[C(CPh2)NPh] 2}TaMe 3 (Ar = Ph, 22; Ar = Mes, 23) ...... 120 5.6 Synthesis of {MesP[C(CPh2)NPh] 2}Ta(I)Me 2 (24) ...... 122 5.7 Synthesis of a bis[N-(trimethyl)iminobenzoyl]phosphanide ligand ...... 124 List of Tables

1.1 Crystallographic Data for (Ar[i-Pr]N)3W(O)Cl (2),

(Ar[i-Pr]N) 3W(=NC[O]CF 3 )(OC[O]CF 3) (4) and (Ar[i-Pr]N) 3W(=NC[O]CF 3 )Cl (5) 51

1.2 Crystallographic Data for PW(N[i-Pr]Ar) 3 (9) ...... 52

2.1 Crystallographic Data for (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10), (Ar[i-Pr]N) 3W(N=PCl 3 )(Cl) 2 (11) and (Ar[i-Pr]N) 3W(Cl) 3 (12) ...... 79

2.2 Crystallographic Data for (Ar[i-Pr]N) 3 W(Cl) 2 (13), 2 [Na(OEt 2)1.5(thf70. 5][W(H)(rj -Me 2C=NAr) 2-cyclo-(N[iPr]Ar)]

([Na(OEt 2 )1.5 (th)0.5 ][14]) and OW(N[i-Pr]Ar) 3 (15) ...... 80

3.1 Enthalpies of oxygen atom transfer reactions ...... 85 3.2 Experimental and computed X-O bond dissociation energy values...... 86 3.3 Rate constants, derived activation parameters and reaction enthalpies for reactions of AnP + M esCNO ...... 91

4.1 Experimental and calculated vCO stretching frequencies (cm-') for

[(P3 0 9)M(CO) 3]n ...... - ...... 99 3 4.2 Experimental and calculated bond lengths for [(P30 9 )Mo(CO) 3 ] - (183-) .... . 102

4.3 Experimental and calculated NMR shifts for (P 30 9)V=O (19) ...... 103

4.4 Crystallographic Data for [PPN]3 [(P 30 9)Mo(CO) 3] ([PPN] 3 [18]) ...... 111

5.1 Crystallographic Data for [Li(OEt 2)]2 {PhP[C(CPh2)NPh] 2} ([Li(OEt2)] 2 [20]), {PhP[C(CPh2)NPh] 2}TaMe3 (22) and {MesP[C(CPh2)NPh] 2}TaMe 3 (23) ..... 132

5.2 Crystallographic Data for {MesP[C(CPh2)NPh] 2}Ta(I)Me 2 (24) ...... 133

A. 1 Crystallographic Data for NW(O-t-Bu)(N[i-Pr]Ar) 2, OW(N[i-Pr]Ar) 2(Cl) 2 , and

[Ar(i-Pr)NBH 2 ]2 - . . . . . --...... - - --. ----...... 145

A.2 Crystallographic Data for [(CH 3CN) 2V(O)(Cl)-p-Cl] 2 ...... 146 26 List of Compounds

1 NW(N[i-Pr]Ar) 3 2 (Ar[i-Pr]N) 3W(O)C1 3 3 (Ar[i-Pr]N) 3W(=N1 C[O]'Ad)Cl 4 (Ar[i-Pr]N) 3W(=NC[O]CF 3)(OC[O]CF 3 ) 5 (Ar[i-Pr]N) 3W(=NC[O]CF 3)C 6m (CH 3C[O]N)W(NMe2) 3 C1 7m (CF 3C[O]N)W(NMe 2) 3 C1 8 (Ar[i-Pr]N) 3W(O)(OTf) 9 PW(N[i-Pr]Ar) 3 10 (Ar[i-Pr]N) 3W(OCN)(Cl) 2 11 (Ar[i-Pr]N) 3W(N=PCl 3)(Cl) 2 12 (Ar[i-Pr]N) 3W(Cl)3 13 (Ar[i-Pr]N) 3W(Cl)2 2 [Na(OEt 2) 1.5(thf)o. 5][14] [Na(OEt 2) 1.5 (thf)0 .5] [W(H)(1 -Me2C=NAr) 2-cyclo-(N[iPr]Ar)] 15 OW(N[i-Pr]Ar) 3 16 (Ar[t-Bu]N) 3MoP 17 (Ar[t-Bu]N) 3MoPO [PPN] [18] 3 [PPN] 3 [P 3 0 9 )Mo(CO) 3] 19 (P3 0 9)V=O [Li(thf)] 2 [20] [Li(thf)] 2 f{PhP[C(CPh2)NPh] 2} [Li(thf)] [21] 2 [Li(thf)] 2 {MesP[C(CPh 2)NPh]2} 22 {PhP[C(CPh2)NPh] 2}TaMe3 23 {MesP[C(CPh2)NPh] 2}TaMe3 24 {MesP[C(CPh 2 )NPh]2}Ta(I)Me 2 28 CHAPTER 1

Synthesis of Nitriles from a Tungsten Nitride Complex

Contents 1.1 Introduction ...... 30 1.2 Background ...... 31 1.3 Nitriles Synthesized from a Tungsten Terminal Nitride ...... 33 1.4 Probing the Mechanism of Nitrile Formation ...... 35 1.5 Treatment of Oxochloride with Triflic Anhydride ...... 40

1.6 Synthesis of a Terminal Phosphide, PW(N[i-Pr]Ar) 3 (9) ...... 40 1.7 Conclusions ...... 41 1.8 Experimental Section ...... 42 1.8.1 General Considerations...... 42

1.8.2 One Pot Synthesis of W2 (0-t-Bu) 6 ...... ------. . 43

1.8.3 Improved Synthesis of Zr(N[i-Pr]Ar) 4 ...... 43

1.8.4 Improved Synthesis of NW(O-t-Bu)(N[i-Pr]Ar) 2 ...... 44

1.8.5 Synthesis of NW(N[i-Pr]Ar) 3 (1) ...... ----- ... 44

1.8.6 Reaction of NW(N[i-Pr]Ar) 3 (1) with pivaloyl chloride ...... 45

1.8.7 General Synthesis of Nitriles from NW(N[i-Pr]Ar) 3(1) ...... 45

1.8.8 Synthesis of (Ar[i-Pr]N) 3W(O)Cl (2) ...... 45 1 13 1.8.9 Reaction of NW(N[i-Pr]Ar) 3 (1) with Ad C(O)C1 ...... 46

1.8.10 Synthesis of (Ar[i-Pr]N) 3W(=NC[O]CF3)(OC[O]CF 3) (4) ..... - .. 46

Reproduced in part with permission from: Clough, C. R.; Greco, J. B.; Figueroa, J. S.; Diaconescu, P. D.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc., 2004, 126, 7742-7743; Copyright 2004 American Chemical Society. Clough, C. R.; MUller, P. M.; Cummins, C. C. Dalton Trans., 2008, 4458-4463; Copyright 2008 The Royal Society of Chemistry. Fox, A. R.; Clough, C. R.; Piro, N. A.; Cummins, C. C. Angew. Chem. Int. Ed. 2007, 46, 973-976; Copyright 2007 Wiley-VCH. 1.8.11 Synthesis of (Ar[i-Pr]N) 3W(=NC[O]CF 3)C1 (5)...... 46

1.8.12 Synthesis of (Ar[i-Pr]N) 3W(O)(OTf) (8)...... 47

1.8.13 Synthesis of PW(N[i-Pr]Ar) 3 (9) - ...... 47 1.9 DFT Calculations ...... 47 1.10 Crystallographic Structure Determinations ...... 48 1.10.1 General Considerations...... 48

1.10.2 X-ray crystal structure of (Ar[i-Pr]N) 3W(O)C1 (2)...... 48

1.10.3 X-ray crystal structure of (Ar[i-Pr]N) 3W(=NC[O)CF3)(OC[O]CF 3) (4) . 48

1.10.4 X-ray crystal structure of (Ar[i-Pr]N) 3W(=NC[O]CF 3)Cl (5) ...... 49

1.10.5 X-ray crystal structure of PW(N[i-Pr]Ar) 3 (9) ...... 49 1.11 References ...... 53

1.1 Introduction

Transition metal, terminal nitride species have long been used as nitrogen atom sources for organic molecules. 1,2 Carreira and co-workers have shown an aziridination/hydroxyamination process with a Mn-salen nitrido species that, when activated with trifluoroacetic anhydride (TFAA), reacts with olefins to form new nitrogen-containing, organic compounds. 3 Carreira's process is similar to one developed by Groves some 14 years earlier where a sterically-encumbered Mn-porphyrin nitride is used to aziridinate cyclooctene. 4 In that process, the Mn-nitride is converted to a transient acylimido species that acts as an electrophile for aziridination.

More recently, Chisholm showed an intriguing example of 15N-label exchange between 15 5 MeC N and PhCN mediated by NW(O-t-Bu) 3 further showing the versatility and reactivity of 5 terminal nitrides. DSM reported the treatment of 1 NMo(N[t-Bu]Ar) 3 (Ar = 3,5-Me 2C 6H3) with 15 TFAA to form H2 NC(O)CF 3 .6 Although the process is ill-defined and results in almost total destruction of the Mo-anilide platform, it is an example of N2 -derived nitrides being used to synthesize nitrogen-containing organics.

The impetus for the work featured herein was to synthesize nitrogen containing organic species from an N2 derived nitride. Activation and cleavage of N2 had already been elucidated extensively in our group using Mo(N[t-Bu]Ar) 3 to form NMo(N[t-Bu]Ar) 3 710 For years after its initial synthesis, there was a seeming lack of reactivity of NMo(N[t-Bu]Ar) 3 (although this has since been proven to be false)."i Nitrile formation from acid chlorides has been accomplished in our group previously using a Nb-platform (Scheme 1.1). 12 Additionally, Scheme 1.1 shows another system employing an anionic Mo-nitride species with an ancillary pincer ligand that reacts metathetically with acid N Na 0 O R CI ArN]%Nb Ar[NpINA %%%%*NbN.ON N -CNAN] N -p] -RaC Ar[Np]N" N NN>

R =Me, HC=CH 2, L j t-Bu, 1-Ad, Ph Figueroa & Cummins

- DMF 0

- 0 /Na-- -- R CI

0.5 MoN-- -RCN O NMe 2 -NaCI -DMF 0I NMe2 R =Me, t-Bu, Ph

2 Sarkar, Abboud & Veige -

Scheme 1.1. Examples of nitrile synthesis from acid chlorides. chlorides to produce nitriles (Scheme 1.1). 13 Van Tamelen has also shown that when Cp2 TiCl2 is 14,15 reduced under N2 and treated with acid chlorides, the corresponding nitriles are formed.

16 Due to theoretical work done by Morokuma, hypothetical W(N[t-Bu]Ar) 3 was proposed to be better at cleaving N2 than its Mo-congener. DFT studies of M(NH 2 )3 (M = Mo, W) predicted that a tungsten(III) trisamide would have a lower activation barrier for cleavage of N2 (vs. Mo) as well as the overall reaction being more exothermic. With Morokuma's work in mind, we set out to develop a tungsten system that would cleave N2 and incorporate nitrogen atoms into organic molecules. With an isolable tungsten nitride in hand, the work highlighted in this chapter is the result of attempts to reverse engineer a tungsten species capable of activating N2 by developing reactions of the W-nitride that create nitrogen-containing organic compounds.

1.2 Background

Jane Brock Greco originally developed the synthesis of NW(N[i-Pr]Ar) 3 (1) and obtained a crystal structure of 1 (Figure 1.1). 17 Unlike the synthesis of the Mo-congener, NMo(N[t-Bu]Ar) 3, 18 1 is synthesized independently of N2 chemistry. To synthesize 1, First NW(O-t-Bu) 3 19 is treated with 0.5 equiv. of Zr(N[i-Pr]Ar) in toluene 4 at elevated temperatures giving NW(O-t-Bu)(N[i-Pr]Ar) 2 and Zr(O-t-Bu) 4 as a byproduct. The synthesis is reminiscent of a Sn ligand transfer reagent used to install a TREN variant onto Mo and V.20 This general method of installing ligands also 21 22 has precedent in our group with syntheses of NMo(S'Ad)3 and NMo(NMe 2)3 from NMo(O- 8 t-Bu) 3 .' Dr. Greco also showed that NW(NMe 2 )3 could be synthesized from NW(O-t-Bu) 3 in 17 a similar fashion using Ti(NMe 2 )4. NW(O-t-Bu)(N[i-Pr]Ar) 2 is then treated with 1 equiv. of 23 [Li-OEt 2 ][N(i-Pr)Ar] to cleanly form 1 with concomitant formation of LiO-t-Bu in 60% isolated yield as colorless crystals (Scheme 1.2).

Figure 1.1. NW(N[i-Pr]Ar) 3 (1) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A): W-N 1.669(5), W-N1 1.972(3).

Throughout the course of this work, the syntheses of starting materials were constantly optimized. The experimental section of this chapter provides improved syntheses of W2 (O-t-Bu) 6, Zr(N[i-Pr]Ar) and NW(O-t-Bu)(N[i-Pr]Ar) 4 2.

N 0.5 Zr(N[i-Pr]Ar)4 N N Li(N[i-Pr]Ar)(Et 2O) t-BuVW O-t-Bu -0.5 Zr(O-t-Bu) -LiO-t-Bu Ar[IP] W t-BuO 4 Ar[i-Pr]N N

Scheme 1.2. Synthesis of NW(N[i-Pr]Ar) 3 (1). 1.3 Nitriles Synthesized from a Tungsten Terminal Nitride

When treated with pivaloyl chloride in Et 20 at room temperature, a colorless solution of 1 rapidly turns blood-red. Analysis of the reaction mixture by 1H NMR spectroscopy shows formation of an intermediate that peaks in concentration after ca. 15 min. After ca. 2 h, the reaction of 1 with pivaloyl chloride is complete. Analysis of the reaction mixture shows near quantitative formation of pivalonitrile and a single N(i-Pr)Ar environment. The reaction mixture was triturated with n- pentane and the volatiles removed under reduced pressure several times to isolate the product of "N for (O)Cl" metathesis, (Ar[i-Pr]N) 3W(O)Cl (2, Scheme 1.3) When a concentrated solution of 2 in pentane is left to sit for several weeks, small red crystals form. An X-ray crystallographic study of those crystals unambiguously shows the W-containing product of 1 + pivaloyl chloride to be the oxochloride species, 2 (Figure 1.2).

N O O

Ar[i-Pr]N W N ' R CI N-W-CI Ar-RCN - IN[i-Pr]Ar -N\ / N [i-Pr]Ar

R = Me, HC=CH 2, i-Pr, t-Bu, 1-Ad, Ph, C6F5

Scheme 1.3. General scheme showing the reaction of NW(N[i-Pr]Ar) 3 (1) with acid chlorides to form nitriles and (Ar[i-Pr]N) 3W(O)Cl (2).

To ascertain the yield of pivalonitrile formed, the reaction of 1 with pivaloyl chloride was also carried out in C6D6 and the volatiles were vacuum transferred into an NMR tube loaded with a known quantity of ferrocene (as an internal standard). The NMR tube was flame sealed and the spectroscopic yield of pivalonitrile determined to be 97 i 5% by IH NMR.

The reaction appears to be rather general as well. Treatment of 1 with a variety of acid chlorides,

RC(O)C1 (R = Me, i-Pr, 'Ad, Ph, C6F5), results in formation of the corresponding nitriles, RCN. 1 also reacts cleanly with acryloyl chloride to form acrylonitrile thus showing some functional group tolerance. In addition, treatment of 0.5 equiv. 1 with 1 equiv. of the bis-acid chloride, adipoyl chloride (Cl(O)C(CH2) 4 C(O)Cl) results in clean formation of adiponitrile (Scheme 1.4). When treated with other carbonyl complexes, however, clean reactivity is not observed. For example, treatment of 1 with trimethylacetamide results in decomposition of the tungsten starting material (free ligand is the predominant identifiable product).

Interestingly, when NMo(N[i-Pr]Ar) 3 is treated with acid chlorides under the same conditions, no discernible reaction is observed. The increased reactivity of the W=N bond compared to the Figure 1.2. ORTEP drawing of (Ar[i-Pr]N) 3W(O)Cl (2) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): WI-01 1.698(3), Wi-Cll 2.5054(12), Ni-Wi-Cll 171.53(10).

O N 0 N-W-C 0.5 CI C1 N[i-Pr]Ar Ar[i-Pr]N *W N 0 Om Ar[i-Pr]N N \ / N[i-Pr]Ar

N 0.5

Scheme 1.4. Treatment of NW(N[i-Pr]Ar) 3 (1) with adipoyl chloride to form adiponitrile. MoaN bond is the predominant reason that a "WL 3'' type species is desirable. If a tungsten trisanilide species could be formed and proves capable of N2 scission, the incorporation of N2 derived nitrogen atoms into organic molecules would be relatively easy to accomplish.

1.4 Probing the Mechanism of Nitrile Formation

As mentioned, the reaction of 1 with acid chlorides proceeds through an intermediate to form 2 and the corresponding nitrile. Since this intermediate is readily visible on the NMR time scale (Scheme 1.5), it was apparent that it could be identified by NMR spectroscopy. By treating compound 1 with 13 a C-labeled acid chloride and tracking the reaction by 13C NMR, it was possible to glean some insight into the bonding nature of the acid chloride carbon (vide infra).

0 R o == N Ar[i-Pr]N " W N Ar[i-Pr]N -W-C Ar[i-Pr]N Ar[i-Pr]N -W-C i -r N[i-Pr]Ar N[i-Pr]Ar N[i-Pr}Ar R -NC

Ar[i-Pr]N -W-C N[i-Pr]Ar N[i-Pr]Ar

N-Ar . - 120 min H

5 min

5.4 5.3 5.2 5.1 5.0 4.9 4.8 4.7 4.6 4.5 ppm Scheme 1.5. Stack plot of the 'H NMR of NW(N[i-Pr]Ar) 3 (1) + R(O)Cl from 15 to 120 minutes. The reaction proceeds through an acylimido intermediate which decays to form (Ar[i-Pr]N)3W(O)C1 (2) and the corresponding nitrile (RCN). The area shown depicts the isopropyl methine region of the 1H NMR spectrum with the peaks for the intermediate, 2 and 1 at S = 5.34, 4.92 and 4.50 ppm, respectively.

Treatment of 1 with 13C labeled 'Ad 13C(O)Cl lent significant insight into the potential mechanism of nitrile formation form 1 (Scheme 1.6). By 13C NMR, the labeled carbon of 1Ad 13C(O)CI has a chemical shift of 8 = 179.7 ppm. Upon treatment of 1Ad13C(O)Cl with 1, an intermediate with the 13C label resonating at 8 = 185.4 ppm with 2jwc = 32 Hz is formed. Based upon the similarity of the shift to the acid chloride starting material, the intermediate is proposed to be the acylimido chloride species, (Ar[i-Pr]N)3W(=N13C[O]'Ad)C1 (3). 3 then cleanly converts to 2 and 'Ad' 3 CN (6 = 125.2 ppm). Comparison of these results to structurally characterized examples of acylimidos supports the structural assignment of 3. The structurally characterized acylimido 3 complex [WCl 2(N-2,6-i-Pr2C6H3)(NC[O]-4-MeC 6H4)(OPMe 3)(PMe 3) exhibits a carbonyl 1 C NMR signal at 8 = 169.0 ppm. 24 Similarly, the acylimido carbon of LWNC(O)Me(SPh)(CO) (L = hydrotris(3,5-dimethylpyrazol-1-yl)borate) has a 13C NMR signal at 8 = 180.5 ppm with 2jwc = 35 Hz. 2 5

N 1-Ad O II13c%:O_ "11 Ar[i-Pr]NU N N-VV-CI A r[i-Pr]N NN- rA 11 JN[i-Pr]Ar Ar[i-Pr]N-W--CI N[i-Pr]Ar

Ar[i-Pr]N_ _ + Ari-Pr]N 1 +

ppm C-N 1dc 5 = 185.4 19-Ad C 2WC = 32 Hz 6= 125.2 ppm 5 179.7 ppm 3 Scheme 1.6. Treatment of NW(N[i-Pr]Ar) 3 (1) with 1C-labeled adamantoyl chloride.

Lending further credence to this assignment is the synthesis of an isolable acylamido complex resulting from the addition of 1 equivalent of TFAA to 1. Interestingly, the product of this reaction is not (Ar[i-Pr]N)3W(O)(OC[O]CF 3) and CF3CN, but instead the acylimido trifluoroacetate (Ar[i- Pr]N) 3W(=NC[O]CF 3)(OC[O]CF 3) (4) (Scheme 1.7). The assignment of 4 was confirmed by 'H NMR, 13C NMR, '9F NMR spectroscopy and most conclusively, X-ray crystallography (Figure 1.3).

CF CF N o O O 3 O= 3 111 7 )j.O A N N 11 AriPrNJ%/'- F3 C o F3 11II neat Me 3SiCI Ar[i-Pr]N N Ar[i-Pr]N-W-OO Ar[i-Pr]N---CI A-r Ar[i-Pr]N' CF3 -Me3SiOC(O)CF 3 Ar[i-Pr]N Ar[i-Pr]N Ar[i-Pr]N

Scheme 1.7. Synthesis of acylimidos (Ar[i-Pr]N) 3W(=NC[O]CF 3)(OC[O]CF 3) (4) and (Ar[i- Pr]N)3W(=NC[O]CF 3)Cl (5).

Formation of 4 confirms the possibility of a 5-coordinate acylimido intermediate; however, it is not well understood why the TFAA does not form the corresponding nitrile. There was speculation as to whether the nitrile was not forming due to the trifluoromethyl acylimido group or because the chloride ligand had been replaced with a trifluoroacetate. To investigate this, the corresponding acylimido chloride complex was synthesized. Treatment of 4 with neat Me 3SiCl leads to clean conversion to the acylimido chloride species (Ar[i-Pr]N) 3W(=NC[O]CF 3)C1 (5).

Neither species, 4 nor 5, releases CF 3CN-even upon prolonged heating. Crystals of 5 can be grown from concentrated Me 3SiCl solutions and have been subjected to an X-ray crystallographic F2 F1 F3 .01

N1 ~ ~ '.F5 N2 02C4 W1 02F6 F4" N3

Figure 1.3. ORTEP drawing of (Ar[i-Pr]N) 3W(=NC[O]CF3)(OC[O]CF 3) (4) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): W1-N4 1.773(4), N4-C41 1.366(7) W1-02 2.162(4), C41-Ol 1.206(7), C43-03, 1.207(7), W1-N4-C41 166.6(4), N2-Wl-02 173.79(16). study (Figure 1.4). To investigate complex 5's inability to extrude CF3CN, DFT studies were performed on several related systems. 26 Whole molecule calculations of the reaction of 1 with 1 AdC(O)C1 showed that the conversion to 3 was thermodynamically favorable (AHxn = -7.8 kcal/mol). Furthermore, the conversion of 3 to 2 and 1AdCN was favorable as well (AHrxn = -9.8 kcal/mol). To investigate the lack of reactivity observed in 4 and 5, calculations were performed on the model complexes (CH3C[O]N)W(NMe 2)3C1 (6m) and (CF 3C[O]N)W(NMe2) 3Cl (7m) to CH 3CN and CF3CN, respectively. Interestingly, extrusion of CH3CN from 6m was calculated to be essentially thermo-neutral (AHrxn = 0.92 kcal/mol) whereas elimination of CF3CN from 7m was thermodynamically uphill by (AHn = 11.78 kcal/mol) suggesting that the trifluoromethyl group on the acylimido prevents nitrile formation.

F1 F2

C11

Figure 1.4. ORTEP drawing of (Ar[i-Pr]N)3W(=NC[O]CF 3)Cl (5) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Wl-N4 1.791(3), N4-C41 1.329(5), Ol-C41 1.199(5), C41-N4-W1 156.8(3).

The identity of the intermediate in nitrile formation from 1 is fairly well understood, but the overall mechanism of nitrile formation from the acylimido chloride intermediate is still ambiguous. The mechanism could consist of bending at the acylimido nitrogen to coordinate the carbonyl oxygen to tungsten, forming a 6-coordinate species that loses nitrile and forms 2. Alternatively, chloride dissociation to give a 5-coordinate transition state is also possible. As sterically encumbering N(i-Pr)Ar ligands are used to engender a low coordination number, it seemed unlikely a 6-coordinate transition state was possible. Through a serendipitous result, 6-coordinate species supported by a tungsten trisanilide platform have been isolated thus lending credibility to the former mechanism. (Synthesis of the resulting 6-coordinate species will be discussed in Chapter 2.) The isolation of 6-coordinate species suggests that the mechanism of nitrile extrusion in the treatment of 1 with acid chlorides does indeed pass through a 6-coordinate transition state (Scheme 1.8). The bending at the acylimido nitrogen position to form a 6-coordinate transition state is similar 3 to a mechanism proposed earlier in our group with the treatment of [NV(N[t-Bu]Ar) 3] with ' CS2 to form SV(N[t-Bu]Ar) 3 and [N13CS] -. 27 It is also similar to the reaction mechanism proposed by 5 15 Chisholm in catalytic 1 N-atom transfer between MeC N and PhCN mediated by NW(O-t-Bu) 3.

R 0 0

R ) CI N Ar[i-Pr]N N Ar[i-Pr]N N 11 Ar[i-Pr]N -W-Cl

N[i-Pr]Ar N[i-Pr]Ar

R-CEN

R N Ar[i-Pr]N -W-CI o ----- W N[i-Pr]Ar N[i-Pr]Ar N[i-Pr]Ar Ar[i-Pr]N NeI N[i-Pr]Ar

Scheme 1.8. Proposed 6-coordinate intermediate in the reaction of NW(N[i-Pr]Ar) 3 (1) + acid chlorides to form nitriles. 1.5 Treatment of Oxochloride with Triflic Anhydride

In an attempt to synthesize a "tungstaziridine hydride" similar to Mo(H)(1 2 -Me2C=NAr)(N[i- 23 Pr]Ar) 2, 2 was treated with triflic anhydride (Tf2O). The desired product was the chloro bis-triflate, (Ar[i-Pr]N) 3W(Cl)(OTf)2. Precedent for this type of reaction has been established within our group in the conversion of (Ar[Np]N)3NbO (Np = neopentyl) to the corresponding 28 bis-triflate, (Ar[Np]N)3Nb(OTf)2- Bis-triflate (Ar[Np]N) 3Nb(OTf)2 can be conveniently reduced 2 29 to form Nb(H)( -t-Bu(H)C=NAr)(N[Np]Ar) 2- Instead of forming a chloro bis-triflate, when 2 is treated with Tf 2O, the chloride ligand is replaced with a triflate forming the oxotriflate (Ar[i- Pr]N)3W(O)(OTf) (8, Scheme 1.9). Preliminary experiments show that 8 does not react with additional Tf2O. Although a simple method to remove the terminal oxo using Tf2O has not been obtained, work presented in Chapter 2 of this thesis unveils a successful method for removal of the oxo ligand.

O O

N-W-C Tf20 N-W-OTf

- N[i-Pr]Ar -TfCI - N[i-Pr]Ar \ / N[i-Pr]Ar \ / N[i-Pr]Ar

O OTf

r[Np] Nb Tf20 Ar[Np]Nlit,''Nb- N Ar[Np]N N Ar[Np]N" / OTf Figueroa & Cummins

Scheme 1.9. Synthesis of (Ar[i-Pr]N) 3W(O)(OTf) (8) and formation of (Ar[Np]N) 3Nb(OTf)2 from ONb(N[Np]Ar) 3-

1.6 Synthesis of a Terminal Phosphide, PW(N[i-Pr]Ar)3 (9)

One unique method to remove the terminal oxo ligand from 2 is to treat the material with 30 3 one equivalent of [Na(OEt 2)][PNb(N[Np]Ar) 3]. ' ' Compound 2 reacts cleanly with the anionic, terminal-phosphide complex [Na(OEt2)][PNb(N[Np]Ar) 3] to produce PW(N[i-Pr]Ar) 3 (9) with concomitant formation of ONb(N[Np]Ar) 3 and NaCl as byproducts (Scheme 1.10). Phosphide 9 is clearly identified using 3 1P NMR spectroscopy by its downfield shift and coupling to 1'3W (8 = 1021, 1Jwc = 193 Hz).32 33 The similar solubility properties of phosphide 9 to ONb(N[Np]Ar) 3 hamper the clean isolation of 9. Despite the inability to isolate the product cleanly, compound 9 was fully characterized spectroscopically. Suitable crystals for an X-ray crystallographic study were grown from a mixture of 9 and ONb(N[Np]Ar) 3 in Et2O which had been stored at -35 'C. Although both phosphide and oxo crystallized from the mixture, it was trivial to separate and select a yellow crystal of 9 instead of an orange crystal of ONb(N[Np]Ar) 3 by using a microscope and the Pasteur method.34-36 The solid state structure of phosphide 9 is similar in overall appearance to the analogous pnictogen congener, nitride 1 (Figure 1.5). The structure of 9 is that of a pseudo- tetrahedral complex with a short W-P bond (2.119(3) A).

P ( ||| Na Ar[Np]N"' NN 4V Ar[Np]N P

N-- CI Ar[i-Pr]N IN + NbN N[i-Pr]Ar Et2 0, -35 to 25 oC Ar[iPr]N Ar[Np]N N[i-Pr]Ar - NaCl Ar[Np]N

31P NMR: 6 =1021 ppm (Oc= 193 Hz) Scheme 1.10. Synthesis of PW(N[i-Pr]Ar) 3 (9)-

In an effort to cleanly isolate phosphide 9, a mixture of 9 and ONb(N[Np]Ar) 3 was treated with one equivalent of Tf2O. Remarkably, the phosphide did not react with the strongly electrophilic triflic anhydride-the Tf 20 was completely consumed by ONb(N[Np]Ar) 3 to form (Ar[Np]N) 3Nb(OTf)2. The stark solubility differences between (Ar[Np]N) 3Nb(OTf)2 and phosphide 9 enabled the clean isolation of the desired compound-a feat which was accomplished by Dr. Alexander Fox. 37

1.7 Conclusions

The work presented in this chapter shows an isovalent N for (O)Cl exchange mediated by a tungsten trisanilide platform. Treatment of 1 with acid chlorides leads to clean formation of the corresponding nitriles as well as oxochloride 2. The reaction appears to be general for acid chlorides and exhibits functional group tolerance that needs to be further explored.

The drawback of this system is that the nitrogen atom being incorporated into organic frameworks is not derived from N2. Future work should focus on developing a tungsten trisanilide system capable of cleaving dinitrogen to form a terminal nitride. The ability to incorporate the Figure 1.5. ORTEP drawing of PW(N[i-Pr]Ar)3 (9) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Wl-P1 2.119(3), W-N 1.939(9) (avg.), P1-Wi-N 104.4(3) (avg.). nitrido nitrogen of 1 is already in place. Further research towards a "tungstaziridine hydride" is presented in later chapters, but as of yet, this molecule remains unsynthesized. In addition, exploration of the reactivity of 1 with other small molecules and organic reagents could lead to interesting and potentially useful applications.

1.8 Experimental Section

1.8.1 General Considerations

Unless stated otherwise, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere. 19 23 38 NW(O-t-Bu) 3, [Li-OEt2][N(i-Pr)Ar] and WC14 were prepared as previously published. 1 13 3 9 3 Ad C(O)Cl was prepared as described in the literature using 1 C0 2 purchased from Cambridge Isotope Laboratories. ZrCl 4 was purchased from Strem Chemicals, Inc. and used as received. t- BuC(O)Cl was purchased from Aldrich and distilled under N2. Trifluoroacetic anhydride (TFAA) and LiO-t-Bu were purchased from Aldrich and used as received. Diethyl ether, n-pentane, and toluene were dried and deoxygenated by the method of Grubbs. 40 THF was distilled from purple

Nalbenzophenone and collected under N2.C6D6 was degassed and dried over 4 A molecular sieves. Other chemicals were purified and dried by standard procedures or were used as received. Celite, alumina and 4 A molecular sieves were dried in vacuo overnight at a temperature above 200 C. 'H, 13C, 19F and 31P NMR spectra were recorded on Varian Mercury-300, Varian INOVA-500 or Bruker AVANCE-400 spectrometers. 1H and 13C chemical shifts are reported with respect to 9 internal solvent (C6D6, 8 = 7.16 and 129.39 ppm, respectively). 1 F chemical shifts are reported 31 with respect to an external reference (CFCl 3, 0.0 ppm). P chemical shifts are reported with respect to an external reference (H3PO4, 0.0 ppm). X-ray data collections were carried out on a Siemens Platform three-circle goniometer with a CCD detector using Mo-Ku radiation, k = 0.71073 A. The data were processed utilizing the program SAINT supplied by Siemens Industrial Automation, Inc. Electron Impact (EI) mass spectra (MS) were recorded on a Bruker Daltronics APEXIV 4.7 FT-ICR Mass Spectrometer. Combustion analyses were performed by either H. Kolbe Mikroanalytisches Laboratorium, Mulheim an der Ruhr, Germany or Midwest Microlabs LLC, Indianapolis, IN.

1.8.2 One Pot Synthesis of W2(0-t-BU)6

WCl 4 (18.67 g, 57.3 mmol, 1.00 equiv) was loaded into a 300 mL RB flask and suspended in ca. 150 mL THF. The grey-blue suspension was cooled in the cold well to -108 'C. A 1%sodium amalgam (130.5 g Hg, 1.318 g Na, 57.3 mmol, 1.00 equiv) was prepared and added directly to the which WCl 4 slurry. The mixture was allowed to warm to 25 'C and vigorously stirred for 3 h at time the reaction mixture had taken on a dark-green color. LiO-t-Bu (13.77 g, 172.0 mmol, 3.00 equiv) was added to the green reaction mixture in three portions over 30 min. The reaction mixture was then stirred for 12 h. After 12 h, the reaction mixture consisted of a blood-red solution atop mercury and precipitated LiCl. The mixture was decanted off of the mercury, filtered through Celite and the volatiles removed under reduced pressure. The red product was extracted with pentane and again filtered through Celite to remove LiCl and NaCl. A very large frit (150 or 350 mL) should be used for this step as the frit tends to clog. The red pentane solution was stored at -35 'C overnight and the product isolated as red crystals in three crops (18.37 g, 22.8 mmol, 79.5% w.r.t. W).

1.8.3 Improved Synthesis of Zr(N[i-Pr]Ar)4

[Li.OEt2][N(i-Pr)Ar] (25.00 g, 102.7 mmol, 4.06 equiv) was suspended in ca. 150 mL Et2O and frozen in the cold well. Solid ZrCl4 (5.90 g, 25.3 mmol, 1.00 equiv) was added to the thawing ethereal slurry. The mixture was allowed to warm to 25 'C and stirred for 12 h. The next day, a homogeneous yellow solution was present and the volatiles were removed under reduced pressure. The product was extracted with ca. 500 mL pentane and the resulting extract was filtered through a pad of Celite. The filtrate was reduced in volume to ca. 150 mL (at which point a light-yellow solid began to precipitate) and frozen in the cold well. The mixture was allowed to partially thaw and solid Zr(N[i-Pr]Ar) 4 was isolated on a large frit, washed with thawing Et20 and dried under reduced pressure (15.95 g, 21.55 mmol, 85.2%). ' H NMR (300 MHz, C6D6): 6 6.80 (s, 8H, ortho), 6.63 (s, 4H, para), 4.17 (septet, 4H, i-Pr methine, J = 6.6 Hz), 2.26 (s, 24H, ArMe), 1.13 (d, 24H, 3 i-Pr methyl, J = 6.6 Hz) ppm. C NMR (75 MHz, C6D6): 8 148.91 (ipso), 138.39 (meta), 125.55 (para), 125.06 (ortho), 52.60 (i-Pr methine), 23.91 (methyl), 22.22 (methyl) ppm. Anal. Caled. for

C44H64N 4Zr: C, 71.38; H, 8.73; N, 7.57. Found: C, 71.80; H, 8.65; N, 7.58.

1.8.4 Improved Synthesis of NW(O-t-Bu)(N[i-Pr]Ar) 2

NW(O-t-Bu) 3 (11.90 g, 28.52 mmol, 1.00 equiv) and Zr(N[i-Pr]Ar) 4 (10.56 g, 14.26 mmol, 0.50 equiv) were loaded into a 300 mL thick-walled glass vessel equipped with a PTFE stopcock and stirbar. The solids were suspended in ca. 200 mL of toluene and the headspace in the vessel was evacuated under reduced pressure. The stopcock was sealed and the vessel was removed from the glovebox. The reaction mixture was heated to 55 'C overnight. After 12 h, the reaction mixture had become a homogeneous, golden-brown solution. The volatiles were removed under reduced pressure and the vessel was brought back into the glovebox. The product was extracted with ca. 200 mL Et20 and the solution filtered through Celite. The filtrate was reduced in volume to ca. 100 mL under reduced pressure and placed in the cold well long enough for the ether to begin to freeze.

A snow-white precipitate had formed with a layer of yellow, frozen Zr(O-t-Bu) 4 atop it. The flask was allowed to warm slightly to redissolve the unwanted Zr-byproduct and the solids were collected on a large frit. The snow white precipitate was washed with thawing Et2O (4 * 20 mL) and dried under reduced pressure (13.45 g, 22.58 mmol, 79.2%). 'H NMR (300 MHz, C6D6 ): 6 6.44 (s, 4H, ortho), 6.36 (septet, 2H, para, J = 0.6 Hz), 4.24 (quintet, 2H, CH(CH 3)2, J = 6.6 Hz), 1.98 (d, 12H,

ArCH 3, J = 0.6 Hz), 1.72 (d, 6H, CH(CH 3)(CH 3 ), J = 6.6 Hz), 1.66 (s, 9H, OC(CH 3 )3 ), 1.55 (d, 6H, CH(CH 3)(CH 3), J = 6.6 Hz) ppm. 13C NMR (75 MHz, C6D6 ): 6 138.78 (ipso), 128.68 (meta), 128.00 (para), 122.78 (ortho), 78.84 (OC(CH 3)3), 63.23 (CH(CH 3 2), 32.38 (OC(C(CH 3)3), 27.80 (CH(CH )(CH )), 26.91 (CH(CH 3 3 3)(CH3)), 21.60 (ArCH 3) ppm. Anal. Calcd. for C26H4 1N30W: C, 52.44; H, 6.94; N, 7.06. Found: C, 52.06; H, 7.34; N, 6.95.

1.8.5 Synthesis of NW(N[i-Pr]Ar)3 (1)

The synthesis of 1 was adapted from J. B. Greco's Ph.D. Thesis17 and published separately 4 1. A solution of [Li.OEt 2][N(i-Pr)Ar](1.749 g, 7.19 mmol, 1.15 equiv) in Et2O (120 mL) was added to solid NW(O-t-Bu)(N[i-Pr]Ar) 2 (3.721 g, 6.25 mmol, 1.00 equiv) and stirred for 12 h, after which time the volatiles were removed under reduced pressure. The solid was re-dissolved in 20 mL

Et20, filtered through a bed of Celite and cooled to -35 'C to yield colorless crystals (2.44 g, 3.72 mmol, 60%). 1H NMR (300 MHz, C6D6): 6 6.47 (s, 3H, para), 6.32 (s, 6H, ortho), 4.50 (septet,

3H, CH(CH 3 )2, J = 6.6 Hz), 1.93 (s, 18H, ArCH 3 ), 1.71 (d, 18H, CH(CH 3) 2, J = 6.6 Hz) ppm. 3 1 C NMR (75 MHz, C6D6): 6 151.25 (ipso), 138.55 (meta), 124.69 (para), 124.53 (ortho), 65.72

(CH(CH3 )2, 27.10 (CH(CH 3)2, 21.64 (ArCH3) ppm. Anal. Calcd. for C33H48N4W: C, 57.88; H, 7.08; N, 8.18. Found: C, 57.90; H, 7.44; N, 8.16. 1.8.6 Reaction of NW(N[i-Pr]Ar)3 (1) with pivaloyl chloride

Pivaloyl chloride (9.1 uL, 0.074 mmol, 1.00 equiv) was added with a microliter syringe to a solution of NW(N[i-Pr]Ar) 3(5 1.0 mg, 0.0745 mmol, 1.01 equiv) in C6D6 and stirred for 3.5 h. The volatiles were vacuum-transferred into a NMR tube over a known amount of ferrocene (13.4 mg, 0.0720 mmol, 0.97 equiv) to be used as an internal standard. 1H NMR (300 MHz, C6D6) 8 4.01 (s, 1OH, ferrocene), 0.77 (s, 9H, t-BuCN) ppm. The NMR spectrometer was shimmed and locked, a spectrum of the sample was taken to insure purity, and the system was allowed to relax for 10 min to insure accurate integrals. After 10 min, a spectrum was acquired with a single pulse. The ferrocene and pivalonitrile resonances were integrated and found to be in a ratio of 10.00:9.01 (Cp2Fe:t- BuCN). Based upon the known amount of ferrocene, the yield was calculated to be 97 ± 5 %. The 'H NMR, 3C NMR and FT-IR spectra of the vac-transferred pivalonitrile matched those of commercially available pivalonitrile. A drop of commercially available pivalonitrile added to the sample introduced no new resonances into the 1H NMR spectrum. A 'H NMR spectrum of the non-volatiles confirmed the presence of (Ar[i-Pr]N) 3W(O)Cl.

1.8.7 General Synthesis of Nitriles from NW(N[i-Pr]Ar)3(1)

A solution of acid chloride (ca. 2.05 mmol, 1.02 equiv) in 2 mL Et20 is added to a 10 mL Et2O solution of NW(N[i-Pr]Ar) 3 (1.369 g, 2.00 mmol, 1.00 equiv) and stirred for 3 h. An aliquot of the reaction mixture can be taken, stripped to dryness and the reaction checked for completeness by 'H

NMR in C6D6. The i-Pr methine region is the most convenient location in the proton spectrum to observe. A septet at 6 = 4.50 ppm indicates presence of 1. A septet at 6 = 4.92 ppm corresponds to 2. A septet at 8 = ca. 5.3 ppm is due to acylimido intermediate. When only the septet at 8 = 4.92 ppm remains, the reaction is complete. Volatile nitriles can be isolated from the reaction mixture by vacuum transfer. 2 can be isolated as described (vide infra) or the reaction mixture can be stripped to dryness under reduced pressure, triturated with pentane and 2 used without further purification.

1.8.8 Synthesis of (Ar[i-Pr]N)3W(O)CI (2)

A solution of pivaloyl chloride (261 mg, 2.16 mmol, 1.06 equiv) in 2 mL Et20 was added to a 10 mL Et 2O solution of NW(N[i-Pr]Ar) 3 (1.40 g, 2.04 mmol, 1.00 equiv) and stirred for 3 h. After this time, the solvent was removed under reduced pressure. The blood-red, oily solid was dissolved in minimum of pentane, filtered through Celite and the solution cooled to -35 'C overnight to yield an orange-red precipitate of spectroscopically pure material in two crops (500 mg and 515 mg, respectively, 1.41 mmol, 69 %). 'H NMR (300 MHz, C6D6): 6 6.74 (s, 6H, ortho), 6.58 (s, 3H, para), 4.92 (septet, 3H, CH(CH 3)2, J = 6.6 Hz), 2.06 (s, 18H, ArCH 3), 1.26 (d, 18H, CH(CH 3)2, J 3 = 6.6 Hz) ppm. 1 C NMR (75 MHz, C6D6): 6 149.60 (ipso), 138.17 (meta), 126.17 (ortho), 63.32 (CH(CH3)2), 22.82 (methyl), 21.78 (methyl) ppm. Para carbon was not observed. Anal. Calcd. for

C33 H48N30C1W: C, 54.88; H, 6.71; N, 5.82. Found: C, 54.85; H, 7.09; N, 5.84.

3 1.8.9 Reaction of NW(N[i-Pr]Ar)3 (1) with 'Ad1 C(O)CI

3 'Ad C(O)Cl (6.0 mg, 0.030 mmol, 1.03 equiv) was dissolved in 0.5 mL C6D6 and added to solid NW(N[i-Pr]Ar) 3 (20.0 mg, 0.029 mmol, 1.00 equiv). The solution was loaded into an NMR tube and monitored by 'H NMR and 13C NMR. Spectra were taken every ca. 5 min for 1 h. The labeled carbon resonance was observed to decrease over time at 8 = 179.7 ppm ('Ad13C(O)Cl), increase and subsequently decrease at 6 = 185.4 ppm (3, 2jwC = 32 Hz), and increase at 6 = 125.2 ppm ('Ad13CN). The resonances assigned to the acid chloride and nitrile were identical to authentic samples. The resonance at 6 = 185.4 ppm was assigned to 3 due to its chemical shift and two-bond coupling to 14.4% 183W.

1.8.10 Synthesis of (Ar[i-Pr]N)3W(=NC[O]CF 3)(OC[O]CF 3) (4)

TFAA (47.0 pL, 0.333 mmol, 1.01 equiv) was added with a microliter syringe to a solution of

NW(N[i-Pr]Ar) 3 (226 mg, 0.330 mmol, 1.00 equiv) in 7 mL Et20. The solution immediately turned blood-red and was stirred for 30 min. Volatiles were removed under reduced pressure and the red, greasy solid was dissolved in 2 mL of pentane. The solution was filtered through Celite and pentane was removed under reduced pressure, whereupon 4 was obtained as a red, oily solid (257 mg, 0.287 mmol, 87%). Complex 4 can be purified by crystallization from an extremely concentrated Et 2O solution at -35 'C, although its high lipophilicity diminishes the overall yield. ' H NMR (300 MHz,

C6D6): 8 6.71 (s, 6H, ortho), 6.67 (s, 3H, para), 5.21 (septet, 3H, CH(CH 3)2, J = 6.6 Hz), 2.15 (s, 3 18H, ArCH 3), 1.16 (d, 18H, CH(CH 3)2, J = 6.6 Hz) ppm. 1 C NMR (125MHz, C6D6): 6 164.38 (q, carbonyl or acetate, 2 JCF = 39 Hz), 159.12 (q, carbonyl or acetate, 2 JCF = 38 Hz), 148.57 (ipso),

138.57 (meta), 129.54 (para), 126.61 (ortho), 117.06 (q, CF3, 'JCF = 290 Hz), 118.79 (q, CF3, 'JCF = 290 Hz), 9 67.83 (CH(CH 3 )2, 23.38 (methyl), 21.63 (methyl) ppm. ' F NMR (282 MHz, C6D6): 6 -73.1 (s, 3F, CF 3 ), -76.7 (s, 3F, CF 3) ppm. Anal. Calcd. for C37H48N4 0 3F6W: C, 49.67; H, 5.42; N, 6.26. Found: C, 47.89; H, 6.41; N, 5.88.

1.8.11 Synthesis of (Ar[i-Pr]N)3W(=NC[O]CF 3)CI (5)

4 (525 mg, 0.767 mmol) was dissolved in a minimum of Me 3SiCl ( 5 mL) and the resulting red solution was stirred for 5 min, filtered through Celite and the filtrate cooled to -35 'C overnight. 5 was obtained as a red precipitate which was washed with cold pentane and dried under reduced pressure (160 mg, 0.196 mmol, 25.5%). X-ray quality crystals can also be grown by following the above procedure using smaller amounts of 4 (ca. 100 mg) in more dilute solutions with Me 3SiC1 as the solvent. 'H NMR (500 MHz, C6D6): 6 6.57 (s, 3H, para), 6.52 (s, 6H, ortho), 5.33 (septet, 13 3H, i-Pr methine), 2.05 (s, 18H, ArCH 3), 1.14 (d, 18H, i-Pr methyl) ppm. C NMR (100 MHz,

C6 D6 ): 6 149.3 (ipso), 138.2 (meta), 129.1 (para), 126.3 (ortho), 65.6 (i-Pr methine), 23.0 (methyl), 9 21.6 (methyl) ppm. ' F NMR (376 MHz, C6D6 ): 6 -72.7 (s, 3F, CF 3) ppm. Anal. Caled. for 6.16; N, 6.73. C35 H48 ClF 3N4 0W: C, 51.45; H, 5.92; N, 6.86. Found: C, 50.95; H,

1.8.12 Synthesis of (Ar[i-Pr]N)3W(O)(OTf) (8)

2 (250 mg, 0.346 mmol, 1.00 equiv) was dissolved in 3 mL Et2 O and the solution was stirred. Tf2O

(104 mg, 0.369 mmol, 1.06 equiv) was dissolved in I mL Et 20 and the solution added to the solution of 2. The reaction mixture immediately changed from a blood-red solution to an orange solution with yellow precipitate. The reaction mixture was stirred for 15 minutes, after which time the volatiles were removed under reduced pressure. The yellow solid residue was suspended in pentane and isolated by filtration on a fritted glass funnel. The yellow solids were washed with pentane and dried under reduced pressure (222 nig, 0.266 mmol, 77%). 'H NMR (300 MHz, C6D6 ): 6 6.53 (s, 3H, para), 6.27 (s, 6H, ortho), 4.85 (septet, 3H, i-Pr methine), 1.98 (s, 18H, ArMe), 1.17 (d, 9 18H, i-Pr methyl) ppm. " F NMR (282 MHz, C6D6 ) 6 -77.74 (s, 3F, OTf) ppm. Anal. Calcd. for

C34 H48N30 4SF 3W: C, 48.87; H, 5.79; N, 5.03. Found: C, 48.15; H, 6.23; N, 4.73.

1.8.13 Synthesis of PW(N[i-Pr]Ar)3 (9)

As initial syntheses of 9 undertaken by the author did not result in clean isolation of the desired product, the preparation accomplished by Dr. Alexander Fox is cited. 37

1.9 DFT Calculations

All calculations were carried out using ADF 2004.01 from Scientific Computing and Modeling (http: //www. scm. com). 26,42 In all cases the LDA functional employed was that of Vosko, Wilk and Nusair (VWN) 43 while the GGA part was handled using the functionals of Becke and Perdew (BP86).44'45 In addition, all calculations were carried out using the Zero Order Regular Approximation (ZORA) for relativistic effects. 46,47 In all cases the basis sets were triple-zeta with two polarization functions (TZ2P) as supplied by ADF. Frozen core approximations were utilized according to the following atom types: F, N, C and 0: 1s frozen; Cl: core frozen through and including 2p; W: core frozen through and including 4f. Calculations were carried out on a four- or an eight-processor Quantum Cube workstation from Parallel Quantum Solutions

(http: \www . pqs-chem. com). All results reported are with reference to fully optimized geometries 48 49 with no imaginary frequencies. , 1.10 Crystallographic Structure Determinations

1.10.1 General Considerations

X-ray data collections were carried out on a Siemens Platform three-circle diffractometer mounted with a CCD or APEX-CCD detector and outfitted with a low-temperature, nitrogen-stream aperture. Graphite-monochromated Mo-Ka radiation (X = 0.71073 A) was used in all cases. All software for diffraction data processing and crystal-structure solution and refinement are contained in the SHELXTL (v6.14) program suite (G. Sheldrick, Bruker XRD, Madison, WI). 50

1.10.2 X-ray crystal structure of (Ar[i-Pr]N)3W(O)CI (2)

Inside the glove box, crystals of 2, obtained from an extremely saturated pentane solution stored at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A red block of approximate dimensions 0.38 * 0.30 * 0.28 mm3 was selected and mounted on a glass fiber. A total of 12310 reflections (-9 < h < 10, -18 < k < 16, -21 < 1 < 21) were collected at 193(2) K in the 0 range of 2.30 to 22.50', of which 4287 were unique (Ri, = 0.0256). The structure was solved by direct methods (SHELXTL V6.14, G. Sheldrick, Bruker XRD, Madison, WI) 50 in conjunction with standard difference Fourier techniques. The systematic absences in the diffraction data are uniquely consistent with the assigned space group of P212121 (Flack parameter = -0.016(7)). All non- hydrogen atoms were refined anisotropically and hydrogen atoms were placed in calculated (dc-H = 0.96 A) positions. An empirical absorption correction ($-scans) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.2885 and 0.2203, respectively. The residual peak and hole electron density were 0.243 and -0.314 e.A -3, respectively. The least squares refinement converged normally with residuals of R1 = 0.0175, wR2 = 0.0422 based upon I > 2-(I), and GOF = 1.116 (based on F 2). No extinction coefficient was applied to the refinement.

Crystal and refinement data: formula C33H48N30C1W, space group P212 121, a = 9.8866(10) A, b = 16.8619(16) A, c = 19.8088(19) A, V = 3302.3(6) A3, Z = 4, p = 3.607 mm-1 , Dcale = 1.452 3 g-cm- , F(000) = 1464, R1 (based on F) = 0.0181, and wR 2 (based on F) = 0.0413.

1.10.3 X-ray crystal structure of (Ar[i-Pr]N)3W(=NC[0]CF 3)(OC[0]CF3) (4)

Inside the glovebox, crystals of 4, obtained from a saturated Et20 solution stored at -35 'C, were coated with Paratone N (an Exxon product) on a microscope slide. A red rod of approximate dimensions 0.38 * 0.18 * 0.14 mm 3 was selected and mounted on a glass fiber. A total of 14095 reflections (-12 < h < 12, -16 < k < 17, -17 < l < 21) were collected at 193(2) K in the 0 range of 1.24 to 22.000, of which 9401 were unique (Rin = 0.566). The structure was solved by calculated Super-Sharpe Patterson vectors employing the Single Isomorphous Replacement (SIR) method as supplied in the SHELXTL program suite. 50 Acceptable solutions were obtained for the PI option in the triclinic setting using the XM program routing (SHELXTL V6.14, G. Sheldrick, Bruker XRD, Madison, WI). No acceptable solutions could be obtained using traditional direct methods or Patterson techniques. Two crystallographically independent, but chemically equivalent molecules are present in the asymmetric unit. All non-hydrogen atoms were refined anisotropically and hydrogen atoms were placed in calculated (dc-H = 0.96 A) positions. An empirical absorption corrections (SADABS) was applied to the diffraction data. The residual peak and hole electron density were 1.760 and -1.951 e-A~3, respectively. These large values are most likely due to the scattering ability of the heavy tungsten atom. The least squares refinement converged normally with 2 residuals of R1 = 0.0475, wR2 = 0.1235 based upon I > 27(I) and GOF = 1.017 (based on F ).

Crystal and refinement data: formula C37H48F6N40 3W, space group P!, a 12.0919(10) A, b = 16.4768(13) A, c = 20.0774(16) A, x = 91.902(2)0, P = 100.9770(10)0, y = 90.6950(10)0, V 3 1 3 = 3924.1(5) A , Z = 4, p= 3.011 mm~ , Dcaic = 1.514 g.cm- , F(000) = 1800, R1 (based on F) = 2 0.0516, and wR 2 (based on F ) = 0.1268.

1.10.4 X-ray crystal structure of (Ar[i-Pr]N)3W(=NC[O]CF 3)CI (5)

Inside the glovebox, crystals of 5, obtained from a saturated Me3SiCl solution stored at -35 'C, were coated with Paratone N oil (an Exxon Product) on a microscope slide. A dark red plate of approximate dimensions 0.14 * 0.08 * 0.02 mm 3 was selected and mounted on a glass fiber. A total of 73315 reflections (-24 < h < 23, 0 < k < 13, 0 < I < 25) were collected at 100(2) K using $- and o-scans in the 0 range of 1.83 to 28.280, of which 8376 were unique (Rint = 0.0907). The structure was solved by Patterson methods using SHELXTL50 and refined against F 2 on all data by full-matrix least squares with SHELXTL.50 The systematic absences in the diffraction data were uniquely consistent with the assigned space group of P21/n. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.9352 and 0.6477, respectively. The residual peak and hole electron density were 1.302 and -1.122 e-A-3, respectively.

The least squares refinement converged normally with residual values of R1 = 0.0359 for I > 2a(I), 2 wR 2 = 0.0791 for all data and GOF = 1.075 (based on F ).

Crystal data: formula C35H48F3N40ClW, space group P21/n, a = 18.1298(6) A, b = 10.4918(3) 3 A, emphc = 19.4084(5) A, p = 107.0410(10)0, V = 3529.67(18) A , Z = 4, p = 3.397 mmi, Dcalc = 1.538 g-cm-3 , F(000) = 1648.

1.10.5 X-ray crystal structure of PW(N[i-Pr]Ar)3 (9)

Inside the glovebox, crystals of 9, obtained from a saturated diethyl ether solution stored at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A yellow block of approximate dimensions 0.12 * 0.08 * 0.08 mm3 was selected and mounted on a glass fiber. A total of 114712 reflections (-13 < h < 31, 0 < k < 14, 0 < I < 76) were collected at 100(2) K using 0- and o-scans in the 0 range of 0.71 to 28.34', of which 16406 were unique (Rint = 0.0824). The structure was solved by direct methods using SHELXS 50 and refined against F2 on all data by full-matrix least squares with SHELXTL. 50 The systematic absences in the diffraction data are consistent with the assigned space group of P21212 1. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.7628 and 0.6733, respectively. The residual peak and hole electron density were 2.606 and -3.683 eA -3, respectively. The least squares refinement converged normally with residuals of R, = 0.0741 for I > 27(I), wR2 = 0.1458 for all data, and GOF = 1.275 (based on F 2).

Crystal data: formula C33H4gN3PW, space group P212121, a = 10.4856(7) A, b = 10.9071(7) A, c = 57.616(4) A, a y= 9 0 ', V = 6589.4(7) A3. Z = 8, p = 3.579 mm 1, Dcale = 1.414 g.cm-3, F(000) = 2848. Table 1.1. Crystallographic Data for (Ar[i-Pr]N),W(O)C1(2), (Ar[i-Pr]N),W(=NC[O]CF3)(OC[O]CF3) (4) and (Ar[i-Pr]N)3W(=NC[0]CF3)C1 (5) 2 4 5 Reciprocal Net code / CCDC code 03165 / ABAKOE 04019 / ABAKIY 04171 / AGIPUC Empirical formula, FW (g/mol) C H F N 0 W, 894.64 C33H48C1N30W, 722.04 37 4 8 6 4 3 C35 H4 8C1F3N4 0W, 817.07 Color / Morphology Red / Block Red /Rod Dark red / Plate Crystal size (mm3 ) 0.38 * 0.30 * 0.28 0.38 *0.18 * 0.14 0.14* 0.08 * 0.02 Temperature (K) 193(2) 193(2) 100(2) 1 Crystal system, Space group Orthorhombic, P2 1212 Triclinic, P1 Monoclinic, P2 1 /n Unit cell dimensions (A, 0) a = 9.8966(10), a 90 a = 12.0919(10), a = 91.902(2) a = 18.1298(6), a = 90 b = 16.8619(16), =90 b = 16.4768(13), P = 100.9779(10) b = 10.5918(3), P = 107.0410(10) c = 19.8088(19), 7=90 c = 20.0774(16), y = 90.6950(10) c = 19.4084(5),y = 90 Volume(A 3) 3302(6) 3924.1(5) 3529.67(18) Z 4 4 4 Density (calc., Mg/m3 ) 1.452 1.514 1.538 Absorption coefficient (mm-') 3.607 3.011 3.397 F(000) 1464 1800 1648 Theta range for data collection (0) 2.30 to 22.50 1.24 to 22.00 1.83 to 28.28 Index ranges -9

Final R indicesb [I> 2(I)] R1 = 0.0175, wR 2 = 0.0411 R1 = 0.0475, wR2 = 0.01235 R1 = 0.0359, wR2 = 0.0727 R indicesb (all data) R1 = 0.0181, wR 2 = 0.0413 R = 0.0516, wR2 = 0.1268 R1 = 0.0522, wR2 = 0.0791 Largest diff. peak and hole (e -l- 3 ) 0.243 and -0.314 1.760 and -1.951 1.302 and -- 1.122

2 . _ 1 .P __ 2F +max(F,O) aGoo-F [ E[w( _ )2] 2 b R 1 = E |F|-Fc| ;wR 2 = w(F - ,-F)2 I (n--p) ' I|o L[w(F,,)']I =G(F,)+(aP)2+bP' 3 -3 Table1.2. Crystallographic Data for PW(N[i-Pr]Ar),(9) 9 Reciprocal Net code / CCDC code 05140 / YEVTID Empirical formula, FW (g/mol) C33H48N3PW, 701.56 Color / Morphology Yellow / Block Crystal size (mm3) 0.12*0.08 *0.08 Temperature (K) 100(2) 2 Crystal system, Space group Orthorhombic, P2 1 121 Unit cell dimensions (A, 0) a = 10.4856(7), a = 90 b = 10.9071(7), = 90 c = 57.616(4), y= 90 Volume(A 3) 6589.4(7) Z 8 Density (calc., Mg/m3) 1.414 Absorption coefficient (mm'1) 3.579 F(O0O) 2848 Theta range for data collection (0) 0.71 to 28.34 Index ranges -13 2a(I)] R = 0.0741, wR2 = 0.1434 R indicesb (alldata) R1 = 0.0808, wR2 = 0. 1458 Largest diff. peak and hole (e -E-3) 2.606 and -3.683

2 2 2 2 aG F [w(b F )]1 D _ £IoIc _ [wF _-F ) 12 aGooF =R -[+F R1= () F ; wR2,4-

_ 1 .P _ 2F,2+ max (F,,0) 2 (F,) + (aP) 2+bP' 3 1.11 References

[1] Dehnicke, K.; Strahle, J. Angew. Chem. Int. Ed. Eng. 1992, 31, 955-978. [2] Eikey, R. A.; Abu-Omar, M. M. Coord. Chem. Rev. 2003, 243, 83-124. [3] DuBois, J.; Tomooka, C. S.; Hong, J.; Carreira, E. M. Acc. Chem. Res. 1997, 30, 364-372. [4] Groves, J. T.; Takahashi, T. J. Am. Chem. Soc. 1983, 105, 2073-2074. [5] Chisholm, M. H.; Delbridge, E. E.; Kidwell, A. R.; Quinlan, K. B. Chem. Commun. 2003, 126-127. [6] Henderickx, H.; Kwakkenbos, G.; Peters, A.; van der Spoel, J.; de Vries, K. Chem. Commun. 2003, 2050-2051. [7] Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861-863. [8] Laplaza, C. E.; Johnson, M. J. A.; Peters, J. C.; Odom, A. L.; Kim, E.; Cummins, C. C.; George, G. N.; Pickering, I. J. J.Am. Chem. Soc. 1996, 118, 8623-8638. [9] Peters, J. C.; Cherry, J. P. F.; Thomas, J. C.; Baraldo, L.; Mindiola, D. J.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc. 1999, 121, 10053-10067. [10] Tsai, Y. C.; Cummins, C. C. Inorg. Chim. Acta 2003, 345, 63-69. [11] Curley, J. J.; Murahashi, T.; Cununins, C. C. J. Am. Chem. Soc. 2006, 128, 7181-7183. [12] Figueroa, J. S.; Piro, N. A.; Clough, C. R.; Cummins, C. C. J. Am. Chem. Soc. 2006, 128, 940-950. [13] Sarkar, S.; Abboud, K. A.; Veige, A. S. J. Am. Chem. Soc. 2008, 130, 16128-16129. [14] Van Tamelen, E. E. Acc. Chem. Res. 1970, 3, 361-367. [15] Van Tamelen, E. E. J. Am. Chem. Soc. 1970, 92, 5253-5254. [16] Cui, A.; Musaev, D. G.; Svensson, M.; Sieber, S.; Morokuma, K. J. Am. Chem. Soc. 1995, 117, 12366-12367. [17] Greco, J. B.; Ph.D. thesis; Massachusetts Institute of Technology; 2001. [18] Cherry, J. P. F.; Stephens, F. H.; Johnson, M. J. A.; Diaconescu, P. L.; Cummins, C. C. Inorg. Chem. 2001, 40, 6860-6861. [19] Schrock, R. R.; Listemann, M. L.; Sturgeoff, L. G. J. Am. Chem. Soc. 1982, 104, 4291-4293. [20] Plass, W.; Verkade, J. G. J.Am. Chem. Soc. 1992, 114, 2275-2276. [21] Agapie, T.; Odom, A. L.; Cummins, C. C. Inorg. Chem. 2000, 39, 174-179. [22] Johnson, M. J. A.; Lee, P. M.; Odom, A. L.; Davis, W. M.; Cummins, C. C. Angew. Chem. Int. Ed. Eng. 1997, 36, 87-91. [23] Tsai, Y. C.; Johnson, M. J. A.; Mindiola, D. J.; Cummins, C. C.; Klooster, W. T.; Koetzle, T. F. J. Am. Chem. Soc. 1999, 121, 10426-10427. [24] Nielson, A. J.; Hunt, P. A.; Rickard, C. E. F.; Schwerdfeger, P. J Chem. Soc. Dalton Trans. 1997, 3311-3317. [25] Thomas, S.; Lim, P. J.; Gable, R. W.; Young, C. G. Inorg. Chem. 1998, 37, 590-594. [26] te Velde, G.; Bickelhaupt, F.M.; Baerends, E. J.; Guerra, C. F.;van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931-967. [27] Brask, J. K.; Dura-Via, V.; Diaconescu, P. L.; Cummins, C. C. Chem. Commun. 2002, 902-903. [28] Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126, 13916-13917. [29] Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2003, 125, 4020-4021. [30] Figueroa, J. S.; Cummins, C. C. Angew. Chem. Int. Ed. 2004, 43, 984-088. [31] Figueroa, J. S.; Cummins, C. C. Dalton Trans. 2006, 2161-2168. [32] Laplaza, C. E.; Davis, W. M.; Cummins, C. C. Angew. Chem. Int. Ed. Eng. 1995, 34, 2042-2044. [33] Zanetti, N. C.; Schrock, R. R.; Davis, W. M. Angew. Chem. Int. Ed. 1995, 34, 2044-2046. [34] Pasteur, L. C. R. Acad. Sci. Paris 1848, 26, 535-538. [35] Pasteur, L. Anal. Chim. Phys. 1848, 24, 442-459. [36] Flack, H. D. Acta Cryst. A 2009, 65, 371-389. [37] Fox, A. R.; Clough, C. R.; Piro, N. A.; Cummins, C. C. Angew. Chem. Int. Ed. 2007, 46, 973-976. [38] Santure, D. J.; Sattelberger, A. P.; Cotton, F. A.; Wang, W. Inorg. Synth. 1989, 26, 219-225. [39] Molle, G.; Bauer, P.; DuBois, J. E. J. Org. Chem. 1982, 47, 4120-4128. [40] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. [41] Clough, C. R.; Greco, J. B.; Figueroa, J. S.; Diaconescu, P. L.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126, 7742-7743. [42] Guerra, C. F.; Sniders, J. G.; te Velde, G.; Baerends, E. J. Theor Chem. Acc. 1998, 99, 391-403. [43] Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200-1211. [44] Becke, A. D. J. Phys. Rev. A 1988, 38, 3098-3 100. [45] Perdew, J. P. Phys. Rev. B 1986, 34, 7406. [46] van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99, 4597-4610. [47] van Lenthe, E.; Ehlers, A.; Baerends, E. J. J. Chem. Phys. 1999, 110, 8943-8953. [48] Fan, L. Y.; Ziegler, T. J. Chem. Phys. 1992, 96, 9005-9012. [49] Fan, L. Y.; Ziegler, T. J. Chem. Phys. 1992, 96, 9005-9012. [50] Sheldrick, G. M. Acta Cryst. A 2008, 64, 112-122. CHAPTER 2

Synthesis and Reactivity of Pseudo-Octahedral Tungsten Trisanilide Species

Contents 2.1 Introduction ...... 56 2.2 Synthesis of a Pseudo-Octahedral Tungsten Trisanilide Species ...... 56

2.3 Syntheses of Pseudo-Octahedral Tungsten Trisanilide Species using PC 5 . 59

2.4 Reduction Studies of (Ar[i-Pr]N)3 W(Cl) 3 (12) ...... 62

2.5 Synthesis of OW(N[i-Pr]Ar) 3 (15) ...... 67 2.6 Conclusions ...... 70 2.7 Experimental Section ...... 70 2.7.1 General Considerations...... 70

2.7.2 Synthesis of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10)...... 70

2.7.3 Synthesis of (Ar[i-Pr]N) 3W(N=PCl3)(C) 2 (11) ...... 71

2.7.4 Synthesis of (Ar[i-Pr]N)3 W(Cl)3 (12) ...... 72

2.7.5 Synthesis of (Ar[i-Pr]N) 3W(Cl) 2 (13) with KC8 ...... 72

2.7.6 Synthesis of (Ar[i-Pr]N) 3W(Cl) 2 (13) with LiBH4 ...... 73 2 2.7.7 Synthesis of [Na(OEt 2)1.5 (thf)o.5 ][W(H)( -Me 2 C=NAr)2-cyclo- (N[iPr]Ar)] ([Na(OEt 2) 1.5 (th)o.5][14])...... 73

2.7.8 Synthesis of OW(N[i-Pr]Ar) 3 (15)...... 74 2.8 Crystallographic Structure Determinations ...... 74 2.8.1 General Considerations ...... 74

2.8.2 X-ray crystal structure of (Ar[i-Pr]N)3W(OCN)(C) 2 (10) ...... 74

2.8.3 X-ray crystal structure of (Ar[i-Pr]N)3W(N=PCl3)(Cl) 2 (11) -- ..... 75 Reproduced in part with permission from: Clough, C. R.; MUller, P. M.; Cummins, C. C. Dalton Trans., 2008, 4458-4463; Copyright 2008 The Royal Society of Chemistry. 2.8.4 X-ray crystal structure of (Ar[i-Pr]N)3W(Cl) 3 (12)...... 76

2.8.5 X-ray crystal structure of (Ar[i-Pr]N)3W(Cl)2 (13)...... 76 2.8.6 X-ray 2 crystal structure of [Na(OEt2) 1.5(thf).5] [W(H)(T -Me 2C=NAr)2- cyclo-(N[iPr]Ar)] ([Na(OEt 2)1.5(th17o. 5][14]) - ...... - . - . 77

2.8.7 X-ray crystal structure of OW(N[i-Pr]Ar) 3 (15)...... 77 2.9 References ...... 81

2.1 Introduction

In exploring the reactivity of the terminal tungsten nitride, NW(N[i-Pr]Ar) 3 (1, Ar = 3,5-Me2C6H3), pseudo-octahedral complexes supported by three bulky N(i-Pr)Ar ligands have been prepared. This result was surprising considering previous work in our group has shown that five-coordinate complexes of "Mo(N[R]Ar) 3" (R = i-Pr, t-Bu) species are rare and, in most cases, either unstable or very reactive. 1,2 Isolation of six-coordinate species using three ancillary N(i-Pr)Ar ligands lends credence to the mechanism proposed for nitrile formation presented in Chapter One. Finally, some of the six-coordinate species could prove useful for future work focused on synthesis of a 2 tungstaziridine hydride species, W(H)(Tj -Me2C=NAr)(N[i-Pr]Ar) 2. Some initial work towards the synthesis of a tungstaziridine complex is presented at the end of this chapter.

2.2 Synthesis of a Pseudo-Octahedral Tungsten Trisanilide Species

In the process of elaborating upon research described in Chapter one of this thesis, 1 was treated with 0.5 equivalents of oxalyl chloride in an attempt to synthesize cyanogen. A IH NMR spectrum of the reaction mixture showed a 1:1 mixture of starting material, 1, and a new product indicating a preferred stoichiometry of 1:1 1:oxalyl chloride. Interestingly, production of oxochloride (Ar[i-

Pr]N) 3W(O)Cl (2) was not observed. Upon repeating the reaction with the correct stoichiometry, effervescence (attributed to CO formation) was observed. Only one complex was observed by 'H NMR spectroscopy (Figure 2.1). The product exhibits three distinct N(i-Pr)Ar ligand environments in a 1:1:1 ratio. Furthermore, the diastereotopic nature of the ligand environments suggests formation of a C1-symmetric complex. It was only upon determination of the X-ray crystal structure combined with FT-IR data that the identity of the product of 1 + oxalyl chloride was determined to be the six-coordinate cyanate dichloride, (Ar[i-Pr]N) 3W(OCN)(Cl)2 (10, Scheme 2.1).

The IH NMR spectrum fully supports the facial arrangement of the (OCN)(Cl) 2 ligands. Most trisanilide systems synthesized in our group exhibit a single, non-diastereotopic ligand environment ortho-H

ortho-H

para-H para-H Ar-methyl

C6DH i-Pr methine

Ar-methyl

7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 ppm i-Pr methyl L-Pr th l THF J,THF L K_ -7--F i I . I I i I I 7 6 5 4 3 2 1 ppm Figure 2.1. IH NMR spectrum of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10)-

N 0 0 C 0 CI Cl /CI Arqi-Pr]Nm-W-C('I - CO Ar[i-Pr]N* ~-rA Et0, 25 *C N[i-Pr]Ar 2 25%

Scheme 2.1. Synthesis of the 6-coordinate, tungsten-trisanilide species (Ar[i-Pr]N)3W(OCN)(Cl) 2 (10). ------

consistent with time-averaged C3, symmetry by 1H NMR spectroscopy at 25 'C. The severe steric crowding of 10 (evident from the solid-state structure) leads to hindered rotation of the anilide ligands as observed in the 'H NMR spectrum. The combination of the Cs-symmetric (OCN)(Cl) 2 grouping with the C3-symmetric anilides leads to the overall CI symmetry of 10. One methyl group from each i-Pr residue is forced into close proximity with the adjacent aryl face. The close contact between the methyl group with the it-cloud of the neighboring aryl group results in the extremely upfield location of the i-Pr methyl resonances (8 ~~-0.5 ppm). Finally, the IR spectrum further confirms the presence of a cyanate ligand with a stretch at v = 2200 cm-1 (Figure 2.2). However, the IR spectroscopy data cannot make a distinction between 0-bound and N-bound connectivity. 3

2237-e0dm.dat'

-5

-19

-15 -2 0c -

-20

-25

-se

d35o I I I I I I a 48e 3509 3099 2500 2008 1500 1888 see Havenunbers (cn-i) Figure 2.2. Solution FT-IR spectrum of (Ar[i-Pr1N)3W(OCN)(Cl) (10) in C6D6 which exhibits the very strong cyanate stretch at 2200 cm- 1. 2

Crystals of compound 10 grown from a tetrahydrofuran solution layered with n-pentane, chilled to -35 'C, were subjected to an X-ray crystallographic study (Figure 2.3). The crystallographic data was not easily modeled. Complex 10 crystallizes in the space group R3 with a crystallographically imposed C3 axis passing through the tungsten center, resulting in disorder of the cyanate and chloride ligands over three crystallographically equivalent positions. The structural model is further complicated by ambiguous atom assignments for the cyanate ligand. The crystallographic model featuring an 0-bound cyanate ligand exhibits greater stability upon least- squares refinement than does the corresponding model containing an N-bound isocyanate ligand, providing support for the assignment of 10 as presented. The structural parameters of compound 10 reflect the extreme steric crowding exhibited when a trisanilide species is forced into a 6-coordinate geometry. The CIA-WI-CiB angle of 840 shows the steric pressure that the N(i-Pr)Ar ligands exert on the chlorides when forced into a facial arrangement in a pseudo-octahedralspecies.

) N~x

Clx

CI1A C11 B 01x

Figure 2.3. ORTEP drawing of (Ar[i-Pr]N) 3W(OCN)(Cl) 2 (10) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Wi-Cl 2.514(5), Wl-Olx 2.01(2), G1x-Cix 1.35(3), Clx-Nlx 1.11(3), CIA-WI-CIB 84.3(2), W1-Olx-Clx 129(2). P1-Clx-Nlx 171(3).

2.3 Syntheses of Pseudo-Octahedral Tungsten Trisanilide Species

using PCI5

After a successful synthesis of the 6-coordinate species supported by three N(i-Pr)Ar ligands, we sought to synthesize more examples of pseudo-octahedral species. The idea of treating the terminal- nitride complex 1 with phosphorus pentachloride comes from the precedent of PC 5 reacting with N- containing species to form complexes of the type R-N=PCl 3.4 For example, t-BuNH 2 is known to react with PCl5 to form the iminophosphorane, t-BuN=PCl 3-5 Perhaps the example most pertinent 6 to this work is the reaction of Cl3C-C=N with PC15 to form CCl3CCl2N=PCl 3. When treated with one equivalent of PCl5 , terminal nitride 1 cleanly forms the six-coordinate phosphinimide, (Ar[i-Pr]N) 3W(N=PC 3)(Cl) 2 (11, Scheme 2.2). The 'H NMR spectrum of 11 is very similar to that of cyanate 10, exhibiting the same C1 symmetry as shown by the three distinct, diastereotopic i-Pr anilide resonances. The 31P NMR spectrum lends further credence to the assignment with the phosphinimide phosphorus resonating at 8 = -49.8 ppm (2jwP = 85 Hz). Metal bound 7 trichlorophosphinimides are known for tungsten (Cl 5W-N=PCl 3, 31P NMR: 8 = 42.6 ppm) and tantalum (Cl Ta-N=PCl 31 8 4 3, P NMR: 8 = 16.2 ppm) resulting from the reactions of Cl3P=NSiMe3 with WC16 or TaC15, respectively. Perhaps the preparation most similar to the synthesis of 11 is the formation of [PPh4][Cl5Mo-N=PCl 3] from the reaction of terminal nitride [PPh 4][Cl 4MoN] with 9 PCl3/PCl 5.

PCI3 III N

Ar[i-Pr]N N 5 ] Ar[i-Pr]N-W--CI Ar[i-Pr]N Et20 Ai N I -116 to 25 C Ar[i-Pr]N N[i-Pr]Ar 77%

Scheme 2.2. Synthesis of (Ar[i-Pr]N) 3W(N=PC 3)(Cl) 2 (1)'

X-ray quality crystals of 11 were grown from a saturated diethyl ether solution. The solid state structure of 11 is reminiscent of the structure of 10. Fortunately, the phosphinimide crystallizes in the space group P21/c. As a result, there is no crystallographically imposed C3 axis and therefore no positional disorder between the phosphinimide and chloride ligands. The W-N-P angle is slightly bent at 1560 and the W-N distance has increased from 1.669(5) A in the terminal nitride (1) to 2.047(3) A. The N-P bond length of 1.45 A shows the multiple bond character of the phosphinimide when compared to the sum of N-P covalent radii (1.78 A).1 0 Calling to mind the structure of 10, the Cl-W-Cl and Cl-W-N angles of well below 90' again as a result of the steric pressure provided by the anilide ligands.

After successfully forming two pseudo-octahedral species from nitride 1, oxochloride 2 was treated with PCl5 to see if it exhibited similar reactivity. When a thawing solution of 2 in

Et20 is introduced to a frozen slurry of PC15 in Et20 and allowed to warm to room temperature while stirring, a canary-yellow precipitate appears in the reaction vessel. 'H NMR spectroscopy indicated a complex with a diastereotopic N(i-Pr)Ar ligand environment, this being consistent with C14 C13

C12 C11 C15

Figure 2.4. ORTEP drawing of (Ar[i-Pr]N)3W(N=PCl3)(Cl) 2 (11) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): W1-N4 2.047(3), N4-P1 1.449(3), W1-C11 2.476(1), W1-Cl2 2.453(1), Cll-W1-Cl2 82.45(3), Cll-W1-N4 84.6(1), C12-W1-N4 85.0(1). a C3-symmetric species with the ligands frozen into a propeller arrangement. The canary-yellow solid was assigned as the predicted, C3-symmetric product (Ar[i-Pr]N) 3W(C) 3 (12, Scheme 2.3). The optimized synthesis of 12 is a convenient preparation where the oxyphosphorus trichloride byproduct is volatile and the desired product precipitates out of the reaction mixture. Compound 12 can be prepared in moderate (70%) yield on multigram scales making it an attractive starting material. The synthesis of 12 is further simplified by the fact that its immediate precursor need not be isolated. Treatment of nitride 1 with pivaloyl chloride cleanly forms oxochloride 2 and pivalonitrile in quantitative yield. The volatile nitrile can be removed under reduced pressure, the oxochloride redissolved and treated with PC15 to synthesize trichloride 12.

O C1 CIC

N--W-C PCI5 Ar[i-Pr]N- W-CI

- I N[i-Pr]Ar Et20 Ar[i-Pr]N \ / N[i-Pr]Ar -116 to 25 *C N[i-Pr]Ar - O=PC 3 70%

Scheme 2.3. Synthesis of (Ar[i-Pr]N) 3W(C1) 3 (12).

Single crystals of 12 were grown from a saturated methylene chloride solution at -35 'C and subject to X-ray diffraction studies (Figure 2.5). Interestingly enough, the C3-symmetric compound 12 does not crystallize in a space group with a three-fold axis of rotation-unlike the Ci -symmetric compound 10 which crystallized in R3. Instead, compound 12 crystallizes in the triclinic space group Pl. As determined by 'H NMR spectroscopy, the solid-state structure of 12 conforms to a

C3-symmetric, 6-coordinate trisanilide species. The metrical parameters of 12 are similar to that of the other two 6-coordinate trisanilide species shown in the present work with a rather acute Cl-W-Cl angle (average Cl-W-Cl angle 82.45(3)0).

2.4 Reduction Studies of (Ar[i-Pr]N)3W(CI) 3 (12)

The isolation of compound 12 was an exciting result as it was hoped that the three chloride ligands would prove labile under reducing conditions allowing for the synthesis of a low-coordinate tungsten trisanilide species. The anticipated product of reduction studies of 12 was the three-electron reduction product "W(N[i-Pr]Ar)3". It was assumed that if the hypothetical W(N[i-Pr]Ar) 3 could be generated, it would quickly isomerize to a tungstaziridine hydride by abstracting a hydrogen 2 atom from one of the isopropyl groups (analogous to Mo(H)(T1 -Me 2C=NAr)(N[i-Pr]Ar)2).- All attempts at reducing compound 12 led to the same result, formation of the tungsten(V) species,

(Ar[i-Pr]N) 3W(Cl) 2 (13, Scheme 2.4). A variety of reductants (e.g. Na/Hg, Na-mirror, KC8 , LiBH 4) and conditions were tried, all leading to the same result. Wolczanski has seen similar results with his Figure 2.5. ORTEP drawing of (Ar[i-Pr]N) 3W(Cl) 3 (12) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Wi-Cl 2.441(1) (avg.), Cl-Wi-Cl 82.45(3) (avg.).

2 tris-silox chemistry in efforts to reduce (silox) 3NbCl 2 by two electrons.1 In the case of the niobium silox complex, complete 2 e reduction was effected by addition of pyridine. All attempts to reduce trichloride 12 in the presence of coordinating ligands did not lead to species with oxidation states lower than tungsten(V).

To date, the cleanest routes to forming dichloride 13 are from the reduction of trichloride 12 with excess KC8 or with one equivalent of LiBH 4 . Both methods result in clean formation of the desired product, but isolated yields of compound 13 are relatively low (29-37%) as a result of the product's lipophilicity. Crystals of compound 13 were grown from a saturated diethyl ether solution and subjected to an X-ray crystallographic study (Figure 2.6). The solid-state structure of 13 is that of a slightly disordered trigonal bipyramid with the chloride ligands in the axial positions. Although the sum of the equatorial angles is nearly ideal (Eequatorial = 359.69(16)0), the Cl-W-Cl bond angle is slightly bent from linear at 173.59(4)'. Also of interest are the W-Cl bond lengths of 13. Wl-C12 (2.4499(12) A) is nearly identical to those in the trichloride 12 (Wl-Clavg. = 2.441() A), but the W1-Cll bond length of 2.3941(12) is considerably shorter than in the trichloride complex.

To date, the only successful reduction of a trisanilide tungsten species to beyond tungsten(V) resulted by treatment of trichloride 12 with sodium amalgam in the presence of 1.5 equivalents of CI C' 1.5 KC 8 Ar[i-Pr]N Ar[i-Pr]N -W-ci OP ".W -N[i-Pr]Ar Et20 Aii-Pr]N eNII[i-Pr]Ar -116 to 25 *C Ar[i-Pr]N I - KCI CI 51%

Scheme 2.4. Synthesis of (Ar[i-Pr]N)3W(Cl)2 (13).

C12

C11

Figure 2.6. ORTEP drawing of (Ar[i-Pr]N)3W(Cl)2 (13) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): W1-Cll 2.3941(12), Wi-Cl2 2.4499(12), C1l-W1-Cl2 173.59(4), Eequatorial = 359.69(16) hydrogen (Scheme 2.5). Unfortunately, the result of this reaction was not the formation of a tungsten 2 trisanilide (W(N[i-Pr]Ar)3 ) or the expected tungstaziridine hydride (W(H)(q -Me 2 C=NAr)(N[i- Pr]Ar)2). The result was the formation of a complex mixture in which the salt of a seven-coordinate anion species [W(H)(T12-Me2C=NAr) 2-cyclo-(N[iPr]Ar)] - ([14] -) was the major product. The salt of anion [14]- was isolated from the reaction mixture, in pure form, in 14% yield as

[Na(OEt 2 ) 1.5 (thf)0.5][14]. The structure of anion [14]- was determined by a combination of X-ray crystallographic and 1H NMR spectroscopic techniques. The solid-state structure of anion [14]~ shows what is best described as a "bis-tungstaziridine-aza-tungstacyclobutane hydride anion". The 1 H NMR spectrum of [Na(OEt 2)1.5 (thf)0 .5][14] confirms the presence of the hydride moiety as a singlet at 6 = 10.54 ppm showing 183W-satellites (1JwH = 45 Hz). In the solid-state structure of anion [14]-, the hydride was not located in the electron difference map. The solid state structure of [Na(OEt 2)1.5(thf)0. 5][14] also shows a close contact between the two counterions. There are two independent molecules within the asymmetric unit in the structure of [Na(OEt 2 )1.5 (thD0.5][14]. In one, the sodium cation is coordinated to two diethyl ether molecules and in the other, the sodium is coordinated to one diethyl ether and one tetrahydrofuran.

It would seem as if the formation of [Na(OEt 2)1.5(thf) 0 .5][14] is a result of reduction to tungsten(III) whereupon the tungsten complex cannibalizes its ligands. Formation of anion [14]- could also occur in a stepwise process whereupon tungsten(VI) is reduced to tungsten(IV) leading to hydrogen abstraction from a ligand reforming tungsten(VI) (which is then reduced again to tungsten(IV) and the process repeats). Of significant interest is the presence of nitride 1 in the reaction mixture of compound 12 with H2/Na/Hg. As the starting material, compound 12, is spectroscopically pure and free of nitride, it is certainly plausible that some low-coordinate tungsten complexes are being formed fleetingly and activate N2 to form nitride 1 (albeit in very low yield). The formation of nitride in these reaction mixtures warrants reinvestigation of this system.

Na(solvent)2 C1H

xs Na/Hg, 1.5 H2 NIm'-Wo Ar[i-Pr]N - C1 THF, 25 C Ar N

N[i-Pr]Ar < 1 atm N2 Ar 14% 2 Scheme 2.5. Synthesis of [Na(OEt 2) L5(thD0.5][W(H)(1 -Me2C=NAr)2-cyclo-(N[iPr]Ar)] ([Na(OEt2)1.5(thD0. 5][14]). Na2

Figure 2.7. ORTEP drawing of [Na(OEt 2)1.5(thf) 0.5][W(H)(T12-Me 2C=NAr)2-cyclo-(N[iPr]Ar)] ([Na(OEt 2)1.5(thf)0. 5][14]) with thermal ellipsoids at the 50% probability level. Hydrogen atoms and solvent molecules have been omitted for clarity. 2.5 Synthesis of OW(N[i-Pr]Ar)3 (15)

Another reduction study carried out involved treatment of oxochloride 2 with sodium amalgam to synthesize the tungsten(V) trisanilide oxo species, OW(N[i-Pr]Ar) 3 (15). As oxochloride 2 reacts with triflic anhydride (Tf2O) to give oxotriflate 8 (see Section 1.5), it was deemed plausible that the tungsten(V) oxo species, 15, may react favorably with Tf2O to form the bis-triflate analogous 13 to the reaction involving ONb(N[Np]Ar) 3- To synthesize 15, oxochloride 2 is generated in situ and treated with Na/Hg in THF in a one-pot procedure (Scheme 2.6). Unfortunately, compound 15 is isolated in low yield as a result of its extreme lipophilicity. The complex appears to be a d' tungsten(V) monomer in solution as evinced by its paramagnetic nature (Evans Method measurement gave peff = 1.49 pB).

N-W-C Na/Hg Ar[i-Pr]N \W N. N[i-Pr]Ar THF, 25 C Ar[i-Pr]N \/ N[i-Pr]Ar 5% generated in situ Scheme 2.6. Synthesis of OW(N[i-Pr]Ar) 3 (15).

Crystals of compound 15 were grown from a saturated diethyl ether solution and used for an X-ray crystallography study. The solid-state structure reveals a bridging bis-p-oxo dimer (Figure 2.8). The asymmetric unit consists of only half of the molecule which is crystallographically related to its other half by a C2 axis. The bonding metrics reveal a short-long bonding motif from each tungsten to each oxygen (1.853(3) and 2.068(2) A, respectively). There is also a close contact between the two tungsten atoms of the dimer of 2.6938(3) A. Attempts were made to determine if compound 15 is diamagnetic in the solid state through SQuID measurements, but unfortunately the complex's extreme air sensitivity (even in the solid state) caused it to to decompose within seconds while attempting to prepare the SQuID sample. Figure 2.9 highlights the tungsten oxo dimer core showing the slight pucker from planar giving the four-membered ring a butterfly-like geometry.

Treatment of compound 15 with Tf2O did not lead to clean formation of the desired bis-triflate complex. Instead, what is observed is a complex reaction mixture where oxotriflate 8 is observed as the major product (Scheme 2.7). Several reaction conditions were attempted, all yielding to the same result. mpg,

Figure 2.8. ORTEP drawing of OW(N[i-Pr]Ar)3 (15) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (0): W1-01 1.853(3), W1-OA 2.068(2), WI-WIA 2.6938(3), 01-WI-OA 88.98(10), Wi-01-W1A 86.61(10).

O O

OTf AiP]Tf 20 N-W -

Ar[i-Pr]N N Et20, 25 *C - [i-Pr]Ar \ / N[i-Pr]Ar major

Scheme 2.7. Reaction of OW(N[i-Pr]Ar) 3 (15) with triflic anhydride. 2A

N3 W1 W1A NI

N20 N2A t

Figure 2.9. ORTEP drawing of OW(N[i-Pr]Ar) 3 (15) with thermal ellipsoids at the 50% probability level highlighting the tungsten oxo dimer core. All atoms not directly bound to tungsten have been omitted for clarity. 2.6 Conclusions

It has been shown that the tungsten trisanilide platform, despite being developed to engender low coordination numbers at the metal, is capable of expanding to a pseudo-octahedral coordination environment. Nitride 1 can be converted into the dichloride cyanate 10 or the phosphinimide

11 through treatment with oxalyl chloride or PCl5, respectively. Similarly, oxochloride 2 can be converted to the trichloride species 12 by treatment with PCl5 . Reduction to lower oxidation states of the trichloride species has proven difficult whereupon most reductants lead to formation of the tungsten(V) species 13. In the presence of hydrogen, however, trichloride 12 is presumably reduced beyond tungsten(V), whereupon it cannibalizes its ancillary trisanilide framework. Formation of nitride 1 in the H2/NafHg reduction of compound 12 requires further investigation to determine if the tungsten trisanilide framework is indeed capable of nitrogen fixation.

2.7 Experimental Section

2.7.1 General Considerations

Unless stated otherwise, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere. NW(N[i-

Pr]Ar)3 (1) and (Ar[i-Pr]N) 3W(O)Cl (2) were prepared as previously published. 14 Oxalyl chloride was purchased from Aldrich and distilled under N2. PCl 5 was purchased from Aldrich and used as received. Diethyl ether, n-pentane and toluene were dried and deoxygenated by the method of Grubbs.15 THF was distilled from purple Na/benzophenone and collected under nitrogen. C6D6 was degassed and dried over 4 A molecular sieves. Other chemicals were purified and dried by standard procedures or were used as received. Celite, alumina and 4 A molecular sieves were dried in vacuo overnight at a temperature above 200 C. 1H, 13C, I9F and 31P NMR spectra were recorded on Varian Mercury-300, Varian INOVA-500 or Bruker AVANCE-400 spectrometers. 'H and 13C NMR chemical shifts are reported with respect to internal solvent (C6D6 , 8 = 7.16 and 128.39 ppm, respectively). 31P NMR chemical shifts are reported with respect to an external 19 reference (85% H3P0 4, 8 = 0.0 ppm). F NMR chemical shifts are reported with respect to an external reference (CFCl3, 6 = 0.0 ppm). Infrared spectra were recorded on a Bio-Rad 135 Series FTIR spectrometer. Combustion analyses were performed by either H. Kolbe Mikroanalytisches Laboratorium, Miheim an der Ruhr, Germany or Midwest Microlabs LLC, Indianapolis, IN.

2.7.2 Synthesis of (Ar[i-Pr]N)3W(OCN)(CI) 2 (10)

Oxalyl chloride (39.0 pL, 0.447 mmol) was added to a colorless solution of 1 (304 mg, 0.444 mmol) in Et20 (10 mL) using a microliter syringe. The reaction mixture turned blood red upon addition and shortly changed color to brown. The reaction mixture was stirred for 20 minutes at which point the volatiles were removed under reduced pressure giving a brownish-yellow solid. The crude material was scraped onto a fritted glass filter and washed with Et20 revealing a bright-yellow solid. The bright-yellow solid was dissolved in minimal THF, the solution filtered through a plug of Celite and the filtrate cooled to -35 'C overnight. From the THF solution, small yellow crystals were harvested, washed with pentane and dried in vacuo (87 mg, 0.11 mmol, 25%). Recrystallized samples of 10 contain ~ 0.5 eq. of THF that persists even after drying in vacuo as evinced by the 'H and 13C NMR spectra. X-ray quality crystals of 10 can be grown from a saturated THF solution 0 layered with pentane and stored at -35 C. 'H NMR (500 MHz, C6D6) 6 7.65 (s, 1H, para), 7.61 (s, 1H, para), 7.16 (s, 1H, para), 6.56 (s, 3H, ortho), 6.43 (septet, 1H, i-Pr methine), 6.36 (s, 3H, ortho), 6.26 (septet, 1H, i-Pr methine), 6.01 (septet, 1H, i-Pr methine), 2.15 (s, 9H, ArCH 3), 2.054 (s, 3H, ArCH 3), 2.049 (s, 3H, ArCH 3), 2.02 (s, 3H, ArCH3), 1.56 (d, 3H, i-Pr methyl), 1.52 (d, 3H, i-Pr methyl), 1.39 (d, 3H, i-Pr methyl), -0.27 (d, 3H, i-Pr methyl), -0.30 (d, 3H, i-Pr methyl), 13 -0.31 (d, 3H, i-Pr methyl) ppm. C NMR(125 MHz, C6D6) 6 152.1 (ipso), 151.9 (ipso), 151.7 (ipso), 138.51 (meta), 138.4 (meta), 137.3 (meta), 137.1 (meta), 137.0 (meta), 136.8 (OCN), 129.05 (ortho), 129.00 (ortho), 128.8 (ortho), 124.2 (ortho), 124.01 (ortho), 123.92 (para), 123.85 (para), 123.82 (ortho), 123.6 (para), 68.0 (i-Pr methine), 67.7 (i-Pr methine), 67.1 (i-Pr methine), 23.4 (i-Pr methyl), 23.3 (i-Pr methyl), 23.2 (i-Pr methyl), 23.0 (i-Pr methyl), 22.5(i-Pr methyl), 22.08 (2C,

ArCH3 ), 22.06 (ArCH3), 21.74 (i-Pr methyl), 21.72 (i-Pr methyl), 21.70 (i-Pr methyl), 21.5 (i-Pr 1 methyl) ppm. FTIR (C6 D6 , KBr) VNCO = 2200 cm- (vs). Anal. Calcd. for C34 H4 8Cl2N4 0W: C, 52.12; H, 6.17; N, 7.15. Found: C, 51.75; H, 6.18; N, 6.93.

2.7.3 Synthesis of (Ar[i-Pr]N)3W(N=PCla)(CI) 2 (11)

A thawing, colorless solution of 1 (505 mg, 0.738 mmol) in Et2 0 (5 mL) was added to a thawing suspension of PCl5 (154 mg, 0.740 mmol) in Et 2O (5 mL) resulting in an orange-red reaction mixture upon addition. The reaction mixture was allowed to warm to room temperature and stirred for 1 h, after which time the volatiles were removed in vacuo leaving an orange-yellow solid. The solid was collected on a fritted glass filter, washed with pentane and dried under vacuum (505 mg, 0.566 mmol, 76.6%). The sample used to collect the NMR spectra was recrystallized from THF layered with Et 20 at -35 'C. The sample contains ~1 equiv of THF of co-crystallization as seen in the 'H and 13C NMR spectra. The THF remains in the sample even after prolonged periods of drying in vacuo. 'H NMR (400 MHz, C6D6 ) 6 7.74 (s, 1H, para), 7.72 (s, 1H, para), 7.32 (s, 1H, para), 6.61 to 6.47 (7H, ortho and i-Pr methine), 6.19 (septet, 1H, i-Pr methine), 5.88 (septet, 1H, i- Pr methine), 2.20 (s, 9H, ArCH3), 2.16 (s, 3H, ArCH3), 2.09 (s, 3H, ArCH3), 2.08 (s, 3H, ArCH 3), 1.65 (d, 3H, i-Pr methyl), 1.55 (d, 3H, i-Pr methyl), 1.44 (d, 3H, i-Pr methyl), -0.22 to -0.25 (9H, 3 i-Pr methyl) ppm. C NMR (100 MHz, C6D6) 6 153.0 (ipso), 152.5 (ipso), 151.9 (ipso), 138.1 (meta), 137.9 (meta), 137.8 (meta), 137.1 (meta), 136.6 (meta), 136.4 (meta), 124.3 (2C, ortho), 124.5 (2C, ortho), 124.3 (para), 124.2 (para), 124.1 (para), 68.9 (i-Pr methine), (i-Pr methine), 68.1 (i-Pr methine), 66.1 (i-Pr methine), 23.7 (methyl), 23.4 (methyl), 23.2 (methyl), 23.1 (methyl), 23.0 (methyl), 22.7 (methyl), 22.02 (methyl), 21.95 (methyl), 21.70 (methyl), 21.66 (methyl) ppm. 31p 2 NMR (162 MHz, C6D6) 8 -49.8 ( jwc = 85 Hz) ppm. Anal. Calcd. for C33H48Cl5N4PW: C, 44.39; H, 5.42; N, 6.27. Calcd. for C37H56C15N40PW (1.0 eq. THF-as observed in the crystal structure): C, 46.05; N, 5.85; N, 5.81. Found: C, 45.95; H, 5.84; N, 5.72.

2.7.4 Synthesis of (Ar[i-Pr]N)3W(C) 3 (12)

A thawing, red solution of 2 (794 mg, 1.10 mmol) in Et2O (5 mL) was added to a thawing suspension of PCl5 (228 mg, 1.10 mmol) in Et2O (3 mL) resulting in an orange reaction mixture upon addition. The reaction mixture was allowed to warm to room temperature and was stirred for 0.5 h. A canary yellow solid precipitated out of solution and was collected on a fritted glass filter. The solids were washed with pentane (3 * 20 mL) and dried under vacuum (595 mg, 0.766 mmol, 69.6%). Samples 3 contain - 1 equiv of Et2O as evinced by 'H and 1 C NMR. The Et2O remains even after prolonged periods of drying in vacuo. X-ray quality crystals of 12 can be grown from a saturated methylene 0 chloride solution layered with diethyl ether and stored at -35 C. 'H NMR (300 MHz, C6D6) 6 7.77 (s, 3H, para), 6.55 (s, 3H, ortho), 6.44 (s coincident with septet, 6H, ortho and i-Pr methine, respectively), 2.15 (s, 9H, ArCH 3), 2.05, (s, 9H, ArCH 3), 1.16 (d, 9H, i-Pr methyl), -0.27 (d, H, i-Pr methyl) ppm. 13C NMR (100 MHz, C6D6) 6 152.3 (ipso), 138.2 (meta), 136.8 (meta), 124.0 (2C, ortho), 123.9 (para), 68.5 (i-Pr methine), 23.4 (methyl), 22.1 (methyl), 22.0 (methyl), 21.6 (methyl) ppm. Anal. Calcd. for C33H48C13N3W: C, 51.01; H, 6.23; N, 5.41. Calcd. for C33.5H49Cl4N3W (0.5 eq. methylene chloride-as observed in the crystal structure): C, 49.10; H, 6.04; N, 5.13. Found: C, 48.58; H, 6.38; N, 4.32.

2.7.5 Synthesis of (Ar[i-Pr]N)3W(CI) 2 (13) with KC8

In the glovebox, compound 12 (500 mg, 0.644 mmol) was suspended in 10 mL of THF and frozen in the cold well. Solid KC8 (130 mg, 0.962 mmol, 1.49 equiv.) was added to the thawing suspension of 12 and the reaction mixture stirred. The color of the reaction mixture immediately changed from yellow to dark-brown. The mixture was stirred for 2 h and filtered through a bed of Celite to remove any graphite and unreacted KC 8. The Celite pad was washed with THF until the filtrates were colorless. The resulting green filtrate was reduced to dryness under reduced pressure and triturated with hexane. The green residue was extracted with diethyl ether, filtered through Celite and the volatiles removed under reduced pressure. The forest-green solid was dissolved in a minimum Et2O, filtered through Celite and cooled to -35 'C overnight. The following day, a green, crystalline solid was isolated by vacuum filtration and washed with cold pentane. The crystalline product, 13, was dried under reduced pressure until constant mass was obtained (242 mg, 0.326 mmol, 50.7%, 3 crops). 'H NMR (400 MHz, C6D6) 6 12.5 (v. br.), 10.6 (v. br.) 5.52 (s), 2.41 (s) ppm. An Evans Method measurement gave a value of peffg = 1.07 PB. Caled. for C33 H4 8 Cl2N3W: C, 53.45; H, 6.52; N, 5.67, 9.56. Found: C, 53.90; H, 6.91; N, 5.88; Cl, 9.14.

2.7.6 Synthesis of (Ar[i-Pr]N)3W(CI) 2 (13) with LiBH4

In the glovebox, compound 12 (434 mg, 0.559 mmol) was suspended in 10 mL of THF and frozen in the cold well. Solid LiBH 4 (12.2 mg, 0.560 mmol) was slurried in 2 mL THF and added to the thawing solution of 12. The reaction mixture immediately changed in color from yellow to dark- brown. The mixture was stirred for 2 h, after which time it was filtered through a bed of Celite. The Celite pad was washed with THF until the filtrates were colorless. The resulting green filtrate was stripped to dryness under reduced pressure and triturated with pentane. The green residue was extracted with pentane, filtered through Celite and again stripped to dryness. The forest-green solids were dissolved in Et 2 O, filtered through Celite and cooled to -35 'C. The following day, green, crystalline 13 was isolated by vacuum filtration, washed with cold pentane and dried under reduced pressure until constant mass was obtained (150 mg, 0.202 mmol, 36%). The I H NMR of the product was identical to that of 13 synthesized from KC8.

2 2.7.7 Synthesis of [Na(OEt 2)1.5(thf)o. 5][W(H)(T9 -Me2C=NAr) 2-cyco-(N[iPr]Ar)] ([Na(OEt2)1.s(thf)o.slE141)

In the glove box, 12 (250 mg, 0.322 mmol) was suspended in 10 mL THF and loaded into a 100 mL Schlenk flask. The reaction flask was equipped with a stirbar, sealed with a septum and removed from the glovebox. The headspace was evacuated and H2 (11.8 mL, 0.483 mmol) was added via syringe. The remaining headspace was backfilled with N2 at 1 atm. 0.4% Na/Hg (32.2 mg, 1.40 mmol, in 8.02 g Hg) was added to the stirring suspension via syringe. The mixture quickly turned green with concomitant formation of NaCl. The reaction mixture was stirred for 1.5 h and the volatiles removed under reduced pressure. The flask was returned to the box and the green product extracted with n-pentane, the extract filtered through Celite, and the filtrate was reduced to dryness under reduced pressure. The solids were dissolved in minimal Et 20 and cooled to -35 'C overnight. The next day, bright orange, crystalline [Na(OEt 2)1.5 (thf)0 s][14] had formed in the vial

(37 mg, 0.044 mmol, 14%). 'H NMR (300 MHz, C6 D6 ) 8 10.54 (s, 1H, hydride, 'JWH = 45 Hz), 7.49 (s, 2H, ortho), 7.07 (s, 2H, ortho), 6.81 (br. s, 2H, ortho), 6.59 (s, 1H, para), 6.32 (s, lH, para),

6.26 (s, 1H, para), 5.29 (q, 1H, methine), 2.94 (q, Et 20), 2.67 (s, 3H, Me), 2.40 (s, 6H, Me), 2.18 (s, 3H, Me), 2.16 (s, 3H, Me), 2.10 (s, 6H, Me), 2.06 (s, 3H, Me), 2.03 (s, 6H, Me), 1.21 (d, 3H, Me),

0.87 (t, Et 2O) ppm. 2.7.8 Synthesis of OW(N[i-Pr]Ar)3 (15)

In the glovebox, 1 (515 mg, 0.752 mmol) was dissolved in 2 mL Et20. t-BuC(O)Cl (93 pL, 0.755 mmol) was added to the stirring tungsten solution via a microliter pipette. The reaction mixture immediately began to turn blood red and was stirred for 2 h. The volatiles were removed under reduced pressure and triturated with n-pentane (2 * 5 mL). The red residue (presumably 100% oxochloride 2) was dissolved in 5 mL THF and transferred to a vial loaded with 1%Na/Hg (24.0 mg Na, 1.04 mmol, in 2.39 g Hg). The reaction mixture was stirred for 1.5 h, after which time the reaction mixture had changed from blood red to a reddish-brown color with concomitant formation of NaCl. The reaction mixture was decanted off of the mercury and filtered through a bed of Celite. The volatiles were removed under reduced pressure and the brownish-red 15 was extracted with pentane (ca. 10 mL) and the extract filtered through Celite. The filtrate was stripped to dryness and the product dissolved in minimal Et20 (ca. 1 mL) and cooled to -35 'C overnight. The next day, small, dark, reddish-brown crystals had formed in the vial. The mother liquor was removed by pipette and the crystals washed with cold pentane. The crystals were dried under reduced pressure until constant mass was obtained giving 15 as reddish-brown crystals (23 mg, 0.033 mmol, 4.5%)

'H NMR (500 MHz, C6D6) 8 9.5 (s, v. br., 18H, Me), 6.5 (s, v. br., 3H, para), 2.2 (s, br., 3H, i-Pr methine), 0.3 (s, br., 18H, Me), -2.6 (s, v. br., 6H, ortho) ppm. An Evans' method measurement gave a value of peff = 1.49 pB. Calcd. for C33H4 8N3 0W: C, 57.73; H, 7.05; N, 6.12. Found: C, 57.69; H, 7.15; N, 6.03.

2.8 Crystallographic Structure Determinations

2.8.1 General Considerations

X-ray data collections were carried out on a Siemens Platform three-circle diffractometer mounted with an APEX-CCD detector and outfitted with a low-temperature, nitrogen-stream aperture. Graphite monochromated Mo-Ka radiation (k = 0.71073 A) was used in all cases. All software for diffraction data processing and crystal-structure solution and refinement are contained in the SHELXTL (v6.14) program suite (G. Sheldrick, Bruker XRD, Madison, WI).16

2.8.2 X-ray crystal structure of (Ar[i-Pr]N)3W(OCN)(Cl) 2 (10)

Inside the glovebox, crystals of 10, obtained from a saturated tetrahydrofuran solution layered with pentane and stored at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A yellow plate of approximate dimensions 0.27 * 0.23 * 0.04 mm 3 was selected and mounted on a glass fiber. A total of 10803 reflections (-15 < h < 7, 0 < k < 15, 0 < 1 < 47) were collected at 100(2) K using <- and co-scans in the e range of 1.54 to 25.14', of which 2442 were unique (Rint = 0.0634). The structure was solved by direct methods using SHELXS1 6 and refined against F2 on all data by full-matrix least squares with SHELXTL. 16 The systematic absences in the diffraction data are consistent with the assigned space group of R3. Two heavily disordered molecules of tetrahydrofuran are present in the asymmetric unit. The disorders were refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. The ratios were refined freely, while constraining the total occupancy of both components to unity. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.8891 and 0.4972, respectively. The residual peak and hole electron density were 2.017 and - 1.728 e-A- 3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0450 for I > 2a(I), wR2 = 0.1394 for all data, and GOF = 1.152 (based on F 2).

Crystal data: formula C34H48Cl2N40W (with two equivalents of C4H80), space group R3, a = 13.3600(12) A, c = 39.556(6) A, V = 6114.5(14) A3. Z = 6, p = 3.206 mm-1, Dcalc = 1.512 g-cm- 3, F(000) = 2856.

2.8.3 X-ray crystal structure of (Ar[i-Pr]N)3W(N=PCl 3)(Cl) 2 (11)

Inside the glovebox, crystals of 11, obtained from a saturated tetrahydrofuran solution layered with diethyl ether and stored at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. An orange shard of approximate dimensions 0.10 * 0.09 * 0.06 mm3 was selected and mounted on a glass fiber. A total of 85345 reflections (-20 < h < 20, -18 < k < 18, -24 < 1 < 24) were collected at 100(2) K using $- and or-scans in the 6 range of 1.84 to 27.88', of which 9806 were unique (Rint = 0.0533). The structure was solved by Patterson methods using SHELXS1 6 and refined against F 2 on all data by full-matrix least squares with SHELXTL.1 6 The systematic absences in the diffraction data are consistent with the assigned space group of P21/c. A disordered molecule of tetrahydrofuran is present in the asymmetric unit. The disorder was refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. The ratios were refined freely, while constraining the total occupancy of both components to unity. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.8309 and 0.7399, respectively. The residual peak and hole electron density were 2.006 and -0.720 e.A-3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0332 for I > 2G(I), wR2 = 0.0895 for all data, and GOF = 1.068 (based on F2 ).

Crystal data: formula C33H48Cl 5N4PW (with one equivalent of C4H80, space group P21/c, a = 15.9389(6) A, b = 16.7357(5) A, c = 18.9527(6) A, = 97.4620(10)0, V = 4114.2(3) A3 . Z = 4, p = 3.206 mm- 1, Dcalc = 1.558 g-cm- 3 , F(000) = 1952. 2.8.4 X-ray crystal structure of (Ar[i-Pr]N)3W(Cl) 3 (12)

Inside the glovebox, crystals of 12, obtained from a saturated methylene chloride solution layered with diethyl ether stored at -35 'C, were coated with Paratone N (an Exxon product) on a microscope slide. A yellow shard of approximate dimensions 0.13 * 0.10 * 0.09 mm 3 was selected and mounted on a glass fiber. A total of 38272 reflections (-13 < h < 14, -- 17 < k < 17, 0 < 1 < 17) were collected at 100(2) K using $- and (o-scans in the 0 range of 1.53 to 28.28', of which 9143 were unique (Rint = 0.0321). The structure was solved by Patterson methods using SHELXS16 and refined against F2 on all data by full-matrix least squares with SHELXTL. 16 One half of a heavily disordered molecule of methylene chloride is present in the asymmetric unit which results in a non-integer number of carbon atoms in the empirical formula. The fourfold disorder involves a crystallographic inversion center and was refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. The ratios were refined freely, while constraining the total occupancy of all four components to unity. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi- empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.7465 and 0.6626, respectively. The residual peak and hole electron density were 2.061 and -0.416 e-A-3 , respectively. The least squares refinement converged normally with residuals of R1 = 0.0293 for I > 2a(I). wR2 = 0.0847 for all data, and GOF = 1.103 (based on F 2).

Crystal data: formula C33H48Cl3N3W (with 0.5 equivalents of CH 2Cl2), space group P1, a = 10.7751(3) A, b = 13.3987(3) A, c = 13.5032(4) A, x = 88.8520(10)', P = 71.9870(10)0, y = 3 83.6570(10)0, V = 1842.39(9) A , Z = 2, p = 3.451 mm 1, Deale = 1.477 g.Cm-3, F(000) = 826.

2.8.5 X-ray crystal structure of (Ar[i-Pr]N)3W(CI) 2 (13)

Inside the glovebox, crystals of 13, obtained from a saturated diethyl ether solution stored at -35 'C, were coated with Paratone N (an Exxon product) on a microscope slide. A light-green plate of approximate dimensions 0.13 * 0.10 * 0.04 mm 3 was selected and mounted on a glass fiber. A total of 67886 reflections (-21 < h < 20, 0 < k < 13, 0 K 1 < 28) were collected at 100(2) K using 0- and cy-scans in the 0 range of 1.97 to 28.28', of which 8583 were unique (Rint = 0.1193). The structure was solved by direct methods using SHELXS1 6 and refined against F2 on all data by full-matrix least squares with SHELXTL. 16 The systematic absences in the diffraction data are uniquely consistent with the assigned space group of P21/c. One half of a disordered molecule of diethyl ether is present in the asymmetric unit. The disorder was refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. The ratios were refined freely, while constraining the total occupancy of both components to unity. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.8722 and 0.6578, respectively. The residual peak and hole electron density were 1.764 and - 1.105 e.A- 3, respectively. The least squares refinement converged normally with residuals of R, = 0.0409 for I > 2a(I), wR 2 = 0.0758 for all data, and GOF = 1.039 (based on F2).

Crystal data: formula C33H48Cl2N2W (with 0.5 equivalents of Et2O), space group P21/c, a = 16.127(2) A, b = 10.4213(12) A, c = 21.654(3) A, = 107.747(4)0, V = 2466.1(8) A3 , Z = 4, p = 3.516 mm-1, Dcalc = 1.492 g-cm-3, F(000) = 1584.

2 2.8.6 X-ray crystal structure of [Na(OEt 2) 1.5(thf)0.5][W(H)(T1 -Me2C=NAr)2-cyclo- (N[iPr]Ar)] ([Na(OEt 2)1.5(thf)0.5 [14])

Inside the glovebox, crystals of 14, obtained from a saturated diethyl ether solution stored at -35 C, were coated with Paratone N oil (an Exxon product) on a microscope slide. An amber block of approximate dimensions 0.36*0.28*0.16 mm 3 was selected and mounted on a glass fiber. A total of 165119 reflections (-28 < h < 26, 0 < k < 28, 0 < 1 < 24) were collected at 100(2) K using $- and )-scans in the 0 range of 1.01 to 27.880, of which 19604 were unique (Rint = 0.0396). The structure was solved using direct methods using SHELXS 16 and refined against F2 on all data by full- matrix least squares with SHELXTL. 16 The systematic absences in the diffraction data are uniquely consistent with the assigned space group of P21/c. There are two independent molecules present in the asymmetric unit. Each tungsten species has a sodium countercation in close contact with it. Each sodium countercation's coordination sphere is completed by two ethereal solvent molecules of co- crystallization-one has two ether molecules, the other has one ether and one THE All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.6580 and 0.4264, respectively. The residual peak and hole electron density were 3.609 and -0.785 e- A, respectively. The least squares refinement converged normally with residuals of R1 = 0.0392 for 2 I > 20-(I), wR 2 = 0.1061 for all data, and GOF = 1.046 (based on F ).

Crystal data: formula C82H130N6Na2O4W2, space group P21/c, a = 21.8199(9) A, b = 21.5109(9) A, c = 18.9705(8) A, p = 112.5330(10)0, V = 8224.4(6) A3 , Z = 4, p = 2.855 mm- 1, Dcaic = 1.345 g-cm-3, F(000) = 3416.

2.8.7 X-ray crystal structure of OW(N[i-Pr]Ar) 3 (15)

Inside the glovebox, crystals of 15, obtained from a saturated diethyl ether solution stored at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A red block of approximate dimensions 0.12 * 0.10 * 0.08 mm 3 was selected and mounted on a glass fiber. A total of 28080 reflections (-20 < h < 22, -23 < k < 13, -17 < 1 < 17) were collected at 194(2) K using <- and o-scans in the 0 range of 1.60 to 28.700, of which 9427 were unique (Rint = 0.0373). The structure was solved using direct methods using SHELXS1 6 and refined against F 2 on all data by full-matrix least squares with SHELXTL. 16 The systematic absences in the diffraction data are uniquely consistent with the assigned space group of P21212. One heavily disordered molecule of tetrahydrofuran of co-crystallization was removed from the structure using the crystallographic routine SQUEEZE. 17 All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi- empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.7842 and 0.7005, respectively. The residual peak and hole electron density were 1.619 and -0.473 e-A-3, respectively. The least squares refinement converged normally with residuals of R1 = 0.0278 for I > 2a(I), wR 2 = 0.0558 for all data, and GOF = 0.985 (based on F 2 ).

Crystal data: formula C33H4 8N30W (with one equivalent of THF, thus giving a total empirical formula of C37H56N30 2W), space group P21212, a = 16.4058(6) A, b = 17.4809(7) A, c = 3 12.7447(5) A, c =p y = 900, V = 3655.0(2) A , Z = 4, p = 3.194 mm- 1, Dcaic = 1.379 gcm- 3 , F(000) = 1556. Table2.1. Crystallographic Data for (Ar[i-Pr]N),W(OCN)(C), (10), (Ar[i-Pr]N),W(N=PCl3 )(Cl), (11) and (Ar[i-Pr]N)3W(C), (12) 10 11 12 Reciprocal Net code / CCDC code 04181 / AGIQAJ 04212 / AGIQEN 04202 / AGIQIR Empirical formula, FW (g/mol) C H C1 N 0PW,b 964.93b C4 2H64 Cl2N4 0 3W,a 927.72a 37 56 5 4 C33.soH49 Cl4N3W, 819.41c Color / Morphology Yellow / Plate Orange / Shard Yellow / Shard Crystal size (mm 3) 0.27 * 0.24 * 0.04 0. 10 * 0.09 * 0.06 0.13 * 0.10 *0.09 Temperature (K) 100(2) 100(2) 100(2) Crystal system, Space group Rhombohedral, R3 Monoclinic, P2 1 /c Triclinic, P1 Unit cell dimensions (A, 0) a = 13.3600(12), a = 90 a = 15.9389(6), X= 90 a = 10.7751(3), a = 88.8520(10) b = 13.3600(12), p = 90 b = 13.7357(5), P = 97.4620(10) b = 13.3987(3), = 71.9870(10) c = 39.556(8), y = 120 c = 18.9527(6), y = 90 c = 13.5032(4), y =83.6570(10) Volume (A3) 6114.5(14) 4114.2(3) 1842.38(9) Z Density (calc., Mg/m 3) 1.512 1.558 1.477 Absorption coefficient (mm-') 3.008 3.206 3.451 F(000) 2856 1952 826 Theta range for data collection (0) 1.54 to 25.15 1.84 to 27.88 1.53 to 28.28 Index ranges -15 2a(I)] R, = 0.0450, wi 2 = 0.1157 R1 = 0.0332, wi 2 = 0.0839 I = 0.0293, wi 2 = 0.0826 R indices' (all data) R, = 0.0650, wi 2 = 0.1394 RI = 0.0436, wR 2 = 0.0895 = 0.0328, wi 2 = 0.0847 Largest duff, peak and hole (e -A-3) 2.017 and - 1.728 2.006 and -0.720 2.061 and -0.416 aTwo heavily disordered molecules of tetrahydrofuran are present in the asymmetric unit. b A disordered moleculeof tetrahydrofuran is present in the assymmetric unit.

COne half of a heavily disordered molecule ofmethylene chloride is present in the asymmetric unit. d GooF [ew(R-]) 1;_~lO~I~ wR [[(F (n-p) E IF,,I [( ) tmax(Fo5,O) 0 S 1=1 2F. 08033 2 (13), [Na(OEt ) (thf)0.5][W(H)(11-Me2C=NAr) 2-cyclo-(N[iPr]Ar)] Table 2.2. Crystallographic Data for (Ar[i-Pr]N)3W(Cl)2 2 15 ([Na(OEt,) 1 s(thf),,][14]) and OW(N[i-Pr]Ar)3 (15) [Na(OEt2) 1.5(th00 .5][14] 15 Reciprocal Net code / CCDC code 05012 / 823808 05087 / 823809 04104/823810 .5W,a 778.55a N Na O W ,b 6 5 .5 5 ' N 0 W,c 758.70c Empirical formula, FW (g/mol) C35H53C12N30 0 C82H118 6 2 4 2 1 6 C37H56 3 2 Color / Morphology Light green / Plate Amber / Prism Red I Block Crystal size (mm3) 0.13 * 0.10 *0.04 0.36 * 0.28 * 0.16 0. 12 *0. 10 * 0.08 Temperature (K) 100(2) 100(2) 194(2) Crystal system, Space group Monoclinic, P2 1/c Monoclinic, P21/c Orthorhombic, P2 1212 Unit cell dimensions (A, 0) a = 16.127(2), =90 a = 21.8199(9), a = 90 a = 16.4058(6), X= 90 b = 10.4213(12), = 107.747(4) b = 21.5109(9), = 112.5330(10) b = 17.4809(7), P = 90 c = 21.654(3), y = 90 c = 18.9705(8), 7=90 c = 12.7447(5), y = 90 Volume(A 3 ) 3466.1(8) 8224.4(6) 3655.0(2) 4 Density (calc., Mg/m3) 1.492 1.345 1.379 Absorption coefficient (mm-) 3.516 2.855 3.194 F(000) 1584 3416 1556 Theta range for data collection (0) 1.97 to 28.28 1.01 to 27.88 1.60 to 28.70 Index ranges -21 25(I)] R1 = 0.0409, wR2 = 0.0758 = 0.0392, wR2 = 0.1003 R1 = 0.0278, wR2 0.0550 R indices' (all data) R, = 0.0734,wR 2 = 0.0867 RI = 0.0496, wR2 = 0.1061 R1 = 0.0307, wR2 = 0.0558 3 Largest duff,peak and hole (e -A- ) 1.764 and - 1. 105 3.609 and -0.785 1.619 and -0.473 is in close aOne half of a disordered molecule of diethyl ether is present in the assymmetric unit. b Two unique molecules are present within the asymmetric unit. One anion [Na(thfX(OEt2)]+countercation. CA heavily disordered molecule of contact with a [Na(OEt 2 ]2+ countercation, the other independent anion is in close contact with a

tetahdrfurn asremoved from the asmer unit using the crystallographic routine SQUEEZE.17 d GF - zwFI12eR =- IF-IL

2 S; W 02F2+ P1.9 2F1 +max(.2,O) 2.9 References

[1] Tsai, Y.-C.; Stephens, F. H.; Meyer, K.; Mendiratta, A.; Gheorghiu, M. D.; Cummins, C. C. Organometallics2003, 22, 2902-2913. [2] Tsai, Y. C.; Cummins, C. C. Inorg. Chim. Acta 2003, 345, 63-69. [3] Bailey, R. A.; Kozak, S. L.; Michelsen, T. W.; Mills, W. N. Coord. Chem. Rev. 1971, 6, 407-445. [4] Becke-Goehring, M. Fortschr Chem. Forsch. 1968, 10, 207-237. [5] Bell, S. A.; Meyer, T. Y.; Geib, S. J. J. Am. Chem. Soc. 2002, 124, 10698-10705. [6] Bradshaw, J. S.; Nielsen, R. B.; Tse, P.-K.; Arena, G.; Wilson, B. E.; Dalley, N. K.; Lamb, J. D.; Christensen, J. J.; Izatt, R. M. J. Heterocycl. Chem. 1986, 23, 361-368. [7] Honeyman, C. H.; Lough, A. J.; Manners, I. Inorg. Chem. 1994, 33, 2988-2993. [8] Rivard, E.; Honeyman, C. H.; McWilliams, A. R.; Lough, A. J.; Manners, I. Inorg. Chem. 2001, 40, 1489-1495. [9] Schmidt, I.; Kynast, U.; Hanich, J.; Dehnicke, K. Z. Naturforsch., B: J. Chem. Sci. 1984, 39, 1248-1251. [10] Cordero, B.; G6mez, V.; Platero-Prats, A. E.; Rev6s, M.; Echeverrfa, J.; Cremades, E.; Barragin, F.; Alvarez, S. Dalton Trans. 2008, 2832-2838. [11] Tsai, Y. C.; Johnson, M. J. A.; Mindiola, D. J.; Cummins, C. C.; Klooster, W. T.; Koetzle, T. F. J. Am. Chem. Soc. 1999, 121, 10426-10427. [12] Veige, A. S.; Kleckley, T. S.; Chamberlin, R. M.; Neithamer, D. R.; Lee, C. E.; Wolczanski, P. T.; Lobkovskky, E. B.; Glassey, W. V. J. Organomet. Chem. 1999, 591, 194-203. [13] Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126, 13916-13917. [14] Clough, C. R.; Greco, J. B.; Figueroa, J. S.; Diaconescu, P. L.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126, 7742-7743. [15] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. [16] Sheldrick, G. M. Acta Cryst. A 2008, 64, 112-122. [17] van der Sluis, P.; Spek, A. L. Acta Cryst. A 1990, 46, 194-201. 82 CHAPTER 3

Oxygen Atom Transfer from MesCNO to (Ar[t-Bu]N)3MoP to form (Ar[t-Bu]N)3MoPO.

Contents 3.1 Introduction ...... 3.2 Results and Discussion ...... 3.2.1 Thermodynamic Experiments......

3.2.2 Kinetic Studies of MesCNO with (Ar[t-Bu]N) 3MoP (16) ......

3.2.3 Computed Mechanism for the Reaction of PhCNO and PMe3 ..... - 3.3 Conclusions ...... 3.4 Experimental Section ...... 3.4.1 General Considerations .. ...

3.4.2 Synthesis of (Ar[t-Bu]N) 3MoPO (17) using MesCNO ...... 3.5 References ......

3.1 Introduction

In 1997, it was found that the terminal phosphide complex (Ar[t-Bu]N)3MoP (16) reacts with dimethyldioxirane (DMDO) to form the unique, terminal phosphorus monoxide complex (Ar[t-Bu]N)3MoPO (17).1 Since then, studies of compound 17 have been severely limited by the properties of DMDO. DMDO is synthesized by reaction of Oxone®

(KHSOs -0.5KHSO 4 -0.5K 2SO 4 ) with acetone to form a dilute solution (ca. 0.05 to 0.20 mmol/g) of DMDO in acetone. 2 Solutions of DMDO are invariably wet and attempts to dry them affect the potency of DMDO. Adding to the issues with DMDO is the fact that it readily decomposes even when stored at -35 'C. Compound 17 has similar stability issues-decomposing over a period of weeks in the solid state at low temperature. In order to study compound 17, one is required to make DMDO frequently to make small batches of the desired complex.

As phosphide 16 does not react favorably with more common oxygen atom transfer (OAT) reagents (e.g. pyridine N-oxide (py-+O), trimethylamine N-oxide (Me 3N- O), propylene oxide, ozone, iodosylbenzene, etc.), 3 we sought a new route to 17. While py-+O does not react with phosphines, despite the OAT reaction being favorable by over 70 kcal/mol, it is known that MesCNO will oxidize tertiary phosphines.4,5 MesCNO is also an attractive OAT reagent as it is relatively easy to prepare and recrystallize 6 and is stable below 30 'C for a period of months. The bond dissociation energy (BDE) and enthalpy of formation of MesCNO in the solid state and gas phase have all been accurately determined.7-9 The N-O BDE in MesCNO has been measured to be 52.3 kcal/mol, making it a more potent OAT reagent that py-+O which has a BDE value of 63.3 kcal/mol. 0 Indeed, MesCNO is a more potent OAT reagent than oxygen itself (Reaction 3.1).

2MesCNO - 2MesCN+0 2 AH -14.5kcal/mol (3.1)

When treated with mesitylnitrile oxide (MesCNO), 6 16 reacts to cleanly form 17 in 57% isolated yield (Scheme 3.1). As MesCNO has seen only limited use in the literature as an OAT reagent,4,5 we decided to collaborate with other groups to learn as much about this potent reagent as possible. The work in this chapter was done in collaboration with Carl Hoff and coworkers at the University of Miami (thermochemical measurements), Elena Rybak-Akimova and coworkers at Tufts University (kinetic measurements) and Manuel Temprado and coworkers at the University of Alcali (Density Functional Theory (DFT) calculations).

0 Il PP MesCNO Mo ~.Mo Ar[t-Bu]N '" Nt CH CI Ar[t-Bu]N '/ Nt j N[t-Bu)Ar 2 2 i N[t-Bu]Ar Ar[t-Bu]N -97 to 25 "C Ar[t-Bu]N -MesCN 57%

Scheme 3.1. Synthesis of (Ar[t-Bu]N)3MoPO (17) with MesCNO 3.2 Results and Discussion

3.2.1 Thermodynamic Experiments

MesCNO reacts rapidly with compound 16 in toluene to yield the dark-purple, phosphorus monoxide complex 17 and MesCN in quantitative yield (as determined by NMR spectroscopy). The experimental enthalpies of reaction of MesCNO with 16 and several phosphines were determined by solution calorimetry as described in Scheme 3.2. Results for thermochemical measurements of

AnP + MesCNO are summarized in Table 3.1. The BDE of Ph3P=O determined by Hoff et al. is in good agreement with the gas phase value reported by Domalski " of 135.4 ± 2.8 kcal/mol, thus lending credence to the validity of the values.

00(a o-N-C N-C

An P AP=On Scheme 3.2. Generic reaction of AnP with MesCNO

Table 3.1. Enthalpies of oxygen atom transfer (OAT) reactions of AnP with MesCNO in toluene solution and derived P=O bond dissociation (BDE) energies. AnP AH (kcal/mol)a P=O BDE (kcal/mol)ab 16 -56.6 ± 0.8 108.9 PPh 3 -79.9± 1.7 132.2

PCy3 -85.3 ±1.8 137.6 PMe 3 -86.2±1.7 138.5 a Data collected by Hoff and coworkers b BDE values are considered accurate to ±3 kcal/mol.

The experimental calorimetric values derived from A.P + MesCNO were compared to computational studies performed by Temprado et al. DFT calculations were performed using Gaussian at the M05-2X/6-31 1G(3df,2p) level for all species investigated. Where applicable for smaller molecules, ab initio calculations using the G3 methodology were performed as well. In all cases where both DFT and ab initio calculations were performed, the results were in good agreement. To determine the X-0 BDE's, the enthalpy of reaction (AHx/xo) were calculated for all compounds using reaction 3.2. From there, the X-0 BDE corresponding to reaction 3.3 was calculated using Eqn. 3.4, where AfH0 (O(g)) = 59.55 kcal/mol.12

X(g) + 1/202(g) -> XO(g) AHx/xo (3.2)

XO(g) -+ X(g) + O(g) BDEx-o (3.3) BDEX-O = -AHx/xo + AfHm(O(g)) (3.4)

Calculated X-0 BDE's are collected in Table 3.2 and compared with available experimental data where available.

Table 3.2. Experimental and computed X-0 bond dissociation energy (BDE) values. BDEcalc - BDEexp in parentheses. X BDEcalc (kcal/mol)" BDEcaic (kcal/mol)a BDEexp (kcal/mol) M05-2X/6-31 1G(3df,2p) G3 NN 38.0 (-2.0) 40.8 (0.8) 40.013 PhCN 48.5 50.0 MesCN 48.1 (-5.0) 48.8 (-4.3) 53.110 Py 61.7 (-1.6) 63.6 (0.3) 63.310 NP 74.6 78.3 - Me2S 88.3 (1.7) 86.0 (-0.6) 86.6 13 PP 90.1 90.1 16 105.9 (-3.0) - 108.9" CO 128.2 (1.0) 128.7 (1.5) 127.2 13 PhNC 131.8 131.0 -

PPh 3 130.2 (-2.0) 132.2' 137.6" PCy3 135.3 (-2.3) PMe 3 133.0 (-5.5) 134.9 (-3.6) 138.5" a Calculations performed by Manuel Temprado and coworkers b Data collected by Hoff and coworkers

In all cases, calculated values for BDE's were in good agreement with available experimental data. The greatest discrepancy for calculated vs. experimental was observed with the DFT calculations for MesCNO (difference of 5.0 kcal/mol). The combination of experimental and calculated data shows that the P=O bond in 17 is significantly weaker than for phosphine oxides

(23.3 kcal/mol weaker vs. Ph3P=O). Although compound 17 is thermodynamically competent to oxidize PPh3, PMe3 or PCy 3 to their corresponding phosphine oxides, this reaction does not 3 proceed. 17 oxidizes Mo(N[t-Bu]Ar) 3 to O=Mo(N[t-Bu]Ar) 3 as the Mo-O BDE in O=Mo(N[t-

Bu]Ar)3 is 155.6(16) kcal/mol. 1 Since the enthalpy of formation of PO is known,1 4 by combining the Mo-P strength in 16 15 with the oxidation of 16 to 17, one can calculate the enthalpy of coordination of PO to Mo(N[t-Bu]Ar) 3 as shown in Scheme 3.3. Scheme 3.3 also includes data for the lighter pnictogen congeners.16 t-Bu t-Bu t-Bu ... t-Bu Mo-N t-Bu Mo-N Ar, No Ar -140.8 E= P Ar - Ar -150.9 E= N + E+ 0 + EEO

-92.2 E =P -60.2 E= P ii. iv. -155.3 E=N -82.5 E= N

O E -108.9 E =P t-Bu III t-Bu -78.1 E =N t-Bu \ II /t-Bu N pp ,N N Mo / t-Bu -N Ar Ar t-Bu N Ar Ar \ Ar Ar +0

Scheme 3.3. Measured and derived enthalpies for (Ar[t-Bu]N) 3MoEO (E = P, N). All data are reported in kcal/mol. Data for E = P were collected by Hoff and coworkers. For data relevant to (Ar[t- Bu)N)3MoNO binding and the (Ar[t-Bu]N) 3MoEN bond strength, see Cherry, et al. 16 The OAT thermochemical data shown in Scheme 3.3 tell nothing of the bonding nature of the species, only the net energy involved. The BDE's determined for (Ar[t-Bu]N) 3MoEO give no insight into whether the compounds should be viewed as (Ar[t-Bu]N) 3Mo=E=O or (Ar[t- Bu]N) 3Mo-E=O. The formal bonding in (Ar[t-Bu]N) 3MoEO is certainly up for debate as both models may be correct to varying degrees. Crystallography shows that the Mo-N bond distance in

(Ar[t-Bu]N)3Mo=N is shorter than in the corresponding nitrosyl complex, (Ar[t-Bu]N) 3MoNO.16 Conversely, the Mo-P distance in phosphide (Ar[t-Bu]N) 3MoP is longer than the PO complex 17.1 Computations by Frenking and coworkers support the bonding description in 16/17 suggesting that the Mo-P bond is stronger in compound 17 vs. the phosphide 16.17 The calorimetric results compiled by Hoff, et al. are in direct conflict with those reported by Frenking as shown in Reactions 3.5 and 3.6.

16 -+ Mo(N[t-Bu]Ar) 3 + P AH = +92.2 kcal/mol (3.5)

17 -+ Mo(N[t-Bu]Ar) 3 + P=O AH = +60.2 kcal/mol (3.6)

Additionally, the P-O bond length in complex 17 was determined to be 1.49(2) A by X-ray crystallography. The above bond length is similar to the d(P-O) of gas phase P=O which has been determined by IR diode laser spectroscopy to be 1.476370(15) A.18', 9 The P-O bond length 20 for Cy3PO is 1.490(2) A as determined by X-ray crystallography. In light of the structural comparisons involving compound 17 (BDEpO = 108.9 kcal/mol) vs. P=O (BDEpO = 140.8 4 kcal/mol)1 and Cy 3PO (BDEpO = 137.6 kcal/mol), it appears that bond lengths are not always indicative of bond strength.

Lastly, it is of note that despite being favorable by ca. 25 kcal/mol (based upon data by Hoff and coworkers), Reaction 3.7 does not proceed.

(Ar[t-Bu]N)3Mo=N +MesCNO -4 (Ar[t-Bu]N) 3MoNO +MesCN (3.7)

Oxidation of pyridine and tetrahydrothiophene (Reactions 3.8 and 3.9, respectively) does not proceed either, despite being thermodynamically favorable as well.

C5H5N +MesCNO -- py-MO +MesCN (3.8)

THT +MesCNO -M THT--+O +MesCN (3.9)

Hoff speculates that "successful OAT chemistry with MesCNO by non-metals may require a strong nucleophile capable of forming an adduct, as well as the presence of additional vacant orbitals for utilization in valence expansion to form the cyclic intermediate necessary for OAT." 21 At present, the only successful OAT reactions found for MesCNO include oxidation of tertiary phosphines, compound 16 to form 17. Further reaction studies are currently underway and preliminary data suggest that MesCNO will oxidize V(N[t-Bu]Ar) 3 to the corresponding vanadium trisanlide oxo species.22 3.2.2 Kinetic Studies of MesCNO with (Ar[t-Bu]N)3MoP (16)

Kinetic studies of the reaction of compound 16 with MesCNO were undertaken by Elena Rybak- Akimova and Meaghan Germain at Tufts University. An absorption at ?= 550 nm in the UV/Vis spectrum was attributed to compound 17 and was used to monitor formation of 17 (Figure 3.1). The experiments were conducted under pseudo-first order conditions using a large excess of MesCNO in CH 2Cl2. The kinetic traces were fit to a single exponential function and rate constants were obtained, kobs = ki [MesCNO] (Figure 3.2). Identical OAT kinetic experiments were repeated in toluene ([MesCNO] = 50mM and [16] = 0.6 mM) at five different temperatures (15, 22, 30, 38 and 45 'C) to obtain activation parameters. For the experiments at the same temperature, results for kobs were very similar for both CH2Cl2 and toluene.

1.5

1.0

-C

0.51. 450 500 550 600 \(nm) Figure 3.1. Final UV/Vis spectrum taken at 15 'C with 25 mM MesCNO, [16] 0.75 mM), t = 60 min showing formation of 17 at k = 550 nm. Data were collected by Rybak-Akimova and Germain.

An Eyring plot was generated using the temperature dependent data in toluene solution and from which, the activation parameters of the reaction 16 + MesCNO were calculated (Figure 3.3) providing AH* = 11 kcal/mol and AS* = -27 cal/mol-K.

Reactions of MesCNO and various phosphines were also studied by FT-IR spectroscopy at 1:1 ratios. Attempts to identify intermediate species in the reactions of MesCNO with AnP were unsuccessful, even when the experiments were carried out at -40 'C. "FT-IR spectral data for the reaction of PCy 3 and MesCNO displayed isosbestic points and the rate of decay of MesCNO 2 and the rate of buildup of Cy 3PO were equal and opposite in sign." 1 The inability to identify an intermediate is indicative of either a single-step OAT, or a multi-step OAT reaction where the 15x10 -I. * 15 C * 22 *C A 30 *C

0 5x10

0 | I | | | | I 0.00 0.02 0.04 0.06 0.08 0.10 0.12 [MesCNO] M Figure 3.2. The plot of the observed rate constant, kobs, vs. the concentration of MesCNO (taken in excess). The data were acquired for various temperatures (15-30 C) over a range of [MesCNO] = 25-100 mM, [16] = 0.6 mM in cuvette. Second-order rate constants: k15 oc = 0.11 M- s-1; k22oC = 0.15 M-ts- ; k30 oC = 0.20 M 1s-1. Data were collected by Rybak-Akimova and Germain.

-6

-7

-8

-9-"

3.1 3.2 3.3 3.4 3.5x1 0-3

1 /T (K') Figure 3.3. Eyring plot for the reaction between MesCNO and compound 16 in toluene. Data were collected by Rybak-Akimova and Germain. formation of the first transition state is the rate limiting step. Calculations performed by Manuel Temprado and coworkers suggest that the latter mechanism is in play for the OAT to phosphines (see Section 3.2.3).

The kinetic values obtained by Rybak-Akimova, et al. are typical for activation parameters of oxygen atom transfer reactions to phosphines. Published activation parameters of AH* = 10 kcal/mol and AS* = -21.6 cal/mol-K were reported for an oxygen-atom transfer from Mes 3 Ir=O to

PPh3 .23 A Re V=O complex transferred an oxygen atom to PPh3 with activation parameters of AH = 11.7 kcal/mol and ASt = -27.5 cal/mol-K. 24 Similarly, AHt = 10.8 kcal/mol and AS* = -27.2 cal/mol-K were reported for Mo VIL 20 2 (L = N,N'-disubstituted dithiocarbamate) transferring an 25 O-atom to PPh3 - The negative activation entropy values in the reaction of compound 16 with MesCNO describe a rate limiting binding process in which degrees of freedom are lost in the 26 27 transition state. ,

Table 3.3. Rate constants at 20.3 'C, derived activation parameters and reaction enthalpies for reactions of AnP + MesCNO AnP k(M-Is-1)a AH* (kcal/mol)a AS: (cal/mol. K)a AHO (kcal/mol)b

Cy 3 P 0.45 8.9 -30 -85.311.8 (p-tolyl)3 P 0.20 9.3 -30 -79.9 ± 1.7c 16 0.07d 10.9d -27d -56.6 ±0.8

0.15,ef 0 .0 4 e,g 1ie -27e a Data were collected by Rybak-Akimova and Germain. b Data were collected by Hoff, Cai, Majumdar and Captain. c The value for Ph3P is given as it is expected to be similar to (p-tolyl)3P. d These values were obtained by FT-IR spectroscopy. e These values were obtained by UV/Vis spectrophotometry. f These data were measured in CH 2Cl 2 solution at 22 'C. g These data were measured in toluene solution at 22 'C.

3.2.3 Computed Mechanism for the Reaction of PhCNO and PMea

As a model for the OAT reaction of MesCNO with compound 17, Manuel Temprado and coworkers investigated the hypothetical Reaction 3.10 computationally.

PMe3 + PhCNO -- Me 3P=O + PhCN (3.10)

Investigation of the frontier molecular orbitals for PMe 3 and PhCNO calculated at the M05-2X/6- 311 G(3df,2p) level (Figure 3.4) suggest that a possible reaction pathway would consist of attack of the HOMO of PMe3 on the LUMO of PhCNO. The HOMO of PMe3 as shown in Figure 3.4 consists of essentially a lone pair on phosphorus, while the LUMO on PhCNO appears to be a delocalized 7t* orbital with a significant lobe on the carbon of the nitrile oxide functional group. As a logical starting point, nucleophilic attack of PMe3 on the carbon of PhCNO was investigated. U',r

C D Figure 3.4. Computed frontier orbitals for PhCNO and PMe3. A. HOMO of PhCNO; B. LUMO of PhCNO; C. HOMO of PMe 3; D. LUMO of PMe3. Figure provided by M. Temprado.

An intrinsic reaction coordinate (IRC) calculated by Temprado, et al. supports the hypothesis of nucleophilic attack of the phosphine on the "CNO" carbon (Figure 3.5). The first calculated transition state (TS1, see Figure 3.6) has the highest energy along the IRC, this being consistent with the inability to observe any intermediates by FT-IR in the reaction of PMe 3 with MesCNO. The calculated kinetic terms for PhCNO + PMe3 , AHe = 11.6 kcal/mol and AS*c = -38.1 cal/mol.K, are in good agreement with those for the experimental FT-IR data for the reaction of MesCNO with

(p-tolyl)3P (AH* = 9.3 kcal/mol and AS* = -30 cal/mol-K). The first intermediate (Int1, Figure 3.6) is essentially an adduct between PMe 3 and PhCNO bound at the "CNO" carbon. There is also an electrostatic interaction between the electropositive phosphorus (formal charge = +1) and electronegative oxygen of the NO functional group (formal charge = -- 1). from Intl, the reaction proceeds through a shallow transition state (TS2) to a second, four-membered ring intermediate

(Int2). The overall reaction is driven by the formation of the strong P=O bond in Me3P0 (AHe = -84.5 kcal/mol).

3.3 Conclusions

MesCNO has been shown to be a powerful OAT reagent that as of yet has seen limited use in that role. The use of MesCNO offers a convenient route to the synthesis of 17, which until now was a laborious process due to the need for DMDO. Through our collaboration with Carl Hoff, Elena Rybak-Akimova and Manuel Temprado, we have gained significant insight into the mechanism of how MesCNO acts as an OAT reagent. TS1 11.6 (-38.1) -860.82- [23.0] TS3 t TS2-2.8 -860.84 -1028 1043 R -(-50.8) = -860.86 0.0 [A (d (0.0) nt 0 -860.88- [0.0 Intl -12.7 0 >%-16.3 (-45.0) G (-46.5) [0.7] 0 0 -860.90- [-2.5]

-860.92-

-860.94-

-860.96 P -84.5 -860.98 (-2.1) -860.98-[-83.9] -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 Reaction Coordinate / amu' 2 bohr Figure 3.5. Intrinsic reaction coordinates calculated at the M05-2X/6-31 1G(3df,2p) level for PhCNO + PMe 3. Relative enthalpies (in kcal/mol, basis set superposition error (BSSE) corrected), entropies between parentheses (in cal/mol-K) and Gibbs energies between brackets (kcal/mol) at T = 298 K. Figure provided by M. Temprado.

1.217 A 1.238 A 1.408 A 1.699 A 2.739 A 1 - 2.531 A 1.761 A 191A1.574 A 2.5A

TS1 TS2 TS3

1.300 A 1.326 A 1.449A1274 A 2.044 A

.834 A 1.688A 1.99oA -

Intl Int2 Figure 3.6. Computed structures of intermediates and transition states for the reaction of PhCNO and PMe 3 at the M05-2X/6-31 1G(3df,2p) level. See Figure 3.5 for relative energies. Figure provided by M. Temprado. 3.4 Experimental Section

3.4.1 General Considerations

Unless otherwise stated, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere. 28 6 (Ar[t-Bu]N) 3MoP and MesCNO were prepared as previously published. Methylene chloride, acetonitrile and THF were dried and deoxygenated using a system built by SG Water USA, LLC 29 (http://www.glasscontoursolventsystems.com). C6D6 was purchased from Cambridge

Isotope Laboratories, purified by distillation off of CaH 2 and stored over 4A molecular sieves. Celite, alumina and 4 A molecular sieves were dried under reduced pressure overnight at a temperature above 200 0C. 'H, 13C and 31P NMR spectra were recorded on Varian Mercury-300, Varian INOVA-500 or Bruker AVANCE-400 spectrometers. 'H and 13C NMR chemical shifts are 31 reported with respect to internal solvent (C6D6 , 6 = 7.16 and 128.39 ppm, respectively). P NMR chemical shifts are reported with respect to an external reference (85% H3P0 4, 6= 0.00 ppm).

3.4.2 Synthesis of (Ar[t-Bu]N)3MoPO (17) using MesCNO

Compound 16 (299 mg, 0.456 mmol) was dissolved in 5 mL CH 2Cl2 and cooled to -35 'C. A solution of MesCNO (84 mg, 0.52 mmol, 1.1 equiv.) in 2 mL CH2Cl2 was cooled to -35 'C as well. The cooled MesCNO solution was added rapidly to the golden-brown, stirring solution of 16.

An additional 1 mL of chilled CH 2Cl2 was used to insure that all MesCNO had been transferred to the reaction mixture. Upon warming to room temperature, the reaction mixture changed from golden brown to deep purple in color. The mixture was stirred for one hour after mixing to insure complete consumption of 16. After one hour of reaction time, the mixture was filtered through a bed of Celite, the Celite pad washed with 5 mL of -35 'C CH 2C12 and the filtrate reduced in volume under reduced pressure to ca. 50% of its original volume. The purple reaction mixture was then partially frozen and 30 mL of thawing acetonitrile was added to precipitate the desired product. The cold mixture was filtered through a medium-porosity, fritted-glass funnel and the purple solids were washed with 10 mL of thawing acetonitrile. The solids were dried to constant volume under reduced pressure yielding compound 17 as a deep-purple powder (175 mg, 0.261 mmol, 57%). 'H and 31p NMR data agreed with previously published data. I 3.5 References

[1] Johnson, M. J. A.; Odom, A. L.; Cummins, C. C. Chem. Commun. 1997, 1523-1524. [2] Adam, W.; Bialas, J.; Hadjiarapoglou, L. Chem. Ber 1991, 124, 2377. [3] Johnson, M. J. A.; Ph.D. thesis; Massachusetts Institute of Technology; 1998. [4] Grundmann, C.; Frommeld, H.-D. J. Org. Chem. 1965, 30, 2077-2078. [5] Sicard, G.; Baceiredo, A.; Crocco, G.; Bertrand, G. Angew. Chem. Int. Ed. 1988, 27, 301-302. [6] Barybin, M. V.; Diaconescu, P. L.; Cummins, C. C. Inorg. Chem. 2001, 40, 2892-2897. [7] Acree Jr., W. E.; Tucker, S. A.; Zvaigzne, A. I.; Meng-Yan, Y.; Pilcher, G.; Ribeiro Da Silva, M. D. M. C. J. Chem. Thermo. 1991, 23, 31-36. [8] Acree Jr., W. E.; Simirsky, V. V.; Kozyro, A. A.; Krasulin, A. P.; Kabo, G. J.; Frenkel, M. L. J. Chem. Eng. Data 1992, 37, 131-133. [9] Acree Jr., W. E.; Sevruk, V. M.; Kozyro, A. A.; Krasulin, A. P.; Kabo, G. J.; Frenkel, M. L. J. Chem. Eng. Data 1993, 38, 101-104. [10] Acree, W. E.; Pilcher, G.; Ribeiro da Silva, M. D. M. C. J. Phys. Chem. Ref Data 2005, 34, 553-572.

[11] Kirklin, D. R.; Domalski, E. S. J Chem. Thermo. 1988, 20, 743-754; O=PPh3 = 135.4 + 2.8 kcal/mol. [12] Cox, J. D.; Wagman, D. D.; Medvedev, V. A. DODATA Key Valuesfor Thermodynamics; Hemisphere, New York, 1989. [13] Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571-589. [14] http://webbook.nist.gov/chemistry. [15] Stephens, F. H.; Johnson, M. J. A.; Cummins, C. C.; Kryatov, 0. P.; Kryatov, S. V.; Rybak-Akimova, E. V.; McDonough, J. E.; Hoff, C. D. J. Am. Chem. Soc. 2005, 127, 15191-15200. [16] Cherry, J. P. F; Johnson, A. R.; Baraldo, L. M.; Tsai, Y.-C.; Cummins, C. C.; Kryatov, S. V.; Rybak-Akimova, E. V.; Capps, K. B.; Hoff, C. D.; Haar, C. M.; Nolan, S. P. J. Am. Chem. Soc. 2001, 123, 7271-7286. [17] Caramori, G. F.; Frenking, G. Theor. Chem. Acc. 2008, 120, 351-361. [18] Butler, J. E.; Kawaguchi, K.; Hirota, E. J. Mol. Spectrosc. 1983, 101, 161-166. [19] Qian, H.-B. J. Mol. Spectrosc. 1995, 174, 161-166. [20] Davies, J. A.; Dutremez, S.; Pinkerton, A. A. Inorg. Chem. 1991, 30, 2380-2387. [21] Cai, X.; Majumdar, S.; Frutos, L. M.; Temprado, M.; Clough, C. R.; Vai, A. T.; Cummins, C. C.; Germain, M. E.; Rybak-Akimova, E. V.; Captain, B.; Hoff, C. D.; Manuscript in preparation. [22] Cai, X.; Majumdar, S.; Silvia, J. S.; Cummins, C. C.; Germain, M. E.; Rybak-Akimova, E. V.; Captain, B.; Hoff, C. D.; unpublished results. [23] Jacobi, B. G.; Laitar, D. S.; Pu, L. H.; Wargocki, M. F.; DiPasquale, A. G.; Fortner, K. C.; Schuck, S. M.; Brown, S. N. Inorg. Chem. 2002, 41, 4815-4823. [24] Das, S.; Chakravorty, A. Eur. J. Inorg. Chem. 2006, 2285-2291. [25] Unoura, K.; Yamazaki, A.; Nagasawa, A.; Kato, Y.; Itoh, H.; Fukuda, Y. Inorg. Chim. Acta 1998, 269, 260-268. [26] Eyring, H. Chem. Rev. 1935, 17, 65-77. [27] Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry, Part A: Structure and Mechanisms, 4th ed.; Kluwer Academic/Plenum Publishers, New York, 2000. [28] Laplaza, C. E.; Davis, W. M.; Cummins, C. C. Angew. Chem. Int. Ed. Eng. 1995, 34, 2042-2044. [29] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. 96 CHAPTER 4

Synthesis of Early Metal Trimetaphosphate Complexes

Contents 4.1 Introduction ...... 97

4.2 Synthesis and Characterization of [PPN] 3[(P 30 9)Mo(CO)3] ([PPN] 3 [18]) . . 98

4.3 Synthesis and Characterization of (P30 9)VaO ...... 102 4.4 Conclusions ...... 105 4.5 Experimental Section ...... 106 4.5.1 General Considerations...... 106

4.5.2 Synthesis of [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3 [18]) from (CH 3CN) 3Mo(CO) 3...... 107

4.5.3 Synthesis of [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3[18]) from Mo(CO) 6 . . .. 10 7

4.5.4 Attempted synthesis of (P30 9)V O (19) ...... 108 4.6 DFT Calculations ...... 108 4.7 Crystallographic Structure Determinations ...... 108

4.7.1 X-ray Crystal Structure of [PPN] 3 [(P3 0 9 )Mo(CO) 3] ([PPN 3][18]) .... 108 4.8 References ...... 112

4.1 Introduction

3 Trimetaphosphate (P 30 9 -) is an attractive, inorganic, tripodal ligand for transition-metal chemistry. 1-15 Like others before us, we sought to develop trimetaphosphate, transition- metal complexes as homogeneous, model complexes for heterogeneous metal-supporting oxide catalysts. 16-20 The first isolable metal trimetaphosphate complexes ([N(n-Bu) 4] 2[(P30 9)M(CO)31, M = Re, Mn) were reported by Klemperer and co-workers in 1981 (Figure 4.1).' Although not structurally characterized, the [(P30 9)M(CO) 3]2- anions were hypothesized to conform to C3v symmetry based upon infrared spectroscopy. Since that initial report, there have been a handful of other reported trimetaphosphate transition-metal complexes-both from the Klemperer lab2-7 and others.8-15 Despite being a readily available, tridentate, trianionic ligand, P309 seems to be underutilized as reflected by the chemical literature. Perhaps one of the reasons trimetaphosphate coordination chemistry is underrepresented in the literature is due to its propensity to hydrolyze forming tripolyphosphate in water. 21 In seeking to introduce P303- into molybdenum chemistry, we accordingly employed an anhydrous synthesis procedure modeled after that of Klemperer (Figure 4.1). Reported herein is the synthesis 3 and structure of [PPN] 3[(P 30 9)Mo(CO) 3] ([PPN] = [Ph3P=N-PPh3]+), an example of a K - trimetaphosphate, tricarbonyl, transition-metal complex.

] O [N(n-Bu) 4 2 0 11 - CO

[(MeCN) 3M(CO) 3][PF6]

+ 0 0______[N(n-Bu) 4]3[P309] 1,2-dichloroethane 1 0 790 CP -[N(n-Bu) 4][PF6] O \0 I\7QO

M=MnorRe 0 Scheme 4.1. Synthesis of [N(n-Bu)4]2 (P30 9)M(CO) 3]-1

4.2 Synthesis and Characterization of [PPN] 3((P30g)Mo(CO) 3 ([PPN]3[18])

22 Initially, (CH 3CN)3Mo(CO) 3 was slurried in methylene chloride and treated with 23 [PPN] 3 [P3 09] -H20. After stirring the solution for one hour, the golden-yellow mixture formed a homogeneous solution yielding [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3[18]) in 91% isolated yield by precipitation with diethyl ether. Alternatively, a streamlined synthesis has been developed by simply refluxing Mo(CO) 6 and [PPN] 3 [P3 09] -H20 in acetonitrile for four hours (Figure 4.2). (CH CN) Mo(CO) is presumably 3 3 3 generated in situ where it reacts with [PPN]3 [P3 09] -H20 to form the desired product in 77% yield. A 31P NMR spectrum of the product in methylene chloride shows two peaks at 6 = 21.7 and - 11.3 ppm in a 2:1 ratio, corresponding to [PPN] + and 183-, respectively. These spectroscopic data [PPN] 3 Na3P309 H20 Na3P309-6 H 0 NaCI 3[PPN]CI 2 33% 0 Co H20, 60 C - 3 NaCI 67%

[PPN ]3[P309]-H20 Mo(CO) 6 refl uxing CH3CN 4h 0 -3CO orange powder, 77% Scheme 4.2. Syntheses of Na3 P 0 H2024, 23 3 9 .6 [PPN] 3 [P30 9] H 20 and [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3[18]).

3 1 are in good agreement with trimetaphosphate P NMR shifts of [(P30 9)M(CO) 3]2- complexes reported by Klemperer (M = Re, 6 = -11.1 ppm; M = Mn, 6 = -9.8 ppm). 183- gives the expected infrared spectrum for a C3, symmetric tricarbonyl species with the A1 -carbonyl stretch at VCO = 1883 cm- 1 (s) and the E stretch at vCo = 1723 cm- 1 (s, br). Values for 183- differ from those for = [(P 3 0 9 )M(CO) 3] 2- (M = Re, VCo = 2018 cm-1 (s), 1885 cm- 1 (s, br); M = Mn, VCo 2034 cm- (s), 1913 cm-1 (s, br)) due to electronic differences between the group 6 trianion vs. the group 7 dianions. The less energetic VCO stretches observed in 183- are expected from the more electron rich molybdenum center when compared to the group 7 congeners. 25 A recent article written by Earley reported calculated structural aspects and vCo stretching frequencies for a series of hypothetical 6 d [(p 30 9)M(CO) 3]"n compounds (where M = the twelve metals that make up group 6-9).26 The 3 calculated vCo stretching frequencies for 18 - were given as 1699 (A1 ) and 1822 (E) cm-1. The differences between calculated and experimental vCo frequencies in 183~ are very similar to those for [(P3 0 9 )M(CO) 3] 2- where M = Mn and Re (Table 4.1).

Table 4.1. Experimental and calculated vCo stretching frequencies (cm-') for [(P0 9 )M(CO),]"- Observeda Calculatedb Difference

18'- A1 1883 1822 61 E 1723 1699 24 2 [(P3 0 9)Mn(CO) 3 ] - A1 2034 1973 61 E 1918 1886 32 2 [(P3 0 9 )Re(CO) 3 ] - A 1 2018 1960 58 E 1885 1848 37

a [(p30 9)Mn(CO) 3] 2- and [(P3 0 9)Re(CO) 3 ]2~ data taken from Besecker, et al. 1 b calculated data taken from Earley 26 Crystals of [PPN]3[18] were grown via vapor diffusion of tetrahydrofuran into a saturated acetonitrile solution at room temperature and subjected to an X-ray crystallographic study (Figure

4.1). The trimetaphosphate salt crystallizes in P21 /c as a non-merohedral twin. The asymmetric unit consists of the salt as well as a disordered tetrahydrofuran molecule. There is also a positional disorder at the 183- anion caused by a C 2 rotation. The two components of the positional disorder exist in a ca. 95:5 ratio (see Section 4.7.1 for full crystallographic details). 01

03 C1 C3 02 C2 Mol

06 05 04 P3 08P 012 T\

09 07 011 010

Figure 4.1. Solid-state structure of anion [(P30 9)Mo(CO) 3 3- (183-) with thermal ellipsoids at the 50% probability level. Averaged bond lengths (A): Mo-C 1.950(3), C-O 1.152(5), Mo-OGM 2.283(3), P-OM 1.493(3), P-Oring 1.625(3), P-Oterminal 1.465(3).

100 03

01 C\ Mol

P1 P3 P2 P5

IN1 P4

P6 N2 ~ P9 N3 P7

Figure 4.2. Solid-state structure of [PPN] 3 (P30 9)Mo(CO) 3] ([PPN] 3[18]) with thermal ellipsoids at the 50% probability level. Hydrogen atoms and a disordered THF molecule have been omitted for clarity.

101 As expected from 31P NMR and FT-IR spectroscopies, the solid-state structure of 183- exhibits near-perfect C3, symmetry with all equivalent bond lengths varying by no more than 0.028 A. To the best of our knowledge, [PPN] 3 [18] is the first example of a crystallographically characterized, 3 transition metal [(P30 9)M(CO) 3]' species. Other K -trimetaphosphate complexes are known, but they are relatively rare. A search of the Cambridge Structural Database and the chemical literature 3 only returns thirteen such K -P30- complexes utilizing seven different transition metals (Hf 9, V 12, Fe 8, Ru 2,5,10,14,15, Rh2, Ir 3, and Pt11.

We compared the experimentally derived metrical parameters from our structure of 183- to those predicted by Earley. 26 Earley has calculated the C-O, Mo-C, Mo-OM and P-OM bond lengths for complexes of the type [(P30 9)M(CO) 3]"- (where "Om" refers to the oxygen atoms of the trimetaphosphate that are bound directly to the metal).

As shown in Table 4.2, with the exception of Mo-C bond lengths, the theoretical results overestimate the bond lengths that are revealed by the crystal structure. Earley contends 26 that as the [(P30 9)M(CO) 3] - complexes become less electron rich (going from n = 3 to n = 0) that C- O bond lengths decrease in accordance with a decreasing amount of back bonding to the K*orbital of CO. A second reason for this effect stems from decreasing d-orbital energies going from group 6 to group 9 leading to poorer overlap with the CO K* orbitals. 26 M-OM bond lengths are simply predicted to be dependent upon the atomic radius of the metal. Predictions of M-C bond lengths were more complicated as they are dependent upon all of the aforementioned factors. M-C bond lengths show an increase within a period for second- and third-row metals vs. first-row as well as a local minimum within each row at group 7.

Table 4.2. Experimental and calculated bond lengths (A) for anion 183- Experimental Calculateda C-O 1.152(5) 1.210 Mo-C 1.950(4) 1.942 MO-OM 2.283(3) 2.348 P-OM 1.493(3) 1.521 26 a calculated data taken from Earley

4.3 Synthesis and Characterization of (P30)V--O (19)

In order to expand the coordination chemistry of trimetaphosphate, attempts were made to synthesize a vanadium species. Described herein are the attempts to generate P3VO10 which is essentially an analogue of P4010 where one phosphorus atom has been replaced with a vanadium. When VOCl3 is treated with [PPN]3[P309] -H20 in acetonitrile at room temperature, the color of

102 31 the dark red solution of VOCl 3 lightens slightly. P NMR spectroscopy of an aliquot of the reaction mixture reveals complete consumption of starting material (8,o3- = -19.9 ppm) and formation of 3 1 a new product (,o3 ,) = -26.0 ppm, Figure 4.3). The singlet resonance of the P30 9 P NMR signal combined with its upfield location suggest C3 symmetry of a complex with the trimetaphosphate ring still intact. The proposed outcome of VOCl 3 + [PPN] 3 [P3 09] -H20 is the formation of the neutral, 3 K -trimetaphosphate vanadium oxo species, (P3 0 9 )V=O (19, Scheme 4.3). Further investigation of the reaction mixture by 51V NMR spectroscopy showed the presence of one major resonance at = -370 ppm (Figure 4.4).

v0c13 ______+ - 3[PPN]CI O [PPN]3[P 309]-H20 p ....p

O O 0

Scheme 4.3. Proposed synthesis of (P30 9)V-O (19)

Density functional theory (DFT) calculations performed with ADF 27 ,28 support the assignment of 19 as the product of the reaction of VOCl 3 with [PPN] 3 [P3 0 9]- H20. The proposed structure of 19 was input into ADF, constrained to C3, geometry, and a geometry optimization calculation was performed using all QZ4P basis sets while utilizing the OLYP functional 29,3 0 (Figure 4.5). The results of the geometry optimization calculation of 19 were then subjected to NMR shift calculations for 3 1P and 5 1V NMR spectroscopy. The calculated values for 19 were as follows: 31P NMR, 6 = -22 ppm; 5 V NMR, 8= -363 ppm (Table 4.3). For reference, 5 1V NMR spectra are referenced to neat VOCl 3 (6 = 0.0 ppm) and VOCl 3 comes at 6= -122 ppm when dissolved in acetonitrile.

Table 4.3. Experimental and calculated NMR shifts (ppm) for 19 Experimental Calculated 31p -26.0 -23 51v -370 -363

All attempts to date have not led to isolation of the compound 19 in pure form. All crystallization/precipitation attempts to date have failed to isolate 19. Due to the 3:1 stoichiometric ratio of byproduct:desired product, only isolation of pure [PPN]C1 in substoichiometric amounts has been achieved. Attempts to sublime the product 19 away from [PPN]C1 led to no observed

103 -29 ppm 2.60

30 20 10 0 -10 -20 -30 ppm 6.00 2.60 Figure 4.3. 31P NMR spectrum of VOC13 + [PPN] 3[P30 9] -H20

-370 ppm

D h i|

- 100 - 200 - 300 -400 -500 Ippm] Figure 4.4. 51V NMR spectrum of VOC13 + [PPN]31P30 9 ]- H20

104 sublimation up to 180 'C at ca. 0.5 Torr (decomposition was instead observed). When reaction mixtures are passed through an alumina plug, only [PPN]C1 is detected in the eluent. Synthesis of the desired complex in solvents other than acetonitrile lead to the formation of complex and intractable product mixtures.

Other attempts to generate 19 have similarly failed. Treatment of VOCl 3 with Ag 3P3O 10 shows complete consumption of the starting materials, but no formation of any discernible 31P or 51V containing species by NMR spectroscopy. Attempts at salt metathesis by treating reaction mixtures of VOCl 3/[PPN] 3 [P309] -H20 with NaBPh4 in order to form NaCl and [PPN][BPh 4] (which might be washed away from the desired product more easily) led to decomposition of 19. Other attempts, such as treating neat VOCl 3 with neat P4 0 10 did not lead to the formation of the desired product- which was not surprising, as by DFT calculations, Reaction 4.1 is 22 kcal/mol uphill (see Section 4.6 for full computational details).

VOCl 3 + P40 10 -+ 19 + POCl3 (4.1)

Silylation of trimetaphosphate was also attempted to in order to try the Reaction 4.2.

VOC13 + P3(OSiR 3)3 0 6 -- 19 + 3 R 3SiCI (4.2)

To date, all efforts to generate P3 (OSiR 3)30 6 (where R = Me, Ph) have led to the formation of complex mixtures of products.

Furthermore, DFT calculations predict a vanadium oxo stretch at vv0 = 1076 cm-1. Attempts to locate the Vvo stretch by FT-IR in reaction mixtures have not been successful as the presence of [PPN]C1 obscures the entire spectrum. Synthetic pursuits of compound 19 and its isolation are still currently underway.

4.4 Conclusions

Reported here is the synthesis and characterization of the first trimetaphosphate molybdenum complex. [PPN] 3 [18] is synthesized in a total of three steps from commercially available materials. 3 1 The structure of the 183- anion exhibits C3, symmetry as inferred from P NMR and FT-IR spectroscopic data. The species is also the first structurally characterized example of a complex of the type [(P30 9 )M(CO) 3] n.

Attempts to develop the trimetaphosphate system for vanadium chemistry have to date yielded inconclusive results. 31P and 5 1V NMR spectroscopy (in conjunction with NMR shift calculations using DFT) suggest that the vanadium oxo species, 19, is being formed. As of yet, the material has not been isolated in its pure form.

105 ------

/ 'm 0%0 pV116.6

126.2 7

'U'60 /7l.607

Figure 4.5. Calculated structure of (P30 9)V-O (19) with calculated bond lengths (A) and angles (O).

4.5 Experimental Section

4.5.1 General Considerations

Unless stated otherwise, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere. Mo(CO) 6 was purchased from Strem and used without further purification. [PPN]3[P309]. H20 23 22 was prepared as previously reported. (CH 3CN) 3Mo(CO) 3 was prepared as previously reported.

VOC13 was purchased from Strem and used without further purification. Diethyl ether, acetonitrile and methylene chloride were dried and deoxygenated using a system built by SG Water USA, LLC 3 1 (www.glasscontoursolventsystems .com). CDC13 was purchased from Cambridge Isotope Labs, purified by distillation off of CaH2 and stored over 4 A molecular sieves. Celite, alumina and 4 A molecular sieves were dried under reduced pressure overnight at a temperature above 200 0C. 'H, 13C and 31P NMR spectra were recorded on Varian Mercury-300, Varian INOVA- 500 or Bruker AVANCE-400 spectrometers. 'H and 13C chemical shifts are reported with respect 31 to internal solvent (CDCl3, 8 = 7.24 and 77.23 ppm, respectively). P NMR chemical shifts are 51 reported with respect to an external reference (85% H3PO4, 6= 0.0 ppm). V NMR chemical shifts are reported with respect to an external reference (neat VOCl 3, 6 = 0.0 ppm). Infrared spectra were recorded on a Bruker TENSOR37 FT-IR Spectrophotometer. X-ray data collections were carried

106 out on a Siemens Platform three-circle goniometer with a CCD detector using Mo-Ka radiation, X = 0.71073 A. Combustion analyses were performed by Midwest Microlabs LLC, Indianapolis, IN.

4.5.2 Synthesis of [PPN] 3[(P30 9)Mo(CO) 3] ([PPN] 3[18]) from (CH3CN) 3Mo(CO) 3-

In the glovebox, (CH 3CN) 3Mo(CO) 3 (1.176 g, 3.880 mmol) was dissolved in 10 mL of acetonitrile and stirred. [PPN] 3[P30 9]- H20 (7.257 g, 3.879 mmol) was dissolved in 20 mL of acetonitrile and added to the stirring, greenish-brown (CH 3CN)3Mo(CO) 3 solution over a period of 5 min. Upon addition of the trimetaphosphate salt, the reaction mixture attained a golden-brown color. The mixture was stirred for 1 h, after which time an aliquot was taken for analysis by 31P NMR 31 spectroscopy. The P NMR (202.5 MHz, CH 3CN) spectrum showed complete consumption of starting material with concomitant formation of the desired product: 8 22.12 (s, 6P, [PPN]+),

-10.70 (s, 3P, P309 ). The reaction mixture was filtered through a bed of Celite. The filtrate was reduced in volume to ca. 10 mL under reduced pressure. The golden-yellow reaction mixture was then added dropwise to 100 mL of diethyl ether while stirring. The desired product initially precipitated out of solution, forming a flocculent, yellow solid before oiling out to an dark-orange tar. The volatiles were then removed under reduced pressure forming a foamy, bright-orange solid. The solids were slurried in ca. 50 mL of diethyl ether and the larger chunks crushed with a spatula. The material was isolated on a medium-porosity, fritted-glass filter and washed with 2* 20 mL of diethyl ether. The bright-orange product was dried under reduced pressure for 24 hours yielding the desired product as an orange powder (7.170 g, 3.527 mmol, 90.9%). 'H NMR (500 MHz, CDCl3) 6 7.64 (br. 13 s, 6H, phenyl), 7.45-7.36 (m, 24H, phenyl) ppm. C NMR (125.8 MHz, CH 3CN/CDCl 3) 6 230.57 (s, 3C, CO), 133.04 (in, 6C, para), 131.44 (in, 6C, ortho), 128.76 (in, 6C, meta), 126.33 3 3 1 (dd, 'Jpc = 107.7 Hz, JPc = 1.7 Hz, 6C, ipso) ppm. P NMR (202.5 MHz, CH2 Cl2/CDC13 ) 21.81 (s, 6P, [PPN]+), 11.08 (s, 3P, P30 3) ppm. FT-IR (KBr, thin film) vco = 1883 (s, A,), 1723 (s, E) 1 cm- . Anal Calcd. for Cjj1H90N3P9012Mo: C, 65.59; H, 4.46; N, 2.07; P, 13.71. Found: C, 63.92; H, 4.68; N, 1.96; P, 13.20.

4.5.3 Synthesis of [PPN] 3[(P309 )Mo(CO) 3] ([PPN] 3[18]) from Mo(CO) 6.

Mo(CO)6 (685 mg, 2.59 mmol) and [PPN] 3 [P309] -H20 (4.857 g, 2.596 mmol) were suspended in 40 mL MeCN and stirred at reflux for 4 h. Upon reaching reflux temperature, the reaction mixture went from a colorless suspension to a homogeneous, light yellow solution. The volatiles were removed under reduced pressure, the golden-yellow solids dissolved in 10 mL CH 2Cl2 and the solution filtered through a bed of Celite. The Celite pad was washed with 5 mL CH2Cl2, the filtrates were combined and added dropwise to 75 mL Et20. Addition to ether caused the formation of a pale yellow precipitate. The volatiles were again removed under reduced pressure. The bright- orange solids were slurried in 30 mL diethyl ether, isolated on a medium-porosity, fritted-glass filter and washed with 3 * 20 mL diethyl ether. The solids were dried under reduced pressure for 12

107 hours yielding the desired product as a free-flowing, orange powder (4.056 g, 1.995 mmol, 77.0%). Spectroscopic data for the sample prepared in this manner were identical to those obtained according to Section 4.5.2.

4.5.4 Attempted synthesis of (P30)V-O (19)

solution VOCl3 (100 mg, 0.577 mmol) was dissolved in 3 mL of MeCN and stirred. A colorless of [PPN] 3[P30 9] H20 (1.079 g, 0.577 mmol) in 3 mL MeCN was added to the dark-red vanadium solution leading the color of the reaction mixture to lighten slightly. The mixture was allowed to stir for 1 h, after which time a fine white precipitate had formed. An aliquot of the reaction mixture was taken, filtered through a bed of Celite and loaded into an NMR tube. The 31P NMR spectrum showed only two phosphorus environments: S[PPN] + = 22.1 ppm and a new trimetaphosphate product 51 at 8p3o9 = -26.0 ppm (presumably compound 19). The V NMR spectrum showed the formation of one major product at 6 = -370 ppm. All attempts to isolate 19 in pure form have not as of yet been successful.

4.6 DFT Calculations

All calculations were carried out using ADF 2004.01 from Scientific Computing and Modeling (http: //www. scm. com).27,28 In all cases the OLYP functional was employed,29 30 while the GGA part was handled using the functionals of Becke and Perdew (BP86).3 2,33 In addition, all calculations were carried out using the Zero Order Regular Approximation (ZORA) for relativistic effects. 34,35 In all cases the basis sets were quadruple-zeta with four polarization functions (QZ4P) as supplied by ADF. No frozen core approximations were used. Chemical shielding tensors were calculated for 31P and 51V nuclei in the optimized structures by the GIAO method using the ADF package. 36-39 Siso was calculated in the following fashion: 6 iso = astd,calc - acalc + 8std,expt where the standards for 31 51 P and V were P4 (6 = -520 ppm) and VOCl3 (6 = 0.0 ppm), respectively. Calculations were carried out on a thirty-two-processor Quantum Cube workstation from Parallel Quantum Solutions (http: \www .pqs - chem. com). All results reported are with reference to fully optimized geometries with no imaginary frequencies.40'41

4.7 Crystallographic Structure Determinations

4.7.1 X-ray Crystal Structure of [PPN]3[(P30,)Mo(CO) 3] ([PPN3][18])

X-ray diffraction data were collected on a Siemens Platform three-circle diffractometer equipped with a Bruker-AXS APEX-CCD detector and an Oxford Cryosystems low-temperature device. Graphite monochromated Mo-Ka radiation (X = 0.71073 A) was used for data collection.

108 Inside the glovebox, crystals of [Ph 3P=N-PPh3][PPN3][18], obtained from a vapor diffusion of tetrahydrofuran into a saturated solution of [PPN 3][18] in acetonitrile at room temperature, were coated with mineral oil on a microscope slide. A yellow plate of approximate dimensions 0.32 * 0.31 * 0.18 mm3 was selected and mounted on a glass fiber. A total of 424991 reflections (-28 < h < 27, 0 < k < 32, 0 < 1 < 24) were collected at 100(2) K using <- and on-scans in the 0 range of 1.39 to 27.58', of which 25269 were unique (Rint = 0.0938). The structure was solved with direct methods using SHELXS 42 and refined against F 2 on all data by full-matrix least squares with SHELXL-97, 43 following established refinement strategies. 44 The systematic absences in the diffraction data are consistent with the assigned space group of P2 1/c. The highest residual density maximum (4.53 e -A-3) was found to be located near the molybdenum atom and, upon careful examination of the difference Fourier synthesis, alternate positions for most atoms of the anion could be distinguished (see Figure 4.6). Unfortunately, this disorder was not stable and the final model contains split positions only for the molybdenum atom, while the other atoms of the anion are treated as not disordered and fully occupied. Anisotropic displacement parameters of the two molybdenum atoms were constrained to be identical: the disorder ratio was refined freely and converged at 0.9510(8).

One disordered molecule of tetrahydrofuran is present in the asymmetric unit. The disorder was refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. The ratio was refined freely (converging to 0.835(5)), while constraining the total occupancy of both components to unity. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the U value of atoms they are linked to. The crystal was non-merohedrally twinned. Two independent orientation matrices for the unit cell were found using the program CELLNOW,45 and data reduction taking into account the twinning was performed with SAINT. 46 The program TWINABS 47 was used to perform absorption correction and to set up the HKLF5 format file for structure refinement. The twin ratio was refined freely and converged at a value of 0.4549(8). The residual peak and hole electron density were 0.935 and -0.746 e-A-3, respectively. The least squares refinement converged normally with residuals of 2 R1 = 0.0594 for I > 2a(I), wR2 = 0.1521 for all data, and GOF = 1.034 (based on F ).

Crystal data: formula C11 H90N3P9O12Mo (with one equivalent of C4 H80, thus giving the formula C115H98N3P9O13Mo), space group P21/c, a = 29.8909(11) A, b = 25.0489(13), c = 18.5013(9) A, p = 102.4190(10)0, V = 9907.7(9) A3. Z = 4, p = 0.345 mm- 1, Deaic = 1.411 g.cm- 3, F(000) = 4360.

109 ......

011 C1B 07 )C3B 07

06 C2B

069

Mol 04B

P38/ C3

012B PB0

09B 01013 P2B 08B3 07B 02

0118 Figure 4.6. Solid-state structure of [(P30 9)Mo(CO) 3 3- (183-) showing the ca. 95:5 positional disorder. The major component is shown with thermal ellipsoids at the 50% probability level. The minor component could not be modeled anisotropically without most of the atoms refining as non-positive definite. The minor component is isotropic and represented using empty spheres. The minor component could not be satisfactorily modeled without using an unreasonable number of restraints and constraints and as a result was modeled as only a molybdenum atom for the complete structure.

110 Table 4.4. Crystallographic Data for {MesP[C(CPhq)NPh],}Ta(I)Meq (24)

[PPN] 3 [18] Reciprocal Net code / CCDC code 11062 / 823192 Empirical formula, FW (g/mol) CjjiH98N3 P9O 13Mo, 2104.63 Color / Morphology Orange / Block Crystal size (mm3 ) 0.32* 0.31 * 0.18 Temperature (K) 100(2) Crystal system, Space group Monoclinic, P21 /c

Unit cell dimensions (A, 0) a = 21.8909(11), X = 90 b = 25.0489(13), p = 102.4190(10) c = 18.5013(9), y= 90 Volume (A3) 9907.7(9) Z 4 Density (calc., Mg/m 3 ) 1.411 Absorption coefficient (mm-') 0.345 F(O0O) 4360 Theta range for data collection (0) 1.39 to 27.58 Index ranges -28 < h < 27, 0 < k < 32, 0 < 1 < 24 Reflections collected 424991 Independent reflections, Rint 25269 (0.0938) Completeness to 0 max (%) 99.6 Max. and min. transmission 0.9404 and 0.8975 Data / restraints / parameters 22854 / 172 / 1321 Goodness-of-fita 1.034

Final R indicesb [I > 2(I)] R = 0.0594, wR2 = 0.1300 R indicesb (all data) R = 0.0949, wR 2 = 0.1521 Largest diff. peak and hole (e -A- 3 ) 0.935 and -0.746

2 2 2 2 a GooF [ E[w(F, -F ) ] bR _ E|F|-IWE2 = r[w(F -F) 1 -L (n-p) i |F '[wwR 2 W - FT +(aP)2+bP' p _ 2F2+max(F,,0) 4.8 References

[1] Besecker, C. J.; Klemperer, W. G. J. Organomet. Chem. 1981, 205, C3 1-C32. [2] Besecker, C. J.; Day, V. W.; Klemperer, W. G. Organometallics1985, 4, 564-570. [3] Day, V. W.; Klemperer, W. G.; Main, D. J. Inorg. Chem. 1990, 29, 2345-2355. [4] Day, V. W.; Klemperer, W. G.; Lockledge, S. P.; Main, D. J. J. Am. Chem. Soc. 1990, 112, 2031-2033. [5] Day, V. W.; Eberspacher, T. A.; Klemperer, W. G.; Planalp, R. P.; Schiller, P. W.; Yagasaki, A.; Zhong, B. Inorg. Chem. 1993, 32, 1629-1637. [6] Klemperer, W. G.; Zhong, B. Inorg. Chem. 1993, 32, 5821-5826. [7] Day, V. W.; Eberspacher, T. A.; Klemperer, W. G.; Zhong, B. J. Am. Chem. Soc. 1994, 116, 3119-3120. [8] Han, K.-N.; Whang, D.; Lee, H.-J.; Do, Y.; Kim, K. Inorg. Chem. 1993, 32, 2597-2599. [9] Ryu, S.; Whang, D.; Kim, J.; Yeo, W.; Kim, K. J. Chem. Soc., Dalton Trans. 1993, 205-209. [10] Attanasio, D.; Bachechi, F.; Suber, L. J. Chem. Soc., Dalton Trans. 1993, 2373-2378. [11] Kamimura, S.; Kuwata, S.; Iwasaki, M.; Ishii, Y. Dalton Trans. 2003, 2666-2673. [12] Suh, I.-H.; Lee, J.-H.; Lim, S.-S.; Park, J.-R.; Choo, G. H.; Kwak, S.-H.; Do, Y.; Lee, H. J. Chung. Kwa. Yong. (Chungnam J. Sci.) 1994, 21, 124. [13] Kamimura, S.; Matsunaga, T.; Kuwata, S.; Iwasaki, M.; Ishii, Y Inorg. Chem. 2004, 43, 6127-6129. [14] Kamimura, S.; lida, T.; Kanao, K.; Nogawa, C.; Tanabe, Y.; Oh-ishi, K.; Fukuzawa, S.; Ishii, Y. Journal of the Institute of Science and Engineering, Chuo University 2005, 11, 1. [15] Ikeda, Y.; Yamaguchi, T.; Kanao, K.; Kimura, K.; Kamimura, S.; Mutoh, Y.; Tanabe, Y.; Ishii, Y. J. Am. Chem. Soc. 2008, 130, 16856-16857. [16] Bafiares, M. A. Catal. Today 1999, 51, 319-348. [17] Royer, S.; Duprez, D. ChemCatChem 2011, 3, 24-65. [18] Ratnasamy, C.; Wagner, J. P. Cat. Rev.-Sci. Eng. 2009, 51, 325-440. [19] Freund, H.-J.; Pacchioni, G. Chem. Soc. Rev. 2008, 37, 2224-2242. [20] Fu, Q.; Wagner, T. Surf Sci. Rep. 2007, 62, 431-498. [21] Montag, M.; Clough, C. R.; MUller, P.; Cummins, C. C. Chem. Commun. 2011, 47, 662-664. [22] Tate, D. P.; Knipple, W. R.; Augl, J. M. Inorg. Chem. 1962, 1, 433-434. [23] Klemperer, W. G.; Main, D. J. Inorg. Chem. 1990, 29, 2355-2360. [24] Bell, R. N. Inorg. Synth. 1950, 3, 103-106. [25] Hocking, R. K.; Hambley, T. W. Organometallics2007, 26, 2815-2823. [26] Earley, C. W. J. Phys. Chem. A 2010, 114, 987-993. [27] te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931-967. [28] Guerra, C. F; Sniders, J. G.; te Velde, G.; Baerends, E. J. Theor Chem. Acc. 1998, 99, 391-403. [29] Handy, N. C.; Cohen, A. J. Mol. Phys. 2001, 99, 403-412. [30] Baker, J.; Pulay, P. J Comp. Chem. 2003,24, 1184-1191. [31] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. [32] Becke, A. D. J Phys. Rev. A 1988, 38, 3098-3 100. [33] Perdew, J. P. Phys. Rev. B 1986, 34, 7406. [34] van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99, 4597-4610. [35] van Lenthe, E.; Ehlers, A.; Baerends, E. J. J. Chem. Phys. 1999, 110, 8943-8953. [36] Schreckenbach, G.; Ziegler, T. J. Phys. Chem. 1995, 99, 606-611. [37] Schreckenbach, G.; Ziegler, T. Int. J. Quantum Chem. 1997, 61, 899-918.

112 [38] Wolff, S. K.; Ziegler, T. J. Chem. Phys. 1998, 109, 895-905. [39] Wolff, S. K.; Ziegler, T.; van Lenthe, E.; Baerends, E. J. J. Chem. Phys. 1999, 110, 7689-7698. [40] Fan, L. Y.; Ziegler, T. J. Chem. Phys. 1992, 96, 9005-9012. [41] Fan, L. Y.; Ziegler, T. J. Chem. Phys. 1992, 96, 9005-9012. [42] Sheldrick, G. M. Acta Cryst. A 1990, 46, 467-473. [43] Sheldrick, G. M. Acta Cryst. A 2008, 64, 112-122. [44] MUller, P. Crystallogr Rev. 2009, 15, 57-83. [45] Sheldrick, G. M. 2008, CELLNOW, University of Gbttingen, Germany. [46] Bruker 2010, SAINT, Bruker-AXS Inc., Madison, WI, USA. [47] Sheldrick, G. M. 2008, TWINABS, University of G6ttingen, Germany.

113 114 CHAPTER 5

Chemistry of a new bis-Enamide Ligand

Contents 5.1 Introduction ...... 116 5.2 Synthesis of bis-Enamide Ligands ...... 118 5.3 Metallation of bis-Enamide Complexes with Tantalum ...... 119 5.4 Discussion ...... 123 5.5 Conclusions ...... 124 5.6 Experimental Section ...... 125 5.6.1 General Considerations...... 125

5.6.2 Synthesis of Ph2 CHC(O)NHPh...... 125

5.6.3 Synthesis of Ph2 CCNPh...... 126

5.6.4 Synthesis of [Li(thf)] 2 {PhP[C(CPh2 )NPh] 2} ([Li(thf)] 2 [20]) ...... 126

5.6.5 Synthesis of Li2PMes...... 127

5.6.6 Synthesis of [Li(thf)] 2{MesP[C(CPh 2)NPh]2} ([Li(thf)]2 [21])...... 127

5.6.7 Synthesis of {PhP[C(CPh2)NPh] 2}TaMe3 (22)...... 127

5.6.8 Synthesis of {MesP[C(CPh2)NPh] 2}TaMe 3 (23)...... 128

5.6.9 Synthesis of {MesP[C(CPh2)NPh] 2}Ta(I)Me2 (24)...... 128 5.7 Crystallographic Structure Determinations ...... 129 5.7.1 General Considerations...... 129 5.7.2 X-ray crystal structure of [Li(OEt 2)]2 {PhP[C(CPh 2)NPh]2 } ([Li(OEt2)] 2[20])...... 129

5.7.3 X-ray crystal structure of {PhP[C(CPh2)NPh] 2}TaMe 3 (22) ...... 130

5.7.4 X-ray crystal structure of {MesP[C(CPh2)NPh] 2}TaMe 3 (23) ...... 130

5.7.5 X-ray crystal structure of {MesP[C(CPh2)NPh]2}Ta(I)Me 2 (24) ..... 131 5.8 References ...... 134

115 5.1 Introduction

The use of primary dilithium phosphanides to generate multidentate ligands is not unprecedented.

Bi- and tridentate phosphines have been synthesized by SN2 type attack of Li2PPh on chloro- substituted tertiary phosphines 1 (5. lA) or mesyl-substituted cyclodextrins 2 (5.1B). Water-soluble 3 phosphonate ligands have also been synthesized using Li2PPh. In the phosphonate case, two equivalents of 2-FC6H4P(O)(NR2) 2 (R = Me, Et) are treated with one equivalent of Li2PPh in a salt metathesis reaction to give PhP[2-C 6H4P(O)(NR2)21 2 and two equivalents of LiF (5. lC). The desired phosphonate is then synthesized by hydrolyzing PhP[3-C 6H4P(O)(NR2)212 with aqueous 3 HCl to give PhP[3-C 6H4P(O)(OH) 212- Perhaps the most interesting example (and most similar to the work presented here) is the synthesis of C-P-C bridged bis-cyclopentadienides by the treatment of two equivalents of 6,6-dimethylfulvalene with one equivalent of either Li2PPh or Li2PCy 4 (Cy = cyclohexyl, 5. iD). Unlike other salt-metathesis reactions of Li 2PR referenced above, the phosphanide acts as an nucleophile and attacks two equivalents of the unsaturated fulvalene at the 6-position to form a dilithiated ligand without formation of a byproduct.

Previous research in our group has involved treatment of the niobium phosphide complex, [Na(OEt2)][PNb(N[Np]Ar) 3] (Np = neopentyl) with the three equivalents of the unsaturated electrophile, diphenylketene. 5 In that case, the phosphide acted as a nucleophile whereupon it attacked each ketene at the unsaturated carbon. This led us to attempt the reaction of dilithium phenylphosphanide with two equivalents of diphenylketene where the expected bidentate ligand was formed (5.2).5 The natural progression of this research was to explore the reactivity of phosphanides with ketenimines leading to the formation of the bis-enamide salts presented herein.

116 i-Pr A i-Pr Li PPh 2 CI P 2 Ph-P -2 LiCI i-Pr

hi-Pr Goikham, et a.

MeO 2 Li2PPh -4 LiOSO Me 2 MeO

Engeldinger, etaL.

F 0

Li2PPh / P(O)(NR ) 2 "#NR2 2 2 -2 LiF NR2 P(O)(NR 2)2

Kant and Bischoff R = Me, Et

2 Li*

2 Li2PR R -P

H6cher, et aL. R = Cy, Ph Scheme 5.1. Existing ligand syntheses utilizing arylphosphanides. 1-4

117 Ph

OLi Li PPh 2 C 2 Ph -P

Oi Ph Ph Ph

Ph Scheme 5.2. Reaction of dilithium phenylphopshanide with two equivalents of diphenylketene. 5

5.2 Synthesis of bis-Enamide Ligands

Treatment of dilithium arylphosphanide (Li2PAr) with two equivalents of triphenylketenimine (Ph2C=C=NPh) results in clean formation of a new dilithiated, bis-enamide ligand,

[Li(thf)]2{PhP[C(CPh2)NPh]2} ([Li(thf)] 2120]) (5.4). Li2PPh is synthesized in one step from PhPH2 in hexane by the addition of excess "BuLi, resulting in a golden-yellow, pyrophoric solid that precipitates from solution. 6 Diphenylacetyl chloride is treated with aniline to form the carboxamide

Ph2CHC(O)NHPh in near-quantitative yield (vide infra). The resulting carboxamide is then refluxed with triphenylphosphine, bromine and excess triethylamine in methylene chloride to form the 7 desired ketenimine, Ph2C=C=NPh (5.3).

ArPH 2 2.4 "BuLi Li2PAr Ar = Ph, Mes n-hexane 97% -95 to 25 *C - butane

Ph 0 PPh 3, Br2 excess NEt Ph2CHC(O)C1 2 PhNH 2 Ph Ph 3 Et 0, 0 'C CHCI , reflux 2 H 2 2 - [PhNH 3]CI - 2 [HNEt 3]Br - OPPh 3 Ph Ph 99.8% 70-85%

Scheme 5.3. Syntheses of bis-enamide starting materials.6-10

When a thawing solution of two equivalents of Ph2C=C=NPh in THF at -108 'C is added to a thawing solution of Li2PPh in THF and allowed to warm to room temperature, the desired ligand, [Li(thf)]2 {PhP[C(CPh2)NPh] 2} ([Li(thf)] 2[20]), is formed by a double addition of the phosphanide to two ketenimine molecules. In tetrahydrofuran solution, [Li(thf)]2[20] shows

118 a clean 31P NMR spectrum with one resonance at 8 = 8.9 ppm. The desired product can be isolated by precipitation from n-hexane and isolated as a "ready-to-use" dilithiated ligand

(5.4). [Li(thf)]2 {MesP[C(CPh 2 )NPh]2}([Li(thf)]2 [21]) is synthesized analogously to [Li(thf)]2 [20] by substituting dilithium mesitylphosphanide, Li 2PMes, for Li2 PPh (5.4).8,9 [Li(thf)]2 [21] is synthesized in 72% yield and isolated as an analytically-pure, golden-yellow powder by precipitation from hexane (3 1P NMR in THF, 8 = 11.4 ppm; 5.4)

thf Ph i. Ph Ph N--L..- N Ph thf Ph

Li2PAr + 2 c Ph Ar Ph Ph Ph -108 to 25C Ar Ph,70%

Ar = Mes, 72%

Scheme 5.4. Synthesis of [Li(thf)] 2 {PhP[C(CPh2 )NPh]2} ([Li(thf)] 2 [20]) and [Li(thf)]2{MesP[C(CPh2)NPh] 2 } ([Li(thf)] 2[21])

[Li(thf)] 2 [20] was dissolved in Et 2 0 and cooled to -35 'C whereupon crystals of the diethyl ether solvate, [Li(OEt2)]2[20], were recovered and used for an X-ray crystallographic study. The solid-state structure of [Li(OEt2)] 2 [20] exhibits a bidentate binding moiety whereby the two lithium cations are primarily bound to NI and N2 (5.1). In the thermal-ellipsoid plot shown in 5.1, only a weak interaction between the phosphorus atom and one of the lithium cations exists (Pl-Li2 = 3.002(4) A). Each lithium cation is bound to one diethyl ether of solvation. The coordinated ether molecules have been omitted from the thermal-ellipsoid plot of shown in 5.1 to clearly illustrate the bidentate binding pocket formed by [20] 2-

5.3 Metallation of bis-Enamide Complexes with Tantalum

When treated with TaMe 3Cl2 ", [Li(thf)]2 [20] cleanly metallates to form the pseudo-trigonal bipyramidal complex {PhP[C(CPh2)NPh] 2}TaMe3 (22) with concomitant loss of two equivalents of LiCl. Compound 22 is isolated in 91% yield (5.5). The 31P NMR spectrum shows one product at 8 = 0.85 ppm in benzene. When compound 22 was dissolved in a mixture of

Et20 and hexamethyldisiloxane and stored at -35 'C, crystals of sufficient quality for an X-ray crystallographic study were obtained. The solid-state structure of 22 reveals a pseudo-trigonal bipyramidal geometry at tantalum with two methyl groups in the equatorial position and one in the axial position (5.2). Although the equatorial ligands from a near-perfect plane (Eequatorial =

119 ~ 4

Figure 5.1. ORTEP drawing of [Li(OEt2)] 2{PhP[C(CPh2)NPh] 2} ([Li(OEt2)] 2[20]) with thermal ellipsoids at the 50% probability level. Hydrogen atoms and coordinated ether molecules have been omitted for clarity

359.83(9)'), the compound is perturbed from an ideal trigonal bipyramid with the Cl-Tal-N2 angle being 165.37(9)0. As a result of the geometry of the ligand, one of the methyl groups (Cl) is forced into a position roughly trans to an enamide nitrogen (N2). From 3D modeling of 22 we do not believe steric influence to be causing the perturbation of the Cl-Tal-N2 angle away from 180'. Instead, we attribute this deviation from linearity as a result of the strong trans-directing properties of the methyl group. The Ta-C bond lengths also support this supposition with equatorial ligands having an average bond distance of 2.169(3) A, whereas the axial carbon is 2.255(3) A from the tantalum center.

Me Me thf PhMemTa,- Ph N NN hh .j.. Ph TaMe 3Cl2 N-.-Li..-N* Ph Ph C 6H 6p Ph thf Ph 6 to 25-C - 2LiCI Ph Ar Ph Ph Ar Ph Ar = Ph, 91% Ar=Mes,96%

Scheme 5.5. Synthesis of {PhP[C(CPh2)NPh]2}TaMe3 (22) and {MesP[C(CPh 2)NPh]2}TaMe 3 (23).

[Li(thf)] 2 [21] was treated with TaMe 3Cl 2 in benzene and {MesP[C(CPh2 )NPh]2}TaMe 3 (23) was formed in 96% isolated yield (5.5). Compound 23 shows a 31P NMR resonance at 3 = -1.7 ppm in THE. Complex 23 is similar to 22 in its overall geometry (5.3). As in 22, the equatorial ligands in 23 are nearly planar (Eequatorial = 359.5(6)') and the axial methyl group is perturbed from

120 ......

Figure 5.2. ORTEP drawing of {PhP[C(CPh2 )NPh]2 }TaMe 3 (22) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Tal-Nl 1.961(2), Tal-N2 2.060(2), Tal-Cl 2.255(3), Tal-C2 2.166(3), Tal-C3 2.172(2), Cl- Tal-N2 16 5 .3 7 (9 ),Eequatorial 359.83(9).

ideal trigonal bipyramidal geometry (Cl-Tal-N2 = 163.5(2)0). For reasons not understood, the Tal-Cl bond length (2.1820(4) A) in 23 is much closer in length to the average of the Ta-Cequatorial bond lengths (2.1708(2) A) than in the parent septaphenyl complex 22.

When 23 is treated with 0.5 equivalents of elemental iodine,

{MesP[C(CPh 2)NPh] 2}Ta(I)Me 2 (24) is formed in 70% yield with concomitant loss of methyl iodide (5.6). Compound 24 shows the furthest 3 1P NMR upfield shift of any of the complexes described presently (8 = - 17 ppm). Crystals of compound 24 were grown from a saturated diethyl ether solution stored at -35 'C and used for an X-ray crystallographic study. The X-ray study confirms the replacement of a methyl group with an iodide ligand residing in the axial position.

The solid-state structure of 24 forms a pseudo-trigonalbipyramid like the (NPN)TaMe3 complexes.

Again, a nearly planar equatorial ligand set is seen (Eequatorial = 359.3(1) 0) and the structure is closer to an ideal trigonal bipyramid with Il-Tal-N2 = 172.29(6)0 than in the trimethyltantalum variants. Figure 5.3. ORTEP drawing of {MesP[C(CPh2)NPh] 2}TaMe3 (23) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Tal-NI 1.9598(3), Tal-N2 2.0788(3), Tal-C1 2.1820(4), Tal-C2 2.1746(2), Tal-C3 2.1669(2), C1-Tal-N2 163.5(2) Eequatorial 359.5(6).

Me Me 11 Me PhMe TaMe Ph PhMefTa'1 Ph N N/ N N Ph Ph Et20 -116 to 25 C p -Mel Ph Mes Ph

Scheme 5.6. Synthesis of {MesP[C(CPh2)NPh]2 }Ta(I)Me2 (24).

122 Figure 5.4. ORTEP drawing of {MesP[C(CPh2)NPh] 2}Ta(I)Me 2 (24) with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles (deg): Tal-NI 1.968(2), Tal-N2 2.032(2), Tal-Ci 2.160(3), Tal-C2 2.185(3), Tal-Il 2.8026(4), Il- Tal-N2 172.29(6), E-equatorial 359.3(1).

5.4 Discussion

The starting materials for the synthesis of [Li(thf)] 2 [20] are made in a total of three steps from commercially available chemicals. Superficially, the ligand frameworks exhibited by [20]2- and [21]2- are similar in structure to other "NPN" diamido-phosphine ligands.12-15 However, known NPN ligands in the literature usually serve as tridentate, dianionic ligands. 12-15 The dianionic- ligand framework exhibited by [20]2 favors a bidentate binding mode due to the formation of a six-membered ring when a metal is bound within the binding pocket. The tridentate form of the NPN ligands is obviated by the unfavorable conformation of two, four-membered metallacyles that the K3-coordination would contain.

The closest example to the bis-enamide ligands described presently that we could find in the literature is the bis[N-(trimethylsilyl)iminobenzoyl]phosphanide complex synthesized by Becker, et al. (5.7). 16 In that case, the lithium bis-trimethylsilylphosphanide is treated with two equivalents of benzonitrile to form the bidentate, monoanionic complex described in 5.7. Becker's ligand shows a similar binding motif to our "NPNTa" complexes when metallated with zinc. The bis[N- (trimethylsilyl)iminobenzoyl]phosphanide forms a six-membered ring with the zinc center where the phosphorus atom is not coordinated to the metal.

The placement of the phosphine moiety in the backbones of bis-enamide ligands [20]2- and [21]2- gives a convenient spectroscopic handle. Reaction mixtures can easily be monitored by 3 1 P NMR spectroscopy in situ without the need of deuterated solvents. This benefit is not only

123 Ph P Ph dme 0.5

(thf)2LiP(SiMe3)2 Me Si 3 Li SiMe 3 0.5 ZnCl Me Si 2 3 N,,. 'N -SiMe 3 + DME N 0 30 N - DME Me 3Si 1 Zn N-SiMe-N. 3 2 PhCN -50 to 20 C -LiCI N N - 2 THF Ph P Ph Ph P Ph

Scheme 5.7. Synthesis of bis[N-(trimethylsilyl)iminobenzoyl]phosphanide and the corresponding metallation with zinc. 16 economical, but environmentally friendly as solvent waste is minimized. Furthermore, it saves time by allowing the chemist to check reaction progress in near real time without having to remove the solvent of an aliquot and then prepare an NMR sample in a deuterated solvent.

The parent ligand can be modified by using different starting phosphanides or ketenimines.

[Li(thf)]2[21] was developed as a proof of principle for ligand modification. Despite needing more steps to synthesize the requisite MesPH2, 17 [Li(thf)] 2[21] exhibits some spectroscopic advantages over the septaphenyl variant [Li(thf)]2[20]. The placement of the methyl groups of the mesityl ring give easily identifiable signals in the alkyl region of the 'H NMR spectrum. The aryl regions of the 'H NMR spectra of these materials are rather crowded and difficult to analyze. Similarly, the 1H NMR spectrum of 23 is much more useful than that of 22 due to the spectroscopic handle the mesityl moiety provides.

5.5 Conclusions

We have successfully synthesized a new class of ligands from the elegant double addition of arylphosphanides to ketenimines. These bis-enamide ligands are readily customizable by modification of either the arylphosphanide or ketenimine. The ligand scaffolds of [20]2- and [21]2- contain phosphorus atoms in the backbone which act as convenient spectroscopic handles for analyses of reaction mixtures by 31P NMR spectroscopy. The ligands have also been shown to be capable of cleanly forming metal complexes with "TaMe 3" fragments. Furthermore, in the case of 24, replacing the methyl ligands with iodides allows for further exploration of "(NPN)Ta" complexes. Future directions of this project will focus on further modification of the tantalum platform by replacing the iodide ligand via salt-metathesis reaction to investigate metal-ligand multiply bonded species.

124 5.6 Experimental Section

5.6.1 General Considerations

Unless stated otherwise, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere.

Ph2CHC(O)C1 was purchased from Aldrich and purified by recrystallization from boiling hexane before use. Triphenylphosphine, triethylamine and 1.6 M "BuLi in hexanes were purchased from Aldrich and used without further purification. Bromine was purchased from Alfa Aesar and used without further purification. Aniline was purchased from Aldrich and distilled off of CaH 2 under reduced pressure before use. Phenylphosphine was purchased from Strem and used without further 6 purification. Li2PPh was prepared as previously published. MesPH 2 was prepared as previously published. 17 TaMe3 Cl2 was prepared as previously published. 11 Hexane, diethyl ether, n-pentane, toluene and THF were dried and deoxygenated using a system built by SG Water USA, LLC 18 (www. glass contours olvent systems. com). C6D6 andCDC 3 were purchased from Cambridge

Isotope Labs, purified by distillation off of CaH 2 and stored over 4 A molecular sieves. Celite, alumina and 4 A molecular sieves were dried under reduced pressure overnight at a temperature above 200 0C. 'H, 13C and 31P NMR spectra were recorded on Varian Mercury-300, Varian INOVA- 500 or Bruker AVANCE-400 spectrometers. 'H and 13C chemical shifts are reported with respect 31 to internal solvent (C6D6, 6 = 7.16 and 128.39 ppm, respectively). P NMR chemical shifts are reported with respect to an external reference (85% H3PO4, 3 = 0.0 ppm). X-ray data collections were carried out on a Siemens Platform three-circle goniometer with a CCD detector using Mo- Ka radiation, k = 0.71073 A. Combustion analyses were performed by Midwest Microlabs LLC, Indianapolis, IN.

5.6.2 Synthesis of Ph2CHC(O)NHPh.

Ph 2CHC(O)NHPh was prepared by modification of a literature preparation for PhCH2 C(O)NHPh.10 In a fumehood, in air, aniline (18.22 g, 195.8 mmol) was loaded into a 500 mL RB and dissolved in 50 mL Et20. The flask was equipped with a stirbar and cooled to 0 'C in an ice bath. A solution of diphenylacetyl chloride (22.67 g, 98.54 mmol) in 100 mL Et2O was added dropwise to the stirring aniline solution over the course of 45 minutes. A large amount of a flocculant white precipitate formed upon addition of the acid chloride to the aniline. After the addition was finished, the reaction mixture was filtered through a medium-porosity fritted glass funnel and the filtrate set aside. The solids were washed with 1.5 L CH2Cl2 and the mixture was again filtered. The filtrates were combined and washed (in succession) with 500 mL of the following: water, 10% w/w Na2CO3, water, 1 M HCl, and water. The CH 2Cl2 fraction was dried with Na 2SO 4 and filtered. The product was obtained as a white powder by removal of the solvent under reduced pressure and used without further purification (26.47 g, 98.38 mmol, 99.8%).

125 5.6.3 Synthesis of Ph2CCNPh.

7 Ph2 CCNPh was prepared as previously published and further purified by recrystallization from dry, degassed Et 2O in the drybox. In a fumehood, in air, a 1 L RB flask was equipped with triphenylphosphine (26.05 g, 110.7 mmol), bromine (15.87 g, 99.31 mmol), triethylamine (75 mL), and then Ph2CHC(O)NHPh (26.73 g, 99.32 mmol). The mixture was refluxed with stirring for lh, after which time the reaction mixture changed color from yellow to dark-brown in color. The volatiles were removed from the reaction mixture and the bright-yellow product extracted with petroleum ether. The petroleum ether was removed under reduced pressure for 4 h (18.73 g, 70%). This crude material is suitable for most applications, but can be further purified by recrystallization from dry, degassed Et 2O at -35 'C in a glovebox (10.92 g, 40.9%).

5.6.4 Synthesis of [Li(thf)]2{ PhP[C(CPh 2)NPh] 2} ([Li(thf)]2[20]).

A thawing solution of Ph2CCNPh (2.00 g, 7.43 mmol) in 20 mL THF was added to a thawing solution of Li2PPh (444 mg, 3.64 mmol) in 16 mL THF. The reaction mixture was stirred and allowed to warm to room temperature. The reaction mixture darkened from bright-yellow to orange upon addition of the ketenimine. Over the course of 45 minutes, the reaction mixture changed color from orange, to dark-orange and finally to a blood-red, homogeneous mixture. At which point, the mixture was filtered through a bed of Celite, the Celite pad was washed with THF and the blood- red filtrate stripped to dryness under reduced pressure. The red, oily solid was dissolved in 8 mL THF and 80 mL of hexane was added while stirring. A red oil formed at the bottom of the flask which was converted to a yellow precipitate with repeated stirring and scratching with a spatula. The solids were isolated on a fritted glass funnel, washed with 4 * 25 mL hexane and dried under reduced pressure for 8 hours yielding the product as a golden-yellow powder (2.056 g, 2.55 mmol, 70%). X-ray quality crystals were grown from a concentrated diethyl ether solution at -35 0C yielding [Li(OEt2)] 2 [20]. 1H NMR (300 MHz, C6D6 ) 6 6.47-7.85 (in, 35H, phenyl), 3.32 (br. s, 8H, THF, v i = 105 Hz), 1.18 (br. s, 8H, THF, vi = 31 Hz) ppm. 'H NMR (500 22 MHz, THF-d8) 6 7.55 (2H, phenyl), 7.10-7.01 (in, 10H, phenyl), 6.86 (in, 1H, phenyl), 6.78-6.71 (in, 6H, phenyl), 6.63 (t, 4H, phenyl), 6.51-6.46 (in, 8H, phenyl), 6.39 (in, 2H, phenyl), 5.81 (in, 2H, phenyl), 3.62 (m, 4H, THF), 1.77 (in, 4H, THF) ppm. 3 C NMR (125.8 MHz, THF-d8) 6 158.77, 158.2 (d, 14.4 Hz), 149.34 (d, 9.4 Hz), 148.58 (d, 10.1 Hz), 147.04 (d, 23.1 Hz), 133.33, 133.30 (d, 18.7 Hz), 130.96, 127.81, 127.39, 126.94 (d, 26.6 Hz), 124.45 (d, 2.9 Hz), 124.01, 123.64, 122.76, 121.72, 31 118.34, 111.99, 68.38 (THF), 26.48 (THF) ppm. P NMR (121.5 MHz, C6D6) 6 7.69 ppm. 31p NMR (202.5 MHz, THF-d8) 6 8.94 ppm. Anal Calcd. for C54H51N2PO2Li2: C, 80.58; H, 6.39; N, 3.48; P, 3.85. Found: C, 78.82; H, 6.74; N, 3.24; P, 3.22.

126 5.6.5 Synthesis of Li2PMes.

6 Li2PMes was prepared by modification of a literature preparation for Li2PPh. MesPH 2 (2.00 g, 13.1 mmol) was dissolved in 20 mL hexane and frozen. The phosphine solution was allowed to partially thaw and a solution of 1.6 M "BuLi (20 mL, 32 mmol) in hexanes was slowly added to the stirring phosphine solution. Upon warming to room temperature, a vibrant-yellow powder spontaneously precipitated out of solution. The yellow solids were isolated on a medium-porosity, fritted-glass filter and washed with 3 *50 mL of hexane. The yellow solids were then dried under reduced pressure until constant mass was obtained. The product was isolated as a bright-yellow powder (2.092 g, 12.75 mmol, 97%) and used without further purification.

5.6.6 Synthesis of [Li(thf)]2{MesP[C(CPh 2)NPh] 2 } ([Li(thf)]2[21]).

MesPLi2 (299 mg, 1.82 mmol) was slurried in 5 mL THF and frozen. A solution of Ph2C=C=NPh (1.00 g, 3.71 mmol) was dissolved in 10 mL THF and frozen as well. The thawing solution of

Ph2C=C=NPh was slowly added to the thawing MesPLi2 slurry while stirring. The stirring reaction mixture was allowed to warm to room temperature as it gradually changed color from bright-yellow to a blood-red, homogeneous solution. After 25 minutes, an aliquot was taken for 31P NMR analysis showing complete consumption of starting material and formation of the desired product at S= 16.7 ppm (in proteo-THF).45 minutes after the reagents were combined, the reaction mixture was filtered through a bed of Celite and the Celite pad was washed with 2*5 mL of THE. The combined filtrates were stripped to dryness under reduced pressure and the blood-red residue was triturated with 4*50 mL of hexane. Product was isolated on a medium-porosity, fritted-glass filter and washed with 3*20 mL of hexane. The solids were then dried under reduced pressure until constant mass was obtained. The desired product was isolated as a bright-yellow powder (1.114 g, 1.315 mmol, 72%). IH NMR

(300 MHz, C6D6) 8 7.32 (in, 4H), 7.31 (d, 4H), 7.07-6.92 (in, 14H), 6.82 (in, 6H) 6.54 (t, 2H), 6.40 (d, 2H), 3.25 (br. m, 4H, THF), 2.75 (s, 6H, o-Me), 1.85(s, 3H, p-Me), 1.20 (br. m, 4H, THF) ppm. 3 1 C NMR (125.8 MHz, C6D6) 6 159.73 (d, 14.4 Hz), 157.54, 149.35 (d, 14.5 Hz), 148.34 (d, 10.8 Hz), 144.22 (d, 12.2 Hz), 137.13 (d, 34.6 Hz), 133.98 (d, 1.4 Hz), 138.72, 130.98, 129.20, 127.55, 126.69 (d, 13.7 Hz), 122.58, 121.48. 121.15 (br. s), 120.80 (br. s), 118.23 (br. s), 111.10, 24.141 4 31 31 (d, JPH = 9.4 Hz, o-Me), 21.04 (p-Me) ppm. P NMR (121.5 MHz, C6D6) 6 5.92 ppm. P NMR (202.5 MHz, THF-d8) 6 11.44 ppm. Anal Calcd. for C57H57N2PO2Li 2: C, 80.83; H, 6.78; N, 3.31; P, 3.66. Found: C, 79.95; H, 7.26; N, 2.98; P, 2.69.

5.6.7 Synthesis of {PhP[C(CPh 2)NPh] 2}TaMe 3 (22).

[Li(thf)]2[20] (461 mg, 0.573 mmol) was dissolved in 5 mL C6H6 and frozen. A solution of TaMe 3Cl2 (170 mg, 0.572 mmol) in 2 mL C6H6 was slowly added to the thawing solution of [Li(thf)]220 which was stirred and allowed to warm to room temperature. No color change was

127 observed. The reaction mixture was stirred for 15 min, after which time the volatiles were removed under reduced pressure. The red residue was redissolved in 5 mL benzene and the volatiles again removed under reduced pressure. The residue was again dissolved in benzene, filtered through a bed of Celite and the filtrate was stripped to dryness under reduced pressure. The product was obtained as a brick-red solid (455 mg, 0.521 mmol, 91%). 1H NMR (300 MHz, C6 D6) 8 7.20-7.90 (in, 5H, phenyl), 6.40-7.10 (in, 30H, phenyl), 1.23 (s, 9H, methyl) ppm. 'H NMR (500 MHz, CDCl3) 7.39 (s, 2H, phenyl), 7.31 (t, 2H, phenyl), 7.14 (s, 8H, phenyl), 7.10-7.02 (in, 9H, phenyl), 6.97 (t, 2H, 3 phenyl), 6.93-6.82 (in, 12H, phenyl), 0.81 (s, 9H, TaMe 3) ppm. C NMR (125.8 MHz, CDCl3) 6 154.38 (d, 33.8 Hz), 150.47, 142.76 (d, 5.8 Hz), 141.36 (d, 5.0 Hz), 140.63 (d, 28.8 Hz), 134.73 (d, 22.3 Hz), 130.46 (d, 5.05 Hz), 129.51 (d, 1.4 Hz), 128.80, 128.53. 128.45, 127.71, 127.65, 127.53, 31 127.50. 127.44, 127.38, 123.09, 121.70 (d, 1.4 Hz), 74.50 (TaMe3) ppm. P NMR (121.5 MHz, 31 Anal Calcd for C H N PTa: C, C6D6) 6 0.85 ppm. P NMR (202.5 MHz, CDCl3) 6 1.23 ppm. 49 44 2 67.43; H, 5.08; N, 3.21; P, 3.55. Found: C, 65.92; H, 5.52; N, 2.97; P, 3.79.

5.6.8 Synthesis of {MesP[C(CPh 2)NPh] 2}TaMe 3 (23).

A solution of TaMe 3Cl2 was dissolved in 2 mL C6H6 and added to a stirring solution of [Li(thf)]2[21] (500 mg, 0.590 mmol) in 6 mL of C6H6 . The mixture was stirred for 10 minutes and then filtered through a bed of Celite. The Celite pad was washed with 5 mL of C6H6 and the combined filtrates were reduced to dryness under reduced pressure. The blood-red residue was triturated with 5 mL benzene and again stripped to dryness. The residue was then dissolved in 5 mL of C6H6 , filtered through a bed of Celite, frozen and freeze dried. The product was obtained as a brick-red powder (520 mg, 0.568 mmol, 96%). 1H NMR (300 MHz, C6D6) 6 = 7.5-6.6 (in, 32H, phenyl), 2.42 (s, 3H, methyl), 2.17 (s, 3H, methyl), 1.86 (s, 3H, methyl), 1.5-0.8 (v. br. s, 9H, methyl) ppm. 'H NMR (500 MHz, THF-d') 6 7.26-6.71 (br. m, 30H, phenyl), 6.37 (br. s, IH, m-Mes), 5.96 (br. s, 1H, m-Mes). 2.13 (br. s, 3H, o-Me), 1.95 (s, 311, p-Me), 1.87 (br. s, 3H, o-Me), 31 31 1.2-0.4 (v. br. s, 9H, TaMe 3) ppm. P NMR (121.5 MHz, C6 D6) -1.24 ppm. P NMR (202.5 8 MHz, THF-d ) 6 -1.70 ppm. Anal Calcd for C52H50N2PTa: C, 68.27; H, 5.51; N, 3.06; P, 3.39. Found: C, 68.08; H, 5.80; N, 2.90; P, 3.36.

5.6.9 Synthesis of {MesP[C(CPh 2)NPh] 2}Ta(I)Me 2 (24).

Compound 23 (200 mg, 0.219 mmol) was dissolved in 7 mL of diethyl ether and frozen. Iodine (56 mg, 0.22 mmol) was dissolved in 3 mL of diethyl ether and frozen as well. The two solutions were allowed to partially thaw, at which point the thawing iodine solution was added to the thawing tantalum solution while stirring. The mixture was allowed to warm to room temperature and stirred for 1 h. The volatiles were removed from the cloudy, dark-red reaction mixture under reduced pressure. The solids were redissolved in 10 mL diethyl ether and filtered through a bed of Celite. The filtrate was again stripped to dryness, redissolved in 10 mL diethyl ether and filtered again. The

128 volatiles were removed from the filtrate under reduced pressure yielding the desired product as a dark-red powder (157 mg, 0.153 mmol, 69.8%). 'H NMR (500 MHz, CDC13) 6 7.43-6.73 (br. m, 30H, aryl), 6.41 (d, 5.2 Hz, 1H, aryl), 5.97 (s, lH, aryl), 1.96 (s, 3H, Me), 1.88 (d, 4 JPH = 2.3 Hz, 3H, o-Me), 1.82 (s, 3H, Me), 1.70 (s, 3H, Me), 0.59 (s, 3H, TaMe) ppm. 3C NMR (125.8 MHz,

CDCl3) 6 169.06 (br. s), 157.10 (br. s), 151.75 (br. s), 148.33 (br. s), 144.56, 144.25, 143.65, 143.61, 143.04, 142.27, 139.82, 138.62, 138.40, 137.87, 131.12, 130.43, 129.94, 129.62, 129.14, 128.62, 128.52, 128.45, 128.16, 127.96, 127.87, 127.74, 127.49, 127.11, 126.73, 126.52, 124.54 (br. s), 123.85 (br. s), 123.08 (br. s), 120.97 (br. s), 120.34 (br. s), 83.13 (s, TaMe), 80.12 (s, TaMe), 3 31 24.22 (s, o-Me), 23.52 (d, JPc, o-Me), 20.84 (p-Me) ppm. P NMR (121.5 MHz, CDCl3) 3 6 -17.12 ppm. 'P NMR (121.5 MHz, C6D6) 6 -17.2 ppm. Anal Caled for C5jH47N2PITa: C, 59.66; H, 4.61; N, 2.73; P, 3.02. Found: C, 59.65; H, 4.79; N, 2.72; P, 2.84.

5.7 Crystallographic Structure Determinations

5.7.1 General Considerations

X-ray data collections were carried out on a Siemens Platform three-circle diffractometer mounted with an APEX-CCD detector and outfitted with a low-temperature, nitrogen-stream aperture. Graphite monochromated Mo-Ka radiation (X = 0.71073 A) was used in all cases. All software for diffraction data processing and crystal-structure solution and refinement are contained in the SHELXTL (v6.14) program suite (G. Sheldrick, Bruker XRD, Madison, WI).19

5.7.2 X-ray crystal structure of [Li(OEt2)]2{ PhP[C(CPh 2)NPh] 2} ([Li(OEt 2)]2[20])

Inside the glovebox, crystals of [Li(thf)] 2[20] obtained from a saturated diethyl ether solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A yellow block of approximate dimensions 0.40 * 0.20 * 0.15 mm 3 was selected and mounted on a glass fiber. A total of 65075 reflections (-25 < h < 25, -14 < k < 14, -22 < 1 < 22) were collected at 100(2) K using $- and co-scans in the 8 range of 1.05 to 24.71', of which 7887 were unique (Rit = 0.0580). The structure was solved by direct methods using SHELXS1 9 and refined against F2 on all data by full-matrix least squares with SHELXTL. 19 The systematic absences in the diffraction data are consistent with the assigned space group of P21/c. There was a non-crystallographically imposed positional disorder of roughly half of the [Li(thf)] 2 [20] structure. The disorder had two positions in a ca. 80:20 ratio of occupancy-the ratios were refined freely, while constraining the total occupancy of both components to unity. The disorders were refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in

129 maximum and minimum transmissions equal to 0.9849 and 0.9605, respectively. The residual peak and hole electron density were 0.714 and -0.326 e-A-3, respectively. The least squares refinement converged normally with residuals of R1 = 0.0490 for I > 2i(I), wR 2 = 0.1265 for all data, and GOF = 1.036 (based on F 2).

Crystal data: formula C54 H55N20 2Li2 , space group P21/c, a = 21.6250(15) A, b = 12.1378(8), c = 19.6341(14) A, p = 116.2980(10)0, V = 4620.2(6) A3 . Z = 4, p = 0.102 mm- 1, Dcaic = 1.163 g-cm- 3, F(000) = 1720.

5.7.3 X-ray crystal structure of {PhP[C(CPh 2)NPh] 2}TaMe 3 (22)

Inside the glovebox, crystals of 22 obtained from a saturated diethyl ether/hexamethyldisiloxane solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A red block of approximate dimensions 0.29 * 0.29 * 0.18 mm3 was selected and mounted on a glass fiber. A total of 117624 reflections (-26 < h K 25, -18 < k K 18, -25 K 1 < 25) were collected at 100(2) K using $- and o-scans in the 0 range of 1.12 to 29.57', of which 12612 were unique (Rit = 0.0480). The structure was solved by direct methods using SHELXS1 9 and refined against F 2 on all data by full-matrix least squares with SHELXTL. 19 The systematic absences in the diffraction data are consistent with the assigned space group of P21/c. There was a one molecule of diethyl ether of co-crystallization present within the asymmetric unit. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.6601 and 0.5289, respectively. The residual peak and hole electron density were 1.277 and -0.918 e-A~3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0270 for I > 20-(I), wR 2 0.0645 for all data, and GOF = 1.078 (based on F2 ).

Crystal data: formula C49H44N 2PTa (with 1 equiv. of Et20, thus giving a total empirical formula of C53H54N2POTa), space group P21/c, a = 19.3194(19) A, b = 13.6727(13), c = 18.1043(18) A, = 109.730(2)', V = 4501.1(8) A3 . Z = 4, p 2.517 mm-1, Dcalc = 1.397 g.cm-3, F(000) = 1928.

5.7.4 X-ray crystal structure of {MesP[C(CPh 2)NPh] 2}TaMe3 (23)

Inside the glovebox, crystals of 23 obtained from a saturated diethyl ether solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A red block of approximate dimensions 0.20 *0.20 *0.10 mm 3 was selected and mounted on a glass fiber. A total of 82343 reflections (-24 < h < 24, -19 K k K 19, -23 K 1 < 23) were collected at 100(2) K using $- and o-scans in the 0 range of 1.17 to 28.280, of which 11210 were unique (Rint = 0.1064). The structure was solved by direct methods using SHELXS1 9 and refined against F 2 on all data by full-matrix least squares with SHELXTL. 19 The systematic absences in the diffraction data are

130 consistent with the assigned space group of P21 /c. One heavily disordered molecule of diethyl ether of co-crystallization with ca. 50% occupancy located upon a special position was present in the asymmetric unit. The disordered solvent could not be modeled satisfactorily and was removed from the structure using the crystallographic routine SQUEEZE.20 All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.7875 and 0.6338, respectively. The residual peak and hole electron density were 1.770 and - 1.090 e-A- 3 , respectively.

The least squares refinement converged normally with residuals of R1 = 0.0429 for I > 2a(I), wR 2 0.0981 for all data, and GOF = 0.947 (based on F2 ).

Crystal data: formula C52H50N2 PTa (with 0.5 equiv. of Et2 0, thus giving a total empirical formula of C54H55N2PO0 .5Ta), space group P2 1/c, a = 18.087(3) A, b = 14.715(3), c = 17.660(3) A, p = 106.145(3)0, V = 4514.9(14) A3 . Z = 4, p = 2.509 mm-1, Deaic 1.400 g.c m-3, F(000) 1940.

5.7.5 X-ray crystal structure of {MesP[C(CPh 2)NPh] 2}Ta()Me 2 (24)

Inside the glovebox, crystals of 24 obtained from a saturated diethyl ether solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A red shard of approximate dimensions 0.31 * 0.25 * 0.17 mm 3 was selected and mounted on a glass fiber. A total of 128304 reflections (-41 < h < 41, -19 < k < 19, -33 < 1 < 33) were collected at 100(2) K using $- and o-scans in the 0 range of 1.42 to 30.030, of which 13987 were unique (Ri,, = 0.0553). The structure was solved by direct methods using SHELXS1 9 and refined against F2 on all data by full-matrix least squares with SHELXTL. 19 The systematic absences in the diffraction data are consistent with the assigned space group of C2/c. The asymmetric unit contained one molecule of diethyl ether of co-crystallization. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.6279 and 0.4547, respectively. The residual peak and hole electron density were 1.399 and -2.489 e-A-3 , respectively. The least squares refinement converged normally with residuals of R1 = 0.0308 for I > 2a(I), wR 2 = 0.0816 for all data, and GOF = 1.054 (based on F 2).

Crystal data: formula C51H47N2 POTa (with 1 equiv. of Et 2 0, thus giving a total empirical formula of C55H57N2PIOTa, space group C2/c, a = 19.786(4) A, b = 14.015(2), c = 23.808(3) A, = 105.774(2)0, V = 9564(2) A3. Z = 8, p = 3.018 mm-1, Dcale = 1.529 g-cm-3, F(000) = 4400. Table 5.1. Crystallographic Data for [Li(OEt2]2 {PhP[C(CPh2)NPh]2} ([Li(OEt2)]2[20]), {PhP[C(CPh2)NPh]2}TaMe3 (22) and {MesP[C(CPh9)NPh],}TaMe,(23) [Li(OEt2)]2[20] 22 23 Reciprocal Net code / CCDC code 09118 / 823185 09198 / 823186 09271 / 823187 Empirical formula, FW (g/mol) 808.85 C54H55Li2N2O2P, C53H54N2OPTa, 946.90 C54H55N2O0 .5PTaa 951.62a Color / Morphology Yellow / Block Red / Block Red / Prism Crystal size (mm3) 0.40 *0.20 * 0.15 0.29 * 0.29 * 0. 18 0.20 * 0.20 * 0. 10 Temperature (K) 100(2) 100(2) 100(2) Crystal system, Space group Monoclinic, P21/c Monoclinic, P21i/c Monoclinic, P21/c Unit cell dimensions (A,0) a = 21.6250(15), ac = 90 a = 19.3194(19), a = 90 a = 18.087(3), a = 90 b = 12.1379(8), P = 116.2980(10) b = 13.6727(13), P = 109.730(2) b = 14.715(3), s = 106.145(3) c = 19.6341(14), y = 90 c = 18.1043(18),y = 90 c = 17.660(3), y = 90 Volume(A 3 ) 4620.2(6) 4501.5(8) 4514.9(14) Z 4 4 4 Density (calc., Mg/m3) 1.163 1.397 1.400 Absorption coefficient (mm-1) 2.517 3.011 2.509 F(O00) 1720 1928 1940 Theta range for data collection (0) 1.05 to 24.71 1.12 to 29.57 1.17 to 28.28 Index ranges -25

Final R indices' [I > 2c(I)] R = 0.0490, wR 2 = 0.1142 R1 = 0.0270, wR2 = 0.0589 R1 = 0.0429, wR 2 = 0.0911 R indices' (all data) R1 = 0.0672, wR 2 = 0.1265 R1 = 0.0365, wR2 = 0.0645 R1 = 0.0678, wR 2 = 0.0981 Largest diff. peak and hole (e .- 3) 0.714 and -0.326 1.277 and -0.918 1.770 and -1.090

Halfan equivalentof heavilydisordered diethyl ether was removedfrom the asymmetricunit using the crystallographicroutine SQUEEZE. 20 b GooF = np) 2 . _ 2F2+max(F ,O) cR,_ E||F|-|F . wR2 = 2 ;1 p Y IFoI '[W(F 2 ' " -- 2(F2)+(aP)2+bP' 3 Table 5.2. Crystallographic Data for {MesP[C(CPh2)NPh] 2}Ta(I)Me 2 (24)

Reciprocal Net code / CCDC code 09351 / 823188 Empirical formula, FW (g/mol) C55 H57 N2OPITa, 1100.85 Color / Morphology Red / Shard Crystal size (mm3) 0.31 *0.25 * 0.17 Temperature (K) 100(2) Crystal system, Space group Monoclinic, C2e

Unit cell dimensions (A, 0) a = 19.786(4), a = 90 b = 14.015(2), P = 105.774(2) c = 23.808(3), y = 90 Volume (A3) 9564(2) Z Density (calc., Mg/m 3 ) 1.529 Absorption coefficient (mm-1 ) 3.018 F(000) 4400 Theta range for data collection (0) 1.42 to 30.03 Index ranges -41

Max. and min. transmission3. 0.62791.4 1 to 30.0and 0.4547 Data'GooF= / restraints[Y[w(F-41F2h141,-19 / parameters = 13987 / 0 / 555 k119, 2a(I)] R, 0.0308, wR2 0.0765 R indicesb (all data) R, = 0.0384, wR 2 = 0.0816 Largest duff. peak and hole (e A-3) 1.399 and -2.489 R1 = 0.030, wR2 0.076

Cy(F,2)+(aP)2+b-P 3 5.8 References

[1] Goikhman, R.; Aizenberg, M.; Ben-David, Y.; Shimon, L. J. W.; Milstein, D. Organometallics 2002, 21, 5060- 5065. [2] Engeldinger, E.; Poorters, L.; Armspach, D.; Matt, D.; Toupet, L. Chem. Commun. 2004, 634-635. [3] Kant, M.; Bischoff, S. Z Anorg. Allg. Chem. 1999, 625, 707-708. [4] Hcher, T.; Cinquantini, A.; Zanello, P.; Hey-Hawkins, E. Polyhedron 2005, 24, 1340-1346. [5] Krummenacher, I.; Clough, C. R.; Cummins, C. C. Manuscriptin preparatiton. [6] Kister, R.; Seidel, G.; MUller, G.; Boese, R.; Wrackmeyer, B. Chem. Ber 1988, 121, 1381-1392. [7] Bestmann, H. J.; Lienert, J.; Mott, L. Liebigs Ann. Chem. 1968, 718, 24-32.

[8] Synthesis of MesPH 2: Masuda, J. D.; Jantunen, K. C.; Ozerov, 0. V.; Noonan, K. J. T.; Gates, D. P.; Scott, B. L.; Kiplinger, J. L. J. Am. Chem. Soc. 2008, 130, 2408-2409.

[9] Li2PMes was synthesized by modification of a published preparation for Li2PPh: Kbster, R.; Seidel, G.; Muller, G.; Boese, R.; Wrackmeyer, B. Chem. Ber 1988, 121, 1381-1392. [10] Meth-Cohn, 0.; Rhouati, S.; Tarnowski, B.; Robinson, A. J. Chem. Soc., Perkin Trans. 1 1981, 1537-1543. [11] Schrock, R. R.; Sharp, P. R. J.Am. Chem. Soc. 1978, 100, 2389-2399. [12] Schrock, R. R.; Seidel, S. W.; Schrodi, Y; Davis, W. M. Organometallics 1999, 18, 428-437. [13] MacLachlan, E. A.; Fryzuk, M. A. Organometallics2005, 24, 1112-1118. [14] MacKay, B. A.; Munha, R. F.; Fryzuk, M. D. J. Am. Chem. Soc. 2006, 128, 9472-9483. [15] Liang, L.-C. Coord. Chem. Rev. 2006, 250, 1152-1177. [16] Becker, G.; Heck, J. R.; Hubler, U.; Schwarz, W.; WUrthwein, E.-U. Z Anorg. Allg. Chem. 1999, 625, 2008-2024. [17] Masuda, J. D.; Jantunen, K. C.; Ozerov, 0. V.; Noonan, K. J. T.; Gates, D. P.; Scott, B. L.; Kiplinger, J. L. J. Am. Chem. Soc. 2008, 130, 2408-2409. [18] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. [19] Sheldrick, G. M. Acta Cryst. A 2008, 64, 112-122. [20] van der Sluis, P.; Spek, A. L. Acta Cryst. A 1990, 46, 194-201.

134 APPENDIX A

Unpublished Crystal Structures

A.1 Introduction

This Appendix consists of four crystal structures that, for whatever reason, did not fit into the

"cohesive story" that is this thesis. A structure of NW(O-t-Bu)(N[i-Pr]Ar) 2 is presented in this Appendix, where it holds the unique distinction of being the only complex presented here that was synthesized deliberately. The other structures described here (OW(N[i-Pr]Ar) 2(Cl) 2, [Ar(i- Pr)NBH 2] 2 and [(CH 3CN)2V(O)(C)-p-Cl] 2) were all unexpected byproducts mounted on the diffractometer in crystal fishing expeditions. The syntheses of all compounds presented here are described in as much detail as is available. The purpose of this Appendix is to have a written record of unwanted byproducts so that future researchers can avoid collecting their structures again. To this end, all the structures in this Appendix have been submitted to the Cambridge Structural Database (http: / /www. ccdc. cam. ac. uk/) so that the unit cell parameters are contained in an online, searchable format.

A.2 Discussion

A.2.1 Solid-state structure of NW(O-t-Bu)(N[i-Pr]Ar)2

A full discussion of NW(O-t-Bu)(N[i-Pr]Ar) 2 is described in Chapter One, Section 1.2. The solid- state structure of NW(O-t-Bu)(N[i-Pr]Ar) 2 is shown in Figure A.l. The structure displays a pseudo- tetrahedral ligand environment around tungsten. The tungsten-nitride bond length of 1.680(2) A is similar to nitride 1 (dW-N = 1.669(5) A).

135 Figure A.1. Solid-state structure of NW(O-t-Bu)(N[i-Pr]Ar) 2 with thermal ellipsoids at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A): Wl-N3 1.680(2), Wl-N1 1.9598(18), Wl-N2 1.9391(19), WI-01 1.8900(15).

A.2.2 Solid-state structure of OW(N[i-Pr]Ar) 2(CI) 2

OW(N[i-Pr]Ar) 2(Cl) 2 was an unexpected byproduct in the reaction of oxochloride (Ar[i- Pr]N)3W(O)Cl (2) with triflic anhydride (see Chapter One, Section 1.5). No rational synthesis has been developed for this complex-it is presumably the result of a disproportion reaction between two equivalents of compound 2 in the presence of triflic anhydride. The crystals of OW(N[i-Pr]Ar) 2(Cl) 2 were grown from a saturated diethyl ether solution at -35 'C while trying to isolate crystalline oxotriflate 8 (Figure A.2). The structure of OW(N[i-Pr]Ar) 2(Cl) 2is that of a distorted, pseudo- trigonal bipyramid with the anilide and oxo ligands in the equatorial plane. The equatorial plane is near perfect with Eequatorial = 360.0(3)0; however, the Cl-W-Cl bond angle is slightly distorted from linear at 177-178'. The W-O bond length of 1.691(8) A in OW(N[i-Pr]Ar) 2(Cl) 2 is statistically identical to the corresponding bond in oxochloride (Ar[i-Pr]N) 3W(O)Cl (dw-o = 1.698(3) A).

The asymmetric unit in the structure of OW(N[i-Pr]Ar) 2(Cl) 2 is rather unique. It consists of two independent fragments where each fragment consists of half of the entire molecule. The full molecules are generated by a C2 axis contained within the space group P21/c.

136 02

C12A C12

W2 N2A N2

Figure A.2. Solid-state structure of OW(N[i-Pr]Ar) 2(Cl)2 with thermal ellipsoids at the 50% probability level. The asymmetric unit contains two independent, half molecules, which are both shown here. Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and angles ('): WI-O1 1.691(8), W1-Nl 1.924(7), Wi-ClI 2.434(2), W2-02 1.691(8), W2-N2 1.931(7), W2-C12 2.432(2), C1l-W1-CllA 177.32(12), Eequatorial = 360.0(3) (around WI), C12-W2-Cl2A 178.01(11), Eequaorial= 360.0(3) (around W2)

137 A.2.3 Solid-state structure of [Ar(i-Pr)NBH212

[Ar(i-Pr)NBH ] was synthesized as 2 2 a byproduct in the reaction of trichloride 12 with LiBH 4 (see Chapter Two, Section 2.4) The bridging-borane dimer of "N(i-Pr)Ar" was an unexpected product when crystals were grown from diethyl ether (in an attempt to crystallize dichloride 13). The structure of [Ar(i-Pr)NBH ] is 2 2 that of two isopropyl anilide ligands bridged by two "BH 2" fragments (Figure A.4). The core of the molecule forms a roughly square shape with B-N-B = 86.04(12)', N-B-N = 93.96(12)0 and all B-N bonds = 1.6184(16) A. The solid-state structure of [Ar(i-Pr)NBH 2] 2 exhibits extremely high symmetry (crystallizing in the tetragonal space group 14/m). Only one quarter of the molecule is present in the asymmetric unit-the full molecule is generated by a crystallographic mirror plane and a C2 axis contained within the space group I4/m (Figure A.3).

C14 C13 C11 C1 31 N1 C12

C18 C7B

Figure A.3. Solid-state structure of the asymmetric unit of [Ar(i-Pr)NBH 2 ]2 with thermal ellipsoids at the 50% probability level. The asymmetric unit consists of ca. 25% of the entire molecule. The entire molecule is generated by a mirror plane and a C2 axis contained within the tetragonal space group I4/m.

138 C17A

Figure A.4. Solid-state structure of [Ar(i-Pr)NBH 2 ]2 with thermal ellipsoids at the 50% probability level. Selected bond lengths (A) and angles (0): Ni-B1 1.6184(16), BI-NI-BlA 86.04(12), Ni-BI- NIA 93.96(12).

A.2.4 Solid-state structure of [(CH 3CN) 2V(O)(Cl)-p-Cl] 2

The dimeric [(CH 3CN) 2V(O)(Cl)-U-Cl] 2 was an unexpected product formed in the reaction of

VOC13 with [PPN]3 [P 3 09] -H 20 in an attempt to synthesize and crystallize compound 19 (see Chapter Four, Section 4.3). The unexpected product was obtained as a green, crystalline solid from an acetonitrile solution of the reaction mixture layered with diethyl ether. [(CH 3 CN)2V(O)(C)- p-Cl] 2 was presumably formed by direct reaction of VOC13 with acetonitrile. In the solid-state

139 of [(CH 3CN) 2V(O)(Cl)-p-Cl] 2 is best described as two "(MeCN) 2V(O)Cl" fragments bridged by two chloride ligands (Figure A.5). Each vanadium center exists in a distorted, pseudo-octahedral coordination environment. The asymmetric unit contained only half of the molecule with the other half being generated by a C2 axis contained within the space group P21/n.

C12A C11

C21 A

N1A C11A O1A

Figure A.5. Solid-state structure of of [(CH 3CN) 2V(O)(Cl)-p-Cl] 2 with thermal ellipsoids at the 50% probability level. The asymmetric unit consists of ca. 50% of the entire molecule. The entire molecule is generated by a C2 axis contained within the space group P21/n. Selected bond lengths (A) and angles (0): V1-01 1.5873(14), VI-Cll 2.3488(6), V1-Cl2 2.3895(6), V1-Cl2A 2.6474(6), Nl-V1-N2 170.42(6), C12-Vl-ClA 78.77(2), V1-Cl2-VlA 101.23(2).

A.3 Experimental Section

A.3.1 General Considerations

Unless stated otherwise, all operations were performed in a Vacuum Atmospheres drybox under an atmosphere of purified nitrogen or using Schlenk techniques under an argon atmosphere.

[PPN]3 [P3 09]- H20 was prepared as previously reported. 1 Tf 2O and LiBH4 were purchased from Aldrich and used without further purification. VOCl 3 was purchased from Strem and used without further purification. Diethyl ether, n-pentane, acetonitrile and methylene chloride were dried and deoxygenated using a system built by SG Water USA, LLC (www. glasscontoursolventsystems. 2 com). Tetrahydrofuran was distilled off of benzophenone ketyl. C6D6 was purchased from

140 Cambridge Isotope Labs, purified by distillation off of CaH 2 and stored over 4 A molecular sieves. Celite, alumina and 4 A molecular sieves were dried under reduced pressure overnight at a temperature above 200 'C. 'H NMR spectra were recorded on Varian Mercury-300 spectrometer.

1H chemical shifts are reported with respect to internal solvent (C6D6, 8 = 7.16 ppm). X-ray data collections were carried out on a Siemens Platform three-circle goniometer with a CCD detector using Mo-Ka radiation, k = 0.71073 A.

A.3.2 Synthesis of NW(O-t-Bu)(N[i-Pr]Ar) 2

The synthesis of NW(O-t-Bu)(N[i-Pr]Ar) 2 is described fully in Chapter One, Section 1.8.4.

A.3.3 Synthesis of OW(N[i-Pr]Ar)2(C) 2

OW(N[i-Pr]Ar) 2(Cl) 2 was synthesized in an attempt to synthesize and purify the tungsten oxotriflate complex 8 (as described in Chapter One, Section 1.8.12). Compound 2 (173 mg, 0.240 mmol) was dissolved in 2 mL of n-pentane and stirred. A solution of Tf 2O (69 mg, 0.24 mmol) in 2 mL of n-pentane was added to the red, stirring solution of 2. Within minutes, the solution lightened and a yellow precipitate formed. The yellow precipitate (107 mg) was isolated by filtration and re-dissolved in minimal diethyl ether. The ethereal solution was filtered through Celite and stored at -35 'C. Orange crystals of OW(N[i-Pr]Ar) 2(Cl) 2 were collected and used for an X-ray crystallographic study. No further spectroscopic details were collected as OW(N[i-Pr]Ar) 2(Cl) 2 was an unwanted byproduct.

A.3.4 Synthesis of [Ar(i-Pr)NBH2]2

[Ar(i-Pr)NBH 2]2 was synthesized in an attempt to synthesize and purify the tungsten dichloride complex 13 (as described in Chapter Two, Section 2.7.6). Trichloride 12 (302 mg, 0.389 mmol) was slurried in 10 mL of tetrahydrofuran and added to solid LiBH 4 (28.8 mg, 1.32 mmol). The reaction mixture immediately turned brown with concomitant effervescence. The mixture was stirred for 10 min, after which time it was filtered through a bed of Celite. The volatiles from the brown filtrate were removed under reduced pressure. The solid, brown residue was triturated with n-pentane, extracted with minimal diethyl ether, filtered through Celite and cooled to -35 'C overnight. The following day, colorless crystals of [Ar(i-Pr)NBH 2]2 had grown and were promptly used for an X- ray diffraction study. No yield was obtained as a significant portion of the product was used for the

X-ray study. As product [Ar(i-Pr)NBH 2]2 was an unwanted byproduct, the reaction conditions were not repeated and formation of [Ar(i-Pr)NBH 2]2 was avoided. 'H NMR (300 MHz, C6D6) 8 7.317 (s, 4H, ortho), 6.67 (s, 2H, para), 3.49 (septet, 2H, i-Pr methine), 2.14 (s, 12H, ArMe), 0.94 (d, 12H, i-Pr methyl) ppm. The hydrogens bound to boron were not observed in the 'H NMR spectrum. A.3.5 Synthesis of [(CH 3CN) 2V(O)(C)-p-Cl] 2

[(CH 3CN) 2V(O)(Cl)-p-Cl] 2 was synthesized in an attempt to synthesize and isolate compound 19 (as described in Chapter Four, Section 4.5.4). [PPN]3[P30 9]- H20 (385 mg, 0.206 mmol) was dissolved in 1 mL acetonitrile and loaded into an NMR tube. The solution was then layered with ca. 1 mL of neat acetonitrile followed by ca. 0.1 mL diethyl ether. VOCl 3 (36 mg, 0.21 mmol) was dissolved in 1.5 mL of diethyl ether and carefully layered atop the [PPN] 3 [P3 09] H20 solution. The NMR tube was capped and sealed with electrical tape. The NMR tube was taped to the faceplate of the glovebox where it could be easily observed without disturbance and checked regularly for crystal growth. At the offset, the experiment consisted of a colorless [PPN] 3 [P309] -H20 solution with a blood-red VOCl 3 solution atop it. After two days, a small amount of crystalline material began forming in the upper half of the tube (in the Et20 layer). After four days, the color of the tube had fully bleached from dark-red to light-blue. On the fifth day, diffractometer time was available and small, light-green crystals were harvested and subject to an X-ray crystallographic study providing the solid-state structure of

[(CH 3CN) 2V(O)(Cl)-p-Cl] 2 presented in this appendix. Again, as [(CH 3CN) 2V(O)(Cl)-p-Cl] 2 was an unwanted byproduct, synthesis of the material was not attempted again and the material was not fully characterized.

A.4 Crystallographic Structure Determinations

A.4.1 General Considerations

A.4.2 X-ray crystal structure of NW(O-t-Bu)(N[i-Pr]Ar)2

Inside the glovebox, crystals of NW(O-t-Bu)(N[i-Pr]Ar) 2obtained from a saturated diethyl ether solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A colorless block of approximate dimensions 0.07 * 0.07 * 0.07 mm 3 was selected and mounted on a glass fiber. A total of 19941 reflections (-13 < h < 5, -22 < k < 21, -22 < 1 < 22) were collected at 193(2) K using 0- and w-scans in the 0 range of 1.73 to 28.28', of which 6788 were unique (Rint = 0.0191). The structure was solved by direct methods using SHELXS 3 and refined against F2 on all data by full-matrix least squares with SHELXTL.3 The systematic absences in

142 the diffraction data are consistent with the assigned space group of P21 /c. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.7557 and 0.7557, respectively. The residual peak and hole electron density were 0.971 and -0.532 e-A-3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0205 for I > 2a(I), wR 2 = 0.0501 for all data, and GOF = 1.040 (based on F 2).

Crystal data: formula C26H41N 3 0W, space group P2 1/c, a = 9.9242(5) A, b = 16.6504(8), c = 3 17.1598(8) A, = 105.0700(10)0, V = 2738.0(2) A . Z = 4, p = 4.239 mm-1, Dcaic = 1.445 g-cm F(000) = 1200.

A.4.3 X-ray crystal structure of OW(N[i-Pr]Ar)2(CI) 2

Inside the glovebox, crystals of OW(N[i-Pr]Ar) 2(C1)2 obtained from a saturated diethyl ether solution at -35 0C, were coated with Paratone N oil (an Exxon product) on a microscope slide. An orange block of approximate dimensions 0.40 * 0.23 * 0.20 mm 3 was selected and mounted on a glass fiber. A total of 11757 reflections (-19 < h < 15, -11 < k < 11, -17 < 1 < 15) were collected at 193(2) K using $- and o-scans in the 0 range of 1.25 to 24.40', of which 4003 were unique (Rint = 0.0277). The structure was solved by direct methods using SHELXS 3 and refined against F 2 on all data by full-matrix least squares with SHELXTL. 3 The systematic absences in the diffraction data are consistent with the assigned space group of P21/c. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.4352 and 0.2401, respectively. The residual peak and hole electron density were 2.458 and -2.702 e-A- 3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0373 for I > 2G(I), wR 2 = 0.0988 for all data, and GOF = 1.175 (based on F 2).

Crystal data: formula C22H32N2Cl2 0W, space group P21/c, a = 16.819(5) A, b = 9.989(3), c = 14.982(4) A, p = 105.202(5)0, V = 2429.0(12) A3 . Z = 4, p = 4.990 mm- 1, Dcalc = 1.628 gcm-3, F(000) = 1176.

A.4.4 X-ray crystal structure of [Ar(i-Pr)NBH 2]2

Inside the glovebox, crystals of [Ar(i-Pr)NBH 2]2 obtained from a saturated diethyl ether solution at -35 'C, were coated with Paratone N oil (an Exxon product) on a microscope slide. A colorless prism of approximate dimensions 0.20 * 0.09 * 0.09 mm 3 was selected and mounted on a glass fiber. A total of 21915 reflections (-17 < h < 17, -- 17 < k < 17, -19< 1 < 19) were collected at 100(2) K using $- and o-scans in the 0 range of 2.20 to 30.03', of which 1650 were unique (Rin

143 = 0.0315). The structure was solved by direct methods using SHELXS 3 and refined against F 2 on all data by full-matrix least squares with SHELXTL.3 The systematic absences in the diffraction data are consistent with the assigned space group of I4/m. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were placed in geometrically calculated positions and refined using a riding model. A semi-empirical absorption correction (SADABS) was applied to the diffraction data resulting in maximum and minimum transmissions equal to 0.9946 and 0.9880, respectively. The residual peak and hole electron density were 0.568 and -0.227 e-A 3, respectively.

The least squares refinement converged normally with residuals of R1 = 0.0592 for I > 2a(I), wR2 = 0.1693 for all data, and GOF = 1.078 (based on F 2 ).

Crystal data: formula C22H36B2 N2, space group 14/m, a = 12.5551(8) A, b = 12.5551(8), c = 13.7118(9) A, a = = 'y= 90', V = 2161.4(2) A3. Z = 4, p = 0.061 mm- 1, Dcalc = 1.076 g.cm-3, F(000) = 768.

A.4.5 X-ray crystal structure of [(CH 3CN) 2V(O)(Cl)-p-Cl] 2

Inside the glovebox, crystals of [(CH 3CN) 2V(O)(Cl)-p-Cl] 2obtained from an acetonitrile solution layered with diethyl ether solution at room temperature, were coated with Paratone N oil (an Exxon product) on a microscope slide. A green shard of approximate dimensions 0.18 * 0.11 * 0.10 3 mm was selected and mounted on a glass fiber. A total of 21182 reflections (-12 < h < 12, - 11 2a(I), wR 2 = 0.0658 for all data, and GOF = 1.076 (based on F ).

Crystal data: formula C8H12N4Cl40 2V2, space group P21/n, a = 9.1485(14) A, b = 8.7230(13), c = 10.8184(17) A, = 102.299(3)0, V = 843.5(2) A3 . Z = 2, p = 1.744 mm- 1, Dcaic = 1.732 g-cm- 3 , F(000) = 436.

144 TableA.1. Crystallographic Data for NW(O-t-Bu)(N[i-Pr]Ar)q,OW(N[i-Pr]Ar),(C),, and [Ar(i-Pr)NBH,],

NW(O-t-Bu)(N[i-Pr]Ar) 2 OW(N[i-Pr]Ar) 2 (C1)2 [Ar(i-Pr)NBH 2 ]2 Reciprocal Net code / CCDC code 04083 / 823804 03249 / 823805 05077 / 823806 Empirical formula, FW (g/mol) C26 H41N30W, 595.47 C22H32N2 0C12 W, 595.25 C22 H36B2N2 , 350.15 Color / Morphology Colorless / Block Orange / Block Colorless / Prism Crystal size (mm3) 0.07 * 0.07 * 0.07 0.40 *0.23 * 0.20 0.20 * 0.09 * 0.09 Temperature (K) 193(2) 193(2) 100(2) Crystal system, Space group Monoclinic, P21/c Monoclinic, P2/c Tetragonal, 14/m Unit cell dimensions (A, 0) a = 9.9242(5), a = 90 a = 16.819(5), a = 90 a = 12.5551(8), a = 90 b = 16.6503(8), P = 105.0700(10) b = 9.989(3), P = 105.202(5) b = 12.5551(8), P = 90 c = 17.1598(8),y = 90 c = 14.982(4),y = 90 c = 13.7118(9), y = 90 Volume (A3 ) 2738.0(2) 2429.0(12) 2161.4(2) Z 4 4 4 Density (calc., Mg/m3) 1.445 1.628 1.076 Absorption coefficient (mm-1) 4.239 4.990 0.061 F(000) 1200 1176 768 Theta range for data collection (0) 1.73 to 28.28 1.25 to 24.40 2.20 to 30.03 Index ranges -13

Final R indicesb [I> 2()] Ri = 0.0205, wR 2 = 0.0488 Ri = 0.0373, wR 2 = 0.0964 R1 = 0.0592, wR2 = 0.1618 R indicesb (all data) Ri = 0.0241, wR 2 = 0.0501 R1 = 0.0441, wR 2 = 0.0988 R1 = 0.0660, wR 2 0.1693 Largest diff. peak and hole (e -A-3) 0.971 and -0.532 2.458 and -2.702 0.568 and -0.227

2 a [X[w(F- F 2)1 i b R |F|-|F| R [w(F -F)222. . __ __ 2Fa+max(F,0) GooF (n-p)wRF =.FY)+(aP2+bP')2 [' [w(F]) | - TableA.2. Crystallographic Data for [(CH3CN)9V(O)(Cl)-p-Cl], [(CH3CN)2V(O)(Cl)-p-Cl]2 Reciprocal Net code / CCDC code 09058 / 823807 Empirical formula, FW(g/mol) C8H12N402 C14V2, 439.90 Color / Morphology Green / Shard Crystal size (mm3) 0.18 *0.11 *0.10 Temperature (K) 100(2) Crystal system,Space group Monoclinic, P21/n Unit cell dimensions (A, 0) a = 9.1485(14), a = 90 b = 8.7230(13), s= 102.299(3) c = 10.8184(17), y = 90 Volume (A3) 843.5(2) Z 2 Density (calc., Mg/m3) 1.732 Absorption coefficient (mm-') 1.744 F(000) 436 Theta range for data collection (0) 2.65 to 29.13 Index ranges - 12 2T(I)] R1 = 0.0269, wR2 = 0.0603 R indicesb (all data) R1 = 0.0359, wR2 = 0.0658 Largest diff. peak and hole (e -A-3) 0.407 and -0.303

2 aGooF= [FC] R |F|wR 2 = [w(F- W (n-p) Z|Fi '[w(Fm)2a, __ I -. p. _ 2F,2+max(F,2,0) &;(F,1)+(aP)2+bP' 3 A.5 References

[1] Klemperer, W. G.; Main, D. J. Inorg. Chem. 1990, 29, 2355-2360. [2] Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organometallics 1996, 15, 1518- 1520. [3] Sheldrick, G. M. Acta Cryst. A 2008, 64, 112-122.

147 148 Christopher R. Clough Massachusetts Institute of Technology Phone: (815) 236-5972 Department of Chemistry 77 Massachusetts Avenue, Room 6-327 E-mail: [email protected] Cambridge, MA 02139

Education

Ph.D. in Inorganic Chemistry, Massachusetts Institute of Technology, Cambridge, MA, 2011. M.S. in Chemistry, The University of Chicago, Chicago, IL, 2002. B.S. with Honors in Chemistry, The University of Chicago, Chicago, IL, 2002.

Research Experience

2008-2011 Investigation of Trimetaphosphate Complexes and Bis-Enamide Ligands Advisor: Prof. Christopher C. Cummins, Massachusetts Institute of Technology 2006-2008 Synthesis of High Quantum Yield Semi-Conducting Nanocrystals QD Vision, Inc., Watertown, MA 2002-2005 Reactivity Studies of a Tungsten Terminal Nitride Advisor: Prof. Christopher C. Cummins, Massachusetts Institute of Technology 2000-2002 Mechanistic Studies of Oxidative Additions and Reductive Eliminations of N-tosylaziridines to and from Nickel Complexes Advisor: Gregory L. Hillhouse, The University of Chicago

Awards

" Massachusetts Institute of Technology, Dept. of Chemistry Teaching Award, 2003 e Elected to Sigma Xi as an associate member, 2002 " F & I. Scherer Memorial Undergraduate Award for Oustanding Research in Chemistry, 2001 " C. W. Chang Book Prize for Undergraduate Studies of Chemistry

Publications

C. R. Clough, C. C. Cummins. A Bidentate Bis-Enamide Ligand Preparation by Double Ketenimine Addition to Dilithium Arylphosphanides: Synthesis of Lithium and Tantalum Complexes. 2011, Manuscriptin preparation. C. R. Clough, J. S. Silvia, P. M6ller, C. C. Cummins. Synthesis and characterization 3 of the trimetaphosphate molybdenum tricarbonyl anion [(P30 9)Mo(CO) 3] - as its tris(bis(triphenylphosphine)iminium) salt. 2011, Manuscript in preparation. X. Cai, S. Majumdar, L. M. Frutos, M. Temprado, C. R. Clough, A. T. Vai, C. C. Cummins, M. E. Germain, E. V. Rybak-Akimova, B. Captain, C. D. Hoff. The Mechanism

149 of Oxygen Atom Transfer Reactions Involving Mesityl Nitrile Oxide to Phosphines, Metal Phosphido Complexes, Metal Complexes and N-Heterocyclic Carbenes. 2011, Manuscriptin preparation M. Montag, C. R. Clough, P. Muller, C. C. Cummins. Cyclophosphates as ligands for cobalt(III) in water. Chem. Commun. 2011, 47, 662-664. C. R. Clough, P. MUller, C. C. Cummins. 6-Coordinate tungsten(VI) tris-n-isopropylanilide complexes: products of terminal oxo and nitrido transformations effected by main group electrophiles. Dalton Trans. 2008, 4458-4463. A. R. Fox, C. R. Clough, N. A. Piro, C. C. Cummins. A Terminal Nitride-to- Phosphide Conversion Sequence Followed by Tungsten Phosphide Functionalization Using a Diphenylphosphenium Synthon. Angew. Chem., Int. Ed. 2007, 46, 973-976. J. S. Figueroa, N. A. Piro, C. R. Clough, C. C. Cummins. A Nitridoniobium(V) Reagent That Effects Acid Chloride to Organic Nitrile Conversion: Synthesis via Heterodinuclear (Nb/Mo) Dinitrogen Cleavage, Mechanistic Insights, and Recycling. J. Am. Chem. Soc. 2006, 128, 940-950. T. Murahashi, C. R. Clough, J. S. Figueroa, C. C. Cummins. A ligand composed of dinitrogen and methyldiphenylphosphane in a cationic molybdenum complex. Angew. Chem., Int. Ed. 2005, 44, 2560-2563. J. E. McDonough, J. J. Weir, M. J. Carlson, C. D. Hoff, 0. P. Kryatova, E. V. Rybak-Akimova, C. R. Clough, C. C. Cummins. Solution Calorimetric and Stopped-Flow Kinetic Studies of the Reaction of -Cr(CO)3C5Me5 with PhSe-SePh and PhTe-TePh. Experimental and Theoretical Estimates of the Se-Se, Te-Te, H-Se, and H-Te Bond Strengths. Inorg. Chem. 2005,44, 3127-3136. C. R. Clough, J. B. Greco, J. S. Figueroa, P. L. Diaconescu, W. M. Davis, C. C. Cummins. Organic Nitriles from Acid Chlorides: An Isovalent N for O(Cl) Exchange Reaction Mediated by a Tungsten Nitride Complex. J.Am. Chem. Soc. 2004, 126, 7742-7743. B. L. Lin, C. R. Clough, G. L. Hillhouse. Interactions of Aziridines with Nickel Complexes: Oxidative-Addition and Reductive-Elimination Reactions that Break and Make C-N Bonds. J.Am. Chem. Soc. 2002, 124, 2890-2891.

Patents

C. R. Clough, C. Breen, A. Thamban, J. S. Steckel. Luminescent Nanocrystals Including a Group IA Element and a Group VA Element, Method, Composition, Device and Other Products. Int. Pat. Appl. WO 2008133660. Nov. 6, 2008.

Presentations

" Bruker/MIT Symposium, Cambridge, MA, January 2005 (Poster) " American Chemical Society 229h National Meeting, New York, NY, September 2003 (Poster)

* American Chemical Society 2 22nd National Meeting, Chicago, IL, August 2001

150 Teaching Experience

2010- 2011 Mentoring of an undergraduate researcher in the Cummins Group 2005 5.33-Advanced Chemical Experimentation and Instrumentation Laboratory Teaching Assistant 2003 5.32-Intermediate Chemical Experimentation Teaching Assistant 2002 5.310-Laboratory Chemistry Teaching Assistant 2001 & 2002 Chemistry 227-Advanced Organic/Inorganic Laboratory Teaching Assistant 2001 Chemistry 220-Organic Chemistry I Teaching Assistant 2001 Chemistry 221-Organic Chemistry II Teaching Assistant

151 152 This thesis was proudly written using IATEX on a computer box running Ubuntu Linux.

153 154 155