Ligand variations in new sulfonamide-supported group 4 ring-opening polymerization catalysts

Organometallics 2010, 29, 4171–4188 4171

DOI: 10.1021/om100508j

Ligand Variations in New Sulfonamide-Supported Group 4 Ring-Opening

Polymerization Catalysts

Andrew D. Schwarz, Katherine R. Herbert, Cristina Paniagua, and Philip Mountford*

Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Mansfield Road,

Oxford OX1 3TA, U.K.

Received May 24, 2010

The synthesis, structures, and ring-opening polymerization (ROP) capability of a wide range of

sulfonamide-supported group 4 amide, alkyl, and alkoxide complexes, varying in sulfonamide

N-substituent, metal, coordination number, and geometry, are reported. Reaction of Ti(NMe₂)₄

or Ti(NMe₂)₂(OⁱPr)₂with MeOCH₂CH₂N(CH₂CH₂NHSO₂Me)₂(12, H₂N₂^MsN^OMe) or PhCH₂N-

(CH₂CH₂NHSO₂R)₂(R = Tol (10, H₂N₂^TsN^Ph) or Me (11, H₂N₂^MsN^Ph)) afforded Ti(N₂^MsN^OMe)-

(NMe₂)₂(18), Ti(N₂^TsN^Ph)(NMe₂)₂(19), Ti(N₂^MsN^Ph)(NMe₂)₂(20), Ti(N₂^MsN^OMe)(OⁱPr)₂(21),

Ti(N₂^TsN^Ph)(OⁱPr)₂(22), Ti(N₂^MsN^Ph)(OⁱPr)₂(23), and Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24). Reaction of

N(CH₂CH₂NHSO₂R)₃(R = Tol (13, H₃N₃^TsN), Me (14, H₃N₃^MsN), or Ar^F(15, H₃N₃^ArFN, Ar^F=

3,5-C₆H₃(CF₃)₂)) with Zr(CH₂SiMe₃)₄formed Zr(N₃^RN)(CH₂SiMe₃) (R = Ts (30), Ms (31), or Ar^F

(32)). Reaction of 15 with Zr(NMe₂)₄gave Zr(N₃^ArFN)(NMe₂) (33). Complexes 19, 21, 24, 30, 32, and

33 were crystallographically characterized. Monomeric six- or five-coordinate structures were found

for the titanium complexes 19, 21, and 24, whereas the zirconium alkyls 30 and 32 were dimeric in the

solid state with terminal and bridging κ²(N,O)-bound sulfonamides. Complexes 18-24 and 30-33,

the previously reported Ti(CyN₂^R)(OⁱPr)₂(25 or 26; CyN₂^R=1,2-C₆H₁₀(NSO₂Tol)₂or 1,2-C₆H₁₀-

(NSO₂Mes)₂), and in situ generated isopropoxide initiators derived from 30-32 were investigated for

the ROP of ε-caprolactone (ε-CL). The four-coordinate 25 was the most active, forming poly(ε-CL)

with a relatively narrow PDI and well-controlled M_n. Compounds 22, 23, 25, and 26 and isoprop-

oxides generated in situ from 30-32 were all active for the ROP of rac-lactide. Of these, the initiators

based on Zr(N₃^RN)(CH₂SiMe₃) (30-32) with ⁱPrOH co-initiator gave good activities and excellent

PDIs (1.08-1.11) and agreement between measured and predicted M_n.

Introduction

extensively studied in academia and are often based on a

Lewis acidic metal, feature one or more alkoxide initiating

groups, and are supported by a polydentate ancillary ligand

(set), which influences the catalyst nuclearity and controls

the coordination-insertion chain growth mechanism. Re-

search into ROP initiators has been extensively reviewed

recently.^9-12Many metals have been studied in this context,

in particular magnesium,^13-17zinc,^13,15,17-21calcium,^16,17,22,23

aluminum,^21,24-27yttrium,^28-30the lanthanides,^31-34tin,³⁵

Over the past decade the most extensively studied bio-

compatible and biodegradeable polyesters have been those

derived from ε-caprolactone and lactide. These materials

have attracted much attention as replacements for oil-

based plastics and have found extensive applications

in the areas of medicine, packaging, film, and thermo-

plastics.^1-7Industrially these polyesters are synthesized by

ring-opening polymerization (ROP) of the cyclic esters

lactide (LA) or ε-caprolactone (CL) catalyzed by Lewis

acidic metal complexes.⁸ROP initiators have also been

(9) Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem. Rev.

2004, 104, 6147.

(10) O’Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. J. Chem. Soc.,

Dalton Trans. 2001, 2215.

(11) Platel, R. H.; Hodgson, L. M.; Williams, C. K. Polym. Rev. 2008,

48, 11.

*To whom correspondence should be addressed. E-mail: philip.

mountford@chem.ox.ac.uk.

(1) Chiellini, E.; Solaro, R. Adv. Mater. 1996, 8, 305.

(2) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff,

K. M. Chem. Rev. 1999, 99, 3181.

(3) Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000,

12, 1841.

(12) Kamber, N. E.; Jeong, W.; Waymouth, R. M.; Pratt, R. C.;

Lohmeijer, B. G. G.; Hedrick, J. L. Chem. Rev. 2007, 107, 5813.

(13) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt,

T. M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001,

123, 3229.

(4) Yoshito, I.; Hideto, T. Macromol. Rapid Commun. 2000, 21, 117.

(5) Stefan, M. Angew. Chem., Int. Ed. 2004, 43, 1078.

(6) Williams, C. K.; Hillmyer, M. A. Polym. Rev. 2008, 48, 1.

(7) Place, E. S.; George, J. H.; Williams, C. K.; Stevens, M. M. Chem.

Soc. Rev. 2009, 38, 1139.

(8) Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000,

12, 1841.

(14) Chisholm, M. H.; Eilerts, N. W. Chem. Commun. 1996, 853.

(15) Chisholm, M. H.; Huffman, J. C.; Phomphrai, K. J. Chem. Soc.,

Dalton Trans. 2001, 222.

(16) Wheaton, C. A.; Hayes, P. G.; Ireland, B. J. Dalton Trans. 2009,

4832.

(17) Poirier, V.; Roisnel, T.; Carpentier, J.-F.; Sarazin, Y. Dalton

Trans. 2009, 9820.

r

2010 American Chemical Society

Published on Web 08/31/2010

pubs.acs.org/Organometallics

4172 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

Figure 1. Group 4 ROP initiators employing polydentate N,O-donor ligands.^46,53,62-65

.

the group 4 elements,^21,36-45germanium,⁴⁶indium,⁴⁷and

iron.^48,49

Figure 2. Recently reported group 4 and aluminum ROP

initiators supported by tetradentate bis(sulfonamide)amine

ligands.^68,69

most widely exploited across the periodic table. This pro-

gress has been summarized in recent reviews.^{9-11,16,50-52}

Within group 4, Davidson, Aida, Okuda, and Harada have

used titanium complexes of bi- and tridentate bis(phenolate)

ligands for the ROP of ε-CL and ^L-LA.^53-56Davidson found

that polymerization of ε-CL with five-coordinate 1 (Figure 1)

gave polyesters with narrow PDIs (polydispersity indices,

M_w/M_n) and well-defined molecular weights, M_n.⁵³The

polymerization was much more effective than with the

analogous six-coordinate complex 2 with a tetradentate bis-

(phenolate)amine, which showed no activity under the same

Among the many supporting ligands that have been used,

polydentate alkoxides and phenolates have been among the

(18) Cheng, M.; Attygalle, A. B.; Lobkovsky, E. B.; Coates, G. W.

J. Am. Chem. Soc. 1999, 121, 11583.

(19) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.;

Pacold, M.; Phomphrai, K. J. Am. Chem. Soc. 2000, 122, 11845.

(20) Wheaton, C. A.; Ireland, B. J.; Hayes, P. G. Organometallics

2009, 28, 1282.

(21) Chisholm, M. H.; Lin, C.-C.; Gallucci, J. C.; Ko, B.-T. Dalton

Trans. 2003, 406.

(22) Zhong, Z.; Dijkstra, P. J.; Birg, C.; Westerhausen, M.; Feijen, J.

Macromolecules 2001, 34, 3863.

(23) Antelmann, B.; Chisholm, M. H.; Iyer, S. S.; Huffman, J. C.;

Navarro-Llobet, D.; Pagel, M.; Simonsick, W. J.; Zhong, W. Macro-

molecules 2001, 34, 3159.

(40) Mogstad, A.-L.; Waymouth, R. M. Macromolecules 1994, 27,

2313.

(41) Miola Delaite, C.; Hamaide, T.; Spitz, R. Macromol. Chem.

Phys. 1999, 200, 1771.

(24) Dubois, P.; Jerome, R.; Teyssie, P. Macromolecules 1991, 24,

977.

(25) Kowalski, A.; Duda, A.; Penczek, S. Macromolecules 1998, 31,

2114.

(42) Hsieh, K. C.; Lee, W. Y.; Hsueh, L. F.; Lee, H. M.; Huang, J. H.

Eur. J. Inorg. Chem. 2006, 2306.

(43) Zelikoff, A. L.; Kopilov, J.; Goldberg, I.; Coates, G. W.; Kol, M.

Chem. Commun. 2009, 6804.

(26) Aida, T.; Inoue, S. Acc. Chem. Res. 1996, 29, 39.

(27) Cameron, P. A.; Jhurry, D.; Gibson, V. C.; White, A. J. P.;

Williams, D. J.; Williams, S. Macromol. Rapid Commun. 1999, 20, 616.

(28) McLain, S. J.; Ford, T. M.; Drysdale, N. E. Polym. Prepr. (Am.

Chem. Soc., Div. Polym. Sci.) 1992, 33, 463.

(29) Chamberlain, B. M.; Sun, Y.; Hagadorn, J. R.; Hemmesch,

E. W.; Young, V. G., Jr.; Hillmyer, M. A.; Tolman, W. B. Macro-

molecules 1999, 32, 2400.

(44) Sergeeva, E.; Kopilov, J.; Goldberg, I.; Kol, M. Inorg. Chem.

2010, 49, 3977.

(45) Umare, P. S.; Tembe, G. L.; Rao, K. V.; Satpathy, U. S.; Trivedi,

B. J. Mol. Catal. A: Chem. 2007, 268, 235.

(46) Amanda, J. C.; Christopher, J. C.; Matthew, G. D.; Matthew,

D. J.; Matthew, D. L.; Steven, D. B.; Mary, F. M. Angew. Chem., Int. Ed.

2007, 46, 2280.

(47) Douglas, A F.; Patrick, B.; Mehrkhodavandi, P. Angew. Chem.,

(30) Cai, C.-X.; Amgoune, A.; Lehmann, C. W.; Carpentier, J.-F.

Int. Ed. 2008, 47, 2290.

Chem. Commun. 2004, 330.

(48) O’Keefe, B. J.; Monnier, S. M.; Hillmyer, M. A.; Tolman, W. B.

J. Am. Chem. Soc. 2001, 123, 339.

(49) O’Keefe, B. J.; Breyfogle, L. E.; Hillmyer, M. A.; Tolman, W. B.

(31) Giesbrecht, G. R.; Whitener, G. D.; Arnold, J. J. Chem. Soc.,

Dalton Trans. 2001, 923.

(32) Dyer, H. E.; Huijser, S.; Schwarz, A. D.; Wang, C.; Duchateau,

J. Am. Chem. Soc. 2002, 124, 4384.

R.; Mountford, P. Dalton Trans. 2008, 32.

(33) Bonnet, F.; Cowley, A. R.; Mountford, P. Inorg. Chem. 2005, 44,

9046.

(50) Labet, M.; Thielemans, W. Chem. Soc. Rev. 2009, 38, 3484.

(51) Wu, J.; Yu, T.-L.; Chen, C.-T.; Lin, C.-C. Coord. Chem. Rev.

2006, 250, 602.

(34) Ma, H.; Okuda, J. Macromolecules 2005, 38, 2665.

(35) Kricheldorf, H. R.; Kreiser-Saunders, I.; Boettcher, C. Polymer

1995, 36, 1253.

(36) Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules

1988, 21, 286.

(52) Amgoune, A.; Thomas, C. M.; Carpentier, J.-F. Pure Appl.

Chem. 2007, 79, 2013.

(53) Chmura, A. J.; Davidson, M. G.; Jones, M. D.; Lunn, M. D.;

Mahon, M. F. Dalton Trans. 2006, 887.

(54) Takashima, Y.; Nakayama, Y.; Hirao, T.; Yasuda, H.; Harada,

(37) Kim, Y.; Jnaneshwara, G. K.; Verkade, J. G. Inorg. Chem. 2003,

42, 1437.

(38) Thomas, D.; Arndt, P.; Peulecke, N.; Spannenberg, A.; Kempe,

R.; Rosenthal, U. Eur. J. Inorg. Chem. 1998, 1998, 1351.

(39) Dobrzynski, P.; Li, S.; Kasperczyk, J.; Bero, M.; Gasc, F.; Vert,

M. Biomacromolecules 2005, 6, 483.

A. J. Organomet. Chem. 2004, 689, 612.

(55) Takashima, Y.; Nakayama, Y.; Watanabe, K.; Itono, T.; Ueyama,

N.; Nakamura, A.; Yasuda, H.; Harada, A.; Okuda, J. Macromolecules

2002, 35, 7538.

(56) Takeuchi, D.; Nakamura, T.; Aida, T. Macromolecules 2000, 33,

725.

Article

Organometallics, Vol. 29, No. 18, 2010 4173

Figure 3. Protio versions of the ligands used in this study (Ar^F= 3,5-C₆H₃(CF₃)₂).^69-72

conditions.⁵⁷Indeed, of all those studied, the only tetra-

dentate bis(phenolate)amine complex found to be active gave

a very poorly controlled polymerization, forming poly(ε-CL)s

with very broad PDIs and poorly controlled M_nvalues.⁵⁷

Other four- or five-coordinate titanium complexes of bi- or

tridentate bis(phenolate) ligands with either CH₂or chalco-

gen bridges were studied for the ROP of ε-CL by Aida,

Okuda, and Harada.^54-56,58Good yields, narrow PDIs, and

close agreement between measured and predicted M_nwere

achieved with these initiators. Recently, Kol and Coates

described group 4 salophan complexes, which demonstrate

remarkable activity, with titanium complexes found to be

more active than zirconium. This is in contrast to all other

studies of the activities of the group 4 metals for a given

ligand.^43,57,59Arnold and Dagorne employed five-coordi-

nate titanium complexes of alkoxide or phenolate donors

containing a linked N-heterocyclic carbene. Good conver-

sions of lactide to poly(LA) with narrow PDIs are observed,

indicative of living behavior.^60,61

Very effective group 4 initiators supported by C₃-sym-

metric alkyl or aryl oxides have been developed by Verkade,

Davidson, and Kol.^{37,59,63,65,66}Verkade was the first to

employ titanatranes of various structures (e.g., 3, Figure 1)

for the ROP of rac-LA.^37,63,64,66Polymerizations conducted

under melt conditions (130 °C, 2 h) gave poly(LA) with a

range of PDIs and poorly controlled M_n. In contrast, the

same systems in toluene at 70 °C yielded an extremely well

controlled polymerization of ^L- and rac-LA for 3.^{37,62-64,66,67}

Kol subsequently compared six-coordinate titanium and

zirconium complexes of tetradentate bis(phenolate)amines

to five-coordinate ones of the type 4 (Figure 1) with C₃-

symmetric tris(phenolate)amine ligands.⁵⁹For both tita-

nium and zirconium, the five-coordinate tripodal ligand

complexes were more active for the ROP of ^L-LA. There

was also a marked dependence of performance on the

phenolate R⁰ring substituents. Davidson has also evaluated

group 4 complexes of the type 4 for the ROP of rac-LA.⁵⁷

Activity generally increased down the triad, and the titanium

i

t

complex (4: R = Pr, R⁰= Bu) gave a ROP activity and

control that were similar to those observed previously by

Verkade and Kol. In contrast, the zirconium and hafnium

analogues effected the extremely well controlled ROP of rac-

LA (as judged by narrow PDIs and predictable M_n), giving

excellent degrees of heterotactic enrichment (P_r= 0.98)

and high conversions after 30 min under melt condi-

tions. Solution polymerizations gave similar results, al-

though the conversions were low (50% after 48 h at room

temperature).⁶⁵

We very recently reported a series of new aluminum, titanium,

and zirconium isopropoxide complexes (L)M(OⁱPr)_n(n = 1 or

2, M = Al, Ti, Zr) supported by dianionic bis(sulfon-

amide)amine ligands, L (Figure 2 shows selected examp-

les).^68,69These thermally robust complexes allowed for the

well-controlled ROP of ε-CL and rac-LA by a coordination-

insertion mechanism initiated by the respective M-OⁱPr

groups. These were the first detailed studies of any sulfon-

amide-supported ROP initiators. The rates of ROP were

significantly faster for the zirconium complexes 6 and 8 than

for their titanium congeners (5 and 7). For 6 and 8, both the

ROP activity and control were comparable to zirconium

complexes of the extensively exploited tetradentate bis-

(phenolate)amine ligands. Similar results were found for

the aluminum-based initiators, leading us to propose that,

in these regards at least, the bis(sulfonamide)amine ligands

are able to act as “phenolate mimics”.

In these earlier studies for group 4 we focused only

on tetradentate bis(sulfonamide)amine ligands with tosyl

N-substituents (i.e., -NSO₂Tol donors). Encouraged by

the performance of complexes 5-8 and given the current

intense interest in the ROP of cyclic esters, we decided to

further develop this new class of initiator using several

different approaches. In light of the success of bi- and

tridentate phenolate ligands (vide supra), it was of interest

to target four- and five-coordinate bis(sulfonamide) initia-

tors since they might give more accessible metal centers and

faster ROP. Five-coordinate complexes bearing C₃-sym-

metric tris(sulfonamide)amine ligands (analogous to the

tris(phenolate)amines) were also clearly of interest. We also

reasoned that introducing sulfonamide ligands with steri-

cally less encumbering groups such as mesyl (-SO₂Me)

instead of tosyl (-SO₂Tol) could improve the activity of

the resulting initiators. Therefore, the synthesis of tri- and

tetradentate sulfonamide complexes bearing -NSO₂Me do-

nors was also undertaken. For the new tetradentate ligands,

methoxy pendant donors were targeted (cf. 5 and 6) rather

(57) Chmura, A. J.; Davidson, M. G.; Jones, M. D.; Lunn, M. D.;

Mahon, M. F.; Johnson, A. F.; Khunkamchoo, P.; Roberts, S. L.;

Wong, S. S. F. Macromolecules 2006, 39, 7250.

(58) Nakayama, Y.; Watanabe, K.; Ueyama, N.; Nakamura, A.;

Harada, A.; Okuda, J. Organometallics 2000, 19, 2498.

(59) Gendler, S.; Segal, S.; Goldberg, I.; Goldschmidt, Z.; Kol, M.

Inorg. Chem. 2006, 45, 4783.

(60) Romain, C.; Brelot, L.; Bellemin-Laponnaz, S. p.; Dagorne, S.

Organometallics 2010, 1191.

(61) Patel, D.; Liddle, S. T.; Mungur, S. A.; Rodden, M.; Blake, A. J.;

Arnold, P. L. Chem. Commun. 2006, 1124.

(62) Kim, Y.; Kapoor, P. N.; Verkade, J. G. Inorg. Chem. 2002, 41,

4834.

(63) Kim, Y.; Verkade, J. G. Organometallics 2002, 21, 2395.

(64) Kim, Y.; Verkade, J. G. Macromol. Symp. 2005, 224, 105.

(65) Chmura, A. J.; Davidson, M. G.; Frankis, C. J.; Jones, M. D.;

Lunn, M. D. Chem. Commun. 2008, 1293.

(66) Kim, Y.; Verkade, J. G. Macromol. Rapid Commun. 2002, 23,

917.

(67) Kim, Y.; Verkade, J. G. Inorg. Chem. 2003, 42, 4804.

4174 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

than pyridyl ones (cf. 7 and 8) because it had been found

previously that this gave more active initiators.⁶⁸

molecular structures, and selected distances and angles are

given in the Supporting Information.

Results and Discussion

Synthesis of Protio-ligand Precursors. Figure 3 shows the

protio versions of the ligands used in this study, along with

their general abbreviations. H₂N₂^TsN^Ph(10),⁶⁹H₂N₂^MsN^Ph

SO₂Mes

(11),⁶⁹H₃N₃^TsN (13),⁷⁰H₂CyN₂^Ts(16),⁷¹and H₂CyN₂

(17)⁷²were prepared according to the literature methods.

The new tetradentate protio-ligand H₂N₂^MsN^OMe(12) with

-NSO₂Me donors was synthesized in good yield by reaction

of MeOCH₂CH₂NH₂with N-mesyl aziridine (eq 1). The

other new protio-ligands, N(CH₂CH₂NHSO₂Me)₃(14,

H₃N₃^MsN) and N(CH₂CH₂NHSO₂Ar^F)₃(15, H₃N₃^ArFN),

were prepared by reaction of tris(2-aminoethyl)amine with

SO₂ClR (R=Me or Ar^F) in a biphasic solution of diethyl

ether and water (eq 2). For 15 this proceeded satisfactorily in

high yield. However, the reaction with SO₂ClMe gave only

poor yields (ca. 20%) of 14 due to its high solubility in water,

even at high pH. So as to make comparisons with their metal

complexes (vide infra) and the previously reported 16,^73,74

Titanium Complexes of Tri- and Tetradentate Bis(sulfon-

amide)amine Ligands. Protonolysis reactions of Ti(NMe₂)₄

with H₂N₂^MsN^OMe(12) H₂N₂^TsN^Ph(10), or H₂N₂^MsN^Ph

(11) gave the corresponding six- and five-coordinate bis-

(dimethylamide) complexes Ti(N₂^MsN^OMe)(NMe₂)₂(18), Ti-

(N₂^TsN^Ph)(NMe₂)₂(19), and Ti(N₂^MsN^Ph)(NMe₂)₂(20), res-

pectively, in moderate to good yield after recrystallization

(Scheme 1). When the reactions were followed in CD₂Cl₂on

the NMR tube scale, the conversions were quantitative and

the expected HNMe₂side product was observed. The lower

isolated yields of the mesyl sulfonamide complexes (45% and

39% for 18 and 20, respectively) compared to the tosyl one

(60%, 19) are a consequence of their high solubility.

H₂N₂^TsN^{OMe 68}and H₂N₂^TsN^py,⁷⁵the X-ray structures of 14

,

and 17 were determined. Further details, a discussion of the

Scheme 1. Synthesis of the New Five- and Six-Coordinate

Bis(sulfonamide)amine Complexes

Metal alkoxide complexes (L)M-OR are often superior

initiators compared to their amide counterparts (L)M-

NR₂.^11,54,55,68In our previous work,⁶⁸we also found that

Ti(N₂^TsN^X)(OⁱPr)₂(X = OMe (5) or py (7)) were more eff-

ective in ROP than the corresponding amides, Ti(N₂^TsN^X)-

(NMe₂)₂. To explore further these structure-activity rela-

tionships, we therefore prepared the bis(isopropoxide)

compounds Ti(N₂^MsN^OMe)(OⁱPr)₂(21), Ti(N₂^TsN^Ph)(OⁱPr)₂

(22), and Ti(N₂^MsN^Ph)(OⁱPr)₂(23) from Ti(NMe₂)₂(OⁱPr)₂

and the respective protio-ligand (Scheme 1). NMR tube scale

investigations showed quantitative formation of 21, 22, or 23

and no evidence for the bis(dimethylamide) complexes 18,

19, and 20. On the preparative scale, 21-23 were obtained in

44-56% yield after 2 h reaction time. Regrettably, attempts

to prepare zirconium analogues of 21-23 from Zr(OⁱPr)₄ⁱPrOH

3

Scheme 2. Syntheses of Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24)

Article

Organometallics, Vol. 29, No. 18, 2010 4175

using analogous methods to those successfully applied for 6

and 8 (Figure 2) gave ill-defined mixtures. The correspond-

ing reactions with Zr(NMe₂)₄were also unsuccessful. There-

fore further work with the N₂^MsN^OMeand N₂^RN^Phligands

focused exclusively on titanium systems.

As discussed later, both the bis(dimethylamide) com-

pound 19 and its bis(isopropoxide) analogue 22 were good

initiators for the ROP of ε-CL, but gave poly(ε-CL)s of quite

different molecular weights. To explore these differences

further, we prepared the mixed amide-alkoxide complex Ti-

(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24). As shown in Scheme 2, NMR

tube scale experiments in C₆D₆found that 24 could be

quantitatively formed via a redistribution reaction between

19 and 22. However, this reaction is rather slow, requiring

heating for 4 days at 80 °C in benzene solution to achieve

completion. A more convenient route was the reaction of

H₂N₂^TsN^Ph(10) with 3 equiv of Ti(NMe₂)₂(OⁱPr)₂, which

formed 24 after 16 h at room temperature in CH₂Cl₂. Sepa-

ration of 24 from excess Ti(NMe₂)₂(OⁱPr)₂and the Ti-

(NMe₂)(OⁱPr)₃side product was straightforward, giving 24

in 74% yield.

Figure 4. Displacement ellipsoid plot of Ti(N₂^MsN^OMe)(OⁱPr)₂

(21). Ellipsoids are drawn at the 20% probability level, and H

atoms are omitted for clarity.

NMR spectra of 19 and 20 show equivalent environments

for the two pairs of ligand CH₂CH₂protons along with a

single resonance for the axial and equatorial NMe₂groups.

Cooling to 248 or 213 K, respectively, gives four CH₂CH₂

proton environments and separate signals for the axial

and equatorial NMe₂ligands. These fluxional processes are

likely to be of the type illustrated in eq 3 for 19 and 20, as

discussed elsewhere for diamide-amine-supported titanium

complexes.⁷⁷

Finally, so as to compare the ROP performance of four-,

five-, and six-coordinate titanium bis(sulfonamide) initia-

tors, the previously reported complexes Ti(CyN₂^Ts)(OⁱPr)₂

(25) and Ti(CyN₂^{SO Mes})(OⁱPr)₂(26) were also synthesized

2

(CyN₂^R= 1,2-C₆H₁₀(NSO₂Tol)₂or 1,2-C₆H₁₀(NSO₂Mes)₂).

These contain a (rac)-trans-cyclohexyl diamine backbone

and were prepared according to literature methods.⁷⁶

The solid-state structures of 19, 21, and 24 have been

determined (see below) and confirm those shown in

Schemes 1 and 2. In solution, all seven compounds 18-24

show dynamic behavior. For 18-23 this involves exchange

of the equatorial and axial Ti-X groups (X = NMe₂(18, 19,

and 20) or OⁱPr (21, 22, and 23)) and in all cases pairwise

exchange of the two sets of R⁰CH₂N(CH₂CH₂NSO₂R)₂

ligand methylene protons (R = Ms or Ts; R⁰= Ph or

CH₂OMe). For example, at 90 °C in toluene-d₈the ¹H

The solid-state structure of Ti(N₂^MsN^OMe)(OⁱPr)₂(21) is

shown in Figure 4, and selected distances and angles are

listed in Table 1. Compound 21 has a pseudo-octahedral

titanium and overall C_smolecular symmetry. Its geometry is

analogous to those of the tosyl analogues Ti(N₂^TsN^OMe)-

(OⁱPr)₂(5) and Zr(N₂^TsN^OMe)(OⁱPr)₂(6),⁶⁸and the principal

metric parameters in 21 are similar to those of 5. For

(68) Schwarz, A. D.; Thompson, A. L.; Mountford, P. Inorg. Chem.

2009, 48, 10442.

(69) Schwarz, A. D.; Chu, Z.; Mountford, P. Organometallics 2010,

29, 1246.

(70) Motekaitis, R. J.; Martell, A. E.; Muraset, I. Inorg. Chem. 1986,

938.

(71) Kitagawa, O.; Yotsumoto, K.; Kohriyama, M.; Dobashi, Y.;

Taguchi, T. Org. Lett. 2004, 6, 3605.

(72) Takahashi, H.; Kawakita, T.; Ohno, M.; Yoshioka, M.;

Kobayashi, S. Tetrahedron 1992, 48, 5691.

(73) Nieger, M.; Josten, W.; Vogtle, F. Personal communication to the

University of Edinburgh (cited in the Cambridge Structural Database), 2004.

(74) Pritchett, S.; Gantzel, P.; Walsh, P. J. Organometallics 1999, 18,

823.

(75) Skinner, M. E. G.; Li, Y.; Mountford, P. Inorg. Chem. 2002, 41,

1110.

(76) Pritchett, S.; Woodmansee, D. H.; Gantzel, P.; Walsh, P. J.

˚

example, the Ti-N_{SO R}(av 2.099(4) in 21 vs 2.089(2) A in

2

i

˚

5) and Ti-O Pr (av 1.770(3) in 21 vs 1.773(2) A in 5) distances

are equivalent within error, while the distances to the dative

donors, namely, Ti-OMe and Ti-N_aminein 21 (2.271(4) and

2.248(4) A) are somewhat shorter than in 5 (2.327(2) and

˚

2.277(2) A).

The molecular structures of Ti(N₂^TsN^Ph)(NMe₂)₂(19) and

Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24) are shown in Figure 5, and

selected distances and angles are listed in Tables 2 and 3. The

geometry around Ti(1) in 19 is best described as distorted

(77) Clark, H. C. S.; Cloke, F. G. N.; Hitchcock, P. B.; Love, J. B.;

Wainwright, A. P. J. Organomet. Chem. 1995, 501, 333.

J. Am. Chem. Soc. 1998, 120, 6423.

4176 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

˚

Table 1. Selected Bond Distances (A) and Angles (deg) for

Table 2. Selected Bond Distances (A) and Angles (deg) for

Ti(N₂^TsN^Ph)(NMe₂)₂(19)

Ti(N₂^MsN^OMe)(OⁱPr)₂(21)

Ti(1)-N(1)

2.103(5)

2.096(5)

1.762(4)

75.4(2)

76.2(2)

94.5(2)

102.0(2)

103.6(2)

84.9(2)

82.4(2)

88.7(2)

Ti(1)-N(2)

2.248(4)

1.778(4)

2.271(4)

151.1(2)

94.2(2)

Ti(1)-N(1)

2.278(2)

2.041(2)

1.878(2)

73.4(1)

131.0(1)

93.1(1)

99.4(1)

Ti(1)-N(2)

Ti(1)-N(4)

2.081(2)

1.889(2)

Ti(1)-N(3)

Ti(1)-O(1)

Ti(1)-N(3)

Ti(1)-O(2)

Ti(1)-O(3)

Ti(1)-N(5)

N(1)-Ti(1)-N(2)

N(2)-Ti(1)-N(3)

N(2)-Ti(1)-O(1)

N(1)-Ti(1)-O(2)

N(3)-Ti(1)-O(2)

N(1)-Ti(1)-O(3)

N(3)-Ti(1)-O(3)

O(2)-Ti(1)-O(3)

N(1)-Ti(1)-N(3)

N(1)-Ti(1)-O(1)

N(3)-Ti(1)-O(1)

N(2)-Ti(1)-O(2)

O(1)-Ti(1)-O(2)

N(2)-Ti(1)-O(3)

O(1)-Ti(1)-O(3)

N(1)-Ti(1)-N(2)

N(2)-Ti(1)-N(3)

N(2)-Ti(1)-N(4)

N(1)-Ti(1)-N(5)

N(3)-Ti(1)-N(5)

N(1)-Ti(1)-N(3)

N(1)-Ti(1)-N(4)

N(3)-Ti(1)-N(4)

N(2)-Ti(1)-N(5)

N(4)-Ti(1)-N(5)

76.4(1)

156.4(1)

100.0(1)

115.2(1)

103.9(1)

93.2(2)

163.0(2)

102.6(2)

74.3(2)

107.1(1)

168.6(2)

˚

Table 3. Selected Bond Distances (A) and Angles (deg) for

Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24)

Ti(1)-N(1)

2.267(2)

2.073(2)

1.766(2)

74.1(1)

144.9(1)

98.57(8)

114.6(1)

102.8(1)

Ti(1)-N(2)

Ti(1)-N(4)

2.086(2)

1.855(2)

Ti(1)-N(3)

Ti(1)-O(5)

N(1)-Ti(1)-N(2)

N(2)-Ti(1)-N(3)

N(2)-Ti(1)-N(4)

N(1)-Ti(1)-O(5)

N(3)-Ti(1)-O(5)

N(1)-Ti(1)-N(3)

N(1)-Ti(1)-N(4)

N(3)-Ti(1)-N(4)

N(2)-Ti(1)-O(5)

N(4)-Ti(1)-O(5)

73.9(1)

139.5 (1)

96.37(9)

103.4(1)

105.8(1)

to N₂^TsN^Ph(but with O in place of NCH₂Ph).⁷⁸In this

compound the NMe₂groups occupy the equatorial sites of

a distorted TBP (N_Ts-Ti-N_Ts=143.5(1)^o), in contrast to 19,

in which one NMe₂is equatorial and one is axial.

The Ti-N_Tsdistances in 19 and 24 (av 2.070, range

˚

2.041(2)-2.086(2) A) are shorter than in their six-coordinate

counterparts Ti(N₂^TsN^X)(NMe₂)₂(X = py (28) or OMe (29),

˚

av Ti-N_Ts= 2.105 A), as expected, and comparable to those

˚

in 27. Likewise, the average Ti-NMe₂distance of 1.884(2) A

in the bis(dimethylamide) 19 is shorter than those in Ti-

Ts

X

68

˚

(N₂N )(NMe₂)₂(av 1.927, range 1.898(6)-1.981(7) A).

It is interesting to note the decrease in Ti-NMe₂distance

˚

from 1.884(2) A (av) in 19 to 1.855(2) A in 24, which has only

one NMe₂ligand. This shortening is attributed to improved

σ and π (2p_π-3d_π) bonding between titanium and the re-

maining NMe₂ligand in 24 since alkoxides are not as good

donors as amides in both of these regards.^79-82The implica-

tions of these structural changes for the ROP characteristics

of 19, 24, and their homologues are discussed later.

Figure 5. Displacement ellipsoid plots of Ti(N₂^TsN^Ph)(NMe₂)₂

(19, top) and Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24, bottom). Ellip-

soids are drawn at the 20% probability level, and H atoms are

omitted for clarity.

trigonal bipyramidal with N(1) and N(4) occupying the

apical positions. The N(1)-Ti(1)-N(4) angle of 156.4(1)^o

is somewhat less than that for an ideal trigonal bipyramid

(TBP), while N(2)-Ti(1)-N(3) (131.0(1)^o) is greater than

the expected 120^obetween equatorial groups. These devia-

tions from an ideal TBP increase in 24 with N(1)-Ti(1)-N-

(4) decreasing to 139.5(1)^oand N(2)-Ti(1)-N(3) increasing

to 144.9(1)^o. Indeed, 24 could also be described as a square-

based pyramid (SBP) with O(1) occupying the axial site since

the four angles O(1)-Ti(1)-N(x) (x = 1, 2, 3, 4) lie in the

narrow range 102.8(1)-114.6(1)^o. We note that Nagashima

recently described Ti{O(CH₂CH₂NTs)₂}(NMe₂)₂(27), con-

taining a tridentate bis(sulfonamide)ether ligand analogous

Zirconium Complexes of C₃-Symmetric Tris(sulfonamide)-

amine Ligands. We turn now to the synthesis of complexes

with tetradentate trianionic sulfonamide-amine ligands. Cer-

tain titanium and aluminum complexes of this class of ligand

have been reported previously, but not with the ligands used

herein and not in the context of polymerization catalysis.⁸³

Tris(amido)amine (“tren”) ligands in general have been

widely used in transition metal chemistry, but again not in

(79) Gade, L. H.; Mountford, P. Coord. Chem. Rev. 2001, 216-217,

65.

(80) Gade, L. H. Chem. Commun. 2000, 173.

(81) Kempe, R. Angew. Chem., Int. Ed. 2000, 39, 468.

(82) Mountford, P.; Ward, B. D. Chem. Commun. 2003, 1797.

(83) Massimiliano, F.; Fredrik, L.; Myriam Mba, B.; Patrick, R.;

Marco, C.; Lutz, H. G.; Giulia, L.; Christina, M. Eur. J. Inorg. Chem.

2006, 1032.

(78) Hamura, S.; Oda, T.; Shimizu, Y.; Matsubara, K.; Nagashima,

H. J. Chem. Soc., Dalton Trans. 2002, 1521.

Article

Organometallics, Vol. 29, No. 18, 2010 4177

Scheme 3. Synthesis of the New Tris(sulfonamide)amine Complexes 30-33

the context of ROP catalysis.^80,81,84,85We initially attempted

the direct synthesis of alkoxide complexes from the protio-

ligands H₃N^TsN (13), H₃N^MsN (14), and H₃N₃^ArFN (15) and

shown in eq 4. Further details are given in the Supporting

Information.

either Ti(OⁱPr)₄or Zr(OⁱPr)₄ⁱPrOH by analogy with the suc-

3

cessful synthesis of M(N₂^TsN^OMe)(OⁱPr)₂and M(N₂^TsN^py)-

(OⁱPr)₂(5-8, Figure 2).⁶⁸Unfortunately, no reaction was

observed between these metal isopropoxides and 13-15,

even under forcing solution conditions or in the melt. Reac-

tions were therefore carried out with Ti(NMe₂)₄, Zr(NMe₂)₄,

and Zr(CH₂SiMe₃)₄in the hope that the so-formed metal-

amide or -alkyl products could in turn be converted to the

more desirable metal-isopropoxides. Reaction of 13-15 with

Ti(NMe₂)₄(by analogy with the successful syntheses of

18-20 and certain literature precedents⁸³) led to complicated

mixtures from which no single product could be isolated.

However, the reactions with Zr(CH₂SiMe₃)₄and Zr(NMe₂)₄

were successful, as shown in Scheme 3.

Reaction of Zr(CH₂SiMe₃)₄with 1 equiv of 13, 14, or

15 gave the corresponding alkyl complexes Zr(N₃^TsN)-

(CH₂SiMe₃) (30), Zr(N₃^MsN)(CH₂SiMe₃) (31), and Zr-

(N₃^ArFN)(CH₂SiMe₃) (32), respectively, in 60-89% yield

after recrystallization. When followed in C₆D₆on the NMR

tube scale, the yields were effectively quantitative and the

expected SiMe₄side product was observed. These alkyl

compounds were readily handled using normal Schlenk line

and drybox techniques. Reaction of Zr(NMe₂)₄with 1 equiv

of 13-15 in CD₂Cl₂likewise showed quantitative conver-

sions and the expected HNMe₂side product. However, scale

up of these reactions met with considerable difficulties

associated with simultaneous formation of hydrolysis pro-

ducts, apparently arising from adventitious water. Only in

one instance, the reaction of Zr(NMe₂)₄with the electron-

deficient H₃N₃^ArFN (15), could a pure product be obtained

on scale up. Zr(N₃^ArFN)(NMe₂) (33) was obtained in 82%

yield after careful recrystallization from rigorously dried

THF and crystallographically authenticated (vide infra).

The hydrolysis product of 33 was identified as the μ-oxo

dimer {Zr(N₃^ArFN)}₂(μ-O) (34). This could be prepared by

repeated slurrying of pure 33 in “wet” Et₂O (eq 4). The solid-

state structure of 34 has been determined and confirms that

The solid-state structures of the alkyl complexes 30 and 32

as well as the amide complex 33 have been determined. The

molecular structures of 30 and 33 are given in Figures 6 and 7,

and selected distances and angles are given in Tables 4 and 5.

The structure of 32 is analogous to that of 30. A displacement

elipsoid and selected metric parameters are given in the

Supporting Information.

Compound 30 (and its analogue 32) exists as a centrosym-

metric, eight-coordinate dimer in the solid state. In addition

to the four Zr-N bonds, two -NSO₂Ts groups (N(2), N(3))

of the N₃^TsN ligand also act as κ²(N,O) donors to one

zirconium, and for one of these there is an additional

bridging SdO Zr interaction to the other metal center

3 3 3

(O(2), O(2A)). The eighth coordination site is occupied by a

terminal CH₂SiMe₃ligand with Zr-CH₂and Zr-CH₂-

SiMe₃parameters within the expected ranges.⁸⁶The geo-

metry around the zirconium centers (see Figure 6 (bottom)

(86) Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem. Inf.

Comput. Sci. 1996, 36, 746 (The UK Chemical Database Service: CSD

version 5.31 updated Feb 2010).

(84) Schrock, R. R. Acc. Chem. Res. 1997, 30, 9.

(85) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955.

4178 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

Figure 6. Displacement ellipsoid plots (25% probability) of

Zr(N₃^TsN)(CH₂SiMe₃) (30). H atoms are omitted for clarity.

Top: full structure. Bottom: details of the metal coordination

environment.

Figure 7. Displacement ellipsoid plots (25% probability) of Zr-

(N₃^ArF)(NMe₂) (33). H atoms omitted for clarity. Top: full struc-

ture. Bottom: details of the metal coordination environment.

˚

Table 4. Selected Bond Distances (A) and Angles (deg) for

Zr(N₃^TsN)(CH₂SiMe₃) (30)

for a clearer view) is intermediate between a distorted square

antiprism (the two planes being {N(1)-N(3), O(3)} and {C-

(28), N(4), O(1), O(2A)}) and a distorted, face-capped penta-

gonal bipyramid. In this latter description, C(28) and N(2)

occupy the axial sites, the equatorial donors are O(1)-O(3),

N(3), and N(4), and the capping atom is the tertiary amine

Zr(1)-C(28)

2.251(2)

2.232(2)

2.238(2)

2.292(2)

1.452(2)

1.486(2)

1.440(2)

107.62(9)

119.9(1)

127.8(1)

129.8(1)

112.1(1)

Zr(1)-N(1)

2.670(2)

2.201(2)

2.672(2)

2.272(2)

1.468(2)

1.433(2)

1.446(2)

130.6(1)

100.6(9)

121.9(2)

113.9(1)

134.4(1)

Zr(1)-N(2)

Zr(1)-N(3)

Zr(1)-N(4)

Zr(1)-O(1)

Zr(1)-O(2A)

Zr(1)-O(3)

S(1)-O(1)

S(1)-O(2)

S(2)-O(3)

S(2)-O(4)

nitrogen N(1) with a relatively long Zr(1) N(1) bond of

3 3 3

S(3)-O(5)

S(3)-O(6)

2.670(2) A. As mentioned, Zr(N₃^ArFN)(CH₂SiMe₃) (32) is

˚

Zr(1)-N(2)-S(1)

S(1)-N(2)-C(2)

Zr(1)-N(3)-C(11)

Zr(1)-N(4)-S(3)

S(3)-N(4)-C(20)

Zr(1)-N(2)-C(2)

Zr(1)-N(3)-S(2)

S(2)-N(3)-C(11)

Zr(1)-N(4)-C(20)

Zr(1)-C(28)-Si(1)

also eight-coordinate and dimeric in the solid state (see the

Supporting Information) with an analogous geometry to

that of 30. The only structural difference is that while 30

features κ²(N,O) coordination of one terminal and one

bridging -NSO₂Ts group, in 32 κ²(N,O) coordination is

found for the two nonbridging -NSO₂Ar^Fgroups and not

for the bridging one. The Zr-NSO₂Ts distances in 30 are

˚

Table 5. Selected Bond Distances (A) and Angles (deg) for

Zr(N₃^ArF)(NMe₂) (33)

Zr(1)-N(1)

2.231(5)

2.198(4)

2.038(5)

2.322(4)

1.467(4)

1.481(4)

1.482(4)

101.9(2)

119.3(4)

124.6(4)

102.4(2)

124.8(4)

119.2(4)

Zr(1)-N(2)

2.280(5)

2.725(4)

2.471(4)

2.365(4)

1.441(4)

1.437(4)

1.430(4)

127.3(4)

96.3(2)

117.8(4)

132.7(4)

131.3(4)

109.5(5)

˚

marginally shorter (av 2.223 A, range 2.201(2)-2.238(2) A)

Zr(1)-N(3)

Zr(1)-N(4)

than the Zr-NSO₂Ar^Fdistances in 32 (av 2.245 A, range

˚

Zr(1)-N(5)

Zr(1)-O(1)

˚

2.217(4)-2.272(4) A).

Zr(1)-O(3)

Zr(1)-O(5)

S(1)-O(1)

S(1)-O(2)

In contrast, the dimethylamido complex 33 is monomeric

in the solid state but still possesses an eight-coordinate

zirconium atom. In addition to the NMe₂group, the N₃^ArFN

ligand occupies seven coordination sites, with each SO₂Ar^F

donor being κ²(N,O) coordinated. The coordination geo-

metry is best described as a distorted pentagonal bipyramid

(cf. Figure 7, bottom), with N(5) and N(3) being in the axial

sites and the capping interaction again being to the tertiary

S(2)-O(3)

S(2)-O(4)

S(3)-O(5)

S(3)-O(6)

Zr(1)-N(1)-S(1)

S(1)-N(1)-C(3)

Zr(1)-N(2)-C(5)

Zr(1)-N(3)-S(3)

S(3)-N(3)-C(7)

Zr(1)-N(5)-C(2)

Zr(1)-N(1)-C(3)

Zr(1)-N(2)-S(2)

S(2)-N(2)-C(5)

Zr(1)-N(3)-C(7)

Zr(1)-N(5)-C(1)

C(1)-N(5)-C(2)

amine donor of N₃^ArFN ligand (Zr(1) (N4) = 2.725(4) A).

˚

angles subtended at N(5) = 360(1)^o) as expected due to 2p_π-4d_π

bonding interactions,^79-82and the Zr-NMe₂distance of

3 3 3

The N atom of the NMe₂ligand is sp²hybridized (sum of

Article

Organometallics, Vol. 29, No. 18, 2010 4179

86

˚

2.038(5) A is within the usual general ranges. A more

detailed comparison of 33 can be made with Scott’s structu-

rally authenticated tris(amido)amine complexes Zr{N(CH₂-

sum of angles at NSO₂R = 348^o) than in the other com-

pounds (range of averages 354-360^o). In particular N(1) and

N(2) (sum of angles 348(1)° and 336(1)^o), which lie in the

pentagonal-bipyramid equatorial plane of 33, are parti-

cularly distorted. In the μ-oxo hydrolysis product, 34, the sums

of the angles at NSO₂Ar^F(351(1)-360(1)^o) again lie in the

normal range. We propose that the relief of this distortion on

going from 33 to 34 might provide a thermodynamic driving

force for hydrolysis and that the distortion is partially driven

by effective 2p_πf4d_ππ-donation from the NMe₂ligand into

the remaining 4d_πacceptor orbital of the eight-coordinate

metal center. The NSO₂Ar^Fdistortions from planarity

would relieve unfavorable filled-orbital-filled-orbital inter-

actions between the strongly π-donating NMe₂lone pair and

(more weakly) π-donating sulfonamide nitrogens. Such π-

donating ligand conflicts are well recognized in early metal

chemistry.^80,81

t

CH₂NR)₃}(NMe₂) (R = SiMe₂Bu or 2,4,6-C₆H₂Me₃), hav-

ing more “innocent” (or conventional) N-substituents than

the sulfonyl groups in N₃^SO2RN.^87,88In these compounds the

five-coordinate Zr atoms have trigonal-bipyramidal geo-

metries with average Zr-NR distances to the tren amide

˚

donors of ca. 2.116 A in each case. These are considerably

shorter than in 33 (av Zr-N_ArF= 2.236(3), range 2.198(4)-

˚

2.280(5) A). This reflects both the electron-withdrawing

nature of the sulfonyl groups (as noted elsewhere⁶⁸) and

the higher coordination number in the case of 33. Despite this

higher coordination number, the Zr-NMe₂distance (2.038(5)

˚

A) is marginally shorter that in Zr{N(CH₂CH₂NR)₃}(NMe₂)

˚

(2.063(2) and 2.044(3) A).

The κ²(N,O) coordination of the sulfonamide groups in 30

and 32-34 is well precedented in the literature for transition

and main group metals in general.^74,76,89-91With regard to

zirconium, three related sulfonamide complexes have been

structurally authenticated previously. These are Zr(CyN₂^Ts)-

(NMe₂)₂(NHMe₂) (the analogue of 25 and 26 possessing two

κ²(N,O)-bound -NTs donors and a trigonal-bipyramidal

Zr),⁸⁹Zr(O₂N^TsN^py)(NMe₂)₂(one κ¹(N,O) and one κ²(N,O)

-NTs), and Zr(O₂N^TsN^OMe)(NMe₂)₂(twoκ¹(N,O) -NTs)).⁶⁸

The Zr-NSO₂R distances in 30 and 32-34 are compar-

able to those in these previous examples. As noted pre-

viously, the SdO distances of the bridging sulfonyl oxygens

in 30 and 32-34 are significantly longer than for the terminal

ones.

Although the alkyls Zr(N₃^RN)(CH₂SiMe₃) (30-32) have

complex structures in the solid state, the solution NMR

spectra in the range 20-40 °C are relatively simple, indicative

of monomeric, C₃-symmetric species in solution (or dimeric

structures in the fast exchange regime). Cooling these solu-

tions leads to more complex spectra. The NMR spectra of 33

and 34 are consistent with the solid-state structures, indicat-

ing C₃and C₁molecular symmetry, respectively.

As mentioned, the N atom of the NMe₂ligand in 33 is

trigonal planar (sum of the angles subtended at N(5) 360-

(1)^o), which is the usual case for early transition metal amide

ligands.⁸⁰In contrast, the angles subtended at the sulfon-

amide donor nitrogens in the four complexes 30 and 32-34

are much more varied, spanning the range 336(1)-360(1)^o.

Such behavior has been seen elsewhere in the literature (e.g.,

for Zr(CyN₂^Ts)(NMe₂)₂(NHMe₂) the corresponding sums of

angles are 340° and 357^o) and may reflect the diminished

2p_π-4d_πbonding in sulfonamides compared to conven-

tional amide ligands. In general, across the four structures,

there is no overall correlation between Zr-NSO₂R distance

and degree of pyramidalization of the respective nitrogen.

At first sight, it is not clear why Zr(N₃^ArFN)(NMe₂) (33) is

so moisture sensitive compared to the alkyls Zr(N₃^RN)-

(CH₂SiMe₃) (30-32), readily eliminating HNMe₂. For ex-

ample, 33, 30, and 32 all contain coordinatively saturated,

eight-coordinate metals centers and, as mentioned, the

Zr-NMe₂bond length in 33 is normal. One reason might

be connected with the aforementioned distortions of the

sulfonamide donor nitrogens from planarity. Although the

average Zr-NSO₂R distances in 30 and 32-34 are very

As mentioned, metal-alkoxide compounds are superior

initiators compared to their amide and alkyl counterparts

due to more favorably balanced rates of initiation and pro-

pagation for the former. Since zirconium alkoxides of the

type Zr(N₃^RN)(OⁱPr) were not accessible from the protio-

ligands H₃N₃^RN and Zr(OⁱPr)₄ⁱPrOH (vide supra), we

3

attempted to prepare them by protonolysis starting from

the alkyl or amide compounds 30-33. As shown in eq 5 for

Zr(N₃^TsN)(CH₂SiMe₃) (30), reaction with PrOH on the

i

NMR scale in C₆D₆quantitatively yielded the corresponding

alkoxide species Zr(N₃^TsN)(OⁱPr) (35). Likewise, reaction of

one equivalent of ArOH with Zr(N₃^ArFN)(NMe₂) (33) on the

NMR scale in C₆D₆gave the aryloxide species Zr(N₃^ArFN)-

(OAr) (Ar = 2,6-C₆H₃Me₂) (36). However, once again with

these π-donor alkoxide/aryloxide ligands the compounds

were difficult to obtain pure on the preparative scale. For

the purposes of the ROP studies discussed below we there-

fore generated the isopropoxides Zr(N₃^RN)(OⁱPr) in situ

from the well-defined alkyl compounds. In situ generation

of alkoxide initiators from stable alkyl or related intermedi-

ates has been widely employed in the literature.^17,22,92-98

˚

similar (range of averages 2.224-2.241 A) and comparable

˚

to those in 6 and 8 (av 2.238 A), the sulfonamide nitrogens in

33 appear to be somewhat more distorted from planarity (av

(87) Morton, C.; Munslow, I. J.; Sanders, C. J.; Alcock, N. W.; Scott,

P. Organometallics 1999, 18, 4608.

(88) Morton, C.; Gillespie, K. M.; Sanders, C. J.; Scott, P.

(92) Pietrangelo, A.; Hillmyer, M. A.; Tolman, W. B. Chem. Com-

mun. 2009, 2736.

(93) Du, H.; Velders, A. H.; Dijkstra, P. J.; Zhong, Z.; Chen, X.;

J. Organomet. Chem. 2000, 606, 141.

(89) Ackermann, L.; Bergman, R. G.; Loy, R. N. J. Am. Chem. Soc.

2003, 125, 11956.

Feijen, J. Macromolecules 2009, 42, 1058.

(94) Nomura, N.; Ishii, R.; Yamamoto, Y.; Kondo, T. Chem.—Eur.

J. 2007, 13, 4433.

(90) Nanthakumar, A.; Miura, J.; Diltz, S.; Lee, C. K.; Aguirre, G.;

Ortega, F.; Ziller, J. W.; Walsh, P. J. Inorg. Chem. 1999, 38, 3010.

(91) Pritchett, S.; Gantzel, P.; Walsh, P. J. Organometallics 1997, 16,

5130.

(95) Gamer, M. T.; Roesky, P. W.; Palard, I.; LeHellaye, M.;

Guillaume, S. M. Organometallics 2007, 26, 651.

(96) Amgoune, A.; Thomas, C. M.; Roisnel, T.; Carpentier, J.-F.

Chem.—Eur. J. 2006, 12, 169.

4180 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

Table 6. Solution Polymerization of ε-CL by Ti(N₂^MsN^OMe)(NMe₂)₂(18), Ti(N₂^TsN^Ph)(NMe₂)₂(19), Ti(N₂^MsN^Ph)(NMe₂)₂(20),

Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24), Ti(N₂^MsN^OMe)(OⁱPr)₂(21), Ti(N₂^TsN^Ph)(OⁱPr)₂(22), Ti(N₂^MsN^Ph)(OⁱPr)₂(23), Ti(CyN₂^Ts)(OⁱPr)₂

(25), and Ti(CyN₂^{SO Mes})(OⁱPr)₂(26)^a

2

entry

initiator

yield (%)^b

time (h)

k_app(min^-1

)

M_n(GPC)^c

M_n(calcd)^d

M_w/M_n

ref

e

1

2

3

4

5

6

7

8

Ti(N₂^TsN^OMe)(NMe₂)₂(29)

Ti(N₂^MsN^OMe)(NMe₂)₂(18)

Ti(N₂^TsN^Ph)(NMe₂)₂(19)

Ti(N₂^MsN^Ph)(NMe₂)₂(20)

Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24)

Ti(N₂^TsN^py)(OⁱPr)₂(7)

95

83

90

94

90

97

87

89

96

94

96

91

22

8

6

4

9

1

3

1.5

0.5

0.2

1

14 260

12 650

12 010

11 050

12 370

5720

6390

7390

6370

6500

5750

5770

1.60

1.46

1.35

1.39

2.08

1.80

1.39

1.57

1.73

1.41

1.28

1.87

68

f

0.005(3)

0.0130(4)

0.007(1)

0.065(3)

0.023(1)

0.042(2)

0.105(8)

0.510(30)

0.039(2)

68

Ti(N₂^TsN^OMe)(OⁱPr)₂(5)

Ti(N₂^MsN^OMe)(OⁱPr)₂(21)

Ti(N₂^TsN^Ph)(OⁱPr)₂(22)

Ti(N₂^MsN^Ph)(OⁱPr)₂(23)

Ti(CyN₂^Ts)(OⁱPr)₂(25)

f

9

10

11

12

5080

12 640

Ti(CyN₂^{SO Mes})(OⁱPr)₂(26)

2

^aData for the previously studied compounds are given for comparison.⁶⁸Conditions: [ε-CL]:[initiator] = 100:1, 6.8 mL of toluene at 100 °C. See

Experimental Section for other details. ^bIsolated yield at 100% NMR conversion. ^cMolecular weights (g mol^-1) determined from GPC using the

appropriate Mark-Houwink corrections. ^dExpected M_n(g mol^-1) for two chains growing per initiator at 100% conversion. ^eNot determined.

^fThis work.

Table 7. Solution Polymerization of ε-CL by Zr(N₃^TsN)(CH₂SiMe₃) (30), Zr(N₃^MsN)(CH₂SiMe₃) (31), Zr(N₃^ArFN)(CH₂SiMe₃) (32),

and Zr(N₃^ArFN)(NMe₂) (33)^a

entry

initiator

co-initiator

yield (%)^b

time (h)

k_app.(min^-1

)

M_n(GPC)^c

M_n(calcd)^d

M_w/M_n

ref

e

1

2

3

4

5

6

7

8

9

Zr(N₃^TsN)(CH₂SiMe₃) (30)

Zr(N₃^MsN)(CH₂SiMe₃) (31)

Zr(N₃^ArFN)(CH₂SiMe₃) (32)

Zr(N₃^ArFN)(NMe₂) (33)

91

96

93

92

93

91

96

89

1

1.5

1.4

1

0.11(1)

0.057(1)

22 270

15 090

17 600

13 720

7770

11 500

11 450

11 470

11,470

1.39

1.33

1.19

1.45

1.18

1.19

1.17

1.16

1.18

f

0.069(1)

0.121(4)

0.108(5)

0.071(1)

Zr(N₂^TsN^OMe)(OⁱPr)₂(6)

Zr(N₂^TsN^py)(OⁱPr)₂(8)

68

1

5180

e

Zr(N₃^TsN)(CH₂SiMe₃) (30)

Zr(N₃^MsN)(CH₂SiMe₃) (31)

Zr(N₃^ArFN)(CH₂SiMe₃) (32)

ⁱPrOH

0.9

0.7

0.6

11 630

10 850

11 280

e

0.093(2)

f

^aData for the previously studied compounds are given for comparison.⁶⁸Conditions: [ε-CL]:[initiator] = 100:1, 6.8 mL of toluene at 100 °C. See

Experimental Section for other details. ^bIsolated yield at 100% NMR conversion. ^cMolecular weights (g mol^-1) determined from GPC using the

appropriate Mark-Houwink corrections. ^dExpected M_n(g mol^-1) for one chain growing per initiator at 100% conversion. ^eThis work. ^fNeither first

nor second order in ε-CL.

Ring-Opening Polymerization of ε-Caprolactone. A princi-

pal aim of this work was to determine the effect of structural

variations of poly(sulfonamide)amine-supported group 4

complexes on the ROP of ε-CL and rac-LA. The titanium

amide and alkoxide complexes 18-26, the zirconium alkyl

and amide complexes 30-33, and the corresponding in situ

generated zirconium isopropoxide complexes were evalu-

ated. Polymerizations were performed in toluene at 100 °C

so as to allow direct comparison with the bis(sulfonamide)-

amine initiators 5-8 and 29 previously investigated.^68,69The

progress of each reaction was monitored by regular sam-

pling, and the results summarized in Tables 6 and 7 corres-

pond to the point at which the polymerizations eventually

reached 100% conversion. The molecular weights and PDIs

were determined by GPC using the appropriate Mark-

Houwink corrections,^99-101and the stated yields refer to

the amount of poly(ε-CL) isolated. The M_n(calcd) values are

those expected for either one or two polymer chains growing

per initiator at 100% conversion of monomer. The results for

titanium and zirconium are discussed in turn.

Titanium Initiators. Table 6 lists the results for the ROP of

ε-CL using the new titanium initiators. The results pre-

Ts

viously obtained with Ti(N₂^TsN^OMe)(NMe₂)₂(29), Ti(N₂

N

-

^OMe)(OⁱPr)₂(5), and Ti(N₂^TsN^py)(OⁱPr)₂(7) are given for

comparison. Of the nine new titanium initiators studied, the

bis(dimethylamide) systems 18-20 were the slowest (entries

2-4). The GPC data are indicative of only one poly(ε-CL)

chain forming per metal on average (M_n(GPC) = 11 050-

12 650 g mol^-1vs M_n(calcd)=11 450 g mol^-1), consistent with

the six-coordinate bis(dimethylamide) complexes studied

previously (cf. entry 1 (29)).⁶⁸The polymerization is rela-

tively well controlled, as indicated by PDIs in the range

1.35-1.46 and good agreement between found and calcu-

lated M_n(assuming one chain per metal). In order to evaluate

these bis(dimethyl)amide initiators further, the most efficient

one (20, entry 4) was monitored throughout the course of

polymerization. The consumption of monomer followed

first-order kinetics, and a reasonably linear plot of M_n(GPC)

vs percent conversion was obtained (see the Supporting In-

formation). There was also an induction period of around 60

min, as expected,^22,53,102owing to the slow first ε-CL inser-

tion into the Ti-NMe₂bond.

(97) Ma, H.; Okuda, J. Macromolecules 2005, 38, 2665.

(98) Martin, E.; Dubois, P.; Jerome, R. Macromolecules 2000, 33,

1530.

(99) Rudin, A.; Hoegy, H.; L., W. J. Polym. Sci., Part A: Polym.

Chem. 1972, 10, 217.

(100) Barakat, I.; Dubois, P.; Jerome, R.; Teyssie, P. J. Polym. Sci.,

ꢀ ^

ꢀ

The apparent first-order rate constant (k_app) of 0.005(3)

Part A: Polym. Chem. 1993, 31, 505.

s

^-1for the ROP using 20 is 200 times less than that for the

(101) Dorgan, J. R.; Janzen, J.; Knauss, D. M.; Hait, S. B.; Limoges,

B. R.; Hutchinson, M. H. J. Polym. Sci., Part B: Polym. Phys. 2005, 43,

3100.

(102) Duda, A.; Penczek, S. Macromolecules 1995, 28, 5981.

Article

Organometallics, Vol. 29, No. 18, 2010 4181

for the related bis(amide) 20. However the bis(alkoxide) ana-

logue 22 showed a negligible induction period of ca. 10 min

(Figure 8). Consistent with the apparent detrimental effect of

a spectator NMe₂group, the propagation rate constant for

24 of 0.0130(4) s^-1was significantly slower than for the

otherwise identical 22 (k_app= 0.042 s^-1, entry 9).

As had been observed previously for 5 and 7 (entries 7 and

6),⁶⁸changing from the bis(amide) initiators to the corres-

ponding bis(alkoxides) (entries 8-12) gives a clear switch

from ca. one to ca. two poly(ε-CL) chains per metal center

and also a significant increase in activity. The behavior of all

these bis(alkoxides) was investigated in detail (see Figures 8

and 9 for representative plots and the Supporting Informa-

tion for remaining plots; k_appvalues are given in Table 6). All

the initiators showed a fairly short induction period followed

by a polymerization process that was first order with respect

to ε-CL concentration (Figure 8). Similar induction periods

have been observed for main group and transition metal

isopropoxide initiators.^22,53,102Linear relationships between

M_nand percent conversion were found, and in the case of

Ti(CyN₂^Ts)(OⁱPr)₂(25, entry 11) there was excellent agree-

ment between the measured and predicted M_n(5080 vs 5720

g mol^-1) and the PDI was only 1.28. Compound 25 was also

the fastest initiator, with an induction period of just 2 min,

and polymerized 100 equiv of ε-CL to completion within

10 min after the induction period.

Figure 8. First-order plot for ε-CL consumption using Ti-

(N₂^TsN^Ph)(OⁱPr)₂(22). Conditions: [ε-CL₀]:[22] = 100:1, 6.8

mL of toluene, 100 °C, 0.1 mL aliquots taken at the given inter-

vals. See Experimental Section for other details. Linear fit (r²=

0.995) shown is for the first-order region after the induction

period.

There are some fairly well-defined relationships between

bis(sulfonamide) supporting ligand type and the rates of

polymerization. For a given sulfonyl substituent -SO₂R,

moving from a tetradentate ligand (N₂^RN^OMeor N₂^RN^py) to

a tridentate one generally gave a noticeable increase in

activity (cf. 29 vs 19; 18 vs 20; 21 vs 23) and also a narrower

Figure 9. Plots of M_nand PDI (determined by GPC) vs con-

version for the ROP of ε-CL using Ti(N₂^TsN^Ph)(OⁱPr)₂(22).

Conditions: [ε-CL₀]:[22] = 100:1, 6.8 mL of toluene, 100 °C, 0.1

mL aliquots taken at the given intervals. Hollow diamonds

correspond to M_nand hollow circles to PDI.

PDI and better agreement between measured and predicted

M_n. An apparent exception is 5 vs 22 (N₂^TsN^OMevs N₂

Ts

-

N

^Ph), where the latter was slightly slower and had a poorer

PDI. On the other hand, 22 is certainly much faster than the

previously reported 7 (N₂^TsN^Phvs N₂^TsN^py), the latter hav-

ing the better (pyridyl) donor. Compound 25 was the fastest

of all the N-tosyl sulfonamide initiators (7, 5, 22, 25). Altho-

ugh 25 is not formally four-coordinate, since both -NSO₂Ts

donors have κ(N,O) bidentate coordination,⁷⁶the change

corresponding bis(isopropoxide) initiator (23, entry 10) with

the same supporting ligand. This would suggest that the

spectator NMe₂in the propagating species Ti(N₂^RN^X)-

(NMe₂){poly(ε-CL)} has a detrimental effect on the rate

on ε-CL insertion into the Ti-OCH₂bond of the growing

polymeryl chain. This is attributed to a combination of steric

and strong π-donation effects of NMe₂(inhibiting monomer

coordination) in comparison with those from a second

growing polymeryl chain (i.e., the situation found starting

from the bis(isopropoxide) initiators). The slower rates of

propagation for the bis(dimethylamide) systems 18-20 com-

pared to their bis(isopropoxide) analogues (21-23, entries

8-10) parallels the behavior of the six-coordinate initiators

29 and 5 (entries 1 and 7) reported previously.⁶⁸

To test this interpretation further, the mixed isopropoxide

bis(dimethylamide) complex Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24,

entry 5) was also evaluated. The measured M_nwas in close

agreement with one polymeryl chain growing per metal

center. The ¹H NMR spectrum of the poly(ε-CL) was domi-

nated by -OⁱPr end groups, consistent with preferred initial

insertion into the Ti-OⁱPr bond of 24. The MALDI-ToF-

MS spectra were also consistent with this, but also showed a

minor distribution indicative of amide initiation (see the

Supporting Information). When the polymerization process

was followed by regular sampling, it was found that ROP

using 24 had an induction period of ca. 30 min (see the Sup-

porting Information), which is significantly shorter than that

Ts

from the bis(sulfonamide)amine ligands N₂^TsN^Rto CyN₂

clearly has a beneficial effect.

For a given ligand type (N₂^RN^OMe, N₂^RN^Ph, CyN₂^R) there

are also some reasonably consistent relationships between

activity and sulfonyl substituent -SO₂R (R=Tol, Me, or

Mes). The most noticeable differences are between 25 and 26

(CyN₂^Tsvs CyN₂^{SO Mes}) in terms of improved activity (k_app

2

of 0.510(3) and 0.039(2) s^-1) and a narrower PDI in the case

of 25. Furthermore, the measured M_nvalues of 5080 and

12 640 g mol^-1show that the greater steric constraints as-

SO₂Mes

sociated with CyN₂

prevent the formation of two

polymer chains per metal center. Comparison of 19 and 20,

and 22 and 23 (N₂^TsN^Phvs N₂^MsN^Ph) shows a smaller in-

crease in activity on changing from -SO₂Tol to -SO₂Me,

which may also be due to steric factors. Comparison of 29

and 18 (N₂^TsN^OMevs N₂^MsN^OMe) also reveals an increase in

activity on changing from -SO₂Tol to -SO₂Me. In contrast,

comparing 5 and 21 with the same N₂^TsN^OMeand N₂^MsN^OMe

ligands revealed that 21 was the slowest with somewhat

poorer control.

Zirconium Initiators. The zirconium complexes Zr-

(N₃^TsN)(CH₂SiMe₃) (30), Zr(N₃^MsN)(CH₂SiMe₃) (31), Zr-

(N₃^ArFN)(CH₂SiMe₃) (32), and Zr(N₃^ArFN)(NMe₂) (33)

4182 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

Table 8. Solution Polymerization of rac-Lactide by Ti(N₂^TsN^OMe)(OⁱPr)₂(5), Ti(N₂^TsN^py)(OⁱPr)₂(7), Ti(N₂^TsN^Ph)(OⁱPr)₂(22),

Ti(N₂^MsN^Ph)(OⁱPr)₂(23), Ti(CyN₂^Ts)(OⁱPr)₂(25), and Ti(CyN₂^{SO Mes})(OⁱPr)₂(26)^a

2

entry

initiator

co-initiator

conversion (%)^b

time (h)

k_app(h^-1

)

M_n(GPC)^c

M_n(calcd)^d

M_w/M_n

ref

e

f

1

2

3

4

5

6

Ti(N₂^TsN^OMe)(OⁱPr)₂(5)

Ti(N₂^TsN^py)(OⁱPr)₂(7)

Ti(N₂^TsN^Ph)(OⁱPr)₂(22)

Ti(N₂^MsN^Ph)(OⁱPr)₂(23)

Ti(CyN₂^Ts)(OⁱPr)₂(25)

90

73

71

44

32

46

24

11

2

4

8990

7230

7900

3980

3620

4880

6540

5320

5160

3230

2370

3370

1.21

1.16

1.13

1.09

1.16

1.22

0.119(4)

0.29(2)

0.11(1)

0.15(1)

Ti(CyN₂^{SO Mes})(OⁱPr)₂(26)

2

^aConditions: [rac-LA]:[initiator] = 100:1, 6.8 mL of toluene at 70 °C. See Experimental Section for other details. ^bNMR conversion. ^cMolecular

weights (g mol^-1) determined from GPC using the appropriate Mark-Houwink corrections. ^dExpected M_n(g mol^-1) for two chains growing per

initiator at the given conversion. ^eNot determined. ^fThis work.

Table 9. Solution Polymerization of rac-Lactide by Zr(N₃^TsN)(CH₂SiMe₃) (30), Zr(N₃^MsN)(CH₂SiMe₃) (31), Zr(N₃^ArFN)(CH₂SiMe₃)

(32), and Zr(N₃^ArFN)(NMe₂) (33)^a

entry

initiator

co-initiator

conversion (%)^b

time (h)

k_app(h^-1

)

M_n(GPC)^c

M_n(calcd)^d

M_w/M_n

ref.

1

2

3

4

5

6

7

Zr(N₂^TsN^OMe)(OⁱPr)₂(6)

Zr(N₃^TsN)(CH₂SiMe₃) (30)

Zr(N₃^MsN)(CH₂SiMe₃) (31)

Zr(N₃^ArFN)(NMe₂) (33)

94

84

83

66

96

95

89

6

61

65

66

20

24

0.468(2)

0.033(2)

0.021(1)

0.018(1)

0.154(2)

0.170(3)

0.134(4)

8290

23 690

18 920

29 260

14 340

15 600

13 005

13 610

12 190

12 050

9550

13 890

13 750

12 880

1.15

1.19

1.38

1.19

1.08

1.11

1.08

68

e

Zr(N₃^TsN)(CH₂SiMe₃) (30)

Zr(N₃^MsN)(CH₂SiMe₃) (31

Zr(N₃^ArFN)(CH₂SiMe₃) (32)

ⁱPrOH

^aData for the previously studied compound 6 are given for comparison.⁶⁸Conditions: [rac-LA]:[initiator]:[co-initiator] = 100:1:1, 6.8 mL of toluene

at 70 °C. See Experimental Section for other details. ^bNMR conversion. ^cMolecular weights (g mol^-1) determined from GPC using the appropriate

Mark-Houwink corrections. ^dExpected M_n(g mol^-1) for one chain growing per initiator at the given conversion. ^eThis work.

were also all evaluated for the ROP of ε-CL. In situ genera-

tion of zirconium alkoxides was performed by reaction of

(Zr-OⁱPr being a more efficient initiating group than

CH₂SiMe₃or NMe₂), and consequently, the PDIs of the

polymers were also narrower (range 1.16-1.18). The experi-

mental M_nvalues (10 850-11 630 g mol^-1) were also in much

better agreement with the calculated ones (11 470 g mol^-1).

Zr(N₃^RN)(CH₂SiMe₃) with PrOH immediately prior to

i

monomer addition. The results are summarized in Table 7,

along with the corresponding data for the previously reported

bis(sulfonamide)amine-supported initiators Zr(N₂^TsN^OMe)-

(OⁱPr)₂(6) and Zr(N₂^TsN^py)(OⁱPr)₂(8). Plots of monomer

conversion vs time and the corresponding first-order ki-

netic plots, and plots of M_nand PDI vs conversion, are

given in the Supporting Information for all four initiators.

Entries 1-4 of Table 7 show the ROP results for 30-33 in

1

The H NMR and MALDI-ToF mass spectra of the poly-

mers showed the expected -OⁱPr chain ends. There was no

specific effect on the apparent propagation rate constant

i

compared to the PrOH-free systems. This is as expected

since the propagating species Zr(N₃^RN){poly(ε-CL)} are

indentical for each N₃^RN ligand (except for the polymeryl

i

the absence of PrOH co-initiator. As expected for metal-

chain end). For 32 þ PrOH the consumption of ε-CL was

alkyl and -amide initiating groups, the experimental M_n

values (13 720-22 270 g mol^-1) are all somewhat higher than

once again neither first nor second order. Whereas Zr-

(N₃^ArFN)(CH₂SiMe₃) (32) þ PrOH gave poly(ε-CL) with

i

expected at 100% conversion of monomer (11 410 g mol^-1

)

exclusively -OⁱPr chain ends, initial results showed that use

of Zr(N₃^ArFN)(NMe₂) (33) þ PrOH gave polymers with

i

and one chain growing per inititing group, and the PDIs

are moderately broad. Despite the higher than expected

M_nvalues, the M_n(GPC) vs conversion plots show good

linearity. This may indicate that there is some instability of

the preinitiators (or some other reason for not all the com-

plexes initiating ROP), but once they enter the catalytic cycle,

the polymerization proceeds in a living fashion. The MAL-

DI-ToF mass spectra of the poly(ε-CL)s showed the presence

of the expected CH₂SiMe₃or NMe₂end-groups. For 32 the

consumption of ε-CL was neither first nor second order,

whereas that for the amide homologue 33 did follow first-

order kinetics (see the Supporting Information). Compound

30 showed an induction period of ca. 20-30 min, whereas 31

and 32 had shorter ones. The induction period for 33 of ca.

20-30 min is consistent with Zr-NMe₂being a less efficient

initating group than Zr-CH₂SiMe₃, and the broader PDI

for the poly(ε-CL) formed with 33 (1.45 vs 1.19 with 32) also

supports this assertion.

both -NMe₂and -OⁱPr end-groups. Since NMR tube

studies showed that 33 was immediately and quantitatively

convered to the corresponding isopropoxide complex, the

-NMe₂-terminated polymer may arise through an activated

monomer mechanism (initiated by HNMe₂), as we have

discussed recently.¹⁰³

Although the ROP activity using 30-32 (with or without

co-initiator) was fairly efficient for a group 4 system in

general,¹¹there was no improvement on the previous systems

6 and 8 (entries 5 and 6) as judged by the propagation rate

constants. Furthermore, the activities of 30-32 are not

significantly better on average than the five-coordinate titanium

initiators 22 and 23. At first sight this seems surprising, given the

results of Kol and Davidson with the tris(phenolate)amine

systems 4 (Figure 1, vide supra).^57,59However, the solid-state

structures of 30, 32, and 33 show that the systems Zr(N₃^RN)X

(X = alkyl, amide, alkoxide) are not five-coordinate analogues

of 4, and in fact all possess eight-coordinate metal centers (and

Entries 7-9 of Table 7 show the effects of adding 1 equiv

i

of PrOH co-initiator to 30-32 prior to introducing the

ε-CL. As expected, this had a beneficial effect on several

aspects of the ROP. The induction periods were all reduced

(103) Clark, L.; Cushion, M. G.; Dyer, H. E.; Schwarz, A. D.;

Duchateau, R.; Mountford, P. Chem. Commun. 2010, 46, 273.

Article

Organometallics, Vol. 29, No. 18, 2010 4183

Of the new titanium compounds reported herein, the five-

and four-coordinate initiators 22, 23, 25, and 26 were chosen

for rac-LA since these had demonstrated the best behavior

for the ROP of ε-CL. Table 8 also gives ROP data for the pre-

viously reported Ti(N₂^TsN^OMe)(OⁱPr)₂(5) and Ti(N₂^TsN^py)-

(OⁱPr)₂(7) for comparison (entries 1 and 2) since these had

not been included in our earlier studies. These two six-

coordinate compounds gave 90% and 73% conversion of

100 equiv of LA to poly(rac-LA) after 24 h. The polymers

formed had reasonably narrow PDIs, although the experi-

mental M_nvalues were somewhat higher than expected for

two chains per initiator (i.e., one chain per Ti-OⁱPr bond).

Analogous behavior was found previously for Zr(N₂^TsN^OMe)-

(OⁱPr)₂(6)⁶⁸(Table9, entry1), with7achieving 94% conversion

after 6 h (cf. 90% after 24 h for the titanium congener 5). The

higher conversions achieved with 6 are consistent with literature

trends and more facile access to the metal center for the larger

metal. The lower conversion with 7 compared to 5 is in line with

the ε-CL ROP results (Table 6, entries 6 and 7) and may reflect

the stronger binding of the pyridyl pendant donor in 7. In all

cases, the ¹H NMR and MALDI-ToF mass spectra showed the

presence of -OⁱPr chain ends. The MALDI-ToF spectra

showed peak envelopes separated by m/z values of 72 (i.e.,

one-half of a LA unit), consistent with extensive transesterifica-

tion, as was found with 6 previously. Disappointingly (as was

also found for 6), all of the polymers produced were atactic

(P_r≈ 0.50), as judged by the selectively homonuclear decoupled

¹H NMR spectra.¹⁰⁴

Figure 10. First-order plot for rac-LA consumption using Zr-

(N₃^MsN)(CH₂SiMe₃) (31) with PrOH co-initiator. Conditions

i

[rac-LA₀]:[31]:ⁱPrOH = 100:1:1, 6 mL of toluene, 70 °C, 0.1 mL

aliquots taken at the given intervals. See Experimental Section

for other details. Linear fit (r²=0.998).

The five- and four-coordinate initiators 22, 23, 25, and 26

were apparently unstable for extended polymerization

times (see the Supporting Information for detailed con-

version vs time plots). Each compound initially catalyzed

the well-controlled ROP of rac-LA, as judged by well-

behaved first-order kinetics, linear M_nvs time plots, and

relatively narrow PDIs in the range ca. 1.1-1.2 (entries 3-6).

All four gave polymers with M_nvalues slightly higher than

expected for two chains growing per metal center (cf. com-

pounds 5-7). However, after the times listed in Table 8

polymerization activity halted. Nonetheless, some structure-

-activity trends can be inferred from the data and the

apparent propagation rate constants (k_app) listed in Table 8,

which correspond to the kinetically well-behaved parts of the

ROP process. As found for the ε-CL studies, changing from

six-coordinate 5 and 7 to five-coordinate 22 and 23 (N₂^RN^X

vs N₂^TsN^Ph) gave an increase in activity (e.g., 73% and 71%

conversion after 24 and 11 h for 5 and 22, respectively) and

reduction in PDI. Changing from 22 to 23 (-NSO₂Tol vs

-NSO₂Me) also gave an increase in activity, as judged by

k_app, and again a decrease in PDI. However, there was

surprisingly little difference between 25 and 26 (-NSO₂Tol

vs -NSO₂Mes) in terms of chains per metal and rate, despite

Figure 11. Plots of M_nand PDI (determined by GPC) vs con-

version for the ROP of rac-LA using Zr(N₃^MsN)(CH₂SiMe₃)

(31) with ⁱPrOH co-initiator. Conditions: [rac-LA₀]:[31]:ⁱPrOH =

100:1:1, 6.0 mL of toluene, 70 °C, 0.1 mL aliquots taken at the

given intervals. Hollow diamonds correspond to M_nand hollow

circles to PDI.

indeed might be dimeric under polymerization conditions,

although this seems less likely). The ability of the -ΝSO₂R

donors to act as bidentate κ(N,O) donors saturates the zirco-

nium coordination spheres, and presumably one or more

sulfonamide groups need to change to a monodentate coordi-

nation mode to accept the incoming ε-CL prior to insertion into

the growing Zr-polymeryl chain.

Ring-Opening Polymerization of rac-Lactide. Polymeriza-

tion studies with rac-LA were carried out in toluene at 70 °C

as for the previously studied complex Zr(N₂^TsN^OMe)(OⁱPr)₂

(6).⁶⁸Reactions were monitored by regular sampling for the

determination of rac-LA conversion and experimental M_n

values. The data are summarized in Tables 8 and 9 for a

number of titanium and zirconium initiators, respectively.

Some representative plots using Zr(N₃^MsN)(CH₂SiMe₃) (31)

R

the different steric demands of the two CyN₂ligands.

Overall, in terms of rate, PDI, and agreement between found

and predicted M_nthe five-coordinate Ti(N₂^MsN^Ph)(OⁱPr)₂

(23) gave the best ROP performance, although the longer-

term stability was poor (44% conversion when the ROP

ceased after 2 h).

Zirconium Initiators. Table 9 summarizes the ROP results

for 30, 31, and 33 (entries 2-4) and also the previously stud-

ied Zr(N₂^TsN^OMe)(OⁱPr)₂(6, entry 1), which, as mentioned,

i

with PrOH co-initiator are given in Figures 10 and 11 and

will be discussed below. Further details are given in the

Supporting Information (conversion vs time, M_nand PDI

vs conversion, and first-order log plots).

Titanium Initiators. Table 8 summarizes the ROP perfor-

mance of a number of titanium bis(isopropoxide) complexes.

(104) Zell, M. T.; Padden, B. E.; Paterick, A. J.; Thakur, K. A. M.;

Kean, R. T.; Hillmyer, M. A.; Munson, E. J. Macromolecules 2002, 35,

7700.

4184 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

forms between one and two poly(rac-LA) chains per metal

center with 94% conversion of 100 equiv of rac-LA after 6 h.

As with the ε-CL study, the alkyl and amide complexes were

active for ROP of rac-LA. Although the polymerization

proceeded very slowly (66-84% conversion after 61-66 h)

and formed poly(rac-LA) with much higher than expected

M_nvalues, the M_n(GPC) vs percent conversion plots were

fairly linear. This may indicate that there is some instability

or side reactions of the complexes prior to initiation, but that

once they have entered the catalytic cycle the polymerization

proceeds in a living fashion, as had been observed for ε-CL.

The MALDI-ToF mass spectra of low molecular weight

polymer samples showed the expected -CH₂SiMe₃or -NMe₂

end groups, as well as evidence of extensive transesterification,

with peak envelopes of m/z = 72 apart.

titanium, with lower coordination number and less ster-

ically encumbered ligands giving the most active and best

controlled initiators. The previously reported Ti(CyN₂^Ts)-

(OⁱPr)₂(25) was the most active initiator of those surveyed

for the ROP of ε-CL. All of the titanium complexes were

poor initiators for the ROP of rac-lactide, giving poor con-

versions (30-70%), although bulkier groups on a given

ligand backbone generally gave higher conversions.

C₃-Symmetric zirconium initiators were also investigated.

The protio-ligands H₃N₃^RN reacted readily with Zr(CH₂-

SiMe₃)₄to form Zr(N₃^RN)(CH₂SiMe₃) (R=Ts (30), Ms (31),

or Ar^F(32)). Reaction of Zr(NMe₂)₄with H₃N₃^ArFN gave

the corresponding amide complex Zr(N₃^ArFN)(NMe₂) (33),

which was found to be highly moisture sensitive, giving

{Zr(N₃^ArFN)}₂(μ-O) (34) in the presence of trace amounts

of water. In situ protonolysis reactions with Zr(N₃^TsN)(CH₂-

SiMe₃) (30) or 33 yielded the corresponding alkoxide or

aryloxide complexes. As has been observed elsewhere, the

zirconium complexes generally provided more active ROP

initiators than the titanium analogues, although Ti(CyN₂^Ts)-

(OⁱPr)₂was the most active of all those surveyed for ε-CL. In

situ generation of alkoxide initiators from the zirconium

alkyl complexes gave well-controlled rac-LA initiators with

narrow polymer PDIs and good agreement between experi-

mental and predicted M_nvalues. Of these compounds, the

least bulky sulfonyl substituents yielded the most active

initiator for rac-lactide. However, these initiators were gen-

erally found to be less active than the six-coordinate zirco-

nium complexes investigated previously,⁶⁸which is attri-

buted to the saturating effects of the κ²(N,O) coordination.

As for the ε-CL studies, in situ generation of alkoxide

initiators from 31-32 (entries 5-7) gave a much more ef-

ficient and better controlled polymerization process. Linear

relationships between M_nand conversion were obtained

that remained effectively constant throughout the reaction.

(Figures 10 and 11 and the Supporting Information).

The resulting polymers had extremely narrow PDIs

(1.08-1.11), and there was very good agreement between

experimental and predicted molecular weights. Consistent

with this, the gradients of the M_nvs percent conversion plots

were in the range 144(2)-159(4) g mol^-1(% conversion)^-1

,

in agreement with that expected (144.1 g mol^-1(% con-

version)^-1) for one poly(rac-LA) chain growing per metal

center. The MALDI-ToF mass spectra showed exclusively

isopropoxide-terminated poly(rac-LA) chains with peak en-

velope separations of m/z=72. Regrettably, all of the poly-

(rac-LA)s produced with these initiators were atactic (P_r≈

0.5), as has been observed for all other poly(sulfonamide)-

supported group 4 and aluminum ROP initiators. All of

these data are consistent with these species acting as

initiators for the living ROP of rac-lactide. The k_appvalues

show a dependence on the sulfonamide group with the

general order of Ms (31) > Ts (30) > Ar^F(32). This may

indicate easier access of the monomer to metal center as steric

demands of the sulfonyl S-substituent diminish. However, in

comparison to six-coordinate 6, the alkoxide initiators de-

rived from 30-32 were 3 to 4 times slower. This is consistent

with the slightly slower rates found for 30-32 for ε-CL ROP

(vide supra). Again, the bidentate κ(N,O) coordination of the

-ΝSO₂R donors presumably inhibits monomer coordina-

tion to the metal center.

Experimental Section

General Methods and Instrumentation. All manipulations

were carried out using standard Schlenk line or drybox techni-

ques under an atmosphere of argon or dinitrogen. Solvents were

degassed by sparging with dinitrogen and dried by passing

through a column of the appropriate drying agent. Toluene

was refluxed over sodium and distilled. Deuterated solvents

were dried over sodium (C₆H₆) or P₂O₅(CDCl₃and CD₂Cl₂),

distilled under reduced pressure, and stored under dinitrogen in

Teflon valve ampules. NMR samples were prepared under di-

nitrogen in 5 mm Wilmad 507-PP tubes fitted with J. Young

Teflon valves. ¹H and ¹³C{¹H} NMR spectra were recorded on

Varian Mercury-VX 300 and Varian Unity Plus 500 spectro-

meters at ambient temperature unless stated otherwise and

referenced internally to residual protio-solvent (¹H) or solvent

(¹³C) resonances and are reported relative to tetramethylsilane

(δ = 0 ppm). Assignments were confirmed using two-dimen-

sional ¹H-¹H and ¹³C-¹H NMR correlation experiments. Che-

mical shifts are quoted in δ (ppm) and coupling constants in Hz.

IR spectra were recorded on a Nicolet Magna 560 ESP FTIR

spectrometer. Samples were prepared in a drybox as Nujol mulls

between NaCl plates, and the data are quoted in wavenumbers

(cm^-1). Elemental analyses were carried out by the Elemental

Analysis Service at the London Metropolitan University.

MALDI-ToF-MS analysis was performed on a Waters

MALDI micro equipped with a 337 nm nitrogen laser. An

accelerating voltage of 25 kV was applied. The polymer samples

were dissolved in THF at a concentration of 1 mg mL^-1. The

cationization agent used was potassium trifluoroacetate (Fluka,

>99%) dissolved in THF at a concentration of 5 mg mL^-1. The

matrix used was trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-pro-

penylidene]malononitrile (DCTB) (Fluka) and was dissolved in

THF at a concentration of 40 mg mL^-1. Solutions of matrix,

salt, and polymer were mixed in a volume ratio of 4:1:4, res-

pectively. The mixed solution was hand-spotted on a stainless

Conclusion

In this contribution we have carried out systematic bis-

(sulfonamide) ligand variations leading to improved ROP

activity and control for the thermally robust group 4 com-

plexes of these ligands. The new bis(dimethylamide) com-

plexes Ti(N₂^MsN^OMe)(NMe₂)₂(18), Ti(N₂^TsN^Ph)(NMe₂)₂

(19), and Ti(N₂^MsN^Ph)(NMe₂)₂(20) were easily prepared

under mild conditions from Ti(NMe₂)₄. Similarly, the bis-

(isopropoxide) complexes Ti(N₂^MsN^OMe)(OⁱPr)₂(21), Ti-

(N₂^TsN^Ph)(OⁱPr)₂(22), and Ti(N₂^MsN^Ph)(OⁱPr)₂(23) could

be synthesized from Ti(OⁱPr)₂(NMe₂)₂and the appropriate

protio-ligand. Reaction between an excess of H₂N₂^TsN^Ph

and Ti(OⁱPr)₂(NMe₂)₂gave Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24).

In accordance with our previous study, the amides were

poorer initiators than the alkoxides toward the ROP of ε-CL.

Certain structure-activity relationships were observed with

Article

Organometallics, Vol. 29, No. 18, 2010 4185

steel MALDI target and left to dry. The spectra were recorded in

the refectron mode.

(trifluoromethyl)benzene sulfonyl chloride (15.00 g, 48.1 mmol)

in diethyl ether (210 mL), and the mixture was stirred at RT for

2 h. The white solid that formed was filtered and washed with

H₂O (3 ꢀ 15 mL) and diethyl ether (3 ꢀ 20 mL). The product was

recrystallized from acetonitrile (40 mL) and methanol (50 mL).

Yield: 12.37 g (79%). ¹H NMR ((CD₃)₂SO, 299.8 MHz, 298 K):

δ 8.43 (3H, s, 4-C₆H₃(CF₃)₂), 8.31 (6H, s, 2-C₆H₃(CF₃)₂), 7.90

(3H, br s, NH), 2.76 (6H, t, ³J = 6.1 Hz, SNCH₂CH₂N), 2.29

(6H, t, ³J = 6.1 Hz, SNCH₂CH₂N) ppm. ¹³C{¹H} NMR

((CD₃)₂SO, 75.4 MHz, 298 K): δ 143.5 (1-C₆H₃(CF₃)₂), 131.4

(CF₃, ¹J = 33.2 Hz), 127.0 (2-C₆H₃(CF₃)₂), 124.4 (3-C₆H₃-

(CF₃)₂), 120.8 (4-C₆H₃(CF₃)₂), 53.2 (SNCH₂CH₂N), 40.4

(SNCH₂CH₂N) ppm. ¹⁹F NMR ((CD₃)₂SO, 282.1 MHz, 298

K): -61.7 (18F, s, CF₃) ppm. IR (KBr plates, Nujol mull cm^-1):

3292 (s), 3227 (m), 3091 (w), 1627 (w), 1408 (w), 1336 (m), 1298

(s), 1282 (s), 1268 (w), 1195 (m), 1139 (s), 1111 (m), 1089 (w),

1046 (w), 1018 (w), 959 (w), 950 (w), 906 (s), 947 (m), 838 (w), 794

(w), 726 (w), 697 (m), 682 (m), 629 (w). Anal. Found (calcd for

C₃₀H₂₄F₁₈N₄O₆S₃): C, 37.03 (36.97); H, 2.48 (2.50); N, 5.84

(5.75).

Polymer molecular weights (M_n, M_w) were determined by

GPC using a Polymer Laboratories Plgel Mixed-D column (300

mm length, 7.5 mm diameter) and a Polymer Laboratories PL-

GPC50 Plus instrument equipped with a refractive index detec-

tor. THF (HPLC grade) was used as an eluent at 30 °C with a

rate of 1 mL min^-1. Linear polystyrenes were used as primary

calibration standards, and Mark-Houwink corrections for

poly(ε-CL) or poly(rac-LA) in THF were applied for the ex-

perimental samples.^99-101

Starting Materials. Ti(NMe₂)₄,^105,106Ti(NMe₂)₂(OⁱPr)₂,^107,108

SO₂Mes 72

,

Zr(CH₂SiMe₃)₄,¹⁰⁹Zr(NMe₂)₄,¹¹⁰H₂CyN₂

,

^{Ts 71}H₂CyN₂

H₃N₃^TsN,⁷⁰N-mesylaziridine,¹¹¹H₂N₂^TsN^{Ph 69}H₂N₂^MsN^{Ph 69}

,

and N-tosylaziridine⁷⁵were synthesized according to published

procedures. ε-CL was dried over freshly ground CaH₂and distilled

before use. rac-LA was recrystallized twice from toluene and then

sublimed twice prior to use. Other reagents were purchased from

Sigma-Aldrich and used without further purification.

H₂N₂^MsN^OMe(12). To a solution 1-(methylsulfonyl)aziridine

(4.42 g, 0.036 mol) in EtOH (200 mL) at room temperature was

added 2-methoxyethylamine (1.44 g, 0.017 mol) dropwise. The

resulting orange solution was stirred at 40 °C for 16 h, filtered,

and concentrated under reduced pressure. An analytically pure

sample was obtained by column chromatography (SiO₂, eluent

system ethyl acetate-pentane-NH₄OH(aq), 50:10:1) to give a

light orange oil. Yield: 5.64 g (50%). ¹H NMR (CDCl₃, 293 K,

300.7 MHz): δ 3.40 (2H, t, ³J = 4.9 Hz, CH₂OMe), 3.10 (4H, t,

³J = 5.4 Hz, CH₂NHS), 3.36 (3H, s, OMe), 2.90 (6H, s, SO₂Me),

2.68 (4H, t, ³J = 5.4 Hz, CH₂CH₂NHS), 2.60 (2H, t, ³J = 4.9

Hz, CH₂CH₂OMe). ¹³C{¹H} NMR (CDCl₃, 293 K, 75.4 MHz):

δ 70.8 (CH₂CH₂OMe), 59.0 (OMe), 54.0 (CH₂CH₂NS), 52.8

(SO₂Me), 42.0 (CH₂NHS), 40.0 (CH₂OMe). IR (KBr plates,

Nujol mull cm^-1): 3285 (m), 3015 (w), 2847 (s), 2361 (w), 2340

(w), 1736 (w), 1718 (w), 1700 (w), 1684 (w), 1653 (m), 1645 (m),

1636 (m), 1559 (w), 1378 (s), 1199 (w), 1057 (m), 1012 (w), 973

(m), 769 (m), 668 (w), 523 (w). ES-HRMS: m/z = 318.1142

([M þ 1]^þ; calcd for C₉H₂₃N₃O₅S₂318.1157).

Ti(N₂^MsN^OMe)(NMe₂)₂(18). A solution of H₂N₂^MsN^OMe

(0.50 g, 0.002 mol) in benzene (20 mL) was added dropwise to

Ti(NMe₂)₄(1.06 g, 0.005 mol) in benzene (20 mL) and stirred for

4 h. The resulting red solution was filtered and volatiles were

removed under reduced pressure to give 18 as a red solid.

Further solid was obtained by recrystallization from toluene

(50 mL) at -80 °C, combined, and dried in vacuo. Yield: 0.11 g

1

(40%). H NMR (C₆D₆, 298 K, 300.7 MHz): δ 3.40 (12H, s,

NMe₂), 3.40 (2H, m, CH₂NS), 2.90 (3H, s, OMe), 2.80 (2H, t, ³J =

4.3 Hz, CH₂OMe), 2.50 (6H, s, SO₂Me), 2.40 (2H, t, ³J = 7.2 Hz,

NCH₂CH₂OMe), 2.20 (4H, t, ³J = 6.4 Hz, NCH₂CH₂NS).

¹³C{¹H} (C₆D₆, 298 K, 75.4 MHz): δ68.0 (NMe₂), 60.0(CH₂OMe),

53.0 (CH₂CH₂NS), 52.0 (CH₂CH₂O),49.0 (CH₂CH₂NS). IR data

(KBr plates, Nujol mull, cm^-1): 1734 (w), 1717 (w), 1700 (w),

1684 (w), 1653 (w), 1646 (w), 1636 (w), 1616 (w), 1576 (w), 1559

(w), 1540 (w),1521 (w), 1507 (w), 1418 (m), 1376 (m), 1259 (s),

1181 (w), 1154 (s), 1118 (m), 1020 (m), 949 (s), 788 (s), 667 (w),

593 (m), 502 (s). EI-HRMS: m/z = 407.0904 ([M - NMe₂]^þ;

calcd for C₁₃H₃₃N₅O₅S₂Ti 407.0902). Anal. Found (calcd for

C₁₃H₃₃N₅O₅S₂Ti): C, 34.47 (34.59); H, 7.19 (7.37); N, 15.64

(15.51).

H₃N₃^MsN (14). To a solution of tris(2-aminoethyl)amine (25.2

g, 26.0 mL, 0.170 mol) and NaOH (22.0 g, 0.550 mol) in H₂O

(100 mL) was added dropwise a solution of methanesulfonyl

chloride (59.2 g, 40.0 mL, 0.520 mol) in diethyl ether (300 mL),

and the mixture was stirred at RT for 2 h. The resulting solution

was evaporated under reduced pressure, and the resulting white

solid recrystallized from H₂O (30 mL) at 100 °C. The resulting

white crystalline solid was filtered and dried in vacuo. Yield: 12.0

g (18%). ¹H NMR ((CD₃)₂SO, 299.8 MHz, 298 K): δ 7.00 (3H,

br s, NH), 3.15 (6H, br s, SNCH₂CH₂N), 3.07 (9H, s, SO₂Me),

2.65 (6H, br s, SNCH₂CH₂N) ppm. ¹³C{¹H} NMR ((CD₃)₂SO,

75.4 MHz, 298 K): δ 67.8 (SNCH₂CH₂N), 65.5 (SNCH₂CH₂N),

54.7 (SO₂Me) ppm. IR (KBr plates, Nujol mull cm^-1): 3301 (m),

3242 (s), 3224 (s), 2725 (w), 1419 (w), 1348 (w), 1305 (s), 1261

(m), 1234 (w), 1224 (m), 1162 (s), 1137 (s), 1091 (m), 1068 (w),

1046 (w), 1026 (w), 1000 (w), 986 (m), 962 (s), 926 (m), 909 (w),

807 (m), 772 (s), 721 (w), 668 (w). ES-HRMS: m/z = 381.0926

([M þ H]^þ; calcd for C₉H₂₅N₄O₆S₃381.0936). Anal. Found

(calcd for C₉H₂₄N₄O₆S₃): C, 28.41 (28.41); H, 6.43 (6.36); N,

14.65 (14.72).

Ti(N₂^TsN^Ph)(NMe₂)₂(19). To a solution of H₂N₂^TsN^Ph(2.00

g, 4.0 mmol) in benzene (50 mL) was added a solution of

Ti(NMe₂)₄(0.94 mL, 12.0 mmol) in benzene (30 mL). The

mixture was stirred for 2 h, resulting in a red solution. The

solution was concentrated under reduced pressure to yield a

yellow solid, which was recrystallized from a concentrated

benzene solution (20 mL) layered with pentane (30 mL) to yield

19 as red diffraction-quality crystals, which were washed with

pentane (3 ꢀ 20 mL) and dried in vacuo. Yield: 2.18 g (86%). ¹H

NMR (CD₂Cl₂, 499.9 MHz, 248 K): δ 7.73 (4H, d, ³J = 10.0 Hz,

2-C₆H₄Me), 7.36-7.32 (3H, m, overlapping 3-C₆H₅and

3

4-C₆H₅), 7.29 (4H, d, J = 10.0 Hz, 3-C₆H₄Me), 7.06 (2H, d,

³J = 9.5 Hz, 2-C₆H₅), 3.72 (2H, s, CH₂C₆H₅), 3.57 (8H, s,

overlapping NMe₂and TsNCH₂CH₂N), 3.31 (8H, s, overlap-

ping NMe₂and TsNCH₂CH₂N), 2.68 (2H, m, TsNCH₂CH₂N),

2.57 (2H, m, TsNCH₂CH₂N), 2.37 (6H, s, C₆H₄Me) ppm.

¹³C{¹H} NMR (CD₂Cl₂, 248 K, 75.4 MHz): δ 142.0 (1-

C₆H₄Me), 139.0 (4-C₆H₄Me), 131.3 (4-C₆H₅), 130.0 (1-C₆H₅),

129.3 (3-C₆H₄Me), 128.7 (4-C₆H₅), 128.5 (3-C₆H₅), 127.1

(2-C₆H₄Me), 54.6 (CH₂C₆H₅), 50.9 (NMe₂), 47.8 (TsNCH₂-

CH₂N), 47.7 (NMe₂), 45.6 (TsNCH₂CH₂N), 21.3 (C₆H₄Me)

ppm. IR (KBr plates, Nujol mull cm^-1): 2774 (w), 1342 (w), 1294

(m), 1280 (s), 1268 (m), 1249 (m), 1205 (w), 1144 (s), 1082 (s), 909

(m), 868 (w), 813 (m), 797 (w), 773 (w), 730 (w), 712 (w), 703

(w), 676 (m), 663 (m), 650 (w). Anal. Found (calcd for

C₂₉H₄₁N₅O₄S₂Ti): C, 55.18 (54.79); H, 6.56 (6.50); N, 11.02

(10.70).

H₃N₃^ArFN (15). To a solution of tris(2-aminoethyl)amine

(2.34 g, 2.40 mL, 16.0 mmol) and NaOH (2.00 g, 34.2 mmol)

in H₂O (105 mL) was added dropwise a solution of 3,5-bis-

(105) Bradley, D. C.; Thomas, I. M. J. Chem. Soc. 1960, 3857.

(106) Diamond, G. M.; Jordan, R. F.; Petersen, J. L. Organometallics

1996, 15, 4030.

(107) Benzing, E.; Kornicker, W. Chem. Ber. 1961, 2263.

(108) Kempe, R.; Arndt, P. Inorg. Chem. 1996, 35, 2644.

(109) Collier, M. R.; Lappert, M. F.; Pearce, R. J. Chem. Soc., Dalton

Trans. 1973, 445.

Ti(N₂^MsN^Ph)(NMe₂)₂(20). A solution of H₂N₂^MsN^Ph(0.500 g,

1.43 mmol) in benzene (50 mL) was added dropwise to a solution

(110) Bradley, D. C.; Thomas, I. M. J. Chem. Soc. 1960, 3857.

(111) Herbert, B. Justus Liebigs Ann. Chem. 1950, 566, 210.

4186 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

of Ti(NMe₂)₄(0.231 g, 1.43 mmol) in benzene (30 mL), and the

mixture was stirred for 2.5 h. The volatiles were removed under

reduced pressure, and the residue was dissolved in toluene and

cooled to -80 °C to yield a red crystalline product, which was filte-

red and dried in vacuo. Yield: 0.30 g, (45%). ¹H NMR (CD₂Cl₂,

213 K, 499.9 MHz): δ 7.16 (2H, s, 3-C₆H₅), 7.08 (1H, s, 4-C₆H₅),

6.58 (2H, d, ³J = 8.3 Hz, 2-C₆H₅), 3.79 (2H, s, CH₂C₆H₅), 3.73

solution of Ti(NMe₂)₂(OⁱPr)₂(0.300 g, 1.18 mmol) in benzene

(20 mL). The resulting solution was stirred for 2 h, and volatiles

were removed under reduced pressure. The residue was ex-

tracted into benzene, and volatiles were removed under reduced

pressure. The resulting red solid was washed with hexane (3 ꢀ

15 mL) and dried in vacuo, yielding 23 as a red powder. Yield:

1

0.30 g, (56%). H NMR (CD₂Cl₂, 223 K, 449.9 MHz): δ 7.16

3

(2H, t, J = 6.1 Hz, CH₂NS), 3.36 (6H, s, NMe₂), 3.32 (6H, s,

(2H, m, 3-C₆H₅), 7.10 (1H, s, 4-C₆H₅), 6.77 (2H, d, ³J = 7.0 Hz,

2-C₆H₅), 5.23 (1H, q, ³J = 5.4 Hz, OCHMe₂), 5.14 (1H, q, ³J =

6.3 Hz, OCHMe₂), 4.11 (2H, s, CH₂C₆H₅), 3.72 (2H, q, ³J = 5.6

Hz, CH₂NS), 2.93 (2H, t, ³J = 6.9 Hz, NCH₂CH₂NS), 2.64 (6H,

s, SO₂Me), 2.21 (2H, t, ³J = 5.4 Hz, CH₂NS), 1.93 (2H, t, ³J =

NMe₂), 2.63 (2H, t, ³J = 6.1 Hz, CH₂NS), 2.47 (6H, s, SO₂Me),

1.95 (2H, t, ³J = 6.0 Hz, NCH₂CH₂NS), 1.68 (2H, t, ³J = 6.0 Hz,

NCH₂CH₂NS) ppm. ¹³C{¹H} (CD₂Cl₂, 213 K, 75.5 MHz): δ131.9

(2-C₆H₅), 131.4 (1-C₆H₅), 130.0 (4-C₆H₅), 129.0 (3-C₆H₅), 57.3

(CH₂C₆H₅), 51.4 (NMe₂), 48.9 (NCH₂CH₂NS), 44.1 (SO₂Me),

40.5 (CH₂NS). IR (KBr plates, Nujol mull cm^-1): 2924 (s), 2361

(m), 2338 (w), 1734 (w), 1684 (w), 1636 (m), 1559 (w), 1540 (w),

1521 (w), 1507 (m), 1377(m), 1260 (m), 805 (w), 666 (w), 477 (s).

Anal. Found (calcd for C₁₇H₃₃N₅O₄S₂Ti): C, 42.18 (42.23); H, 6.91

(6.88); N, 14.39 (14.49).

3

6.2 Hz, NCH₂CH₂NS), 1.47 (6H, d, J = 6.3 Hz, OCHMe₂),

1.16 (6H, d, ³J = 5.4 Hz, OCHMe₂) ppm. ¹³C{¹H} (CD₂Cl₂, 293

K, 75.5 MHz): δ 132.4 (1-C₆H₅). 132.0 (2-C₆H₅), 129.0 (4-

C₆H₅), 128.9, (3-C₆H₅), 83.5 (OCHMe₂), 58.6 (CH₂C₆H₅), 51.5

(NCH₂CH₂NS), 48.7 (SO₂Me), 39.4 (CH₂NS), 26.1 (OCHMe₂).

IR (KBr plates, Nujol mull, cm^-1): 3853 (m), 3838 (w), 3821 (s),

3801 (w), 3751 (m), 3745 (w), 3735 (m), 3676 (w), 3649 (m), 3629

(m), 3583 (w), 2361 (s), 2341 (m), 2275 (s), 2120 (m), 2050 (m),

1772 (w), 1740 (w), 1717 (w), 1700 (w), 1684 (w), 1670 (w), 1653

(w), 1636 (w), 1617 (w), 1570 (m), 1559 (w), 1541 (w), 1521 (s),

1507 (m), 1377 (s), 1260 (s), 1019 (s), 802 (m), 721 (w), 666 (s),

542 (s), 446 (s). EI-HRMS: m/z = 453.0570, ([M - OⁱPr]^þ;

calcd for C₁₉H₃₅N₃O₆S₂Ti: 453.0570), m/z = 348.0798 ([L]^þ,

calcd for C₁₉H₃₅N₃O₆S₂Ti 348.1130). Anal. Found (calcd for

C₁₉H₃₅N₃O₆S₂Ti): C, 44.35 (44.44); H, 6.91 (6.87); N, 8.14

(8.18).

Ti(N₂^MsN^OMe)(OⁱPr)₂(21). A solution of H₂N₂^MsN^OMe

(0.340 g, 0.0011 mol) in benzene (20 mL) was added dropwise

to a solution of Ti(NMe₂)₂(OⁱPr)₂(0.550 g, 0.0022 mol) in ben-

zene (20 mL). The resulting bright orange solution was stirred

for 2 h at room temperature, and volatiles were removed under

reduced pressure. The residue was extracted into toluene (50

mL) and cooled to -80 °C for 16 h. The resulting orange solu-

tion was filtered and dried in vacuo to yield 21 as an orange

powder. Yield: 0.23 g (44%). Red single crystals suitable for an

X-ray diffraction study were obtained by recrystallization from

Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24). To a solution of Ti(NMe₂)₂-

(OⁱPr)₂(0.840 g, 3.30 mmol) in dichloromethane (30 mL) was

added a solution of H₂N₂^TsN^Ph(0.500 g, 1.00 mmol) in dichloro-

methane (30 mL). The mixture was stirred for 16 h, resulting in

an orange solution. The solution was concentrated to ca. 15 mL,

and hexane (40 mL) was added to precipitate the product, which

was filtered. The orange solid was recrystallized from a con-

centrated dichloromethane solution layered with pentane to

yield 24 as an orange solid, which was washed with pentane

(3 ꢀ 20 mL) and dried in vacuo. The recrystallized product yielded

crystals suitable for an X-ray diffraction study. Yield: 0.48 g

(74%). ¹H NMR(CD₂Cl₂, 299.9 MHz, 248 K): δ7.73 (4H, d, ³J =

9.0 Hz, 2-C₆H₄Me), 7.38 (2H, m, 2-C₆H₅), 7.29 (4H, d, ³J = 9.0

Hz, 3-C₆H₄Me), 7.00 (3H, m,overlapping 3-C₆H₅and 4-C₆H₅),

5.16 (1H, sept., ³J = 6.1 Hz, OCHMe₂), 3.73 (2H, s, CH₂C₆H₅),

3.43 (10H, s, overlapping NMe₂and TsNCH₂CH₂N), 2.74

(4H, m, TsNCH₂CH₂N), 2.37 (6H, s, C₆H₄Me), 1.61 (6H, d,

³J = 6.1 Hz, OCHMe₂) ppm. ¹³C{¹H} NMR (CD₂Cl₂, 248 K,

75.4 MHz): δ 141.9 (1-C₆H₄Me), 139.6 (4-C₆H₄Me), 131.4 (4-

C₆H₅), 129.8 (1-C₆H₅), 129.2 (3-C₆H₄Me), 128.9 (4-C₆H₅), 128.5

(3-C₆H₅), 126.9 (2-C₆H₄Me), 83.1 (OCHMe₂), 54.7 (CH₂C₆H₅),

52.7 (NMe₂), 47.6 (TsNCH₂CH₂N), 47.5 (TsNCH₂CH₂N), 25.8

(OCHMe₂),21.3(C₆H₄Me) ppm. IR (KBr plates, Nujol mull cm^-1):

1623 (w), 1297 (s), 1168 (w), 1143 (m), 1085 (m), 1022 (m),

952 (s), 903 (w), 820 (w), 771 (w), 723 (m), 668 (m). Anal. Found

(calcd for C₃₀H₄₂N₄O₅S₂Ti): C, 55.68 (55.92); H, 6.51 (6.58); N, 8.61

(8.44).

1

a concentrated toluene-hexane solution (1:3) at -80 °C. H

NMR (C₆D₆, 353 K, 499.2 MHz): δ 5.01 (2H, sep, ³J = 6.1 Hz,

OCHMe₂), 3.19 (3H, s, OMe), 3.05 (2H, t, ³J = 5.1 Hz, CH₂NS),

2.77 (4H, app. q, J = 5.7 Hz, overlapping NCH₂CH₂NS and

3

CH₂CH₂OMe), 2.54 (6H, s, SO₂Me), 2.18 (2H, t, J = 5.5 Hz,

CH₂CH₂OMe), 1.27 (12H, d, ³J = 6.1 Hz, OCHMe₂) ppm.

¹³C{¹H} (C₆D₅CD₃, 353 K, 75.5 MHz): δ 70.5 (OCHMe₂), 63.0

(OMe), 59.0 (CH₂CH₂NS), 58.0 (CH₂CH₂OMe), 54 (CH₂OMe),

49.5 (SO₂Me), 39.5 (CH₂NS), 26.0 (OCHMe₂) ppm. IR (KBr

plates, Nujol mull, cm^-1): 1653 (w), 1559 (w), 1321 (s), 1363 (m),

1261 (w), 1162 (s), 1007 (m), 851 (w), 805 (w), 666 (w), 617 (w), 488

(m). EI-HRMS: m/z = 422.0889 ([M - OiPr]^þ; calcd for C₂₁H₂₉-

N₃O₇S₂Ti 422.0899). Anal. Found (calcd for C₂₁H₂₉N₃O₇S₂Ti): C,

37.51 (37.42); H, 7.27 (7.25); N, 8.69 (8.73).

Ti(N₂^TsN^Ph)(OⁱPr)₂(22). To a solution of H₂N₂^TsN^Ph(0.500

g, 1.00 mmol) in chlorobenzene (30 mL) was added Ti(NMe₂)₂-

(OⁱPr)₂(0.254 g, 1.00 mmol) in chlorobenzene (30 mL). The

resulting light yellow solution mixture was stirred for 2 h and

concentrated under reduced pressure to yield a yellow solid,

which was recrystallized from a concentrated dichloromethane

(10 mL) solution layered with pentane (30 mL) to yield 22 as a

light yellow solid, which was washed with pentane (3ꢀ20 mL)

1

and dried in vacuo. Yield: 0.31 g (47%). H NMR (CD₂Cl₂,

299.9 MHz, 246 K): δ 8.16 (4H, d, ³J = 9.2 Hz, 2-C₆H₄Me), 6.98

(4H, m, overlapping 3-C₆H₅and 4-C₆H₅), 6.95 (4H, d, ³J = 9.2

Hz, 3-C₆H₄Me), 6.70 (2H, d, J = 8.5 Hz, 2-C₆H₅), 5.51 (1H,

3

sept., J = 6.2 Hz, OCHMe₂), 5.31 (1H, sept., J = 5.9 Hz,

OCHMe₂), 4.15 (2H, s, CH₂C₆H₅), 3.34 (4H, app. t, ³J = 6.3 Hz,

TsNCH₂CH₂N), 2.22 (4H, app. t, ³J = 6.3 Hz, TsNCH₂CH₂N),

1.98 (6H, s, C₆H₄Me), 1.61 (12H, 2 ꢀ overlapping d, ³J = 6.2 and

5.9 Hz, OCHMe₂) ppm. ¹³C{¹H} NMR (CD₂Cl₂, 246 K, 75.4

MHz): δ 142.5 (1-C₆H₄Me), 139.0 (4-C₆H₄Me), 131.5 (4-C₆H₅),

131.0 (1-C₆H₅), 129.5 (3-C₆H₄Me), 129.0 (4-C₆H₅), 128.8 (3-

C₆H₅), 127.4 (2-C₆H₄Me), 84.4 (OCHMe₂), 83.4 (OCHMe₂),

57.1 (CH₂C₆H₅), 50.8 (TsNCH₂CH₂N), 47.7 (TsNCH₂CH₂N),

25.4 (OCHMe₂), 24.9 (OCHMe₂), 21.3 (C₆H₄Me) ppm. IR (KBr

plates, Nujol mull cm^-1): 2723 (w), 1298 (m), 1152 (s), 1109 (m),

1086 (m), 1028 (w), 997 (w), 946 (m), 906 (w), 814 (m), 772 (w), 732

(m), 667 (m). Anal. Found (calcd for C₃₁H₄₃N₃O₆S₂Ti): C, 55.63

(55.29); H, 6.39 (6.34); N, 6.19 (6.45).

Alternative NMR Tube Scale Synthesis of Ti(N₂^TsN^Ph)(OⁱPr)-

(NMe₂) (24). To a solution of Ti(N₂^TsN^Ph)(NMe₂)₂(10 mg, 15.7

μmol) in C₆D₆(0.2 mL) was added a solution of Ti(N₂^TsN^Ph)-

(OⁱPr)₂(10.5 mg, 15.7 μmol) in C₆D₆(0.2 mL). Initial ¹H NMR

indicated that no reaction had taken place. The reaction mixture

was then heated at 80 °C for 4 days. A color change of the

initially dark orange to light orange was observed. Analysis by

¹H NMR indicated that 24 had been formed quantitatively.

Zr(N₃^TsN)(CH₂SiMe₃) (30). To a solution of Zr(CH₂SiMe₃)₄

(0.520 g, 1.20 mmol) in THF (15 mL) was added a solution of

H₃N₃^TsN (0.710 g, 1.20 mmol) in THF (15 mL). The reaction

was stirred at RT for 16 h, after which time the resulting light

yellow solution was concentrated to ca. half the original volume

and cooled to -78 °C. The resulting white crystalline solid was

filtered, washed with pentane (3ꢀ15 mL), and dried in vacuo.

Ti(N₂^MsN^Ph)(OⁱPr)₂(23). A soultion of H₂N₂^MsN^Ph(0.370 g,

1.073 mmol) in benzene (50 mL) was added dropwise to a

Article

Organometallics, Vol. 29, No. 18, 2010 4187

Yield: 0.63 g (67%). Crystals suitable for X-ray diffraction

studies were obtained from slow evaporation of a THF solution

at RT. ¹H NMR (CD₂Cl₂, 299.9 MHz, 298 K): δ 7.90 (6H,

d, ³J = 6.4 Hz, 2-C₆H₄Me), 7.31(6H, d, ³J = 6.4Hz, 3-C₆H₄Me),

3.53 (6H, m, TsNCH₂CH₂N), 2.82 (6H, m, TsNCH₂CH₂N), 2.41

(9H, s, C₆H₄Me), 0.26 (2H, br s, CH₂Si), -0.30 (9H, s, SiMe₃) ppm.

¹³C{¹H} NMR (CD₂Cl₂, 75.4 MHz, 298 K): δ 141.9 (1-C₆H₄Me),

139.1 (4-C₆H₄Me), 129.8 (3-C₆H₄Me), 127.5 (2-C₆H₄Me), 60.8

(TsNCH₂CH₂N), 47.6 (TsNCH₂CH₂N), 21.5 (C₆H₄Me), 3.8

(CH₂Si), 2.4 (SiMe₃) ppm. IR (KBr plates, Nujol mull cm^-1):

1262 (s), 1161 (m), 1110 (m), 1019 (w), 968 (w), 814 (s), 668 (m),

635 (w). Anal. Found (calcd for C₂₉H₄₁N₅O₄S₂Ti): C, 47.42 (47.48);

H, 5.76 (5.66); N, 7.09 (7.14).

{Zr(N₃^ArFN)}₂(μ-O) (34). Zr(N₃^ArFN)(NMe₂) (33) (0.100 g,

0.100 mmol) was slurried three times in “wet” diethyl ether (5 ꢀ

15 mL) and dried in vacuo to yield 34 as a white solid. Yield: 0.12

g (61%). Crystals suitable for X-ray diffraction studies were

obtained from slow evaporation of a concentrated diethyl ether

solution at RT. ¹H NMR (C₆D₆, 499.9 MHz, 298 K): δ 8.79 (2H,

s, 2-C₆H₃(CF₃)₂), 8.60 (2H, s, 2-C₆H₃(CF₃)₂), 8.38 (2H, s, 2-

C₆H₃(CF₃)₂), 7.90 (1H, s, 4-C₆H₃(CF₃)₂), 7.82 (1H, s, 4-C₆H₃-

(CF₃)₂), 7.56 (1H, s, 4-C₆H₃(CF₃)₂), 3.80 (1H, m, SNCH₂-

CH₂N), 3.20 (2H, m, SNCH₂CH₂N), 2.77 (1H, m, SNCH₂CH₂-

N), 2.67 (1H, m, SNCH₂CH₂N), 2.53 (1H, m, SNCH₂CH₂N),

2.28 (1H, m, SNCH₂CH₂N), 2.06 (1H, m, SNCH₂CH₂N), 1.87

(1H, m, SNCH₂CH₂N), 1.52 (1H, m, SNCH₂CH₂N), 1.36 (1H,

m, SNCH₂CH₂N), 0.95 (1H, m, SNCH₂CH₂N) ppm. A satis-

factory ¹³C NMR spectrum could not be obtained due to the

insoluble nature of this compound. ¹⁹F NMR (C₆D₆, 125.8

MHz, 298 K): δ -128.2 (6F, s, CF₃), -128.1 (6F, s, CF₃), -127.9

(6F, s, CF₃) ppm. IR (KBr plates, Nujol mull cm^-1): 1281 (s),

1139 (s), 1006 (w), 992 (w), 966 (w), 905 (m), 844 (m), 808 (w),

722 (m), 698 (m), 682 (m), 648 (w). Anal. Found (calcd for

C₆₀H₄₂F₃₆N₈O₁₃S₆Zr): C, 35.50 (35.14); H, 1.93 (2.06); N, 5.19

(5.46).

Zr(N₃^MsN)(CH₂SiMe₃) (31). To a solution of Zr(CH₂SiMe₃)₄

(0.500 g, 0.100 mmol) in toluene (10 mL) was added a slurry of

H₃N₃^MsN (0.440 g, 0.100 mmol) in toluene (15 mL), and the

mixture was heated to 100 °C for 3 h, after which time the solu-

tion was cooled to RT and filtered. The resulting light yellow

solution was concentrated to ca. 5 mL, giving a white crystalline

solid, which was filtered and washed with pentane (3 ꢀ 15 mL)

1

and dried in vacuo. Yield: 0.35 g (60%). H NMR (CD₂Cl₂,

299.9 MHz, 298 K): δ 3.47 (6H, t, ³J = 6.0 Hz, SNCH₂CH₂N),

3.08 (9H, s, SO₂Me), 2.74 (6H, t, ³J = 6.0 Hz, SNCH₂CH₂N),

0.29 (2H, br s, CH₂Si), -0.30 (9H, s, SiMe₃) ppm. ¹³C{¹H}

NMR (CD₂Cl₂, 75.4 MHz, 298 K): 60.8 (SNCH₂CH₂N), 46.8

(SO₂Me), 40.7 (SNCH₂CH₂N), 3.1 (CH₂SiMe₃), 2.8 (SiMe₃)

ppm. IR (KBr plates, Nujol mull cm^-1): 1324 (w), 1249 (s), 1150

(w), 1075 (m), 982 (s), 863 (m), 936 (m), 761 (w), 722 (w), 697 (w),

668 (w). Anal. Found (calcd for C₁₃H₃₂N₄O₆S₃SiZr): C, 27.96

(28.09); H, 5.82 (5.80); N, 9.99 (10.08).

NMR Tube Scale Synthesis of Zr(N₃^TsN)(OⁱPr) (35). To a

solution of Zr(N₃^TsN)(CH₂SiMe₃) (30) (10.0 mg, 12.8 μmol) in

CD₂Cl₂(0.2 mL) was added a solution of ⁱPrOH (0.770 mg, 1.00

1

μL, 12.8 μmol) in CD₂Cl₂(0.2 mL). After 1 h analysis by H

NMR indicated that 35 had been formed quantitatively. ¹H

NMR (CD₂Cl₂, 299.9 MHz, 298 K): δ 9.97 (6H, d, ³J = 9.0 Hz,

3

2-C₆H₄Me), 7.29 (6H, d, J = 9.0 Hz, 3-C₆H₄Me), 4.40 (1H,

sept., ³J = 6.0, OCHMe₂) 3.39 (6H, t, ³J = 5.8, TsNCH₂CH₂N),

2.89 (6H, t, ³J = 5.8, TsNCH₂CH₂N), 2.42 (9H, s, C₆H₄Me), 0.98

(1H, sept., ³J = 6.0, OCHMe₂) ppm.

Zr(N₃^ArFN)(CH₂SiMe₃) (32). To a solution of Zr(CH₂SiMe₃)₄

(0.190 g, 0.200 mmol) in toluene (5 mL) was added a slurry of

H₃N₃^ArFN (0.400 g, 0.200 mmol) in toluene (15 mL), and the mixture

was heated to 100 °C for 2 h, after which time the light yellow solution

was cooled to 0 °C. A white crystalline product was isolated by

filtration and dried in vacuo. Yield: 0.21 g (89%). Crystals suitable for

X-ray diffraction studies were obtained from a concentrated benzene

NMR Tube Scale Synthesis of Zr(N₃^ArFN)(O-2,6-C₆H₃Me₂)

(36). To a solution of Zr(N₃^ArFN)(NMe₂) (33) (10 mg, 8.7 μmol)

in CD₂Cl₂(0.2 mL) was added a solution of 2,6-dimethylphenol

(1.1 mg, 8.7 μmol) in CD₂Cl₂(0.2 mL). After 1 h, analysis by ¹H

NMR indicated that 36 had been formed quantitatively. ¹H

NMR (CD₂Cl₂, 299.9 MHz, 298 K): δ 8.34 (6H, s, 2-C₆H₃-

(CF₃)₂), 7.51 (6H, s, 4-C₆H₃(CF₃)₂), 6.64 (2H, d, ³J = 9.0 Hz,

4-C₆H₃Me₂), 6.48 (2H, d, ³J = 9.0 Hz, 3-C₆H₃Me₂), 2.97 (6H, br

s, SNCH₂CH₂N), 2.21 (6H, br s, SNCH₂CH₂N), 2.14 (2H, s,

C₆H₃Me₂) ppm.

1

solution. H NMR (CD₂Cl₂, 299.9 MHz, 313 K): δ 8.52 (6H, s,

2-C₆H₃(CF₃)₂), 8.14 (3H, s, 4-C₆H₃(CF₃)₂), 3.69 (6H, br s,

SNCH₂CH₂N), 2.98 (6H, br s, SNCH₂CH₂N), 0.34 (2H, br s,

CH₂Si), 0.36 (9H, s, SiMe₃) ppm. ¹³C{¹H} NMR (CD₂Cl₂, 75.4

1

MHz, 313 K): 145.8 (1-C₆H₃(CF₃)₂), 133.5 (CF₃, J = 34.3 Hz),

127.8 (3-C₆H₃(CF₃)₂), 127.0 (2-C₆H₃(CF₃)₂), 125.0 (4-C₆H₃(CF₃)₂),

61.0 (SNCH₂CH₂N), 58.0 (SNCH₂CH₂N), 1.8 (SiMe₃) ppm. ¹⁹F

NMR (CD₂Cl₂, 282.1 MHz, 313 K): -61.5 (18F, s, CF₃) ppm. IR

(KBr plates, Nujol mull cm^-1): 3102 (w), 2726 (w), 1624 (w), 1365

(s), 1282 (s), 1193 (s), 1141 (s), 1084 (m), 1044 (w), 1007 (w), 994 (w),

970 (m), 937 (m), 908 (s), 845 (s), 823 (m), 808 (m), 747 (w), 720 (w),

698 (m), 682 (s), 651 (w), 644 (m). Anal. Found (calcd for

C₃₄H₃₂F₁₈N₄O₆S₃Zr):C,35.47(35.51);H,2.79(2.80);N,4.78(4.87).

Zr(N₃^ArFN)(NMe₂) (33). To a solution of Zr(NMe₂)₄(0.110 g,

0.400 mmol) in THF (15 mL) was added a solution of H₃N₃^ArFN

(0.400 g, 0.410 mmol) in THF (15 mL), and the mixture was

stirred at RT for 16 h, after which time the light yellow solution

was concentrated to ca. half its original volume and cooled to

-78 °C. The white crystalline product was isolated by filtration and

dried in vacuo. Yield: 0.37 g (82%). Crystals suitable for X-ray

diffraction studies were obtained from a concentrated THF solu-

tion. ¹H NMR (C₆D₆, 499.9 MHz, 298 K): δ 8.50 (6H, s, 2-C₆H₃-

(CF₃)₂), 7.63 (3H, s, 4-C₆H₃(CF₃)₂), 2.99 (6H, br s, SNCH₂CH₂N),

2.86 (6H, s, NMe₂), 2.22 (6H, br s, SNCH₂CH₂N) ppm. ¹³C{¹H}

NMR (CD₂Cl₂, 75.4 MHz, 298 K): 145.35 (1-C₆H₃(CF₃)₂), 133.1

(CF₃, ¹J = 33.9 Hz), 126.2 (3-C₆H₃(CF₃)₂), 125.2 (2-C₆H₃(CF₃)₂),

121.6 (4-C₆H₃(CF₃)₂), 60.45(SNCH₂CH₂N), 46.2 (SNCH₂CH₂N),

44.1 (NMe₂) ppm. ¹⁹F NMR (CD₂Cl₂, 282.1 MHz, 298 K): -62.7

(18F, s, CF₃) ppm. IR (KBr plates, Nujol mull cm^-1): 1365 (m),

1280(s),1191(m),1125(s),1053(w),1016(w),956(w),937(m),924

(m), 901 (m), 780 (s), 738 (w), 720 (w), 698 (m), 682 (m), 644 (w).

Anal. Found (calcd for C₃₂H₂₇F₁₈N₅O₆S₃Zr): C, 34.66 (34.72); H,

2.42 (2.46); N, 6.38 (6.33).

General Procedure for Polymerization of ε-CL. Parallel dupli-

cate experiments were carried out in each of which a solution of

ε-CL (6.6 mmol) in toluene (3.0 mL) was heated to 100 °C and

added to a solution of initiator (0.066 mmol) in toluene (3.8 mL)

also at 100 °C. For one sample aliquots were taken at the

respective time. Upon completion, the reaction was quenched

by addition of wet THF (10 mL) and the solution evaporated to

dryness to give the crude polymer. Isolated yields were obtained

from the parallel experiment for which the polymer was

quenched by wet THF (10 mL) and precipitated by addition

to ethanol (250 mL) with vigorous stirring, filtered, and dried to

constant weight in vacuo.

Procedure for Polymerization of ε-CL in the Presence of

Alcohol Co-initiators. Parallel duplicate experiments were car-

ried out in each of which to initiator (0.066 mmol) was added a

solution of HOR (0.066 mmol) in toluene (3.8 mL). This

solution was stirred for 15 min at room temperature, then heated

to 100 °C, and a solution of ε-CL (6.6 mmol) in toluene (3.0 mL)

also at 100 °C was added. For one sample aliquots were taken at

the respective time. Upon completion, the reaction was quenched

by addition of wet THF (10 mL) and the solution evaporated to

dryness to give the crude polymer. Isolated yields were obtained

from the parallel experiment for which the polymer was quenched

by wet THF (10 mL) and precipitated by addition to ethanol

(250 mL) with vigorous stirring, filtered, and dried to constant

weight in vacuo.

General Procedure for Solution Polymerization of rac-LA. rac-

LA (6.00 mmol) and initiator (0.06 mmol) were added to a

4188 Organometallics, Vol. 29, No. 18, 2010

Schwarz et al.

Schlenk flask and heated to 70 °C. To this was added hot (70 °C)

toluene (6.0 mL), rapidly dissolving both solids. The resultant

solution was heated at 70 °C, and aliquots were taken at the

respective time. Upon completion of the reaction, wet THF

(10 mL) was added and the solution evaporated to dryness to

give the poly(rac-LA). Conversions were determined by ¹H

NMR integration of the OCHMe resonance relative intensities

of the residual rac-LA and poly(rac-LA).

and equivalent reflections merged.¹¹²The structures were solved

with SIR92¹¹³or SHELXS-97,¹¹⁴and further refinements and all

other crystallographic calculations were performed using either

the CRYSTALS program suite¹¹⁵or SHELXS-97.¹¹⁶Other de-

tails of the structure solution and refinements are given in the

Supporting Information (CIF data). A full listing of atomic

coordinates, bond lengths and angles, and displacement para-

meters for all the structures has been deposited at the Cambridge

Crystallographic Data Centre. See Notice to Authors on the

Web.

Procedure for Solution Polymerization of rac-LA in the Pre-

sence of Alcohol Co-initiators. rac-LA (6.00 mmol) and initiator

(0.06 mmol) were added to a Schlenk flask and heated to 70 °C.

To this was added a hot (70 °C) solution of HOR (0.06 mmol) in

toluene (6.0 mL), rapidly dissolving both solids. The resultant

solution was heated at 70 °C, and aliquots were taken at the

respective time. Upon completion of the reaction, wet THF

(10 mL) was added and the solution evaporated to dryness to

give the poly(rac-LA). Conversions were determined by ¹H

NMR integration of the OCHMe resonance relative intensities

of the residual rac-LA and poly(rac-LA).

Acknowledgment. We thank the EPSRC for support

to A.D.S. and Dr. A. L. Thompson for advice with regard

to the crystallography. C.P. acknowledges MICINN

(MAT2007-60997) for financial support and CAM

ꢀ

and Universidad de Alcala for fellow funding.

Supporting Information Available: X-ray crystallographic

data in CIF format for the structure determinations of

Crystal Structure Determinations of H₂N₂^MsN^OMe(12),

SO₂Mes

H₂CyN₂

(17), Ti(N₂^TsN^Ph)(NMe₂)₂(19), Ti(N₂^MsN^OMe)-

SO₂Mes

H₂N₂^MsN^OMe(12), H₂CyN₂

(17), Ti(N₂^TsN^Ph)(NMe₂)₂

(OⁱPr)₂(21), Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24), Zr(N₃^TsN)(CH₂-

SiMe₃) (30), Zr(N₃^ArFN)(CH₂SiMe₃) (32), Zr(N₃^ArFN)(NMe₂)

(33), and {Zr(N₃^ArFN)}₂(μ-O) (34). X-ray data collection and

processing parameters are given in the Supporting Information.

Crystals were mounted on glass fibers using perfluoropolyether

oil and cooled rapidly in a stream of cold N₂using an Oxford

Cryosystems Cryostream unit. Diffraction data were measured

using an Enraf-Nonius KappaCCD diffractometer. As appro-

priate, absorption and decay corrections were applied to the data

(19), Ti(N₂^MsN^OMe)(OⁱPr)₂(21), Ti(N₂^TsN^Ph)(OⁱPr)(NMe₂) (24),

Zr(N₃^TsN)(CH₂SiMe₃) (30), Zr(N₃^ArFN)(CH₂SiMe₃) (32), Zr-

(N₃^ArFN)(NMe₂) (33), and {Zr(N₃^ArFN)}₂(μ-O) (34). Data con-

cerning the ROP catalysis; X-ray data collection and processing

parameters; additional crystallographic data (displacement ellipsoid

plots, selected distances and angles, and discussion) for 12, 17, 32,

and 34. This information is available free of charge via the Internet at

http://pubs.acs.org.

(114) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.

(115) Betteridge, P. W.; Cooper, J. R.; Cooper, R. I.; Prout, K.;

Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487.

(116) Sheldrick, G. M.; Schneider, T. R. Methods Enzymol. 1997, 277,

319.

(112) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction

Data Collected in Oscillation Mode; Academic Press: New York, 1997.

(113) Altomare, A.; Cascarano, G.; Giacovazzo, G.; Guagliardi, A.;

Burla, M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.

Article Doi

Ligand variations in new sulfonamide-supported group 4 ring-opening polymerization catalysts

DOI: 10.1021/om100508j

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om100508j

Full text of DOI:10.1021/om100508j

Products guided by the article

R&D Labs maybe for 1245029-78-0

Relevant to this article

Hot Product