Inorganica Chimica Acta 380 (2012) 213

Contents lists available at SciVerse ScienceDirect

Inorganica Chimica Acta

journal homepage: www.elsevier.com/locate/ica

Well-dened, solvent-free cationic barium complexes: Synthetic strategies and catalytic activity in the ring-opening polymerization of lactide
Bo Liu, Thierry Roisnel, Yann Sarazin
UMR CNRS 6226 Sciences Chimiques de Rennes Universit de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France

a r t i c l e

i n f o

a b s t r a c t
Well-dened, solvent-free cationic barium complexes of the type [{LnX}Ba]+[H2N{B(C6F5)3}2] stabilized by multidentate amino-ether phenoxide or uorinated amino-ether alkoxide ligands {LnX} are available according to original, general and high-yield protocols. These cations have been prepared by (i) hydrolysis of heteroleptic complexes {LnX}BaN(SiMe2H)2 stabilized by BaHSi interactions with [H(OEt2)2]+[H2N{B(C6F5)3}2], or (ii) reaction of {Ba[N(SiMe2H)2]2}n with the doubly acidic pro-ligands [{LnX}HH]+[H2N{B(C6F5)3}2]. The solid-state structures of [{LO2}Ba(THF)2]+[H2N{B(C6F5)3}2] ({LO2} H = 2-[(1,4,7,10-tetraoxa-13-azacyclopentadecan-13-yl)methyl]-4,6-di-tert-butylphenol) and [{RO2}Ba]+ [H2N{B(C6F5)3}2] {RO2}H = 2-[(1,4,7,10-tetraoxa-13-azacyclopentadecan-13-yl)methyl]-1,1,1,3,3,3hexauoropropan-2-ol) are described, highlighting the key role of internal secondary BaFC interactions in these highly electrophilic species. In combination with an excess of an external nucleophile (chosen from benzyl alcohol, 1,3-propanediol, benzyl amine or an hydroxyl-functionalized alkoxy-amine) as a co-initiator, some of these Ba cations provide extremely efcient catalysts for the immortal ring-opening polymerization of L-lactide in the temperature range 030 C, converting rapidly up to 5000 equiv. of monomer in a controlled fashion and with excellent end-group delity. 2011 Elsevier B.V. All rights reserved.

Article history: Available online 16 September 2011 Young Investigator Award Special Issue Keywords: Barium Well-dened cations Secondary interactions Ring-opening polymerization

1. Introduction The polymerization of enantiomerically pure L-lactide (L-LA), a fully bio-resourced monomer derived from the fermentation of sugars or starch [1], has attracted a great deal of attention since the turn of the century, not least because poly(L-lactide) (PLLA) is a biodegradable thermoplastic with mechanical properties comparable to those of polystyrene [2]. A broad range of well-dened {LnX}MetNu complexes (where {LnX}n is a bulky ancillary ligand and Nu is a reactive nucleophilic group such as alkyl, amide or alkoxide) have been developed for the controlled ring-opening polymerization (ROP) of L-LA [3,4]. Many initiators based on zinc [5], aluminum [6] or group III and lanthanide metals [7] allow for the living ROP of L-LA, as well as that of other cyclic esters such as rac-lactide (the equimolar mixture of the D- and L-isomers of lactide), e-caprolactone and b-butyrolactone [8], and a good understanding of ROP by coordination-insertion mechanism was gained through the use of these initiators. Magnesium complexes are also known to be competent ROP initiators [3,5b,9], but because of their higher sensitivity they are comparatively less common than their Zn analogues. The development of catalytic systems for the immortal ROP (iROP) of cyclic esters, rst proposed by Inoue for the ROP of epoxides [10], represents one of the persist Corresponding author. Fax: +33 (0)2 23 23 69 39.
E-mail address: yann.sarazin@univ-rennes1.fr (Y. Sarazin). 0020-1693/$ - see front matter 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.ica.2011.09.020

ing challenges in this eld: whereas a living system generates only one polymer chain per metal, the use of an excess of external protic co-initiator (typically an alcohol) with the metal initiator enables the production of hundreds of polymer chains per metal center [11]. Besides, we and others have recently demonstrated that cationic well-dened complexes, with their exacerbated Lewis acidity, promoted the polymerization of cyclic esters with great efcacy and were worth considering as a new generation of ROP catalysts [1216]. In stark contrast with Zn or even Mg, only a handful of efcient ROP single-site initiators based on the larger alkaline-earth metals (calcium, strontium and barium) have been reported to date [5f,1618]. Chisholm et al. pioneered the rst effective calcium initiators supported by highly encumbered tris(pyrazolyl)borate and b-diketiminate ancillary ligands [17cd], while Feijen et al. showed that Ca[N(SiMe3)2]2(THF)2 [18a] and the dimeric {Ca(THF)}2 (thmd)2(l-thmd)(l-N(SiMe3)2 (thmd-H = tetramethylheptanedione) [17a] also constituted moderately active initiators. Hill and co-workers reported the syntheses of stable heteroleptic bis(phosphinimino)methyl derivatives of Ca and Sr which showed promising ROP catalytic ability [17b], but surprisingly they did not elaborate on these initial results. If two other examples of ill-dened Sr compounds able to promote the ROP of cyclic esters have been reported [19], the only molecular Ba ROP initiator known until very recently was the aminebis(phenolate) complex disclosed by Davidson et al. [18d].

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

Such paucity of well-dened Ae-based initiators (Ae = Ca, Sr, and Ba) is undoubtedly related to the difculties encountered in taming the high reactivity of these very large (rionic = 1.00, 1.18 and 1.35 for Ca, Sr and Ba, respectively) [20] and electropositive metals. This is especially the case with the largest element, barium. Due to their high kinetic lability, the preparation of stable heteroleptic complexes {LnX}BaNu (where {LnX} is a monoanionic ancillary ligand) is very troublesome, as their synthesis is often hampered by deleterious Schlenk-type equilibria. Starting from {LnX}BaNu, these side-reactions lead to the formation of aggregates and poorly reactive species such as {LnX}2Ba and {BaNu2}n. As the pronounced ionic nature of the bonding increases with the ionic radius of the element, the propensity for ligand scrambling is most detrimental in the case of Ba complexes. For instance, heteroleptic complexes of Ca and Sr supported by bis(phosphinimino)methyl [17b] or b-diketiminate ligands [21] are stable, but the same is not true of their Ba parents. Efcient ROP initiators based on Ae metals (and especially Ba) can only be developed if convenient strategies aimed at stabilizing heteroleptic complexes against ligand scrambling are devised. Although steric bulk can sometimes impart sufcient stability to the complexes [17b,22], it is hard to rationalize it for all ligand frameworks, especially as the size of the metal varies considerably from Ca to Ba. Hill and co-workers reported an interesting strategy based on the dearomatization of bulky aromatic ligands [23]. Methods relying on stabilization by internal secondary interactions seemed to us to offer real scope as a general way to stabilize labile Ae species. We have recently described Ae[N(SiMe2H)2]2(THF)x homoleptic precursors (Ae = Ca, x = 1; Sr, x = 0.66; Ba, x = 0), and showed that the presence of internal stabilizing b-SiH agostic interactions constituted a key factor in the isolation of the heteroleptic complexes {LnX}AeN(SiMe2H)2 [17h]. Besides, we have also shown that the extremely electrophilic cations in {RO}Ae+X ion pairs (where {RO} is a tertiary uorinated alkoxide and X is a weakly-coordinating anion) were stabilized by internal AeFC secondary interactions in the solid-state [16c]. In the present study, the syntheses and solid-state structures of discrete cationic complexes of barium supported by bulky aminoether phenoxide and alkoxide ligands are described. A synthetic strategy for the isolation of such cations is detailed, and cases of BaHSi and BaFC secondary interactions in charge-neutral and cationic complexes are discussed. The remarkable catalytic activity of these cations in the immortal ROP of L-LA is also presented.

standards in the range of 580380 000 g mol1. The molecular weights of all poly(lactide)s were corrected by the recommended factor of 0.58 [24]. MALDI-ToF-MS spectra were obtained with a Bruker Daltonic MicroFlex LT, using a nitrogen laser source (337 nm, 3 ns) in linear mode with a positive acceleration voltage of 20 kV. Samples were prepared as follow: 1 lL of a 2:1 mixture of a saturated solution of a-cyano-4-hydroxycinnamic acid (Bruker Care) in HPLC quality acetonitrile and a 0.1% solution of triuoroacetic acid in ultrapure water was deposited on the sample plate. After total evaporation, 1 lL of a 5 to 10 mg mL1 solution of the polymers in HPLC-quality THF were deposited. Bruker Care Peptide Calibration Standard and Protein Calibration Standard I were used for external calibration. Elemental analyses were performed on a Carlo Erba 1108 Elemental Analyser instrument at the London Metropolitan University by Stephen Boyer and were the average of a minimum of two independent measurements. FTIR spectra were recorded at room temperature as Nujol mulls in KBr plates on a Shimadzu Afnity-IR spectrometer. 2.2. Materials Benzyl alcohol (VWR) and 1,3-propanediol (Acros) were dried and distilled over dry magnesium turnings and then stored over activated 3 molecular sieves. Benzyl amine (Acros) was distilled from CaH2 and kept over molecular sieves. BaI2 (anhydrous beads, 99.995%) was purchased from Aldrich and used as received. 1-Aza15-crown-5 (IBC) and 3,3,3-triuoro-2-(triuoromethyl)-1,2-propenoxide (Apollo) were used without purication. HN(SiMe3)2 (Acros) and HN(SiMe2H)2 (ABCR) were dried over activated 3 molecular sieves and distilled under reduced pressure prior to use. Technical grade L-lactide (provided by Total Petrochemicals) was puried by recrystallization from a hot, concentrated iPrOH solution (80 C), followed by two subsequent recrystallizations in hot toluene (105 C). After purication, L-lactide was stored at 30 C under the inert atmosphere of the glove-box. Toluene was distilled under Argon from melted sodium prior to use. THF was rst pre-dried over sodium hydroxide and distilled under argon over CaH2, and then freshly distilled a second time under argon from Na/benzophenone prior to use. Et2O, dichloromethane and pentane were distilled under argon from Na/benzophenone, CaH2 and Na/benzophenone/tetraglyme, respectively. All deuterated solvents (Eurisotop, Saclay, France) were stored in sealed ampoules over activated 3 molecular sieves and were thoroughly degassed by several freezethaw cycles. The synthetic precursors [H(OEt2)2]+[H2N{B(C6F5)3}2] [25], Ba[N(SiMe3)2]2(THF)2 [26], [{LO1}HH]+[H2N{B(C6F5)3}2] [16c] and {Ba[N(SiMe2H)2]2}n [17h], the complexes [{LO2}Ba]+ [H2N{B(C6F5)3}2] (5) [16a] and [{RO2}Ba]+ [H2N{B(C6F5)3}2] (6) [16c], and the pro-ligands {LO1}H [5j], {LO2}H [5j] and {RO2}H [16c] were all prepared as described elsewhere. 2.3. Syntheses and characterization 2.3.1. Synthesis of {LO1}BaN(SiMe3)2 (1) At room temperature, a solution of {LO1}H (0.25 g, 0.71 mmol) in Et2O (5 mL) was slowly added to a solution of Ba[N(SiMe3)2]2(THF)2 (0.45 g, 0.75 mmol) in Et2O (10 mL). A white precipitate formed instantly. The suspension was stirred for 2 h, and the supernatant was eliminated by ltration to afford 1 as white powder after drying in vacuo. Yield 0.38 g (83%). 1H NMR (C6D6, 298 K, 500.13 MHz): d 7.54 (d, 4JHH = 2.2 Hz, 1H, m-H), 7.06 (d, 4 JHH = 2.2 Hz, 1H, m-H), 3.30 (s, 2H, ArCH2N), 3.582.89 (br, 10H, CH2CH2O and 6H, OCH3), 2.21 (br, 4H, NCH2CH2), 1.69 (s, 9H, o-C(CH3)3), 1.42 (s, 9H, p-C(CH3)3), 0.14 (d, 4JHH = 3.0 Hz, 18H, Si(CH3)3) ppm. 13C{1H} NMR (C6D6, 298 K, 125.76 MHz): d

2. Experimental 2.1. General procedures All manipulations were performed under inert atmosphere using standard Schlenk techniques or in a Jacomex glove-box (O2 < 1 ppm, H2O < 5 ppm) for catalyst loading. NMR spectra were recorded on Bruker AC-300, AC-400 and AM500 spectrometers. All chemicals shifts were determined using residual signals of the deuterated solvents and were calibrated versus SiMe4. Assignment of the signals was carried out using 1D (1H, 13 C{1H}) and 2D (COSY, HMBC, and HMQC) NMR experiments. Coupling constants are given in Hertz. 19F{1H} chemical shifts were determined by external reference to an aqueous solution of NaBF4. 11 B chemical shifts are reported relative to BF3Et2O. Size Exclusion Chromatography (SEC) measurements were performed on a Polymer Laboratories PL-GPC 50 instrument equipped with a PLgel 5 MIXED-C column and a refractive index detector. The GPC column was eluted with THF at room temperature at 1 mL/min and was calibrated using 11 monodisperse polystyrene

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

166.7 (i-C), 138.2 (o-C), 136.8 (p-C), 129.5 (o-C), 125.8 (m-C), 121.6 (m-C), 69.6 (CH2CH2O), 59.8 (ArCH2N), 59.2 (OCH3), 54.6 (N CH2CH2), 36.1 (o-C(CH3)3), 34.8 (p-C(CH3)3), 32.5 (p-C(CH3)3),30.6 (o-C(CH3)3), 6.3 (Si(CH3)3) ppm. Repeated attempts to obtain X-ray quality crystals of 1 by recrystallization in benzene were unsuccessful, but yielded instead micro-crystals of {LO1}2Ba (2). Attempts to obtain satisfactory elemental analysis for 1 were unsuccessful, most probably because of contamination with various amounts of 2. 2.3.2. Synthesis of {LO1}2Ba (2) Compound 2 was prepared in the same fashion as that described for 1 by reaction of Ba[N(SiMe3)2]2(THF)2 (0.13 g, 0.21 mmol) with {LO1}H (0.15 g, 0.43 mmol). Yield 0.18 g (99%). Single-crystals of 2[C6H6]0.5 suitable for X-ray diffraction crystallography were obtained by recrystallization from a concentrated C6H6 solution at room temperature. 1H NMR (C6D6, 298 K, 500.13 MHz): d 7.59 (d, 4JHH = 2.7 Hz, 2H, m-H), 7.07 (d, 4 JHH = 2.7 Hz, 2H, m-H), 3.30 (s, 4H, ArCH2N), 3.10 (m, 4H, CH2C(H)HO), 3.04 (m, 4H, CH2CH(H)O), 3.01 (s, 12H, OCH3), 2.30 (br, 4H, NC(H)HCH2), 2.22 (br, 4H, NCH(H)-CH2), 1.80 (s, 18H, o-C(CH3)3), 1.55 (s, 18H, p-C(CH3)3) ppm. 13C{1H} NMR (C6D6, 298 K, 125.76 MHz): d 167.2 (i-C), 136.7 (o-C), 130.4 (p-C), 127.6 (o-C), 124.0 (m-C), 122. 6 (m-C), 69.5 (CH2CH2O), 59.6 (ArCH2N), 59.2 (OCH3), 53.3 (NCH2CH2), 36.2 (o-C(CH3)3), 34.5 (p-C(CH3)3), 33.1 (p-C(CH3)3), 30.8 (o-C(CH3)3) ppm. Anal. Calc. for C42H72BaN2O6 (838.89 g mol1): C, 60.2; H, 8.7; N, 3.3. Found: C, 60.1; H, 8.8; N, 3.3%. 2.3.3. Synthesis of {LO1}BaN(SiMe2H)2 (3) A solution of {LO1}H (0.20 g, 0.57 mmol) in Et2O (5 mL) was added slowly at room temperature to a solution of {Ba[N(SiMe2H)2]2}n (0.24 g, 0.60 mmol) in Et2O (10 mL). A white precipitate formed slowly upon stirring at room temperature. The suspension was stirred for 6 h, and the precipitate was then isolated by ltration. Following drying to constant weight, 3 was obtained as a colorless powder. Yield 0.28 g (79%). Recrystallization from a concentrated benzene solution afforded X-ray quality crystals of 3. 1H NMR (C6D6, 298 K, 500.13 MHz): d 7.60 (d, 4JHH = 2.7 Hz, 1H, m-H), 7.06 (d, 4JHH = 2.7 Hz, 1H, m-H), 4.72 (m, 1JSiH = 160 Hz, 2H, Si(CH3)2H), 3.30 (s, 2H, ArCH2N), 3.09 (m, 2H, CH2C(H)HO), 3.04 (m, 2H, CH2CH(H)O), 3.00 (s, 6H, OCH3), 2.30 (m, 2H, N C(H)HCH2), 2.22 (m, 2H, NCH(H)CH2), 1.80 (s, 9H, o-C(CH3)3), 1.55 (s, 9H, p-C(CH3)3), 0.11 (d, 4JHH = 3.0 Hz, 12H, Si(CH3)2H) ppm. 13C{1H} NMR (C6D6, 298 K, 125.76 MHz): d 166.9 (i-C), 136.5 (o-C), 130.7 (p-C), 127.2 (o-C), 123.8 (m-C), 122.5 (m-C), 69.5 (CH2CH2O), 59.3 (ArCH2N), 59.2 (OCH3), 53.4 (NCH2 CH2), 36.0 (o-C(CH3)3), 34.3 (p-C(CH3)3), 32.6 (p-C(CH3)3), 30.9 (oC(CH3)3), 5.3 (Si(CH3)2H) ppm. 29Si{1H} NMR (C6D6, 298 K, 79.49 MHz): d 30.4 ppm. FTIR (Nujol, KBr plates): m = 1987 (br s), 1910 (w sh), 1804 (w), 1770 (w), 1686 (w), 1604 (m) cm1. Anal. Calc. for C25H50BaN2O3Si2 (620.20 g mol1): C, 48.4; H, 8.1; N, 4.5. Found: C, 48.5; H, 8.2; N, 4.6%. 2.3.4. Synthesis of [{LO1}Ba]+[H2N{B(C6F5)3}2] (4) A solution of [{LO1}HH]+[H2N{B(C6F5)3}2] (0.40 g, 0.29 mmol) in Et2O (5 mL) was added dropwise at room temperature to a solution of {Ba[N(SiMe2H)2]2}n (0.12 g, 0.30 mmol) in Et2O (10 mL). The resulting colorless solution was stirred at room temperature for 2 h. The volatiles were then removed under vacuum to yield an
1 Ion pairs such as 46 can usually be obtained free of Et2O or THF by dissolving them in CH2Cl2 and re-precipitating them by very slow addition of pentane upon vigorous stirring, see Ref. [16a]. However, this method proved ineffective in the case of 4, as the yields were then dramatically decreased (from 93% to ca. 2030%) and the product still contained traces of Lewis base.

off-white solid. Repeated washings with pentane (5 20 mL) followed by stripping with CH2Cl2 (5 5 mL) yielded [{LO1}Ba]+[H2N{B(C6F5)3}2] (4) as a white powder which always contained residual Et2O; attempts to obtain the complex free of solvent were thwarted by the solubility properties of 4.1 Yield 0.43 g (93%). Alternatively, 4 could be obtained upon treatment of 3 (0.12 g, 0.19 mmol) with [H(OEt2)2]+[H2N{B(C6F5)3}2] (0.23 g, 0.19 mmol) in Et2O (15 mL). 1H NMR (CD2Cl2, 298 K, 500.13 MHz): d 7.48 (d, 4JHH = 2.5 Hz, 1H, arom. H), 7.14 (d, 4JHH = 2.5 Hz, 1H, arom. H), 5.70 (br, 2H, NH2), 4.212.53 (m, 16H, ArCH2N, NCH2CH2, CH2O and OCH3), 1.53 (s, 9H, C(CH3)3), 1.30 (s, 9H, C(CH3)3) ppm. 13C{1H} NMR (CD2Cl2, 298 K, 125.76 MHz): d 148.8, 147.0, 140.0, 138.1, 137.6, 135.6 (all C6F5), 157.6 (i-C), 139.9 (p-C), 137.2 (o-C), 128.5 (m-C), 126.0 (m-C), 124.0 (o-C), 71.6, 62.1, 60.0, 59.1, 58.3, 51.6 (all br., weak signals, ArCH2N, NCH2CH2, CH2O and OCH3), 35.6 (C(CH3)3), 34.4 (C(CH3)3), 31.6 (br, C(CH3)3) ppm. 19 1 F{ H} NMR (CD2Cl2, 188.29 MHz, 298 K): d 132.8 (d, 3JFF = 18.9, 12F, o-F), 160.2 (t, 3JFF = 18.9, 6F, p-F), 165.7 (d, 3JFF = 18.9, 12F, m-F) ppm. 11B NMR (CD2Cl2, 96.29 MHz, 298 K): d 8.4 ppm. Satisfactory elemental analysis for C57H38B2BaF30N2O3 (1527.84 g mol1) could not be obtained in a reproducible fashion as the samples submitted contained various amounts of residual Et2O. 2.3.5. Synthesis of [{LO2}Ba(THF)2]+[H2N{B(C6F5)3}2] (5(THF)2) 5(THF)2 was synthesized quantitatively by dissolving 5 (0.28 g, 0.17 mmol) in THF (3 mL), stirring at room temperature for 10 min followed by removal of the solvent under vacuum. Yield 0.30 g (99%). X-ray quality crystals of 5(THF)2 were obtained by recrystallization in CH2Cl2 and pentane in the presence of a small amount of THF at room temperature. 1H NMR (CD2Cl2, 298 K, 500.13 MHz): d 7.42 (d, 4JHH = 2.2 Hz, 1H, arom. H), 7.08 (d, 4JHH = 2.2 Hz, 1H, arom. H), 5.70 (br, 2H, NH2), 4.37 (d, 2JHH = 11.3 Hz, 1H, Ar CH(H)N), 3.96 (m, 2H, Hmacrocycle), 3.83 (m, 1H, Hmacrocycle), 3.71 (br, 8 + 3H, THF + Hmacrocycle), 3.58 (m, 1H, Hmacrocycle), 3.533.32 (m, 6H, Hmacrocycle), 3.26 (d, 2JHH = 11.5 Hz, 1H, ArC(H)HN), 3.213.08 (m, 3H, Hmacrocycle), 2.87 (m, 1H, Hmacrocycle), 2.76 (m, 1H, Hmacrocycle), 2.63(m, 1H, Hmacrocycle), 2.36 (m, 1H, Hmacrocycle), 1.85 (br, 8H, THF), 1.52 (s, 9H, C(CH3)3), 1.29 (s, 9H, C(CH3)3) ppm. 13C{1H} NMR (CD2Cl2, 298 K, 125.76 MHz): d 149.5, 147.6, 140.7, 138.3, 138.1, 136.3 (all C6F5), 158.3 (i-C), 138.7 (p-C), 138.0 (o-C), 127.8 (m-C), 126.0 (m-C), 124.3 (o-C), 70.8, 70.0, 69.8, 69.6, 69.5, 69.5, 68.8, 67.1, 59.3 (all Cmacrocycle), 68.4 (O-CH2 (THF)), 64.0 (ArCH2N), 35.6 (C(CH3)3), 34.8 (C(CH3)3), 34.4 (C(CH3)3), 31.9 (C(CH3)3), 26.1 (OCH2CH2 (THF)) ppm. 19F{1H} NMR (CD2Cl2, 188.29 MHz, 298 K): d 133.0 (d, 3JFF = 18.9, 12F, o-F), 160.1 (t, 3 JFF = 18.9, 6F, p-F), 165.6 (d, 3JFF = 18.9, 12F, m-F) ppm. 11B NMR (CD2Cl2, 96.29 MHz, 298 K): d 8.4 ppm. Anal. Calc. for C69H60BaF30N2O7 (1758.15 gmol1): C, 47.1; H, 3.4; N, 1.6. Found: C, 47.4; H, 3.3; N, 1.8%. 2.3.6. Re-crystallization of [{RO2}Ba]+[H2N{B(C6F5)3}2] (6) Single-crystals of 6 [16c] suitable for X-ray studies were obtained by very slow diffusion (over a period of several weeks) at room temperature of vapors of pentane into a solution of 6 in CH2Cl2 (using two Schlenk vessels connected by a long U-shape glass tube, the former containing the solution in CH2Cl2 while the latter contained dry pentane). 2.3.7. Typical polymerization procedure In the glove-box, the metal-based initiators and the puried monomer were placed at once in a Schlenk ask. The vessel was sealed and removed from the glove-box. All subsequent operations were carried out using standard Schlenk techniques. The required amount of dry, degassed solvent was added with a syringe to the Schlenk ask containing the initiator and monomer. The nucleophilic co-initiator was then added rapidly, the Schlenk vessel was

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

immerged in a water or oil bath set at the desired temperature and the polymerisation time was measured from this point. The reaction was terminated by addition of acidied MeOH (HCl, 1%) and the polymer was precipitated in methanol. It was puried by reprecipitation, using dichloromethane or THF as solvent and methanol as a non-solvent. The polymer was then dried to constant weight under dynamic vacuum. 2.3.8. X-ray diffraction crystallography Suitable crystals for X-ray diffraction analysis of 2[C6H6]0.5, 3, 5(THF)2 and 6 (CCDC numbers 826134826137) were obtained by recrystallization of the puried products. Diffraction data were collected at 150 K on a Bruker APEX CCD diffractometer with graphite-monochromated Mo Ka radiation (k = 0.71073 ). A combination of x and U scans was carried out to obtain at least a unique data set. The crystal structures were solved by direct methods, remaining atoms were located from difference Fourier synthesis followed by full-matrix least-squares renement based on F2 (programs SIR97 and SHELXL-97) [27]. The hydrogen atom contributions were calculated but not rened. All non-hydrogen atoms were rened with anisotropic displacement parameters. Summary of crystal and renement data for compounds 2[C6H6]0.5, 3, 5(THF)2 and 6 are collected in Table 1.

Fig. 1. Pro-ligands employed in this study.

3. Results and discussion 3.1. Synthesis of neutral and cationic complexes Discrete cationic complexes have recently emerged as a new family of highly active catalysts for the ROP of cyclic esters [12 16], and our group has for instance developed various cations of the large Ae metals (Ae = Ca, Sr, and Ba) [16a,c]. We have reported the syntheses of the Ba species [{LO2}Ba]+[H2N{B(C6F5)3}2] (5; {LO 2 }H = 2-[(1,4,7,10-tetraoxa-13-azacyclopentadecan-13-yl)methyl]-4,6-di-tert-butylphenol) and [{RO2}Ba]+[H2N{B(C6F5)3}2] (6; {RO2}H = 2-[(1,4,7,10-tetraoxa-13-azacyclopentadecan-13-yl)-

methyl]-1,1,1,3,3,3-hexauoropropan-2-ol), but their X-ray structures remained elusive. Both ligands contain a highly chelating aza-crown-ether fragment, but while {LO2} is based on a bulky phenoxide, the {RO2} platform is supported by a tertiary uorinated alkoxide (Fig. 1). Because of the electron-withdrawing substituents in a position to the oxygen atom, alkoxides such as {RO2} are far less basic and p-donating, and therefore have a signicantly lower bridging ability than conventional alkoxides which readily lead to the formation of aggregated and poorly-dened species [28]. However, both {LO2} and {RO2} are highly chelating, and the preparation of the less hindered [{LO1}Ba]+[H2N{B(C6F5)3}2] (4; {LO 1 }H = 2-{[bis(2-methoxyethyl)amino]methyl}-4,6-di-tertbutylphenol) was carried out, with the rationale that 4 would display higher catalytic activities than its derivatives 5 and 6 (Scheme 1) on account of the lower coordination number and higher Lewis acidity of the metal center in 4. The reaction of {LO1}H with Ba[N(SiMe3)2]2(THF)2 in Et2O yielded the heteroleptic {LO1}BaN(SiMe3)2 (1) in good yield. The complex was characterized by 1H and 13C{1H} NMR spectroscopy, which indicated the presence of some contamination by the homoleptic derivative {LO1}2Ba (2), and indeed only crystals of the less

Table 1 Crystal and renement data for compounds 2[C6H6]0.5, 3, 5(THF)2 and 6. 2[C6H6]0.5 Empirical formula Formula weight Crystal system Space group a () b () c () a () b () c () Volume (3) Z Dcalc (g cm3) Abs. Coeff. (mm1) F(0 0 0) Crystal size, mm h Range () Limiting indices C42H72BaN2O6, 0.5(C6H6) 877.41 monoclinic P21/n 11.9756(3) 14.7336(3) 26.2381(5) 90 92.8460(10) 90 4623.84(17) 4 1.260 0.904 1852 0.3 0.16 0.1 1.5527.47 15 < h < 10 19 < k < 16 34 < l < 33 0.0283 38 079 10 537 10537/6/480 1.181 0.0288 (0.0369) 0.0787 (0.0954) 1.509 and 1.059 3 C25H49BaN2O3Si2 619.18 triclinic  P1 11.3434(3) 12.5243(3) 12.6446(3) 62.8210(10) 80.6770(10) 89.2680(10) 1573.10(7) 2 1.307 1.364 642 0.25 0.24 0.23 1.8227.47 13 < h < 14 16 < k < 16 16 < l < 16 0.0241 21 577 7137 7137/1/309 1.193 0.0299 (0.0333) 0.0861 (0.1007) 2.883 and 1.303 5(THF)2 C69H60B2BaF30N2O7 1758.15 triclinic  P1 9.9479(8) 19.7011(18 20.266(2) 89.753(6) 80.164(6) 82.094(5) 3875.5(6) 2 1.507 0.635 1760 0.15 0.15 0.1 1.0227.52 12 < h < 12 25 < k < 25 26 < l < 25 0.0653 42 387 17 484 17484/0/1010 0.959 0.0605 (0.1037) 0.1583 (0.1785) 0.796 and 1.270 6 C50H24B2BaF36N2O5 1575.67 monoclinic P21/c 12.5284(5) 20.9798(11) 20.9607(9) 90 92.238(2) 90 5505.2(4) 4 1.901 0.894 3072 0.6 0.6 0.6 1.3727.47 15 < h < 16 27 < k < 27 27 < l < 27 0.0445 72 870 12 562 12562/0/827 1.048 0.0460 (0.0585) 0.1192 (0.1309) 2.058 and 1.353

Rint Reections collected Unique reections [I > 2r(I)] Data/restraints/parameters Goodness-of-t (GOF) on F2 R1 [I > 2r(I)] (all data) wR2 [I > 2r(I)] (all data) Largest difference (e A3)

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

Scheme 1. Syntheses of complexes 3 and 4.

soluble 2 were obtained upon attempts to recrystallize 1 from benzene. Also, slow decomposition of 1 was observed in solution, leading to the formation of 2 and an unidentied barium-amide species. Compound 1 was fully soluble in Et2O and THF, moderately so in toluene and benzene, and insoluble in aliphatic hydrocarbons. The homoleptic 2 could be independently prepared quantitatively upon treatment of Ba[N(SiMe3)2]2(THF)2 with 2 equiv. of {LO1}H in Et2O. The solubility of 2 in aromatic solvents was limited, but it was fairly soluble in ethers (THF, Et2O). X-ray quality crystals of 2[C6H6]0.5 were obtained by recrystallization from a solution in benzene. As the potential contamination of 1 and its questionable stability in solution could prove troublesome for subsequent reactions, the synthesis of the heteroleptic species {LO1}BaN(SiMe2H)2 (3) was carried out. By contrast to 1, the kinetically stable 3 was obtained free of impurity in satisfactory yield (79%) by treatment of {Ba[N(SiMe2H)2]2}n with a stoichiometric amount of {LO1}H in Et2O (Scheme 1). The solubility properties of 3 were comparable to those of 1, and single-crystals suitable for X-ray diffraction crystallography were readily grown from benzene. The identity and purity of 3 were conrmed by 1H, 29Si{1H} and 13C{1H} NMR spectroscopy, FTIR and elemental analysis. Its stability in solution was established by 1H NMR, as no sign of evolution (and formation of 2) was observed during the monitoring of solution of 3 in C6D6. In comparison to 1, the greater stability of 3 against Schlenk-type equilibria was attributed to the presence of internal BaHSi agostic interactions, identied both in solution (NMR) and in the solid-state (FTIR). Indeed, in the 1H NMR spectrum of 3, the 1JSiH coupling constant of 160 Hz was indicative of weak agostic interactions between Ba and the SiH moieties [17h]. In the 29Si{1H} NMR spectrum of 3, the high eld resonance (dSi = 30.4) was also characteristic of BaN(SiMe2H)2 fragments with BaHSi secondary interactions [17h]. In addition, the ms(SiH) bands at 1987 (s) and 1910 (sh) cm1 in the FTIR spectrum of 3 recorded in Nujol were diagnostic of BaHSi interactions of low intensity (Fig. 2); by contrast, free HN(SiMe2H)2 and non-interacting SiH moieties in a metal complex typically exhibit vibrations at 2122 and ca. 20002080 cm1, respectively. The addition of 1 equiv. of Bochmanns acid [H(OEt2)2]+[H2N{B(C6F5)3}2] to a solution of 3 in Et2O afforded

the well-dened ion pair [{LO1}Ba]+[H2N{B(C6F5)3}2] (4). Gratifyingly, NMR spectroscopy indicated that the cationic complex [{LO1}Ba]+ in 4 (4+) was devoid of coordinated solvent molecule, even if traces of residual Et2O could be detected. The choice of the somewhat unusual uorinated weakly-coordinating anion (WCA) [H2N{B(C6F5)3}2] in our protocols instead of the more traditional [B(C6F5)4] was motivated because of its far better crystallization properties. Indeed, the latter can be considered spherical and very often leads to the formation of oily materials. On the other hand, the bent [H2N{B(C6F5)3}2] counterion exhibits a signicant dipolar moment and is therefore oriented towards the cation, which eventually renders crystallization processes more favorable [16a,c,25,29]. The 1H NMR spectra of 4 recorded in CD2Cl2 at room temperature displayed very broad signals for all CH2 and CH3 groups, most likely as the result of a dynamic behavior in solution; attempts to freeze the uxionality by recording the spectrum at low temperature (down to 60 C) did not afford a substantial improvement. In the 19F{1H} spectrum of 4, the usual resonances of the non-coordinated anion were typically located at 132.8, 160.2 and 165.7 ppm, while characteristically a single resonance at 8.4 ppm was found in its 11B NMR spectrum. The near-quantitative synthesis of 4 could also be conveniently achieved by reacting the doubly-protonated pro-ligand [{LO1}HH]+ [H2N{B(C6F5)3}2] (prepared almost quantitatively by stoichiometric reaction of [H(OEt2)2]+[H2N{B(C6F5)3}2] and {LO1H} [16c]) with {Ba[N(SiMe2H)2]2}n in Et2O.2 This second procedure offered the advantage of yielding the targeted solvent-free, discrete salts 4 without having to preliminarily synthesize the heteroleptic precursor 3, and it therefore constituted the synthetic method of choice. The syntheses of [{LO2}Ba]+[H2N{B(C6F5)3}2] (5) and [{RO2}Ba]+[H2N{B(C6F5)3}2] (6) were previously carried out following this protocol [16a,c], but the only elucidated solidstate structure was that of the solvent-containing {6}2EtOH. Instead, the structures of 5(THF)2 and 6 are reported here for the rst time
2 If the reaction is performed in THF, or if the traditional Ba[N(SiMe3)2]2(THF)2 is used instead of {Ba[N(SiMe2H)2]2}n, then the nal product contains a signicant amount of THF (possibly coordinated on the metal center) which cannot be removed under vacuum.

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

Fig. 2. 25001500 cm1 region of the FTIR spectrum of 3 recorded in Nujol in KBr plates.

(vide infra). 5(THF)2 was prepared quantitatively by dissolving 5 in THF, followed by simple removal of the volatiles under vacuum. While repeated attempts to grow single-crystals of 5 suitable for X-ray studies were unsuccessful, X-ray quality crystals of 5(THF)2 were isolated by recrystallization in a mixture of CH2Cl2, pentane and THF. 3.2. Structural characterization The solid-state structure of 2[C6H6]0.5 was determined by X-ray diffraction methods. The compound crystallizes in the P21/n space group. There is no contact between the solvent molecule and the complex; there is some disorder in one the tBu groups in para position of the aromatic ring. The metal is 8-coordinated, and sits in a slightly distorted cubic arrangement (Fig. 3). The two nitrogen atoms (both located at ca. 2.92 of the metal center) and two phenoxide O-atoms are in trans position with each other (N(1)Ba(1)N(2) = 175.2, O(1)Ba(1)O(2) = 176.9). The BaO distance to the phenoxide O-atoms (2.59 ) is much shorter than the lengths to the O-atoms on the chelating side-arm (in the range 2.820(2)2.847(2) ); both are comparable to those already observed in related compounds [17c].  Compound 3 crystallizes in the P 1 space group. The compound is a centro-symmetric dimer in the solid-state (Fig. 4), with a central Ba2O2 planar core where the two metal centers are bridged by the Ophenoxide atoms. There is some disorder in one of the SiMe2H fragments, and only the main component is depicted in Fig. 4. Each Ba atom is 6-coordinated, as the ligand is bonded in j4N,O,O,O fashion to the metal. The BaOphenoxide distance of 2.65 is shorter than those to the O-atoms of the ether side-arms (2.772.87 ). Despite the bulkiness of the N(SiMe2H)2 amido group, the Ba(1)N(30) distance of 2.66 is far shorter than the Ba(1)N(1) one (2.93 ). Metallophilic interactions can be ruled out on account of the long intermetallic distance (4.25 ). There is no clear evidence for the presence of BaHSi agostic interaction in the dimer of 3, as the Ba(1)N(30)Si(X) and Si(2)N(30)Si(3A) angles (ca. 114 and 131, respectively) fall in the typical range for such compounds; similarly, the Ba(1)Si(2) and Ba(1)Si(3A) distances of 3.67 are unexceptional. Even if for each SiMe2H group the four Ba, N, Si and H atoms are almost perfectly co-planar as anticipated in the case of BaHSi contacts, this may simply be the outcome of steric repulsion. This is in agreement with the FTIR data, which suggested weak interactions in the solid-state; on the other hand, 1 H and 29Si NMR data were indicative of somewhat stronger contacts in solution, but direct comparison could not be established
Fig. 3. ORTEP diagram of the solid-state structure of 2[C6H6]0.5. Only the main component of the disordered tBu group is represented. The non-interacting solvent molecule and hydrogen atoms are omitted for clarity. Ellipsoids are drawn at the 50% probability level. Selected bond lengths [] and angles []: N(1)Ba(1) = 2.918(2), N(2)Ba(1) = 2.920(2), O(1)Ba(1) = 2.588(2), O(2)Ba(1) = 2.588(2), O(3)Ba(1) = 2.839(2), O(4)Ba(1) = 2.847(2), O(5)Ba(1) = 2.820(2), O(6) Ba(1) = 2.820(2); O(2)Ba(1)O(1) = 176.94(5), O(5)Ba(1)O(3) = 176.91(5), O(6)Ba(1)O(4) = 175.30(5), N(1)Ba(1)N(2) = 175.19(5), O(2)Ba(1)O(6) = 89.09(5), O(1)Ba(1)O(6) = 88.38(5), O(2)Ba(1)O(4) = 90.23(5), O(1)Ba(1) O(4) = 92.43(5).

as there was no evidence that the dimeric structure of 3 was retained in solution. By contrast to 3, the geometric features in the solid-state structure of {LO2}BaN(SiMe2H)2 (narrow BaNSi angle of 102 and short BaSi distance of 3.45 in one of the SiMe2H moieties) were clearly diagnostic of strong BaHSi agostic interactions [17h]. Note that despite the somewhat weak nature of the BaHSi contacts in 3, they nevertheless play a non negligible role in the formation of this complex, as highlighted by its greater kinetic stability with respect to 1. Although all attempts to obtain X-ray quality crystals of the solvent-free 5 were unsuccessful, single-crystals of the THF adduct 5(THF)2 suitable for crystallographic studies were obtained by recrystallization in a mixture of CH2Cl2, pentane and a small amount of THF at room temperature. In the cation (Fig. 5), the Ba atom is 8-coordinated as the coordination sphere around the metal includes the multidentate amino-ether phenoxide ligand bonded in j6 N,O,O,O,O,O manner and the two solvent molecules; the metal sits in a bicapped trigonal prismatic environment. There is considerable disorder in one of the THF molecules, and only the main component is displayed in Fig. 5. There is no contact between the cationic metal center and the uorine atoms of the counterion; besides, the very bulky WCA (V = 538 3 [30]) displays the characteristic pattern of intramolecular FHN hydrogen bonding in the range 1.935 2.320 , i.e. well below the sum of the van der Waals radii (ca. 2.5 ) for H and F [25,29]. By contrast to the related calcium derivative [{LO2}Ca]+[H2N{B(C6F5)3}2] where the complex forms a dicationic bimetallic species in the solid-state [16a], the Ba cation is monomeric in 5(THF)2, undoubtedly owing to the presence of the two

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

Fig. 4. ORTEP diagram of the solid-state structure of 3. Only the main component of the disordered SiMe2H group is represented. Hydrogen atoms (except those on Si atoms) are omitted for clarity. Ellipsoids are drawn at the 50% probability level. Selected bond lengths [] and angles []: Ba(1)O(25) = 2.653(2), Ba(1) O(25)#1 = 2.647(2), Ba(1)N(30) = 2.665(3), Ba(1)O(4) = 2.767(2), Ba(1)O(8) = 2.871(2), Ba(1)N(1) = 2.930(2), Ba(1)Si(2) = 3.6721(11), Ba(1)Si(3A) = 3.6750(13), Ba(1)Ba(1)#1 = 4.2487(3); Si(3A)N(30)Si(2) = 131.09(19), Si(3A) N(30)Ba(1) = 114.40(17), Si(2)N(30)Ba(1) = 113.64(17).

Compound 6 crystallizes in the P21/c space group. The complex forms a centrosymmetric bimetallic dication, with a planar Ba2O2 core where the Ba atoms are bridged by the two Oalkoxide of the ancillary ligand (Fig. 6). There is no contact between the metal centers in the cationic fragments and the surrounding [H2N{B(C6F5)3}2] counterion, which is stabilized by the usual FHN intramolecular interactions (1.9332.353 ) between F atoms in ortho positions of the peruorinated aromatic rings and the NH protons. The long BaBa distance (4.35 ) is indicative of the absence of metallophilic interaction. In the dication, each metal center exhibits one strong internal BaFC secondary interactions with a uorine atom, and the coordination of the ligand is consequently best described as l2:j7,j1. Each Ba atom is therefore 8-coordinated and sits in a distorted bicapped octahedral arrangement. The BaOalkoxide bond lengths (2.632.66 ) are generally substantially shorter than the distances between metal centers and macrocyclic O atoms (2.832.91 ); however, the Ba(1)O(4) distance of 2.70 is surprisingly short in comparison to all other BaOmacrocycle bond length. The Ba(1)F(23) length of 2.92 is far lower than the sum of the van der Waals radii for Ba (2.00 ) and F (1.47 ) [32]. We have recently described similar BaFC stabilizing interactions in {[{RO2}Ba]+[H2N{B(C6F5)3}2]}2EtOH (BaFC = 2.933.08 ), and DFT calculations indicated that the presence of the BaFC resulted in stabilization by ca. 25 kcal mol1 [16c]; nevertheless, the role of the solvent molecule in this latter species was not investigated. On the other hand, 6 is devoid of any coordinated solvent, and it therefore represents the rst case of a solvent-free, well-dened cationic Ba complex structurally characterized. In light of these results, one can legitimately consider that secondary BaFC interactions represent an effective mean to stabilize such highly electrophilic complexes of Ba (and more generally of the alkaline-earth metals), as recently suggested by Ruhlandt-Senge and co-workers [33]. Other examples of such BaFC secondary interactions, albeit in neutral homoleptic complexes, include those found in the hexauoroacetylacetonato complex Ba2{hfacac}4Et2O (2.773.09 ) [34], in [(THF)2Ba{N(H)2,6-F2C6H3}2]1 (2.872.90 ) [35], or in Ba{amak}2 ({amak}H = HOC(CF3)2CH2N(CH2CH2OMe)2, 3.133.21 ) [36].

Fig. 5. ORTEP diagram of the cationic fragment in 5(THF)2. Only the main component of the disordered THF molecule is represented. Hydrogen atoms are omitted for clarity. Ellipsoids are drawn at the 50% probability level. Selected bond lengths [] and angles []: Ba(1)O(4) = 2.833(3), Ba(1)O(7) = 2.841(3), Ba(1)O(11) = 2.937(3), Ba(1)O(14) = 2.844(3), Ba(1)O(22), 2.550(3), Ba(1)O(40) = 2.915(4), Ba(1)O(50) = 2.824(4), Ba(1)N(1) = 2.976(4); C(22)O(22)Ba(1) = 114.2(3).

coordinated solvent molecules. The BaOphenoxide distance (2.55 ) is far shorter than all other BaOmacrocycle (2.832.93 ) and BaOTHF (2.822.91 ) bond lengths. While the N,O,O,O,O core is folded by 80 2 in [{LO2}M]+[H2N{B(C6F5)3}2] (M = Zn, Mg, Ca) [16a,c], the aza-crown-ether tether is not signicantly distorted in 5(THF)2, since the N atom lies only 1.25 above the plane perfectly formed by O(4), O(7), O(11) and O(14). Similarly, the presence of solvent molecules in Itoh and Kitagawas seminal complexes of the smaller Ae elements [{LO2}M(L)]+[BPh4] (M = Mg, Ca, Sr; L = CH3OH, H2O) yielded structures where the macrocycle was mildly or not distorted [31].

Fig. 6. ORTEP diagram of the centrosymmetric bimetallic dication in 6. Only the main component of the disordered aza-crown-ether is represented. Hydrogen atoms are omitted for clarity. Ellipsoids are drawn at the 50% probability level. Selected bond lengths [] and angles []: Ba(1)O(26)#1 = 2.633(2), Ba(1)O(26) = 2.662(3), Ba(1)O(4) = 2.698(4), Ba(1)O(1) = 2.830(4), Ba(1)O(13A) = 2.832(3), Ba(1) O(10) = 2.885(4), Ba(1)N(7) = 2.915(3), Ba(1)F(23) = 2.918(2), Ba(1)Ba(1)#1 = 4.3505(4); O(26)#1Ba(1)O(26) = 69.52(8), Ba(1)#1O(26)Ba(1) = 110.48(8).

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213 Table 2 Polymerization data for the iROP of L-LA catalyzed by 46 and an external nucleophile.a Entry 1 2 3 4e 5 6f 7 8 9 10 11 12 13 14
a b c d e f

Initiator 4 6 5 5 5(THF)2 5 5 5 5 5 5 5 5 5

Co-initiator NuH BnOH BnOH BnOH BnOH BnOH BnOH BnOH BnOH BnOH BnOH BnOH BnNH2 HO(CH2)3OH AA-OH

[LA]0/[M+]0 /[NuH]0 1000:1:10 1000:1:10 1000:1:10 1000:1:10 1000:1:10 1000:1:10 1000:1:10 1000:1:10 1000:1:50 2 500:1:100 5000:1:100 1000:1:50 1000:1:50 1000:1:50

Tre (C) 30 100 30 30 30 30 0 0 30 30 30 30 30 30

t (h) 0.25 6.5 0.5 4 0.5 3.5 4.5 6.5 0.5 5.5 24 6 6 6

Yieldb (%) 96 50 91 94 34 84 79 99 97 97 95 95 90 98

TOF [mol(mol h)1] 3 840 77 1 820 235 680 240 176 152 1 940 441 198 158 150 163

Mn,theoc (g mol1) 12 7 13 13 5 12 11 14 2 3 7 2 2 3 600 300 300 700 000 200 500 300 900 600 000 800 700 100

Mn,SECd (g mol1) 8 5 10 10 3 7 8 8 2 3 5 2 3 4 300 600 200 400 800 400 200 500 600 000 300 900 000 100

Mw/Mnd 1.85 1.23 1.33 1.30 1.09 1.54 1.08 1.08 1.15 1.12 1.08 1.09 1.07 1.13

Polymerizations carried out in CH2Cl2 with [L-LA]0 = 2.0 M. Isolated yield after precipitation. Calculated from Mn,theo = [L-LA]0/[NuH]0 yield 144.13 + MNuH, with MBnOH = 108, MBnNH2 = 107, MHO(CH2)3OH = 76 and MAA-OH = 277 g mol1. Determined by size exclusion chromatography calibrated vs. polystyrene standards, and corrected by a factor of 0.58 according to literature recommendations [24]. Polymerizations carried out in THF. Polymerizations of rac-LA.

Fig. 7. 1H NMR spectrum (500 MHz, CDCl3, 298 K) of a low molecular weight (Mn,SEC = 5300 g mol1, Mw/Mn = 1.08) prepared with 5/BnOH ([L-LA]0/[5]0/ [BnOH]0 = 5000:1:100).

3.3. Catalytic activity in the iROP of L-LA The well-dened, solvent-free barium complexes 46 were employed to catalyze the iROP of L-LA in CH2Cl2 in combination with BnOH as an external co-initiator.3,4 Selected polymerization data are collected in Table 2. Whereas 6, supported by the uorinated alkoxide ancillary ligand, displayed moderate activity and required ele-

3 The polymerization of L-LA proceeds in the absence of external nucleophilic agent such as BnOH, but in a very poorly controlled and non reproducible fashion. The polymerization activity in this case most probably results from the presence of unknown and variable quantities of protic impurities such as lactic acid, water or iPrOH. 4 Please note that the term co-initiator is employed here to designate the external nucleophile, namely BnOH. The co-initiator plays the roles of both the initiating agent (formally, BnO initiates the formation of the polymer chain and is indeed located at the end of the polymer chain at the conclusion of the reaction) and the transfer agent (the proton H+ in BnOH is in this instance the vehicle used for the transfer between dormant and growing (macro)alcohols).

vated temperature (entry 2), 4 proved too active under the selected experimental conditions (entry 1). Indeed, full conversion of 1000 equiv. of L-LA ([L-LA]0/[4]0/[BnOH]0 = 1000:1:10) was typically ensured within 15 min at 30 C, with a high activity (>3800 molL1 ) but a poor control over the polymerization (Mw/ LA(molBa h) Mn = 1.85) which could only be marginally improved by modication of the reaction conditions. The bulkier, multidentate phenoxide analogue 5 offered the best compromise, being very efcient even at temperature as low as 0 C. At room temperature, almost complete conversion of 1000 equiv. of monomer takes place at 30 C upon addition of 10 equiv. of BnOH to 5 (entry 3). Although the agreement between theoretical and experimental molecular weights was satisfactory, the polydispersity index was somewhat large (Mw/Mn = 1.33), an indication that the binary system 5/BnOH was probably hampered by transesterication side-reactions at high monomer conversion. In THF, the reaction proceeded much more slowly than in CH2Cl2 (our solvent of choice with this catalytic system), most likely as

10

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

Fig. 8. MALDI-ToF MS spectrum (major population: Na+; minor population, K+) of a PLLA sample (Mn,SEC = 8 500 g mol1, Mw/Mn = 1.08) prepared with the binary catalytic system 5/BnOH (Mn,calc = 144.13 DP + 108.14 + 22.99, where DP is the degree of polymerization, MBnOH = 108.14 g mol1, MNa = 22.99 g mol1 and ML1 ). Note that the slight discrepancies between Mn,calc and Mn,obs LA = 144.13 g mol are due to the set of standards used for external calibration.

Fig. 10. MALDI-ToF MS spectrum of a PLLA sample (Mn,SEC = 2 900 g mol1, Mw/ Mn = 1.09) prepared with the binary catalytic system 5/BnNH2 (Mn,calc = 144.13 DP + 107.16 + 39.01, where DP is the degree of polymerization, MBnNH2 = 107.16 g mol1, MK = 39.01 g mol1 and ML-LA = 144.13 g mol1).

the result of competitive coordination of the solvent onto the metal center (compare entries 3 and 4); the resulting polymers nevertheless displayed identical macromolecular features. The detrimental inuence of coordinated solvent molecule was further highlighted when the initiator 5(THF)2 was utilized to promote the iROP of 1000 equiv. of L-LA in CH2Cl2, as the presence of only two molecules of coordinated solvent substantially reduced the catalytic activity (entries 3 and 5); however, at lower monomer conversion (34%), the molecular weight distribution was much narrower (entry 5, Mw/Mn = 1.09) than previously at near complete conversion (entry 3, Mw/Mn = 1.33). Unexpectedly, the iROP of rac-LA (entry 6) was much slower and less controlled than that of L-LA, an observation which we cannot rationalize so far; the resulting polymer was essentially atactic, as indicated by examination of the methine region of the homodecoupled 1H NMR spectrum [37]. Interestingly, the catalytic activity was maintained even at 0 C, where complete conversion of 1000 equiv. of monomer could be achieved in 6.5 h in the presence of 10 equiv. of BnOH (entries 7 and 8); although under these con-

ditions there was a very slight discrepancy between the theoretical molecular weights and those determined by SEC, their distribution was extremely narrow (Mw/Mn = 1.08), even at complete conversion. Remarkably, up to 5000 of L-LA could be quantitatively converted (entry 11), and the barium cation could withstand the addition of large excesses of BnOH without signicant loss of catalytic activity or control (entries 911). Thus, complete conversion of 25005000 equiv. of monomer could be readily achieved at room temperature in the presence of up to 100 equiv. of co-initiator; the control over the polymerization parameters remained excellent, and the productivity was satisfactory (ca. 2002000 molL-LA(molBa h)1). Besides, as expected for an iROP activated monomer mechanism with fast and reversible chain transfer between dormant and growing macromolecular chains [11d], at xed monomer loading the molecular weight of the polymers decreases regularly with increasing BnOH contents (compare entries 3 and 9). The controlled nature of the polymerization was illustrated by the anticipated increase of the molecular weight of the materials upon increasing the monomer contents at constant concentration

Fig. 9. 1H NMR spectrum (500 MHz, CDCl3, 298 K) of a low molecular weight (Mn,SEC = 2 900 g mol1, Mw/Mn = 1.09) prepared with 5/BnNH2 ([L-LA]0/[5]0/ [BnNH2]0 = 1000:1:50).

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213

11

Fig. 11. 1H NMR spectrum (500 MHz, CDCl3, 298 K) of a low molecular weight (Mn,SEC = 3000 g mol1, Mw/Mn = 1.07) prepared with 5/HO(CH2)2OH ([L-LA]0/[5]0/ [HO(CH2)3OH]0 = 1000:1:50).

Fig. 12. MALDI-ToF MS spectrum (major population: Na+; minor population, K+) of a PLLA sample (Mn,SEC = 3 000 g mol1, Mw/Mn = 1.07) prepared with the binary catalytic system 5/HO(CH2)3OH (Mn,calc = 144.13 DP + 76.09 + 22.99, where DP is the degree of polymerization, MHO(CH2)3OH = 76.09 g mol1, MNa = 22.99 g mol1 and ML-LA = 144.13 g mol1).

of co-initiator (entries 10 versus 11). This was also substantiated by NMR and MALDI-ToF MS analyses of the resulting low to medium molecular weight materials. Indeed, end-group delity was established by 1H NMR spectroscopy, which indicated that all polymer chains contained the expected CH(CH3)OH and BnOC(O) CH(CH3) termini (Fig. 7); this was further corroborated by MALDI-ToF MS (Fig. 8), where the presence of a single population of macromolecules corresponding to BnO(C6H8O4)nH (often with a signicant amount of product of transesterication sidereactions) was detected.5

A variety of nucleophilic co-initiators could be used in combination with 5, and BnOH was for instance replaced by benzyl amine (entry 12), 1,3-propanediol (entry 13) or Hawkers [38] hydroxyl functionalized alkoxy-amine 1-hydroxy-2-phenyl-2-(20 ,20 ,60 ,60 -tetramethyl-10 -piperidinyloxy)-ethane (AA-OH, entry 14) without signicant detrimental effect. If the polymerizations clearly proceeded more slowly in these cases (6 h, non-optimized reaction time) than with BnOH (0.5 h, entry 9), both the control (Mw/ Mn = 1.071.13) and end-group delity remained excellent during the iROP of 1000 equiv. of L-LA in the presence of 50 equiv. of these external nucleophiles. Upon addition of benzyl amine (entry 12), quantitative end-functionalization of the resulting monodispersed PLLA chains by the expected BnNHC(O) groups was attested by NMR spectroscopy (Fig. 9) and MALDI-ToF MS (Fig. 10) analyses. With 1,3-propanediol (entry 13), an a,x-dihydroxy-telechelic PLLA capped by a CH(CH3)OH terminal group at both extremities of the macromolecule was obtained, and its identity was also fully authenticated by NMR spectroscopy (Fig. 11) and MALDI-ToF MS (Fig. 12); such materials represent for instance convenient building blocks for the preparation of poly(ester-urethane)s containing biodegradable segments [39]. In the presence of AA-OH (entry 14), PLLA end-capped by a reactive alkoxy-amine moiety was obtained with a good level of control; this type of end-functionalized PLLAs can be employed as macro-initiators for the controlled, nitroxymediated polymerization of styrene, yielding poly(L-LA-b-styrene) block copolymers with tunable macromolecular features [11c,38].

4. Conclusion Efcient and general procedures for the preparation of well-dened, solvent-free cationic complexes [{LnX}Ba]+[H2N{B(C6F5)3}2] of the very large and electrophilic barium have been devised, involving either the synthesis of stable heteroleptic, neutral complexes, or the use of doubly-protonated cationic pro-ligands associated to the weakly-coordinating counterion [H2N{B(C6F5)3}2]. While heteroleptic neutral complexes can be stabilized against detrimental Schlenk-type equilibria by internal BaHSi contacts, the

5 MALDI-ToF MS analyses showed that transesterication reactions occurred to a signicant extent with the binary catalytic system 5 or 6 and BnOH, as a second distribution of usually minor intensity (separated from the main one by an increment of 72 Da, i.e. half a lactide unit) was often detected.

12

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213 (b) M.K. Kiesewetter, E. Ji Shin, J.L. Hedrick, R.M. Waymouth, Macromolecules 43 (2010) 2093. (a) M. Cheng, A.B. Attygalle, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 121 (1999) 11583; (b) B.M. Chamberlain, M. Cheng, D.R. Moore, T.M. Ovitt, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 123 (2001) 3229; (c) C.K. Williams, L.E. Breyfogle, S.K. Choi, W. Nam, V.G. Young Jr., M.A. Hillmyer, W.B. Tolman, J. Am. Chem. Soc. 125 (2003) 11350; (d) C.M. Silvernail, L.J. Yao, L.M.R. Hill, M.A. Hillmyer, W.B. Tolman, Inorg. Chem. 46 (2007) 6565; (e) J. Brner, U. Flrke, K. Huber, A. Dring, D. Kuckling, S. Herres-Pawlis, Chem. Eur. J. 15 (2009) 2362; (f) V. Poirier, T. Roisnel, J.-F. Carpentier, Y. Sarazin, Dalton Trans. (2009) 9820; (g) D.J. Darensbourg, O. Karroonnirun, Inorg. Chem. 49 (2010) 2360; (h) D.J. Darensbourg, O. Karroonnirun, Macromolecules 43 (2010) 8880; (i) L. Wang, H. Ma, Dalton Trans. 39 (2010) 7897; (j) V. Poirier, T. Roisnel, J.-F. Carpentier, Y. Sarazin, Dalton Trans. 40 (2011) 523; (k) J. Brner, I. dos Santos Vieira, A. Pawlis, A. Dring, D. Kuckling, S. HerresPawlis, Chem. Eur. J. 17 (2011) 4507. (a) A. Le Borgne, V. Vincens, M. Jouglard, N. Spassky, Makromol. Chem. Macromol. Symp. 73 (1993) 37; (b) M. Wisniewski, A. Le Borgne, N. Spassky, Macromol. Chem. Phys. 198 (1997) 1227; (c) T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 124 (2002) 1316; (d) N. Nomura, R. Ishii, M. Akakura, K. Aoi, J. Am. Chem. Soc. 124 (2002) 5938; (e) N. Nomura, R. Ishii, Y. Yamamoto, T. Kondo, Chem. Eur. J. 13 (2007) 4433; (f) M. Bouyahyi, E. Grunova, N. Marquet, E. Kirillov, C.M. Thomas, T. Roisnel, J.F. Carpentier, Organometallics 27 (2008) 5815; (g) H. Du, A.H. Velders, P.J. Dijkstra, J. Sun, Z. Zhong, X. Chen, J. Feijen, Chem. Eur. J. 15 (2009) 9836; (h) N. Nomura, A. Akita, R. Ishii, M. Mizuno, J. Am. Chem. Soc. 132 (2010) 1750; (i) A.D. Schwarz, Z. Chu, P. Mountford, Organometallics 29 (2010) 1246. (a) S.J. McLain, T.M. Ford, N.E. Drysdale, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 33 (1992) 463; (b) T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 121 (1999) 4072; (c) C.-X. Cai, A. Amgoune, C.W. Lehmann, J.-F. Carpentier, Chem. Commun. (2004) 330; (d) A. Amgoune, C.M. Thomas, T. Roisnel, J.-F. Carpentier, Chem. Eur. J. 12 (2006) 169; (e) A. Otero, J. Fernndez-Baeza, A. Lara-Snchez, C. Alonso-Moreno, I. Mrquez-Segovia, L.F. Snchez-Barba, A.M. Rodrguez, Angew. Chem., Int. Ed. 48 (2009) 2176; (f) T.V. Mahrova, G.K. Fukin, A.V. Cherkasov, A.A. Trifonov, N. Ajellal, J.-F. Carpentier, Inorg. Chem. 48 (2009) 4258; (g) Z. Zhang, X. Xu, S. Sun, Y. Yao, Y. Zhang, Q. Shen, Chem. Commun. (2009) 7414; (h) Z. Zhang, X. Xu, W. Li, Y. Yao, Y. Zhang, Q. Shen, Y. Luo, Inorg. Chem. 48 (2009) 5715; (i) R.H. Platel, A.J.P. White, C.K. Williams, Chem. Commun. (2009) 4115; (j) L. Clark, M.G. Cushion, H.E. Dyer, A.D. Schwarz, R. Duchateau, P. Mountford, Chem. Commun. 46 (2010) 273; (k) H.E. Dyer, S. Huijser, N. Susperregui, F. Bonnet, A.D. Schwarz, R. Duchateau, L. Maron, P. Mountford, Organometallics 29 (2010) 3602; (l) E. Grunova, E. Kirillov, T. Roisnel, J.-F. Carpentier, Dalton Trans. 39 (2010) 6739; (m) M. Bouyahyi, N. Ajellal, E. Kirillov, C.M. Thomas, J.-F. Carpentier, Chem. Eur. J. 17 (2011) 1872. J.-F. Carpentier, Macromol. Rapid Commun. 31 (2010) 1696. and references cited therein. (a) M.H. Chisholm, J.C. Gallucci, K. Phomphrai, Inorg. Chem. 44 (2005) 8004; (b) L.F. Snchez-Barba, A. Garcs, M. Fajardo, C. Alonso-Moreno, J. FernndezBaeza, A. Otero, A. Antiolo, J. Tejeda, A. Lara-Snchez, M.I. Lpez-Solera, Organometallics 26 (2007) 6403; (c) J. Wu, Y.-Z. Chen, W.-C. Hung, C.-C. Lin, Organometallics 27 (2008) 4970; (d) L. Wang, H. Ma, Macromolecules 43 (2010) 6535; (e) L.F. Snchez-Barba, A. Garcs, J. Fernndez-Baeza, A. Otero, C. AlonsoMoreno, A. Lara-Snchez, A.M. Rodrguez, Organometallics 30 (2011) 2775. (a) S. Asano, T. Aida, S.J. Inoue, Chem. Soc., Chem. Commun. (1985) 1148; (b) T. Aida, S. Inoue, Acc. Chem. Res. 29 (1996) 39; (c) S. Inoue, J. Polym. Sci. Part A: Polym. Chem. 38 (2000) 2861. (a) E. Martin, P. Dubois, R. Jrme, Macromolecules 33 (2000) 1530; (b) M.-L. Hsueh, B.-H. Huang, C.-C. Lin, Macromolecules 35 (2002) 5763; (c) V. Poirier, M. Duc, J.-F. Carpentier, Y. Sarazin, ChemSusChem 3 (2010) 579; (d) N. Ajellal, J.-F. Carpentier, C. Guillaume, S.M. Guillaume, M. Helou, V. Poirier, Y. Sarazin, A. Trifonov, Dalton Trans. 39 (2010) 8363. and references therein; (e) W. Zhao, D. Cui, X. Liu, X. Chen, Macromolecules 43 (2010) 6678. For Zn cations, see: (a) M.D. Hannant, M. Schormann, M.J. Bochmann, Chem. Soc., Dalton Trans. (2002) 4071; (b) Y. Sarazin, M. Schormann, M. Bochmann, Organometallics 23 (2004) 3296; (c) C.A. Wheaton, B.J. Ireland, P.G. Hayes, Organometallics 28 (2009) 1282; (d) C.A. Wheaton, P.G. Hayes, Dalton Trans. 39 (2010) 3861; (e) C.A. Wheaton, P.G. Hayes, Chem. Commun. 46 (2010) 8404. For Mg cations, see Ref. [12b] and: B.J. Ireland, C.A. Wheaton, P.G. Hayes, Organometallics 29 (2010) 1079.

solvent-free cation [{RO2}Ba]+ supported by the uorinated alkoxide could be successfully isolated and structurally characterized, showing the presence of strong BaFC secondary interactions. The role of these interactions in stabilizing the cation is crucial; by contrast to [{RO2}Ba]+, the solid-state structure of the putative [{LO2}Ba]+ cation bearing the non-uorinated phenoxide analogue of {RO2} could not be determined, and crystallization of a Ba cation supported by the {LO2} ancillary ligand, namely [{LO2}Ba(THF)2]+, could only be achieved upon deliberate addition of THF. As anticipated, in the presence of alcohol (up to 100 equiv.) as an external nucleophile, the resulting highly electrophilic Ba cations displayed excellent activity in the controlled iROP of L-LA, converting up to 5000 equiv. of monomer at room temperature, and high catalytic activity was maintained even at 0 C. In particular, the catalytic efciency of the binary catalyst 5/BnOH outclasses that of any other cationic system reported to date [1216]. The high activity of 4 in comparison to 5 conrmed the initial assumption that excessive coordination of donor atoms (from the ligand framework) hampered the catalytic efcacy. This was in agreement with the observation that the ROP of L-LA promoted by 5(THF)2 was slower than with the simple 5; similarly, catalysis by 5/BnOH proceeded much faster in CH2Cl2 than in THF. A broad range of protic co-initiators (alcohols, amines) can be used with these barium complexes to yield quantitatively end-functionalized PLLA chains; since these binary catalytic systems can produce several dozen polymer chains per metal center with excellent end-group delity, they are of obvious interest to the synthetic organometallic and polymer chemists. We are currently exploring the syntheses of similar well-dened cations of the alkaline-earth metals stabilized by internal secondary interactions, and are investigating the inuence of the ligand scaffold in a variety of transformations catalyzed by these Ae cations. Acknowledgments Financial support from the CNRS is gratefully acknowledged. Y.S. and B.L. are also thankful to Prof. J.-F. Carpentier (Universit de Rennes 1) for his generous assistance. The authors thank Stephen Boyer (London Metropolitan University) for the elemental analyses and Total Petrochemicals for the gift of L-lactide. Appendix A. Supplementary material

[5]

[6]

[7]

[8]

CCDC 826134, 826135, 826136 and 826137 contain the supplementary crystallographic data for 2[C6H6]0.5, 3, 5(THF)2 and 6, respectively. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/ data_request/cif. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.ica.2011.09.020. References

[9]

[10]

[11] [1] (a) K. E Uhrich, S.M. Cannizzaro, R.S. Langer, K.M. Shakesheff, Chem. Rev. 99 (1999) 3181; (b) S. Mecking, Angew. Chem. Int. Ed. 43 (2004) 1078. [2] D.L. Kaplan, Biopolymers from Renewable Resources, Springer, Berlin, 1998. [3] (a) B.J. OKeefe, M.A. Hillmyer, W.B.J. Tolman, Chem. Soc., Dalton Trans. (2001) 2215; (b) O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem. Rev. 104 (2004) 6147; (c) J. Wu, T.-L. Yu, C.-T. Chen, C.-C. Lin, Coord. Chem. Rev. 250 (2006) 602; (d) C.A. Wheaton, P.G. Hayes, B.J. Ireland, Dalton Trans. (2009) 4832; (e) C.M. Thomas, Chem. Soc. Rev. 39 (2010) 165; (f) M.J. Stanford, A.P. Dove, Chem. Soc. Rev. 39 (2010) 486. [4] For organo-catalysts, see: (a) N.E. Kamber, W. Jeong, R.M. Waymouth, R.C. Pratt, B.G.G. Lohmeijer, J.L. Hedrick, Chem. Rev. 107 (2007) 5813;

[12]

[13]

B. Liu et al. / Inorganica Chimica Acta 380 (2012) 213 [14] For Ln and group III cations, see Ref. [7j] and: (a) H. -T Sheng, H. Zhou, H.-D. Guo, H.-M. Sun, Y.-M. Yao, J.-F. Wang, Y. Zhang, Q. Shen, J. Organomet. Chem. 692 (2007) 1118; (b) D. Robert, M. Kondracka, J. Okuda, Dalton Trans. (2008) 2667; (c) D. Robert, E. Abinet, T.P. Spaniol, J. Okuda, Chem. Eur. J. 15 (2009) 11937. [15] For Al cations, see: (a) N. Nguyen, H. Emig, H. Krautscheid, R. Rau, J.-B. Cazeaux, G. Bertrand, Organometallics 17 (1998) 3599; (b) D.A. Robson, L.H. Rees, P. Mountford, M. Schrder, Chem. Commun. (2000) 1269; (c) J. Lewnski, P. Horeglad, M. Dranka, I. Justyniak, Inorg. Chem. 43 (2004) 5789; (d) S. Milione, F. Grisi, R. Centore, A. Tuzi, Organometallics 25 (2006) 266; (e) S. Dagorne, F. Le Bideau, R. Welter, S. Bellemin-Laponnaz, A. MaisseFranois, Chem. Eur. J. 13 (2007) 3202; (f) M. Haddad, M. Laghzaoui, R. Welter, S. Dagorne, Organometallics 28 (2009) 4584; (g) B. Lian, H. Ma, T.P. Spaniol, J. Okuda, Dalton Trans. (2009) 9033. [16] For Ca, Sr and Ba cations, see: (a) Y. Sarazin, V. Poirier, T. Roisnel, J.-F. Carpentier, Eur. J. Inorg. Chem. (2010) 3423; (b) M.G. Cushion, P. Mountford, Chem. Commun. 47 (2011) 2276; (c) Y. Sarazin, B. Liu, T. Roisnel, L. Maron, J.-F. Carpentier, J. Am. Chem. Soc. 133 (2011) 9069. [17] For heteroleptic Ae initiators, see: (a) M. Westerhausen, S. Schneiderbauer, A.N. Kneifel, Y. Sltl, P. Mayer, H. Nth, Z. Zhong, P.J. Dijkstra, J. Feijen, Eur. J. Inorg. Chem. (2003) 3432; (b) M.S. Hill, P.B. Hitchcock, Chem. Commun. (2003) 1758; (c) M.H. Chisholm, J. Gallucci, K. Phomphrai, Chem. Commun. (2003) 48; (d) M.H. Chisholm, J. Gallucci, K. Phomphrai, Inorg. Chem. 43 (2004) 6717; (e) D.J. Darensbourg, W. Choi, C.P. Richers, Macromolecules 40 (2007) 3521; (f) D.J. Darensbourg, W. Choi, O. Karroonnirun, N. Bhuvanesh, Macromolecules 41 (2008) 3493; (g) X. Xu, Y. Chen, G. Zou, Z. Mac, G. Li, J. Organomet. Chem. 695 (2010) 1155; (h) Y. Sarazin, D. Rosca, V. Poirier, T. Roisnel, A. Silvestru, L. Maron, J.-F. Carpentier, Organometallics 29 (2010) 6569. [18] For molecular Ae ROP initiators, see: (a) Z. Zhong, P.J. Dijkstra, C. Birg, M. Westerhausen, J. Feijen, Macromolecules 34 (2001) 3863; (b) D.J. Darensbourg, W. Choi, P. Ganguly, C.P. Richers, Macromolecules 39 (2006) 4374; (c) Y. Sarazin, R.H. Howard, D.L. Hughes, S.M. Humphrey, M. Bochmann, Dalton Trans. (2006) 340; (d) M.G. Davidson, C.T. OHara, M.D. Jones, C.G. Keir, M.F. Mahon, G. KociokKhn, Inorg. Chem. 46 (2007) 7686. [19] (a) Z. Tang, X. Chen, Q. Liang, X. Bian, L. Yang, L. Piao, X. Jing, J. Polym. Sci. Part A Polym. Chem. 41 (2003) 1934; (b) H. Guana, Z. Xiea, Z. Tanga, X. Xua, X. Chena, X. Jing, Polymer 46 (2005) 2817. [20] R.D. Shannon, Acta Crystallogr., Sect. A 32 (1976) 751.

13

[21] A.G. Avent, M.R. Crimmin, M.S. Hill, P.B. Hitchcock, Dalton Trans. (2005) 278. [22] For the very effective use of diketiminate ligands, see: (a) S. Pillai Sarish, S. Nembenna, S. Nagendran, H.W. Roesky, Acc. Chem. Res. 44 (2011) 157. and references therein; For the use of pyrrolyl-based ligands, see: (b) J. Jenter, R. Kppe, P.W. Roesky, Organometallics 30 (2011) 1404. [23] M. Arrowsmith, M.S. Hill, G. Kociok-Khn, Organometallics 30 (2011) 1291. [24] M. Save, M. Schappacher, A. Soum, Macromol. Chem. Phys. 203 (2002) 889. [25] S.J. Lancaster, A. Rodriguez, A. Lara-Sanchez, M.D. Hannant, D.A. Walker, D.L. Hughes, M. Bochmann, Organometallics 21 (2002) 451. [26] J.M. Boncella, C.J. Coston, J.K. Cammack, Polyhedron 10 (1991) 769. [27] (a) G.M. Sheldrick, SHELXS-97, Program for the Determination of Crystal Structures, University of Goettingen, Germany, 1997; (b) G.M. Sheldrick, SHELXL-97, Program for the Renement of Crystal Structures, University of Goettingen, Germany, 1997. [28] For the use of uorinated alkoxide ligands for the developments of polymerization catalysts, see: J.-F. Carpentier, Dalton Trans. 39 (2010) 37. and references cited therein. [29] (a) Y. Sarazin, J.A. Wright, M. Bochmann, J. Organomet. Chem. 691 (2006) 5680; (b) Y. Sarazin, D.L. Hughes, N. Kaltsoyannis, J.A. Wright, M. Bochmann, J. Am. Chem. Soc. 129 (2007) 881; (c) Y. Sarazin, N. Kaltsoyannis, J.A. Wright, M. Bochmann, Organometallics 26 (2007) 1811; (d) M. Bochmann, Coord. Chem. Rev. 253 (2009) 2000. [30] M.H. Hannant, J.A. Wright, S.J. Lancaster, D.L. Hughes, P.N. Horton, M. Bochmann, Dalton Trans. (2006) 2415. [31] S. Itoh, H. Kumei, S. Nagatomo, T. Kitagawa, S. Fukuzumi, J. Am. Chem. Soc. 123 (2001) 2165. [32] (a) L. Pauling, The Nature of the Chemical Bond, Cornell University Press, Ithaca, NY, 1960; (b) A. Bondi, J. Phys. Chem. 68 (1964) 441. [33] (a) W.D. Buchanan, D.G. Allis, K. Ruhlandt-Senge, Chem. Commun. 46 (2010) 4449; (b) A. Torvisco, A.Y. OBrien, K. Ruhlandt-Senge, Coord. Chem. Rev. 255 (2011) 1268. [34] A. Drozdov, S. Troyanov, J. Chem. Soc., Chem. Commun. (1993) 1619. [35] M. Grtner, H. Grls, M. Westerhausen, Dalton Trans. (2008) 1574. [36] Y. Chi, S. Ranjan, T.-Y. Chou, C.-S. Liu, S.-M. Peng, G.-H. Lee, J. Chem. Soc., Dalton Trans. (2001) 2462. [37] M.H. Chisholm, S.S. Iyer, M.E. Matison, D.G. McCollum, M. Pagel, Chem. Commun. (1997) 1999. [38] (a) C.J. Hawker, G.G. Barclay, A. Orellana, J. Dao, W. Devonport, Macromolecules 29 (1996) 5245; (b) C.J. Hawker, G.G. Barclay, J. Dao, J. Am. Chem. Soc. 118 (1996) 11467. [39] M. Helou, J.-F. Carpentier, S.M. Guillaume, Green Chem. 13 (2011) 266.