Synthesis and structural characterization of cyclic indium thiolate complexes and their utility as initiators for the ring-opening polymerization of cyclic esters

Journal of Organometallic Chemistry 736 (2013) 55e62

Contents lists available at SciVerse ScienceDirect

Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem

Synthesis and structural characterization of cyclic indium thiolate complexes and their utility as initiators for the ring-opening polymerization of cyclic esters Laura E.N. Allan a, Glen G. Briand b, *, Andreas Decken c, Jessica D. Marks b, Michael P. Shaver a, Ryan G. Wareham b a b c

School of Chemistry, University of Edinburgh, Joseph Black Chemistry Building, West Mains Road, Edinburgh EH9 3JJ, United Kingdom Department of Chemistry and Biochemistry, Mount Allison University, Sackville, New Brunswick, Canada E4L 1G8 Department of Chemistry, University of New Brunswick, Fredericton, New Brunswick, Canada E3B 6E2

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 October 2012 Received in revised form 4 March 2013 Accepted 5 March 2013

We have synthesized indium complexes incorporating bifunctional thiolate ligands and examined their utility as initiators for the ring-opening polymerization of rac-lactide and 3-caprolactone. The facile reaction of Me3In with the corresponding bifunctional thiols in diethyl ether or thf resulted in the formation of [MeIn(SCH2C(O)OMe)2]2 (5), [MeIn(SCH2CH2NMe2)2] (6) or [Me2In(SCH2C(O)OMe)] (7). The solid-state structure of 5 is dimeric via short intermolecular In/S interactions, yielding an asymmetric In2S2 core. One pendant and one chelating methyl thioglycolate ligand gives a distorted trigonal bipyramidal S3OC bonding environment for indium. Compound 6 shows a bicyclic monomeric structure with a distorted trigonal bipyramidal S2N2C bonding environment for indium. Compound 5 polymerized bulk rac-lactide rapidly with high conversion, but yielded broad PDIs and low MWs. Solution polymerizations using one equivalent of benzyl alcohol per metal centre were reasonably well controlled at 70
C, though molecular weights were lower than theoretical values. Compound 6 was also an efficient mediator of bulk rac-lactide polymerization when initiated by benzyl alcohol, reaching >90% conversion in 15 min. Molecular weights were in excellent agreement with the theoretical values and the PDIs were narrow. Solution polymerizations utilizing 6 in conjunction with benzyl alcohol were much slower than the analogous reactions using 5. Compound 5 was less efficient at controlling the ROP of 3-caprolactone versus rac-lactide, while 6 was inactive towards 3-caprolactone under a variety of conditions. This work represents the first study of indium thiolate complexes for the ROP of cyclic esters, and contains rare examples of structurally characterized organoindium bis(thiolate) compounds, the first to be prepared via the hydrocarbon elimination reaction. Ó 2013 Elsevier B.V. All rights reserved.

Keywords: Indium Thiolate X-ray diffraction Catalysis ROP Cyclic esters

1. Introduction Aliphatic poly(ester)s such as poly(lactic acid) and poly(3-caprolactone) have been identified as possible candidates as alternative biodegradable and bio-renewable polymers with specialized applications in the pharmaceutical and microelectronics industries [1e6]. The two main routes to the preparation of these materials are polycondensation and ring-opening polymerization (ROP). The latter route has been found to be superior in that it allows for a greater degree of control over the molecular parameters of the resulting polymer, including lower polydispersities, higher molecular weights and higher end-group fidelity [1,5,6]. This means * Corresponding author. Tel.: þ1 506 364 2346; fax: þ1 506 364 2313. E-mail address: [email protected] (G.G. Briand). 0022-328X/$ e see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jorganchem.2013.03.009

greater control of the physical properties of the resulting polymers, such as melting point, solubility and biodegradability. Ultimately, these factors govern potential utility of these materials for commercial applications. Key to the chemical control provided by the ROP synthetic method is the identification of suitable catalysts that facilitate polymerization of the cyclic monomer. Lewis acidic main group metal (e.g. aluminium) compounds have been found to catalyze the ring-opening polymerization of rac-lactide and 3-caprolactone [7,8]. More recently, some compounds of indium have been shown to facilitate the ROP of lactide [9e17], 3-caprolactone [18] and bbutyrolactone [19], though there have been relatively few studies in this area. Indium based catalysts are attractive due to their new reactivity profile, low toxicity and stability in water. All reported compounds involve multidentate ligands with covalent amido (N)

56

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

or aryloxy (O) linkages to the indium centre and secondary dative amine (N), phosphine oxide (O), ether (O) or thioether (S) interactions (e.g. 1e4). To our knowledge, compounds involving covalent indium thiolate (S) ligand bonding have not been examined. These compounds are expected to be air and moisture stable due to the strong covalent bonds formed between the relatively soft sulphur and indium atoms [20]. Although the additional formation of intermolecular IneS interactions in the solid-state yields insoluble materials, these can be precluded through the use of bidentate ligands which contain secondary Lewis basic groups that fill coordination sites on the metal centre. In this context, we have synthesized and structurally characterized the compounds [MeIn(SCH2C(O)OMe)2]2 (5) and [MeIn(SCH2CH2NMe2)2] (6), which possess thiolate ligands containing secondary ester (O) and amine (N) functional groups, respectively, and studied their reactivity and utility as initiators for the polymerization of 3-caprolactone and rac-lactide.

2. Results and discussion 2.1. Synthesis and solution characterization Compounds 5 and 6 were prepared via the hydrocarbon elimination reaction between trimethylindium and two equivalents of the corresponding bifunctional thiol. All reactions occurred rapidly at room temperature with evolution of methane gas. Reaction mixtures were stirred for three (5) or 16 (6) hours, and filtered to remove any precipitated product (6). Crystalline materials were isolated by slow evaporation or cooling of reaction mixtures. Although all reactions were quantitative, as determined from 1H NMR spectra of the reaction mixtures, the reported yields (79% and 87% for 5 and 6, respectively) are of crystalline material obtained from the reaction filtrate. Both compounds were found to be stable in air for several days as confirmed by FT-Raman spectroscopy.

PtBu2 O

t Bu

O In

X

N

Ts

O

t Bu

CH2SiMe3

Ts

N

O

In

Me

O

N

PtBu2

2 1 tBu tBu

t Bu Me2 N H

N

In O

Cl

Me2 iPr

N O

Cl Cl O Et

tBu

In

N

H

S

O O

In O

S

S In S

O O

O iPr

t Bu

tBu tBu

tBu

t Bu

3

MeO O

S

Me

In

tBu

tBu t Bu

4

CH2CO2Me S S

MeO2CH2C

Me In S

S

O

S

NMe2 In

Me

Me

NMe2

Me In

S

O

OMe OMe

5

6

7

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

Attempts to prepare other crystalline MeInSR0 compounds via 1:2 reactions of Me3In with N-(methyl)mercaptoacetamide or 2methoxyethanethiol were unsuccessful, and yielded an insoluble powder and oily substance, respectively. Further, attempts to prepare the alkoxy derivatives BnOIn(SR)2 via the 1:1 reaction with benzyl alcohol in toluene at 23
C, 50
C or under reflux conditions resulted in multiple unidentifiable products for 5, while no reaction was observed with 6, as determined by 1H NMR spectra of reaction mixtures. Previous studies of the 1:2 reaction of triorganylindium with alkyl- and aryl-thiols suggest that the target product RIn(SR0 )2 is in equilibrium with R2InSR0 and In(SR0 )3. The stability of the RIn(SR)2 complex was reported to be determined by the steric bulk of the thiolate ligand and the ReIn group, and the acidity of the thiol reactant [21,22]. Similarly, the successful isolation of [MeIn(SCH2CO2Me)2]2 (5) from the reaction of Me3In with two equivalents of MeO2CCH2SH was found to be dependant on reaction conditions. For example, decreasing the relative the volume of reaction solvent to 80% during a 300% scale-up of the reaction (see Sections 4.2 and 4.4) resulted in the precipitation of a colourless powder that was characterized as the 1:1 product [Me2In(SCH2CO2Me)] (7). Increasing reaction time from 3 h to 24 h did not result in conversion to the desired product (5). The RIn(SR0 )2 compounds isolated from the aforementioned studies of the hydrocarbon elimination reaction were spectroscopically characterized only. One structurally characterized example has been reported previously, namely [(Me3Si)3CIn(SPh)2]2 (8), which is prepared from the reaction of the indium(I) complex In4[C(SiMe3)3]4 with the disulfide PhSSPh [23]. The 1H NMR spectra of 5 in both thf-d8 and CDCl3 show a single set of resonances for the e SCH2C(O)OMe ligand, suggesting a monomeric structure in solution. This was confirmed by cryoscopic molecular weight determination in benzene solution. The 1H NMR spectrum of 6 in CDCl3 shows a number of peaks for the e SCH2CH2NMe2 ligand. This indicates a dynamic association/dissociation process occurring in solution, presumably resulting from the lability of the dative IneN bonding interactions. A similar observation was made previously for a series of dimethylaminoethoxy complexes {[Me2InOCH2CHRNMe2]2} [24].

Table 1 Selected bond distances ( A) and angles (deg) for 5. 5 In1eC17 In1eS1 In1eS3 In1eO10 In1eS2 C17eIn1eS1 C17eIn1eS3 S1eIn1eS3 O10eIn1eS2

2.133(4) 2.549(1) 2.431(1) 2.440(3) 2.730(1) 112.8(1) 138.3(1) 107.33(4) 166.45(7)

In2eC18 In2eS2 In2eS4 In2eO14 In2eS1 C18eIn2eS2 C18eIn2eS4 S2eIn2eS4 S1eIn2eO14

2.130(4) 2.552(1) 2.431(1) 2.432(3) 2.716(1) 111.8(1) 139.3(1) 107.47(4) 165.94(7)

Table 1). The In2S2 ring bond distances In1eS2 and In2eS1 are significantly longer than In1eS1 and In2eS2. This is presumably a result of the trans influence of the ester oxygen atoms (O7 and O10), and yields an asymmetric In2S2 core. In contrast to 5, the structure of MeIn(SCH2CH2NMe2)2 (6) (Fig. 2) shows the compound to be monomeric in the solid-state. It exhibits two chelating eSCH2CH2NMe2 ligands and a five coordinate S2N2C bonding environment for indium [In1eS1 ¼ 2.4704(6)  A; In1eS2 ¼ 2.4774(5)  A; In1eC1 ¼ 2.177(2)  A; IneN1 ¼ 2.419(2)  A; In1eN2 ¼ 2.430(2)  A]. Like 5, bond angles suggest a distorted trigonal bipyramidal geometry at indium, with the two thiolate sulphur atoms and the methyl carbon atom in equatorial positions [C1eIn1eS1 ¼ 122.04(6)
, C1eIn1eS2 ¼ 116.53(6)
, S1eIn1e S ¼ 121.30(2)
], and the amine nitrogen atoms in axial positions [N1eIn1eN2 ¼ 163.57(5)
]. In this case, the second axial coordination site is occupied by the amine donor atom of the chelating thiolate ligand rather than a bridging (intermolecular) IneS interaction. The structure of 6 is similar to those observed for other bicyclic XIn(SCH2CH2NMe2)2 [X ¼ Cl (9), I (10), 4-MeC6H4S (11), 4MeOC6H4S (12)] compounds, though the IneS and IneN bond distances are slightly longer [9e12: IneS ¼ 2.437(2)-2.446(1)  A; IneN ¼ 2.343(7)e2.409(3)  A] and the NeIneN bond angle is slightly more acute [9e12: NeIneN ¼ 166.5(2)e173.7(3)
] [20]. Ph (Me3Si)3C

2.2. X-ray crystal structures Crystals suitable for X-ray crystallographic analysis were isolated by the slow evaporation of the reaction mixture at 23
C (5) or cooling of the reaction mixture at 4
C (6). Selected bond distances and angles are given in Tables 1 and 2. The structure of [MeIn(SCH2C(O)OMe)2]2 (5) (Fig. 1) shows a A; dimer via pendant em-SCH2C(O)OMe groups [In1eS1 ¼ 2.549(1)  In1eS2 ¼ 2.730(1)  A; In2eS1 ¼ 2.716(1)  A; In2eS2 ¼ 2.552(1)  A]. The remaining eSCH2C(O)OMe ligands chelate the metal centres via a thiolate sulphur atom and a carbonyl oxygen atom [In1e S3 ¼ 2.431(1)  A; In1eO10 ¼ 2.440(3)  A; In2eS4 ¼ 2.431(1)  A;  In2eO14 ¼ 2.432(3) A], while a fifth coordination site on each metal centre is occupied by the methyl carbon atom [In1e C17 ¼ 2.133(4)  A; In2eC18 ¼ 2.130(4)  A]. This results in S3OC distorted trigonal bipyramidal bonding environments for indium, with two thiolate sulphur atoms and a methyl carbon atoms in equatorial positions [C17eIn1eS1 ¼ 112.8(1)
, C17eIn1eS3 ¼ 138.3(1)
, S1eIn1eS3 ¼ 107.33(4)
; C18eIn2eS2 ¼ 111.8(1)
, C18eIn2e S4 ¼ 139.3(1)
, S2eIn2eS4 ¼ 107.47(4)
] and an ester oxygen atom and a thiolate sulphur atom in axial positions [O10eIn1e S2 ¼ 166.45(7)
; O14eIn2eS1 ¼ 165.94(7)
]. Because of the arrangements of the pendant SCH2C(O)OMe ligands, the structure is not centrosymmetric and there are two unique indium centres. However, bond distances and angles are similar at both centres (see

57

SPh

S In

In S

PhS

NMe2 S In

C(SiMe3)3

X

S NMe2

Ph

8

X = Cl (9), I (10), 4-MeC6H4S (11), 4-MeOC6H4S (12)

Compounds 5 and 6 are two of the first examples of structurally characterized organoindium bis(thiolate) (RInSR0 2) complexes. Only one other structurally characterized compound has been

Table 2 Selected bond distances ( A) and angles (deg) for 6. 6 In1eC1 In1eS1 In1eS2 In1eN1 In1eN2 C1eIn1eS1 C1eIn1eS2 S1eIn1eS2 N1eIn1eN2

2.177(2) 2.4704(6) 2.4774(5) 2.419(2) 2.430(2) 122.04(6) 116.53(6) 121.30(2) 163.57(5)

58

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

Fig. 1. X-ray structure of 5 (30% probability ellipsoids). Hydrogen atoms are not shown for clarity.

reported previously, namely [(Me3Si)3CIn(SPh)2]2 (8) [24], which possesses a similar dimeric structure to 5. The In2S2 ring bond distances In1eS1 [2.549(1)  A] and In2eS2 [2.552(1)  A] of 5 are similar to those observed in [(Me3Si)3CIn(SPh)2]2 (8) [2.561(1) and 2.626(1)  A], while the In1eS2 [2.730(1)  A] and In2eS1 [2.716(1)  A] distances (i.e. those trans to the O7 and O10, respectively) are significantly longer due to the trans influence of the ester oxygen atom. The In1eS3 [2.431(1)  A] and In2eS4 [2.431(1)  A] bond distances of 5 are similar to the exocyclic IneS bond distances of 8 [2.452(1)  A] [20]. 2.3. DFT computational studies DFT calculations were performed to provide insight into the observed preference for dimeric (5) and monomeric (6) solidstate structures. Structural representations of the geometryoptimized structures MeIn(SCH2C(O)OMe)2, [MeIn(SCH2C(O) OMe)2]2, MeIn(SCH2CH2NMe2)2 and [MeIn(SCH2CH2NMe2)2]2 are shown in Fig. 3, and are very similar to those of the corresponding

Fig. 3. Geometry-optimized structures of a) MeIn(SCH2C(O)OMe)2, b) [MeIn(SCH2C(O) OMe)2]2 (5), c) MeIn(SCH2CH2NMe2)2 (6) and d) [MeIn(SCH2CH2NMe2)2]2. Hydrogen atoms are not shown for clarity.

compounds in the solid-state (see Supplementary information). Energies for geometry-optimized structures are given in the Supplementary information. The relative stabilities of the corresponding monomeric versus dimeric species may be calculated according to Equation (1).

  2 MeInðSR0 Þ2 / MeInðSR0 Þ2 2

(1)

An Edimerization of 42 kJ mol-1 was obtained for R0 ¼ CH2C(O) OMe, thus indicating that dimerization is thermodynamically favourable in the gas phase. Conversely, an Edimerization of þ49 kJ mol1 was obtained for R0 ¼ CH2CH2NMe2, indicating that the monomeric structure is thermodynamically favourable in the gas phase. These results are in accordance with the observed dimeric (5) and monomeric (6) solid-state structures, and suggest that the observed structures are not a result of packing forces. These data also indicate that the IneOester secondary bonding interaction is weaker than the intermolecular In/S bond, which is in turn less favourable than the IneNamine dative interaction. It is worth noting, however, that the axial coordination of the ester oxygen atom in the solid-state structure of 5 (Fig. 1) precludes a possible second axial In/S interaction at the indium centre, which would yield a coordination polymer of the type [MeIn(SCH2C(O)OMe)2]N. This preference may be a result of sterics imposed by the equatorial SCH2C(O)OMe groups. 2.4. Reaction of 5 and 6 as initiators for ROP of cyclic esters

Fig. 2. X-ray structure of 6 (30% probability ellipsoids). Hydrogen atoms are not shown for clarity.

As reported for numerous group 13 complexes [6], a coordination-insertion mechanism is expected for the ROP of raclactide and 3-caprolactone using 5 and 6 as initiators. The polymerization of rac-lactide using 5 at 120
C in bulk proceeded rapidly, with high conversions reached after 15 min (Table 3, entry 1). However, PDIs were broad and the molecular weights of the

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

59

Table 3 Ring-opening polymerization of rac-lactide and 3-caprolactone using complex 5. Entry

Monomer

M:I

Medium

Temp./
C

Time/min

% Conv.

Mn,th

Mn

PDI

1 2 3 4 5 6 7 8 9

rac-LA rac-LA rac-LA rac-LA rac-LA rac-LA rac-LA 3-CL 3-CL

100 100* 100* 100 100 250 500 100 100

Bulk Bulk Tol Tol Tol Tol Tol Tol Tol

120 120 70 70 70 70 70 70 22

15 15 15 15 60 120 240 10 180

96 >99 63 77 98 89 91 98 98

13,950a 14,430a 9100a 7470b 9470b 21,720b 44,290b 7600b 7600b

9480 3020 23,740 1310 5940 16,520 26,540 6390 3470

1.45 1.30 1.08 1.26 1.07 1.07 1.15 1.56 1.42

Benzyl alcohol used as initiator, 1 eq. per metal centre except for * where no alcohol added. 3 mL of solvent used where appropriate, unless otherwise stated. a Mn,th ¼ [M]0/[I]0  MW(monomer)  conv. þ MW(initiator). b Mn,th ¼ [M]0/([In]0 þ [I]0)  MW(monomer)  conv. þ MW(initiator). Conversion determined from 1H NMR spectra of crude samples. Mn and PDI determined from GPCMALS.

polymers obtained were significantly lower than the theoretical molecular weights. This may be attributed to transesterification, but can also be due to the benzyl alcohol acting as a chain transfer agent. Complex 5 is capable of initiating the polymerization of raclactide without the addition of benzyl alcohol (Table 3, entries 2 and 3), both under bulk and solution conditions. The bulk reaction yields polymers with reasonably broad PDIs and molecular weights which are much lower than the theoretical values, indicating significant transesterification occurs. Interestingly, the solution polymerizations, run at the lower temperature of 70
C, result in molecular weights which are much higher than theoretical values, with narrow PDIs (1.08). This indicates inefficient initiation under these conditions, with only a few growing chains which reach higher than expected molecular weights because of the effective increase in monomer concentration. Solution polymerizations of rac-lactide, using 1 equivalent of benzyl alcohol per metal centre, are reasonably well controlled at 70
C (Table 3, entries 4e7). PDIs are as low as 1.07, although molecular weights are still significantly lower than theoretical values, even when the indium-methyl bonds are assumed to initiate, with the benzyl alcohol acting as chain transfer agent. The possibility of initiation at other sites has not been ruled out, with the lability of this dinuclear complex currently under investigation. Complex 5 was less efficient at controlling the ROP of 3-caprolactone; although polymerizations proceeded rapidly, control was poor (Table 3, entries 8 and 9). Molecular weights were lower than the theoretical values and PDIs were broad (>1.4). Complex 6 is also an efficient mediator of bulk rac-lactide polymerization, initiated by benzyl alcohol, reaching >90% conversion in 15 min (Table 4, entry 1). Molecular weights were in excellent agreement with the theoretical values and the PDIs were narrow, 1.10. The alkoxy indium species formed from the in situ reaction of 6 and benzyl alcohol is a much more efficient catalyst than the organometallic species, which only reached 48% conversion in 15 min (Table 4, entry 2). Without the addition of benzyl

alcohol, PDIs are significantly broadened and molecular weights are much higher than the theoretical values, consistent with inefficient initiation of the indium-methyl group. Solution polymerizations utilizing 6 in conjunction with benzyl alcohol are much slower than the analogous reactions using catalyst 5 (Table 4, entries 3 and 4; cf Table 3 entries 4 and 5) and require 24 h at 70
C to reach high conversions. However, control over molecular weight is good and the PDIs are reasonable. As expected, polymerizations conducted in the absence of benzyl alcohol are both slower and less controlled, resulting in molecular weights much higher than the theoretical values and broad PDIs of ca 1.4. Complex 6 was inactive for 3-caprolactone polymerization under a variety of conditions. Even under bulk conditions at 70
C, conversion was less than 20% after 24 h and the polymer obtained was of very high molecular weight, with broad PDIs. In all instances, no tacticity bias was observed in the resultant polymer. Atactic polymer suggests that these ligand frameworks do not impose enough steric constraint on the metal centre to enforce either chain end control or enantiomeric site control on the polymerization reaction. While catalyst 5 was most effective in mediating solution polymerizations at lower temperatures, catalyst 6 was only effective at elevated temperatures in bulk. Future work will involve variation of the steric bulk, inclusion of stereocentres and alteration of donor frameworks on these novel indium complexes to both optimize the activity and induce stereoselectivity in cyclic ester polymerizations. Detailed mechanistic studies including kinetic profiles and MALDI end-group determination will follow with second generation catalysts capable of controlling the reaction rates and tacticity. 3. Conclusions The hydrocarbon elimination reaction of trimethylindium and bifunctional thiols is a simple and high yield route to cyclic indium thiolate complexes 5 and 6. Although the ligands are anchored to

Table 4 Ring-opening polymerization of rac-lactide and 3-caprolactone using complex 6. Entry

Monomer

M:I

Medium

Temp./
C

Time/min

% Conv.

Mn,th

Mn

PDI

1 2 3 4 5 6 8

rac-LA rac-LA rac-LA rac-LA rac-LA rac-LA 3-CL

100 100* 100 100 100* 100* 100

Bulk Bulk Tol Tol Tol Tol Bulk

120 120 70 70 70 70 70

15 15 15 1440 15 1440 1440

91 48 6 96 5 88 17

13,230 7030 970 13,950 830 12,800 20,70

13,110 53,280 6260 11,750 10,660 20,600 32,100

1.10 1.41 1.28 1.15 1.41 1.39 1.38

Benzyl alcohol used as initiator, 1 eq. per metal centre except for * where no alcohol added. 3 mL of solvent used where appropriate, unless otherwise stated. Mn,th ¼ [M]0/ [I]0  MW(monomer)  conv. þ MW(initiator). Conversion determined from 1H NMR spectra of crude samples. Mn and PDI determined from GPC-MALS.

60

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

the indium centre via the thiolate sulphur atom, they show dynamic behaviour in solution, and both monomeric and dimeric structures in the solid-state. This is a result of labile intramolecular IneO/N and intermolecular In/S interactions, all of which highlight the Lewis acidity of the indium centre and availability of coordination sites for reaction with substrates. In light of the demonstrated potential of indium complexes as initiators for the ROP of cyclic esters, we have carried out the first systematic study of organoindium thiolate compounds as initiators for the ROP of rac-lactide and 3-caprolactone. The results of these studies demonstrate the potential of organoindium dithiolates [RIn(SR0 )2]n for this application and provide direction for the design of second generation systems. Notably, the observation of low molecular weight transesterification products highlights the need for increased steric bulk at the indium centre. Further, the observation that solution polymerizations utilizing 6 in conjunction with benzyl alcohol were much slower than the analogous reactions using 5, and that compound 5 was active towards 3-caprolactone while 6 was not, suggests that the presence of weakly coordinating secondary ester oxygen donor groups yields a more reactive indium centre versus amine nitrogen groups. We are currently preparing organoindium dithiolate compounds incorporating polyfunctional thiolate ligands and screening their ability to initiate cyclic ester polymerization reactions. 4. Experimental 4.1. General considerations Solution 1H and 13C{1H} spectra were recorded at 23
C on either a JEOL GMX 270 MHz þ spectrometer (270 and 67.9 MHz, respectively), or a Varian Mercury 200MHz þ spectrometer (200 and 50 MHz, respectively), and chemical shifts are calibrated to the residual solvent signal. FT-IR spectra were recorded as Nujol mulls with NaCl plates on a Mattson Genesis II FT-IR spectrometer in the range of 4000e 400 cm1. FT-Raman spectra were recorded on a Thermo Nicolet NXR 9600 Series FT-Raman spectrometer in the range 3900e70 cm1. Melting points were recorded on an Electrothermal MEL-TEMP melting point apparatus and are uncorrected. Elemental analyses were performed by Chemisar Laboratories Inc., Guelph, Ontario. Cryoscopic molecular weight determination experiments were performed in benzene using literature methods [25]. GPC analyses were carried out on a Polymer Laboratories PL-GPC 50 Plus system equipped with two Jordi Gel DVB mixed bed columns (300 mm  7.8 mm), a refractive index detector (880 nm) and a Wyatt Technology miniDAWNÔ TREOSÒ multiple angle light scattering (MALS) detector operating at 658 nm. Samples were dissolved and eluted in HPLCgrade THF at a flow rate of 1 mL min1 at 50
C. dn/dc values of 0.051 for poly(lactide) [26] and 0.079 for poly(caprolactone) [27] were used to calculate molecular weights. Polymerizations were set up under inert atmosphere using an MBraun LABmaster sp glovebox equipped with a 35
C freezer, [O2] and [H2O] analyzers and a builtin Siemens Simatic Touch Panel. Methyl thioglycolate 95%, 2-(dimethylamino)ethanethiol hydrochloride 95%, N-(methyl)mercaptoacetamide 97% and sodium hydride 95% were used as received from SigmaeAldrich. 2Methyoxyethanthiol was prepared according to literature methods [28]. Trimethylindium was used as received from Strem. PURASORB DL-lactide was obtained from PURAC Biochem by Gorinchem and sublimed 3 times under vacuum prior to use. 3-Caprolactone and benzyl alcohol were obtained from SigmaeAldrich, dried over calcium hydride and distilled under inert atmosphere prior to use. Tetrahydrofuran (thf) was dried using an MBraun SPS column solvent purification system. Diethyl ether, anhydrous 99%þ was used as received from SigmaeAldrich. All reactions were performed

under an atmosphere of inert dinitrogen using standard Schlenk techniques unless otherwise indicated. 4.2. Preparation of [MeIn(SCH2C(O)OMe)2]2 (5) Methyl thioglycolate (0.266 g, 2.5 mmol) was added to a solution of In(CH3)3 (0.200 g, 1.25 mmol) in diethyl ether (5 mL) to give a clear solution. The reaction mixture was stirred for 3 h and concentrated under vacuo to yield 5 as a colourless crystals (0.370 g, 1.25 mmol, 87%). Anal. Calc. for C14H26In2O8S4: C, 24.71; H, 3.86; N, 0.00. Found: C, 24.49; H, 3.60; N, <0.20. M.p. ¼ 140e143
C. FT-IR (cm1): 700s, 778w, 880m, 894w, 985m, 1129m, 1172m, 1205s, 1278m, 1319m, 1571w, 1680s, 1737m. FT-Raman (cm1): 126vs, 197w, 272w, 341w, 401w, 484vs, 518m, 568vw, 686vw, 778w, 897w, 994w, 1159s, 1401w, 2920m, 2958w. NMR data (thf-d8, ppm), 1H NMR: 0.00 [s, 3H, MeIn(SCH2C(O)OMe)2], 3.34 [s, 4H, MeIn(SCH2C(O)OMe)2], 3.60 [s, 6H, MeIn(SCH2C(O)OMe)2]; 13C{1H} NMR: 14.0 [MeIn(SCH2C(O) OMe)2], 22.5 [MeIn(SCH2C(O)OMe)2], 59.4 [MeIn(SCH2C(O)OMe)2], 165.4 [MeIn(SCH2C(O)OMe)2]. NMR data (CDCl3, ppm), 1H NMR: 0.29 [s, 3H, MeIn(SCH2C(O)OMe)2], 3.61 [s, 4H, MeIn(SCH2C(O) OMe)2], 3.81 [s, 6H, MeIn(SCH2C(O)OMe)2]. 4.3. Preparation of [MeIn(SCH2CH2NMe2)2] (6) Sodium hydride (0.081 g, 3.38 mmol) was added to a solution of 2(dimethylamino)ethanethiol hydrochloride (0.448 g, 3.16 mmol) in thf (8 mL). The resulting cloudy solution was stirred for 1 h then combined with Me3In (0.251 g, 1.57 mmol) in thf (3 mL). After 16 h, the reaction mixture was filtered, and the filtrate was concentrated to 1 mL in vacuo and allowed to sit at 4
C. After 1 d, the solution was filtered to yield 6 as colourless crystals (0.198 g, 0.585 mmol, 79%). Anal. Calc. for C9H23InN2S2: C, 31.96; H, 6.87; N, 8.28. Found: C, 32.19; H, 6.53; N, 8.01. Mp ¼ 143e145
C. FT-IR (cm1): 664m, 724w, 766s, 891m, 949m, 1030m, 1102vw, 1123w, 1167w, 1224, 1247w. FT-Raman (cm1): 456w, 484vs, 515w, 535w, 669m, 767m, 900w, 952w, 999w, 1035w, 1126w, 1149s, 1226w, 1246w, 1297w, 1437m, 1462m, 2794m, 2839s, 2919vs, 2934vs, 2988m. NMR data (CDCl3, ppm), 1H NMR: 0.01 (s, 3H, MeIn(SCH2CH2NMe2)2), 2.29 (m, 12H, MeIn(SCH2CH2NMe2)2), 2.64 (m, 4H, MeIn(SCH2CH2NMe2)2), 2.84 (m, 4H, MeIn(SCH2CH2NMe2)2); 13C {1H} NMR: 23.7 [MeIn(SCH2CH2NMe2)2], 44.0 [MeIn(SCH2CH2NMe2)2], 46.2 [MeIn(SCH2CH2NMe2)2], 60.6 [MeIn(SCH2CH2NMe2)2]. 4.4. Preparation of [Me2In(SCH2C(O)OMe)] (7) Methyl thioglycolate (0.798 g, 7.50 mmol) was added to a solution of In(CH3)3 (0.600 g, 3.75 mmol) in diethyl ether (12 mL) to give a clear solution which turns cloudy after a few minutes. The reaction mixture was stirred for 3 h and filtered to yield 7 as a colourless powder (0.424 g, 1.70 mmol, 45%). Anal. Calc. for C5H11InO2S: C, 24.02; H, 4.44; N, 0.00. Found: C, 23.41; H, 4.18; N, <0.3. M.p. ¼ 111e113
C. FT-IR (cm1): 703s, 878w, 897m, 994m, 1169m, 1196s, 1303s, 1397w, 1434m, 1572w, 1647w, 1699m. FTRaman (cm1): 126s, 196m, 272w, 339w, 403w, 484vs, 517m, 562vw, 691vw, 778w, 897w, 994w, 1159m, 1401w, 1437vw, 1452vw, 2920m, 2957w. NMR data (CDCl3, ppm), 1H NMR: 0.04 [s, 6H, Me2In(SCH2C(O)OMe)], 3.48 [s, 2H, Me2In(SCH2C(O)OMe)], 3.75 [s, 3H, Me2In(SCH2C(O)OMe)]; 13C{1H} NMR: 4.5 [Me2In(SCH2C(O) OMe)], 29.2 [Me2In(SCH2C(O)OMe)], 53.7 [Me2In(SCH2C(O)OMe)], 176.1 [Me2In(SCH2C(O)OMe)]. 4.5. Polymerization experiments 0.012 g of the desired precatalyst (5 or 6), 3.6 mL of benzyl alcohol and either 0.500 g rac-lactide or 0.400 g 3-caprolactone (1:1:100 indium:benzyl alcohol:monomer ratio) were added to an

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

Acknowledgements

Table 5 Crystallographic data for 5 and 6.

Formula fw Crystal system Space group a ( A) b ( A) c ( A) a (deg) b (deg) g (deg) V ( A3) Z F(000) rcalcd, g cm3 m, mm1 T, K l,  A Goodness-of-fit on F2 R1a wR2b a b

61

5

6

C14H26In2O8S4 680.23 monoclinic Cc 19.897(7) 8.362(3) 14.943(5) 90 93.793(6) 90 2480.6(14) 4 1344 1.821 2.229 188(1) 0.71073 1.083 0.0213 0.0534

C9H23InN2S2 338.23 monoclinic P2(1)/n 10.6081(14) 9.5405(13) 13.9120(18) 90 93.233 90 1405.7(3) 4 688 1.598 1.949 188(1) 0.71073 1.078 0.0210 0.0553

R1 ¼ [SjjFoj  jFcjj]/[SjFoj] for [F2o > 2s(F2o)]. wR2 ¼ {[Sw(F2o  F2c )2]/[Sw(F4o)]}½.

oven-dried ampoule charged with a magnetic stir bar and 3 mL of toluene when appropriate. The ampoule was sealed and heated to 120
C (bulk polymerization) or 70
C (solution polymerization) with stirring for the desired period of time. The ampoule was cooled to room temperature and the resulting viscous mixture was dissolved in a 10:1 v:v dichloromethane:methanol solution. Once fully dissolved, the solution was left to stir at ambient temperature for 30 min and a sample was removed for determination of conversion before the polymer was precipitated by dropwise addition of the solution into 100 mL of cold methanol. The resulting white precipitate was filtered and dried in vacuo to constant weight. Low molecular weight samples which did not precipitate were concentrated in vacuo and then dried to constant weight. Samples were analyzed by 1H NMR spectroscopy and gel-permeation chromatography.

We thank the following: Dan Durant for assistance in collecting solution NMR data; and the Natural Sciences and Engineering Research Council of Canada, the New Brunswick Innovation Foundation, the Canadian Foundation for Innovation and Mount Allison University for financial support. Appendix A. Supplementary material CCDC 905629 and 905630 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www. ccdc.cam.ac.uk/data_request/cif. Appendix B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jorganchem.2013. 03.009. References [1] [2] [3] [4] [5] [6] [7]

[8] [9] [10]

4.6. X-ray crystallography Crystals of 5 and 6 were isolated from the reaction mixtures as indicated above. Single crystals were coated with Paratone-N oil, mounted using a 20 mm cryo-loop and frozen in the cold nitrogen stream of the goniometer. A hemisphere of data was collected on a Bruker AXS P4/SMART 1000 diffractometer using u and q scans with a scan width of 0.3
and 10 s exposure times. The detector distance was 5 cm. The data were reduced (SAINT) [29] and corrected for absorption (SADABS) [30]. The structures were solved by direct methods and refined by full-matrix least squares on F2 (SHELXTL) [31]. All non-hydrogen atoms were refined using anisotropic displacement parameters. Hydrogen atoms were included in calculated positions and refined using a riding model. Crystallographic data are given in Table 5. 4.7. Computational methods DFT calculations were performed using Gaussian 09 at the B3LYP 6-31G* level of theory for all atoms except In, for which Stuttgart electron core pseudo-potentials (sdd) were employed [32]. All structures were geometry optimized and structural parameters for input files were derived from crystal structure data where possible. Frequency calculations were performed on all structures and gave no imaginary frequencies. Structural parameters are given in the Supplementary data.

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]

[28] [29] [30] [31]

C.K. Williams, Chem. Soc. Rev. 36 (2007) 1573. J. Wu, T.L. Yu, C.-T. Chen, C.-C. Lin, Coord. Chem. Rev. 250 (2006) 602. D. Garlotta, J. Polym. Environ. 9 (2001) 63. O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem. Rev. 104 (2004) 6147. Y. Cheng, S. Deng, P. Chen, R. Ruan, Front. Chem. China 4 (2009) 259. M.J. Stanford, A.P. Dove, Chem. Soc. Rev. 39 (2010) 486. (a) M.-H. Thibault, F.-G. Fontaine, Dalton Trans. 39 (2010) 5688; (b) M.H. Chisholm, J.C. Gallucci, K.T. Quisenberry, Z. Zhou, Inorg. Chem. 47 (2008) 2613; (c) H. Du, A.H. Velders, P.J. Dijkstra, J. Sun, Z. Zhong, X. Chen, J. Feijen, Chem. Eur. J. 15 (2009) 9836; (d) A. Arbaoui, C. Redshaw, D.L. Hughes, Chem. Commun. (2008) 4717; (e) M. Haddad, M. Laghzaoui, R. Welter, S. Dagorne, Organometallics 28 (2009) 4584; (f) S. Milione, F. Grisi, R. Centore, A. Tuzi, Organometallics 25 (2006) 266. A.P. Dove, V.C. Gibson, E.L. Marshall, A.J.P. White, D.J. Williams, Chem. Commun. (2001) 283. M.P. Blake, A.D. Schwarz, P. Mountford, Organometallics 30 (2011) 1202. (a) A.F. Douglas, B.O. Patrick, P. Mehrkhodavandi, Angew. Chem. Int. Ed. 47 (2008) 2290; (b) K.M. Osten, I. Yu, I.R. Duffy, P.O. Lagaditis, J.C.-C. Yu, C.J. Wallis, P. Mehrkhodavandi, Dalton Trans. 41 (2012) 8123; (c) I. Yu, A. Acosta-Ramírez, P. Mehrkhodavandi, J. Am. Chem. Soc. 134 (2012) 12758. D.C. Aluthge, B.O. Patrick, P. Mehrkhodavandi, Chem. Commun. http://dx.doi. org/10.1039/c2cc33519k. I. Peckermann, A. Kapelski, T.P. Spaniol, J. Okuda, Inorg. Chem. 48 (2009) 5526. M. Hu, M. Wang, P. Zhang, L. Wang, F. Zhu, L. Sun, Inorg. Chem. Commun. 13 (2010) 968. J.-C. Buffet, J. Okuda, P.L. Arnold, Inorg. Chem. 49 (2010) 419. A. Pietrangelo, M.A. Hillmyer, W.B. Tolman, Chem. Commun. (2009) 2736. A. Pietrangelo, S.C. Knight, A.K. Gupta, L.J. Yao, M.A. Hillmyer, W.B. Tolman, J. Am. Chem. Soc. 132 (2010) 11649. M. Normand, E. Kirillov, T. Roisnel, J.-F. Carpentier, Organometallics 31 (2012) 1448. I.-P. Hsieh, C.-H. Huang, H.M. Lee, P.-C. Kuo, J.-H. Huang, H.-I. Lee, J.-T. Cheng, G.-H. Lee, Inorg. Chim. Acta 359 (2006) 497. C. Xu, I. Yu, P. Mehrkhodavandi, Chem. Commun. 48 (2012) 6808. G.G. Briand, B.F.T. Cooper, D.B.S. MacDonald, C.D. Martin, G. Schatte, Inorg. Chem. 45 (2006) 8423. R. Nomura, S. Inazawa, K. Kanaya, H. Matsuda, Polyhedron 8 (1989) 763. G.G. Hoffmann, J. Organomet. Chem. 338 (1988) 305. C. Peppe, F. Molinos de Andrade, W. Uhl, J. Organomet. Chem. 694 (2009) 1918. H. Schumann, J. Kaufmann, B.C. Wassermann, F. Girgsdies, N. Jaber, J. Blum, Z. Anorg. Allg. Chem. 628 (2002) 971. R. Mikulak, O. Runquist, J. Chem. Educ. 38 (1961) 557. W. Zhao, Y. Wang, X. Liu, D. Cui, Chem. Commun. 48 (2012) 4588. Z.A.A. Hamid, A. Blencowe, B. Ozcelik, J.A. Palmer, G.W. Stevens, K.M. Abberton, W.A. Morrison, A.J. Penington, G.G. Qiao, Biomaterials 31 (2010) 6454. F. Tisato, T. Maina, L.-R. Shao, M.J. Heeg, E. Deutsch, J. Med. Chem. 39 (1996) 1253. SAINT 7.23A, Bruker AXS, Inc., Madison, Wisconsin, USA, 2006. G.M. Sheldrick, SADABS, Bruker AXS, Inc., Madison, Wisconsin, USA, 2008. G.M. Sheldrick, Acta Cryst. A64 (2008) 112.

62

L.E.N. Allan et al. / Journal of Organometallic Chemistry 736 (2013) 55e62

[32] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J. Normand,

K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09. Revision B.01, Gaussian, Inc., Wallingford CT, 2010.