Kinetics of Fe-Mediated ATRP with Triarylphosphines - ACS Publications

Jun 29, 2015 - The strongly Lewis basic TTMPP may coordinate to FeII species, but primarily it acts as a reducing agent for [FeIIIBr4]− in ATRP abov...
3 downloads 7 Views 593KB Size
Download PDF
Article pubs.acs.org/Macromolecules

Kinetics of Fe-Mediated ATRP with Triarylphosphines Hendrik Schroeder,† Krzysztof Matyjaszewski,‡ and Michael Buback*,† †

Institut für Physikalische Chemie, Georg-August-Universität Göttingen, Tammannstraße 6, 37077 Göttingen, Germany Center for Macromolecular Engineering, Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213, United States

‡

S Supporting Information *

ABSTRACT: Phosphines have been successfully used as additives for Fe-based ATRP. To understand their role in detail, the kinetics of Febased ATRP in the presence of tris(2,4,6-trimethoxyphenyl)phosphine (TTMPP) and of triphenylphosphine (TPP) was studied via online vis/near-IR and 31P NMR spectroscopy. No significant effect on the ATRP equilibrium constant, KATRP, was observed upon the addition of up to 1.5 equiv of phosphine relative to Fe. The strongly Lewis basic TTMPP may coordinate to FeII species, but primarily it acts as a reducing agent for [FeIIIBr4]− in ATRP above 100 °C, thus transforming TTMPP to TTMPP-Br+. TPP is a weaker Lewis base and consequently a less effective reducing agent.

■

INTRODUCTION

additive for GAMA ATRP, since FeBr3 by itself retards radical polymerization.13 However, full understanding of the ability of different types of phosphines to act as ligand,10,11 reducing agent,11,13 or assistant for chain initiation10,11,20 has not yet been attained. It is a matter of priority to clarify the specific roles of phosphine and to achieve further mechanistic insight. The present study aims at improving the situation by monitoring phosphineassisted Fe-mediated ATRPs and associated monomer-free model systems via online vis/near-IR (NIR) spectroscopy. In addition, equilibrium constants, KATRP, were measured for the ATRP of styrene and BA and compared to our previous data for MMA.34 Of primary interest among the triarylphosphine additives is tris(2,4,6-trimethoxyphenyl)phosphine (TTMPP), which has been reported to reduce iron(III) halides to iron(II) compounds and is highly Lewis basic35 (pKa = 11.2).11,36,37 The rate of this reaction was also studied via online spectroscopic monitoring.

Reversible-deactivation radical polymerization (RDRP) allows for the synthesis of polymeric materials with precisely tailored topology, architecture, chain length, functionality, and narrow molar-mass distribution.1−7 The development of Fe catalysts is an attractive target for RDRP procedures, including atomtransfer radical polymerization (ATRP), due to the low toxicity and broad availability of Fe.8,9 Phosphines are among the most frequently used additives in iron-mediated ATRP10,11 and generate economic and robust Fe catalysts. Their influence on the rate and degree of control over an ATRP has been studied for several monomers,12−16 such as styrene13,15,17−21 and methyl methacrylate (MMA).13,15,19−28 It is recommendable to add the iron catalyst to the polymerization medium in the less air-sensitive, higher oxidation state, FeIII, which requires its reduction to start ATRP. Iron-mediated ATRP with phosphines has been effectively carried out in the presence11,14,22,24,26−29 and in the absence18,23 of external reducing agents. The results suggest that phosphines themselves may reduce FeIII.13,29 It was found that substituted, electron-rich phosphines accelerate polymerization12,13 and improve functional group tolerance which allowed for copolymerizations with HEMA12 and enabled conducting an ATRP in the presence of methanol.14 Iron/phosphine-catalyzed ATRP even operates in the absence of an alkyl halide or a thermal initiator.11,20,22,30−33 Dass et al. reported that dipolar interactions between the carbonyl oxygen of MMA and the phosphorus atom are mediated by the presence of the iron(III) halide complex.31,32 The so-induced catalytic formation of 1,2-dihaloisobutyrate was detected via NMR spectroscopy.20 Such in situ initiator formation is referred to as generation of activators by monomer addition (GAMA) ATRP.11 Phosphine presents an essential © XXXX American Chemical Society

■

EXPERIMENTAL SECTION

Chemicals. FeBr3 (ABCR, anhydrous, 99%), FeBr2 (ABCR, ultra dry, 99.995% metals basis), tris(2,4,6-trimethoxyphenyl)phosphine (TTMPP, STREM, ≥97.0%), triphenylphosphine (TPP, ABCR, 99%), tetrabutylammonium bromide (TBA-Br, Fluka, ≥99.0%), ethyl αbromophenylacetate (EBrPA, ABCR, 97%), methyl 2-bromoisobutyrate (MBriB, Fluka, ≥99%), N,N-dimethylformamide (DMF, extra dry over molecular sieve, 99.8%), DMF-d7 (ABCR, 99.5 atom % D), and anisole (Aldrich, anhydrous, 99.7%) were used as received. Styrene (Aldrich, ≥99.0%), methyl methacrylate (MMA, Fluka, >99.0%), butyl acrylate (BA, Aldrich, ≥99%), and N-methylpyrrolidin-2-one (Acros, Received: May 15, 2015 Revised: June 12, 2015

A

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules ultra dry, 99.5%) were passed through a column filled with Al2O3 (Acros, neutral, Brockmann I, 50−200 μm). The monomers and solvents were degassed by several freeze−pump−thaw cycles. All solutions were prepared under an argon atmosphere. Spectroscopic Measurements. ATRP was carried out in sealed thermostated quartz cells (117-100-10-40/QS, Hellma). The concentration of FeIII was monitored via online UV/vis (Cary 300, Agilent) spectroscopy. Monomer conversion was measured via online FT-NIR spectroscopy (IFS 88, Bruker). The NMR spectra (Avance DPX 300, Bruker) were recorded at ambient temperature and frequencies of 121.49 MHz (31P) and 300.13 MHz (1H). The spectra were calibrated against 85% phosphoric acid (31P) and tetramethylsilane (1H) as external standards and against the residual proton signals of DMF-d7. SEC Analysis. Molar mass distributions were determined with an SEC device consisting of an Agilent 1260 ALS G1329B autosampler, an Agilent 1260 Infinity ISO pump G1310B, three columns (PSSSDV, 8 × 300 mm each, particle diameter 5 μm, pore sizes of 106, 105, and 103 Å), an Agilent 1260 refractive index (RI) detector G1362A, and tetrahydrofuran (THF) at 35 °C as the eluent at a flow rate of 1 mL min−1. The setup was calibrated against PMMA and PS standards of narrow dispersity from Polymer Standard Services (PSS) with molar masses ranging from 800 to 2 000 000 g mol−1.

stage, ATRP is significantly accelerated upon increasing the amount of added TTMPP. This observation is in agreement with reported data for ATRP carried out in heterogeneous systems with anisole under otherwise similar conditions.13 Br−FeIII/L was monitored between wavelengths of 15 000 and 9000 cm−1 as illustrated in Figure 1C. This absorption is highest at time zero, prior to adding TTMPP. The Br−FeIII/L absorption band almost completely disappears upon addition of 1 equiv of TTMPP due to the rapid reaction of Br−FeIII/L with TTMPP. The associated formation of the FeII redox partner may be evidenced via the absorption between 9000 and 5000 cm−1 (Figure S1), underlying the intense monomer band in Figure 1A.38 No monomer conversion was observed prior to the formation of FeII which indicates that ATRP is initiated via the FeII catalyst complex. The subsequent increase in Br−FeIII/ L absorption is consistent with the persistent radical effect.39 The measured Br−FeIII/L concentration vs time curve is plotted in Figure 1D. The absorption pattern in Figure 1C, measured prior to adding TTMPP, was tentatively assigned to Br−FeIII/L and is actually due to [FeBr4]−,38,40 the expected primary species present in the absence of additives: [FeBr4]−:[Fe(Solv)6]3+ ≈ 3:1.38,41 Quantitative formation of [FeBr4]− from FeBr3 requires additional bromide. The FeII bromide complex, i.e., the redox partner in ATRP, may serve as a bromide source.38 [FeBr4]− should be the only FeIII species as soon as the formation of the FeII bromide complex allows for bromide transfer to FeIII.38 Interestingly, no major change in the shape of the FeIII absorption band was observed after adding TTMPP, which suggests that [FeBr4]− remains the single FeIII species present in the reaction medium. FeIII/TTMPP complexes have not been detected, probably because the reduction of FeIII presents the preferred reaction pathway in ATRPs conducted at elevated temperature. In the following discussion, we assume [FeBr4]− to be the single FeIII species present in the reaction medium, thus specifying the structure of the Br−FeIII/L complex given in Scheme 1 and Figure 1. We used the reported protocol38 for calibration of the [FeBr4]− concentration vs time traces recorded during the TTMPP-assisted ATRPs (cf. Figure 1D). Our online spectroscopic measurements indicate that a higher TTMPP content yields progressively higher FeII:FeIII ratios, thus enhancing the rate of the ATRP, Rp, according to eq 1. The simultaneous measurement of both monomer conversion and [FeBr4]− concentration allows for the quantitative analysis of KATRP via eq 1.42

■

RESULTS AND DISCUSSION The ATRPs were started in the reverse fashion with the Fe catalyst added to the reaction in the higher oxidation state, FeIII, as illustrated in Scheme 1. The reagents marked in red, i.e., the Scheme 1. Iron-Mediated ATRP Involving FeII and FeIII Complexesa

a

TTMPP acts as the reducing agent for the FeIII complex. The type of ligand, L, coordinated to iron will be specified below. Rn-Br refers to deactivated chains, Rn• to growing radicals of chain length n, M to monomer, and kp and kt to the propagation and the termination rate coefficient, respectively.

alkyl halide initiator, the Br−FeIII/L complex, TTMPP, and monomer, M, are the initial components. The formation of the FeII/L activator complex is expected to occur in situ via the reduction of Br−FeIII/L with TTMPP. The mechanism and the kinetics of this reduction will be discussed in detail below, as will be the type of ligand, L, including Br, the phosphine, or solvent. ATRPs of styrene were carried out at 100 °C in DMF solution (33 vol %) using the following initial molar ratios of reagents: [Sty]:[MBriB]:[FeBr3] = 200:1.25:1.00. The polymerizations were carried out with different levels of TTMPP: 0.5, 1.0, and 1.5 equiv with respect to FeBr3. Spectroscopic measurements were started just before adding TTMPP to the reaction mixture at the target polymerization temperature. Monomer conversion was monitored between 6250 and 6000 cm−1 via online NIR spectroscopy as illustrated in Figure 1A for styrene ATRP with 1.0 equiv of TTMPP. Only seven out of a multitude of spectra recorded over 21 h are shown. The ln([M]0/[M]) vs time traces for all three levels of TTMPP are illustrated in Figure 1B. Especially in the early polymerization

Rp = −

[Fe II][R nX] d[M] [M] = k p[R •n][M] = k pKATRP dt [Fe III] (1)

The propagation rate coefficient is precisely known from pulsed-laser polymerization experiments43−45 for systems of similar monomer-to-solvent composition.46,47 The time-dependent concentrations of [FeII] and [Rn−X] are calculated from the following relationships: [FeII] = [FeIII]0−[FeIII] and [R n −X] = [R n −X] 0 −[Fe III ]. K ATRP was found to be independent of TTMPP content, and no significant change of KATRP was observed within the range of monomer conversions under investigation. The so-obtained mean value for styrene at 100 °C amounts to KATRP = 7.6 × 10−9 (Table 1). This value is by about a factor of 2 above the one found when using 1 equiv of triphenylphosphine (TPP), KATRP = 3.6 × 10−9, instead of B

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 1. Kinetics of ATRP monitored via online NIR spectroscopy: (A) monomer conversion measured between 6250 and 6000 cm−1; (B) ln([M]0/[M]) vs time traces measured at three levels of TTMPP; (C) Br−FeIII/L monitored between 15 000 and 9000 cm−1; (D) Br−FeIII/L is first reduced by TTMPP and accumulates with reaction time, resulting from radical termination and following the persistent radical effect.

Table 1. KATRP for Fe-Mediated ATRPs of Different Monomers with or without TTMPP as the Additive entry

monomer/solvent

T (°C)

1 2 3 4 5

MMA/NMP (1:1) MMA/NMP (1:1) Sty/DMF (2:1) Sty/DMF (2:1) BA/NMP (2:3)

60 60 100 100 100

equiv of TTMPPa 1.5 0.5−1.5 1.1b 1

MMA reflects the expectations based on the relative bond strengths of the associated alkyl halides.48−50 The polymerization rate, as visualized by the ln([M]0/[M]) vs time plots, for homogeneous solutions in either DMF or NMP is enhanced as compared to heterogeneous polymerization with anisole.13 Despite the rate enhancement in homogeneous polymerization, the ATRPs of styrene and MMA occurred in a controlled fashion. The dispersity values were Đ = 1.49 (PS) at 44% and Đ = 1.34 (PMMA) at 76% monomer conversion, respectively, with 1.5 equiv of TTMPP being added in each case (Figure S3). While the heterogeneous system may be the preferred option for synthesizing polymer due to the lower dispersity values13 and the minor toxicity concerns with the solvent anisole, the homogeneous system was selected as it enables the online NIR spectroscopic studies. The measured molar masses were in good agreement with the theoretical values calculated on the basis of the amount of initially added alkyl halide (MBriB or EBrPA), e.g., Mn,SEC = 8900 g mol−1 vs Mn,theo = 9200 g mol−1 for polystyrene. This observation confirms that control over the polymerization was attained. We did not observe a significant impact of initiation via the GAMA ATRP mechanism, which would have resulted in a lower than predicted molar mass due to the formation of an additional amount of initiator. Under the conditions investigated, the reduction of [FeBr4]− by TTMPP is the dominant process, as detailed in the following section. Reduction of [FeBr4]− by TTMPP. The rate of reduction of the [FeBr4]− complex by TTMPP was measured in

KATRP 1.4 3.2 7.6 3.6 9.0

× × × × ×

10−8 34 10−8 34 10−9 10−9 10−10

a

Equivalents of TTMPP with respect to FeBr3. Polymerization without TTMPP (entry 1) was started with FeBr2 instead of FeBr3. b TPP was used instead of TTMPP at the following initial molar ratios: [FeBr2]:[FeBr3]:[TPP] = 0.86:0.14:1.10.

TTMPP. Despite the level of uncertainty of about ±30%, the higher KATRP suggests a weak interaction between FeII and TTMPP. Also listed in Table 1 are the associated values for BA polymerization at 100 °C, KATRP = 9.0 × 10−10, and for MMA polymerization at 60 °C, KATRP = 3.2 × 10−8,34 which were obtained via the same methodology using NMP instead of DMF as the highly polar solvent. The value of KATRP (entry 2) is slightly enhanced by the presence of TTMPP compared to the value measured with FeBr2 in the absence of any additive, KATRP = 1.4 × 10−8 (entry 1).34 The reported reaction enthalphies34 suggest that KATRP for the TTMPP-assisted ATRP of MMA should be around 6.0 × 10−7 at 100 °C, i.e., by about 2 orders of magnitude above the value for styrene. The increase of KATRP upon passing from BA to styrene and to C

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 2. (A) Reduction of [FeBr4]− with 0.5 equiv of TTMPP in DMF at 25, 60, 100, and 140 °C monitored by online spectroscopy between 15 000 and 9000 cm−1; see the inset. (B) Reduction of [FeBr4]− by 1.5 equiv of TTMPP or by 1.5 equiv of TPP monitored at 25 °C.

monomer-free systems between 25 and 140 °C via online VIS spectroscopy (Figure 2A). An equivalent amount of TBA-Br was added to a solution of 30 mM FeBr3 in DMF to allow for quantitative formation of [FeBr4]−. After recording the first spectrum, 0.5 equiv of TTMPP was added. The subsequent reduction of [FeBr4]− in the presence of TTMPP was monitored via the associated absorption bands between 15 000 and 9000 cm−1 as illustrated in the inset in Figure 2A (see ref 41 for details). At 25 °C, about 33% of the total [FeBr4]− is immediately reduced after injecting 0.5 equiv of TTMPP into the reaction medium. However, no further reaction occurs for several hours. Figure 2B illustrates that 1.5 equiv of TTMPP is needed for complete reduction of [FeBr4]− at 25 °C (cf. Figure S2). Interestingly, the analogous reaction using TPP as the reducing agent requires more than 100 h for the reduction of ca. 85% [FeBr4]− at 25 °C. The lower rate of reduction may be related to the weaker Lewis basic character of the unsubstituted triphenylphosphine. The resulting lower ratio of FeII:FeIII in TPP-mediated ATRP would yield a lower polymerization rate than with TTMPP (cf. eq 1), although KATRP is similar (see Table 1). This observation is consistent with the reported lowering of the ATRP rate with decreasing electron-donating character of the phosphine.12,13 As shown in Figure 2A, only 0.5 equiv of TTMPP is sufficient to almost quantitatively reduce [FeBr4]− at 140 °C within 4−5 h. TTMPP thus predominantly acts as a reducing agent for [FeBr4]− at elevated temperatures. At ambient temperature, more than stoichiometric amounts of the reducing agent are needed for complete reduction, which suggests that coordination of TTMPP to Fe also takes place. Because of the ability to reduce FeIII, the coordination of TTMPP should almost exclusively occur to the FeII complex. Support for such coordination is provided by the fact that an equimolar mixture of FeBr2 and TTMPP is insoluble in less polar environments, such as 2-butanone and MMA, even though the individual components are soluble.38 31 P NMR is particularly useful for deducing further information on the interaction of TTMPP and Fe. The spectra shown in Figure 3 were recorded in DMF-d7 at 25 °C; the 1H NMR spectra are shown in Figure S4. The chemical shift of TTMPP in the absence of Fe is −66.7 ppm.51 This signal disappears in a solution containing equivalent amounts of FeBr2 and TTMPP. It is assumed that the paramagnetism of the soformed FeII/TTMPP complex does not allow for the detection

Figure 3. 31P NMR spectra of TTMPP in a metal-free DMF-d7 solution (black), of 1 equiv of TTMPP with FeBr2 (red), of 2 equiv of TTMPP with FeBr2 (blue), and of 1.5 equiv of TTMPP with FeBr3 (green).

of the associated 31P NMR signal. The signal of the free ligand at −66.7 ppm returns when 2 equiv of TTMPP is added to FeBr 2 , indicating that only one TTMPP molecule is coordinated to FeII, most likely due to steric limitations. A fourth spectrum has been recorded for a mixture of FeBr3 and 1.5 equiv of TTMPP. A signal at −53.7 ppm appears which is due to the presence of TTMPP−Br+,52 i.e., the oxidized product of TTMPP associated with the reduction of FeIII. Again no signal is observed for the FeII/TTMPP complex, which should also be present. Upon the addition of more than 1.5 equiv TTMPP to FeBr3, the signal of the free ligand at −66.7 ppm is additionally observed (not shown). Dunbar and Quillevéré investigated the analogous reaction of FeCl3 with TTMPP and characterized the X-ray structure of the TTMPP-Cl+ (PV) product.36 In addition to the NMR analysis, we identified the bromine analogue TTMPP−Br+ via ESI-MS (MeCN), m/z (%) = 611.1; 613.1 (10) for the reduction of FeBr3 with an excess of TTMPP, m/z (%) = 533.2 (88). The reported X-ray structure for the FeII/TTMPP complex is [HTTMPP]2[Fe2Cl6].36 The protonation of TTMPP (PIII) occurs due to the presence of HCl, which is present in iron chloride as a contaminant.36 In agreement with the X-ray structure, the peak position of the NIR absorbance, determined for the dissolved FeII/TTMPP complex, is also indicative of a tetrahedral FeII complex (Figure S1).38 It is assumed that TTMPP predominantly occurs as nonprotonated ligand in our D

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

with 1 equiv of TTMPP at T > 100 °C, all [FeBr4]− is reduced in less than 2 min to a mixture of [FeIIBr3(Solv)]− and [FeIIBr3(TTMPP)]− species. The TTMPP ligand can be replaced by solvent and may reduce [FeBr4]−, resulting from radical termination and following the persistent radical effect. Such a continuous decrease of the [FeBr4]− deactivator content may, according to Scheme 1, produce large amounts of dead polymer, unless substoichiometric amounts of the catalyst are used. Such an unfavorable scenario may occur despite the formation of polymers with acceptable dispersity values well below 1.5. The consequences of catalyst regeneration in the phosphine-assisted ATRPs should be considered for the selection of suitable reaction conditions (vide inf ra). ATRP of acrylates is feasible with phosphine-containing Fe systems (cf. KATRP in Table 1). ATRPs with monomer conversion up to 86% and dispersities below 1.2 have been achieved4,6,7 although the kinetics may be complicated due to intramolecular transfer reactions, such as backbiting,59−62 and by the formation of macromonomer.63 An even more important factor is that acrylate-type radicals terminate predominantly via an FeII halide-catalyzed pathway, i.e., at a much faster rate than via conventional radical−radical termination.64 Therefore, improving the livingness of the polymerization is particularly important with this class of monomers. ATRP of acrylates should definitely be operated at lower Fe content to avoid FeII-catalyzed radical termination (FeII-CRT). We modified the overall ratio of reagents to achieve a high degree of chain-end functionality (CEF) despite the catalyst regeneration via TTMPP. ATRP of BA was carried out in anisole (25 vol %) at 100 °C and initial molar ratios of [BA]: [MBriB]:[FeBr3]:[TTMPP] = 100:1.00:0.10:0.15, i.e., with just 5 mM FeBr3. The [FeBr4]− deactivator may be reduced and subsequently be regenerated twice, but according to Scheme 1, each regeneration is accompanied by an equivalent loss in CEF. However, due to the presence of excess R-Br initiator, MBriB, compared to overall Fe, the loss of CEF remains below 20% of total Rn-Br according to stoichiometric considerations. We achieved 40% monomer conversion and poly(butyl acrylate) with a dispersity of Đ = 1.56. The higher dispersity is a consequence of lowering the Fe content. More active Fe catalysts with a higher KATRP are required for achieving lower dispersity and higher degrees of BA conversion at a similarly high level of preserved chain-end functionality and at possibly even lower Fe concentrations. An improved mechanistic insight into phosphine-based reducing agents is important for these potential further improvements. ATRP of MMA and styrene at ppm levels of the iron bromide catalyst is already feasible.23,24,65−67

solutions, since HCl is present only in subequivalent amounts. The nonprotonated TTMPP may coordinate directly to FeII, which most likely yields monomeric complexes of otherwise analogous composition (cf. eq 2a) as reported for the dimeric X-ray structure. Monomeric complexes have already been described for the structurally similar [FeCl3(NMP)]− and the [FeCl3(PMe3)]− species.53,54 Equation 2a illustrates the suggested mechanism for the reduction of [FeBr4]− by TTMPP at ambient temperature, assuming coordination of TTMPP to FeII. 2[Fe IIIBr4]− + 3TTMPP → 2[Fe IIBr3(TTMPP)]− + 1TTMPP−Br + + 1Br − (2a)

At higher temperatures, TTMPP decoordinates from [FeIIBr3(TTMPP)]− (see eq 2b). At 140 °C, the complex is apparently sufficiently labile, so only 0.5 equiv of TTMPP is required for complete reduction of [FeIIIBr4]− (see net eq 3). 2[Fe IIIBr4]− + 1[Fe IIBr3(TTMPP)]− solvent

⎯⎯⎯⎯⎯⎯→ 3[Fe IIBr3(Solv)]− + 1TTMPP−Br + + 1Br − (2b)

2[Fe IIIBr4]− + 1TTMPP solvent

⎯⎯⎯⎯⎯⎯→ 2[Fe IIBr3(Solv)]− + 1TTMPP−Br + + 1Br −

(3)

It is evident that the transformation of TTMPP to TTMPP− Br+ is a two-electron process with TTMPP−Br• probably being the intermediate radical. EPR spectra of TPP−Br• radicals were reported for annealed glasses and single crystals.55,56 We observed similar EPR features, which suggest the occurrence of TTMPP−Br• intermediates during the reaction of [FeBr4]− and TTMPP by recording EPR spectra of flash-frozen solutions at 140 K (Figure S5). An interaction of both TTMPP and TTMPP−Br• with FeII may contribute to the curvature of [FeBr4]− vs time trace observed at 140 °C (cf. Figure 2A) and may in turn make the determination of rate coefficients for the reduction of [FeBr4]− rather complicated. One may, however, conclude that TTMPP primarily acts as a reducing agent at the elevated temperatures typically used in Fe-mediated ATRPs. In principle, similar mechanistic scenarios should also apply to other phosphines. TPP also reduces [FeIIIX4]−, even though the reaction is much slower than with TTMPP. The reported coordination of TPP to FeII 11 may be beneficial in enhancing the tolerance of the FeII catalyst to the presence of functional groups.12 Since TPP is a weaker base, even FeIII/TPP complexes are sufficiently stable for detection by both X-ray and low-temperature EPR spectroscopy.57 In the absence of an alkyl halide, the TPP-mediated generation of activators by monomer addition (GAMA) should be considered.11 Consequences for the Livingness of Fe-Mediated ATRP. TTMPP appears to be an attractive additive for Febased ATRP, in particular, as it serves as a convenient reducing agent for the stable FeIII precursor. TTMPP is associated with minor toxicity concerns as opposed to the commonly used tin(II)-based reducing agents58 and exhibits better solubility in the organic phase than ascorbic acid. However, the amount of reducing agent needs to be carefully selected: 0.5 equiv of TTMPP may be sufficient to reduce [FeBr4]−, although mostly equimolar amounts of alkyl halide, FeBr3, and the phosphinebased reducing agent are used in ATRP. When ATRP is started

■

CONCLUSION Online vis/NIR spectroscopy in conjunction with 31P NMR analysis was applied to the study of Fe-based ATRP in the presence of phosphine additives. No significant effect on the ATRP equilibrium constant, KATRP, was observed upon addition of up to 1.5 equiv of phosphine relative to Fe. The focus of this study was on clarifying the role of highly Lewis basic TTMPP, which may coordinate to FeII, but primarily acts as a reducing agent for [FeIIIBr4]− when ATRP is carried out above 100 °C. The oxidized product is TTMPP-Br+ with the P(III) species thus being transformed into P(V). TPP is a weaker Lewis base and consequently a less effective reducing agent. Such mechanistic insights are important for selecting suitable E

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

(25) Xue, Z.; Noh, S. K.; Lyoo, W. S. Macromol. Res. 2007, 15, 302− 307. (26) Zhang, L. F.; Cheng, Z. P.; Tang, F.; Li, Q.; Zhu, X. L. Macromol. Chem. Phys. 2008, 209, 1705−1713. (27) Moineau, G.; Dubois, P.; Jerome, R.; Senninger, T.; Teyssie, P. Macromolecules 1998, 31, 545−547. (28) Ando, T.; Kamigaito, M.; Sawamoto, M. Macromolecules 1997, 30, 4507−4510. (29) di Lena, F.; Matyjaszewski, K. Prog. Polym. Sci. 2010, 35, 959− 1021. (30) Bedri, B.; Cikilmazkaya, M.; Tunca, U.; Hizal, G. Des. Monomers Polym. 2003, 6, 299−307. (31) Dutta, A.; Mahato, P. K.; Dass, N. N. J. Macromol. Sci., Chem. 1991, A28, 1001−1008. (32) Dutta, A.; Mahato, P. K.; Dass, N. N. Eur. Polym. J. 1991, 27, 465−469. (33) Xue, Z. G.; Linh, N. T. B.; Noh, S. K.; Lyoo, W. S. Angew. Chem., Int. Ed. 2008, 47, 6426−6429. (34) Schroeder, H.; Buback, M.; Matyjaszewski, K. Macromol. Chem. Phys. 2013, 215, 44−53. (35) Wada, M.; Higashizaki, S. J. Chem. Soc., Chem. Commun. 1984, 482−483. (36) Dunbar, K. R.; Quillevéré, A. Polyhedron 1993, 12, 807−819. (37) Wang, Y.; Kwak, Y.; Buback, J.; Buback, M.; Matyjaszewski, K. ACS Macro Lett. 2012, 1, 1367−1370. (38) Schroeder, H.; Buback, J.; Demeshko, S.; Matyjaszewski, K.; Meyer, F.; Buback, M. Macromolecules 2015, 48, 1981−1990. (39) Fischer, H. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 1885− 1901. (40) Gill, N. S. J. Chem. Soc. 1961, 3512−3515. (41) Tang, W.; Tsarevsky, N. V.; Matyjaszewski, K. J. Am. Chem. Soc. 2006, 128, 1598−1604. (42) Buback, M.; Kuchta, F. D. Macromol. Chem. Phys. 1995, 196, 1887−1898. (43) Buback, M.; Kurz, C. H.; Schmaltz, C. Macromol. Chem. Phys. 1998, 199, 1721−1727. (44) Beuermann, S.; Buback, M.; Russell, G. T. Macromol. Rapid Commun. 1994, 15, 351−355. (45) Zammit, M. D.; Davis, T. P.; Willett, G. D.; O’Driscoll, K. F. J. Polym. Sci., Part A: Polym. Chem. 1997, 35, 2311−2321. (46) Olaj, O. F.; Schnoll-Bitai, I. Monatsh. Chem. 1999, 130, 731− 740. (47) Tang, W.; Matyjaszewski, K. Macromolecules 2007, 40, 1858− 1863. (48) Gillies, M. B.; Matyjaszewski, K.; Norrby, P. O.; Pintauer, T.; Poli, R.; Richard, P. Macromolecules 2003, 36, 8551−8559. (49) Nanda, A. K.; Matyjaszewski, K. Macromolecules 2003, 36, 8222−8224. (50) Schroeder, H.; Yalalov, D.; Buback, M.; Matyjaszewski, K. Macromol. Chem. Phys. 2012, 213, 2019−2026. (51) Tschan, M. J. L.; Garcia-Suarez, E. J.; Freixa, Z.; Launay, H.; Hagen, H.; Benet-Buchholz, J.; van Leeuwen, P. W. N. M. J. Am. Chem. Soc. 2010, 132, 6463−6473. (52) Godfrey, S. M.; McAuliffe, C. A.; Pritchard, R. G.; Sheffield, J. M. J. Chem. Soc., Dalton Trans. 1998, 1919−1923. (53) Ding, K. Y.; Zannat, F.; Morris, J. C.; Brennessel, W. W.; Holland, P. L. J. Organomet. Chem. 2009, 694, 4204−4208. (54) Bleeke, J. R.; Hays, M. K. Organometallics 1987, 6, 486−491. (55) Berclaz, T.; Geoffroy, M.; Ginet, L.; Lucken, E. A. C. Chem. Phys. Lett. 1979, 62, 515−518. (56) Symons, M. C. R.; Petersen, R. L. J. Chem. Soc., Faraday Trans. 1979, 75, 210−219. (57) Walker, J. D.; Poli, R. Inorg. Chem. 1989, 28, 1793−1801. (58) Yang, D. F.; He, D.; Liao, Y. G.; Xue, Z. G.; Zhou, X. P.; Xie, X. L. J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 1020−1027. (59) Schroeder, H.; Buback, M. Macromolecules 2014, 47, 6645− 6651. (60) Plessis, C.; Arzamendi, G.; Alberdi, J. M.; van Herk, A. M.; Leiza, J. R.; Asua, J. M. Macromol. Rapid Commun. 2003, 24, 173−177.

ratios of initiator:Fe:phosphine to yield polymers with a high degree of CEF.

■

ASSOCIATED CONTENT

* Supporting Information S

Figures S1−S5. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/ acs.macromol.5b01049.

■

AUTHOR INFORMATION

Corresponding Author

*Fax +49 551 3912709, e-mail [email protected] (M.B.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS H.S. gratefully acknowledges a Ph.D. fellowship granted by the Fonds der Chemischen Industrie. Funding by the DFG (CHE 10-26060) and NSF (14-00052) is gratefully acknowledged.

■

REFERENCES

(1) Matyjaszewski, K. Macromolecules 2012, 45, 4015−4039. (2) Ouchi, M.; Terashima, T.; Sawamoto, M. Chem. Rev. 2009, 109, 4963−5050. (3) Tsarevsky, N. V.; Matyjaszewski, K. Chem. Rev. 2007, 107, 2270− 2299. (4) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101, 3689−3745. (5) Ouchi, M.; Terashima, T.; Sawamoto, M. Acc. Chem. Res. 2008, 41, 1120−1132. (6) Matyjaszewski, K.; Tsarevsky, N. V. J. Am. Chem. Soc. 2014, 136, 6513−6533. (7) Matyjaszewski, K.; Xia, J. H. Chem. Rev. 2001, 101, 2921−2990. (8) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. Rev. 2004, 104, 6217−6254. (9) Gopalaiah, K. Chem. Rev. 2013, 113, 3248−3296. (10) Poli, R.; Allan, L. E. N.; Shaver, M. P. Prog. Polym. Sci. 2014, 39, 1827−1845. (11) Xue, Z.; He, D.; Xie, X. Polym. Chem. 2015, 6, 1660−1687. (12) Nishizawa, K.; Ouchi, M.; Sawamoto, M. Macromolecules 2013, 46, 3342−3349. (13) Wang, Y.; Kwak, Y.; Matyjaszewski, K. Macromolecules 2012, 45, 5911−5915. (14) Xue, Z. G.; He, D.; Noh, S. K.; Lyoo, W. S. Macromolecules 2009, 42, 2949−2957. (15) Aoshima, H.; Satoh, K.; Umemura, T.; Kamigaito, M. Polym. Chem. 2013, 4, 3554−3562. (16) Xue, Z.; Linh, N. T. B.; Noh, S. K.; Lyoo, W. S. Angew. Chem., Int. Ed. 2008, 47, 6426−429. (17) He, D.; Xue, Z.; Khan, M. Y.; Noh, S. K.; Lyoo, W. S. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 144−151. (18) Zhang, L. F.; Cheng, Z. P.; Zhang, Z. B.; Xu, D. Y.; Zhu, X. L. Polym. Bull. 2010, 64, 233−244. (19) Matyjaszewski, K.; Wei, M. L.; Xia, J. H.; McDermott, N. E. Macromolecules 1997, 30, 8161−8164. (20) Khan, M. Y.; Chen, X. X.; Lee, S. W.; Noh, S. K. Macromol. Rapid Commun. 2013, 34, 1225−1230. (21) Satoh, K.; Aoshima, H.; Kamigaito, M. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 6358−6363. (22) Xue, Z. G.; Oh, H. S.; Noh, S. K.; Lyoo, W. S. Macromol. Rapid Commun. 2008, 29, 1887−1894. (23) Zhu, G. H.; Zhang, L. F.; Zhang, Z. B.; Zhu, J.; Tu, Y. F.; Cheng, Z. P.; Zhu, X. L. Macromolecules 2011, 44, 3233−3239. (24) Chen, X. X.; Khan, M. Y.; Noh, S. K. Polym. Chem. 2012, 3, 1971−1974. F

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (61) Asua, J. M.; Beuermann, S.; Buback, M.; Castignolles, P.; Charleux, B.; Gilbert, R. G.; Hutchinson, R. A.; Leiza, J. R.; Nikitin, A. N.; Vairon, J. P.; van Herk, A. M. Macromol. Chem. Phys. 2004, 205, 2151−2160. (62) Willemse, R. X. E.; van Herk, A. M.; Panchenko, E.; Junkers, T.; Buback, M. Macromolecules 2005, 38, 5098−5103. (63) Nikitin, A. N.; Hutchinson, R. A. Macromolecules 2005, 38, 1581−1590. (64) Wang, W.; Nikitin, A. N.; Hutchinson, R. A. Macromol. Rapid Commun. 2009, 30, 2022−2027. (65) Wang, Y.; Zhang, Y. Z.; Parker, B.; Matyjaszewski, K. Macromolecules 2011, 44, 4022−4025. (66) Mukumoto, K.; Wang, Y.; Matyjaszewski, K. ACS Macro Lett. 2012, 1, 599−602. (67) Zhang, L. F.; Miao, J.; Cheng, Z. P.; Zhu, X. L. Macromol. Rapid Commun. 2010, 31, 275−280.

G

DOI: 10.1021/acs.macromol.5b01049 Macromolecules XXXX, XXX, XXX−XXX