2-Mediated RDRP in Aqueous Solution - ACS Publications - American


2-Mediated RDRP in Aqueous Solution - ACS Publications - American...

1 downloads 71 Views 1MB Size

Article pubs.acs.org/Macromolecules

Kinetics of Fe−Mesohemin−(MPEG500)2‑Mediated RDRP in Aqueous Solution Sebastian Smolne,† Michael Buback,*,† Serhiy Demeshko,‡ Krzysztof Matyjaszewski,§ Franc Meyer,‡ Hendrik Schroeder,† and Antonina Simakova§ †

Institut für Physikalische Chemie, Georg-August-Universität Göttingen, Tammannstraße 6, D-37077 Göttingen, Germany Institut für Anorganische Chemie, Georg-August-Universität Göttingen, Tammannstraße 4, D-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: A speciation analysis for Fe−mesohemin−(MPEG500)2-mediated reversible-deactivation radical polymerization (RDRP) in aqueous solution was carried out by a combination of visible (vis) and 57Fe Mössbauer spectroscopy. The results were used within kinetic studies of ATRP and OMRP reactions via highly time-resolved EPR spectroscopy. ATRP control was effective with the rate coefficient for deactivation clearly exceeding the one for formation of organometallic species. Deactivation rate coefficients increase by more than 1 order of magnitude in passing from polymerization in 30 to 90 wt % H2O. Media with water contents of and above 70 wt % are well suited for controlled ATRP. The Fe−mesohemin catalyst provides an exceptionally high ATRP equilibrium constant even at ambient temperature, which approaches the one of highly active Cu catalysts.



INTRODUCTION

Fe-mediated atom-transfer radical polymerization (ATRP) is an attractive alternative to the extensively used Cu-catalyzed ATRP1,2 due to low potential toxicity, good biocompatibility, and the availability of iron.3−7 Fe-based ATRP in organic solvents has mostly been applied not only to styrene and methacrylates6−8 but also to nitriles7,9 and, with some limitations, to acrylates.10−14 The iron catalysts reported so far are mostly based on halides, phosphines, amines, and imines used as ligands.6−8,15 Iron-halide-mediated ATRP may even be performed in organic solvents without additional ligands.6−8,14,16 These ATRPs are however restricted to the monomers mentioned above and to reactions in organic phase.6 It was recently reported that porphyrin-based Fe catalysts allow for reversible-deactivation radical polymerization (RDRP) under biorelevant conditions, i.e., in aqueous phase and under ambient conditions.17−20 Protein-based ATRP ligands with iron−heme centers, such as horseradish peroxidase,20 catalase,18 and hemoglobin,21 were used to prepare high-molarmass polymers with narrow molar-mass distribution (MMD). Matyjaszewski et al. reported a modified FeIII protoporphyrin complex with an additional axial bromide ligand (Figure 1).17 Attached to this mesohemin−(MPEG500)2 complex was a methoxy poly(ethylene glycol) (MPEG) side arm for better solubility in aqueous phase. The two unsaturated alkene groups were hydrogenated. In the present study, RDRP with such Fe© XXXX American Chemical Society

Figure 1. Structure of the Fe-based mesohemin−(MPEG500) complex.

based mesohemin−(MPEG500)2 catalysts was the subject of a detailed mechanistic and kinetic analysis. Illustrated in Scheme 1 are the relevant equilibria involved in Fe-mediated RDRP. The blue box illustrates the ATRP mechanism. The reaction of the FeII complex, FeII/L, with an alkyl halide initiator, R-Br, yields the deactivator complex, BrFeIII/L, and a radical, R•. The scheme includes propagation of radicals and their termination to yield dead polymer. Received: August 14, 2016 Revised: September 30, 2016

A

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

FeIII/Br−mesohemin−(MPEG500)2 and FeIII/Cl−mesohemin− (MPEG500)2 were prepared as reported.17 Monomers, solvents, and liquid alkyl halide initiators were degassed by several freeze−pump−thaw cycles. The solutions for all experiments were prepared under an argon atmosphere. Size Exclusion Chromatography. The SEC analysis of poly(PEGMA) was performed at 35 °C with dimethylacetamide (DMAc) as the eluent (0.8 mL min−1 flow rate) and with toluene as the flowrate marker on an SEC system consisting of a 1260 Iso pump G1310B, an Agilent 1260 ALS G1329B autosampler, three PSS SDV columns (5 μm particle size; 2 × 103 and 30 Å pore sizes), and a refractive index detector (1260 RID G1262A). The SEC setup was calibrated against narrowly distributed poly(MMA) and poly(styrene) standards (M = 800−2 × 106 g mol−1, PSS). Vis Measurements. The samples for the vis measurements (on a Cary 300, Agilent) were prepared under an argon atmosphere and were carried out in sealed quartz cells of 5 or 10 mm path length. The extinction coefficients at 640 nm, required for estimating the catalyst concentration, were determined for solutions with known FeIII catalyst concentration, between 0.3 and 4.0 mM, but without the ATRP initiator, for each solvent composition under investigation. Mö ssbauer Spectroscopy. The samples were flash-frozen in liquid nitrogen. The spectra were obtained with a 57Co source embedded in a Rh matrix using an alternating constant acceleration Wissel Mössbauer spectrometer operated in the transmission mode and equipped with a closed-cycle helium cryostat (SHI 850, Janis). Isomer shifts are given relative to iron metal at ambient temperature. Lorentzian doublets have been fitted to the zero-field spectra using the Mfit program.41 SP-PLP-EPR Experiments. Full EPR spectra and radical concentration vs time traces measured at constant magnetic field after laser pulsing were recorded on a Bruker Elexsys E 500 series CW EPR spectrometer consisting of a microwave bridge, a microwave source, a detector, a cavity (ER 4122SHQE-LC, Version V1.1, Bruker), a console (Spectrometer Electronics) for electronic data processing, and two tunable magnets. Temperature control was achieved via an ER 4131VT unit (Bruker) by purging the sample cavity with nitrogen. The sample was irradiated by a XeF laser (LPX 210 iCC, Lambda Physik) at 351 nm with about 80 mJ/pulse. The EPR spectrometer and the laser source were synchronized by a Quantum Composers 9314 pulse generator (Scientific Instruments). Further details of the setup are described elsewhere.42 Sample Preparation for the SP-PLP-EPR Measurements. The purified monomer PEGMA and ultrapure water were degassed by several freeze−pump−thaw cycles. The required monomer−solvent mixtures containing about 20 mM Darocur 1173 and the ATRP catalyst FeIII−mesohemin−(MPEG500)2 at concentrations between 1 and 3 mM were prepared under an argon atmosphere. The highly polar samples were investigated in EPR flat cells. To prevent halide dissociation, an excess of 250 equiv of NaBr was added to solutions containing up to 50 wt % H2O, and 500 equiv of NaBr was added at higher water content. This addition of NaBr also applied to the vis measurements described below. Vis Measurements with Stopped-Flow Injection. The samples were prepared under an argon atmosphere and stored in sealed syringes. Prior to the experiment, the tubing and the reaction cell were purged with nitrogen. The stopped-flow experiments were performed in a setup with two syringes. Their contents (see below) were injected by a syringe driver (Bio-Logic μ-SFM 20) into a ball mixer (Berger-Ball technology mixers) and subsequently into the UV−vis cell (10 mm path length). The reaction temperature was controlled by an external cryostat (Huber CC-75 cryostat). The absorbance signal was measured between 400 and 800 nm via a diode array UV−vis spectrometer (J&M MCS-UVNIR500, range 190−1015 nm with a resolution of about 1 nm). The minimum integration time of 12 ms per spectrum was chosen. The setup was operated via the Biokine software, whereas data analysis was performed via the Specfit 32 global analysis software from Bio-Logic. The stopped-flow experiments were performed at

Scheme 1. Iron-Mediated Radical Polymerization with a Concurrent ATRP and OMRP Equilibriuma

Both reaction pathways involve the FeII−hemin activator complex and growing radicals, R•. The potential subsequent reaction of R-FeIII/ L (with another radical) also yields FeII/L and thus denotes an overall catalytic radical termination (CRT) process. a

Also, the FeII/L species may react with propagating radicals. This step is relevant for Fe-mediated organometallic radical polymerization (OMRP), as illustrated by the red box in Scheme 1. The resulting R-FeIII/L species may dissociate back to form FeII/L or may be part of catalytic radical termination (CRT) by reaction with another radical to also yield FeII/ L.22−25 Scheme 1 illustrates that the FeII/L species can participate in both reaction pathways. A potential interplay of ATRP and OMRP has already been discussed for several Fe-based systems6,22,26−32 as well as for Cu-based systems33 and was investigated by spectroscopic27 and computational means.34,35 Iron−porphyrin complexes are known to form stable organometallic species in organic solvents by the reaction of alkyl radicals with FeII or by other reactions.36−39 Because of the potential mechanistic and kinetic complexity induced by the simultaneous occurrence of ATRP and OMRP, the precise knowledge of the mechanism and of the individual rate coefficients is necessary to foster Fe-based RDRP mediated by mesohemin−(MPEG500)2 catalysts in aqueous solution. The present study addresses the detailed speciation analysis of the relevant Fe species via visible (vis) and Mössbauer spectroscopy40 and the determination of the relevant ATRP and OMRP parameters. The ATRP deactivation process was investigated via highly time-resolved EPR spectroscopy. By applying vis spectroscopy in conjunction with stopped-flow injection, the three parameters kdeact, kact (and thus KATRP), and the rate coefficient for the addition of radicals to FeII, kadd, were studied for monomer−water mixtures with variable water content and for monomer-free model systems. The kinetic experiments were accompanied by measuring molar-mass dispersity to identify suitable reaction conditions for wellcontrolled RDRP.



EXPERIMENTAL SECTION

Chemicals. The photoinitiator 2-hydroxy-2-methylpropiophenone (Darocur 1173, Aldrich) was used as received. Poly(ethylene glycol) methyl ether methacrylate (PEGMA, Mn ∼ 500 g mol−1, Aldrich) and poly(ethylene glycol) dimethyl ether (PEO) with molar mass Mn ∼ 500 g mol−1 were purified by passing through a flash column with neutral aluminum oxide (Type CG-20, Aldrich). The solutions were prepared with ultrapure (type I) water (resistivity 18.2 MΩ cm at 25 °C, total organic carbon 85%), NaBr (Aldrich, ≥99%), ethyl αchlorophenylacetate (EClPA, Aldrich, 97%), 2-hydroxyethyl 2bromoisobutyrate (HEMA-Br, Aldrich, 95%), ethyl α-bromophenylacetate (EBrPA, ABCR, 97%), and 2,2′-azobis[2-(2-imidazolin-2yl)propane] dihydrochloride (VA-44, Wako) were used as received. B

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules different flow rates to identify the optimum condition for mixing the two solutions. To achieve efficient mixing in the monomer-free model system, both syringes contained the solvent at identical composition, including the same amount of NaBr. As solvents, mixtures of 50 and 70 wt % H2O in PEO were used. A typical stopped-flow experiment was carried out with one syringe containing 0.9 mM of the mesohemin− (MPEG500)2 catalyst and 0.5 equiv of Na2S2O4. The Na2S2O4 was added to the aqueous solution as the last component to avoid decomposition into [SO2]•− radicals prior to the addition of the catalyst. The catalyst was reduced in situ before the kinetic measurement started. The second syringe contained ca. 20 mM of HEMA-Br.

Na2S2O4 has completely disappeared once the ATRP process starts. The measured absorbance spectra demonstrate that FeIII is completely reduced to Fe II even at the Na 2 S 2 O 4 concentration of 0.5 equiv used in the kinetic experiments. The black line in Figure 2 is assigned to the spectrum of the FeII/L complex, which was obtained by reduction of Br-FeIII/L with 0.6 equiv of Na2S2O4. The absorbance of the Br-FeIII/L complex at 577 nm completely disappeared and was replaced by the characteristic bands for FeII porphyrin complexes at around 466 and 500 nm.49,50 The ATRP activation of FeII/L with the alkyl halide initiator HEMA-Br yielded the green-line spectrum, which is almost identical to the one recorded for the initial Br-FeIII/L species. The immediate oxidation suggests that FeII/L is a very active ATRP complex, which will be more closely examined further below. To demonstrate the formation of the organometallic R-FeIII/ L species, the thermal initiator VA-44, which decomposes and produces propagating PEGMA radicals in the aqueous monomer solution, was added in a separate experiment instead of the alkyl halide. The reaction of the FeII/L complex with radicals was monitored at 65 °C (cf. Figure S2). The resulting spectrum, red line in Figure 2, was associated with the R-FeIII/L species. This species was stable at 65 °C for at least 15 min and much longer at 20 °C (cf. Figure S2). To confirm the spectroscopic assignments and to obtain direct information on the oxidation and spin states of the Fe species, 57Fe Mössbauer spectroscopy was applied.40 The experimental procedures for preparing the polymer samples subjected to Mössbauer spectroscopy were similar to the ones for vis spectroscopy, except that higher concentrations of each component were used to improve signal-to-noise quality. All Mössbauer spectra were recorded at 80 K on flash-frozen solutions (for Cl-FeIII/L, spectra were also recorded for solid material, see Figure S4), which should represent the Fe/L composition at ambient temperature. Because of the unfavorably large γ-capture cross section of bromide, the chloride derivative of the complex was used for the measurements. Previous investigations suggest that the chloride catalyst undergoes the same reactions as the bromide species.27,51,52 Unfortunately, the detection of the initial Cl-FeIII/L species posed some problems (see Figure S3), most likely due to the occurrence of intermediate spin relaxation.53,54 We were, however, able to detect the associated asymmetric signal for the solid substance (see Figure S4). In solution, the conversion of Cl-FeIII/L was monitored via vis spectroscopy, as discussed above. Illustrated in Figure 3A is the Mössbauer spectrum recorded at 80 K after the reduction of Cl-FeIII/L with a 2-fold excess of Na2S2O4 at ambient conditions. The associated Mössbauer parameters: isomeric shift, δ, quadrupole splitting, ΔEQ, line width, Γ, and relative intensity, are listed in Table 1. The overall spectrum was fitted by two subfunctions, which are assigned both to the FeII/L species, but to two different spin states, highspin S = 2 (gray function, δ = 0.99 mm s−1 and ΔEQ = 2.50 mm s−1) and low-spin S = 0 (black function, δ = 0.46 mm s−1 and ΔEQ = 0.27 mm s−1). While FeII porphyrin complexes with intermediate S = 1 spin state exhibit similar isomer shifts of 0.4−0.6 mm s−1, their quadrupole splitting is usually larger (0.4−0.6 mm s−1);55 thus, the presence of the intermediate spin species is unlikely. We assume that the occurrence of these two spin states is due to partial spin crossover at 80 K.56,57 As highspin complexes are favored at elevated temperature, it may be



RESULTS AND DISCUSSION Speciation Analysis. The speciation analysis was carried out to identify the relevant ATRP and OMRP species which may be formed during RDRP with the mesohemin− (MPEG500)2 catalyst. An excess of the associated sodium halide was added to prevent halide displacement from the Br-FeIII/L complex by H2O.43−47 The individual Fe species may be distinguished via vis spectroscopy of the so-called Q-bands centered at around 400 and 700 nm.36,48 Shown in Figure 2 are the vis spectra of Fe/L

Figure 2. Absorbance spectra in the visible range of individual mesohemin−(MPEG500)2 species in solution of PEGMA containing 50 wt % water at 22 °C. The blue line indicates the initial Br−FeIII/L species which was reduced by Na2S2O4 to yield the FeII/L spectrum (black line). The reaction of FeII/L with the thermal initiator VA-44 at 65 °C leads to the R-FeIII/L species (red line). The green curve was measured after oxidation of FeII/L with HEMA-Br to yield Br-FeIII/L. The resulting absorbance spectrum is not easily seen in the figure because of the strong overlap with the blue curve of the initial complex. The green curve clearly suggests that no significant amount of non-ATRP product (red curve) is formed. The reproducibility of the spectroscopic data is better than the difference between the green and blue curve in the region from 500 to 550 nm.

species in solution of PEGMA (Mn ∼ 500 g mol−1) with 50 wt % water at 22 °C. The blue curve represents the Br-FeIII/L species with the characteristic absorption band around 577 nm. To produce the ATRP-activating FeII/L complex from BrFeIII/L requires an efficient reducing agent. As illustrated in Figure S1 of the Supporting Information, ascorbic acid, which has already been successfully used for RDRP with the catalyst under investigation,17 only slowly and partially reduces mesohemin−(MPEG500)2. The stronger and water-soluble Na2S2O4 performed more efficiently as a reducing agent. The FeIII complex was dissolved in water first, before adding the dithionite. The reaction conditions were chosen such that C

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Illustrated in Scheme 2 is the procedure for measuring the rate coefficient kdeact. The experiment started in the reverse Scheme 2. SP-PLP-EPR Measurement of kdeacta

a

The starting components, i.e., the photoinitiator Darocur 1173, monomer M, and the Br-FeIII/L complex, are marked in red. The photoinitiator-derived radicals generated by the laser pulse rapidly add to monomer, thus producing primary PEGMA radicals, R•1, that may grow to propagating radicals, R•n, of chain length n. FeII/L and Rn-Br result from deactivation. Termination of radicals to dead polymer is also occurring.

Figure 3. 57Fe Mössbauer spectra recorded at 80 K on a flash-frozen solution of PEGMA containing 50 wt % H2O. (A) Spectrum of the FeII/L low-spin species (black) and the FeII/L high-spin species (gray) after reaction of Cl-FeIII/L with 2 equiv of Na2S2O4. (B) Spectrum of R-FeIII/L recorded after the reaction of FeII/L with VA-44 for 20 min at 65 °C. The associated Mössbauer parameters are listed in Table 1.

ATRP fashion58 with the catalyst in the higher oxidation state, Br-FeIII/L.52,59 The starting materials are marked in red. Darocur 1173 acts as the water-soluble photoinitiator for producing primary radicals which rapidly add to monomer molecules, M. The propagating radicals, Rn•, react with BrFeIII/L to generate alkyl halide, Rn-Br, and FeII/L. As detailed elsewhere,59 the technique is favorable in that the laser pulse intensity and the number of applied laser pulses may be selected such as to generate only small amounts of Rn-Br and FeII/L. Thus, the reverse reaction of ATRP activation and organometallic reactions of FeII/L play no significant role. The measurement of kdeact is particularly sensitive in PEGMA polymerization, as deactivation occurs at a much faster rate than termination. The system under investigation contains 1.0 mM FeIII porphyrin bromide complex, Br−FeIII/L, and 20 mM Darocur for PEGMA−H2O mixtures with water contents from 30 to 90 wt %. The NaBr content was chosen depending on water concentration as described in the Experimental Section. The optimum field position for monitoring PEGMA radical concentration was identified from stationary experiments prior to the time-resolved analysis as detailed elsewhere (cf. Figure S5).60 Shown in Figure 4 are four [PEGMA•] vs time traces measured at 20 °C in solutions of PEGMA with 30, 50, and 70 wt % water and of one trace recorded in the absence of the iron complex for PEGMA in a mixture with 50 wt % water. The concentration of PEGMA radicals increased almost instantaneously at t = 0, when the single laser pulse was applied. The radical decay in the absence of Fe (black line) was entirely due to radical−radical termination. The decrease of radical concentration was by up to 10 times faster in the presence of Br-FeIII/L, indicating efficient ATRP deactivation. ATRP deactivation may be treated as a pseudo-first-order reaction, since the Br-FeIII/L concentration largely exceeds the one of PEGMA radicals.59 The slope of the ln([PEGMA•]0/ [PEGMA•]) vs time data from the SP-PLP-EPR measurement thus yields the product kdeact[Br-FeIII/L] according to eq 1. Illustrated in Figure 5 is the ln([PEGMA•]0/[PEGMA•]) vs time correlation measured during polymerization of PEGMA in a mixture with 50 wt % H2O. A straight line may be fitted to the data except for the very initial time period, where the radical concentration is high and radical−radical termination may interfere with deactivation (for details see ref 59).

Table 1. Mössbauer Parameters for FeII/L Associated with the Spectra Shown in Figure 3a

(A) (A) (B)

iron species

δ (mm s−1)

ΔEQ (mm s−1)

Γ (mm s−1)

rel conc (%)

FeII/L S = 0 FeII/L S = 2 R-FeIII/L

0.46 0.99 0.36

0.27 2.5 0.53

0.35 0.37 0.69

66 34 100

δ, ΔEQ, and Γ refer to isomeric shift, quadrupole splitting, and line width, respectively. The spectra were recorded at 80 K. a

anticipated that the FeII/L high-spin species is the dominant one, which is also why only a single species was observed by vis spectroscopy at 20 °C (cf. Figure 2). Moreover, subsequent reactions using this FeII/L solution provided only a single Fe species, i.e., the R-FeIII/L species described above. Shown in Figure 3B is the Mössbauer spectrum assigned to R-FeIII/L as the single Fe species obtained by reaction for 20 min of FeII/L with propagating radicals from initiation by VA44 (t1/2 ≈ 1 h) at 65 °C. The fitted parameters of δ = 0.36 mm s−1 and ΔEQ = 0.53 mm s−1 and the asymmetric peak shape clearly differ from the ones of the Cl-FeIII/L species recorded in bulk (cf. Figure S4). R-FeIII/L should be a high-spin S = 5/2 species because the isomer shift for low-spin S = 1/2 iron(II) porphyrin complexes is typically below 0.3 mm−1, and the quadrupole splitting for the intermediate S = 3/2 species is typically very high (up to 4.3 mm s−1).55 The speciation analysis demonstrates that both ATRP and organometallic species may occur. The purpose of the subsequent kinetic analysis was to check for a preference of either of the two mechanisms presented in Scheme 1. Measurement of ATRP Deactivation. To study the effect of water concentration on the rate of ATRP deactivation is crucial for well-controlled ATRP. Highly time-resolved EPR spectroscopy in conjunction with laser-pulse-induced radical production, i.e., the so-called SP-PLP-EPR technique, is well suited for measuring this fast deactivation step.42,52 D

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 6. Dependence on water content (in wt %) of the deactivation rate coefficient in mesohemin−(MPEG500)2-mediated deactivation of PEGMA radicals at 20 °C.

Figure 4. Normalized PEGMA radical concentration vs time profiles at 20 °C measured via the SP-PLP-EPR method with the laser pulse being applied at t = 0. The black line represents the radical decay by conventional radical−radical termination. The colored lines represent the radical decay in the presence of 1 mM mesohemin−(MPEG500)2 for solutions of PEGMA containing 30, 50, and 70 wt % H2O.

is in agreement with what has been reported for other types of Fe and Cu catalysts.59,61,62 The highest value of kdeact = 3.2 × 105 L mol−1 s−1, measured at a water content of 90 wt %, is close to the one reported for Cu-mediated ATRP deactivation with 1,1,4,7,10,10hexamethyltriethylenetetramine (HMTETA) as the ligand to Cu.61 The associated value measured for dodecyl methacrylate (DMA) in solution of 15 wt % acetonitrile is kdeact = 8 × 105 L mol−1 s−1.61 The kdeact value for mesohemin−(MPEG500)2 with PEGMA radicals in 90 wt % H2O is also close to the value reported for the [FeIIIBr4]− catalyst with MMA radicals, kdeact = 5.0 × 105 L mol−1 s−1, measured at 60 °C.63 Similarly, with amine−bis(phenolate)iron catalysts elevated temperatures have to be applied to reach sufficiently high deactivation rates.59 The mesohemin−(MPEG500)2 catalyst is the first Fe-based catalyst, for which adequately high values of kdeact for well-controlled ATRPs were attained already at ambient temperature. As indicated by the data in Figure 6, water contents of at least 70% are required to yield suitably high kdeact values of about 1.0 × 105 L mol−1 s−1. Most of the reported RDRPs of PEGMA were indeed carried out at such high water contents,17,47,64,65 perhaps for the reason that the lower viscosity in the water-rich regime facilitates processing of the polymerization mixture. A striking effect is the strong dependency of kdeact on the water content, which was not observed in studies into Cu-catalyzed ATRP in aqueous phase.44 Previous studies into the redox potential of porphyrin-type FeII/FeIII complexes indicate that the reduction is facilitated in solvents with a high dielectric constant,66−68 which is in agreement with the observed increase in ATRP deactivation rate toward higher water content, in particular at water concentrations above 60 wt %. The SP-PLP-EPR technique may also be used for studies into OMRP deactivation, i.e., to the reaction of FeII/L with propagating radicals. Such measurements are easily performed whenever FeII/L species are available as the starting material. To deduce such rate coefficients, monomer-free model systems were examined by vis spectroscopy. Simultaneous Measurement of Kinetic Parameters for ATRP and OMRP. Vis spectroscopy was applied to measure within a single experiment both the ATRP-specific coefficients kact and kdeact as well as the rate coefficient for the formation of the organometallic species via addition of growing radicals to FeII/L, kadd. The measurement was based on monitoring the reaction of the alkyl halide, R-Br, with FeII/L. The measure-

Figure 5. ln([R•]0/[R•]) vs time trace from SP-PLP-EPR measured on the deactivation by the mesohemin−(MPEG500)2 complex in a H2O−PEGMA mixture containing 50 wt % H2O at 20 °C. The signal in the very early time region indicates a significant contribution of radical−radical termination. A straight line has been fitted to the data for the extended time regime in which ATRP deactivation controls the decay of PEGMA radical concentration.

d([R•]0 /[R•]) = kdeact[Br−Fe III /L] dt

(1)

The Br-FeIII/L concentration required for estimating kdeact was deduced from vis spectroscopy, as this species could not be detected via X-band EPR spectroscopy at ambient temperatures, most likely because of fast spin relaxation. The Br-FeIII/ L concentration measured before and after applying the sequence of laser pulses differs by less than 10%. The arithmetic mean of these two measured Br-FeIII/L concentrations is used for estimating kdeact. The value deduced from the data in Figure 5 is kdeact = 1.3 × 104 L mol−1 s−1 for 20 °C in an aqueous solution of PEGMA in 50 wt % water. Depicted in Figure 6 are the kdeact values for mesohemin− (MPEG500)2-mediated deactivation at 20 °C for PEGMA in solutions containing 30, 50, 70, 80, and 90 wt % H2O. kdeact increases by more than 1 order of magnitude, i.e., from 9.5 × 103 to 3.2 × 105 L mol−1 s−1, upon passing from 30 to 90 wt % H2O. The absolute values of kdeact suggest that deactivation is chemically controlled, even at high water content. This finding E

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

in absorbance at 640 nm was used for measuring the Br-FeIII/L concentration vs time traces. Shown in Figure 8 is the Br-FeIII/L concentration vs time trace (black line) for the reaction of 0.59 mm FeII−

ment is more sensitive to kact and kdeact upon using an excess of alkyl halide, R-Br, as higher R-Br content favors the ATRP over the OMRP pathway according to Scheme 1. Such an excess of R-Br is more easily achieved in a monomer-free model system,51 where PEGMA was replaced by poly(ethylene glycol) dimethyl ether (PEO, Mn ∼ 500 g mol−1), which may be considered as a monomer model without reactive methacrylate moiety. The absence of monomer also simplifies the kinetics to eqs 2−5, i.e., to activation, deactivation, termination, and the organometallic addition reaction, assuming this latter reaction to be irreversible on the time scale of the experiment. kact

Fe II /L + R−Br ⎯→ ⎯ Br−Fe III /L + R• kdeact

Br−Fe III /L + R• ⎯⎯⎯⎯→ Fe II /L + R−Br kt

R• + R• → R−R

(2) (3) (4)

kadd

Fe II /L + R• ⎯⎯→ R−Fe III /L

(5) II

Prior to the experiment, the Fe /L complex was prepared in situ via the reduction with 0.5 equiv of Na2S2O4. The vis monitoring was on the reaction of FeII/L with HEMA-Br acting as the alkyl halide initiator, R-Br. The so-produced Br-FeIII/L may be quantified via the absorbance of the d−d transition at around 640 nm (cf. Figure S6). As the formation of Br-FeIII/L with the highly active mesohemin−(MPEG500)2 complex was expected to take less than 1 min, a stopped-flow rapid-mixing device was used in conjunction with vis-spectroscopic monitoring. One injection syringe contains the solution of the reduced FeII/L complex, the other contains the HEMA-Br solution. The low viscosity of the monomer-free (and thus polymer-free) model system further assists efficient mixing. Shown in Figure 7 are the vis spectra (Q-band region) for the reaction of FeII/L with HEMA-Br in solution of water (50 wt %) with PEO at 20 °C. The red curve represents the FeII/L species at the beginning of the reaction, and the black curve is measured after formation of the Br-FeIII/L species. The increase

Figure 8. PREDICI modeling of the [Br-FeIII/L] vs time profiles for a mixture of PEO with 70 wt % water at 20 °C assuming different values of kadd; kdeact and Kmodel were kept constant.

mesohemin−(MPEG500)2 and 18 mM HEMA-Br in a mixture of PEO with 70 wt % water at 20 °C. kact, kdeact, and kadd were estimated in a stepwise fashion based on the three steps detailed below. Within the first two steps, it was assumed that the [Br-FeIII/L] vs time trace was controlled entirely by kact, kdeact, and kt whereas kadd could be neglected. Two reaction stages can be distinguished: an initial pre-equilibrium period and an equilibrium state toward the end of the reaction (cf. ref 63). First, the activation−deactivation equilibrium constant, referred to as Kmodel, was estimated from the equilibrium state using the literature procedure,69 in which Kmodel was deduced via the modified Fischer eq 6. To remain consistent with previous notations, I0 refers to [HEMA-Br]0, C0 to [FeII/L]0, and Y to [Br-FeIII/L].69 The associated plot of F([Y]) vs t is shown in Figure S7). The estimate of kt for eq 6 was based on diffusion rate70 which was assumed to be proportional to the measured inverse viscosity, i.e., to fluidity.71 F (Y ) =

∫0

Y

Y2 dY = 2ktK model 2t (I0 − Y )2 (C0 − Y )2

(6)

The so-obtained Kmodel values should be below the true values, since the F[Y] function did not consider the formation of RFeIII/L via kadd. For the system under investigation, the associated discrepancy turned out to be below a factor of 2. As detailed in the Supporting Information, the initial value of Kmodel = 7 × 10−5, deduced from the F([Y]) function, was refined within the subsequent steps of the modeling procedure. In the second step, kact and kdeact were estimated from the pre-equilibrium data. kact was substituted by Kmodelkdeact such that kdeact was the only parameter to be fitted via the program package PREDICI. The pre-equilibrium data shown in Figure 8 were best represented by kdeact = 3 × 105 L mol−1 s−1 and thus by kact = Kmodelkdeact = 35 L mol−1 s−1 (see Supporting Information for further details). The third evaluation step took the overall Fe concentration into account. In case of a reaction which was entirely ATRP-

Figure 7. Vis spectra recorded during the reaction of 0.4 mm FeII/L with HEMA-Br in solution of water (50 wt %) with PEO at 20 °C. The red curve represents the initial spectrum of FeII/L whereas the black spectrum of the Br-FeIII/L species was taken after the reaction. The arrows at 535 and 640 nm indicate the direction of change in absorption with time. The absorbance at around 640 nm was used for deducing the Br-FeIII/L concentration vs time profile. F

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

based complexes.52 This high catalytic activity is favorable for ATRP at low amounts of the catalyst. Moreover, Kmodel is similar to the values reported for methacrylate-type radicals and active Cu catalysts such as Cu/Me6TREN (Me6TREN = tris(2dimethylaminoethyl)amine) in organic phase74 and Cu catalysts in aqueous solution.41 The final section presents the potential of the mesohemin−(MPEG500)2 catalyst and the impact on the RDRP kinetics. Impact of Water Content on the Polymerization. To determine the influence of water content and of the associated variation of kdeact on dispersity, a set of RDRPs was carried out with PEGMA. The reaction conditions were similar to the ones reported by Simakova et al.;17 i.e., ascorbic acid was used as the reducing agent to allow for slow and continuous regeneration of the FeII/L species. The polymerizations were carried out at 25 °C with 2 mM FeIII−mesohemin−(MPEG500)2, 2 mM ethyl α-bromophenylacetate as the alkyl halide, and 4 mM ascorbic acid in various PEGMA/H2O mixtures. The experiments were carried out with an excess of 50 equiv of NaBr. Shown in the lower part of Figure 9 are the dispersities measured via SEC at about 80% monomer conversion. The

controlled, all FeII/L should be rapidly transformed into BrFeIII/L because of the high Kmodel. As illustrated in Figure 8, the Br-FeIII/L concentration reached a maximum concentration of 0.50 mM, which was, however, slightly below the overall Fe concentration of 0.59 mM. The discrepancy of about 20% should be due to the formation of R-FeIII/L. This additional piece of information may be used for the refinement of kact and kdeact and to estimate kadd during the final modeling of the [BrFeIII/L] vs t trace. Illustrated in Figure 8 is the close agreement of modeled and measured data by lowering and increasing kadd by a factor of 5 and 2.2, respectively. Further modeling experiments confirmed that the iterative analysis is similarly sensitive toward the variation of kact or kdeact as toward kadd (cf. Figure S8). The uncertainties adopted for the values given in Table 2 also Table 2. Values for Kmodel, kact, kdeact, and kadd Estimated for the Mesohemin−(MPEG500)2 at 20 °C via PREDICI Modeling Kmodel kact/L mol−1 s−1 kdeact/L mol−1 s−1 kadd/L mol−1 s−1

50 wt % H2O/PEO

70 wt % H2O/PEO

(1.2 ± 0.5) × 10−4 7.4 ± 1.9 (6.2 ± 1.7) × 104 (1.3 ± 0.9) × 104

(1.0 ± 0.3) × 10−4 21 ± 6 (2.1 ± 0.6) × 105 (4.9 ± 1.2) × 104

account for the uncertainties due to the calibration and to the mixing of the two solutions. The final parameters resulting from the three-step fitting procedure are kact = 21 L mol−1 s−1 and kdeact = 2.1 × 105 L mol−1 s−1 (see Table 2) is obtained with kadd = 4.9 × 104 L mol−1 s−1 for the PEO model system with 70 wt % water. Irrespective of the water content, kadd is by about a factor of 5 lower than kdeact. Moreover, the ATRP equilibrium is strongly shifted to the Br-FeIII/L side such that the ratio of Br-FeIII/L to FeII/L becomes as large as 100:1, thus further enhancing BrFeIII/L-mediated ATRP deactivation relative to OMRP deactivation mediated by FeII/L. Under polymerization conditions such as the ones discussed in the next section, ATRP deactivation is by up to 500 times faster than OMRP deactivation. With the catalyst under investigation, ATRP control is thus far more efficient than OMRP control, which is an important piece of information for the proper selection of reaction conditions (see below). Listed in Table 2 are the Kmodel, kact, kdeact, and kadd values estimated for mixtures with 50 and 70 wt % H2O in PEO at 20 °C. The values of kdeact increase from 6.2 × 104 L mol−1 s−1 at 50 wt % H2O to 2.1 × 105 L mol−1 s−1 at 70 wt % H2O. The absolute values of kdeact for the monomer-free model system exceed the ones for the polymerization system, obtained via SPPLP-EPR, by a factor of 4. The discrepancy is assigned to the so-called back-strain effect.72,73 As a consequence of the penultimate α-methyl group on the polymeric backbone, a steric strain is induced which hinders the addition of bromine from Br-FeIII/L to the radical and thus reduces kdeact. In the model system, the methacrylate ATRP initiator has no penultimate unit to induce such steric strain. Interestingly, the Kmodel values for 50 and 70 wt % H2O are almost identical, which indicates that the variation of kdeact discussed above is counterbalanced by a similar effect on kact. It should further be noted that Kmodel for the mesohemin− (MPEG500)2 system measured at ambient conditions is by orders of magnitude above the values reported for other Fe-

Figure 9. Upper figure presents the ratio of kp[M]/(kdeact[Br-FeIII/L]) as a function of water content of the reaction mixture. The lower figure displays the variation with water content of dispersity for the mesohemin−(MPEG500)2-mediated ATRP of PEGMA. The PEGMA polymerization was carried out with 2 mM mesohemin−(MPEG500)2, 2 mM ethyl α-bromophenylacetate, and 4 mM ascorbic acid at 25 °C.

upper part of Figure 9 illustrates the associated data for kp[M]/ (kdeact[Br-FeIII/L]), i.e., the number of propagation steps prior to the occurrence of a deactivation step. With the exception of the value for kp,60 the kinetic quantities were experimentally determined within the present study. Dispersity decreases from 1.85 at 30 wt % H2O to 1.25 at 80 wt % H2O, as expected from the measured increase in kdeact upon enhancing water content. kp[M]/(kdeact[Br-FeIII/L]) decreases from 80 at 30 wt % H2O to about unity at 80 wt % H2O. The latter value indicates efficient ATRP control. This finding is consistent with the linear correlation of molar mass and monomer conversion observed with FeIII−mesohemin-catalyzed PEGMA polymerization under similar conditions by Simakova et al.17



CONCLUSIONS The measurement of kdeact, kact, and kadd provides a detailed understanding of Fe−mesohemin−(MPEG 500) 2-mediated RDRP of methacrylates in aqueous solution. Polymerization is predominantly controlled by the ATRP mechanism due to G

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

(8) di Lena, F.; Matyjaszewski, K. Transition metal catalysts for controlled radical polymerization. Prog. Polym. Sci. 2010, 35, 959− 1021. (9) Hou, C.; Qu, R.; Ji, C.; Wang, C.; Wang, C. Synthesis of polyacrylonitrile via reverse atom transfer radial polymerization (ATRP) initiated by diethyl 2,3−dicyano−2,3−diphenylsuccinate, FeCl3, and triphenylphosphine. Polym. Int. 2006, 55, 326−329. (10) Kawamura, M.; Sunada, Y.; Kai, H.; Koike, N.; Hamada, A.; Hayakawa, H.; Jin, R. H.; Nagashima, H. New Iron(II) Complexes for Atom-Transfer Radical Polymerization: The Ligand Design for Triazacyclononane Results in High Reactivity and Catalyst Performance. Adv. Synth. Catal. 2009, 351, 2086−2090. (11) Aoshima, H.; Satoh, K.; Umemura, T.; Kamigaito, M. A simple combination of higher-oxidation-state FeX3 and phosphine or amine ligand for living radical polymerization of styrene, methacrylate, and acrylate. Polym. Chem. 2013, 4, 3554−3562. (12) Schroeder, H.; Matyjaszewski, K.; Buback, M. Kinetics of FeMediated ATRP with Triarylphosphines. Macromolecules 2015, 48, 4431−4437. (13) Wang, Y.; Kwak, Y.; Matyjaszewski, K. Enhanced Activity of ATRP Fe Catalysts with Phosphines Containing Electron Donating Groups. Macromolecules 2012, 45, 5911−5915. (14) Teodorescu, M.; Gaynor, S. G.; Matyjaszewski, K. Halide anions as ligands in iron-mediated atom transfer radical polymerization. Macromolecules 2000, 33, 2335−2339. (15) Ouchi, M.; Terashima, T.; Sawamoto, M. Transition MetalCatalyzed Living Radical Polymerization: Toward Perfection in Catalysis and Precision Polymer Synthesis. Chem. Rev. 2009, 109, 4963−5050. (16) Wang, Y.; Matyjaszewski, K. ATRP of MMA in Polar Solvents Catalyzed by FeBr2 without Additional Ligand. Macromolecules 2010, 43, 4003−4005. (17) Simakova, A.; Mackenzie, M.; Averick, S. E.; Park, S.; Matyjaszewski, K. Bioinspired Iron-Based Catalyst for Atom Transfer Radical Polymerization. Angew. Chem., Int. Ed. 2013, 52, 12148− 12151. (18) Ng, Y. H.; di Lena, F.; Chai, C. L. L. PolyPEGA with predetermined molecular weights from enzyme-mediated radical polymerization in water. Chem. Commun. 2011, 47, 6464−6466. (19) Ng, Y. H.; di Lena, F.; Chai, C. L. L. Metalloenzymatic radical polymerization using alkyl halides as initiators. Polym. Chem. 2011, 2, 589−594. (20) Sigg, S. J.; Seidi, F.; Renggli, K.; Silva, T. B.; Kali, G.; Bruns, N. Horseradish Peroxidase as a Catalyst for Atom Transfer Radical Polymerization. Macromol. Rapid Commun. 2011, 32, 1710−1715. (21) Silva, T. B.; Spulber, M.; Kocik, M. K.; Seidi, F.; Charan, H.; Rother, M.; Sigg, S. J.; Renggli, K.; Kali, G.; Bruns, N. Hemoglobin and Red Blood Cells Catalyze Atom Transfer Radical Polymerization. Biomacromolecules 2013, 14, 2703−2712. (22) Poli, R. New Phenomena in Organometallic-Mediated Radical Polymerization (OMRP) and Perspectives for Control of Less Active Monomers. Chem. - Eur. J. 2015, 21, 6988−7001. (23) Poli, R. Relationship between one-electron transition-metal reactivity and radical polymerization processes. Angew. Chem., Int. Ed. 2006, 45, 5058−5070. (24) Schroeder, H.; Buback, M. SP−PLP−EPR Measurement of Iron-Mediated Radical Termination in ATRP. Macromolecules 2014, 47, 6645−6651. (25) Wang, Y.; Soerensen, N.; Zhong, M. J.; Schroeder, H.; Buback, M.; Matyjaszewski, K. Improving the “Livingness” of ATRP by Reducing Cu Catalyst Concentration. Macromolecules 2013, 46, 683− 691. (26) Poli, R. Radical Coordination Chemistry and Its Relevance to Metal-Mediated Radical Polymerization. Eur. J. Inorg. Chem. 2011, 10, 1513−1530. (27) Schroeder, H.; Lake, B. R. M.; Demeshko, S.; Shaver, M. P.; Buback, M. A Synthetic and Multispectroscopic Speciation Analysis of Controlled Radical Polymerization Mediated by Amine−Bis(phenolate)iron Complexes. Macromolecules 2015, 48, 4329−4338.

the high ATRP equilibrium constant and to the relatively high deactivation rate coefficient, kdeact, which clearly exceeds the rate coefficient for organometallic addition, kadd. ATRP additionally benefits from an excess of R-Br with respect to Fe thus preventing the formation of significant amounts of organometallic species. The high KATRP allows for the application of advanced ATRP techniques, which operate at very low iron levels and use slow catalyst regeneration, e.g., by ascorbic acid. In contrast, Na2S2O4 quickly reduces the associated Br-FeIII/L species, which is useful for the effective in situ formation of FeII/L for kinetic experiments. RDRP with the Fe−mesohemin−(MPEG500)2 complex should be preferably carried out at water contents of at least 70 wt % due to the beneficial increase of kdeact in water-rich solutions. The high KATRP of about 1 × 10−4 found for the monomerfree model system should be even higher for PEGMA polymerizations due to the back-strain effect.72,73 Because of the high catalyst activity for methacrylates, Fe−mesohemin− (MPEG500)2-mediated ATRP of acrylates should also be possible.17 High activity and fast deactivation together with the excellent stability of porphyrins may even enable ATRP of acidic monomers.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.6b01774. Further visible, Mössbauer, and EPR spectroscopic data, analysis of the F([Y]) function, and further modeling results (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] (M.B.). Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors gratefully acknowledge financial support by the Deutsche Forschungsgemeinschaft, the State of Lower Saxony (International PhD Program “Catalysis for Sustainable Synthesis CaSuS”), and NSF CHE 1400052 and 1026060.



REFERENCES

(1) Matyjaszewski, K. Atom Transfer Radical Polymerization (ATRP): Current Status and Future Perspectives. Macromolecules 2012, 45, 4015−4039. (2) Matyjaszewski, K.; Xia, J. H. Atom transfer radical polymerization. Chem. Rev. 2001, 101, 2921−2990. (3) Enthaler, S.; Junge, K.; Beller, M. Sustainable metal catalysis with iron: From rust to a rising star? Angew. Chem., Int. Ed. 2008, 47, 3317− 3321. (4) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Iron−catalyzed reactions in organic synthesis. Chem. Rev. 2004, 104, 6217−6254. (5) Gopalaiah, K. Chiral Iron Catalysts for Asymmetric Synthesis. Chem. Rev. 2013, 113, 3248−3296. (6) Poli, R.; Allan, L. E. N.; Shaver, M. P. Iron-mediated reversible deactivation controlled radical polymerization. Prog. Polym. Sci. 2014, 39, 1827−1845. (7) Xue, Z.; He, D.; Xie, X. Iron-catalyzed atom transfer radical polymerization. Polym. Chem. 2015, 6, 1660−1687. H

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (28) Shaver, M. P.; Allan, L. E. N.; Rzepa, H. S.; Gibson, V. C. Correlation of metal spin state with catalytic reactivity: Polymerizations mediated by alpha-diimine-iron complexes. Angew. Chem., Int. Ed. 2006, 45, 1241−1244. (29) Shaver, M. P.; Allan, L. E. N.; Gibson, V. C. Organometallic intermediates in the controlled radical polymerization of styrene by alpha-diimine iron catalysts. Organometallics 2007, 26, 4725−4730. (30) Gibson, V. C.; O’Reilly, R. K.; Wass, D. F.; White, A. J. P.; Williams, D. J. Polymerization of methyl methacrylate using fourcoordinate (alpha-diimine)iron catalysts: Atom transfer radical polymerization vs catalytic chain transfer. Macromolecules 2003, 36, 2591−2593. (31) Allan, L. E. N.; MacDonald, J. P.; Reckling, A. M.; Kozak, C. M.; Shaver, M. P. Controlled Radical Polymerization Mediated by Amine− Bis(phenolate) Iron(III) Complexes. Macromol. Rapid Commun. 2012, 33, 414−418. (32) Schroeder, H.; Buback, M.; Shaver, M. P. Kinetics of Amine− Bis(phenolate) Iron-Mediated ATRP Up to High Pressure. Macromolecules 2015, 48, 6114−6120. (33) Schröder, K.; Konkolewicz, D.; Poli, R.; Matyjaszewski, K. Formation and Possible Reactions of Organometallic Intermediates with Active Copper(I) Catalysts in ATRP. Organometallics 2012, 31, 7994−7999. (34) Poli, R.; Shaver, M. P. Atom Transfer Radical Polymerization (ATRP) and Organometallic Mediated Radical Polymerization (OMRP) of Styrene Mediated by Diaminobis(phenolato)iron(II) Complexes: A DFT Study. Inorg. Chem. 2014, 53, 7580−7590. (35) Poli, R.; Shaver, M. P. ATRP/OMRP/CCT Interplay in Styrene Polymerization Mediated by Iron(II) Complexes: A DFT Study of the alpha−Diimine System. Chem. - Eur. J. 2014, 20, 17530−17540. (36) Guilard, R.; Boisseliercocolios, B.; Tabard, A.; Cocolios, P.; Simonet, B.; Kadish, K. M. Electrochemistry and Spectroelectrochemistry of Sigma−Bonded Aryliron Porphyrins 0.3. Synthesis and Characterization of High, Low, and Variable Spin State 5-Coordinate Sigma-Bonded Aryliron and Perfluoroaryliron(Iii) Complexes. Inorg. Chem. 1985, 24, 2509−2520. (37) Wade, R. S.; Castro, C. E. Oxidation of iron(II) porphyrins by alkyl halides. J. Am. Chem. Soc. 1973, 95, 226−230. (38) Brault, D.; Bizet, C.; Morliere, P.; Rougee, M.; Land, E. J.; Santus, R.; Swallow, A. J. One-electron reduction of ferrideuterioporphyrin IX and reaction of the oxidized and reduced forms with chlorinated methyl radicals. J. Am. Chem. Soc. 1980, 102, 1015−1020. (39) Brault, D.; Neta, P. Reactions of iron porphyrins with methyl radicals. J. Am. Chem. Soc. 1981, 103, 2705−2710. (40) Gütlich, P. Fifty Years of Mossbauer Spectroscopy in Solid State Research − Remarkable Achievements, Future Perspectives. Z. Anorg. Allg. Chem. 2012, 638, 15−43. (41) Bill, E. Mfit; Max-Planck Institute for Chemical Energy Conversion: Mülheim/Ruhr, Germany, 2008. (42) Kattner, H.; Buback, M. Detailed Investigations into Radical Polymerization Kinetics by Highly Time-Resolved SP−PLP−EPR. Macromol. Symp. 2013, 333, 11−23. (43) Tsarevsky, N. V.; Pintauer, T.; Matyjaszewski, K. Deactivation Efficiency and Degree of Control over Polymerization in ATRP in Protic Solvents. Macromolecules 2004, 37, 9768−9778. (44) Smolne, S.; Buback, M. Kinetic Investigations of Cu-Mediated ATRP in Aqueous Solution. Macromol. Chem. Phys. 2015, 216, 894− 902. (45) Averick, S.; Simakova, A.; Park, S.; Konkolewicz, D.; Magenau, A. J. D.; Mehl, R. A.; Matyjaszewski, K. ATRP under Biologically Relevant Conditions: Grafting from a Protein. ACS Macro Lett. 2012, 1, 6−10. (46) Konkolewicz, D.; Magenau, A. J. D.; Averick, S. E.; Simakova, A.; He, H. K.; Matyjaszewski, K. ICAR ATRP with ppm Cu Catalyst in Water. Macromolecules 2012, 45, 4461−4468. (47) Simakova, A.; Averick, S.; Konkolewicz, D.; Matyjaszewski, K. Aqueous ARGET ATRP. Macromolecules 2012, 45, 6371−6379. (48) The Handbook of Porphyrin Science; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; World Scientific Publishing Co. Pte. Ltd: 2012.

(49) Donohoe, R. J.; Atamian, M.; Bocian, D. F. Characterization of singly reduced iron(II) porphyrins. J. Am. Chem. Soc. 1987, 109, 5593−5599. (50) Chang, C. K.; Sotiriou, C. Migratory aptitudes in pinacol rearrangement of vic−dihydroxychlorins. J. Heterocycl. Chem. 1985, 22, 1739−1741. (51) Schroeder, H.; Buback, J.; Demeshko, S.; Matyjaszewski, K.; Meyer, F.; Buback, M. Speciation Analysis in Iron-Mediated ATRP Studied via FT−Near−IR and Mossbauer Spectroscopy. Macromolecules 2015, 48, 1981−1990. (52) Buback, M.; Schroeder, H.; Kattner, H. Detailed Kinetic and Mechanistic Insight into Radical Polymerization by Spectroscopic Techniques. Macromolecules 2016, 49, 3193−3213. (53) Moss, T. H.; Bearden, A. J.; Caughey, W. S. Mössbauer Studies of Bonding in Iron Porphyrin-Ligand Systems. J. Chem. Phys. 1969, 51, 2624−2631. (54) König, E.; Ritter, G.; Kanellakopulos, B. Low-Temperature Magnetism of Low-Spin Iron(II) Cation High-Spin Iron(III) Anion Complexes − Supporting Evidence for the True Nature of So-Called Spin-Triplet Iron(II) Compounds. Inorg. Chim. Acta 1982, 59, 285− 291. (55) Debrunner, P. G. Mössbauer spectroscopy of iron porphyrins. In Iron Porphyrin, Part III; Lever, A. B. P., Gray, H. B., Eds.; Physical Bioinorganic Chemistry; Wiley-VCH: New York, 1989; Vol. 4, pp 139−234. (56) Ali, E.; Sanyal, B.; Oppeneer, P. M. Electronic Structure, SpinStates, and Spin-Crossover Reaction of Heme-Related Fe-Porphyrins: A Theoretical Perspective. J. Phys. Chem. B 2012, 116, 5849−5859. (57) Scheidt, R. W.; Reed, C. A. Spin-state/stereochemical relationships in iron porphyrins: implications for the hemoproteins. Chem. Rev. 1981, 81, 543−555. (58) Xia, J.; Matyjaszewski, K. Controlled/″Living” Radical Polymerization. Homogeneous Reverse Atom Transfer Radical Polymerization Using AIBN as the Initiator. Macromolecules 1997, 30, 7692−7696. (59) Schroeder, H.; Buback, M. SP−PLP−EPR Measurement of Iron-Mediated ATRP Deactivation Rate. Macromolecules 2015, 48, 6108−6113. (60) Smolne, S.; Weber, S.; Buback, M. Propagation and Termination Kinetics of Poly ethylene glycol) Methyl Ether Methacrylate in Aqueous Solution. Macromol. Chem. Phys. 2016, DOI: 10.1002/ macp.201600302. (61) Soerensen, N.; Barth, J.; Buback, M.; Morick, J.; Schroeder, H.; Matyjaszewski, K. SP−PLP−EPR Measurement of ATRP Deactivation Rate. Macromolecules 2012, 45, 3797−3801. (62) Soerensen, N.; Schroeder, H.; Buback, M. SP−PLP−EPR Measurement of CuII-Mediated ATRP Deactivation and CuI-Mediated Organometallic Reactions in Butyl Acrylate Polymerization. Macromolecules 2016, 49, 4732−4738. (63) Schroeder, H.; Buback, J.; Demeshko, S.; Matyjaszewski, K.; Meyer, F.; Buback, M. Macromolecules 2015, 48, 1981−1990. (64) Miao, J.; He, W.; Zhang, L.; Wang, Y.; Cheng, Z.; Zhu, X. AGET ATRP of water-soluble PEGMA: Fast living radical polymerization mediated by iron catalyst. J. Polym. Sci., Part A: Polym. Chem. 2012, 50, 2194−2200. (65) He, W.; Zhang, L.; Miao, J.; Cheng, Z.; Zhu, X. Facile IronMediated AGET ATRP for Water-Soluble Poly(ethylene glycol) Monomethyl Ether Methacrylate in Water. Macromol. Rapid Commun. 2012, 33, 1067−1073. (66) Koerner, R.; Wright, J. L.; Ding, X. D.; Nesset, M. J. M.; Aubrecht, K.; Watson, R. A.; Barber, R. A.; Mink, L. M.; Tipton, A. R.; Norvell, C. J.; Skidmore, K.; Simonis, U. Electronic Effects in Transition Metal Porphyrins. 9. Effect of Phenyl Ortho Substituents on the Spectroscopic and Redox Properties and Axial Ligand Binding Constants of Iron(III) Tetraphenylporphyrinates. Inorg. Chem. 1998, 37, 733−745. (67) Kadish, K. M.; Bottomley, L. A. Substituent effects on the formation constants of iron(III) and iron(II) tetraphenylporphyrin− pyridine complexes. J. Am. Chem. Soc. 1977, 99, 2380−2382. I

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (68) Bottomley, L. A.; Kadish, K. M. Counterion and solvent effects on the electrode reactions of iron porphyrins. Inorg. Chem. 1981, 20, 1348−1357. (69) Tang, W.; Tsarevsky, N. V.; Matyjaszewski, K. Determination of equilibrium constants for atom transfer radical polymerization. J. Am. Chem. Soc. 2006, 128, 1598−1604. (70) Schroeder, H.; Yalalov, D.; Buback, M.; Matyjaszewski, K. Activation−Deactivation Equilibrium Associated With Iron-Mediated Atom−Transfer Radical Polymerization up to High Pressure. Macromol. Chem. Phys. 2012, 213, 2019−2026. (71) Habibullah, M.; Rahman, I. M. M.; Uddin, M. A.; Iwakabe, K.; Azam, A.; Hasegawa, H. Densities and Viscosities of the Binary Mixtures of Phenylmethanol with 2-Butanone. J. Chem. Eng. Data 2011, 56, 3323−3327. (72) Nanda, A. K.; Matyjaszewski, K. Effect of penultimate unit on the activation process in ATRP. Macromolecules 2003, 36, 8222−8224. (73) Lin, C. Y.; Coote, M. L.; Petit, A.; Richard, P.; Poli, R.; Matyjaszewski, K. Ab Initio Study of the Penultimate Effect for the ATRP Activation Step Using Propylene, Methyl Acrylate, and Methyl Methacrylate Monomers. Macromolecules 2007, 40, 5985−5994. (74) Tang, W.; Kwak, Y.; Braunecker, W.; Tsarevsky, N. V.; Coote, M. L.; Matyjaszewski, K. Understanding atom transfer radical polymerization: Effect of ligand and initiator structures on the equilibrium constants. J. Am. Chem. Soc. 2008, 130, 10702−10713.

J

DOI: 10.1021/acs.macromol.6b01774 Macromolecules XXXX, XXX, XXX−XXX