eATRP-Matyjaszewski Polymer Group - Carnegie Mellon University

Electrochemical control over an ATRP

When one considers the status of ATRP at the end of 2010, after exemplification of ARGET and ICAR ATRP (1)  and a true emulsion ATRP process, (2)  one could ask what targets remain in order to make ATRP more industrially compatible?  The answers include, as always: reduce costs, elimination/reduction of reagents, define conditions that allow simplified reaction conditions e.g. low levels of O2 permissible and, if possible provide additional degrees of polymerization control.  

So what was next?   

It is a new environmentally friendly procedure for controlling the ratio of activator to deactivator in an ATRP, specifically by electrochemical procedures.  Electrochemical methods offer multiple readily adjustable parameters, e.g. applied current, potential, and total passed charge, that can be applied to manipulate polymerization rates by selective targeting of specific relative concentrations of redox-active catalytic species which will provide industrially scalable procedures that allow for the creation of more complex specialized materials.  

Cyclic voltammetric (CV) studies of copper complexes suitable for catalyzing atom transfer radical polymerization have been used since ~2000 to measure the relative activity of copper based catalyst complexes in an ATRP. (3, 4)   It was found that the E1/2 value for the redox couple CuI/II/L strongly depends on the nature of the ligand and the selected transferable halogen, Figure 1.

eATRP F1

This would indicate that as the number of coordination sites on the ligand increases, i.e the electron donating ability of the ligand increases the reducing power of the corresponding CuI complex also increases. With the same ligand, CuCl complexes usually have a more negative potential, lower redox potential, than CuBr complexes.  The possibility of using CV electrochemistry as a screening method for selecting ATRP catalysts was subsequently discussed and implemented. (5, 6)  Electrochemistry has even been suggested as a means to remove the copper from the reaction medium by electrodeposition. (7)   However application of an electrochemical stimulus (i.e. electrolysis), which can be uniquely paired with redox based catalyst systems employed for an atom transfer radical polymerization (ATRP), had been generally overlooked.  

An initial report exploring the possibility of using an electrochemically-triggered initiation of controlled polymerization sequence during a CV cycle involving activation of alkyl- and benzyl halide initiators by an electrogenerated FeIISalen complex was reported in 2009. (8)  The reactions between the formed FeIISalen and several organic halides were investigated within the time scale of cyclic voltammetry measurements. A fast reaction was observed between electrogenerated FeIISalen catalyst complex and ATRP initiator molecule in the presence of styrene at 110 °C.  Polymerization was initiated and the polydispersity of the resulting low molecular weight polystyrene appeared to depend on the initial FeIIISalen/FeIISalen ratio.  However this work was apparently not continued or expanded.

The proposed mechanism of ATRP mediated through electrochemical (re)generation of activators is shown below in Figure 2.

eATRP F2

Figure 2. Schematic of proposed mechanism for electrochemical control over an ATRP

The schematic demonstrates the dynamic modulation of polymerization rates through electrochemical means, essentially a specific targeted fraction of the air stable CuIIBr2/Me6TREN catalyst complexes can be electrochemically reduced at the working electrode under cathodic currents to form the CuIBr/Me6TREN activator that can invoke or trigger a controlled polymerization in the presence of the remaining CuIIBr2/Me6TREN catalyst complexes.  In absence of mass transport limitations, the rate of reduction is dictated by the applied potential (Eapp) allowing one to finely tune the polymerization rate by balancing the ratio of [CuI]/[CuII] and thereby fine tuning of the polymerization rate. Further to this point, electrochemical methods allow the lower oxidation state catalyst, exemplified in Scheme 2 by CuIBr/Me6TREN, to be reverted back to its original higher oxidation state, simply by shifting Eapp to more positive values and in doing so providing a means to deactivate an ongoing polymerization and thereby control any temperature exotherms.  This dormant time in the polymerization may also be used for modification of the polymer or injection of additional reagents/monomers.  This procedure for an electrochemically mediated ATRP has been given the abbreviation eATRP.

The initial paper that actually studied procedures for electrochemical control over an ATRP (9) demonstrated that the dynamic modulation of polymerization rates through electrochemistry provided a means to allow "on demand" polymerization initiation, cessation of polymerization, and rejuvenation of a controlled/living radical polymerization process. Applying electricity to the system provided more precise control over the reaction. Electrochemical methods provide almost instantaneous control over an ATRP by offering additional adjustable control parameters, e.g. current, potential, and total passed charge, to manipulate polymerization activation/deactivation and polymerization rates by selective targeting of a desired ratio of redox-active catalytic species. A computer-controlled battery allows one to manipulate the ATRP process in real-time by changing the current or voltage.  Moreover, all the advantages of low catalyst concentration ATRP procedures, such as ARGET ATRP, are maintained by achieving ppm concentrations of catalyst (1, 10)  and tolerance to ambient O2, (11)  while simultaneously offering an environmentally friendly alternative by elimination of chemical reducing agents in addition to providing a convenient catalyst removal/recycle process through electrodeposition. (7)

A schematic and an image of the equipment used for this initial eATRP study are shown in Figure 3.

eATRP F3

Figure 3. Five necked flask with three electrodes used for e-ATRP: the Pt working electrode, the counter electrode, and the reference electrode which were a 3 mm Pt disk (Gamry Instruments), a Pt mesh, and an Ag|Ag+, which contained a 0.1 M AgNO3 and TBAPF6 filling solution in MeCN separated from the working solutions by a porous Vycor tip, respectively.

In an eATRP, a desired amount of the catalyst complex (X-CuII/L) can be electrochemically reduced to activators (CuI/L) to start a controlled radical polymerization. (9)  Typically, the reaction mixture initially contained solvents, monomers, initiators, supporting electrolyte and X-CuII/L. Due to absent of CuI/L activators, polymerization has not occurred. The polymerization begins only when a sufficient cathodic electrical current was applied to the working electrode (WE) for reduction of a fraction of the added X-CuII/L to CuI/L at the WE surface. The reduced activators then spreads out into the working solution by vigorous stirring, and reacts with initiators (e.g., alkyl halide, Pn-X) to form radicals (Pn) and is oxidized back to deactivators (X-CuII/L). The radical species propagated to form polymeric chains by reacting with monomers (M), or were deactivated back to the dormant species (Pn-X), Figure 2. Continuous (re)generation of activators (CuI/L) and modulation of the rate of polymerization (Rp) can be achieved by an appropriately selected Eapp.

As noted above, applying the appropriate charge is a requirement for a well-controlled electrochemical ATRP reaction. In the initial example reduction of the CuIIBr2/Me6TREN complex was accomplished through application of a cathodic current.  Prior to the electrolysis experiments, cyclic voltammograms (CVs) of methyl acrylate (MA) and acetonitrile (MeCN) over a potential range of 1.5 volts, from 0.0 to -1.5 V versus Ag|Ag+ reference electrode were acquired to ensure the absence of any redox process that might interfere with the catalyst reduction.  After confirming the electrochemical stability of the reaction medium an additional CV in the presence of added CuIIBr2/Me6TREN was acquired to identify the potential window appropriate for accurate manipulation of the oxidation states of the redox-active catalyst in electrolysis experiments.  A reversible peak couple attributed to the single electron transfer reduction of the X-CuII/L complex was observed, as shown in Figure 4.  The value of the half wave potential of the copper catalyst (CuI/II/L) was calculated to be E1/2 = -0.69 V versus Ag/AgI/I-.

Figure 4: Cyclic voltammetry of 1.0 mM CuIIBr2/Me6TREN recorded at v = 0.5 V/s-1 in 50 % (v/v) MA in MeCN with 0.1 M supporting electrolyte.

The application of a -0.66 V potential to the reaction medium resulted in the rapid consumption of monomer, reaching nearly 80 % conversion within four hours, Figure 5A blue points.  Linear first order kinetic behavior was observed during the polymerization, Figure 5B, indicating a constant concentration of propagating species, and an excellent correlation persisted between theoretical and experimental molecular weight values as the polymerization progressed.  Molecular weight values increased linearly with conversion and molecular weight distributions decreased gradually with increasing monomer conversion reaching a minimum value of approximately Mw/Mn = 1.06. Since the rate of an ATRP is defined by ratio of [CuIBr/Me6TREN] and [CuIIBr2/Me6TREN], and therefore depends on the extent to which CuIIBr2/Me6TREN is reduced and consequently the Eapp.  
The choice of the Eapp is of primary importance in a polymerization conducted under potentiostatic conditions. Specifically, two conditions can be used to control the reaction; the first is electrolysis under diffusive control, where Eapp is more negative than Epc and the overall rate of the reaction is controlled by the rate of the diffusion of the reactants to the electrode surface rather than the rate of the reaction itself. The second is electrolysis controlled by the rate of electron transfer, in this case Eapp is more positive than Epc and the overall rate of the reaction is controlled by the rate of the reaction itself rather than the rate of the mass transport of the reactants to the electrode surface. In addition to these considerations, the selected Eapp should avoid other undesired electron transfer (ET) reactions. In ATRP systems, one of these undesired ET reactions is the further reduction of CuI/L to Cu0. This is expected to occur at potentials more negative than that of the CuI/II/L redox coupling potential when disproportionation is not thermodynamically favoured. (12)  It should be selected only when one desires to remove the copper from the reaction medium.
Two additional and separate polymerizations were conducted with a 30 and 60 mV increase of the Eapp to -0.69 and -0.72 V, respectively.  An enhanced polymerization rate was observed at more negative potentials whereas more positive potentials provided a slower rate of polymerization, Figure 5A.

eATRP F5

Figure 5:(A) Monomer conversion with respect to time, and (B) number average molecular weight (Mn) and Mw/Mn with respect to conversion as a function of applied potential. Polymerizations conducted in 50% (v/v) MA in MeCN at 25° C with a total reaction volume of approximately 12 mL. [MA]0 = 5.55 M; [MA]0:[EBP]0:[Me6TREN]0:[CuIIBr2]0 = 500:1:0.025:0.025.

A high cathodic decay has been observed in the early stages of a polymerization conducted under potentiostatic conditions (constant potential), representing the reaction of the reduced CuI/L complex with an initiator followed by an EC' (electrolysis-catalysis) mechanism. More negative η produces larger initial currents, indicating a faster rate of reduction. Subsequently, a steady-state condition, equilibria of [Pn-X], [Pn], [CuI/L], and [X-CuII/L] is established, hence constant current is observed. (13)  This paper also examined the effect of stirring and the effect of distance from the reducing electrode as the polymerization kinetics depend on the ratio of [CuII]/[CuI]. The ratio can be adjusted by varying the applied potentials with fast stirring or the distance (d) between the working electrode and the initiator modified surface (without stirring). In the first case, the reduction of the catalysts only occurs at the WE surfaces and fast stirring efficiently spreads out the reduced activators and prevents passivation of the growing polymer chains at the WE surfaces if stirring is diminished or stopped. The influence of the distance will be addressed below when discussing the preparation of polymeric surfaces by surface-initiated eATRP.  
As with all other ATRP procedures, catalyst complexes with higher reducing power provide faster polymerizations, due to larger value for KATRP.  eATRP is also influences by the supporting electrolyte and a decrease in rate was observed when the concentration of electrolyte was decreased.  This decrease can be explained as a consequence of an increase in the resistance of the contacting solution.  When all processes involved in the reaction are fast then electron transfer is the limiting factor in the reaction. (12)
Typically eATRP reactions were carried out in DMF and MeCN as solvents with polymerizations fasted in DMF due to the influence of polarity on KATRP.  For instance KATRP is ~103 times higher in aqueous media than in MeCN.  As noted elsewhere on the site eATRP overcomes many of the drawbacks typically associated with conducting an ATRP protic media (14) and recently water soluble methacrylates and acrylamides have been successfully polymerized by eATRP providing polymers with narrow molecular weight distribution and targeted degrees of polymerization.  (14-18)   The X-CuII/L bond can easily dissociate in an aqueous ATRP, therefore either high concentrations of catalysts or addition of salts (halide anions, X-) were required.  (14, 19)   The addition of excess X- promotes the (re)formation of X-CuII/L and a noticeable improvement of MW distribution indeed targeting conditions that result in formation of a constant and high [CuII]/[CuI] ratio is optimal.  Reaction temperature also exerts an influence in an aqueous ATRP as concurrent nucleophilic substitution (solvolysis) of alkyl halides in water can be suppressed at lower temperatures. Thus, aqueous systems that are conducted at relatively low temperature and high concentration of Cu catalysts seem to be most effective.  (15, 20, 21)

Potentiostatic versus galvanostatic conditions

Electrolysis under galvanostatic conditions is attractive from both an industrial and experimental standpoint because it provides a constant current instead of a constant potential and eliminates the need for the reference electrode (RE).  Galvanostatic conditions have been successfully utilized for the eATRP synthesis of homopolymers and copolymers. (13)  

eATRP reactions were conducted under galvanostatic or potentiostatic conditions and Figure 6 shows a comparison of the applied conditions and results. In this case, the selected galvanostatic current values resembled those of the potentiostatic polymerization (Q = A × s), whereby an initial high current value, red section Figure 6(a), and consecutive low current value, blue section Figure 6(a), were employed. This two-stage current program was chosen to rapidly convert the vast majority of X-CuII/L to CuI/L, initiating eATRP, and next lower current step providing compensation for any (re)generated X-CuII/L incurred from termination.

Figure 6.  eATRP of n-butyl acrylate (BA): (a) Current profiles of (black) a potentiostatic (black) and (red/blue) galvanostatic polymerization. (b) First-order kinetic plot of a potentiostatic and galvanostatic eATRP.

As the polymerization took place the potential remained in close proximity to the potential of Cu/L couple, however, near the end of each current stage it began to shift to more negative potentials. This indicates that in the vicinity of the electrode, nearly all X-CuII/L was converted to CuI/L, and therefore the potentiostat began to apply a stronger reducing potential to maintain these current values. The polymerization under galvanostatic conditions was slower than the polymerization under the potentiostatic setup, Figure 6(b), because of a lower applied current, i.e. a smaller amount of charge (Q) passed at the same time period. With regards to polymerization control, the polymerization displayed a linear increase in MW with conversion and low Mw/Mn values during the course of the polymerization

Simplified eATRP (seATRP)

The complexity of an electrochemical reaction setup can be overcome by using sacrificial anode (counter electrode). The simplified eATRP (seATRP) does not require any separation of CE from the reaction medium and can be directly inserted into the reaction mixture, Figure 7.


            
    Figure 7. Comparison of setup required for an eATRP and a seATRP

There are two potential advantages of using seATRP. The first one is the use of an undivided cell, which allows for a simpler, lower cost process, and most of all, the minimization of the ohmic drop, bringing a beneficial saving in energy costs. The second advantage is the possibility of a galvanostatic process using two electrode system and a power supply, which is simpler than a potentiostatic process.  A series of seATRP polymerizations were carried out under different Eapp, with a Pt mesh WE, an aluminum (Al) wire CE, and Ag/AgI/I- RE. (22)  In the ideal case of seATRP, the sacrificial counter electrode as well as the products of its oxidation should not react with the CuI/L catalysts. In this case, aluminum (Al) does not reduce X-CuII/L to CuI/L even though it usually shows a negative standard potential in different media. This is likely because the surface of the Al is passivated by forming stable oxidized layers to prevent reduction of X-CuII/L to CuI/L. Therefore the Al seemed to fit the requirements of a sacrificial electrode. The polymerization results showed similar MW evolution, maintaining a narrow MWD throughout the reactions which are similar to results observed for n-butyl acrylate (BA) polymerization under conventional eATRP. In addition, galvanostatic conditions with a two electrode system were developed for synthesizing PBA. The direct immersion of an Al wire anode in the reaction medium can avoid additional preparation setup steps, and the polymerization results indicated good control of reaction kinetics, providing polymers with molecular weight evolution close to theoretical values and generating polymers with narrow molecular-weight distribution. The rate of the polymerizations Rp was controlled by applying different potentials Eapp, with faster Rp observed using more negative Eapp.  Synthesis of high MW polymers and chain extension reactions indicated good conservation of chain-end functionalities. The seATRP procedure can be further simplified by using only two electrodes and applying a constant current under galvanostatic conditions. The use of a multi-step current procedure showed identical results to polymerizations carried out under potentiostatic conditions, namely linear first-order kinetics and a uniform growth of polymers.

RFERERENCES

(1)    Matyjaszewski, K.;  Jakubowski, W.;  Min, K.;  Tang, W.;  Huang, J.;  Braunecker, W. A.; Tsarevsky, N. V. Proceedings of the National Academy of Sciences 2006, 103, 15309-15314.
(2)    Min, K.;  Gao, H.; Matyjaszewski, K. Journal of the American Chemical Society 2006, 128, 10521-10526.
(3)    Qiu, J.;  Matyjaszewski, K.;  Thouin, L.; Amatore, C. Macromolecular Chemistry and Physics 2000, 201, 1625-1631.
(4)    Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921-2990.
(5)    Tang, W.; Matyjaszewski, K. Macromolecules 2006, 39, 4953-4959.
(6)    Braunecker, W. A.;  Brown, W. C.;  Morelli, B. C.;  Tang, W.;  Poli, R.; Matyjaszewski, K. Macromolecules 2007, 40, 8576-8585.
(7)    Nasser-Eddine, M.;  Delaite, C.;  Dumas, P.;  Vataj, R.; Louati, A. Macromolecular Materials and Engineering 2004, 289, 204-207.
(8)    Bonometti, V.;  Labbe, E.;  Buriez, O.;  Mussini, P.; Amatore, C. J. Electroanal. Chem. 2009, 633, 99-105.
(9)    Magenau, A. J. D.;  Strandwitz, N. C.;  Gennaro, A.; Matyjaszewski, K. Science 2011, 332, 81-84.
(10)    Matyjaszewski, K.;  Dong, H.;  Jakubowski, W.;  Pietrasik, J.; Kusumo, A. Langmuir 2007, 23, 4528-4531.
(11)    Pintauer, T.; Matyjaszewski, K. Chemical Society Reviews 2008, 37, 1087-1097.
(12)    Matyjaszewski, K.;  Bortolamei, N.;  Magenau, A.;  Gennaro, A.; Isse, A. A. In WIPO; (Carnegie Mellon University, USA). Application: WO 2012091965, 2012; p 69pp.
(13)    Magenau, A. J. D.;  Bortolamei, N.;  Frick, E.;  Park, S.;  Gennaro, A.; Matyjaszewski, K. Macromolecules (Washington, DC, United States) 2013, 46, 4346-4353.
(14)    Bortolamei, N.;  Isse, A. A.;  Magenau, A. J. D.;  Gennaro, A.; Matyjaszewski, K. Angewandte Chemie, International Edition 2011, 50, 11391-11394, S11391/11391-S11391/11315.
(15)    Chmielarz, P.;  Park, S.;  Simakova, A.; Matyjaszewski, K. Polymer 2015, 60, 302-307.
(16)    Hosseiny, S. S.; van Rijn, P. Polymers (Basel, Switzerland) 2013, 5, 1229-1240.
(17)    Chmielarz, P.;  Krys, P.;  Park, S.; Matyjaszewski, K. Polymer 2015, 71, 143-147.
(18)    Shida, N.;  Koizumi, Y.;  Nishiyama, H.;  Tomita, I.; Inagi, S. Angewandte Chemie, International Edition 2015, 54, 3922-3926.
(19)    Tsarevsky, N. V.;  Pintauer, T.; Matyjaszewski, K. Macromolecules 2004, 37, 9768-9778.
(20)    Millard, P.-E.;  Mougin, N. C.;  Boeker, A.; Mueller, A. H. E. ACS Symposium Series 2009, 1023, 127-137.
(21)    Zhang, Q.;  Wilson, P.;  Li, Z.;  McHale, R.;  Godfrey, J.;  Anastasaki, A.;  Waldron, C.; Haddleton David, M. J Am Chem Soc 2013, 135, 7355-7363.
(22)    Park, S.;  Chmielarz, P.;  Gennaro, A.; Matyjaszewski, K. Angewandte Chemie, International Edition 2015, 54, 2388-2392.