α-Amido Trifluoromethyl Xanthates: A New Class of RAFT/MADIX Agents

Xanthates have long been described as poor RAFT/MADIX agents for styrene polymerization. Through the determination of chain transfer constants to xanthates, this work demonstrated beneficial capto-dative substituent effects for the leaving group of a new series of α-amido trifluoromethyl xanthates, with the best effect observed with trifluoroacetyl group. The previously observed Z-group activation with a O-trifluoroethyl group compared to the O-ethyl counterpart was quantitatively established with Cex = 2.7 (3–4 fold increase) using the SEC peak resolution method. This study further confirmed the advantageous incorporation of trifluoromethyl substituents to activate xanthates in radical chain transfer processes and contributed to identify the most reactive xanthate reported to date for RAFT/MADIX polymerization of styrene.


Introduction
Reversible addition-fragmentation chain transfer polymerization (RAFT), coined MADIX for Macromolecular Design via the Interchange of Xanthates (MADIX) for the specific use of xanthate chain transfer agents, recently celebrated its 25th anniversary [1][2][3][4].Since its discovery, a significant amount of work has been performed on the design of new thiocarbonylthio chain transfer agents of the general structure R-S(C=S)Z [5].Among them, dithiocarbonates (xanthates, Z=OZ') have received increasing attention over the years and are today firmly established as best suited for the RAFT/MADIX polymerization of non-conjugated monomers [6][7][8].Typical methods for xanthate synthesis are the reaction of a dithiocarbonate salt with an alkylating agent [9][10][11], thioacylation reactions [12,13], the ketoform reaction [14] and the radical substitution of a xanthogen disulfide [15].The effectiveness of RAFT/MADIX agents depends on the nature of both activating and leaving groups, Z and R, respectively.For instance, we reported that O-ethyl xanthates (Z=OEt) with secondary leaving groups such as R = 1-phenylethyl and 1-(methoxycarbonyl)ethyl exhibit a moderate reactivity (C tr = k tr /k p ~0.5-1) in RAFT/MADIX polymerization of styrene (St) [16][17][18].Later studies showed that the level of control over number-average molar masses (M n ) and dispersities (Ð) could be markedly improved by selecting the electron-withdrawing CH 2 CF 3 group as Z' [10,17,18].It appeared to us that there was still a great space for diversification and improvement in the selection of the leaving R group.In 2008, Zard et al. [19] described the use of α-acetamido trifluoromethyl xanthates as a powerful reagent for the introduction of geminal amido and trifluoromethyl groups through radical addition to olefins.Drawing on this work, we here report on the synthesis of a series of xanthates with α-amido trifluoromethyl leaving groups (Scheme 1) and the evaluation of their reactivity in RAFT/MADIX bulk polymerization of styrene.The choice of St was motivated by the moderate reactivity of xanthates to this monomer, thus allowing us to simply establish a structure-reactivity relationship via the determination of distinctive kinetic data.
Molecules 2024, 29, x FOR PEER REVIEW 2 of 11 the synthesis of a series of xanthates with α-amido trifluoromethyl leaving groups (Scheme 1) and the evaluation of their reactivity in RAFT/MADIX bulk polymerization of styrene.The choice of St was motivated by the moderate reactivity of xanthates to this monomer, thus allowing us to simply establish a structure-reactivity relationship via the determination of distinctive kinetic data.

Xanthate Synthesis
The synthesis of the series of α-amido trifluoromethyl O-ethyl xanthates 9-12 was performed as follows (Scheme 2): the condensation between fluoral (trifluoroacetaldehyde methyl hemiacetal) and different amides afforded intermediate N-acyl hemiaminals, which in turn were converted to the corresponding chlorides, and these were eventually transformed to the desired xanthates via nucleophilic substitution with potassium ethylxanthogenate.Four groups of compounds were synthesized, with methyl, phenyl, t-butyl and tbutoxy R substituents corresponding to products 1 to 12 of Table 1.t-BuO was considered as a R group to cope with the failure of the direct condensation between trifluoroacetaldehyde hemiacetal and trifluoroacetamide.Thus, the series of products 4, 8 and 12 was synthesized and the t-Boc group of xanthate 12 was eventually removed to introduce the trifluoroacetyl group and obtain xanthate 13 (Scheme 3).Condensation between the methyl hemiacetal of fluoral and different amides afforded the desired alcohols in moderate yields.In the case of alcohol 4, a longer reaction time was needed, generally 4 h as opposed to 1 h for alcohols 1 to 3. The obtained alcohols were then transformed into the corresponding chlorides via reaction with SOCl2 in heptane.The reaction was fast and clean, the desired chlorides were obtained in good yields and generally no purification was furthermore needed.Finally, the conversion of these to the desired xanthates was carried out by a reaction with the potassium salt of ethylxanthic acid in ethanol, and the xanthates were obtained in good yields.As mentioned before, for α-amido trifluoromethyl xanthates of the study.

Xanthate Synthesis
The synthesis of the series of α-amido trifluoromethyl O-ethyl xanthates 9-12 was performed as follows (Scheme 2): the condensation between fluoral (trifluoroacetaldehyde methyl hemiacetal) and different amides afforded intermediate N-acyl hemiaminals, which in turn were converted to the corresponding chlorides, and these were eventually transformed to the desired xanthates via nucleophilic substitution with potassium ethylxanthogenate.
Molecules 2024, 29, x FOR PEER REVIEW 2 of 11 the synthesis of a series of xanthates with α-amido trifluoromethyl leaving groups (Scheme 1) and the evaluation of their reactivity in RAFT/MADIX bulk polymerization of styrene.The choice of St was motivated by the moderate reactivity of xanthates to this monomer, thus allowing us to simply establish a structure-reactivity relationship via the determination of distinctive kinetic data.

Xanthate Synthesis
The synthesis of the series of α-amido trifluoromethyl O-ethyl xanthates 9-12 was performed as follows (Scheme 2): the condensation between fluoral (trifluoroacetaldehyde methyl hemiacetal) and different amides afforded intermediate N-acyl hemiaminals, which in turn were converted to the corresponding chlorides, and these were eventually transformed to the desired xanthates via nucleophilic substitution with potassium ethylxanthogenate.Four groups of compounds were synthesized, with methyl, phenyl, t-butyl and tbutoxy R substituents corresponding to products 1 to 12 of Table 1.t-BuO was considered as a R group to cope with the failure of the direct condensation between trifluoroacetaldehyde hemiacetal and trifluoroacetamide.Thus, the series of products 4, 8 and 12 was synthesized and the t-Boc group of xanthate 12 was eventually removed to introduce the trifluoroacetyl group and obtain xanthate 13 (Scheme 3).Condensation between the methyl hemiacetal of fluoral and different amides afforded the desired alcohols in moderate yields.In the case of alcohol 4, a longer reaction time was needed, generally 4 h as opposed to 1 h for alcohols 1 to 3. The obtained alcohols were then transformed into the corresponding chlorides via reaction with SOCl2 in heptane.The reaction was fast and clean, the desired chlorides were obtained in good yields and generally no purification was furthermore needed.Finally, the conversion of these to the desired xanthates was carried out by a reaction with the potassium salt of ethylxanthic acid in ethanol, and the xanthates were obtained in good yields.As mentioned before, for Synthetic strategy for α-amido trifluoromethyl xanthates.Xa = S(C=S)OEt.
Four groups of compounds were synthesized, with methyl, phenyl, t-butyl and t-butoxy R substituents corresponding to products 1 to 12 of Table 1.t-BuO was considered as a R group to cope with the failure of the direct condensation between trifluoroacetaldehyde hemiacetal and trifluoroacetamide.Thus, the series of products 4, 8 and 12 was synthesized and the t-Boc group of xanthate 12 was eventually removed to introduce the trifluoroacetyl group and obtain xanthate 13 (Scheme 3).Also, we synthesized xanthate 14 (Scheme 1) with the Z=OEt group being replaced with a 2,2,2-trifluoroethyl group, with R=Me.The commercially available 2,2,2-trifluoroethanol was treated with sodium hydride in the presence of an excess of carbon disulfide in DMF at −40 °C, followed by the addition of the corresponding alkyl halide.The temperature was an important factor because the reaction of sodium trifluoroethoxide with carbon disulfide leading to the desired xanthate salt is quite reversible, due to the high inductive effect of the trifluoromethyl group.Carrying the reaction at higher temperatures favors the reverse process, and under these conditions, the only isolated product after addition of the electrophile is the ether resulting from the substitution of the chloride by Condensation between the methyl hemiacetal of fluoral and different amides afforded the desired alcohols in moderate yields.In the case of alcohol 4, a longer reaction time was needed, generally 4 h as opposed to 1 h for alcohols 1 to 3. The obtained alcohols were then transformed into the corresponding chlorides via reaction with SOCl 2 in heptane.
The reaction was fast and clean, the desired chlorides were obtained in good yields and generally no purification was furthermore needed.Finally, the conversion of these to the desired xanthates was carried out by a reaction with the potassium salt of ethylxanthic acid in ethanol, and the xanthates were obtained in good yields.As mentioned before, for the synthesis of trifluoroacetamido xanthate 13, treatment of xanthate 12 with trifluoroacetic acid and trifluoroacetic anhydride in dichloromethane afforded the desired compound in 61% yield (Scheme 3).Also, we synthesized xanthate 14 (Scheme 1) with the Z=OEt group being replaced with a 2,2,2-trifluoroethyl group, with R=Me.The commercially available 2,2,2-trifluoroethanol was treated with sodium hydride in the presence of an excess of carbon disulfide in DMF at −40 • C, followed by the addition of the corresponding alkyl halide.The temperature was an important factor because the reaction of sodium trifluoroethoxide with carbon disulfide leading to the desired xanthate salt is quite reversible, due to the high inductive effect of the trifluoromethyl group.Carrying the reaction at higher temperatures favors the reverse process, and under these conditions, the only isolated product after addition of the electrophile is the ether resulting from the substitution of the chloride by the trifluoroethoxide anion.In contrast, the formation of the xanthate proceeded in good yield at the reported low temperature.

Structure-Reactivity Relationship of α-Amido Trifluoromethyl Xanthates
Self-initiated bulk polymerization of St was carried out at 110 • C in sealed tubes under vacuum.The transfer constants to xanthates C tr were determined from both the Mayo method [20] and Equation ( 1) derived from the kinetic model established by Müller et al. [21] for living polymerization systems exhibiting slow degenerative transfer: where DP n,th is the theoretical number-average degree of polymerization (equal to ([M] 0 /[X] 0 ).conv.for an instantaneous xanthate consumption, where conv.represents the fractional monomer conversion) and DP n,calc is the calculated DP n when X was not fully reacted (DP n,calc = conv. [M] 0 /∆[X]).DP n,calc can be assimilated to DP n,SEC provided that the contribution of initiator-derived chains is negligible.With this method, C tr may be slightly overestimated when the contribution of initiation (therefore termination) is such that 2f∆[I 2 ] << ∆[X] cannot be applied.As for the Mayo method, based on the Expression (2) giving the evolution of [X]/[M] during polymerization, C tr values are only valid at low monomer conversions and low C tr and tend to be underestimated at greater conversions and C tr .
Thus, it can be propounded that the real C tr value lies between those determined by the two chosen methods.
The C tr values are listed in Table 2.The experimental data used for plotting Mayo regressions for xanthates 9, 10, 11 and 13 are gathered in Table S1.The corresponding Mayo plots can be found in Figures S1-S4.The data used for the calculation of C tr for 13 DP n,calc and 14 from Equation ( 1) are available in Table S2.C tr values are moderate (comprised between 2 and 3) and nearly equal for xanthates 9, 10 and 11, reflecting a similar stabilizing effect of methyl, phenyl and t-butyl R substituents on the leaving group radical.The Mayo method and the one derived from Müller's kinetic model gave similar results.It is worth mentioning that these values are ~3-times greater than those previously reported for O-ethyl xanthates with classically used 1-phenylethyl and 1-(methoxycarbonyl)ethyl leaving groups [16,17].However, at low conversions (~5-10%), M n values are much higher than those expected for a controlled RAFT/MADIX process, and dispersities are in the range 1.85-2.35(Table S1).This observation is typical of RAFT/MADIX systems, for which both C tr (for starting RAFT/MADIX agent) and C ex (for macro-RAFT/MADIX agent PSt-X, C ex = k ex /k p , with k ex the exchange rate constant) are low.This is supported by results of Catala et al., who reported a C ex value of 0.7-0.8 for the St/PSt-S(C=S)OEt system [22].In contrast, the trifluoromethyl R group in 13 further increases the electrophilic character of the leaving group radical and facilitates the fragmentation step, resulting in a greater C tr equal to 6.1.It is worth mentioning that the Mayo regression for 13 was plotted with St conversions of about 1% in order to minimize the error in the determination of C tr .Using DP n , calc values extracted from Figure 1 or Table S2, the same C tr value of 6.1 was obtained from Equation (1).On Figure 1, it can be seen that as a consequence of the high C tr value, the evolution of M n with conversion approaches linearity.Dispersities are high and increase from 1.69 to 2.39 during polymerization.This can be explained by a slow equilibrium between dormant PSt-13 chains and macroradicals (C ex ~0.7-0.8 for O-ethyl xanthates) [22], responsible for the presence of multimodality in molar mass distributions (Figure 1b).reported for O-ethyl xanthates with classically used 1-phenylethyl and 1-(methoxycarbonyl)ethyl leaving groups [16,17].However, at low conversions (~5-10%), Mn values are much higher than those expected for a controlled RAFT/MADIX process, and dispersities are in the range 1.85-2.35(Table S1).This observation is typical of RAFT/MADIX systems, for which both Ctr (for starting RAFT/MADIX agent) and Cex (for macro-RAFT/MADIX agent PSt-X, Cex = kex/kp, with kex the exchange rate constant) are low.This is supported by results of Catala et al., who reported a Cex value of 0.7-0.8 for the St/PSt-S(C=S)OEt system [22].In contrast, the trifluoromethyl R group in 13 further increases the electrophilic character of the leaving group radical and facilitates the fragmentation step, resulting in a greater Ctr equal to 6.1.It is worth mentioning that the Mayo regression for 13 was plotted with St conversions of about 1% in order to minimize the error in the determination of Ctr.Using DPn,calc values extracted from Figure 1 or Table S2, the same Ctr value of 6.1 was obtained from Equation (1).On Figure 1, it can be seen that as a consequence of the high Ctr value, the evolution of Mn with conversion approaches linearity.Dispersities are high and increase from 1.69 to 2.39 during polymerization.This can be explained by a slow equilibrium between dormant PSt-13 chains and macroradicals (Cex~0.7-0.8 for O-ethyl xanthates) [22], responsible for the presence of multimodality in molar mass distributions (Figure 1b).A similar behavior was reported by our group for S-cyanoisopropyl O-ethyl xanthate, for which a Ctr of 6.8 close to that of 13 had been determined [16].Due to a capto-dative stabilizing effect for the leaving group radical, this series of α-amido trifluoromethyl groups represents the best secondary leaving groups ever reported for O-ethyl RAFT/MA-DIX agents.The carbon-sulfur bond is weakened by an anomeric type effect due to the lone pair on the nitrogen (a lone pair-σ* interaction), which may explain the efficiency of these xanthates.In agreement with some of our earlier studies [10,17,18], the reactivity of A similar behavior was reported by our group for S-cyanoisopropyl O-ethyl xanthate, for which a C tr of 6.8 close to that of 13 had been determined [16].Due to a capto-dative stabilizing effect for the leaving group radical, this series of α-amido trifluoromethyl groups represents the best secondary leaving groups ever reported for O-ethyl RAFT/MADIX agents.The carbon-sulfur bond is weakened by an anomeric type effect due to the lone pair on the nitrogen (a lone pair-σ* interaction), which may explain the efficiency of these xanthates.In agreement with some of our earlier studies [10,17,18], the reactivity of the xanthate was greatly increased by substituting the O-ethyl Z group of 13 for a O-CH 2 CF 3 in 14.A C tr of 15 was determined from Equation (1) (Table S2).This was confirmed with the observed good control of M n with conversion (Figure 2).At 85% St conversion, Ð equals 1.49 (Figure 2), which reflects a faster interchain transfer compared to O-ethyl analogs.This increase in reactivity owing to the trifluoroethyl Z group is quantified in the next section through the determination of C ex for xanthate 14.
the xanthate was greatly increased by substituting the O-ethyl Z group of 13 for a O-CH2CF3 in 14.A Ctr of 15 was determined from Equation (1) (Table S2).This was confirmed with the observed good control of Mn with conversion (Figure 2).At 85% St conversion, Đ equals 1.49 (Figure 2), which reflects a faster interchain transfer compared to O-ethyl analogs.This increase in reactivity owing to the trifluoroethyl Z group is quantified in the next section through the determination of Cex for xanthate 14.

Determination of Cex for Z=OCH2CF3
We applied the SEC-based peak resolution method developed by Fukuda et al. [23] to determine the activation rate constant kact for PSt-S(C=S)OCH2CF3 (PSt-14) in St polymerization.A low molar mass PSt-14 was synthesized (Mn = 1850 g/mol, Đ = 1.3).The fraction of dead chains Fdead was estimated to 6.4% from a chain extension test and treatment of the resulting SEC trace (see Appendix A for details).Fdead was taken into account in the data treatment for the determination of kact, as dead chains do not contribute to PSt-14 activation.A series of polymerizations were carried out at 80 °C with [PSt-14]0 = 1.8.10 −2 mol/l and varying values of [BPO]0 (Figure 3a,b).Figure 3a gives the first-order plots of the monomer concentration [M], hence [P•] for each [BPO]0.Here St conversion was directly determined from SEC traces, which directly give the concentration of formed PSt.The resolution of the bimodal SEC curves of each raw polymerization product of the series allowed for the determination of the activation rate coefficient kact from the slope of ln(S/S0) vs. time (Figures 3b and S5) [23].
The plot of kact = kd + kex[P•] against [P•] makes a straight line with its intercept at [P•] = 0 and slope giving kd and kex, respectively.The data points from Figure 3a,b give a straight line with kd~0 and Cex= 2.67 (Figure 3c).These results are in agreement with the experimental Đ values of the present and past studies [10,17] (model predictions give Đ ~1 + 1/Cex at the end of the polymerization) [21].

Determination of C ex for Z=OCH 2 CF 3
We applied the SEC-based peak resolution method developed by Fukuda et al. [23] to determine the activation rate constant k act for PSt-S(C=S)OCH 2 CF 3 (PSt-14) in St polymerization.A low molar mass PSt-14 was synthesized (M n = 1850 g/mol, Ð = 1.3).The fraction of dead chains F dead was estimated to 6.4% from a chain extension test and treatment of the resulting SEC trace (see Appendix A for details).F dead was taken into account in the data treatment for the determination of k act , as dead chains do not contribute to PSt-14 activation.A series of polymerizations were carried out at 80 • C with [PSt-14] 0 = 1.8.10 −2 mol/l and varying values of [BPO] 0 (Figure 3a,b).Figure 3a gives the first-order plots of the monomer concentration [M], hence [P•] for each [BPO] 0 .Here St conversion was directly determined from SEC traces, which directly give the concentration of formed PSt.The resolution of the bimodal SEC curves of each raw polymerization product of the series allowed for the determination of the activation rate coefficient k act from the slope of ln(S/S 0 ) vs. time (Figures 3b and S5) [23].
ln(S 0 /S) = k act t The plot of k act = k d + k ex [P•] against [P•] makes a straight line with its intercept at [P•] = 0 and slope giving k d and k ex , respectively.The data points from Figure 3a,b give a straight line with k d ~0 and C ex = 2.67 (Figure 3c).These results are in agreement with the experimental Ð values of the present and past studies [10,17] (model predictions give Ð ~1 + 1/C ex at the end of the polymerization) [21].

Materials
St was purchased from Aldrich.It was distilled under vacuum and stored over CaH2 prior to any use.All the reactants that took part in the synthesis of the xanthates were purchased from Aldrich and used as received.

Synthesis of the α-Amido Trifluoromethyl O-ethyl Xanthates
General method for the synthesis of trifluoromethyl alcohols.To a solution of fluoral (1 equiv.) in dioxane (1.5 mL/mmol), the corresponding amide (1 equiv.) was added and the mixture was heated at reflux temperature for one hour.After that time, the reaction mixture was concentrated, and the residue was purified by the corresponding method.
General method for the synthesis of trifluoromethyl chlorides.To a suspension of alcohol (1 equiv.) in heptane (2 mL/mmol), SOCl2 (1.1 equiv.) was added and the resulting

Materials
St was purchased from Aldrich.It was distilled under vacuum and stored over CaH 2 prior to any use.All the reactants that took part in the synthesis of the xanthates were purchased from Aldrich and used as received.

Synthesis of the α-Amido Trifluoromethyl O-ethyl Xanthates
General method for the synthesis of trifluoromethyl alcohols.To a solution of fluoral (1 equiv.) in dioxane (1.5 mL/mmol), the corresponding amide (1 equiv.) was added and the mixture was heated at reflux temperature for one hour.After that time, the reaction mixture was concentrated, and the residue was purified by the corresponding method.
General method for the synthesis of trifluoromethyl chlorides.To a suspension of alcohol (1 equiv.) in heptane (2 mL/mmol), SOCl 2 (1.1 equiv.) was added and the resulting mixture was heated to 85 ºC.The reaction mixture turned from cloudy to transparent.It was stirred for a further 15 min and then cooled down.When it was cold, a solid precipitated.It was filtered and washed with cold petrol ether, thus obtaining the desired chloride, which was used in the next step without further purification.
N General method for the synthesis of trifluoromethyl xanthates.To a solution of chloride (1 equiv.) in absolute EtOH (4 mL/mmol) cooled to 0 • C, potassium ethylxanthogenate (1.2 equiv.) was added and the mixture was stirred at room temperature for 2 h.After that time, the reaction mixture was concentrated, the residue redissolved in Et 2 O and washed with H 2 O.The organic phase was dried and evaporated, and the residue purified by the corresponding method.
Polymerizations.Bulk polymerization of styrene was carried out at 110 • C in sealed tubes after degassing by three pump-freeze-thaw cycles.For the determination of C tr values by the Mayo method, polymerizations were run at different initial xanthate concentrations and were stopped at low St conversion (<10%).When Equation (1) was used with xanthates 9, 10 and 11, C tr was determined as the average of all values calculated for different [X] 0 .For xanthates 13 and 14, C tr was calculated from Equation (1) based on one single [X] 0 from different conversion points.Conversion was determined by gravimetry.

Characterization
NMR spectra were recorded in CDCl3 using a Bruker AMX400 operating at 400 MHz for 1 H and 100 MHz for 13 C.The chemical shifts are expressed in parts per million (ppm) referenced to residual chloroform. 1 H NMR data are reported as follows: δ, chemical shift; multiplicity (recorded as: s, singlet; bs, broad singlet; d, doublet; t, triplet; q, quadruplet; dq, double quadruplet; m, multiplet), coupling constants (J are given in Hertz, Hz) and integration.Infrared absorption spectra were recorded as a solution in CCl4 with a Perkin-Elmer 1600 Fourier Transform Spectrophotometer.Mass spectra were recorded with an HP 5989B mass spectrometer via direct introduction for chemical positive ionization (CI) using ammonia as the reagent gas.Melting points were determined by Reichert microscope apparatus and were uncorrected.HRMS were performed on JEOL JMS-GcMate II, GC/MS system spectrometer.

Figure 1 .
Figure 1.(a) Evolution of Mn (closed symbol) and Đ (open symbol) for 13-mediated RAFT/MADIX polymerization of St in bulk at 110 °C.Solid line: theoretical Mn.The dashed line is a guide to the eye.(b) Overlay of SEC traces.

Figure 1 .
Figure 1.(a) Evolution of M n (closed symbol) and Ð (open symbol) for 13-mediated RAFT/MADIX polymerization of St in bulk at 110 • C. Solid line: theoretical M n .The dashed line is a guide to the eye.(b) Overlay of SEC traces.

Figure 2 .
Figure 2. Evolution of Mn (closed symbol) and Đ (open symbol) for X14-mediated RAFT/MADIX polymerization of St in bulk at 110 °C.

Figure 2 .
Figure 2. Evolution of M n (closed symbol) and Ð (open symbol) for X14-mediated RAFT/MADIX polymerization of St in bulk at 110 • C.

Table 1 .
Series of alcohol and chloride precursors and xanthate products.
a Product number; b yield.
a Determined from the Mayo plot.b From Equation (1).
a Determined from the Mayo plot.b From Equation (1).