Electronic vs . Steric Hindrance Effects in Amine Ligands to Ru-Based Initiators for ROMP

Complexos do tipo [RuCl2(PPh3)2(amina)x], com NH2Ph (1; x = 2), NH2Bz (2; x = 2) and NHBuPh (3; x = 1) na presença de etildiazoacetato (EDA), foram aplicados em reações de polimerização via metátese por abertura de anel (ROMP) de norborneno (NBE), norbornadieno (NBD) e diciclopentadieno (DCPD). Rendimentos quantitativos de poliNBE foram obtidos a 50 °C por 30 min usando 1 e por 5 min usando 2. Isso ocorreu a 25 °C por 5 min quando foi usado 3. Os valores do índice de polidispersidade (PDI) variaram de 3,5 a 1,6 (Mw = 10 -10 g mol). O complexo 3 foi ativo para ROMP de NBD e DCPD, bem como para copolimerização de NBE com NBD ou com DCPD. O maior caráter σ-doador de NH2Bz em 2, em relação a NH2Ph em 1, favoreceu a reatividade do complexo 2 que é hexacoordenado como 1. O ângulo de cone de NHBuPh definiu a pentacoordenação em 3 e a sua melhor reatividade para ROMP, apesar do baixo caráter σ-doador como o de NH2Ph.


Introduction
][8][9] A diversity of catalysts have been tested from complexes of Ru, W, Mo, Ti, Ta, Nb, V, Re, Ni, etc. 1,[10][11][12][13][14][15] Besides the electronic nature of the metal center to select an olefin instead of an oxo-compound in the medium, 16 the success of ROMP is strongly influenced by ancillary ligands. 17,18hese ligands directly influence the reaction mechanism via an intramolecular interaction between a coordinated olefin and a metal-carbene moiety to produce a metallocyclobutane intermediate. 1,2The correct geometric arrangement to permit the orbital overlap and the electronic activation are determinant for OM success. 19][22][23][24] Usually, the ancillary ligands in OM with Ru-based complex are bulky and good σ-donors. 20The steric hindrance comes from the cone angle (θ), as defined by Tolman. 25The σ-donor ability is rationalized in terms of pK a value of the molecule. 26e have worked in the development of the [RuCl 2 (phosphine) x (amine) y ]-type complex for ROMP and copolymerization (ROMCP) with the idea of combining phosphine and amine to obtain helpful electronic and steric effects for the occurrence of reactions.A long list of amines has been attained for this proposal, [27][28][29][30][31] with significant changes in the polymer characteristic, such as molecular weight, polydispersity index (PDI), thermal properties, morphological shape, etc. [27][28][29][30][31] In the present study, we focus on the development of Rubased catalysts combining the phosphine PPh 3 with amines of the type NHR 1 R 2 to act as ancillary ligands, where the amine substituents R 1 and R 2 permute their characteristics considering the electronic nature and steric hindrance.The amines phenylamine (NH 2 Ph; aniline) and benzylamine (NH 2 Bz) present R 1 = H and different R 2 groups (Ph or Bz), where the CH 2 unit between the N and the Ph ring in NH 2 Bz promotes an increase in the σ-donor nature as evaluated from the pK a values (Figure 1), without significantly changing the cone angles (θ).On the other hand, changing R 1 = H for the n Bu chain in NH 2 Ph results in the amine N-n butylphenylamine (NHBuPh; N-n Butylaniline), which presents similar σ-donor nature to R 1 = H (Figure 1), but a larger cone angle in NHBuPh is expected.The main purpose of this study is to analyze the more appropriate characteristics (σ-donor and θ) of the amines to combine with PPh 3 in Ru-based-type complex for ROMP and ROMCP of norbornene (NBE), norbornadiene (NBD) and dicyclopentadiene (DCPD).The use of PPh 3 is appropriate to avoid easy oxidation of the Ru(II) metal center (a low spin d 6 configuration) and be helpful with the steric hindrance.In addition, PPh 3 is air-stable in the solid state and cheap.The current amines are non-π-receptor molecules and they act only as σ-donor ligands to promote olefin activation via metathesis.
This complex was obtained following the literature with few changes. 28[RuCl 2 (PPh 3 ) 3 ] (0.52 mmol; 0.50 g) was added to a solution of NH 2 Ph (10.4 mmol; 0.97 g) in acetone (30 mL).The solution became green for 5 min and changed to purple for 12 h under stirring at RT.The volume was reduced to ca. [RuCl 2 (PPh 3 ) 3 ] (0.52 mmol; 0.50 g) was added to a solution of NHBuPh (5.2 mmol; 0.78 g) in acetone:CHCl 3 (25 mL).The mixture was stirred for 24 h at RT and the volume was reduced to ca. 5

Equipment
Elemental analyses were performed in a Perkin-Elmer CHN 2400 at the Institute of Chemistry, USP.ESR measurements of the solid sample were carried out at 77 K, using a Bruker ESR 300C (X-band) equipped with a TE102 cavity and HP 52152A frequency counter.FTIR measurements were performed in CsI pellets on a Bomem FTIR MB 102.The NMR ( 1 H; 13 C{ 1 H}; 31 P{ 1 H}) spectra were obtained in CDCl 3 at 25.0 ± 0.1 °C using a Bruker DRX 400 spectrometer.The obtained chemical shifts were reported in ppm relative to the high frequency of tetramethylsilane (TMS) or PF 6 -ion.Size-exclusion chromatography (SEC) analyses were carried out in a Shimadzu Prominence LC system equipped with an LC-20AD pump, a DGU-20A5 degasser, a CBM-20A communication module, a CTO-20A oven at 27 ºC, and an RID-10A detector connected to three PL gel columns (5 mm MIXED-C: 30 cm, Ø = 7.5 mm).Retention time was calibrated with standard monodispersed polystyrene using HPLC-grade CHCl 3 as eluent.PDI is M w / M n .

Polymerization reactions
In a typical ROMP experiment, 1.1 mmol of catalyst (1, 2 or 3) was dissolved in 2 mL of CHCl 3 and a certain amount of monomer (NBE, NBD or/and DCPD) and 5 mL of EDA were added.The solution was stirred for different periods of time at 25 or 50 ± 1 °C in a silicon oil bath.At RT, ca. 5 mL of methanol were added and the polymer was filtered, washed with methanol and dried in a vacuum oven at 27 °C until a constant weight was achieved.The reported yields are average values from catalytic runs performed at least three times, with maximum 10% deviation.The isolated polyNBEs were dissolved in 2 mL of CHCl 3 for SEC measurements.

ROMP with 1 and 2
Complex 1 was active for NBE in the presence of EDA used as carbene source (Table 1).Yields in the range of 20-40% for 5 min at 25 o C were obtained, with the best result at 50 o C with 5 mL of EDA solution (61%).A tendency of larger amounts of EDA in reducing the reaction progress was observed, as in earlier studies. 27he active metal species was poisoned with excess of EDA, decreasing the propagation rate, or the monomer concentration decreased by cyclopropanation occurrence, as possible reasons to explain that event. 33The volume of 5 mL was used throughout this work.

Reactions for 5 min with 1 as a function of [NBE]/[Ru]
molar ratio in the range from 1000 to 5000 showed polyNBE yield increased either at 25 or 50 °C (Figure 2).However, the maximum value was ca.60% at 50 °C.Semi-quantitative yields were obtained at 50 °C when the reaction time was 30 min.Higher yields at 50°C were probably obtained because of a faster displacement of a ligand from the six-coordinated species in the induction period, as could be expected.This is necessary to form the metal-carbene species with a coordinated olefin, followed by metallocyclobutane formation. 1Thus, the temperature probably facilitated the induction towards initiation, but long periods of reaction time were necessary to obtain better results with 1.On the other hand, semi-quantitative yields (ca.90%) were already obtained for 5 min at 50 °C with complex 2 for different starting [NBE]/[Ru] molar ratios (Table 2).
Decreased results for [NBE]/[1] = 10000 at 25 °C for 5 min can be associated with the gelation of the medium (Figure 2), which hinders the diffusion of the monomer to the catalyst connected to the polymer chain.However, this was not important at 50 °C, either for 5 or 30 min, with 67 and 97% of yield, respectively.The M w values changed from 40 to 88 × 10 4 g mol -1 as the PDI values decreased from 1.92 to 1.73 (Table S1).In the case of 2, good yields were also obtained for molar ratio of [NBE]/[Ru] = 10000, with PDI below 1.9 (Table 2).Considering the catalyst activity in terms of turnover number (TON), it can be observed that the polymer production increases linearly with the monomer load (Table 2).This relationship can also be observed in the case of 1 at 50 °C (Table S1).This is an important result, since simple and stable complexes were able to afford ROMP of a large amount of monomer.
M w values with a magnitude order of 10 4 -10 5 g mol -1 were obtained either with 1 or 2. PDI values were higher than two for low [NBE] loads, but decreased to values lower than two for higher [NBE] loads (Figure 2; Tables 2 and S1).This can be associated with the induction and initiation reactions, where higher temperatures accelerated these steps affording broad dispersion in the molecular weights, but larger monomer loads resulted in the insertion of more units of monomers in the growing chains, with further time consumption.
The initiator 1 was inactive for NBD at 25 °C for periods from 5 up to 240 min and 2 produced only 3-6% of polyNBD under similar conditions.At 50 °C, 13% with 1 and 29% with 2 were obtained up to 240 min, with no dependence on time (Table S2).An explanation for the low yields is the possible double coordination of the monomer to the metal complex, which works as a chelating ligand. 34

ROMP with 3
High yields (79 to 99%) were obtained at 25 °C for 5 min with [NBE]/[Ru] molar ratios in the range of 1000-10000 (Figure 3).This differs from the results with 1 and 2, where semi-quantitative yields were only obtained at 50 o C for 30 and 5 min, respectively.
M w values with an order of magnitude of 10 4 g mol -1 and narrower distributions resulted in high NBE loads (Figure 3).M w increased by one order of magnitude for a run with [NBE]/[Ru] molar ratio of 10000 in 4 mL of solvent (from 10 4 to 10 5 g mol -1 ).In this case, the yield decreased ca.10%, but the PDI improved from 1.96 to 1.60.The increase in the M w value probably occurred because of the better dissolution of the growing polymer in larger volume, providing the accessibility of the monomer to the metal center.Rapid gelation of the solution occurred with 2 mL of solvent preventing short chains from increasing.
Syntheses of polyNBD were sensitive to temperature and period of reaction time, with 90% yield at 50 °C for 30 min (Table 3; entries 2-4).The yield was also high for 5 min at 50 °C (78%), but it differed from the quantitative results for polyNBE at 25 °C for 5 min (Table 3; entry 1).This was probably because the coordination of the two cyclic olefins in the NBD molecule, which could also explain the low yields with DCPD (Table 3; entries 5-6).Higher temperatures prevented the stabilization of the catalyst through double coordination of NBD, 31 with loss of yield in the case of DCPD (26-21% yield at 50 °C for 60-120 min, Table 3; entries 5-6).
PolyDCPD was not produced at 25 °C for 60 min.PolyNBD and polyDCPD were insoluble in CHCl 3 owing to crosslinking occurrence.4).In the case of fixed NBE load, a linear decrease of semi-quantitative yield to 49% was observed when the NBD load increased up to feed ratio of [NBE]/[NBD] = 1 (Figure 4).On the other hand, increasing the NBE load in solutions with fixed amount of NBD greatly reduced the yield compared with that of the homopolymerization for any feed ratio of [NBD]/[NBE] (Table 4).In the latter case, an ascendant linear trend line can be observed with the rising NBE load (Figure 4).
The results are consistent with the production of copolymers, considering that blended polymers will be isolated with yields in correlation with the production of the homopolymers where the reactivities of 3 with NBE and NBD are very high.
Therefore, the profiles can be associated with the reactivity ratios of the monomers to produce the copolymer, defining how much NBE and NBD was inserted in polyNBE (r 1 ), and how much NBD and NBE was inserted in polyNBD (r 2 ), considering the feed ratios [NBE]/[NBD] and [NBD]/[NBE]. 35,36The following equation represents the reactivity ratios (r) described in terms of the rate constants (k) for the propagation steps, 35 where k NBE/NBE is polyNBE inserting NBE, k NBE/NBD is polyNBE inserting NBD, k NBD/NBD is polyNBD inserting NBD, and k NBD/NBE is polyNBD inserting NBE: Unfortunately, the chemical compositions of the isolated materials were not determined due to their very low solubility in CHCl 3 to take NMR measurements.However, a few suggestions with respect to reactivity ratios can be evaluated from the yield results.
First, given that the presence of comonomers decreased the yields relative to that of the homopolymerizations, it is possible to expect values higher than 1 for both r 1 and r 2 .This occurs because the yields for homopolymerizations  were much better than those for copolymerization, indicating that k NBE/NBE > k NBE/NBD and k NBD/NBD > k NBD/NBE , where insertion of NBE in polyNBE would be more effective than insertion of NBD (r 1 > 1), and insertion of NBD in polyNBD would be more effective than insertion of NBE (r 2 > 1).Furthermore, k NBE/NBE >> k NBD/NBD is expected, considering that quantitative polyNBE yield was obtained at 25 o C for 5 min, whereas semi-quantitative yield for polyNBD was only obtained at 50 o C for 30 min (Table 3; entries 1 and 4).Finally, k NBE/NBD << k NBD/NBE is also expected, taking into account that the abrupt drop in the yield for copolymerization of NBD varying NBE is not observed in copolymerization of NBE varying NBD (Table 4).Thus, from these remarks, r 1 is expected to be higher than r 2 , with occurrence of a greater reactivity with NBE in both cases.It is clearly observed that increasing NBE load in the mixtures, when increasing [NBE]/[NBD] or decreasing [NBD]/[NBE], increases the yields (Figure 4).Under this circumstance, it is important to note the match in the yield results for equimolar loads of monomers Yields from copolymerizations NBE-DCPD and NBD-DCPD were ca.35-31% with equimolar monomer loads at 50 °C for 60 min (Table 3; entries 7 and 8).Double coordination of DCPD to the metal center, as already discussed, could explain the low yields. 34mparison of the reactivity with complexes 1, 2 and 3 Yields for polyNBE and polyNBD were dependent on the reaction time, [monomer]/[Ru] molar ratio and temperature with different yield patterns.
Yields for polyNBD were lower for complexes 1 and 2 regardless of the reaction time and temperature.Complex 3 produced quantitative yield for 30 min and semi-quantitative for 5 min, at 50 °C.
Considering the studied conditions, the complexes presented good reactivity for ROMP with variation in the yields, which can be associated with both the structure of the complexes and the characteristics of the amine ligands combined with PPh 3 .
Complexes 1 and 2 were six-coordinated with two amines while complex 3 was five-coordinated with only one amine molecule, considering the analytical data.The ESR spectra were silent, suggesting that the ruthenium centers presented 2+ oxidation states with low spin d 6 electronic configurations.Elementary analyses indicated complexes with two phosphines and two chloride ions.Typical vibration bands in the FTIR spectra confirmed the presence of chlorides, amines and phosphines in the complexes.The 31 P NMR spectrum of 1 in CDCl 3 at 25 ºC showed only one singlet at 2.0 ppm, suggestive of trans-positioned PPh 3 molecules (Figure S1).Thus, a ttt-[RuCl 2 (PPh3) 2 (NH 2 Ph) 2 ] conformation could be proposed (Figure 5), considering that only one sharp vibration peak in the region of ν(Ru-Cl) stretching mode was observed. 29,30The 31 P NMR spectrum did not change for 12 h.The 31 P NMR spectrum of complex 2 showed a double doublet typical of a six-coordinated complex containing unequivalent cis-positioned P-atoms (26.1 and 40.6 ppm; 2 J PP 7.5 Hz; Figure S2).In this case, a ccc-[RuCl 2 (PPh3) 2 (NH 2 Bz) 2 ] conformation could be considered (Figure 5), in agreement with two ν(Ru-Cl) bands. 29,30In addition, observed signals at 40.4 ppm can be associated with five-coordinated complexes upon dissociation of amine or phosphine ligands, which can undergo dimerization (50.6 ppm; 2 J PP 28.5 Hz) .The 31 P NMR spectrum of 3 showed a singlet at 27.8 ppm (Figure S3), suggesting that the two PPh 3 ligands were equivalently located in the complex.Correlating this result with the FTIR spectrum, where only one ν(Ru-Cl) band was observed that suggested that the two chloride ions were trans-positioned, 29,30 a square pyramidal geometry can be defined for the complex with NHBuPh at the apical position (Figure 5).The spectrum did not show the appearance of other signals as a function of time.It is important to observe the absence of signals associated with binuclear species in the solution, as promptly occurred when dissolving the fivecoordinated [RuCl 2 (PPh 3 ) 3 ] complex in CHCl 3 . 29In the latter case, the displacement of one PPh 3 molecule occurred owing to the large steric hindrance around the coordination sphere.On the other hand, the presence of the amine NHBuPh in 3 tuned the steric hindrance around the metal center without occurrence of departure of ligands.The NMR spectrum of 3 remained the same for 12 h at RT.

Ru
From the supposed geometric structures, it can also be concluded that complex 1 did not either undergo dimerization process or geometric rearrangement when in solution, as a function of time.Both complexes 1 and 3 showed stability at RT for 12 h, contrary to complex 2, where the 31 P NMR spectrum showed evidence of other species in solution.In the case of 1 and 3, restriction of PPh 3 departure occurs by absence of steric hindrance on the metal coordination sphere where the phosphines are transpositioned to each other.Complex 1 is six-coordinated, but the amines present low cone angle.In the case of 3, the amine presents a higher cone angle and the complex is five-coordinated.In both cases, the low σ-donor character of the amines is inefficient to disturb the stability of the complexes.Furthermore, the nonexistence of a geometrical rearrangement in the five-coordinated complex 3 could also be attributed to an adjusted steric hindrance in the coordination metal sphere, provided by the butyl group in the amine, in spite of the low σ-donor character as in NH 2 Ph.In the case of 2, the six-coordinated complex lost PPh 3 because the ccc geometric arrangement provoked a steric hindrance in the coordination sphere, in addition to the high σ-donor character of the NH 2 Bz raising the electronic density in the {Cl 2 Ru(II)} moiety.
Therefore, if the complexes 1 and 2 are six-coordinated (Figure 5), where the amines NH 2 Ph (θ = 111°) and NH 2 Bz (θ = 106°) present almost the same cone angles without providing a significant steric hindrance, an increase in temperature is necessary for better reactivity.However, an increase in the reaction time is not necessary in the case of 2, as in the case of 1, since NH 2 Bz is more basic than NH 2 Ph, besides the fact that the NH 2 Bz molecules can be cis-positioned.In addition, in the case of 1, a slow propagation rate can be expected where the reaction time was cooperative with the temperature to achieve better results.This can be concluded taking into account a weak σ-donor effect from NH 2 Ph to forward monomer activation via a synergism amine → Ru → monomer and a low steric hindrance around the metal center to accelerate the replacement of ligands.Difference in the inertness between 1 and 2 can be correlated with the NMR spectra, where the spectrum of 1 did not show changes for hours and the spectrum of 2 changed over time.As semi-quantitative yields with 1 were only achieved for 30 min at 50 °C, the difference in reactivity can be associated with the better synergism amine → Ru → monomer in 2, owing to the higher σ-donor character of NH 2 Bz (pK a = 9.4), as the cone angles are similar (Figure 1).
The temperature increase favors the formation of the metal-carbene moiety and the coordination of the olefin from the discoordination of ligands.In the case of complex 3, NHBuPh presents a σ-donor character similar to that of NH 2 Ph, but a higher cone angle (expected θ > θ NHEtPh = 126°), which enabled a five-coordinated complex (Figure 5).The higher steric hindrance in 3 provided better results in ROMP with no increase in the reaction time or temperature.Therefore, the nature of the amines and geometric arrangements seems to explain the obtained ROMP results.
Data for 13 C NMR spectra of polyNBE Analyses of the 13 C NMR spectra from polyNBE synthesized with complexes 1, 2 and 3 showed that the fractions of the cis structures (σ c ) were between 0.4 and 0.5, considering the C 1,4 peak areas (Figure S4; Table S3).

Conclusions
Three Ru(II) complexes of the type [RuCl 2 (PPh 3 ) 2 (amine) x ], with amine = NH 2 Ph (1) or NH 2 Bz (2), for x = 2, and amine = NHBuPh (3) for x = 1, have been synthesized.The complexes were tested in the ROMP of NBE, NBD, DCPD and copolymerizations of NBE-NBD, NBE-DCPD and NBD-DCPD, in the presence of EDA.In the series, complex 3 is the most active, probably because there is more steric hindrance due to the number and the bulkiness of the ligands, even with a lower σ-donor character.
Concluding, simple, inexpensive and easy-to-handle amines combined with PPh 3 produced active complexes to prepare polymers and copolymers via ROMP, where the steric hindrance in the initiator is very important for its reactivity.Nowadays, Grubbs-type complexes are very important for the continuous advancement in olefin metathesis.][39]

Figure 4 .
Figure 4. Dependence of yield on the [monomer]/[comonomer] feed ratio for ROMCP of NBE and NBD with 3, at 50 °C for 30 min.
mL under flow of argon.A

Table 1 .
Dependence of yield, M w and PDI on the EDA solution volume for ROMP of NBE with 1 for 5 min; [NBE]/[Ru] = 5000

Table 2 .
Dependence of yield, M w and PDI on the [NBE]/[Ru] molar ratio for ROMP with 2 for 5 min at 50 °C

Table 3 .
Dependence of the yield on the temperature and reaction time for ROMP with 3