Synthesis of Tris(trifluoromethyl)nickelates(II)—Coping with “The C 2 F 5 Problem”

: When synthesizing the versatile precursors (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] we recently encountered the problem that marked amounts of C 2 F 5 were incorporated instead of CF 3 under the chosen reaction conditions forming mixed-ligand nickelates [Ni(CF 3 )x(C 2 F 5 )y(MeCN)] − (x + y = 3). We studied the three products with y = 0, 1, or 2, using 19 F nuclear magnetic resonance (NMR) spectroscopy and single-crystal X-ray diffraction. We were able to trace the reaction mechanism and solve the problem by modifying the experimental conditions.

We recently contributed to this topic by exploring the synthesis, structures, and electrochemical properties of tris(trifluoromethyl)nickelates(II) of the type (NMe 4 )[Ni(CF 3 ) 3 (F-NHC)] containing fluorinated N-heterocyclic carbene ligands (F-NHC) of the N,N-di(perfluorophenyl)imidazolylidene type [14].These carbene complexes were synthesized from the previously reported (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] (1) and the corresponding carbene ligands.The precursor 1, as well as its bis-and tetrakis-CF 3 congeners, are prepared through transmetalation of the CF 3 ligands from silver.In the course of our syntheses of 1 and its use as precursor for [Ni(CF 3 ) 3 (L)] − complexes, we observed in 19 F nuclear magnetic resonance (NMR) spectra and crystal structures of the products that CF 3 groups were partially replaced by C 2 F 5 groups [14] as a result of unintended chain elongation-which we called "the C 2 F 5 problem".However, there is increasing interest in the chemistry of the C 2 F 5 group [2,4,12,[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33], including the use of the Ruppert-Prakash reagent Si(CH 3 ) 3 (C 2 F 5 ) (TMS-C 2 F 5 ) [17] for perfluoroethylation reactions with applications in bioactive molecules [27].Compared with the CF 3 group, the far less explored C 2 F 5 provides increased steric demand, higher electronegativity, and lipophilicity [34][35][36][37] that render it an attractive alternative to CF 3 [21,[25][26][27]33].The formation of C 2 F 5 during our syntheses of Ni(CF 3 ) 3 derivatives is, therefore, interesting regarding potential targeted use of the method for the synthesis of C 2 F 5 complexes.On the other hand, understanding the phenomenon of unintended C 2 F 5 formation during the synthesis of tris(trifluoromethyl)nickelates(II) is also relevant to any future work on Ni(CF 3 ) n complexes in order to avoid "the C 2 F 5 problem".While this chemistry is extremely interesting, the synthetic work is not trivial and requires careful consideration and optimization regarding setups, reagents, and reaction conditions, especially when using Schlenk techniques as opposed to a glovebox.(TMS-C2F5) [17] for perfluoroethylation reactions with applications in bioactive molecules [27].Compared with the CF3 group, the far less explored C2F5 provides increased steric demand, higher electronegativity, and lipophilicity [34][35][36][37] that render it an attractive alternative to CF3 [21,[25][26][27]33].The formation of C2F5 during our syntheses of Ni(CF3)3 derivatives is, therefore, interesting regarding potential targeted use of the method for the synthesis of C2F5 complexes.On the other hand, understanding the phenomenon of unintended C2F5 formation during the synthesis of tris(trifluoromethyl)nickelates(II) is also relevant to any future work on Ni(CF3)n complexes in order to avoid "the C2F5 problem".While this chemistry is extremely interesting, the synthetic work is not trivial and requires careful consideration and optimization regarding setups, reagents, and reaction conditions, especially when using Schlenk techniques as opposed to a glovebox.Herein, we report on the characterization of mixed CF3/C2F5-containing Ni(II) complexes formed through perfluoroalkyl chain elongation, and present an assessment of the underlying mechanism, and how we were able to prevent this unwanted side reaction through modification of the reaction conditions.
AgCF3 is reported to exist in solution in a disproportionation equilibrium with Ag I [Ag I (CF3)2], or to decay to Ag I [Ag III (CF3)4] under elimination of 2 equivalents Ag 0 (elemental silver), depending on the temperature and the presence of less noble metal ions, which catalyze the elimination reaction [38,39].As we usually observed a silver mirror in the flask, we assume the formation of Ag I [Ag III (CF3)4] in significant amounts.
Herein, we report on the characterization of mixed CF 3 /C 2 F 5 -containing Ni(II) complexes formed through perfluoroalkyl chain elongation, and present an assessment of the underlying mechanism, and how we were able to prevent this unwanted side reaction through modification of the reaction conditions.

NMR Spectroscopy and Stereoselectivity
The 19 F NMR spectra (Figure 1, complete spectra in the Supplementary Materials) showed the coexistence of (NMe4 AgCF 3 is reported to exist in solution in a disproportionation equilibrium with Ag I [Ag I (CF 3 ) 2 ], or to decay to Ag I [Ag III (CF 3 ) 4 ] under elimination of 2 equivalents Ag 0 (elemental silver), depending on the temperature and the presence of less noble metal ions, which catalyze the elimination reaction [38,39].As we usually observed a silver mirror in the flask, we assume the formation of Ag I [Ag III (CF 3 ) 4 ] in significant amounts.
The solution was decanted from the produced solids after 2 h and added to a suspension of [NiBr 2 (dme)] (dme = 1,2-dimethoxyethane) and NMe 4 I in MeCN.After 2 days reaction time, workup was performed and the product was analyzed primarily via 19 F NMR (full analytics in the Section 3).

NMR Spectroscopy and Stereoselectivity
The 19 F NMR spectra (Figure 1, complete spectra in the Supplementary Materials) showed the coexistence of (NMe

Single-Crystal X-ray Diffraction (sc-XRD)
We were unable to separate the precursor complexes 1 to 3 by crystallization but found that extraction of the precursor mixture with THF yields fractions that contain only minor amounts of 1, suggesting that C2F5 substitution does have a significant effect on solubility [26,27].Recrystallization of such a sample and sc-XRD study showed that 2 is contaminated with approximately 20% 1 in the solid state.Remarkably, CF3 and C2F5 fractionally occupy the same position in the otherwise identical structures (monoclinic, P21/n).The structure of the complex species of 1 was previously reported in the structure of (PPh4)[Ni(CF3)3(MeCN)] (P1) [16] and we refined the structure of 2 with an 80:20 occupancy of 2:1.This explains while recrystallization does not allow separation of 1 and 2. Further, the structure of 2 confirmed our interpretation from 19 F NMR of the C2F5 group as located cis to MeCN (Figure 2).
The Ni(II) center in 2 adopts a distorted square planar coordination with the "trans"

Single-Crystal X-ray Diffraction (sc-XRD)
We were unable to separate the precursor complexes 1 to 3 by crystallization but found that extraction of the precursor mixture with THF yields fractions that contain only minor amounts of 1, suggesting that C 2 F 5 substitution does have a significant effect on solubility [26,27].Recrystallization of such a sample and sc-XRD study showed that 2 is contaminated with approximately 20% 1 in the solid state.Remarkably, CF 3 and C 2 F 5 fractionally occupy the same position in the otherwise identical structures (monoclinic, P2 1 /n).The structure of the complex species of 1 was previously reported in the structure of (PPh 4 )[Ni(CF 3 ) 3 (MeCN)] (P1) [16] and we refined the structure of 2 with an 80:20 occupancy of 2:1.This explains while recrystallization does not allow separation of 1 and 2. Further, the structure of 2 confirmed our interpretation from 19 F NMR of the C 2 F 5 group as located cis to MeCN (Figure 2).The Ni(II) center in 2 adopts a distorted square planar coordination with the "trans" angles N1-Ni-C3 and C4-Ni-C1 of around 171 • deviating markedly from the ideal 180 • (Table 1).This represents a slight distortion towards tetrahedral with a τ 4 value of 0.13 (ideal tetrahedral = 1, ideal square planar = 0) [38].In the previously reported structure (PPh 4 )[Ni(CF 3 ) 3 (MeCN)] (P1) the corresponding angles of about 178 • are much closer to the ideal 180 • and the τ 4 value of 0.03 is much smaller.At the same time, the bonding distances around Ni(II) are very similar for 2 and P1 with Ni-C3 being the shortest in agreement with the weaker MeCN ligand in trans position to this CF 3 group.The corresponding Ni-CF 3 bond in P1 and the Ni-CF 2 CF 3 bond in 2 have virtually the same length.a From sc-XRD.The structure of P1 is reported in ref. [16].
Further, the C 2 F 5 C-C bond with 1.513(3) Å and the Ni-C-C angle of 119.6(1) • in 2 are very similar to those found in the related complexes [39] showing a C-C bond length of around 1.51 Å and Ni-C-C angles ranging from 115 to 120 • .
From this comparison, we conclude that, while the coordination around Ni(II) is flexible concerning the trans angles, the length of the Ni-C bonds are quite invariable, which points to a strong σ-donating power of the CF 3 and C 2 F 5 ligands.The C-C bond length of the C 2 F 5 ligand is also rather invariant and the space requirement for the CF 3 to C 2 F 5 replacement in 1 to 2 is provided through the flexible geometry around Ni(II) in addition to some flexibility of the Ni-C-C angle.

Reaction with N-Heterocyclic Carbene (NHC) Ligands
When the mixed precursor material (NMe 4 )[Ni(CF 3 ) x (C 2 F 5 ) y (MeCN)] (x + y = 3; 1 to 3) was reacted with fluorinated N-heterocyclic carbene (NHC) ligands as reported elsewhere [14], we observed significant amounts of cis-(NMe 4 )[Ni(CF 3 ) 2 (C 2 F 5 )(L)] via 19 F NMR, meaning that the C 2 F 5 -containing precursor readily reacts with carbene ligands.For example, in the case of the fluorinated NHC ligand N,N-bis(2,4-difluorophenyl)imidazolylidene (F-NHC), the spectra (Figure 3) show that even a crystalline sample of the material contains only approximately 75 to 80% (NMe 4 )[Ni(CF 3 ) 3 (F-NHC)] (4) and 20 to 25% cis- sc-XRD of the sample, as well as of other fluorinated NHC complexes from the same precursor as reported previously [14], confirmed the coexistence of 4 and 5 in the form of CF 3 /C 2 F 5 replacement in the crystal structure as seen for the precursor.As for the pair 1 and 2, the CF 3 and C 2 F 5 ligands occupy the same positions in the otherwise identical crystal structure.Refinement of the "disordered" crystal structure of 4/5 yielded an occupancy of approximately 85% CF 3 and 15% C 2 F 5 [14].Therefore, 4 and 5 cannot be separated by crystallization to obtain pure 4 either.This is a somewhat lower content of 5 than we concluded from the NMR spectra for the bulk crystalline sample, though in a similar range.Judging from the varying C 2 F 5 contents in the structures of different carbene complexes from our previously published study [14], it is reasonable to assume that C 2 F 5 incorporation may vary between crystals both statistically and throughout the process of crystal formation from a given solution.Pure 4, including a crystal structure, was previously obtained from pure 1 precursor [14].sc-XRD of the sample, as well as of other fluorinated NHC complexes from the same precursor as reported previously [14], confirmed the coexistence of 4 and 5 in the form of CF3/C2F5 replacement in the crystal structure as seen for the precursor.As for the pair 1 and 2, the CF3 and C2F5 ligands occupy the same positions in the otherwise identical crystal structure.Refinement of the "disordered" crystal structure of 4/5 yielded an occupancy of approximately 85% CF3 and 15% C2F5 [14].Therefore, 4 and 5 cannot be separated by crystallization to obtain pure 4 either.This is a somewhat lower content of 5 than we concluded from the NMR spectra for the bulk crystalline sample, though in a similar range.Judging from the varying C2F5 contents in the structures of different carbene complexes from our previously published study [14], it is reasonable to assume that C2F5 incorporation may vary between crystals both statistically and throughout the process of crystal formation from a given solution.Pure 4, including a crystal structure, was previously obtained from pure 1 precursor [14].

Mechanistic Considerations on the CF3 to C2F5 Conversion and Refined Reaction Procedure
It has long been known that conditions for the preparation of metal-CF3 complexes can also generate metal-C2F5 byproducts, and M=CF2 intermediates have been proposed [29,[40][41][42][43][44].The formation of difluorocarbene requires the elimination of fluoride from CF3,

Mechanistic Considerations on the CF 3 to C 2 F 5 Conversion and Refined Reaction Procedure
It has long been known that conditions for the preparation of metal-CF 3 complexes can also generate metal-C 2 F 5 byproducts, and M=CF 2 intermediates have been proposed [29,[40][41][42][43][44].The formation of difluorocarbene requires the elimination of fluoride from CF 3 , which is promoted by LEWIS acids such as (hard) metal cations [45] or protons.We assume that C 2 F 5 via CF 2 insertion is formed at Ag during the reaction of TMS-CF 3 with AgF since chain elongation during a later step at Ni would probably lead to the formation of Ni-F complexes as significant byproducts.For example, a common side product of moistureinduced decomposition of Ni-CF 3 complexes, where protons promote the elimination of fluoride from CF 3 , is the fluoride-bridged dimer {[Ni(CF 3 ) 2 F]} 2 [14], which can also result from thermal decomposition [39].
Nebra et al. reported that an Ag III =CF 2 complex can be generated from [Ag III (CF 3 ) 4 ] − in the presence of K•••F interactions, though the reaction is highly endothermic and the resulting [Ag III (CF 3 ) 3 (CF 2 )] species is very reactive due to negligible π-backdonation from Ag III to difluorocarbene according to their DFT analysis [45].However, they do not mention C 2 F 5 formation during their trifluoromethylation experiments with [Ag III (CF 3 ) 4 ] − in the presence of K + [45].For the formation of Cu-C 2 F 5 (and even Cu-C 3 F 7 ) from Cu-CF 3 , Hu et al. concluded that the plausible intermediate is a difluorocarbene complex Cu I =CF 2 [29].Combining these two pieces of information, we believe that the chain elongation is occurring at Ag I , rather than Ag III , in our case.Due to the higher π-backdonation ability of Ag I , Ag I =CF 2 formation promoted by M•••F interactions is probably more facile from Ag I -CF 3 .
Thus, considering that both the disproportionation of Ag I -CF 3 into Ag I [Ag III (CF 3 ) 4 ] and Ag 0 and α-fluoride elimination are promoted by LEWIS acids [40,41,45], we are looking at two competing, opposing effects of the presence of trace metals, or moisture, on C 2 F 5 formation during the reaction of TMS-CF 3 with AgF.While the formation of Ag I [Ag III (CF 3 ) 4 ] can occur at any temperature above −30 • C, α-fluoride elimination is additionally promoted thermally.
We thus concluded that the C 2 F 5 -containing by-products are generated when the reaction of TMS-CF 3 with AgF is performed in concentrated MeCN solution at elevated temperatures.The formation of C 2 F 5 was facilitated by the exothermic reaction during upscaling experiments with high reagent concentrations.Based on this, we modified the reaction protocol for the synthesis of (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] (1) (Scheme 3).Reducing the overall concentration of reagents and employing better heat management during the reaction of AgF with TMS-CF 3 , while maintaining the original conditions for the reaction of the resulting Ag-CF 3 with Ni(II), allowed us to eliminate the formation of the C 2 F 5containing 2 and 3.This further supports our assumption that the chain elongation occurs at Ag as opposed to Ni as stated earlier.Furthermore, we note that the quality of the AgF used for the reaction is extremely vital to the success of the synthesis with respect to the "C 2 F 5 problem" due to the complex effects of the presence of free LEWIS acid on Ag I [Ag III (CF 3 ) 4 ] vs. Ag I =CF 2 formation.
tion C2F5 formation during their trifluoromethylation experiments with [Ag III (CF3)4] -in the presence of K + [45].For the formation of Cu-C2F5 (and even Cu-C3F7) from Cu-CF3, Hu et al. concluded that the plausible intermediate is a difluorocarbene complex Cu I =CF2 [29].Combining these two pieces of information, we believe that the chain elongation is occurring at Ag I , rather than Ag III , in our case.Due to the higher π-backdonation ability of Ag I , Ag I =CF2 formation promoted by M•••F interactions is probably more facile from Ag I -CF3.Thus, considering that both the disproportionation of Ag I -CF3 into Ag I [Ag III (CF3)4] and Ag 0 and α-fluoride elimination are promoted by LEWIS acids [40,41,45], we are looking at two competing, opposing effects of the presence of trace metals, or moisture, on C2F5 formation during the reaction of TMS-CF3 with AgF.While the formation of Ag I [Ag III (CF3)4] can occur at any temperature above −30 °C, α-fluoride elimination is additionally promoted thermally.
We thus concluded that the C2F5-containing by-products are generated when the reaction of TMS-CF3 with AgF is performed in concentrated MeCN solution at elevated temperatures.The formation of C2F5 was facilitated by the exothermic reaction during upscaling experiments with high reagent concentrations.Based on this, we modified the reaction protocol for the synthesis of (NMe4)[Ni(CF3)3(MeCN)] (1) (Scheme 3).Reducing the overall concentration of reagents and employing better heat management during the reaction of AgF with TMS-CF3, while maintaining the original conditions for the reaction of the resulting Ag-CF3 with Ni(II), allowed us to eliminate the formation of the C2F5-containing 2 and 3.This further supports our assumption that the chain elongation occurs at Ag as opposed to Ni as stated earlier.Furthermore, we note that the quality of the AgF used for the reaction is extremely vital to the success of the synthesis with respect to the "C2F5 problem" due to the complex effects of the presence of free LEWIS acid on Ag I [Ag III (CF3)4] vs. Ag I =CF2 formation.If Ag-C 2 F 5 species are, thus, the reason for the formation of the observed mixedligand nickelates [Ni(CF 3 ) x (C 2 F 5 ) y (MeCN)] (x + y = 3), the observed stereoselectivity of cis-(NMe 4 )[Ni(CF 3 ) 2 (C 2 F 5 )(MeCN)] (2) and trans-(NMe 4 )[Ni(CF 3 )(C 2 F 5 ) 2 (MeCN)] (3) would then simply be a matter of the increased bulkiness of C 2 F 5 compared with CF 3 .To further support such Ag-C 2 F 5 species, we tried to study them by 19 F NMR in solutions containing TMS-CF 3 and AgF, but failed due to the formation of elemental Ag impeding NMR.

General Synthesis Conditions
All manipulations were performed using standard Schlenk techniques.Solvents were freed of oxygen by purging with Ar and dried using freshly activated 3 Å molecular sieves.
[NiBr 2 (dme)] was prepared according to a reported procedure [50].The reagents, 3.0 g (9.72 mmol; 1.0 eq.) [NiBr 2 (DME)] and 1.96 g (9.72 mmol; 1.0 eq.) NMe 4 I, were weighed under ambient conditions and quickly transferred into a flamedried Schlenk-flask.The solids were dried under reduced pressure at 60 • C for 1 h.AgF (3.7 g; 29.16 mmol; 3.0 eq.) was weighed under ambient conditions, quickly transferred into a separate, flame-dried Schlenk-flask, and dried under reduced pressure at 60 • C for 1 h.After cooling to RT, both flasks were backfilled with Ar.MeCN (25 mL) was added to each flask.(CH 3 ) 3 SiCF 3 (5.02mL; 4.84 g; 34.02 mmol; 3.5 eq.) was quickly added to the suspended AgF.The resulting solution was stirred for 2 h under exclusion of light.Then, the AgCF 3 -containing solution was transferred via a cannula into the [NiBr 2 (DME)]/NMe 4 I suspension.The resulting bright yellow suspension was stirred vigorously for 2 d at RT under exclusion of light.The suspension was filtered over Celite ® and evaporated.A total of 2.22 g of a bright yellow solid was isolated.The solid contained a 5:4:1 mixture of (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] ( 1 The reagents, 2.5 g (8.10 mmol; 1.0 eq.) [NiBr 2 (DME)] and 0.89 g (8.10 mmol; 1.0 eq.) NMe 4 Cl, were weighed under ambient conditions and quickly transferred into a flamedried Schlenk-flask.The solids were dried under reduced pressure at 80 • C for 1 h.AgF (3.19 g; 25.11 mmol; 3.1 eq.) was weighed under ambient conditions, quickly transferred into a separate, flame-dried Schlenk-flask, and dried under reduced pressure at RT for 1 h.After cooling to RT both flasks were backfilled with Ar.MeCN (40 mL) was added to each flask.(CH 3 ) 3 SiCF 3 (3.55mL; 3.69 g; 25.92 mmol; 3.2 eq.) was added to a third flame-dried Schlenk-flask and diluted with 15 mL MeCN.All three flasks were cooled to 0 • C. The diluted (CH 3 ) 3 SiCF 3 was added dropwise to the suspended AgF.The resulting solution was stirred for 15 min under exclusion of light.Afterwards, the AgCF 3 -containing solution was transferred via a cannula into the [NiBr 2 (DME)]/NMe 4 Cl suspension.The resulting bright yellow suspension was stirred vigorously for 2 d at RT under exclusion of light.The suspension was filtered over Celite ® and partially evaporated.A total of 66 mL of a bright yellow MeCN solution was recovered.The concentration of (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] (1) in the solution was determined using quantitative NMR.The solution contained 0.082 mmol/mL, corresponding to 5.41 mmol (67%) (NMe 4 )[Ni(CF 3 ) 3 (MeCN)]. 19

Conclusions
In our recent attempts to synthesize the versatile precursor (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] (1) we had observed that marked amounts of C 2 F 5 were incorporated instead of CF 3 under the chosen reaction conditions with the formation of mixed-ligand nickelates [Ni(CF 3 ) x (C 2 F 5 ) y (MeCN)] − (x + y = 3).The C 2 F 5 -containing precursor derivatives readily form NHC complexes like the tris-CF 3 precursor 1.We studied the precursors with x = 3 (1), x = 2 (cis-2), and x = 1 (trans-3) and their stereochemistry, as well as the corresponding F-NHC complexes (F-NHC = N,N-bis(2,4-difluorophenyl)imidazolylidene) using 19 F nuclear magnetic resonance (NMR) spectroscopy and single-crystal X-ray diffraction, and surprisingly found significant CF 3 /C 2 F 5 replacement in the solid-state structures, explaining our failure to purify the materials through crystallization.We were able to trace the reaction mechanism and assume transmetalating Ag-C 2 F 5 species to be responsible for the C 2 F 5 ligands in the nickelates, and we finally solved the problem by modifying the experimental conditions.We found that the synthesis of (NMe 4 )[Ni(CF 3 ) 3 (MeCN)] (1) must be carried out avoiding high concentrations of TMS-CF 3 and excessive formation of heat.For larger reaction batches, efficient cooling of the reaction vessel must be provided to suppress formation of C 2 F 5 species (NMe 4 )[Ni(CF 3 ) x (C 2 F 5 ) y (MeCN)] (x + y = 3).Furthermore, due to the complex role of LEWIS acid impurities in the reactions of Ag-CF 3 complexes, it should be kept in mind that trace impurities in AgF may also have a notable influence on C 2 F 5 formation during the synthesis of 1.

Figure 1 . 19 F
Figure 1. 19F NMR spectrum (282 MHz, MeCN-d 3 ) of (NMe 4 )[Ni(CF 3 ) x (C 2 F 5 ) y (MeCN)] (x + y = 3), 1, 2, and 3, with partial CF 3 to C 2 F 5 replacement.The signal of the CF 3 ligand trans to the MeCN ligand, which is observed at −25.1 ppm for 1, is shifted slightly and progressively downfield upon replacement of one or both CF 3 ligands cis to MeCN with C 2 F 5 .Interestingly, only the cis CF 3 ligands are prone to replacement by C 2 F 5 , while the formation of trans-[Ni(CF 3 ) 2 (C 2 F 5 )(MeCN)] − with C 2 F 5 located trans to the coordinated MeCN has not yet been observed.This is fully in line with the tris-C 2 F 5 derivative [Ni(C 2 F 5 ) 3 (MeCN)] − not being observed at all, not even in traces by the very sensitive 19 F NMR method.The CF 3 ligand trans to MeCN seems to be far less prone to be replaced by C 2 F 5 .Both the selective trans stereochemistry of

Inorganics 2024 ,
12,  x FOR PEER REVIEW 5 of 12 2,2′;6′,6″-terpyridine)[39] showing a C-C bond length of around 1.51 Å and Ni-C-C angles ranging from 115 to 120°.From this comparison, we conclude that, while the coordination around Ni(II) is flexible concerning the trans angles, the length of the Ni-C bonds are quite invariable, which points to a strong σ-donating power of the CF3 and C2F5 ligands.The C-C bond length of the C2F5 ligand is also rather invariant and the space requirement for the CF3 to C2F5 replacement in 1 to 2 is provided through the flexible geometry around Ni(II) in addition to some flexibility of the Ni-C-C angle.
at ambient temperature in MeCN to form AgCF 3 .