Terminal and Internal Alkyne Complexes and Azide-Alkyne Cycloaddition Chemistry of Copper(I) Supported by a Fluorinated Bis(pyrazolyl)borate.

Copper plays an important role in alkyne coordination chemistry and transformations. This report describes the isolation and full characterization of a thermally stable, copper(I) acetylene complex using a highly fluorinated bis(pyrazolyl)borate ligand support. Details of the related copper(I) complex of HC≡CSiMe3 are also reported. They are three-coordinate copper complexes featuring η2-bound alkynes. Raman data show significant red-shifts in C≡C stretch of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) and [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) relative to those of the corresponding alkynes. Computational analysis using DFT indicates that the Cu(I) alkyne interaction in these molecules is primarily of the electrostatic character. The π-backbonding is the larger component of the orbital contribution to the interaction. The dinuclear complexes such as Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 display similar Cu-alkyne bonding features. The mononuclear [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) complex catalyzes [3 + 2] cycloadditions between tolyl azide and a variety of alkynes including acetylene. It is comparatively less effective than the related trinuclear copper catalyst {μ-[3,5-(CF3)2Pz]Cu}3 involving bridging pyrazolates.

1. Introduction

Copper is an important metal in alkyne chemistry. It mediates a number of transformations of acetylene as well as larger alkynes including cycloaddition chemistry [1,2,3,4,5], cyclopropenation [6,7,8]. Terminal and Internal Alkyne Complexes and Azide-alkyne Cycloaddition Chemistry of Copper(I) Supported by a Fluorinated Bis(pyrazolyl)borate partial hydrogenation, [9,10] hetero atom-hydrogen bond additions [11,12,13,14,15,16], Csp-H bond functionalizations, and alkyne coupling processes [17,18,19,20,21,22,23,24,25,26]. Copper catalyzed azide-alkyne cycloaddition (CuAAC) is perhaps the most popular among the different reaction types due to its virtues of mild reaction conditions, high yields, and regioselectivity, with a major impact on organic and materials chemistry to chemical-biology applications [2,5,27,28,29,30,31,32]. Copper based materials are also useful for the separation of acetylene from CO2 and acetylene storage [33,34,35,36,37]. Copper alkyne or alkynide complexes [38] are believed to be key intermediates in many of these reactions. Copper alkynes are used as precursors for the copper deposition as well [39,40,41].

Over the past few years, we have been working on the coordination chemistry [42,43,44,45,46,47] and transformations [6,47,48,49] of alkynes involving copper. For example, we demonstrated that the trinuclear copper(I) pyrazolate {μ-[3,5-(CF3)2Pz]Cu}3 [50] is an effective catalyst for facile azide-alkyne cycloaddition leading to 1,2,3-triazoles, alkyne C(sp)-H bond carboxylation with CO2, and S-H addition to alkyne moiety [47]. Some of the alkyne complexes of copper pyrazolates show interesting luminescence [46]. We also discovered that the mononuclear, bis(pyrazolyl)borate [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1, Pz = pyrazolyl; Figure 1) is an excellent mediator of cyclopropenation chemistry of alkynes with ethyl diazoacetate.[6]

Figure 1

[H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1) and a selected group of structurally characterized copper complexes of acetylene, Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)2 (2) and Cu4(μ-[3,5-(CF3)2Pz])4(μ-HC≡CH)2 (3).

In this work, we describe the use of [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) as a catalyst in azide-alkyne cycloaddition chemistry between p-tolylazide and several alkynes including acetylene and trimethylsilylacetylene. We also describe the isolation of an acetylene complex as well as larger alkyne complexes involving the copper bis(pyrazolyl)borate moiety [H2B(3,5-(CF3)2Pz)2]Cu. Such compounds are of current interest [38,51,52] and allow a comparison of mono-nuclear copper bis(pyrazolyl)borates to dinuclear copper(I) pyrazolates (e.g., [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) and Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)2 (2)). It is also noteworthy that despite the long history of copper(I)-acetylene chemistry [53,54], copper(I) acetylene complexes with detailed structural and spectroscopic data are surprisingly scarce. For example, apart from the dinuclear and tetranuclear copper complexes 2 and Cu4(μ-[3,5-(CF3)2Pz])4(μ-HC≡CH)2 (3) reported by us recently [47], structurally authenticated copper-C2H2 complexes are limited to [Cu{NH(Py)2}(HC≡CH)]BF4 and [Cu(phen)(HC≡CH)]ClO4 with Cu(η2-HC≡CH) moieties [55,56,57] and polymeric or octanuclear, chloride bridged copper(I) adducts containing μ2−η2,η2-(HC≡CH) moieties [58,59,60].

2. Results and Discussion

The bis(pyrazolyl)borate copper(I) complex [H2B(3,5-(CF3)2Pz)2]Cu(NCMe)[42] reacts with purified acetylene (~1 atm) [61,62] in CH2Cl2, affording [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) as a white solid in >90% yield (Scheme 1), which is quite amenable to detailed spectroscopic and structural studies. The room temperature 1H NMR spectrum of 4 in CDCl3 displayed the acetylenic proton resonances at δ 4.70 ppm. That is a significant downfield shift relative to the corresponding signal of the free acetylene (δ 2.01 ppm) [63]. The 13C resonance of the acetylenic carbons appears at δ 80.2 ppm, which is a downfield shift of 7.0 ppm relative to that of the free acetylene (δ 73.2 ppm) [63]. The ῡC≡C band of solid 4 in the Raman spectrum was observed at 1819 cm−1, representing a 155 cm−1 red shift relative to the corresponding stretching frequency of the free HC≡CH (1974 cm−1) [64]. This red shift is not as high as that observed for Cu4(μ-[3,5-(CF3)2Pz])4(μ-HC≡CH)2 (3 with ῡC≡C of 1638 cm−1) containing a μ2−η2,η2-(HC≡CH) (which is a formally 4e-donor, bridging acetylene). Table 1 shows available, albeit limited, 1H and 13C NMR data and C≡C stretch of structurally characterized copper complexes featuring a formally 2e-donor η2-(HC≡CH). [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) shows the smallest downfield shift of the acetylenic NMR signal, and red-shift of C≡C stretching frequency relative to that of the free HC≡CH among these (although the differences are minor), suggesting relatively weaker σ/π-interaction between the copper(I) and acetylene ligand in terms of the Dewar-Chatt-Duncanson picture [65,66]. A rare, well-authenticated silver(I)-acetylene complex, [HB(3,5-(CF3)2Pz)3]Ag(HC≡CH) is also available for a comparison (although it has a tris(pyrazolyl)borate (not bis(pyrazolyl)borate) ligand support) [67]. It is a system that features relatively low M→alkyne backbonding. The acetylenic 1H signal of this silver(I) complex has been observed at δ 3.48 ppm, which is an even smaller downfield shift from the free acetylene resonance (δ 2.01 ppm), compared to that observed for [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4). Interestingly, alkyne resonance of [HB(3,5-(CF3)2Pz)3]Ag(HC≡CH) in 13C NMR appears at δ 66.3 ppm.

Scheme 1

Synthesis of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) and [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5) from [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1) and the corresponding alkyne.

Table 1

Selected bond distances (Å) and angles (º) for [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4), Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)2 (2), [Cu{NH(Py)2}(HC≡CH)]BF4 and [Cu(phen). (HC≡CH)]ClO4. Specific bond distances and angles of 4 are presented in CIF and Supporting Information section. Data collection temperatures are in Kelvin. Free acetylene C≡C distance, C≡C stretching frequency, 1H and 13C chemical shifts are 1.2033(2) Å (gas phase) and 1.193(6) Å (neutron diffraction—less accurate) [71,72], 1974 cm−1, 2.01 ppm, and 73.2 ppm [63], respectively.

Complex\Parameter	4	2	[Cu{NH(Py)2}(HC≡CH)]BF4	[Cu(phen)(HC≡CH)]ClO4
Temp.	100	100	173	283–303
Cu-C	1.972(3)1.973(3)	1.966(3)1.974(3)	1.971(3)1.971(3)	1.930(5)1.961(5)
C≡C	1.225(5)	1.227(4)	1.188(11)	1.190(7)
Cu-N	1.981(3)1.981(3)	1.9697(18)1.9742(18)	1.968(3)1.968(3)	1.979(4)1.978(4)
C-Cu-C	36.17(14)	36.29(11)	35.1(3)	35.6(2)
N-Cu-N	96.63(10)	98.94(8)	96.8(2)	84.9(2)
ῡ(C≡C)	1819	1811	1795	1800
1H	4.40	4.75	5.21	-
13C	80.2	-	-	-
ref	This work	[47]	[57]	[56]

Overall, acetylenic 1H and 13C resonance of d-block metal complexes bearing 2e-donor, η2-acetylene ligands has been observed (keeping in mind that η2-acetylene can also serve as a formally 4e-donor moiety) at quite a wide chemical shift range [68]. For example, NMR spectra of Ru(II) complex [Cp*Ru(HC≡CH)(PEt3)2][BPh4] [69] and Ni(0) complex (Ph3P)2Ni(HC≡CH) [70] display their signals for the metal bound C2H2 in 1H and 13C at δ 4.38, 66.14 ppm and δ 6.41, 122 ppm, respectively. The latter nickel complex is expected to display more pronounced M→alkyne backbonding than in 4. For comparison, protons of the bridging acetylene group of Cu4(μ-[3,5-(CF3)2Pz])4(μ-HC≡CH)2 (3) have been observed at δ 6.16 ppm [47].

The X-ray crystal structure of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) is illustrated in Figure 2. It is a three coordinate, trigonal planar copper-acetylene complex. The acetylene ligand is oriented parallel to the NCuN plane so as to maximize back-bonding interactions [73]. Selected bond distances and angles of 4 and copper complexes featuring 2e-donor, η2-acetylene ligand in the literature are given in Table 1. The key parameters involving the CuC2 core are remarkably similar between these molecules. This suggests that cationic copper species [Cu{NH(Py)2}(HC≡CH)]BF4 and [Cu(phen)(HC≡CH)]ClO4 featuring relatively electron-rich supporting ligands and neutral copper complexes 4 and 2 involving weakly donating fluorinated ligands have similar effects on the Cu-C2H2 alkyne moiety, or produce effects that are not large enough to be parsed out by routine X-ray crystallography. They both show slightly elongated C≡C bonds relative to the free acetylene (1.181(7) Å) [74], but these changes are overshadowed by the somewhat high esd associated with bond distance measurements.

Figure 2

Molecular structure of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4); ORTEP view with 50% probability ellipsoids are shown. Selected bond distances (Å) and angles (°): Cu-N2 1.981(3), Cu-N4 1.981(3), Cu-C11 1.972(3), Cu-C12 1.973(3), C11-C12 1.225(5), B-N1 1.570(4), B-N3 1.569(4), Cu···B 3.046, C11-Cu-C12 36.17(14), N2-Cu-N4 96.63(10), N4-Cu-C12 113.19(13), N2-Cu-C11 114.01(13), N1-B-N3 107.6(2).

In addition to acetylene, we also tested the use of HC≡CSiMe3 as a substrate in CuACC chemistry. Considering that structurally authenticated metal complexes of η2-HC≡CSiMe3 are rare (a search of Cambridge Structural Database [75] disclosed only three such examples involving transition metal ions) [76,77,78] and unknown for copper to our knowledge [75], we also synthesized [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5) for a detailed study. Treatment of [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) with HC≡CSiMe3 in CH2Cl2 led to 5 in 91% yield (Scheme 1). It is a white solid and was characterized by NMR and Raman spectroscopy and X-ray crystallography. The ῡC≡C band of solid 5 in the Raman spectrum was observed at 1870 cm−1, which is a 237 cm−1 red shift relative to the corresponding stretching frequency of the free HC≡CSiMe3 (2107 cm−1). This ῡ(C≡C) is similar to that reported for [HC{C(CF3)CO}2]Cu(HC≡CSiMe3) [41]. This suggests the presence of an η2-HC≡CSiMe3 bound alkyne moiety on copper(I) [46,51].

X-ray crystal structure of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5) is depicted in Figure 3. It is a monomeric, trigonal planar copper complex with an η2-HC≡CSiMe3 bound alkyne moiety. The HC≡CSiMe3 is bonded slightly asymmetrically as evident from the marginally longer Cu-C12, which is a carbon atom with the larger, silyl group. The alkyne group shows a significant deviation from the ideal 180° as evident from C≡C-Si angle, 160.64(11)°. This is about 19° bending back of the alkyne group due to the metal ion coordination. As noted above, there are no structural data on related copper η2-HC≡CSiMe3 complexes for comparisons. The Cp2Nb(H)(HC≡CSiMe3) [76] and (NMe-Paa)W(CO)F(HC≡CSiMe3) (based on a κ-[C,N,N′] chelator NMe-Paa = 2-(2-dimethylaminoethyl)methylaminomethylphenyl) [78] complexes are known and display significantly smaller C≡C-Si angles of 141.7(5)° and 138.1(1)° (or much larger deviation from linearity), which points to stronger metal-alkyne σ/π-bonding in these W(II) and Nb(III) complexes, in terms of the Dewar-Chatt-Duncanson model [65,66].

Figure 3

Molecular structure of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5); ORTEP view with 50% probability ellipsoids are shown. Selected bond distances (Å) and angles (°): Cu-N2 1.9857(9), Cu-N4 1.9845(9), Cu-C11 1.9600(12), Cu-C12 1.9957(11), C11-C12 1.2343(17), B-N1 1.5778(15), B-N3 1.5737(16), C12-Si 1.8713(12), Cu···B 2.964, C11-Cu-C12 36.35(5), N2-Cu-N4 90.59(4), N4-Cu-C12 120.05(4), N2-Cu-C11 112.92(5), N1-B-N3 106.41(9), C11-C12-Si 160.64(11).

Compounds [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4) and Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)2 (2) (note: non-brominated, Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 analog has been observed but not isolated) allow us to compare the effects of replacing a [H2B]+ with a [(alkyne)Cu]+ moiety in these systems. Although structural features of N2CuC2 core are very comparable (Table 1), 1H and Raman spectroscopic data of the alkyne group suggest that the copper site in the mononuclear 4 is slightly more Lewis acidic than that of dinuclear 5, despite having a 4-bromo pyrazolate in the latter. Table S1 (Supplementary Materials) in supporting information shows two additional sets of molecules, more closely related to each other; [H2B(3,5-(CF3)2Pz)2]Cu(EtC≡CEt) (6) [6] and Cu2(μ-[3,5-(CF3)2Pz])2(EtC≡CEt)2 (7)[46] as well as [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CPh) (8) [42] and Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CPh)2 (9) [47]. To facilitate this analysis, the X-ray crystal structure of [H2B(3,5-(CF3)2Pz)2]Cu(EtC≡CEt) was also investigated and the details are included in the Supporting Information [6]. A comparison of mononuclear bis(pyrazolyl)borate systems to the corresponding dinuclear pyrazolate systems show very similar metrical parameters involving the N2CuC2 cores, except for N-Cu-N angles, which are smaller for the bis(pyrazolyl)borate copper complexes. Closer analysis indicates that this is not because of noticeably longer Cu-N distances in bis(pyrazolyl)borate copper systems but due to their more folded CuN4B cores (compared to the flatter CuN4C rings in the related dinuclear pyrazolates). This is evident from the separation of the pyrazolyl ring carbons at 4-positions (e.g., Pz-C4···Pz-C4 distance of [H2B(3,5-(CF3)2Pz)2]Cu(EtC≡CEt) (6) and Cu2(μ-[3,5-(CF3)2Pz])2(EtC≡CEt)2 (7) is 6.18 and 6.38 Å).

Further analysis of the alkyne-copper(I) interaction is performed using density functional calculations (Table 2) to understand the variations between the mononuclear and dinuclear species as well as the copper and different alkynes. The overall interaction energy (ΔEint) between the alkyne and copper(I) center for mononuclear species [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4), [H2B(3,5-(CF3)2Pz)2]Cu(EtC≡CEt) (6), and [H2B(3,5-(CF3)2Pz)2]Cu(PhC≡CH) (8) is −46.8, −51.5, and −49.4 kcal⸱mol−1, respectively, which is further dissected in different contributions within the Ziegler-Rauk energy decomposition analysis (EDA) [79,80]. In this framework, the interaction energy (ΔEint) exhibits a larger electrostatic character (ΔEelstat) of about ~60% of the stabilizing terms, whereas the orbital contribution to the interaction is about ~35% (ΔEorb), with the remaining ~5% attributable to dispersion-type contributions (ΔEdisp). The ΔEorb involves both π-backdonation and σ-donation, which contributes 57.2% and 26.3% to the bonding stabilization of 4. The π-backbonding contribution in EtC≡CEt and PhC≡CH counterparts is similar to that of 4 (Figure 4). For the dinuclear species, Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 (10), Cu2(μ-[3,5-(CF3)2Pz])2(EtC≡CEt)2 (7), and Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CPh)2 (9), the Cu-alkyne interaction energy (ΔEint) amounts to −43.8, −49.4 and −47.8 kcal⸱mol−1, respectively. These values are slightly lower than the somewhat related mononuclear, bis(pyrazolyl)borate species. They, however, involve similar bonding characteristics. The structurally characterized, brominated species Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)2 (2) with an Cu-alkyne interaction energy of −42.9 kcal⸱mol−1, shows a slight destabilization in comparison to Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 (10).

Table 2

Energy decomposition analysis of the interaction energy in mononuclear and dinuclear Cu(I) species. Values in kcal·mol−1. Vibrational frequencies in cm−1.

Mononuclear, Bis(pyrazolyl)Borate Copper Complexes, L = [H2B(3,5-(CF3)2Pz)2]; L’ = [H2B(4-Br-3,5-(CF3)2Pz)2]
	LCu(HC≡CH) 4		LCu(EtC≡CEt) 6		LCu(PhC≡CH) 8		LCu(HC≡CSiMe3) 5		L’Cu(HC≡CH)
ΔEPauli	121.8		131.0		124.8		123.6		119.9
ΔEElstat	−100.4	59.5%	−112.6	61.7%	−102.3	58.8%	−103.7	59.0%	−99.1	59.5%
ΔEorb	−62.2	36.9%	−59.9	32.8%	−61.8	35.5%	−60.3	34.3%	−61.5	36.9%
ΔEDisp	−6.0	3.6%	−10.0	5.5%	−10.0	5.7%	−11.7	6.7%	−6.0	3.6%
ΔEint	−46.8		−51.5		−49.4		−52.1		−46.7
π→Cu	−35.6	57.2%	−31.0	51.8%	−33.8	54.6%	−31.5	52.2%	−34.8	56.5%
σ←Cu	−16.3	26.3%	−17.5	29.2%	−16.6	26.9%	−16.7	27.7%	−16.5	26.8%
vC≡C, Calc.(Exp.)	1817(1819)		2056(2064)		1954(1927)		1870(1870)		1818
Dinuclear , Copper Pyrazolate Complexes
	Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)210		Cu2(μ-[3,5-(CF3)2Pz])2(EtC≡CEt)27		Cu2(μ-[3,5-(CF3)2Pz])2(PhC≡CH)29			Cu2(μ-[4-Br-3,5-(CF3)2Pz])2(HC≡CH)22
ΔEPauli	128.0		143.2		142.0			124.6
ΔEElstat	−102.7	59.8%	−118.2	61.3%	−109.8	57.8%		−99.9	59.6%
ΔEorb	−62.8	36.5%	−62.2	32.3%	−65.6	34.5%		−61.5	36.7%
ΔEDisp	−6.3	3.7%	−12.2	6.3%	−14.5	7.6%		−6.1	3.7%
ΔEint	−43.8		−49.4		−47.8			−42.9
π→Cu	−35.8	57.0%	−32.3	52.0%	−35.7	54.5%		−34.8	56.6%
σ←Cu	−16.7	26.5%	−17.9	28.7%	−17.0	25.9%		−16.6	27.1%
vC≡C, Calc.(Exp.)	1814		2055(2033, 2066)		1950			1813(1811)

Figure 4

Representative deformation densities accounting for the π-backbonding (left) and σ-donation (right) contribution to the bonding scheme in the formation of alkyl copper complexes (from top to bottom) for (a) [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4), (b) [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5) and (c) the dinuclear Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 (10). Charge flow from red to blue.

We also probed the effect of different substituents on the alkyne moieties by including a phenyl and silyl group –SiMe3 (Table 2). In [H2B(3,5-(CF3)2Pz)2]Cu(PhC≡CH) (8) and [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5), the Cu-alkyne stabilization increases slightly due to a small increase of the electrostatic character in comparison to the HC≡CH counterpart (4). Moreover, if the silicon atom in the silyl derivative is replaced by a carbon atom, that is—CMe3, the stabilization is further improved with an increase in electrostatic contribution. The computational analysis of the Ni(0) complex (Ph3P)2Ni(HC≡CH)[70] was also performed for a comparison (ESI). As expected, it shows a significantly more stabilized Ni-(HC≡CH) bond (ΔEint = −75.9 kcal⸱mol−1) owing to the increase of both electrostatic and orbital stabilizations, as given by a pronounced M→alkyne backbonding, which amounts to −77.6 versus −35.6 kcal⸱mol−1 in 4, which is consistent with the trends of red-shift of the C≡C stretching frequency (or the weakening of the C≡C bond).

Considering the importance of and interest in copper catalyzed azide-alkyne cycloaddition (CuAAC), as well as the rich alkyne chemistry of “[H2B(3,5-(CF3)2Pz)2]Cu” moiety, we also set out to explore the use of fluorinated bis(pyrazolyl)borate copper(I) complex [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) as a catalyst in cycloaddition of organic azides with terminal alkynes. We also compare the effectiveness of [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) with the trinuclear copper(I) pyrazolate {μ-[3,5-(CF3)2Pz]Cu}3 [50], which was found to be quite an effective catalyst for these reactions [47,81]. The Huisgen 1,3-dipolar cycloaddition [82] of organic azides and alkyne received significant attention ever since Sharpless [2] and Meldal [1] independently developed a copper(I) catalyzed reaction. Multidentate nitrogen ligands such as poly(pyrazolyl)borate ligands are known for stabilizing different metal ions including Cu(I) [83,84]. Ruthenium(II) tris(pyrazolyl)borates are reported to catalyze cycloaddition of organic azides and alkyne [85]. There are however very few reports on copper(I) tris(pyrazolyl)borate catalyzed azide-alkyne cycloaddition even though they are known to catalyze several reactions such as cyclopropanation, cyclopropenation, nitrene transfer reactions, etc. [86]. One of the reports concerns the synthesis of N-sulfonyl-1,2,3-triazoles from N-sulfonyl azides and alkynes [87]. That report shows copper(I) supported by tris(pyrazolyl)methanes providing better yields than the tris(pyrazolyl)borate based catalysts. More recent work by Stiriba and co-workers described the use of bis- and tris(pyrazolyl)boratocopper(I) systems to mediate reactions between phenyl- and alkyl-azides with different alkynes to produce 1,4-disubstituted 1,2,3-triazole derivatives moderate to excellent yields [88].

Table 3 summarizes the results of azide-alkyne cycloaddition mediated by the mononuclear [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1) and the trinuclear copper(I) pyrazolate {μ-[3,5-(CF3)2Pz]Cu}3. A reaction of p-tolylazide with different terminal alkynes using 1 mol% of catalyst 1 at 110 °C in toluene gives a very high yield for all substrates. The control reaction of p-tolyl azide with phenylacetylene without catalyst at the same temperature gives a lower (64%) yield. Thus, the temperature alone drives some of these processes albeit less effectively. We then tested the chemistry using a lower temperature but using a higher catalyst load. The use of 10 mol% compound 1 in EtOH at 40 °C gives triazoles in moderate to excellent yield including a rare CuACC reaction involving the acetylene gas. The control reaction without the catalyst for these conditions gives <1% triazole (reaction of phenylacetylene and p-tolylazide). We compared these results to the corresponding {μ-[3,5-(CF3)2Pz]Cu}3 mediated chemistry using 1 mol% catalyst, at room temperature in dichloromethane. This trinuclear catalyst catalyzed alkyne-azide cycloadditions very effectively under milder conditions and generated products in high yield for all substrates except for trimethylsilylacetylene. It is noteworthy that all reactions (except 110 °C reactions) are done in a vial using normal solvents and without using any inert atmosphere. Although we have not probed the mechanistic details, recent work by Larinov and coworkers on {μ-[3,5-(CF3)2Pz]Cu}3 and Stiriba and co-workers on poly(pyrazolyl)borate copper catalyzed cycloadditions suggest different mechanisms, most notably involving tetranuclear and dinuclear intermediates, respectively, for the two systems [88,89]. Better activity of the trinuclear {μ-[3,5-(CF3)2Pz]Cu}3 over mononuclear [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) in copper catalyzed azide-alkyne cycloaddition suggests that these reactions involve multicenter catalytic intermediates, consistent with proposed mechanisms [31] and having pre-assembled copper sites with bridging ligands is an advantage [49].

Table 3

Azide-alkyne cycloaddition mediated by the mononuclear bis(pyrazolyl)borate [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1) and the trinuclear copper(I) pyrazolate {μ-[3,5-(CF3)2Pz]Cu}3. Different Methods are described in the experimental section.

Entry	Alkyne	Yield (%)[H2B(3,5-(CF3)2Pz)2]Cu(NCMe)catalyst		Yield (%) {μ-[3,5-(CF3)2Pz]Cu}3 CatalystMethod III
		Method I	Method II
1	HC≡CH	-	80	99
2	n-PrC≡CH	99	85	99
3	n-BuC≡CH	99	89	99
4	n-C8H17C≡CH	99	91	99
5	PhC≡CH	99	84	99
6	Me3SiC≡CH	99	56	10

In summary, we report the isolation and characterization of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) and [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) supported by a highly fluorinated bis(pyrazolyl)borate ligand. They feature three-coordinate, trigonal planar copper sites in the solid state and exhibit a significant reduction in ῡC≡C value relative to the corresponding free alkyne C≡C stretch. Computational analysis of these molecules and several other related compounds using DFT indicates that the Cu(I)-alkyne interaction in these copper complexes is primarily of electrostatic character. Furthermore, despite the presence of a highly fluorinated ligand in [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) and [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3), the Cu→alkyne π-backbonding component is much larger than the alkyne→Cu σ-bonding interaction. However, the backbonding is not as high as that computed for (Ph3P)2Ni(HC≡CH). The mononuclear and dinuclear complexes such as [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) and Cu2(μ-[3,5-(CF3)2Pz])2(HC≡CH)2 display similar Cu-alkyne bonding features. The bis(pyrazolyl)borate complex [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) catalyzes [3+2] cycloaddition chemistry between tolyl azide and a variety of alkynes including acetylene to produce 1,2,3-triazoles. It is, however, comparatively less effective than the related trinuclear copper catalyst {μ-[3,5-(CF3)2Pz]Cu}3 involving bridging pyrazolates. We are presently exploring the metal mediated alkene and alkyne chemistry supported by these and other fluorinated ligands [90,91].

3. Experimental Details

All manipulations except catalysis were carried out under an atmosphere of purified nitrogen using standard Schlenk techniques or in a MBraun glovebox equipped with a −25 °C refrigerator. Solvents were purchased from commercial sources, purified prior to use. NMR spectra were recorded at 25 °C on a JEOL Eclipse 500 spectrometer (Peabody, MA, USA) (1H, 500.16 MHz; 13C, 125.78 MHz; 19F, 470.62 MHz) unless otherwise noted. Proton and carbon chemical shifts are reported in ppm versus Me4Si. 1H NMR coupling constants (J) are reported in Hertz (Hz) and multiplicities are indicated as follows: s (singlet), d (doublet), t (triplet), m (multiplet). 19F NMR values were referenced to external CFCl3. Melting points were obtained on a Mel-Temp II apparatus (Wayne, PA, USA) and were not corrected. Elemental analyses were performed using a Perkin–Elmer Model 2400 CHN analyzer (Waltham, MA, USA). IR spectra were collected at room temperature on a Shimadzu IR Prestige-21 FTIR (Kyoto, Japan) containing an ATR attachment using pure liquid or solid materials, with instrument resolution at 2 cm−1. Raman data were collected on a Horiba Jobin Yvon LabRAM Aramin Raman spectrometer (Edison, NJ, USA) with a HeNe laser source of 633 nm, by placing pure liquid or solid materials on a glass slide/cuvette. Heating was accomplished by either a heating mantle or a silicone oil bath. Purification of reaction products was carried out by flash column chromatography using silica gel 60 (230–400 mesh). TLC visualization was accompanied by UV light or KMnO4 stains. The [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1)[42] and {μ-[3,5-(CF3)2Pz]Cu}3 [50] were prepared according to reported literature procedures. p-Tolyl azide was prepared according to the literature procedure [92]. All other reactants and reagents were purchased from commercial sources. Acetylene gas was freed from acetone and purified before use [61]. All other reactants and reagents were purchased from commercial sources.

Warning. Due care must be taken when working with acetylene gas. It is known to produce explosive combinations with oxygen, and also form potentially explosive acetylides and other materials with copper salts [62].

[H≡CH) ( [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (0.15 g, 0.29 mmol) was dissolved in 7 mL dichloromethane and stirred for ~10 min while bubbling acetylene. The reaction mixture was concentrated with continuous flow of acetylene and kept at −20 °C to obtain X-ray quality colorless crystals of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH). Yield: >90%. M.P.: 78–79 °C. Anal. Calc. C12H6BCuF12N4: C, 28.34%; H, 1.19%; N, 11.02%. Found: C, 28.18%; H, 1.17%; N, 11.04%. 1H NMR (CDCl3): δ (ppm) 6.93 (s, 2H, PzH), 4.70 (s, 2H, ≡CH), 4.00 (br, 2H, BH). 19F NMR (CDCl3): δ (ppm) −59.8 (s), −60.7 (s). 13C{1H} NMR (CDCl3): δ (ppm) 142.2 (q, 2JC-F = 39.6 Hz, C-3/C-5), 139.9 (q, 2JC-F = 41.2 Hz, C-3/C-5), 120.1 (q, 1JC-F =268.7 Hz, CF3), 119.2 (q, 1JC-F =271.1 Hz, CF3), 106.4 (C-4), 80.2 (C≡C). 13C (1H coupled) NMR (CDCl3): δ (ppm) 142.2 (q, 2JC-F = 38.4 Hz, C-3/C-5), 139.9 (q, 2JC-F = 40.8 Hz, C-3/C-5), 120.1 (q, 1JC-F =268.7 Hz, CF3), 119.2 (q, 1JC-F =271.1 Hz, CF3), 106.4 (d, 1JC-H = 187.1 Hz, C-4), 80.3 (dd, 1JC-H = 251.9 Hz, 2JC-H = 43.2 Hz, C≡C). Raman (cm−1), selected peak: 1819 (C≡C).

[H≡CSiMe [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (0.15 g, 0.29 mmol) was dissolved in 7 mL dichloromethane. Trimethylsilylacetylene (48 μL, 0.35 mmol) was added to the reaction mixture and stirred overnight. The solvent was evaporated to get white powder product. The product was dissolved in dichloromethane and kept at −20 °C to obtain X-ray quality colorless crystals of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3). Yield: 91%. M.P.: 59–61 °C. Anal. Calc. C15H14BCuF12N4Si: C, 31.02%; H, 2.43%; N, 9.65%. Found: C, 30.64%; H, 2.07%; N, 9.98%. 1H NMR (CDCl3): δ (ppm) 6.88 (s, 2H, PzH), 4.81 (s, 1H, ≡CH), 3.92 (br, 2H, BH), 0.24 (s, 9H, Si(CH3)3). 19F NMR (CDCl3): δ (ppm) −59.9 (s), −61.0 (s). 13C{1H} NMR (CDCl3): δ (ppm) 141.8 (q, 2JC-F = 38.4 Hz, C-3/C-5), 139.8 (q, 2JC-F = 43.2 Hz, C-3/C-5), 120.2 (q, 1JC-F =269.9 Hz, CF3), 119.3 (q, 1JC-F =269.9 Hz, CF3), 106.2 (C-4), 97.8 (C≡C), 97.2 (C≡C), −0.2 (SiCH3). Raman (cm−1), selected peak: 1870 (C≡C).

[H≡CEt) ( This was synthesized as reported earlier[6] and crystallized using dichloromethane at −20 °C to obtain crystals suitable for X-ray analysis.

Details of CuACC chemistry involving several alkynes and p-tolyl azide

General method I for the synthesis of triazoles:

A 50 mL Schlenk flask was charged with the selected alkyne (1.0. mmol), [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (1 mol%) and toluene (5.0 mL) under a nitrogen atmosphere. p-tolyl azide (1.0 mmol) was added to the reaction and stirred at 110 °C for 12 h. The solvent was removed under reduced pressure. The residue was dissolved in dichloromethane and filtered through a celite. The dichloromethane was evaporated to get pure product.

General method II for the synthesis of triazoles:

A round bottom flask was charged with the selected alkyne (1.0. mmol), p-tolyl azide (1.0 mmol), and EtOH (5.0 mL). [H2B(3,5-(CF3)2Pz)2]Cu(NCMe) (10 mol%) was added to the reaction and stirred at 40 °C for 12 h. The solvent was evaporated. The product was isolated using column chromatography (Ethyl acetate/hexanes).

General method III for the synthesis of triazoles

A 5 mL vial was charged with selected alkyne (1.0. mmol), p-tolyl azide (1.0 mmol), and dichloromethane (3.0 mL). {μ-[3,5-(CF3)2Pz]Cu}3 (1 mol%) was added to the reaction and stirred at room temperature for 12 h. The yield was calculated using 1,3,5-(trimethoxy)benzene as internal standard.

4. X-ray Data Collection and Structure Determinations

A suitable crystal covered with a layer of hydrocarbon/Paratone-N oil was selected and mounted on a Cryo-loop and immediately placed in the low temperature nitrogen stream. The X-ray intensity data were measured at 100(2) K on a Bruker D8 Quest with a Photon 100 CMOS detector equipped with an Oxford Cryosystems (Billerica, MA, USA) 700 series cooler, a graphite monochromator, and a Mo Kα fine-focus sealed tube (λ = 0.71073 Å). Intensity data were processed using the Bruker Apex program suite. Absorption corrections were applied by using SADABS [93]. Initial atomic positions were located by SHELXT [94] and the structures of the compounds were refined by the least-squares method using SHELXL [95] within Olex2 GUI [96]. All the non-hydrogen atoms were refined anisotropically. The hydrogen atoms of BH2 moieties as well as acetylenic C≡CH were located in difference Fourier maps, included and refined freely with isotropic displacement parameters. The remaining hydrogen atoms were included in their calculated positions and refined as riding on the atoms to which they are joined. X-ray structural figures were generated using Olex2 [96]. The CCDC 2065196-2065198 files contain the supplementary crystallographic data of [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CH) (4), [H2B(3,5-(CF3)2Pz)2]Cu(HC≡CSiMe3) (5) and [H2B(3,5-(CF3)2Pz)2]Cu(EtC≡CEt) (6) [6]. These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html (accessed on 7 December 2021) or from the Cambridge Crystallographic Data Centre (CCDC), 12 Union Road, Cambridge, CB2 1EZ, UK). Additional details are provided in the Supporting Information section.

5. Theoretical Methods

Computational details are given as Supporting information.