Open Access Article
Alessandro
Ponti
a,
Alessandra
Silvani
b and
Giorgio
Molteni
*b
aIstituto di Scienze e Tecnologie Chimiche “Giulio Natta” (SCITEC), Consiglio Nazionale delle Ricerche, via G. Fantoli 16/15, Milan 20138, Italy
bUniversità degli Studi di Milano, Dipartimento di Chimica, via C. Golgi 19, Milan 20133, Italy. E-mail: giorgio.molteni@unimi.it
First published on 27th July 2023
Deuterium incorporation in the 5-substituted pyrazoles arising from the copper(I)-catalysed reaction between chlorohydrazones and terminal acetylenes suggests the intermediacy of copper(I)-complexed alkynylhydrazones. Since the efficiency of such a complexation depends on both the chlorohydrazone and the solvent, the obtainment of the pyrazoles and/or the corresponding alkynylhydrazones is variable depending on the reaction conditions. Copper(I)-complexed alkynylhydrazone intermediates should play a pivotal role in the proposed catalytic cycle.
Access to the pyrazole ring can be pursued via 1,3-dipolar cycloaddition of the nitrilimine intermediate on the carbon–carbon triple bond. Unfortunately, the cycloaddition between nitrilimines and unsymmetrically substituted alkynes very often yields mixtures of regioisomeric pyrazoles. The poor regioselectivity of the reaction applies to both the classical thermal cycloadditions according to Huisgen3 and the more recently introduced metal-catalysed cycloadditions.4
The regioselective synthesis of the pyrazole ring from chlorohydrazones in which the formation of the nitrilimine intermediate is avoided was first performed by us5 and disclosed the copper(I)-catalysed reaction between chlorohydrazones and terminal alkynes as a fruitful, regioselective route to 5-substituted pyrazoles.
The present paper is focused on the mechanistic aspects of such a reaction that can be deduced from the behaviour of chlorohydrazones 1a–c and the deuterated one D-1a towards terminal acetylenes 2 and deutero-phenylacetylene D-2a (Fig. 1).
Alongside the synthetic aspects just mentioned, the extent of deuterium incorporation in pyrazole 3aa was determined. For this purpose, deuterated chlorohydrazone D-1a and deutero-phenylacetylene D-2a were prepared. Both deuterated reagents were obtained in an analytically pure form; the former by exchange with D2O of the hydrazonic proton of 1a, the latter by treatment of phenylacetylene with n-butyllithium, followed by addition of D2O (Scheme 1).
The copper(I)-catalysed reaction between deutero-phenylacetylene D-2a and chlorohydrazone 1a was pursued in CCl4 to eliminate undesirable source of hydrogens due to the solvent that could weaken the connection between the results and the reaction mechanism. A mixture of the two pyrazoles 3aa and D-3aa that could not be separated by chromatographic methods was obtained, which showed 30% deuterium incorporation at 1H NMR (Scheme 2, eqn (1)). The value of deuterium incorporation increases to 65% by reacting deuterated chlorohydrazone D-1a with phenylacetylene (Scheme 2, eqn (2)), and in the reaction between the two deuterated substrates the incorporation reaches 90% (Scheme 2, eqn (3)). The latter value is consistent with similar deuteration experiments. In fact, quantitative deuterium incorporation is quite hard to achieve because of the incomplete purity of all species in the reaction mixture.7
The above results are consistent with the catalytic cycle depicted in Fig. 2; for the sake of clarity the behaviour of chlorohydrazone 1a is shown.
Consistent with typical azide “click” reactions involving terminal acetylenes in the presence of copper(I),8 the first step of the catalytic cycle involves the formation of a copper(I) acetylide, followed by nucleophilic addition to the C
N double bond of chlorohydrazone. The generation of the key intermediate A is fully plausible due to the known propensity of some hydrazones to complexate the copper cation.9 This intermediate accounts for the greater degree of deuterium incorporation into the pyrazole adduct from deuterated hydrazonoyl chloride D-1a (65%) than from deutero-phenylacetylene D-2a (30%), which is clearly dictated by the proximity of the deuterium atom to the 4-position of the pyrazole ring upon its closure. In the case of D-2a, one out of three pyrazole molecules must receive the deuterium atom from triethyldeuterammonium chloride, which is present in the reaction medium as it is released in the first step of the catalytic cycle. The same proportion of hydrogen transfer from ammonium cations is observed in the reaction conducted on deuterated chlorohydrazone D-1a; in this case, one in three pyrazole molecules must receive the 1H atom from triethylammonium chloride. As for the pyrazolium metallated cation B, its intervention has been postulated for the final step of the azide–alkyne “click” cycloaddition8 and the cycloaddition of alkynylhydrazones promoted by Au(I) salts.10
It was not possible to isolate the key intermediate A nor the corresponding alkynylhydrazone 4aa (vide infra) in the reaction between 1a and phenylacetylene. In this regard, it is useful to point out that the attempt to obtain intermediate A by crystallisation of 4aa in the presence of an equimolecular amount of CuCl in dichloromethane yielded the pyrazole 3aa. These results may appear surprising on the basis of a paper published in 2014 in which the facile isolation of alkynylhydrazones from the reaction between C,N-diphenyl chlorohydrazone 1c and terminal alkynes in dimethylformamide is described.11
The contradiction is only apparent as can be seen by comparing the behaviour of chlorohydrazones 1a and 1b, c (Fig. 1) towards phenylacetylene in the presence of catalytic amounts of copper(I) ions.
While the reaction of C-methoxycarbonyl-substituted chlorohydrazone 1a with phenylacetylene in dichloromethane selectively gave the pyrazole 3aa (Table 1, entry 1), mixtures of 3aa and the corresponding alkynylhydrazone 4aa were obtained by using acetonitrile or dimethylformamide as the solvent (Table 2, entries 1 and 2).
| Entry | R1 | Solvent | Time (min) | Products | Ratio 3 : 4 |
Yield (%) 3 + 4 |
|---|---|---|---|---|---|---|
| a Literature datum: 91%.11 | ||||||
| 1 | COOMe | MeCN | 45 | 3aa, 4aa | 86 : 14 |
81 |
| 2 | COOMe | DMF | 45 | 3aa, 4aa | 78 : 22 |
78 |
| 3 | 2-MeO-C6H4 | CH2Cl2 | 60 | 3ba, 4ba | 91 : 9 |
70 |
| 4 | 2-MeO-C6H4 | MeCN | 85 | 3ba, 4ba | 68 : 32 |
66 |
| 5 | 2-MeO-C6H4 | DMF | 120 | 3ba, 4ba | 60 : 40 |
74 |
| 6 | Ph | CH2Cl2 | 60 | 3ca, 4ca | 45 : 55 |
78 |
| 7 | Ph | MeCN | 80 | 3ca, 4ca | 15 : 85 |
80 |
| 8 | Ph | DMF | 120 | 3ca, 4ca | 0 : 100 |
84a |
C-Aryl-substituted chlorohydrazones 1b,c also gave product mixtures (Table 2, entries 3–7) except C-phenyl-substituted 1c in dimethylformamide, which selectively gave the alkynylhydrazone 4ca (Table 2, entry 8). In this latter case the literature datum11 was reproduced in the presence of CuI as the catalyst (see Experimental).
These apparently conflicting results can be rationalised in light of the different complexation extent of copper(I) halides with alkynylhydrazones 4 as a function of solvent. For ease of reading, the three possible intermediates A are illustrated for each hydrazonoyl chloride in Fig. 3, and it can be perceived that the tenacity of the complexation decreases in the order A-1a > A-1b > A-1c, while remaining in the range of a labile complexation.
![]() | ||
| Fig. 3 Key intermediates A for the catalytic cycle depicted in Fig. 2. | ||
In pathway a of the proposed catalytic cycle (Fig. 2), the copper(I) moiety must move to the carbon at 4- position of the pyrazolium metallated cation B. It is likely that this latter intermediate is more easily formed the more stable the corresponding intermediate A is, thus explaining the experimental results shown in Table 1, entry 1 and Table 2, entries 3 and 6.
Interestingly, the experimental outcome markedly depends on the reaction solvent as can be seen by comparing data of Table 1, entry 1, and Table 2. We interpret this dependence with reference to the complexing ability of organic solvents. This is usually expressed by their “donor number” (DN),12 and decreases in the order: DMF (26.6) > MeCN (14.1) CH2Cl2 (0).13
It is plausible that the better complexation of copper(I) chloride by dimethylformamide compared to acetonitrile, and especially dichloromethane, is responsible for the lack of or reduced complexation of the alkynylhydrazone by the copper(I) halide, which necessarily results in greater difficulty in cyclization to the metallated pyrazolium cation B, and thus to the pyrazole product 3. Such competition between solvent and alkynylhydrazone favours the latter over the cyclic product when the solvent has a strong complexing ability towards copper(I) (Fig. 2, pathway b). As can be inferred from the product ratios in entries 1, 4, 7 and 2, 5, 8 of Table 2, the alkynylhydrazone-copper(I) complexation efficiency in acetonitrile and dimethylformamide, respectively, should also decrease in the order A-1a > A-1b > A-1c. In the latter case it must be practically non-existent, effectively preventing the alkynylhydrazone → pyrazole cyclisation, in perfect agreement with the literature data.11 In other words, it is as if the high complexing power of dimethylformamide versus copper(I) halogenide prevented the in situ formation of complex A-1c.
Deutero-phenylacetylene D-2a14 and chlorohydrazones 1a–c,15–17 were prepared according to literature procedures.
1,3,5,-Substituted pyrazoles 3aa,183ba,193ca3a and alkynylhydrazone 4ca11 are known in the literature.
In the case of 1-butyn-2-one 2b, Table 1, entry 2, the crude was filtered over a silica gel pad and the solvent was evaporated under reduced pressure. Crystallisation of the residue with iPr2O gave pure 1-phenyl-3-methoxycarbonyl-5-acetylpyrazole 3ab (0.53 g, 92%) as white powder having mp 109–111 °C; IR (nujol): 1735 (ester
C
O), 1710 (ketone
C
O) (cm-1); 1H-NMR: 7.50 (s, 1H, pyrazole-H4), 7.35–7.46 (m, 5H, aromatics), 3.96 (s, 3H, –COOCH3), 2.51 (s, 3H, –COCH3); 13C-NMR: 187.2 (s, –COCH3), 161.9 (s, –COOCH3), 143.3 (s, pyrazole-C5), 140.9 (s, pyrazole-C3), 140.0 (s, aromatic), 129.3 (d, aromatic), 128.7 (d, aromatic), 126.0 (d, aromatic), 114.8 (d, pyrazole-C4), 52.4 (q, –COOCH3), 28.7 (q, –COCH3). MS: 244 m/z (M+). Anal. calcd for C13H12N2O3: C, 63.93; H, 4.95; N, 11.47. Found: C, 63.89; H, 4.90; N, 11.54.
In the case of cyclopentylacetylene 2c (Table 1, entry 3) and 3-phenoxyprop-1-yne 2d (Table 1, entry 4) the residue was chromatographed on a silica gel column with hexane/EtOAc 1
:
2. First fractions contained starting 1a, further elution followed by crystallisation with iPr2O gave pure 3.
C
O) (cm-1); 1H-NMR: 7.41–7.47 (m, 5H, aromatics), 6.76 (s, 1H, pyrazole-H4), 3.91 (s, 3H, –COOCH3), 2.98–3.02 (m, 1H, cyclopentyl-CH), 1.66–1.80 (m, 2H, cyclopentyl –CH2CH
), 1.66–1.79 (m, 2H, cyclopentyl –CH2CH
), 1.49–1.65 (m, 4H, cyclopentyl –CH2–CH2–); 13C-NMR: 163.0 (s, –COOCH3), 150.7 (s, aromatic), 143.3 (s, pyrazole-C3), 139.3 (s, pyrazole-C5), 129.0 (d, aromatic), 128.8 (d, aromatic), 126.2 (d, aromatic), 105.7 (d, pyrazole-C4), 51.8 (q, –COOCH3), 36.2 (d, –CH2CH
), 33.6 (t, –CH2CH
), 25.1 (t, –CH2–CH2–). MS: 270 m/z (M+). Anal. calcd for C16H18N2O2: C, 71.09; H, 6.71; N, 10.36. Found: C, 71.13; H, 6.68; N, 10.41.
O) (cm−1); 1H-NMR: 7.08–7.47 (m, 10H, aromatics), 6.72 (s, 1H, pyrazole-H4), 4.00 (s, 2H, –CH2Ph), 3.94 (s, 3H, –COOCH3); 13C-NMR: 163.0 (s, –COOCH3), 144.2 (s, aromatic), 143.6 (s, pyrazole-C3), 139.0 (s, aromatic), 137.2 (s, pyrazole-C5), 126.0–129.2 (m, aromatic
C–H), 109.6 (d, pyrazole-C4), 52.1 (q, –COOCH3), 32.4 (t, –CH2Ph). MS: 292 m/z (M+). Anal. calcd for C18H16N2O2: C, 73.95; H, 5.52; N, 9.58. Found: C, 74.03; H, 5.47; N, 9.64.
N–D), 1700 (ester
C
O) (cm−1); 1H-NMR: 7.04–7.39 (m, 5H, aromatics), 3.95 (s, 3H, –COOCH3), 13C-NMR: 160.4 (s, –COOCH3), 141.5 (s,
C
N–), 130.3 (s, aromatic), 129.5 (d, aromatic), 123.3 (d, aromatic), 114.6 (d, aromatic), 53.6 (q, –COOCH3). MS: 213 m/z (M+), exact mass 213.0411. Anal. calcd for C8H8DClN2O2: C, 50.60; H, 4.72; N, 13.11. Found: C, 50.58; H, 4.70; N, 13.18.
In the case of the reaction between 1a and D-2a, 355 mg (85% combined yield) were obtained in the ratio D-3aa
:
3aa = 30
:
70.
In the case of the reaction between D-1a and 2a, 364 mg (87% combined yield) were obtained in the ratioD-3aa
:
3aa = 65
:
35.
In the case of the reaction between D-1a and D-2a, 368 mg (88% combined yield) were obtained in the ratio D-3aa
:
3aa = 90
:
10.
Product ratios D-3aa
:
3aa were deduced on the basis of the 1H NMR spectra of the corresponding reaction crudes (see Scheme 2 and ESI†).
:
2. First fractions contained pyrazoles 3, further elution gave pure alkynylhydrazones 4. Overall reaction yields (3 + 4) and products ratio (3
:
4) are collected in Table 2.
4), 3.98 (s, 3H, –COOCH3); 13C-NMR: 162.8 (s, –COOCH3), 144.6 (s, pyrazole C5), 143.9 (s, pyrazole C3), 139.4 (s, aromatic), 128.9 (d, aromatic CH), 128.7 (d, aromatic CH), 125.6 (d, aromatic CH), 109.8 (d, pyrazole C4), 52.0 (q, –COOCH3). MS: 278 m/z (M+).
N–H), 1735 (ester
C
O) (cm−1); 1H-NMR: 8.55 (br s, 1H, –NH-Ph). 7.19–7.33 (m 10H, aromatics), 3.96 (s, 3H, s, –COOCH3); 13C-NMR: 162.9 (s, –COOCH3), 143.7 (s, aromatic, C–N), 139.5 (s,
C
N–), 135.9 (s, aromatic), 113.6–132.0 (aromatic CH), 103.8 (s,
C–), 78.8 (s,
C-Ph), 52.2 (q, –COOCH3). MS 278 m/z (M+). Anal. calcd for C17H14N2O2: C, 73.37; H, 5.07; N, 10.07. Found: C, 73.31; H, 5.11; N, 10.13.
N–H), 1735 (ester
C
O) (cm−1). 1H-NMR: 8.96 (br s, 1H, –NH-Ph), 7.80 (d, 1H, J = 4.0 Hz, aromatic), 6.95–7.40 (m, 13H, aromatics), 3.97 (s, 3H, CH3O–); 13C-NMR: 156.6 (s, aromatic ≥C-OMe), 151.6 (s, aromatic ≥C-N–), 141.6 (s, aromatic), 132.9 (s,
C
N–), 120.8–131.3 (aromatic CH), 119.6 (s, aromatic), 112.0 (d, aromatic), 106.3 (s,
C–), 78.0 (s,
C-Ph), 54.9 (q, CH3O–). MS 326 m/z 326 (M+). Anal. calcd for C22H18N2O2: C, 80.96; H, 5.56; N, 8.58. Found: C, 81.03; H, 5.55 N, 8.64.
C
N–), 135.9 (s, aromatic), 113.6–132.0 (d, aromatic CH), 103.8 (s,
C–), 79.0 (s,
C-Ph). MS: 296 m/z (M+).
Method b: in a 10 mL test tube fitted with a vent needle, a biphasic mixture of alkynylhydrazone 4aa (8 mg, 29 μmol) and CuCl (3 mg, 29 μmol) in MeCN (0.5 mL) and hexane (0.5 mL) was slowly evaporated at 4 °C in 28 days.
In both methods, the residue was a yellow-green solid of dusty appearance that was not suitable for X-ray analysis. 1H NMR spectrum of the residue revealed the presence of pyrazole 3aa as the only product.
Footnote |
| † Electronic supplementary information (ESI) available: 1H and 13C NMR of all new compounds, IR and MS spectra of deuterated chlorohydrazone D-1a. See DOI: https://doi.org/10.1039/d3nj02712k |
| This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2023 |