INTRODUCTION

Studied [Cu(hfac)2LR1/R2] belong to breathing crystals based on Cu(II) complexes with nitroxide radicals (NRs) which are characterized by reversible structural rearrangements occurring in the solid phase of these compounds upon changes in the temperature and/or pressure. These rearrangements change the spin multiplicity in exchange clusters {>N–•O–Cu2+–O•–N<} or {>N–•O–Cu2+}, which causes the appearance of anomalies in temperature dependence curves of the effective magnetic moment (μeff) of the compound and specific changes in the EPR spectra [1-7]. In the discussed class of compounds, phase transitions often take place without destructing the crystal, and this makes them valuable objects for a detailed analysis of structural transformations induced by the external action. This fact in turn underlies the development of a specific group of sensors to the external action [8, 9]. A representative group of the studied heterospin compounds is composed of Cu(hfac)2 complexes with NRs containing 1-R-pyrazol-4-yl substituent in the second position of the imidazoline ring of the nitronylnitroxide moiety (LR, Scheme 1). The experimental database collected so far for these compounds, which contains information about the magnitudes of exchange integrals and structural parameters of exchange clusters, enables one to predict the appearance of spin transitions in the complexes depending on the size and position of the alkyl substituent R in the pyrazole ring LR [1]. The compounds containing an additional substituent R2 in the pyrazole ring (Scheme 1) have been studied much less, although this substitution can lead to nontrivial effects. Thus, for [Cu(hfac)2LMe/Et] and [Cu(hfac)2LEt/Me] a reversible solid-phase polymerization/depolymerization reaction was detected (for the first time for heterospin compounds) [6, 7]. A change in the position of R1 and R2 in LMe/Et and LEt/Me resulted in the opposite directions of effects when the crystals were cooled: polymerization of [Cu(hfac)2LEt/Me] and depolymerization of [Cu(hfac)2LMe/Et]. This fact stimulated us to expand a series of polymeric chain complexes [Cu(hfac)2LR1/R2] containing alkyl substituents in positions 1 and 3 of the pyrazole ring [10] and examine the temperatureную structure dynamics and magnetic properties of [Cu(hfac)2] complexes with LR1/R2, where R1 and R2 = Me, R1 and R2 = Et, R1 = Me and R2 = p-Pr, R1 = p-Pr and R2 = Et.

Scheme 1
scheme 1

LR and LR1/R2 NRs.

Full size image

EXPERIMENTAL

Commercial reagents and solvents were used without additional purification. The reaction course was monitored by thin layer chromatography on Silica gel 60 F254 (Merck) aluminum plates. The chromatographic analysis was conducted on silica gel (0.063-0.200 mm, Merck) for column chromatography. IR spectra (4000-400 cm–1) of nitroxides were recorded on a VECTOR 22 Bruker instrument in KBr pellets. Microanalysis was carried out on a EA-3000 HEKAtech GmbH analyzer. Using the published procedures we synthesized 2,3-bis(hydroxyamino)-2,3-dimethylbutane, 2,3-bis(hydroxyamino)-2,3-dimethylbutane sulfate monohydrate [11], and 2-(1,3-dimethyl-1H-pyrazol-4-yl)-4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazole-3-oxide-1-oxyl (LMe/Me) [12]. Other nitroxides used for the synthesis of complexes described below were obtained by Scheme 2 described in detail in [6, 7, 10] and common for dialkyl-substituted NRs.

Scheme 2
scheme 2

General synthesis scheme of dialkyl-substituted NRs. 2,3-Bis(hydroxyamino)-2,3-dimethylbutane sulfate monohydrate or 2,3-bis(hydroxyamino)-2,3-dimethylbutane (1); NaIO4/CH2Cl2/H2O or MnO2/O2/CH3OH (2).

Full size image

2-(1,3-diethyl-1H-pyrazol-4-yl)-4,4,5,5-tetramethyl-4,5-dihydro-1Н-imidazole-3-oxide-1-oxyl (LEt/Et) and 2-(1-methyl-3-propyl-1H-pyrazol-4-yl)-4,4,5,5-tetramethyl-4,5-dihydro-1Н-imidazole-3-oxide-1-oxyl (LMe/Pr) was synthesized from 1,3(5)-diethyl-1H-pyrazole and 1-methyl-3-propyl-1H-pyrazole-4-carbaldehyde respectively. The formylation procedure, condensation with 2,3-bis(hydroxyamino)-2,3-dimethylbutane, and oxidation were performed as the previously described procedure for LMe/Et [10, 13].

LEt/Et. Yield: 88%. According to X-ray crystallography and IR spectroscopy data, during the recrystallization of the product from hexane to the solid phase, dark blue rhombic LEt/Et crystals were isolated. M.p. 92 °C. Found (%): C 59.5, H 7.8, N 20.0. C14H23N4O2. Calculated (%): C 60.2, H 8.3, N 20.1. IR spectrum (ν, cm–1): 3444, 3126, 2968, 2931, 2871, 1580, 1506, 1445, 1393, 1352, 1321, 1220, 1173, 1156, 1143, 1111, 1085, 1054, 1014, 956, 871, 847, 802, 761, 728, 698, 654, 608, 540, 500, 464. μeff = 1.8 μB (50-300 K).

LMe/Pr. Yield: 95%. According to X-ray crystallography and IR spectroscopy data, during the recrystallization of the product from hexane to the solid phase, blue-purple LMe/Pr crystals were isolated. M.p. 90 °C. Found (%): C 60.4, H 8.1, N 20.0. C14H23N4O2. Calculated (%): C 60.2, H 8.3, N 20.1. IR spectrum (ν, cm–1): 3418, 3125, 2999, 2961, 2935, 2871, 1583, 1512, 1486, 1458, 1397, 1352, 1291, 1219, 1173, 1154, 1142, 1088, 1004, 895, 871, 839, 774, 747, 710, 654, 642, 638, 609. μeff= 1.75 μB (30-300 K).

LMe/Et and LEt/Me crystals were grown from hexane [6, 7].

[Cu(hfac)2LMe/Me]. A mixture of Cu(hfac)2 (0.0950 g, 0.20 mmol) and LMe/Me (0.0500 g, 0.20 mmol) was dissolved in 15 mL of ether. The obtained dark brown solution was added with 15 mL of hexane, then the solvent excess was removed by a slow air flow (up to ~15 mL). The dark crystals formed (dark blue when ground) were filtered off, washed with hexane, and dried in air. Yield: 83%. M.p. 108 °C. The complex is soluble in most of organic solvents. On keeping in the solution for a day and longer, it gradually decomposes (in few minutes with low heating temperatures). During a long-term storage (more than 10 years) in the crystalline form, the complex gradually transforms into a complex with imino derivatives of this radical. Calculated for C22H21N4O6F12Cu (%): C 36.3, H 2.9, N 7.7, F 31.3. Found (%): C 37.0, H 3.1, N 7.6, F 31.3. This complex also forms when pure benzene or its mixtures with hexane (heptane) are used as solvents.

[Cu(hfac)2LEt/Et]. A mixture of Cu(hfac)2 (0.0477 g, 0.1 mmol) and LEt/Et (0.0279 g, 0.10 mmol) samples was dissolved in 7 mL of hexane. The solvent excess was slowly removed by an air flow at room temperature to a volume of ~2 mL, after which the reaction mixture was cooled to –18 °C. Dark brown crystals formed in 36-38 h were filtered off and dried in air. Yield: 88%. M.p. 86-87 °C. Found (%): C 38.2, H 3.4, F 30.6, N 7.7. C24H25CuF12N4O6. Calculated (%): C 38.1, H 3.3, F 30.1, N 7.4.

[Cu(hfac)2LPr/Et]. A mixture of Cu(hfac)2 (0.0477 g, 0.1 mmol) and mixtures of oil-like LPr/Et and LPr/Et* (0.0300 g, 0.1 mmol) were dissolved in 2.5 mL of Et2O and 4 mL of hexane were added. The reaction mixture was cooled to –30 °C. Large dark-red prismatic crystals formed in ~120 h were filtered off and dried in air. Yield: 77%. M.p. 81 °C. Found (%): C 39.1, H 3.1, F 30.5, N 7.4. C25H27CuF12N4O6. Calculated (%): C 38.9, H 3.5, F 29.6, N 7.3.

[Cu(hfac)2LMe/Pr]. A mixture of Cu(hfac)2 (0.0477 g, 0.1 mmol) and LMe/Pr (0.0280 g, 0.1 mmol) samples was dissolved in 4 mL of hexane at low heating temperatures (47 °C). The solution obtained was added with 0.5 mL of Et2O. The reaction mixture was kept at room temperature for 1 h, after which it was cooled to +10 °C. Brown crystals formed in 24 h were filtered off and dried in air. Yield: 71%. M.p. 105-106 °C. Found (%): C 38.8, H 3.6, F 29.1, N 7.8. C24H25CuF12N4O6. Calculated (%): C 38.1, H 3.3, F 30.1, N 7.4.

X-ray crystallography studies. Sets of reflections from single crystals of the compounds were collected on automated Bruker AXS - SMART APEX diffractometers (MoKα radiation) with a Helix (Oxford Cryosystems) helium flow cooling system and Apex Duo (CuKα radiation) with a Cobra (Oxford Cryosystems) cryosystem using the standard procedures. Structures were solved by direct methods and refined by the full-matrix least squares technique in the anisotropic approximation for non-hydrogen atoms. Hydrogen atoms are partially localized from the difference maps (the other were calculated geometrically) and included in the refinement in the riding model. All calculations were made using the SHELX software [14, 15]. Crystallographic characteristics of the studied compounds and experimental details are summarized in Tables 1-3. Full sets of the crystallographic data have been deposited with the Cambridge Crystallography Data Center (Tables 1-3).

Magnetic measurements. The magnetic susceptibility (χ) of the polycrystalline samples was measured on a Quantum Design MPMSXL SQUID-magnetometer in the temperature range of 2-300 K at the magnetic field strength of 5 kOe. Paramagnetic components of the magnetic susceptibility were determined with regard to a diamagnetic contribution estimated from Pascal constants. The effective magnetic moment was calculated by the formula \({{\mu }_{\text{eff}}}~=~{{[\text{3}k\chi T/({{N}_{\text{A}}}\mu _{\text{B}}^{2})]}^{\text{1}/\text{2}}},\) where NA, μB, and k are the Avogadro number, the Bohr magneton, and the Boltzmann constant respectively.

Table 1 Crystallographic Characteristics of the [Cu(hfac)2LEt/Et] and Experimental Details
Full size table
Table 2 Crystallographic Characteristics of the [Cu(hfac)2LMe/Me] and Experimental Details
Full size table
Table 3 Crystallographic Characteristics of the Compounds and Experimental Details
Full size table

RESULTS AND DISCUSSION

The structure of nitroxides LMe/Me, LEt/Et, and LMe/Pr, which we managed to obtain as single crystals suitable for the single crystal X-ray diffraction (XRD) analysis, is illustrated in Fig. 1. The same figure depicts the structures of LMe/Et and LEt/Me molecules, which are further used to compare the magnetic properties of individual NRs. The lengths of N–O bonds within 1.278(2)-1.284(2) Å, which is typical of nitronylnitroxide radicals [16-18]. Distances between the paramagnetic centers – O atoms of the neighboring molecules – exceed 4 Å. The introduction of the R2 substituent to the third position of the pyrazole ring causes a substantial turn of the pyrazole ring plane relative to the nitronylnitroxide fragment {ONCNO}. In monoalkyl-substituted spin-labeled pyrazoles LMe and LEt, this angle does not exceed 5° whereas in dialkylpyrazoles, it has a value larger than 30° (Table 4).

Results of studies of the magnetic properties of LR1/R2 are illustrated in Fig. 2. High-temperature μeff values are well consistent with a theoretical purely spin value of 1.73 μB. For LMe/Et and LEt/Me, when the temperature lowers, μeff first gradually decreases, and below 100 K this decrease is sharper. This behavior of the μeff(T) dependences indicates the occurrence of antiferromagnetic exchange interactions between NR spins. For LMe/Pr the μeff value remains constant on cooling to 5 K, and for LEt/Et some increase in μeff is observed below 30 K, which evidences the occurrence of ferromagnetic exchange interactions. Experimental μeff(T) dependences are well described within the model of exchange-bound dimers (spin Hamiltonian H = –2J·S1S2). Optimal g-factors and exchange interaction parameters J are listed in Table 4. An increase in the size of substituents in the pyrazole ring leads to a decrease in the energy of exchange interactions between spins of nitroxides.

Fig. 1
figure 1

Structures of LMe/Me (a), LEt/Et (b), LMe/Et (c), LEt/Me (d), and LMe/Pr (e) molecules.

Full size image
Table 4 Selected Stereochemical NR Parameters, g-Factors, and the Exchange Interaction Energy J
Full size table

All [Cu(hfac)2LR1/R2] discussed below have a polymeric chain structure of the solid phase. The environment of all crystallographically independent Cu atoms is a centrosymmetric square bipyramid whose equatorial plane is formed by Ohfac atoms and axial positions are taken either by ONO atoms of the nitronylnitroxide moiety or pyrazole N atoms (Fig. 3). Due to the bridging (ditopic) coordination of paramagnetic ligands there is an alternation of {CuO4(ONO)2} and {CuO4N2} coordination sites, which provides the general head-to-head motif of the chain (Fig. 3). The {CuO4(ONO)2} sites are heterospin exchange clusters {>N–•O–Cu2+–O•–N<} with Cu–ONO distances of 2.280(2)-2.360(1) Å at room temperature (Table 5). Here the second ONO atom of the nitronylnitroxide moiety does not participate in the coordination and interchain distances between these atoms are longer than 3.9 Å.

Fig. 2
figure 2

Dependence μeff(T) for nitronylnitroxides LR1/R2 (dots are the experimental values, solid lines are the theoretical curves).

Full size image
Fig. 3
figure 3

Chain fragment in the [Cu(hfac)2LMe/Me] structure.

Full size image

The investigation of the magnetic properties reveals that for the [Cu(hfac)2LMe/Me] complex the μeff value at 300 K is 2.57 μB (Fig. 4а). With decreasing temperature it gradually increases to 2.69 μB (at 90 K), after which it drops to 2.38 μB (at 83 K). On further cooling a small rise to 2.42 μB is observed (at 45 K) and again an abrupt decrease to 1.83 μB (at 35 K). Below 35 K the μeff value practically does not change and is close to the theoretical purely spin value of 1.86 μB for one paramagnetic center with spin S = 1/2 at a g-factor of 2.15.

The heating and cooling curves almost coincide, which indicates that the observed transitions are reversible. In the temperature range of 300-90 K, the μeff(T) dependence is described by the expression for the magnetic susceptibility with equal contributions from three-spin exchange clusters {>N–•O–Cu2+–O•–N<} and isolated Cu2+ ions of the {CuO4N2} coordination sites (Fig. 4а, curve 1)

χ = (χ(J,T) + C/T)/2,

where \(\chi (J,T)=\frac{N\mu_{\text{B}}^{2}}{3kT}\cdot \frac{15{{(({{g}_{\text{Cu}}}+2{{g}_{\text{R}}})/3)}^{2}}\cdot {{e}^{J/kT}}+1.5{{((4{{g}_{\text{R}}}-{{g}_{\text{Cu}}})/3)}^{2}}\cdot {{e}^{-2J/kT}}+1.5g_{\text{Cu}}^{2}}{4\cdot {{e}^{J/kT}}+2\cdot {{e}^{-2J/kT}}+2}\) is the expression for the magnetic susceptibility of the three-spin exchange cluster (spin Hamiltonian H = –2J(SR1SCu + SCuSR2), SR1 = SR2 = 1/2 – NR spins). The magnetic susceptibility of isolated Cu2+ ions was taken into account using the Curie law χ = C/T, where \(C~=~\,N\mu _{\text{B}}^{2}{{({{g}_{\text{Cu}}})}^{\text{2}}}S(S+\text{1})/\text{3k }\approx {{\text{(}{{g}_{\text{Cu}}}\text{)}}^{\text{2}}}S(S+\text{1})/\text{8}\text{.}\) Optimal gCu and J values are 2.09 cm–1 and 21.5 cm–1 (the value gR = 2 was fixed to avoid reparametrization).

Drastic changes in the μeff(T) dependence are caused by structural transformations in heterospin clusters {>N–•O–Cu2+–O•–N<}. Thus, a change in the nitroxide coordination from axial to equatorial gives rise to a strong antiferromagnetic exchange between Cu2+ ion and NR spins and the compensation of a part of spins. The analysis of the μeff(T) dependence in the range of 85-45 K allows the estimation of a fraction of α clusters, in which a structural transition occurred, by the expression χ = ((1 – α)·χ(J,T) + (1 + α)·C/T)/2, using the values of gCu and J parameters for the high-temperature range. Changes in μeff at 85 K correspond to a transition of ~40% of heterospin clusters {>N–•O–Cu2+–O•–N<} (α = 0.4, Fig. 4a, curve 3). At 40 K a transition takes place in the rest of {CuO6} coordination sites (α = 1, Fig. 4a, curve 3).

Table 5 Slected Bond Lengths, Intermolecular Contacts (Å), and Angles (deg) in [Cu(hfac)2(LR1/R2)]
Full size table
Fig. 4
figure 4

Dependence μeff(T) for [Cu(hfac)2LMe/Me] (a) and [Cu(hfac)2LMe/Pr] (b) complexes; dots are the experimental data, solid lines (1, 2, 3) are the theoretical curves obtained as a result of modeling (comments in the text).

Full size image

For the [Cu(hfac)2LMe/Pr] complex the μeff(T) dependence has another character (Fig. 4b). The μeff value of 2.79 μB at 300 K at first decreases gradually, and below 80 K there is an abrupt decrease reaching 1.78 μB at 5 K. The high-temperature μeff value corresponds to the theoretical purely spin value of 2.45 μB for two paramagnetic centers: one Cu(II) ion and nitroxide with spins S = 1/2. The low-temperature μeff value corresponds to the theoretical purely spin value for one paramagnetic center with spin S = 1/2.

For [Cu(hfac)2LEt/Et] the μeff value at 300 K is 2.65 μB (Fig. 5а), and decreases with temperature, reaching a plateau of ~1.84 μB below 80 K. The high-temperature μeff value corresponds to the theoretical purely spin value of 2.45 μB for two paramagnetic centers: one Cu2+ ion and NR with spins S = 1/2. The low-temperature μeff value is close to the theoretical purely spin value for one paramagnetic center with spin S = 1/2. The decrease in μeff in the temperature range of 300-80 K indicates the occurrence of a smooth spin transition and strong antiferromagnetic interactions in the low-temperature range for three-spin exchange clusters {>N–•O–CuII–O•–N<}, which are characteristic of the equatorial coordination of nitroxides. The analysis of the μeff(T) dependence below 130 K allows the estimation of the antiferromagnetic exchange interaction energy as |J| > 85 cm–1 at gCu = 2.3. In the room temperature range, the μeff(T) dependence approaches the theoretical curve with gCu and J parameters of 2.3 cm–1 and 14 cm–1, which corresponds to weak ferromagnetic interactions in three-spin exchange clusters {>N–•O–CuII–O•–N<} that are characteristic of the axial coordination of nitroxides. For comparison we present the theoretical curve with gCu and J parameters of 2.3 cm–1 and –14 cm–1, which corresponds to weak antiferromagnetic exchange.

For the [Cu(hfac)2LPr/Et] complex the μэeff value at 300 K is 2.55 μB and with decreasing temperature it at first decrease gradually and below 150 K it decreases abruptly (Fig. 5b), reaching 2.27 μB at 80 K. Below 80 K there is an increase in μeff to 2.42 μB at 7 K, which reveals the occurrence of ferromagnetic exchange interactions. The analysis of the μeff(T) dependence in the temperature range of 5-80 K allows the estimation of the exchange interaction energy and a fraction of heterospin clusters {>N–•O–Cu2+–O•–N<} with the equatorial coordination of nitroxides, and consequently, with strong antiferromagnetic exchange. Optimal values of gCu and J parameters are 2.14 cm–1 and 9.7 cm–1 at a fixed value α = 0.5 (Table 5b, curve 3). Theoretical curve 1 corresponds to the value α = 1 at the same gCu and J parameters (2.14 cm–1 and 9.7 cm–1), and in the high-temperature range it approaches the experimental μeff values. In the temperature range of 300-150 K, the experimental μeff(T) dependence is well described by a curve with gCu and J parameters of 2.24 cm–1 and –16.4 cm–1 at α = 1 (Fig. 5b, curve 2).

In [Cu(hfac)2LR1/R2] crystal structures, the polymeric chains are packed similarly (Fig. 6), and the interchain ONO…ONO distances between uncoordinated atoms are sufficiently long (Table 5), therefore, the observed distinctions in the temperature dynamics of magnetic properties are first of all caused by the features of changes in the structure of the complex inside the chains.

Fig. 5
figure 5

Dependence μeff(T) for [Cu(hfac)2LEt/Et] (а) and [Cu(hfac)2LPr/Et] (b) complexes; dots are the experimental data, solid lines (1, 2, 3) are the theoretical curves obtained as a result of modeling (comments in the text).

Full size image
Fig. 6
figure 6

Packing of polymeric chains of [Cu(hfac)2LR1/R2] complexes in the crystal.

Full size image

A series of low-temperature single crystal XRD studies of [Cu(hfac)2(LMe/Me)] has shown that upon cooling from 300 K to 120 K the environment of the Cu atom changes insignificantly: the Cu–ONO bond length decreases only by 0.024 Å (from 2.360 Å to 2.336 Å). Then at 75 K it reduces immediately by 0.152 Å (Table 5), i.e. there is a tendency to replace elongated Jahn Teller’s axis ONO–Cu–ONO by Ohfаc–Cu–Ohfаc in the Cu atom environment in the {CuO6} coordination site. We failed to trace structural changes below this temperature because of crystal fracturing.

The comparison of [Cu(hfac)2(LMe/Me)] and [Cu(hfac)2(LMe/Pr)] structures revealed the similar structures of their fragments and chain packings as well as changes in their structures with decreasing temperature from 300 K to 100 K. A significant distinction was found in Cu–N distances at {CuO4N2} sites: in [Cu(hfac)2(LMe/Pr)] it is larger by 0.1 Å and practically remains unchanged when the crystal is cooled. This is likely to be due to the location of the Pr-substituent in a limited space between hfac ligands of this site and the uncoordinated NR fragment (Fig. 7), which does not allow the Cu and ONO atoms to approach at a distance required for the spin transition. It is also worth noting an appreciable decrease in the distances between NRs of the neighboring ONO…ONO chains by на 0.35 Å in the temperature range of 295-103 K (Table 5), which results in strengthening antiferromagnetic interactions between them and causes a decrease in the μeff value at temperatures below 80 K.

The structural feature of [Cu(hfac)2LEt/Et] and [Cu(hfac)2LPr/Et] is the location of terminal groups of alkyl substituents on one side of the pyrazole ring (in free LEt/Et they are on different sides, Fig 1). In its polymeric chain at 295 K, the Cu–ONO distances in the {CuO6} sites (Table 5) are the shortest among all [Cu(hfаc)2LEt/R2] with the head-to-head motif. When the temperature lowers from 295 K to о 103 K, their smooth shortening is detected from 2.279(2) Å to 2.020(1) Å, i.e. in {CuO6} sites, the elongated Jahn-Teller axis changes from ONO–Cu–ONO to Ohfаc–Cu–Ohfаc. This corresponds to a decrease in the μeff value due to strengthening exchange interactions of the antiferromagnetic character in these sites (Fig. 5а). As shown previously by quantum chemical calculations, this effect is a consequence of a decrease in the fraction of a low-temperature phase in the structure of the high-temperature phase with decreasing temperature [19-21].

Fig. 7
figure 7

[Cu(hfac)2(LMe/Pr)] chain fragment at T = 103 K in the space-filling representation; Pr-substituent atoms are shown by crimson (see electronic version).

Full size image
Fig. 8
figure 8

[Cu(hfac)2(LPr/Et)] chain fragments in the space-filling representation; atoms of alkyl substituents are shown by purple (see electronic version).

Full size image

The [Cu(hfac)2LPr/Et] single crystals turned to be thermally inelastic and decomposed when cooled. Therefore, we failed to trace changes in distances in solid KC upon its cooling. As can be seen in Fig. 8, two bulky substituents in LPr/Et, which are located between hfac ligands and the uncoordinated nitroxide fragment, certainly should hinder the displacement of chain fragment when the crystal is cooled.

CONCLUSIONS

Thus, the comparison of the structure and the temperature structure dynamics of [Cu(hfac)2LMe/Me], [Cu(hfac)2LEt/Et], [Cu(hfac)2LMe/Pr], [Cu(hfac)2LPr/Et] heterospin complexes with [Cu(hfac)2LMe/H], [Cu(hfac)2LEt/H], [Cu(hfac)2LPr/H], [Cu(hfac)2LEt/Me], and [Cu(hfac)2LMe/Et] previously described for complexes with mono- and dialkyl-substituted spin-labeled pyrazoles has shown that changes of both R1 and R2 substituents can significantly affect not only the possibility of the appearance of anomalies in the temperature dependence curve of the effective magnetic moment but also its form.