Aromatic SNF-Approach to Fluorinated Phenyl tert-Butyl Nitroxides.

The interaction of octafluorotoluene (1a), as well as pentafluorobenzonitrile (1b) with tert-butylamine, followed by the oxidation of thus formed tert-butylanilines (2a,b) with meta-chloroperoxybenzoic acid led to functionalized perfluorinated phenyl tert-butyl nitroxides [namely, 4-(N-tert-butyl(oxyl)amino)heptafluorotoluene (3a) and 4-(N-tert-butyl(oxyl)amino)tetrafluorobenzonitrile (3b)] with nearly quantitative total yields. The molecular and crystal structures of nitroxide 3a were proved by single crystal X-ray diffraction analysis. The radical nature of both nitroxides was confirmed by ESR data. The interaction of Cu(hfac)2 with the obtained nitroxides 3a,b gave corresponding trans-bis(1,1,1,5,5,5-hexafluoropentane-2,4-dionato-κ2O,O')bis{4-(N-tert-butyl(oxyl)amino)perfluoroarene-κO}copper (II) complexes ([Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2]). X-ray crystal structure analysis showed square bipyramid coordination of a centrally symmetric Cu polyhedron with the axial positions occupied by oxygen atoms of the nitroxide groups. Magnetic measurements revealed intramolecular ferromagnetic exchange interactions between unpaired electrons of Cu(II) ions and paramagnetic ligands, with exchange interaction parameters JCu-R reaching 53 cm-1.

1. Introduction

The use of organofluorine compounds has had a huge impact on the main areas of modern chemistry, such as the development of new functional materials and pharmaceuticals with unique properties [1,2,3,4]. The ever-increasing importance of fluorine organic compounds brings the development of new directed and high-performance synthetic methods to the forefront [5]. The chemistry of stable organic radicals is among the scientific fields where the need for the creation of new methods for the synthesis of fluorinated compounds is ripe. The number of studies here is extremely limited, although interest in fluorinated organic radicals has been consistently strong [6,7,8,9]. In particular, this interest is due to the unusual magnetic and functional properties that fluorinated paramagnets often manifest [10,11,12,13]. To devise new approaches to their synthesis, we applied nucleophilic substitution reactions of a fluorine atom in a series of polyfluorinated aromatic compounds. This idea has been especially fruitful in nitroxide chemistry. For example, the 4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazole-3-oxide-1-oxyl lithium derivative has been shown to react with polyfluoroaromatic compounds and give rise to new fluorinated nitronyl nitroxides [14,15,16]. It is noteworthy that according to quantum chemical calculations, the reaction follows a concerted pathway without formation of an intermediate [14]. The reason is that at an early stage, the lithium ion facilitates fluoride elimination when the C–C bond between nitroxide and aromatic carbon atoms is formed only partially.

tert-Butyl aryl nitroxides and polynitroxides represent an important class of paramagnets that have been actively used as ligands for the construction of heterospin systems. The design of polyradicals with >N–O groups at the meta position of the benzene ring, which favors intraligand ferromagnetic coupling, and their complexation with M(hfac)2 have allowed Iwamura et al. to obtain a series of molecular magnets capable of cooperative magnetic ordering at 3.4–46 K [17,18,19,20]. In addition, nitroxides containing aryl fragment substituents able to form intermolecular H-bonds are attractive as components of the controlled assembly of high-dimension systems [21,22,23,24,25,26,27].

A typical method for the preparation of tert-butyl aryl nitroxides involves oxidation of the corresponding tert-butyl anilines or phenylhydroxylamines. As a rule, the general synthetic approach starts with a reaction of an appropriate aromatic organometallic compound with tert-nitrosobutane, thus providing aryl tert-butylhydroxylamine, followed by its oxidation to the target radical product [28,29]. The methods for preparation of radicals on the basis of UV irradiation of the corresponding iodo arenes in the presence of tert-nitrosobutane or the interaction of nitro arenes with tert-butyl magnesium chloride (Scheme 1) are also worthy of mention but of no practical value [30,31].

Scheme 1

Synthetic approaches to phenyl tert-butyl nitroxides.

In this study, we have successfully applied the early proposed concept based on nucleophilic substitution of a fluorine atom [14] to the synthesis of fluorinated N-tert-butyl aryl nitroxides. This approach was found to generate highly stable nitroxide radicals in almost quantitative yields. An interaction of bis(hexafluoroacetylacetonato)copper (II) [abbreviated as Cu(hfac)2] with the synthesized nitroxides led to formation of complexes of composition 1:2 containing heterospin clusters (>N–O)2Cu.

2. Results and Discussion

2.1. Synthesis of Fluorinated Nitoxodes

As an approach to phenyl tert-butyl nitroxides, we linked together two transformations: at the first stage, the nucleophilic substitution of a fluorine atom in an activated arene under the action of tert-butylamine, and at the second stage, oxidation of the obtained amine into the target paramagnetic compound. As to the first transformation, it is known that the interaction of perfluorotoluene 1a with methyl-, n-butyl-, tert-butylamine or perfluorobenzonitrile 1b with methylamine in isopropanol at 20–70 °C for 20–90 h leads to selective para-fluorine substitution with the formation of the corresponding N-alkyl anilines in 70–98% yields [32]. We carried out the reaction of 1a,b with tert-butylamine in chloroform at room temperature for 72 and 1.5 h, respectively, and isolated 4-(tert-butylamino)heptafluorotoluene 2a or 4-(tert-butylamino)tetrafluorobenzonitrile 2b in almost quantitative yields. The oxidation of anilines 2a,b with meta-chloroperoxybenzoic acid (m-CPBA) was performed at room temperature and provided target nitroxides 3a,b as viscous red liquids in yields >90% (Scheme 2).

Scheme 2

Synthesis of fluorinated nitroxides 3a,b.

Newly obtained fluorinated nitroxides 3a,b were comprehensively studied both in solution and in a condensed state.

2.2. ESR Measurement of Radicals

The ESR spectra for diluted (~10−4 M) and oxygen-free chloroform solutions of radicals 3a,b showed triplet patterns at g = 2.0057(2) owing to the hfs of an unpaired electron at the nitrogen nucleus (AN = 1.33 mT for 3a and 1.31 mT for 3b; Figure 1 and Figure S18). In the case of nitroxide 3a, a high-resolution ESR spectrum was recorded to obtain more complex splitting of each line of its triplet. The spectrum was well reproduced, taking into account 9 hfs constants on the protons of tert-butyl group (AH = 0.02 mT) and two pairs of hfs constants on the distant fluorine atoms (AFortho = 0.12 mT; AFmeta = 0.06 mT).

Figure 1

Experimental (black curve) and simulated (red curve) ESR spectra of 3a: (a) the whole spectrum, and (b) only the central component.

2.3. Electrocemical Measurements

Electrochemical properties of the studied compounds were evaluated by cyclic voltammetry measurements in a CH2Cl2 solution (Figure S21). Both radicals 3a and 3b demonstrated irreversible oxidation waves with E1/2 ≈ 1 V, which were assigned to the oxidation of the nitroxide radicals to the corresponding oxoammonium cations. For comparison, 4-(N-tert-butyl-N-oxylamino)benzene (4a) and 4-(N-tert-butyl-N-oxylamino)benzotrifluoride (4b) displayed the reversible redox potentials at 0.38 and 0.53 V, respectively (Table 1). The redox reversibility of 4a and 4b could be ascribed to the higher stability of the corresponding oxoammonium cations generated at considerably less oxidative potentials [33].

Table 1

Half-wave potentials for the studied compounds.

Compound	E1/2ox, V	E1/2Red, V
3a *	1.04	−1.44
3b *	1.00	−1.14
[Cu(hfac)2(3a)2] *	1.04	−1.40
[Cu(hfac)2(3b)2] *	1.06	−1.20
4a **	0.38	−1.63
4b **	0.53	−1.37

* Data obtained in a CH2Cl2/0.1 M n-Bu4NPF6 solution at 298 K vs. Fc/Fc+. ** Data obtained in a MeCN/0.1 M n-Bu4NBF4 solution [33].

On the cathodic side, both 3a and 3b exhibited an irreversible redox at −1.44 and −1.14 V, respectively. The non-substituted at the phenyl ring nitroxide 4a also reduced irreversibly, but at a considerably higher potential of −1.63 V, which corresponds to the relative change in the SOMO energy level (Table 1). It is interesting that the trifluoromethylphenyl nitroxide 4b exhibited redox at −1.37 V in a clear reversible manner that results from the stability of the corresponding aminoxy anion [33]. Therefore, the substitution of phenyl ring influences not only on the redox potentials, but also on the stability of the corresponding anionic forms (oxoammonium cations and aminoxy anions).

2.4. Crystal Structure of Radical

Even though both freshly prepared radicals 3a,b were red oils, repeated efforts were made to obtain them in a crystalline form. Finally, by crystallization from a cold hexane solution, we managed to isolate nitroxide 3a as high-quality crystals and determined its molecular and crystalline structure by X-ray diffraction (XRD) analysis. The radical 3a crystallizes in the orthorhombic Pbca space group, and bond lengths of the tert-butyl-nitroxide moiety are completely compatible with those of previously described radicals of this family (see Table 2 for crystallographic data). The nitroxide group in radical 3a is twisted by a large angle (~68°) relative to the aromatic ring (Figure 2), obviously owing to the mutual effects of steric repulsion between the tert-butyl group and phenylene ortho-fluorines and of the electrostatic repulsion of dipoles C–F and N–O. In this regard, it should be noted that a similar dihedral angle in nonfluorinated tert-butyl phenyl nitroxides is twofold smaller and manifests an experimental value of 23–32°. This finding is consistent with the calculated minimum of the heat of formation (MO, B3LYP/6-31G) for the model radical (N-tert-butyl(oxyl)amino)benzene at a 34° angle [34].

Table 2

XRD data on radical 3a and complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2].

Compound	3a	[Cu(hfac)2(3a)2]	[Cu(hfac)2(3b)2]
Empirical formula	C11H9F7NO	C32H20CuF26N2O6	C32H20CuF20N4O6
Formula weight	304.19	1086.04	1000.06
Crystal system	Orthorhombic	Triclinic	Monoclinic
Space group	Pbca	P-1	P21/c
a, Å	11.6131(7)	10.0680(3)	11.3678(8)
b	10.9047(7)	13.8773(5)	8.6913(7)
c	19.7695(12)	16.1651(6)	20.219(1)
α, o	90.00	96.289(2)	90.00
β, o	90.00	102.367(2)	90.158(3)
γ, o	90.00	104.872(2)	90.00
Volume, Å3	2503.6(3)	2099.6(1)	1997.6(2)
Z	8	2	2
Density (calcd.), mg·m−3	1.614	1.718	1.663
Abs. coefficient, mm−1	0.174	0.680	0.687
F(000)	1224	1074	994
Crystal size, mm3	0.40 × 0.70 × 1.00	0.30 × 0.40 × 0.50	0.10 × 0.12 × 0.80
Θ range for data collection, °	2.76–26.07	1.31–26.43	1.79–25.05
Index ranges	−14 ≤ h ≤ 14,−13 ≤ k ≤ 13,−24 ≤ l ≤ 24	−12 ≤ h ≤ 12,−17 ≤ k ≤ 17,−20 ≤ l ≤ 20	−13 ≤ h ≤ 13,−10 ≤ k ≤ 10,−24 ≤ l ≤ 24
Reflections collected	34,642	64,762	26,173
Independent reflections	2473 R(int) 0.0266	8633 R(int) 0.052	3530 R(int) 0.069
Completeness to θ 50º, %	99.6	99.7	99.7
Data/restraints/parameters	2473/0/184	8633/0/613	3530/48/376
Goodness-of-fit on F2	1.044	1.007	1.03
Reflections with I > 2σ(I)	2057	5535	2587
Final R indices at I > 2σ(I)	R1 0.0555, wR2 0.1399	R1 0.0630, wR2 0.1764	R1 0.0530, wR2 0.1466
Final R indices (all data)	R1 0.680, wR2 0.1554	R1 0.1034, wR2 0.2165	R1 0.0791, wR2 0.1716
Largest diff. peak/hole, e·Å−3	0.59/−0.27	0.73/−0.47	0.58/−0.38

Figure 2

Molecular structure (a) and a fragment of the crystal structure (b) of nitroxide 3a (the thermal ellipsoids are drawn at the 50% probability level).

Analysis of the crystal packing of nitroxide 3a revealed C–FCF3…π interactions, with F…Cg and Dpln distances equal to 3.069(2) and 3.049 Å, respectively, and ONO…C short contacts 3.101(3) and 3.130(3) Å (Figure 2), that formally bind molecules into chains along the b axis. The chains in turn are packed into layers parallel to plane (a, b) with contacts F…F equal to 2.650(2) Å, that is, shorter than the sum of van der Waals radii of two fluorine atoms (2.92 Å) [35].

2.5. Synthesis and Crystal Structure of Complexes [Cu(hfac)2()2], [Cu(hfac)2()2] and [Cu(hfac)2()2]

Radicals 3a,b underwent complexation with half equivalents of Cu(hfac)2 in CHCl3 to form 1:2 complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] (Scheme 3), which were recrystallized from n-hexane to obtain crystals suitable for XRD analysis.

Scheme 3

Complexation of Cu(hfac)2 with nitroxides 3a,b.

Figure 3 illustrates the molecular structures of [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] (see Table 2 for crystallographic data). The asymmetric part of [Cu(hfac)2(3a)2] includes two halves of independent units located on the centers of symmetry. In the [Cu(hfac)2(3b)2] complex, CF3 groups and tert-butyl groups are disordered. In the distorted octahedron around the copper ion, two ONO atoms occupy axial positions (dCu–O = 2.398(4), 2.411(5) Å in [Cu(hfac)2(3a)2], and 2.452(4) Å in [Cu(hfac)2(3b)2]). Equatorial Cu–Ohfac distances in both complexes do not exceed 1.946(3) Å (Table 3).

Figure 3

XRD structures of [Cu(hfac)2(3a)2], one of two independent molecules (a), and [Cu(hfac)2(3b)2] with omitted minor positions of CF3 and tert-butyl groups (b).

Table 3

Selected geometrical parameters of complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2].

Parameter	[Cu(hfac)2(3a)2] 1		[Cu(hfac)2(3b)2]
	Bond Lengths, Å
Cu1–O1	2.411(5)	2.398(4)	2.452(4)
Cu1–O2	1.942(3)	1.946(3)	1.944(2)
Cu1–O3	1.935(3)	1.935(3)	1.937(2)
	Bond angles, °
O1–Cu1–O2	89.0(1)	88.1(1)	86.6(1)
O1–Cu1–O3	84.1(1)	82.3(1)	83.9(1)
O2–Cu1–O3	92.4(1)	92.4(1)	92.7(1)

1 For two independent moieties.

Polyfluorinated compounds, including complexes of transition metal ions, are known to have high volatility. Our experiments showed that at reduced pressure (~1 Torr) and a temperature of ~95 °C, at which according to thermal analysis data (Figures S19 and S20) complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] are stable, they almost quantitatively collected on a sublimator finger with the formation of crystalline phases. The infrared (IR) spectra of these phases are fully identical to those of the initial complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2]. Moreover, according to the results of the XRD analysis, the structure of the sublimed complexes coincides with that of the starting compounds [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2]. This finding indicates that the formation constants of these complexes are quite high, and these compounds are transferred to the gas phase in the form of molecules of coordination compounds with subsequent precipitation.

Of note, the interaction of Cu(hfac)2 with radical 3b leads to coordination of the nitroxide group, not the nitrile one. Driven by curiosity, we carried out the reaction of acceptor matrix Cu(hfac)2 with amine 2b (Figure S22). In this case, the nitrile group underwent coordination with the formation of a centrosymmetrical mononuclear complex in which two NCN atoms occupy axial positions [dCu–N = 2.534(3) Å] (Figure 4). This means that the nitrile group in 3b may be also involved in the coordination, as may the nitroxide group, thus affording complexes of chain-polymeric structure.

Figure 4

XRD structure of [Cu(hfac)2(2b)2].

2.6. Magnetic Measurements of Complexes [Cu(hfac)2()2] and [Cu(hfac)2()2]

Temperature dependences of the effective magnetic moment (µeff) for complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] are presented in Figure 5. At 300 K, the µeff values are 3.28 and 3.34 µB for [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2], respectively, and are in agreement with a theoretical spin-only value of 3 µB for three paramagnetic centers: one Cu(II) ion and two nitroxides. The increase in µeff with lowering temperature indicates domination of ferromagnetic exchange interactions between spins of paramagnetic centers. Analysis of experimental µeff(T) dependences using a model of a three-spin exchange cluster [spin Hamiltonian H = −2J] allowed us to estimate exchange interaction parameters. The best-fit values of gCu and exchange interaction parameters J and J are 2.26, 53.1 cm−1, and −17 cm−1 for [Cu(hfac)2(3a)2] and 2.27, 27.0 cm−1, and −0.4 cm−1 for [Cu(hfac)2(3b)2]. The g-factors for nitroxides were fixed at gR = 2 to avoid overparametrization.

Figure 5

The µeff(T) dependences for complexes [Cu(hfac)2(3a)2] (a) and [Cu(hfac)2(3b)2] (b). Solid lines are theoretical curves.

The observed ferromagnetic exchange interactions between the unpaired electrons in [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] are consistent with the XRD data on the axial coordination of the nitroxide groups with the Cu2+ ion. According to the results of experimental [36,37,38,39] and theoretical studies [40,41], such coordination provides the orthogonal arrangement of the spins in the exchange cluster with J values, which in some cases exceed 50 cm−1. With regard to complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2], such a strong ferromagnetic coupling leads to that the half-wave potentials for electrochemical oxidation and reduction are almost the same as those for the free radicals (Table 1).

3. Conclusions

Using octafluorotoluene 1a and pentafluorobenzonitrile 1b as examples, we demonstrated a new synthetic approach for obtaining functionalized fluorinated phenyl tert-butyl nitroxides by sequential substitution of a fluorine atom in polyfluorinated arenes with tert-butylamine and oxidation of resultant tert-butylaniline with m-CPBA. This convenient approach can be useful in the design of nitroxide structures as precursors of new complexes with desired magnetic characteristics. For instance, the interaction of Cu(hfac)2 with newly synthesized nitroxides 3a and 3b yielded isomorphic three-spin complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2]. They are centrosymmetrical square bipyramids with two axial positions occupied by oxygen atoms of the nitroxide groups; this arrangement gives rise to ferromagnetic intramolecular exchange interactions between Cu(II) and radical spins. In complexes with axial coordination of the nitroxyl group observed in compounds [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2], thermo- or photo-induced spin transitions may take place [42,43,44,45,46,47]. To obtain such complexes, it is necessary to modify the structure of tert-butyl nitroxides of type 3 in an appropriate way. We believe that this may be achieved by a decrease in the number of fluorine atoms on the aromatic ring of paramagnetic ligands 3.

4. Materials and Methods

4.1. Reagents and General Methods

Bis(hexafluoroacetylacetonato)copper-(II) [abbreviated as Cu(hfac)2] was prepared and purified by previously described procedures [48]. Other chemicals were of the highest purity commercially available and were used as received. The progress of reactions was monitored by thin-layer chromatography (TLC) on Silica gel 60 F254 aluminum sheets with hexane or CHCl3 as the eluent. Column chromatography was carried out on silica gel (0.063–0.200 mm). NMR spectra were recorded for 2a,b solutions in CDCl3 on Bruker Avance-300 (300.13 MHz for 1H, 282.25 MHz for 19F) and Avance-400 (400.13 MHz for 1H, 100.62 MHz for 13C) spectrometers; chemical shifts (δ) of 1H and 13C{1H} are given in ppm, with the solvent signals serving as the internal standard (δH = 7.24 ppm, δC = 76.9 ppm); the internal standard for 19F spectra was C6F6 (δ = −162.9 ppm). Fourier transform infrared (FT-IR) spectra were acquired in KBr pellets on a Bruker Vector-22 spectrometer. UV-vis spectra were registered on an HP Agilent 8453 spectrophotometer (in 10−5–10−4 M solutions in EtOH). Masses of molecular ions were determined by high-resolution mass spectrometry (HRMS) by means of a DFS Thermo Scientific instrument (EI, 70 eV). Melting points were recorded on a Melter-Toledo FP81 Thermosystem apparatus. Elemental analyses were performed using a Euro EA 3000 elemental analyzer.

4.2. Synthesis of 4-(tert-butylamino)arenes (General Procedure)

A mixture of tert-butyl amine (5.0 mmol, 365 mg) and perfluoroarene 1a or 1b (1.0 mmol) in CHCl3 (5 mL) was stirred at room temperature until the starting substrate was completely consumed: for 72 and 1.5 h for 1a and 1b, respectively. Flash chromatography (SiO2, column 3 × 4 cm, CHCl3 as an eluent) followed by solvent removal under reduced pressure yielded colorless crystals of pure product 2a or 2b.

4-(tert-Butylamino)heptafluorotoluene (2a) [32]: Colorless crystals; yield 99% (286 mg). 1H NMR (300.13 MHz, CDCl3): δ 4.07 (s, 1H), 1.37 (s, 9H). 13C NMR (125.75 MHz, CDCl3): δ 144.94 (dm, J = 254.7 Hz, 2C) 137.86 (dm, J = 239.3 Hz, 2C), 130.22 (tt, J = 12.5 Hz, J = 3.2 Hz, 1C) 121.52 (q, J = 272.4 Hz, 1C), 97.04–98.08 (m, 1C), 53.76 (s, 1C), 30.46 (t, J = 3.7 Hz, 3C). 19F NMR (282.37 MHz, CDCl3): δ 109.58 (t, J = 20.6 Hz, 3F), 21.09–21.48 (m, 2F), 10.67 (d, J = 15.9 Hz, 2F). IR (KBr, cm−1) 3433, 2978, 2945, 2918, 2879, 1659, 1516, 1470, 1433, 1400, 1371, 1335, 1240, 1203, 1180, 1138, 1043, 993, 930, 872, 804, 714, 422. HRMS calcd for C11H10F7N: 289.0696. Found 289.0698.

4-(tert-Butylamino)tetrafluorobenzonitrile (2b): Colorless crystals; yield 99% (244 mg). 1H NMR (400.13 MHz, CDCl3): δ 4.46 (s, 1H), 1.40 (s, 9H). 13C NMR (100.62 MHz, CDCl3): δ 147.98 (dm, J = 256.1 Hz, 2C), 136.13 (dm, J = 240.0 Hz, 2C), 132.22 (tt, J = 11.9 Hz, J = 3.6 Hz, 2C), 108.76 (t, J = 3.7 Hz, 2C), 79.76 (tt, J = 18.0 Hz, J = 1.8 Hz, 1C), 53.85 (s, 1C) 30.60 (s, 1C). 19F NMR (282.37 MHz, CDCl3): δ 26.34 (m, 2F), 7.68 (d, J = 17.4 Hz, 2F). IR (KBr, cm−1) 3427, 3020, 2978, 2943, 2881, 2617, 2420, 2233, 1655, 1525, 1510, 1477, 1435, 1400, 1373, 1313, 1302, 1336, 1201, 1176, 1126, 1047, 993, 978, 889, 806, 721, 661, 634, 523, 494. HRMS calcd for C11H10F4N2: 246.0775. Found 246.0771. The molecular and crystal structures of 2b were refined from single-crystal XRD data (see SI).

4.3. Synthesis of 4-(N-tert-butyl(oxyl)amino)arenes (General Procedure)

A solution of 4-(tert-butylamino)perfluoroarene 2a,b (1 mmol) and m-CPBA (1.2 mmol, 208 mg) in CHCl3 (10 mL) was stirred at room temperature for 48 h. Column chromatography (SiO2, column 3 × 20 cm, CHCl3 as an eluent) afforded a red fraction of radical 3a,b. The solvent was removed under reduced pressure at room temperature to obtain radicals 3a,b.

4-(N-tert-butyl(oxyl)amino)heptafluorotoluene (3a): Red crystals; yield 96% (292 mg). IR (KBr, cm−1) 2987, 2945, 1790, 1766, 1655, 1606, 1504, 1469, 1417, 1358, 1336, 1257, 1232, 1190, 1153, 1197, 991, 904, 831, 791, 715, 588. UV-Vis (EtOH) λmax/nm (lg ε): 383 (2.49), 301 (3.30), 271 (3.51), 221 (3.73), 203 (3.93). HRMS calcd for C11H9O1N1F7: 304.0567. Found 304.0564. Crystals of 3a were grown from hexane at −15 °C. The molecular and crystal structures of 3a were refined from single-crystal XRD data (see main text).

4-(N-tert-butyl(oxyl)amino)tetrafluorobenzonitrile (3b): Red oil; yield 95% (248 mg). IR (neat, cm−1) 3440, 2987, 2945, 1790, 1766, 1655, 1606, 1504, 1469, 1417, 1358, 1336, 1257, 1232, 1190, 1153, 1107, 991, 904, 831, 791, 715, 588. UV-Vis (EtOH) λmax/nm (lg ε): 291 (4.34), 220 (4.02), 204 (4.06). HRMS calcd for C11H9O1N2F4: 261.0646. Found 261.0643.

4.4. Synthesis of Complexes [Cu(hfac)2()2], [Cu(hfac)2(3b)2] and [Cu(hfac)2()2] (General Procedure)

Cu(hfac)2·H2O (248 mg, 0.5 mmol) was added to a solution of radical 3a,b or amine 2b (1.0 mmol) in CHCl3 (10 mL). The reaction mixture was stirred for 30 min and then was incubated at −15 °C for 10 h. The solution was filtered and evaporated. The residue was dissolved in n-hexane (5 mL); the solution was filtered and incubated at −15 °C for 10 h to prepare crystals that were filtered off and air dried.

trans-Bis(1,1,1,5,5,5-hexafluoropentane-2,4-dionato-κ2O,O′)bis{4-(N-tert-butyl(oxyl)amino)heptafluorotoluene-κO}copper (II) ([Cu(hfac)2(3a)2]): Brown crystals; yield 30% (163 mg). Mp 123.6 °C (DSC), mp 137.8–138.6 °C (Melting Point Apparatus). IR (KBr, cm−1) 3140, 2997, 1641, 1604, 1558, 1529, 1504, 1485, 1406, 1362, 1336, 1263, 1211, 1173, 1147, 1107, 1030, 989, 903, 827, 800, 746, 715, 683, 596, 530, 453. Anal. Calcd for C32H20CuF26N2O6: C, 35.39; H, 1.86; F, 45.48; N, 2.58. Found C, 35.07; H, 2.04; F, 45.54; N, 2.45. The molecular and crystal structures of [Cu(hfac)2(3a)2] were refined from single-crystal XRD data (see main text).

trans-Bis(1,1,1,5,5,5-hexafluoropentane-2,4-dionato-κ2O,O′)bis{4-(N-tert-butyl(oxyl)amino)tetrafluorobenzonitrile-κO}copper (II) ([Cu(hfac)2(3b)2]): Brown crystals; yield 30% (151 mg). Mp 68.8 °C (DSC), mp 70.6 °C with decomposition (Melting Point Apparatus). IR (KBr, cm−1) 3431, 3149, 2993, 2927, 2249, 1643, 1606, 1560, 1531, 1500, 1487, 1402, 1354, 1311, 1257, 1213, 1151, 1109, 993, 976, 831, 804, 746, 681, 596, 530, 513. Anal. Calcd for C32H20CuF20N4O6: C, 38.43; H, 2.02; F, 38.00; N, 5.60. Found C, 38.87; H, 2.03; F, 38.05; N, 5.60. The molecular and crystal structures of [Cu(hfac)2(3b)2] were refined from single-crystal XRD data (see main text).

trans-Bis(1,1,1,5,5,5-hexafluoropentane-2,4-dionato-κ2O,O′)bis{4-(N-tert-butyl(oxyl)amino)tetrafluorobenzonitrile-κN}copper (II) [Cu(hfac)2(2b)2]: Green crystals; yield 89% (431 mg). Mp 120.7 °C (Melting Point Apparatus). IR (KBr, cm−1) 3427, 31612980, 2623, 2239, 1657, 1643, 1612, 1562, 1533, 1512, 1487, 1443, 1402, 1373, 1356, 1306, 1259, 1217, 1200, 1153, 1111, 1047, 997, 982, 806, 681, 596, 528, 490. Anal. Calcd for C33H22CuF20N4O4: C, 39.62; H, 2.29; F, 39.17; N, 5.57. Found C, 39.55; H, 2.30; F, 39.11; N, 5.75. The molecular and crystal structures of [Cu(hfac)2(2b)2] were refined from single-crystal XRD data (see main text and SI).

4.5. X-Band ESR Measurements

ESR spectra were acquired in diluted and oxygen-free chloroform solutions at 295 K at the concentrations of ~10−4 M by means of the commercial Bruker X Band (9 GHz) spectrometer Elexsys E 540 (Bruker Corporation, Billerica, MA, USA). For determining the isotropic g-factors (giso), we recorded X-band continuous-wave ESR spectra of a mixture of the investigated radical with the Finland trityl. Then, the known giso of the Finland trityl was used for the spectrum simulation, and the target giso value was excluded. Simulations of the solution ESR lines were carried out in software Easy Spin, which is available at http://www.easypin.org.

4.6. Cyclic Voltammetry Measurements

The analysis of electrochemical behavior of 3a,b, [Cu(hfac)2(3a)2], and [Cu(hfac)2(3b)2] was performed in a CH2Cl2 solution by computer-controlled P-8 nano potentiostat (Elins, Chernogolovka, Russia) in combination with a three-electrode cell (Gamry, Warminster, PA, USA); 0.1 M tetrabutylammonium hexafluorophosphate served as a supporting electrolyte. Pt, a Pt wire, and Ag/AgCl were used as working, counter, and reference electrodes, respectively. The reference electrode was calibrated by measuring the redox potential of ferrocene. The scan rate was 100 mV/s.

4.7. Crystallographic Analysis

XRD data were collected at 200(2) K for 2b and 3a, and at room temperature for [Cu(hfac)2(2b)2], [Cu(hfac)2(3a)2], and [Cu(hfac)2(3b)2] on a Bruker Kappa Apex II CCD diffractometer using φ, ω scans of narrow frames with Mo Kα radiation (λ = 0.71073 Å) and a graphite monochromator. Absorption corrections were applied empirically using SADABS programs [49]. The structures were solved by direct methods and refined by full-matrix least-squares method against all F2 in anisotropic approximation (beside the atoms H) using the SHELX-97 programs set [50]. The H atoms positions, except the amino hydrogen in 2b, were treated with the riding model. The amino hydrogen position of 2b was localized from difference map and refined independently as well as thermal parameter. The analysis of the geometry and the intermolecular interactions for nitroxide 3a and complexes [Cu(hfac)2(3a)2], [Cu(hfac)2(3b)2] was performed using PLATON program. [51,52] CCDC 1962404–1962408 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/cgi-bin/catreq.cgi, or from the Cambridge Crystallographic Data Centre, 12 Union Road, CambridgeCB2 1EZ, UK; fax: (+44) 1223 336 033; or e-mail: .

4.8. Thermal Analysis

Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) measurements were performed on a NETZSCH STA 409 instrument at a heating rate of 10 K/min under He flow 30 mL/min. The onset temperature of decomposition (T0), the melting temperature (Tm) were determined using NETZSCH Proteus Thermal Analysis software.

4.9. Magnetic Measurements

Magnetic susceptibility of the polycrystalline samples of complexes [Cu(hfac)2(3a)2] and [Cu(hfac)2(3b)2] was measured with a Quantum Design MPMSXL SQUID magnetometer in the temperature range 2 to 300 K with a magnetic field of up to 5 kOe. Diamagnetic corrections were made via the Pascal constants. The effective magnetic moment was calculated as μeff(T) = [(3k/NAμB2)χT]1/2 ≈ (8χT)1/2.