Thionitrosyl Complexes of Rhenium and Technetium with PPh₃ and Chelating Ligands—Synthesis and Reactivity

Abstract

In contrast to corresponding nitrosyl compounds, thionitrosyl complexes of rhenium and technetium are rare. Synthetic access to the thionitrosyl core is possible by two main approaches: (i) the treatment of corresponding nitrido complexes with S₂Cl₂ and (ii) by reaction of halide complexes with trithiazyl chloride. The first synthetic route was applied for the synthesis of novel rhenium and technetium thionitrosyls with the metals in their oxidation states “+1” and “+2”. [M^VNCl₂(PPh₃)₂], [M^VNCl(PPh₃)(L^OMe)] and [M^VINCl₂(L^OMe)] (M = Re, Tc; {L^OMe}⁻ = (η⁵-cyclopentadienyl)tris(dimethyl phosphito-P)cobaltate(III)) complexes have been used as starting materials for the synthesis of [Re^II(NS)Cl₃(PPh₃)₂] (1), [Re^II(NS)Cl₃(PPh₃)(OPPh₃)] (2), [Re^II(NS)Cl(PPh₃)(L^OMe)]⁺ (4a), [Re^II(NS)Cl₂(L^OMe)] (5a), [Tc^II(NS)Cl(PPh₃)(L^OMe)]⁺ (4b) and [Tc^II(NS)Cl₂(L^OMe)] (5b). The triphenylphosphine complex 1 is partially suitable as a precursor for ongoing ligand exchange reactions and has been used for the synthesis of [Re^I(NS)(PPh₃)(Et₂btu)₂] (3a) (HEt₂btu = N,N-diethyl-N′-benzoyl thiourea) containing two chelating benzoyl thioureato ligands. The novel compounds have been isolated in crystalline form and studied by X-ray diffraction and spectroscopic methods including IR, NMR and EPR spectroscopy and (where possible) mass spectrometry. A comparison of structurally related rhenium and technetium complexes allows for conclusions about similarities and differences in stability, reaction kinetics and redox behavior between these 4d and 5d transition metals.

1. Introduction

The two “group 7” elements technetium and rhenium possess radioactive isotopes (^99mTc, ^186,188Re), which make them more than interesting for nuclear medical applications [1,2,3,4,5,6,7,8,9,10,11]. The metastable nuclide ^99mTc (pure γ-emitter, half-life 6 h, γ-energy: 411 keV) is sometimes denoted as the “workhorse” of diagnostic nuclear medicine, since more than 80 percent of clinical applications in this field in more than 10,000 hospitals worldwide are performed with ^99mTc-containg pharmaceuticals [12]. This dominating position is mainly due to the favorable nuclear properties of this nuclide: (i) an almost optimal γ-energy without considerable amounts of accompanying particle radiation, (ii) a half-life which is long enough to examine metabolic processes but short enough to avoid a considerable radiation burden to the patient and (iii) the formation of ^99mTc in a so-called generator system from ⁹⁹Mo [13], which ensures a permanent availability of this nuclide in the clinics. A similar generator system starting from ¹⁸⁸W as mother nuclide is also available for the supply of ¹⁸⁸Re (E_β,max: 2.12 MeV, E_β,av: 784 keV, half-life: 16.9 h), a beta-emitting isotope with potential in nuclear medical therapy [14]. For both isotopes, kit-like preparations of the pharmaceuticals have been developed, which allow for a rapid and cost-efficient production of the individual administrations [1,14].

^99mTc and ¹⁸⁸Re are also under discussion as an almost perfect example of a “matched pair” of nuclides for theragnostic applications [15,16], which is a modern approach in personalized medicine, where an optimized dosage of a therapeutic drug becomes possible by the parallel monitoring of the therapeutic effects by a diagnostic counterpart.

Although the syntheses of ^99mTc and ¹⁸⁸Re compounds proceed at an approximate nanomolar level due to the low (chemical) concentrations obtained from the corresponding nuclide generators, fundamental chemical and structural studies concerning the formed compounds are commonly conducted with natural rhenium and the long-lived technetium isotope ⁹⁹Tc, a weak beta emitter (E_max = 297 keV, half-life 2.11 × 10⁵ years), which is also available in macroscopic amounts. During such studies, occasionally at the first glance, unusual solutions for the development of novel radiopharmaceuticals have been found, including several organometallic approaches [17,18,19,20,21,22,23] and drugs on the basis of nitrido [24,25,26,27] or hydrazido compounds [28,29,30]. Finally, the usability of a potential solution is decided by the successful transfer to the low concentration level of ^99mTc and the stability of the product under aqueous conditions. Consequently, the potential of a chemical solution for a final medical application cannot easily be predicted by a simple, intuitive evaluation of a substance or a substance class. A good example is “^99mTc-sestamibi”, a cationic organotechnetium(I) complex with six isocyanide ligands. As a compound with six technetium-carbon bonds, it was most probably not in the first row for candidates for myocardial imaging. Nowadays, it is one of the most used radiopharmaceuticals worldwide and, up to now, more than 40 million patients have been examined with this drug [17]. Also, nitrosyl complexes of technetium were considered for such purposes and related ^99mTc compounds have been tested for their biodistribution [31]. The related structural chemistry was performed with the long-lived isotope ⁹⁹Tc and starting from these early days in the 1980s, our knowledge about nitrosyltechnetium compounds has significantly increased.

In contrast to the relatively large number of nitrosyl complexes of the elements rhenium [32,33,34,35,36,37,38,39,40] and technetium [40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60], there are only a few reports about thionitrosyls of these two elements [61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80]. This should be mainly due to the lack of a readily available monomeric nitrogen sulfide. Although the thionitrosyl radical has been described spectroscopically and some NS⁺ salts have been isolated [81,82,83,84], the synthesis of such compounds remains a challenge and they have only occasionally been used as precursors for the synthesis of thionitrosyl complexes [61,83,84]. More convenient are reactions starting from cyclic compounds such as tetrasulfur tetranitride, S₄N₄, or trithiazyltrichloride, (NSCl)₃ (Figure 1), which readily undergo thermal or photochemical decomposition under release of NS⁺ or NSCl⁺ building blocks. But it should be noted that N₄S₄ is a potentially dangerous compound, which tends to explode on heating or upon shock. Thus, it is not a favored agent at least for reactions with the radioactive technetium. Finally, the most promising synthetic approach to thionitrosyl complexes is the addition of a “sulfur atom” to a nitrido ligand. A variety of “sulfur sources” have been used including elemental sulfur, SOCl₂, dithionite or NCS⁻ ligands. Most successful, however, are reactions of nitrido complexes with S₂Cl₂, which give low-valent thionitrosyl species in good yields. About reactions following the latter approach, we report in the present paper starting from different rhenium and technetium complexes with the metals in the oxidation states “+5” or “+6”.

2. Results and Discussion

2.1. Triphenylphosphine Complexes

Treatment of the rhenium(V) complex [ReNCl₂(PPh₃)₂] with an excess of disulfur dichloride in refluxing dichloromethane results in a subsequent dissolution of the sparingly soluble starting material and a dark-red, almost black solution is formed, from which the thionitrosyl complexes [Re(NS)Cl₃(PPh₃)₂] (1) and [Re(NS)Cl₃(PPh₃)(OPPh₃)] (2) can be isolated. The parallel formation of phosphine and phosphine oxide complexes is not completely surprising with respect to the oxidizing conditions in the reaction mixture and has been observed previously during similar reactions with [TcNCl₂(PMe₂Ph)₃] [74]. The two products can readily be separated during the crystallization procedure since compound 2 is highly soluble in acetone, while complex 1 is almost insoluble in this solvent. During the slow evaporation of CH₂Cl₂/acetone solutions of mixtures of 1 and 2, [Re(NS)Cl₃(PPh₃)₂] precipitates first as dark-red crystals, while the orange-yellow crystals of 2 do not deposit until almost all acetone is evaporated. The best method for the isolation of pure [Re(NS)Cl₃(PPh₃)₂], however, is the addition of an excess of PPh₃ during the synthesis, which obviously avoids the formation of significant amounts of the phosphine oxide complex (Scheme 1).

The infrared spectra of the complexes show the characteristic ν_(NS) stretches at 1213 cm^–1 (phosphine complex) and 1230 cm^–1 (phosphine oxide complex). These values are in agreement with such recorded for other rhenium complexes and those calculated for similar chromium(I) species [62]. The difference of approximately 20 cm^–1 reflects a stronger back-donation in compound 1 compared with 2. Since the oxidation state of rhenium in both compounds is “+2”, the observed difference indicates some changes in the coordination sphere of the transition metal. Particularly the electronic properties of the ligand in the trans position to the thionitrosyl ligand have a direct influence to the donor/acceptor behavior of NS⁺.

Single-crystal X-ray diffraction confirms that the oxidation of one of the PPh₃ ligands is accompanied with a ligand rearrangement and the OPPh₃ ligand in 2 occupies the position trans to NS⁺ instead of Cl⁻ in 1. The molecular structures of both complexes are depicted in Figure 2 and selected bond lengths and angles are summarized in

Table 1. Selected bond lengths (Å) and angles (°) in [Re(NS)Cl₃(PPh₃)₂] (1) and (b) [Re(NS)Cl₃(PPh₃)(OPPh₃)] (2).

	Re1–N10	N10–S10	Re1–Cl1	Re1–Cl2	Re1–Cl3	Re1–P1	Re1–P2/O1	O1–P2	Re1–N10–S10	Re1–O1–P2
1	1.795(5)	1.518(5)	2.353(1)	2.340(1)	2.416(1)	2.540(2)	2.555(2)	-	174.4(3)	-
2 (1)	1.77(1) 1.77(1)	1.53(1) 1.52(1)	2.368(4) 2.357(4)	2.255(3) 2.332(3)	2.338(3) 2.366(3)	2.506(4) 2.584(4)	2.100(8) 2.099(8)	1.509(9) 1.497(9)	167.2(7) 178.4(6)	155.7(6) 164.5(5)

⁽¹⁾ Values for two crystallographically independent molecules.

. It becomes evident that the thionitrosyl units are arranged linearly in both compounds with Re–N–S angles between 167° and 178°. Such a bonding situation has been found for all hitherto structurally characterized thionitrosyl complexes. Also, the direction of a formed phosphine oxide ligand in the trans position to a multiply bonded ligand is not without precedence and has been observed before during the formation of [Tc(NS)Cl₃(PMe₂Ph)(OPMe₂Ph) from cis,mer-[TcNCl₂(PMe₂Ph)₃] [74]. More generally, the presence of one triphenylphosphine oxide ligand in the coordination sphere of a transition metal favors the cis coordination to PPh₃, since the steric and electronic effects, which direct two PPh₃ ligands commonly into trans positions to each other, are significantly lowered.

[Re(NS)Cl₃(PPh₃)₂] (1) as well as [Re(NS)Cl₃(PPh₃)(OPPh₃)] (2) contain rhenium in the oxidation state “+2”. Due to the corresponding 5d⁵ “low spin” configuration, they are paramagnetic with one unpaired electron. This allows for the detection of resolved EPR spectra in solution. Figure 3 shows the spectra of both compounds at various temperatures. The spectra in liquid solutions (Figure 3a) consist of six hyperfine lines due to interactions of the unpaired electron with ^185,187Re, both having a nuclear spin of I = 5/2. Superhyperfine couplings due to interactions of the unpaired electron with the coordinating ³¹P (I = 1/2), ^35,37Cl (I = 3/2) or ¹⁴N (I = 1) nuclei could not be resolved. The frozen solution EPR spectra, which are shown in Figure 3b, confirm the presence of essentially axially symmetric, randomly oriented S = ½ spin systems with resolved parallel and perpendicular sets of ^185,187Re hyperfine lines. The analysis of the spectral parameters indicates the presence of a small rhombic component in the perpendicular part of the spectrum of complex 1, which has been taken into account during the simulation. The simulated spectra, which have been used to derive the spectral parameters, are depicted in the Supporting Material.

As already mentioned, couplings with the ³¹P nuclei of the coordinated phosphine ligands could not be resolved in the spectra of 1 and 2. This finding is in contrast to the previously studied technetium(II) complexes [Tc(NS)Cl₃(PMe₂Ph)₂] and [Tc(NS)Cl₃(PMe₂Ph)(OPMe₂Ph)], where the ⁹⁹Tc hyperfine signals of the parallel part show splittings into triplets (phosphine complex) or doublets (phosphine/phosphine oxide complex) and provide direct information for the occupation of their respective equatorial coordination spheres [74,77,85,86]. The larger line-widths of the EPR spectra obtained for the rhenium complexes prevent the resolutions of such couplings. Nevertheless, a remarkable trend in the EPR spectral parameters should be mentioned: the isotropic ^185,187Re coupling constants a₀^Re show a dependence on the number of chlorido ligands in the equatorial coordination sphere in a way that they increase from compound 1 (2 Cl⁻: 342 × 10^–4 cm^–1) via compound 2 (3 Cl⁻: 372 × 10^–4 cm^–1) to the [Re(NS)Cl₄]⁻ anion (4 Cl⁻: 457 × 10^–4 cm^–1 [66]). In the same sequence of complexes, the corresponding g₀ values decrease. Similar trends are also observed for the g values and coupling constants of the anisotropic spectra, which reflects considerable changes in the MO of the unpaired electron mainly due to the presence or absence of phosphine ligands.

2.2. Complexes with Chelating Ligands

Since the triphenylphosphine complexes 1 and 2 might have potential as precursors for the synthesis of more thionitrosyl complexes of rhenium, we tested some ligand exchange reactions with the chelating ligands shown in Figure 4. HEt₂btu is a potentially bidentate chelator. It readily deprotonates after the addition of a supporting base and stable complexes are known with almost all common transition metals. This also includes rhenium and technetium complexes with the metals in several oxidation states [87,88,89,90,91,92,93,94]. The tripodal, organometallic ligand {L^OMe}⁻ belongs to a classical family of tripodal ligands, which has been designed by Wolfgang Kläui and tested for a plethora of various applications [95,96]. Also, a number of rhenium complexes with this ligand are known and it is the first example of a chelating ligand, which forms stable technetium complexes with the metal in seven different oxidation states [54,95,96,97,98,99,100,101].

Reactions of [Re(NS)Cl₃(PPh₃)₂] with HEt₂btu in a mixture of CH₂Cl₂ and EtOH proceeded at room temperature (Scheme 2). After the addition of a small amount of Et₃N as a supporting base, the red reaction mixture immediately turned dark brown and a red-brown solid could be isolated. Single crystals suitable for X-ray diffraction were obtained by the slow evaporation of a CH₂Cl₂ solution. The deprotonation of HEt₂btu and its coordination as an S,O chelate is strongly indicated by the IR spectrum of the product by the disappearance of the NH stretches of the ligands and a strong bathochromic shift of the ν_C=O band by almost 200 cm⁻¹. This is typical for chelate-bonded R₂btu⁻ ligands [87]. Reduction of the metal ion is frequently observed during reactions with sulfur-containing ligands and, thus, the product of the described reaction with HEt₂btu is the diamagnetic rhenium(I) complex [Re(NS)(PPh₃)(Et₂btu)₂] (3a). This allows for the measurement of NMR spectra. They support the composition of the product as is performed by the mass spectrum.

The results of the spectroscopic studies are confirmed by a single-crystal structure analysis of the product. Figure 5 depicts the molecular structure of compound 3a, which shows the expected linear arrangement of the thionitrosyl ligand (

Table 2. Selected bond lengths (Å) and angles (°) [Re(NS)(PPh₃)(Et₂btu)₂] (3a).

Re1–N10	N10–S10	Re1–P1	Re1–S1	Re1–O1	Re1–S11	Re1–O11	O1–C1	C1–N1	N1–C2
1.749(4)	1.571(4)	2.362(1)	2.401(1)	2.101(3)	2.439(1)	2.103(4)	1.296(6)	1.301(7)	1.353(7)
C2–N2	C2–S1	O11–C11	C11–N11	N11–C12	C12–N12	C12–S11	Re1–N10–S10
1.319(7)	1.755(6)	1.266(6)	1.329(6)	1.338(7)	1.325(7)	1.758(6)	176.5(3)

). The equatorial coordination sphere of rhenium is occupied by an S,O chelate, a PPh₃ and the sulfur atom of the second Et₂btu⁻ ligand. Interestingly, the two Re–O single bonds are almost equal, which indicates similar structural trans influences induced by the NS⁺ and the PPh₃ ligands. The N–S distance in the rhenium(I) complex is markedly longer than in the rhenium(II) complexes 1 and 2, which might be a consequence of a stronger back-donation of the d⁶ metal ion into anti-bonding orbitals of the thionitrosyl ligand. Unfortunately, the corresponding IR stretch, which is normally more indicative, cannot be assigned unambiguously for 3a due to many overlapping bands in the respective range of the spectrum.

With respect to the clean reaction of [Re(NS)Cl₃(PPh₃)₂] with HEt₂btu and the ready formation of the thionitrosylrhenium(I) complex 3a, a surprising result was obtained for an analogous reaction of the corresponding technetium complex [Tc(NS)Cl₃(PPh₃)₂]. It proceeded under complete removal of the NS⁺ ligand and oxidation of the metal ion. The resulting technetium(III) tris complex [Tc(Et₂btu)₃] (Scheme 2) has been reported before (i) as the product of the direct reduction of pertechnetate with SnCl₂ in the presence of excess ligand [102], and (ii) by a subsequent reduction/ligand exchange procedure starting from (NBu₄)[TcOCl₄] [87]. The observed lability of the thionitrosyl unit is a considerable drawback for a potential use of 1 as precursor in ligand exchange procedures and is accompanied by another disappointing experience during attempted reactions with Na{L^OMe}. Although reactions of an entire series of different technetium and rhenium complexes with the “Kläui ligand” give well-defined and stable products [95,96,97,98,99,100,101], attempted reactions with 1 at room temperature did not proceed. Heating of such reaction mixtures finally gave a small amount of the chelate [Re(NS)Cl(PPh₃)(L^OMe)], but the yield was low and the product was accompanied by a number of side-products and impurities, which precluded the isolation of a pure compound in reasonable yields. Consequently, we preferred the second general approach to thionitrosyl compounds: the reaction of nitrido complexes with S₂Cl₂. Corresponding starting materials with {L^OMe}⁻ ligands, [MNCl(PPh₃)(L^OMe)] and [MNCl₂(L^OMe)] complexes with M = Re or Tc, are readily available from simple procedures, and the reactions with S₂Cl₂ (Scheme 3) give the products [M(NS)Cl(PPh₃)(L^OMe)]Cl (M = Re: 4a; M = Tc: 4b) and [M(NS)Cl₂(L^OMe)] (M = Re: 5a; M = Tc: 5b) in reasonable yields and good purities.

The reactions with all four starting complexes are restricted to the nitrido ligands and the remaining coordination spheres of the metals remain unchanged. Cationic thionitrosyls of rhenium(II) and technetium(II) are formed, when [M^VNCl(PPh₃)(L^OMe)] complexes (M = Re, Tc) are used as precursors. The products can be precipitated from the reaction mixture as yellow (Re compound, 4aCl) and red (Tc complex, 4bCl) chloride salts by the addition of n-hexane. Microcrystalline samples were obtained after recrystallization of CH₂Cl₂/n-hexane mixtures. For the rhenium complex, the PF₆⁻ salt was prepared by metathesis with AgPF₆. Orange-yellow needles of [Re(NS)Cl(PPh₃)(L^OMe)](PF₆), 4a(PF₆), were suitable for X-ray diffraction. The structure of the complex cation is shown in Figure 6a and selected bond lengths and angles are summarized in

Table 3. Selected bond lengths (Å) and angles (°) in [Re(NS)Cl(PPh₃)(L^OMe)](PF₆) (4a(PF₆)) and (b) [Tc(NS)Cl₂(L^OMe)] (5b).

	M1–N10	N10–S10	M1–Cl1	M1–Cl2	M1–P1	M1–O2	M1–O3	M1–O4	M1–N10–S10
4a	1.752(5)	1.554(5)	2.337(1)	-	2.475(1)	2.071(3)	2.032(3)	2.083(3)	175.4(3)
5b (1)	1.732(7) 1.75(1)	1.540(7) 1.54(1)	2.339(3) 2.330(3)	2.343(2) 2.344(3)	-	2.060(6) 2.068(5)	2.047(6) 2.057(6)	2.095(5) 2.088(5)	171.7(6) 172.0(8)

⁽¹⁾ Values for two crystallographically independent molecules.

. Expectedly, the {L^OMe}⁻ ligands act as tripods in both complexes, which directs the remaining three ligands into trans positions of their oxygen atoms. The M–N–S angles are close to 180° in both compounds.

Similar reaction patterns are observed for [ReNCl₂((L^OMe)] and [TcNCl₂((L^OMe)]. Both precursors contain the transition metals in their oxidation states “+6”, while the products [Re(NS)Cl₂(L^OMe)] (5a) and [Tc(NS)Cl₂(L^OMe)] (5b) are Re(II) and Tc(II) complexes (Scheme 3). An X-ray crystal structure determination has been performed for the technetium complex and its molecular structure is shown in Figure 6b. The general structural features found for the cation 4a also apply to technetium complex 5b (Table 3).

As compounds 1 and 2, the {L^OMe}⁻ complexes of the present study, are paramagnetic with a d⁵ “low spin” configuration, this results in S = ½ spin systems and allows us to record resolved EPR spectra in liquid and frozen solutions. The spectra have essentially axial symmetry with well-resolved ^185,187Re and ⁹⁹Tc hyperfine couplings in their parallel and perpendicular parts. Figure 7 shows exemplarily the spectra obtained for [Tc(NS)Cl(PPh₃)(L^OMe)]Cl (4bCl).

⁹⁹Tc has a nuclear spin of I = 9/2, which results in a ten-line pattern in the spectrum in liquid solution, while two sets of ten ⁹⁹Tc hyperfine lines (each one in the parallel and one in the perpendicular part of the spectrum) are characteristic for the anisotropic frozen-solution spectrum as can be described by the Spin Hamiltonian (1), where g_‖, g_⊥, A_‖^M. and A_⊥^M are the principal values of the ⁹⁹Tc and ^185,187Re hyperfine tensors A^M. The spectral parameters are summarized in

Table 4. EPR parameters of the Re(II) and Tc(II) complexes with {L^OMe}⁻ ligands, coupling constants are given in 10^–4 cm^–1.

	g0	a0M	g‖	g⊥	A‖M	A⊥M
[Re(NS)Cl(PPh3)(LOMe)]Cl (4aCl)	1.995	420	1.842	1.986	644	330
[Tc(NS)Cl(PPh3)(LOMe)]Cl (4bCl)	2.007	168	1.966	2.023	252	112
[Re(NS)Cl2(LOMe)] (5a)	1.990	455	1.945	1.760	787	398
[Tc(NS)Cl2(LOMe)] (5b)	1.996	177	1.943	2.021	285	129

.

$${\widehat{H}}_{sp}={\beta }_e\left[g_{\Vert }{B}_z{\widehat{S}}_z+g_{\perp }\left(B_x{\widehat{S}}_x+B_y{\widehat{S}}_y\right)\right]+A_{\Vert }^M{\widehat{S}}_z{\widehat{I}}_z+A_{\perp }^M({\widehat{S}}_x{\widehat{I}}_x+{\widehat{S}}_y{\widehat{I}}_y)$$

(1)

Superhyperfine interactions due to couplings of the unpaired electron with the ³¹P nuclei of the PPh₃ ligands in compounds 4 are not visible. This prevents us from obtaining direct information about the extent of spin delocalization into the ligand orbitals. Indirect information, however, can be deduced from the ^185,187Re coupling parameters. As described for compounds 1 and 2, also for the {L^OMe}⁻ thionitrosyl complexes, the isotropic and anisotropic A^Tc and A^Re coupling parameters decrease, when phosphine ligands are present in the equatorial coordination spheres of the metals instead of Cl⁻. This can be understood by the transfer of electron density to the π-acceptor PPh₃. The experimental line-widths of the spectra, however, do not allow for the resolution of such couplings for the complexes of the present study.

3. Materials and Methods

[ReNCl₂(PPh₃)₂] [103], [TcNCl₂(PPh₃)₂] [104], (NBu₄)[TcNCl₄] [105], (NBu₄)[ReNCl₄] [106], [ReNCl(PPh₃)(L^OMe)] [98], [TcNCl(PPh₃)(L^OMe)] [58], [ReNCl₂(L^OMe)] [107], [TcNCl₂(L^OMe)] [58], [Tc(NS)Cl₃(PPh₃)₂] [108] and HEt₂btu [109] were prepared according to literature procedures. All other chemicals were reagent-grade and purchased commercially. Reactions with air- or moisture-sensitive compounds were performed with the standard Schlenk technique. A laboratory approved for the handling of radioactive material was used for the synthesis of the technetium compounds. Normal glassware gives adequate protection against the weak beta radiation of ⁹⁹Tc as long as small amounts of this isotope as in the present paper are used.

3.1. Syntheses

[Re(NS)Cl₃(PPh₃)₂] (1): [ReNCl₂(PPh₃)₂] (2 g, 2.6 mmol) was suspended in 40 mL CH₂Cl₂ and an excess of S₂Cl₂ (0.5 mL, 6.26 mmol) and PPh₃ (680 mg, 5.2 mmol) was added. The mixture was heated under reflux for approximately 30 min, which resulted in a slow dissolution of the starting material and the formation of a dark-red solution. The progress of the reaction can be controlled by TLC. After a complete consumption of [ReNCl₂(PPh₃)₂], the solvent and the remaining S₂Cl₂ were removed in vacuum. The resulting solid was crystallized from a CH₂Cl₂/acetone (1:1) mixture giving dark-red crystals. Yield: 1.49 g (66%). Elemental analysis: Calcd. For C₃₆H₃₀Cl₃NP₂ReS: C, 50.1; H, 3.5; N, 1.6; S, 3.7%. Found: C 49.8; H, 3.5; N, 1.6; S, 4.0%. IR (ATR, cm^–1): 3057 (w), 1981 (w), 1670 (w), 1586 (w), 1574 (w), 1482 (s), 1434 (s), 1316 (m), 1213 (s), 1187 (s), 1159 (m), 1089 (s), 1074 (m), 1028 (m), 998 (m), 928 (w), 855 (w), 758 (m), 744 (s), 711 (m), 692 (s), 617 (w). EPR (77 K, CH₂Cl₂): g_x = 2.110, g_y = 1.991, g_z = 1.943; A_x^Re = 281 × 10^–4 cm^–1, A_y^Re = 283 × 10^–4 cm^–1, A_z^Re = 524 × 10^–4 cm^–1. EPR (RT, CH₂Cl₂): g₀ = 2.025; a₀^Re = 342 × 10^–4 cm^–1. ESI⁺ MS (m/z): 885.002 [M+Na]⁺ (calcd. 885.009), 900.975 [M+K]⁺ (calcd. 900.983).

[Re(NS)Cl₃(PPh₃)(OPPh₃)] (2): The product was obtained as a side product of the synthesis of [Re(NS)Cl₃(PPh₃)₂] when no additional PPh₃ was added. It could be isolated as an orange-red solid following compound 1 during the crystallization procedure. Yield: 0.6 g (26%). IR (ATR, cm^–1): 3056 (w), 1589 (w), 1482 (m), 1435 (s), 1231 (s), 1188 (w), 1140 (vs), 1119 (vs), 1026 (m), 998 (m), 924 (w), 852 (w), 747 (s), 723 (vs), 690 (vs). EPR (77 K, acetone): g_‖ = 1.923, g_⊥ = 2.008; A_‖^Re = 594 × 10^–4 cm^–1, A_⊥^Re = 293 × 10^–4 cm^–1. EPR (RT, acetone): g₀ = 2.017; a₀^Re = 372 × 10^–4 cm^–1.

[Re(NS)(PPh₃)(Et₂btu)₂] (3a): [Re(NS)Cl₃(PPh₃)₂] (100 mg, 0.1 mmol) was dissolved in 30 mL CH₂Cl₂ and a solution of HEt₂btu (60 mg, 0.25 mmol) in 20 mL Ethanol containing 3 drops of triethylamine was added. The mixture was stirred at room temperature for 30 min. During this time, the color changed to pale brown. The volatiles were removed in vacuum and the remaining solid was subsequently washed with cold methanol, diethyl ether and n-hexane. The raw product was dissolved in CH₂Cl₂ for crystallization. Slow evaporation of the solvent gave red-brown crystals. Yield: 31 mg (25%). Elemental analysis: Calcd. for C₄₂H₄₅N₅O₂PReS₃: C, 52.3; H, 4.7; N, 7.3; S, 10.0%. Found: C 52.0; H, 4.9; N, 6.8; S, 10.2%. IR (ATR, cm^–1): 3058 (w), 2972 (w), 2927 (w), 1599 (w), 1587 (w), 1528 (m), 1499 (s), 1446 (m), 1416 (s), 1394 (s), 1353 (s),1311 (m), 1294 (m), 1251 (s),1210 (w), 1140 (m), 1092 (m), 1071 (m), 1025 (m), 997 (m), 895 (m), 883 (m), 827 (m), 797 (m), 752 (w), 742 (m), 702 (s), 692 (s), 671 (m), 661 (m), 619 (w). ¹H NMR (400 MHz, CDCl₃, ppm): 8.10 (d, J = 7.6 Hz, 2H, Ph), 7.65 (d, J = 7.3 Hz, 6H, Ph), 7.37–7.13 (m, 15H), 7.01 (t, J = 7.9 Hz, 2H, Ph), 4.37 (dq, J = 14.2, 7.1 Hz, 1H, CH₂), 4.14–3.80 (m, 4H, CH₂), 3.73 (q, J = 7.3, 6.8 Hz, 1H, CH₂), 3.62 (q, J = 6.3 Hz, 1H, CH₂), 3.51 (dd, J = 12.9, 6.5 Hz, 1H, CH₂), 1.52 (s, 2H), 1.39 (dd, J = 9.6, 4.2 Hz, 3H, CH₃), 1.27 (q, J = 12.3, 9.7 Hz, 3H, CH₃), 1.03 (t, J = 6.5 Hz, 3H, CH₃), 0.90 (d, J = 6.3 Hz, 3H, CH₃). J = 8.0 Hz, 12H, CH₃). ³¹P NMR (CDCl3, ppm): 21.67 (s). ESI⁺ MS (m/z): = 965.237 [M]⁺ (calcd. 965.203) 988.225 [M+Na]⁺ (calcd. 988.192), 1004.202 [M+K]⁺ (calcd. 1004.166).

[Re(NS)Cl(PPh₃)(L^OMe)]Cl (4aCl): [ReNCl(PPh₃)(L^OMe)] (250 mg, 0.25 mmol) was suspended in 30 mL CH₂Cl₂, an excess of S₂Cl₂ (0.1 mL) was added and the mixture was heated under reflux for 30 min. The solvent was reduced to 15 mL and n-hexane (150 mL) was added. The resulting precipitate was washed with hexane and diethyl ether and dried. Recrystallization from a CH₂Cl₂/n-hexane (2:1) mixture gave a pale-yellow solid. Yield: 211 mg (83%). Elemental analysis: Calcd. for C₂₉H₃₈Cl₂CoNO₉P₄ReS: C, 34.3; H, 3.8; N, 1.4; S, 3.2%. Found: C 34.0; H, 4.0; N, 1.4; S, 3.1%. IR (ATR, cm^–1): 3056 (w), 2992 (w), 2948 (w), 1634 (w), 1483 (w), 1457 (w), 1436 (m), 1223 (m), 1177 (m), 1036 (vs), 1014 (vs), 943 (s), 912 (s), 853 (s), 793 (s), 746 (vs), 693 (s), 599 (vs). EPR (77 K, CH₂Cl₂): g_‖ = 1.842, g_⊥ = 1.986; A_‖^Re = 644 × 10^–4 cm^–1, A_⊥^Re = 330 × 10^–4 cm^–1. EPR (RT, CHCl₃): g₀ = 1.995; a₀^Re = 420 × 10^–4 cm^–1. ESI⁺ MS (m/z): 981.013 [M]⁺ (calcd. 980.979).

[Re(NS)Cl(PPh₃)(L^OMe)](PF₆) (4a(PF₆)): [Re(NS)Cl(PPh₃)(L^OMe)]Cl (50 mg, 0.05 mmol) was dissolved in acetonitrile and an excess of AgPF₆ (19 mg, 0.075 mmol) was added. The solution was stirred at room temperature and the formed colorless precipitate was removed by filtration. Orange-yellow needles suitable for X-ray diffraction formed upon the slow evaporation of a solution of the complex in CH₂Cl₂/n-hexane. Yield: 52 mg (93%). The essential spectroscopic data are identical with those of 4aCl.

[Re(NS)Cl₂(L^OMe)] (5a): [ReNCl₂(L^OMe)] (34 mg, 0.05 mmol) was dissolved in 5 mL CH₂Cl₂ and 3 drops of S₂Cl₂ were added. The mixture was stirred at room temperature. The progress of the reaction can readily be monitored by EPR spectroscopy and typically after 15 min the signals of the starting material have disappeared. The solvent and residual S₂Cl₂ were removed in vacuum and the solid residue was washed subsequently with n-hexane and diethyl ether. The product was recrystallized from CH₂Cl₂/diethyl ether giving a pale-brown, microcrystalline solid. Yield: 20 mg (53%). IR (ATR, cm^–1): 3111 (w), 2958 (m), 2874 (w), 1460 (w), 1425 (w), 1381 (w), 1260 (w), 1222 (s), 1175 (s), 1012 (vs), 945 (w), 860 (m), 796 (vs), 743 (s), 595 (m). EPR (77 K, CH₂Cl₂): g_‖ = 1.945, g_⊥ = 1.760; A_‖^Re = 787 × 10^–4 cm^–1, A_⊥^Re = 398 × 10^–4 cm^–1. EPR (RT, CHCl₃): g₀ = 1.990; a₀^Re = 455 × 10^–4 cm^–1. ESI⁺ MS (m/z): 776.846 [M+Na]⁺ (ber.: 776.847); 792.821 [M+K]⁺ (ber.: 792.821).

[Tc(NS)Cl(PPh₃)(L^OMe)]Cl (4bCl): [TcNCl(PPh₃)(L^OMe)] (72 mg, 0.08 mmol) was dissolved in 2 mL CH₂Cl₂ (2 mL) and S₂Cl₂ (11 mg, 0.08 mmol) was added under permanent stirring. The stirring was continued at room temperature for 1 h and the obtained dark-red mixture was filtered to remove some insoluble solid (most probably elemental sulfur). All volatiles were removed in vacuum and the remaining solid crystallized from a CH₂Cl₂/n-pentane (1:1) mixture. Storing such a mixture in a refrigerator for several days gave a red, microcrystalline powder. Yield: 11 mg (15%). IR (KBr, cm^–1): 2910 (vs), 2723 (s), 1726 (m), 1630 (w) 1479 (vs), 1381 (vs), 1227 (vs), 1171 (m), 1040 (w), 949 (s), 928 (s), 851 (w), 802 (w), 723 (m), 610 (w), 542 (m). EPR (77 K, CH₂Cl₂): g_‖ = 1.966, g_⊥ = 2.023; A_‖^Tc = 252 × 10^–4 cm^–1, A_⊥^Tc = 112 × 10^–4 cm^–1. EPR (RT, CH₂Cl₂): g₀ = 2.007; a₀^Tc = 168 × 10^–4 cm^–1.

[Tc(NS)Cl₂(L^OMe)] (5b): [TcNCl₂(L^OMe)] (41 mg, 0.06 mmol) was dissolved in CH₂Cl₂ (10 mL) and treated with an excess of S₂Cl₂ (0.3 mL). After stirring for 30 min at room temperature, all volatiles were removed in vacuum. The remaining solid was washed with n-hexane and diethyl ether and dissolved in 1 mL CH₂Cl₂. Overlayering with n-hexane and storing in a refrigerator for several days gave pale-red crystals. Yield: 32 mg (75%). IR (KBr, cm^–1): 3107 (w), 3086 (w), 2990 (w), 2841 (w), 1462 (w), 1423 (w), 1310 (vs), 1236 (vs), 1182 (vs), 1128 (vs), 1098 (vs), 986 (vs), 847 (m), 795 (s), 747 (s), 638 (m), 608 (s), 530 (w), 476 (w), 426 (w). EPR (77 K, CH₂Cl₂): g_‖ = 1.943, g_⊥ = 2.021, A_‖^Tc = 285 × 10^–4 cm^–1, A_⊥^Tc = 129 × 10^–4 cm^–1. EPR (RT, CH₂Cl₂) g₀ = 1.996; a₀^Tc = 177 × 10^–4 cm^–1.

3.2. Spectroscopic and Analytical Methods

Elemental analyses of carbon, hydrogen, nitrogen and sulfur were determined using a Heraeus vario EL elemental analyzer. IR spectra were measured as KBr pellets on a Shimadzu IR Affinity-1 (technetium compounds) or a Thermo Scientific Nicolet iS10 ATR spectrometer (rhenium complexes). The NMR spectra were recorded on JEOL ECS-400 or ECZ-400 400 MHz spectrometers. ESI TOF mass spectra were measured with an Agilent 6210 ESI TOF (Agilent Technologies, Santa Clara, CA, USA). X-Band EPR spectra were recorded in solution with a Magnettech Miniscope MS400 spectrometer at 300 and 77 K. Simulation and visualization of the EPR spectra were conducted with the EasySpin tool box in MatLab [110,111].

3.3. X-ray Crystallography

The intensities for the X-ray determinations were collected on STOE IPDS-2T or Bruker CCD instruments with Mo/Kα radiation. The various temperatures applied are due to the experimental setup of the different diffractometers. Semi-empirical or numerical absorption corrections were carried out by the SADABS or X-RED32 programs [112,113]. Structure solution and refinement were performed with the SHELX programs [114,115] included in the WinGX [116] program package or OLEX2 [117]. Hydrogen atoms were calculated for idealized positions and treated with the “riding model” option of SHELXL. The solvent mask option of OLEX2 was applied to treat diffuse electron density due to disordered CH₂Cl₂ in the crystals of [Re(NS)Cl₃(PPh₃)(OPPh₃)] (2). Details are given in the Supplementary Materials. The representation of molecular structures was conducted using the program DIAMOND [118].

4. Conclusions

Reactions of rhenium and technetium nitrido compounds with disulfur dichloride give access to novel thionitrosyl complexes of these elements having monodentate and chelating ligands in their coordination spheres. The new compounds are chemically related to corresponding nitrosyl complexes but do not possess their robustness and stability. Their thermal instability and the lack of a suitable one-step synthesis starting from pertechnetate set narrow limits for potential applications in nuclear medical procedures, as is discussed in the introduction.