Accepted Manuscript Title: Metal-Radical Coordination Complexes of Thiazyl and Selenazyl Ligands Author: Kathryn E. Preuss PII: DOI: Reference:
S0010-8545(14)00257-4 http://dx.doi.org/doi:10.1016/j.ccr.2014.09.016 CCR 111932
To appear in:
Coordination Chemistry Reviews
Received date: Revised date: Accepted date:
30-6-2014 15-9-2014 22-9-2014
Please cite this article as: K.E. Preuss, Metal-Radical Coordination Complexes of Thiazyl and Selenazyl Ligands, Coordination Chemistry Reviews (2014), http://dx.doi.org/10.1016/j.ccr.2014.09.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Ac ce p
te
d
M
an
us
cr
ip t
MetalRadical Coordination Complexes of Thiazyl and Selenazyl Ligands Kathryn E. Preuss Department of Chemistry, University of Guelph, Guelph, ON, Canada, N1G 2W1. Tel: 1‐(519) 824‐4120 Email:
[email protected] Abstract The so‐called “metal‐radical approach” to the design of molecule‐based magnetic materials relies on the availability of paramagnetic ligands that can reliably form coordination complexes with paramagnetic metal ions. While the most common radical ligands are based on nitroxides, other paramagnetic building blocks are also gaining attention. Thiazyls and their related selenazyls are promising candidates for the development of radical ligands and, because they have inherently interesting materials properties, their use as ligands has the potential to generate novel materials with unprecedented properties. Significant progress has been made in this field in the past ten years. This review is a timely look back at the development of the field, highlighting the diversity of complexes and designs, and a look forward to a promising future. Keywords: thiazyl; selenazyl; radical ligand; coordination; magnetism; materials Table of Contents 1. Introduction 2. Complexes of 1,2,3,5‐Dithiadiazolyl (DTDA) and 1,2,3,5‐Diselenadiazolyl (DSDA) 2.1 Scoordination Complexes 2.2 π Complexes 2.3 Ncoordination Complexes 2.3.1 Ligand Design 2.3.2 Transition Metal Complexes 2.3.3 Lanthanide Complexes 3. Complexes of 1,3,2‐Dithiazyl (132DTA) 4. A Complex of 1,2,3‐Dithiazyl (123DTA) 5. Complexes of 1,2,4,6‐Thiatriazinyl (TTA) 5.1 Scoordination and π Complexes 5.2 Ncoordination Ligand Design 6. Complexes of 1,2,5‐Thiadiazole (125TDA) and Related Radical Anions 7. Concluding Remarks 8. Acknowledgements 9. References
1
Page 1 of 48
Ac ce p
te
d
M
an
us
cr
ip t
1. Introduction Thiazyls are an interesting class of compound that contain the unsaturated, odd‐ electron –S=N‐ linkage. The documented study of thiazyls dates back at least as far as 1835, to the preparation of S4N4 from the reaction of sulfur monochloride with ammonia [1]. Although S4N4 is shock‐sensitive and can detonate to generate elemental sulfur and nitrogen, controlled decomposition of S4N4 generates poly(sulfur nitride), (SN)x, a fascinating bronze colored material [2] that exhibits metallic [3] and superconducting [4] properties. Thus, in only a few steps from very basic reagents, a thiazyl compound with impressive materials properties can be prepared. With renewed interest in (SN)x in the 1970s came a renaissance in thiazyl chemistry [5]. In particular, the propensity of thiazyls to form isolable paramagnetic species [6] drew attention from theoreticians and synthetic chemists alike. A diverse variety of paramagnetic thiazyl heterocycles can be prepared and stored as pure materials for prolonged periods. The caveat is that many thiazyl species are susceptible to hydrolysis and/or reaction with atmospheric oxygen [7‐9], thus inert atmosphere and anhydrous reaction/storage conditions may be required. In the last fifty years, an extensive body of work has been developed, with ever more challenging designs and synthetic methodologies, and ever more exciting materials properties reported. Examples of the wide range of impressive properties exhibited by thiazyl‐based molecular materials include magnetic ordering as a spin‐canted antiferromagnet at 36 K [10], ferromagnetic ordering at 1.32 K [11], magnetic bistability with a large hysteresis centered around room temperature [12], and metallic conductivity under pressure [13]. By incorporating selenium in the place of sulfur, making structurally analogous selenazyl and thiaselenazyl molecular materials, improved and altered materials properties have been reported, including ferromagnetic ordering temperatures as high as 17.5 K [14] and hysteretic spin‐ crossover [15]. A key structural feature of most thiazyl radicals, strongly correlated to the resulting materials properties, is the high spin density at S and N atoms that are devoid of substituents (Fig. 1). These heteroatoms are free to engage in supramolecular contacts, directing the crystal packing in the solid structure and providing the primary pathways for magnetic and conductive properties. They are also free to participate in chemical reactions, such as hydrolysis [7, 16] and oxidation with O2 [9, 17]. Thus, the structural features that give rise to the most interesting and technologically promising properties of thiazyl radicals are the same features that impart air‐sensitivity, which can make working with these species somewhat challenging. It should be noted, however, that some of the more recently designed paramagnetic thiazyls are air‐stable [18, 19], a welcome advancement in this field. In addition, the network of intermolecular interactions that gives rise to supramolecular materials properties also contributes to the lattice enthalpy in the solid state, decreasing the solubility and volatility of these species, a second challenge which may be overcome by judicious choice of solvent or solvent mixtures and molecular design. Although the reactivity of many thiazyls can be a disadvantage, it is this very feature that may also be of interest. Apart from hydrolysis and oxidation, reactivity
2
Page 2 of 48
Ac ce p
te
d
M
an
us
cr
ip t
with other reagents such as atomic nitrogen [20] and metal‐containing species (vide infra) have been reported. The present review is primarily concerned with the latter, the reactivity of thiazyl and selenazyl radicals with metal‐based reagents and the resulting coordination complexes. Reasons for exploring the reactivity of thiazyl and selenazyl radicals with metal‐based reagents can be roughly divided into two categories: 1) elucidation of the nature and reactivity of this interesting family of paramagnetic compounds and 2) development of paramagnetic ligands that will remain intact upon coordination, generating novel molecular materials with enhanced or unprecedented properties. The former is primarily concerned with the chemistry of thiazyl ring systems, viewing them as a synthetic precursor with the potential to contribute to therapeutic, pharmaceutical or pesticide related applications. Low valent metal components (0 or +1 oxidation states) are typically employed and reaction occurs at the S atoms. In some cases, the heterocycles remain intact, but it is also common for bonds to be broken, possibly with the abstraction of atoms from the system. The latter approach seeks to advance the so‐called “metal‐radical approach” [21] to molecular materials design by exploiting the unique intermolecular interactions available with thiazyl species, thereby adding another facet to the electronic and magnetic properties. Thus, the real advantage of thiazyl‐based ligands compared to other paramagnetic ligands lies in the potential for intermolecular interactions. The N atoms are viewed as σ‐donor coordination sites and harder metal ions (+2 or +3 oxidation state) are typically employed. Ultimately, both approaches have generated valuable insight. As recently as ten years ago, metal‐thiazyl coordination chemistry was still in its infancy and progress in the area was documented in a limited number of reviews [22‐24]. The past decade has witnessed acceleration in the growth and progress of this field, with increased diversity in designs and results. This is a good time to take stock of what has been achieved and look forward to the future. With this in mind, the present review is a timely overview of the field and its current status with the hope that new directions will be encouraged and that the next decade is as exciting as the past ten years have been. [PLACE FIGURE 1 HERE] Figure 1. Generic structures and numbering schemes for the neutral thiazyl heterocycles discussed herein; surfaces of singly occupied molecular orbitals (SOMOs) and total Mulliken spin density (blue = α; green = β) for the thiazyls (R =H), generated from DFT calculations using b3lyp/6‐31G(d,p) [25] and rendered using isovalues of 0.020000 for the MOs and 0.000400 for densities [26]. 2. Complexes of 1,2,3,5Dithiadiazolyl (DTDA) and 1,2,3,5Diselenadiazolyl (DSDA) The 1,2,3,5‐dithiadiazolyl (DTDA) heterocycles and their selenium analogues (DSDA) are arguably the most intensely and thoroughly studied paramagnetic thiazyls. This is, in part, due to valuable and reliable synthetic methodologies that
3
Page 3 of 48
Ac ce p
te
d
M
an
us
cr
ip t
have made it possible to synthesize RDTDA species with a wide range of R substituents (Fig. 1) [27‐33]. As such, it is not surprising that DTDAs were the first class of thiazyl radical to be investigated with respect to metal coordination (vide infra) as well as the species for which the largest number of metal‐thiazyl coordination complexes having been reported to date. 2.1 Scoordination Complexes In 1989, Banister et al. reported the first metal coordination complex of a DTDA, prepared by reacting PhDTDA with either Fe2(CO)9 or Fe3(CO)12 [34]. The product of these reactions was originally assigned as Fe 2(CO)6(PhCN2S2), wherein (CO)3Fe‐Fe(CO)3 is inserted into the S‐S bond of the PhDTDA radical. It was later discovered that the orange product is actually the related diamagnetic imine Fe2(CO)6PhDTDAH (Fig. 2), bearing a hydrogen atom at one of the thiazyl N sites [22, 35, 36]. Reports of reactions between PhDTDA and a variety of other low valent metal precursors soon followed, illustrating a diverse series of S‐coordination motifs. The Ni2Cp2PhDTDA complex is structurally analogous to the original di‐iron complex, and the species remains paramagnetic (Cp = cyclopentadienyl) [37]. Reaction of PhDTDA with Pt(PPh3)3 generates a paramagnetic complex in which a Pt(PPh3)2 fragment is inserted into the S‐S bond, [Pt(PPh3)2]PhDTDA (Fig. 3) [38]. Spectroscopic evidence indicates a similar structure for [Pd(dppe)2]PhDTDA (dppe = 1,2‐bis(diphenylphosphino)ethane) [39]. The [Pt(PPh3)2]PhDTDA complex decomposes in solution over time to generate a trinuclear platinum species, [Pt(PPh3)2]PhDTDAPtPhDTDA[Pt(PPh3)2] (Fig. 3) [38]. Reaction of RDTDA (R = Ph, 3′‐py, 4′‐py) with Pd(PPh3)4 apparently does not generate a mononuclear species that is stable enough to be isolated, and only trinuclear palladium species, [Pd(PPh3)2]RDTDAPdRDTDA[Pd(PPh3)2], have been reported [39, 40]. In 2000, Wong et al. expanded on the reactivity of DTDA radicals with low valent metal reagents, exploring the reactivity of RDTDA (R = 3′‐py, 4′‐py) with both Pd(PPh3)4 and Mn(CO)5Br [41]. A one‐dimensional coordination polymer of trinuclear [Pd(PPh3)2]4′pyDTDAPd4′pyDTDA [Pd(PPh3)2] units bridged by Mn(CO)3Br at the pyridyl N atoms is among the interesting species reported. [PLACE FIGURE 2 HERE] Figure 2. Line drawing of PhDTDA and Fe and Ni S‐coordination complexes; image of Ni2Cp2PhDTDA generated from the cif file as found in the Cambridge Structural Database (CSD Refcode: KIPYEN) [37]. [PLACE FIGURE 3 HERE] Figure 3. Line drawing and crystal structure of [Pt(PPh3)2]PhDTDA (CSD Refcode: WICXAH); line drawing of the decomposition product [Pt(PPh3)2]PhDTDAPt PhDTDA[Pt(PPh3)2] [38].
4
Page 4 of 48
Ac ce p
te
d
M
an
us
cr
ip t
In 1994, Rawson et al. reported the first metal complex of a DSDA radical [42]. Reaction of Pt(PPh3)3 with PhDSDA was monitored via EPR spectroscopy. Decreasing intensity of the EPR spectrum, accompanied by color changes reminiscent of those observed for the analogous PhDTDA reaction, suggest that the same type of trinuclear platinum species is formed from rapid decomposition of an initial mononuclear species. Convincing crystallographic evidence for the analogous palladium trinuclear complex, [Pd(PPh3)2]PhDSDAPdPhDSDA[Pd(PPh3)2], followed shortly thereafter [43]. In 1998, Rawson et al. reported two interesting complexes formed from the decomposition of the reaction product of either PhDTDA or PhDSDA with Pt(dppe)2 in chlorinated organic solvents [44]. Strictly speaking, the resulting 10π metallocycles are not coordination complexes of DTDA/DSDA radicals, however the unique [PhC(NH)2EPt(dppe)][Cl] species (E = S, Se) are worth noting in the present review (Fig. 4). [PLACE FIGURE 4 HERE] Figure 4. Line drawing and crystal structure of [PhC(NH)2SePt(dppe)]+ (CSD Refcode: HIBVOD) [44]. 2.2 π Complexes In 2006, Boeré, Goh, et al. reported a new coordination motif for DTDA radicals [45]. Reaction of RDTDA (R = 4‐MePh, 4‐ClPh, 3‐CN‐5‐tBuPh) with [CpCr(CO)3]2, which readily dissociates into the 17‐electron [CpCr(CO)3]• radical in solution, generates η2 π complexes in which the DTDA heterocycles remain intact. The CpCr(CO)2 fragment is coordinated to both S atoms of the DTDA ring without cleavage of the S‐S bond. Both endo and exo isomers of the resulting diamagnetic complexes are formed (Fig. 5). In 2008, the series was expanded further to include R = 4‐MeOPhe and 4‐F3CPh [46]. [PLACE FIGURE 5 HERE] Figure 5. Line drawings and crystal structures of representative examples of endo and exo coordination in the series of Cr(Cp)(CO)2RDTDA complexes (CSD Refcodes: SEKDOC and SEKDIW) [45]. 2.3 Ncoordination Complexes In 2004, Preuss et al. reported the first N‐coordination complex of a DTDA radical [47]. The 2′‐pyridyl‐1,2,3,5‐dithiadiazolyl radical (pyDTDA) was designed as a chelating ligand, intended to remain paramagnetic, with little disruption to the thiazyl heterocycle upon coordination. This ligand, and several others, has since been employed in developing coordination complexes of a variety of transition metals and lanthanides. The N‐coordinating DTDA and DSDA ligands designed and reported to date, their coordination complexes, and materials properties are presented and discussed in subsections 2.3.1 ‐2.3.3.
5
Page 5 of 48
Ac ce p
te
d
M
an
us
cr
ip t
2.3.1 Ligand Design There are several synthetic routes in the literature for the preparation of 4‐ R‐1,2,3,5‐dithiadiazolyl radicals (RDTDAs) with a wide variety of R groups (e.g., R = Ph, tBu, H, NMe2, F, CF3, etc.) [27, 30, 48‐50]. Common to all these routes is the preparation and reduction of a closed‐shell cationic 1,2,3,5‐dithiadiazolium precursor ([DTDA+][X‐], where X is usually Cl) to generate the desired radical. In order to access the closed‐shell cation [DTDA+][X‐], the most widely applicable methods start with the R‐CN nitrile as a precursor [29, 32, 33, 51]. Essentially, so long as the R‐CN starting material can be made or acquired, the related R‐DTDA radical can nearly always be prepared. This is one feature that makes the DTDA heterocycle an ideal building block for the design of paramagnetic ligands. Another useful feature is that most DTDAs can be readily sublimed on a preparative scale and are thus easily obtained as pure materials in sufficient quantities to be used as off‐the‐shelf ligands. This thermal stability, as well as modest solubility in a variety of aprotic polar and non‐polar solvents (e.g., acetonitrile, methylene chloride, toluene, THF), facilitates the coordination chemistry of DTDA‐based ligands. The DTDA heterocycle has one substituent, at the C atom (position 4). This position is ortho to the two N atoms (positions 3 and 5), thus a natural design strategy for creating an N‐coordinating ligand from a DTDA radical is to use substituents with σ‐donor heteroatoms, such as 2‐pyridyl or 2‐pyrimidyl. In this way, bidentate chelating ligands pyDTDA and pymDTDA respectively (Fig. 6), were designed to be paramagnetic analogs of 2,2′‐bipyridine and 2,2′‐bipyrimidine [47, 52]. The pyridyl and pyrimidyl heterocycles can be decorated with groups such as halides and nitriles in order to promote electrostatic contacts in the solid state and manipulate the crystal packing of the complexes. The 4′ and 5′cyanopyDTDA and the 5′bromopyDTDA ligands, and some of their transition metal complexes, have been reported [53]. Ligand designs need not be restricted to the use of pyridyl and pyrimidyl substituents. Although N atoms typically function to provide a better σ‐donor site than O‐atoms or S‐atoms, it is reasonable to imagine N‐coordinating DTDA ligands with furan, thiophene and other comparable heterocyclic substituents. Indeed, the 4‐(2′‐cyanofur‐5′‐yl)‐1,2,3,5‐dithiadiazolyl (cfDTDA) and 4‐(benzoxazol‐2′‐yl)‐ 1,2,3,5‐dithiadiazolyl (boaDTDA) ligands and many of their complexes have also been reported [54, 55]. With the idea that the DTDA moiety can theoretically be used in place of a pyridyl or pyrimidyl ring, many other paramagnetic analogs of known chelating ligands are also possible. For example, the 4,6‐bis(1,2,3,5‐dithiadiazolyl)pyrimidine (DTDA2pym) ligand is a recently reported biradical analog of 4,6‐di(pyridin‐2‐ yl)pyrimidine [56]. With a triplet ground state (S = 1), this paramagnetic species has been employed as a bis(bidentate) ligand (Fig. 6). [PLACE FIGURE 6 HERE]
6
Page 6 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6. DTDA‐based ligands for which N‐coordinating metal complexes have been reported 2.3.2 Transition Metal Complexes Ultimately, the purpose behind developing a research program based on the design of paramagnetic ligands from thiazyl building blocks is not to simply imitate what has been previously accomplished with other known paramagnetic ligands, such as nitronyl nitroxides [57‐61] and oxoverdazyls [62‐65]. Rather, it is to harness the unique and impressive materials properties of thiazyl radicals that arise from supramolecular contacts in the solid state (see Section 1). Nevertheless, the first steps in developing this program necessitate a proof‐of‐concept. It must first be shown that thiazyl heterocycles can be employed as building blocks for paramagnetic ligands, that they can N‐coordinate to metal ions such that the thiazyl ring remains intact, and that the magnetic exchange coupling between the coordinated paramagnetic metal ion and the thiazyl‐based radical ligand can be understood and predicted based on a simple orbital overlap model. To this end, the first examples of metal‐radical coordination complexes using DTDA‐based paramagnetic ligands were simply intended to demonstrate these attributes. The bidentate pyDTDA ligand was coordinated to the Co(hfac)2 fragment and structural and magnetic properties of the sublimed, crystalline complex (Fig. 7) were reported (where hfac = 1,1,1,5,5,5‐hexafluoroacetylacetonato‐) [47]. [PLACE FIGURE 7 HERE] Figure 7. Line drawing and crystal structure of Co(hfac)2(pyDTDA) (CSD Refcode: SAGMUJ) [47]. There are number of points regarding the Co(hfac)2(pyDTDA) complex that are worth noting. First, the design of the pyDTDA ligand as a bidentate, chelating ligand prevents the possibility of unpredictable rotation of the metal ion with respect to the ligand plane and thus ensures a fixed geometry for the overlap of the metal and ligand magnetic orbitals. A fixed geometry is important if the goal is to promote strong exchange coupling between the metal and ligand magnetic moments and to understand and predict the nature of this exchange coupling based on an orbital overlap model. Essentially, if the magnetic orbitals are orthogonal to one another, the ground state is expected to be high‐spin (according to Hund’s rule), i.e., ferromagnetic (FM) exchange coupling. If the magnetic orbitals are non‐orthogonal, the ground state is predicted to be low‐spin (according to the Pauli principle) giving antiferromagnetic (AF) exchange coupling. In cases where there are multiple magnetic orbitals on the metal ion, some non‐orthogonal and some orthogonal to the ligand magnetic orbital, predicting the nature of the exchange coupling becomes less certain and depends on the relative number of each type of orbital as well as their energy levels relative to the ligand orbital. Second, a dicationic 3d metal ion was selected in order to promote N‐ coordination over other possible interactions, such as insertion into the S‐S bond. Whereas the S‐coordination complexes of DTDAs employ low‐valent, soft and
7
Page 7 of 48
Ac ce p
te
d
M
an
us
cr
ip t
thiophilic metal ions, a harder metal ion, such as CoII, is more likely to form a complex by binding to the harder N atom sites. The choice of electron withdrawing hfac as the auxiliary ligands, providing charge balance and a 6‐coordinate environment, was intended to increase the relative hardness of the metal ion, but also to provide the added benefit of imparting volatility on the resulting coordination complex. All the DTDA coordination complexes of M(hfac)x fragments reported to date can be sublimed in reasonable yields, regardless of whether they are mononuclear, multinuclear or coordination polymers. Finally, although DTDA radicals exhibit a propensity to form π‐dimers (a.k.a. “pancake” bonds) [66, 67], the Co(hfac)2(pyDTDA) is not dimerized in the solid state. In fact, although there are several examples of dimerized coordination complexes of DTDA ligands, examples of undimerized complexes are equally prevalent (vide infra). Steric factors may play a role in suppressing dimerization in the coordination complexes, however it is also likely that donation of electron density to the metal ion serves to decrease the strength of the pancake bond, increasing the role of competing electrostatic and shape‐based packing factors in determining the crystal structure. Since the initial report of Co(hfac)2(pyDTDA), a series of M(hfac)2(pyDTDA) complexes has been synthesized, isolated and characterized (M = Mn, Fe, Ni, Cu). The MnII and CuII complexes both form pancake bonds in the solid state, however the dimerization motifs for the two are different. Whereas the DTDA heterocycles of the [Mn(hfac)2(pyDTDA)]2 pair interact to form a cis‐cofacial pancake bond, the CuII complex dimerizes in a twisted‐cofacial motif, [Cu(hfac)2(pyDTDA)]2 (Fig. 8). Both the MnII and CuII complexes exhibit monomer/dimer equilibria in solution [68]. By contrast, crystalline FeII and NiII complexes are isostructural with the CoII complex, thus are not dimerized in the solid state [69]. Two unusual decomposition products of the FeII complex were also isolated, one from thermolysis resulting in an Fe(hfac)2 complex of 2‐(pyrid‐2‐yl)‐4,6‐bis(trifluoromethyl)pyrimidine, and one from oxidation in air resulting in a tetranuclear FeIII cluster. This type of reactivity is atypical of coordination complexes of DTDA ligands. [PLACE FIGURE 8 HERE] Figure 8. Pancake bonds (π‐dimers) of N‐coordination complexes of 3d divalent metal ions with DTDA‐based paramagnetic ligands: [Mn(hfac)2(pyDTDA)]2 and [Cu(hfac)2(pyDTDA)]2 exhibit the cis‐cofacial and twisted‐cofacial dimer motifs respectively. (CSD Refcodes: UDIPED and UDIPIH) [68]. The magnetic properties of the above series of M(hfac)2(pyDTDA) coordination complexes were investigated. The hs‐MnII and hs‐FeII complexes exhibit AF coupling between metal and ligand magnetic moments, whereas the hs‐ CoII, NiII and CuII complexes exhibit FM exchange coupling. This trend is consistent with a simple orbital overlap model. For hs‐MnII, there are three unpaired electrons in the π* orbitals (nominally the t2g set for perfect Oh symmetry), at least two of which are non‐orthogonal to the π* SOMO of the DTDA moiety, whereas there are
8
Page 8 of 48
Ac ce p
te
d
M
an
us
cr
ip t
only two unpaired electrons in σ* orbitals (eg set) orthogonal the ligand‐based magnetic orbital. For hs‐FeII, there are two unpaired electrons in each set, but it seems that the non‐orthogonal relationship “wins out”. For hs‐CoII, there are now more unpaired electrons in orbitals orthogonal to the ligand magnetic orbital, and FM coupling is observed. For NiII and CuII, all the unpaired electrons are in σ* orbitals and FM exchange coupling with the ligand moment is observed, as anticipated. In a 6‐coordinate environment, crystal field stabilization energy (CFSE) associated with an octahedral arrangement of the ligand coordination sites for FeII, CoII, and NiII contributes some rigidity to the ligand sphere. By contrast, for hs‐MnII, having zero CFSE, the geometry of the ligand coordination positions is often far removed from octahedral and may even be better described as trigonal in many cases. In 6‐coordinate CuII species, an axial elongation (Jahn‐Teller distortion) is common. With these points in mind, it was initially postulated that the rigidity of the octahedral coordination environment, and resulting steric contributions, played a primary role in the crystal packing and lack of dimerization for M(hfac)2(pyDTDA) where M = FeII, CoII, and NiII, whereas distortions in the coordination geometry for M = MnII and CuII decreased the steric contribution enabling the formation of pancake bonds. In order to probe this postulate and attempt to manipulate the crystal packing of the DTDA‐metal complexes through electrostatic contacts, MnII and NiII complexes of the 4′ and 5′cyanopyDTDA ligands, and the MnII complex of 5′ bromopyDTDA where prepared [53]. The results indicate that the steric contribution from the hfac ligand field is not necessarily the major factor in the (non)formation of pancake bonds in the solid state. Whereas the NiII complex of pyDTDA does not dimerize in the solid state, NiII complexes of both 4′ and 5′ cyanopyDTDA crystallize as dimers, the former in a twisted‐cofacial and the latter in a trans‐antarafacial motif. By contrast, whereas the MnII complex of pyDTDA forms pancake bonded π‐dimers in the solid state, the MnII complex of 4′cyano pyDTDA does not. Thus, the incorporation of electrostatic supramolecular contacts can alter the crystal packing sufficiently to change the outcome of dimer formation. The pymDTDA ligand (Fig. 6) was designed as a paramagnetic analog of 2,2′‐ bipyrimidine, and thus a bis(bidentate) ligand capable of bridging two metal ions and mediating strong, predictable magnetic coupling between the metal ion moments. To this end, a series of bimetallic complexes of pymDTDA were reported, (M(hfac)2)2pymDTDA where M = Mn, Co, Ni, Zn (Fig. 9) [52, 70]. None of these species form pancake bonds in the solid state. As anticipated, the AF coupling in the hs‐MnII complex results in S = 9/2 spin ground state, whereas FM coupling in the CoII and NiII complexes gives rise to S = 7/2 and S = 5/2 spin ground state respectively, although it should be noted that modeling magnetic data of CoII species can be problematic owing to large magnetic anisotropy associated with this ion. The magnitude of the exchange coupling is relatively large (e.g. J/kB = +65(2) K for the NiII species, using the Ĥ = ‐2J(Ŝi・Ŝj) convention for the phenomenological spin Hamiltonian), which is a result of significant overlap between the ligand and metal magnetic orbitals, owing to the large spin density at the thiazyl N atoms. As ZnII is
9
Page 9 of 48
Ac ce p
te
d
M
an
us
cr
ip t
diamagnetic, the magnetic properties of the complex arise solely from the paramagnetic ligand, with an S = ½ spin ground state as anticipated. [PLACE FIGURE 9 HERE] Figure 9. The crystal structure of (Ni(hfac)2)2pymDTDA is representative of a series of 3d divalent metal ion complexes of pymDTDA (CSD Refcode: BARXEZ) [70]. By contrast, the crystal structure of (Ni(hfac)2)2pymDSDA is characterized by the formation of dimers via short Se…Se contacts (CSD Refcode: BARXID). The absence or presence of dimerization has clear consequences on the magnetic properties, illustrated at the χT product as a function of temperature. Bimetallic complexes of the selenium‐containing analog, 4‐(pyrimid‐2‐yl)‐ diselenadiazolyl, pymDSDA, are also reported for M = MnII and NiII [70]. Curiously, although the MnII complex of pymDSDA is isomorphous to that of pymDTDA in the solid state, the NiII complex of pymDSDA forms dimers in the trans‐antarafacial motif (Fig. 9). Dimerization has clear consequences for the magnetic properties. When involved in pancake bonding, the ligands can be treated as essentially diamagnetic, resulting in weak AF superexchange coupling between the coordinated NiII ions. Comparison of the magnetic data for (Ni(hfac)2)2pymDTDA with that for (Ni(hfac)2)2pymDSDA (Fig. 9) provides an unambiguous example of the role played by the paramagnetic ligand in mediating strong, predictable exchange coupling between metal ion moments, increasing the spin ground state of the molecule. Despite the relatively rigid octahedral ligand environment around the NiII ions, and thus significant steric demand of the hfac ligands, (Ni(hfac)2)2pymDSDA is the only example of dimerization in the series of analogous bimetallic, bridged complexes, further supporting the notion that steric factors are not necessarily the dominant factor determining whether or not dimers of the metal‐radical complexes are formed in the solid state. Further insight is gained by looking at established work on DSDA species suggesting that the pancake bond in a selenazyl dimer is stronger than that in a corresponding thiazyl dimer [43, 49]. From this perspective, it is the undimerized Mn complex of pymDSDA that may be the outlier, lending evidence to the hypothesis that coordination to a hard M(hfac) n fragment withdraws sufficient electron density from the heterocycle to significantly weaken the pancake bond, such that the lattice energy minimum for a complex is not always achieved through dimerization. Variations on the DTDA‐based ligand design extend beyond simple structural analogs of bipy and bpym (vide supra). An isostructural series of M(hfac)2 complexes of the cfDTDA ligand (M = Mn, Co, Ni) exhibits monodentate N‐coordination to the DTDA ring [54]. Two hfac ligands occupy the equatorial plane of the pseudo‐ octahedral metal coordination sphere, while the N atoms of two DTDA rings are in the axial positions (Fig. 10). Thus a metal ion bridges two ligands. Because the cfDTDA ligands are dimerized via a trans‐cofacial pancake bond, the magnetic properties of these complexes are dictated by essentially non‐interacting metal ion moments. Nevertheless, observation of monodentate N‐coordination is important
10
Page 10 of 48
Ac ce p
te
d
M
an
us
cr
ip t
because it suggests that the σ‐donor capacity of DTDA nitrogen atoms is sufficient to form a complex without chelation. Whether an electronic “assist” is required from the furan O atom, which is oriented toward the metal ion, is still up for debate. [PLACE FIGURE 10 HERE] Figure 10. Coordination polymers of the cfDTDA ligand with M(hfac)2 fragments (M = Mn, Co, Ni) exhibit monodentate N‐coordination involving one of the thiazyl nitrogen atoms. A fragment of the crystal structure for M = Mn is shown (CSD Refcode: LICWUQ) [54]. Given the observed monodentate N‐coordination of the cfDTDA ligand, the boaDTDA ligand (Fig. 6) was designed with the idea that both monodentate and bidentate N‐coordination might be possible in a single system. Indeed, this type of mixed coordination has been reported for a LaIII species (vide infra). An added feature of the boaDTDA ligand is that the shape and size of the benzoxazolyl ring is quite different from that of the DTDA ring. Crystal structures of the pyDTDA and pymDTDA complexes often suffer from disorder in the orientation of the ligand owing to the similarity in shape and size of the pyridyl/pyrimidyl rings and the DTDA ring. Disorder of this kind is avoided by employing a ligand with structurally dissimilar rings. As an important consequence, supramolecular contacts that impart magnetic exchange pathways become apparent in many boaDTDA complexes. The Mn(hfac)2 coordination complex of boaDTDA was the first of the series to be reported [55]. The complex does not form pancake bond π‐dimers in the solid state, but an unusual pairwise interaction, via electrostatic contacts between a DTDA S atom of one molecule and an hfac O atom of a neighboring molecule, is apparent (Fig. 11). As expected, AF coupling between the metal and ligand moments is observed (S = 2 per molecule). Moreover, the electrostatic Sδ+…Oδ‐ contacts provide a pathway for magnetic coupling between the molecules in a pair. The hfac O atoms have spin density related to the metal moment. The S atom has large spin density contributing to the ligand moment. Contact between the S and O atoms results in AF exchange coupling between the ligand moment of one molecule and the metal moment of a neighboring molecule. The result is a high spin ground state (ST = 4 per pair of molecules). This finding provides clear evidence for a supramolecular exchange pathway mediated by the thiazyl radical ligand. It also elucidates the magnetic properties of (Mn(hfac)2)2pymDTDA, which exhibits crystallographic disorder of the type described above as well as an increase in magnetic moment upon decreasing temperature, to values significantly above those expected for isolated complexes [52, 70]. [PLACE FIGURE 11 HERE] Figure 11. The Mn(hfac)2 coordination complex of boaDTDA crystal packs with short Sδ+…Oδ‐ contacts that provide an intermolecular exchange coupling pathway, resulting in a spin ground state of ST = 4 for each [Mn(hfac)2boaDTDA]2 pair (χT
11
Page 11 of 48
Ac ce p
te
d
M
an
us
cr
ip t
presented per pair of molecules). The Ni(hfac)2 complex of boaDTDA exhibits FM intermolecular exchange coupling, best understood in light of a McConnell I mechanism. (CSD Refcodes: FUWREV and XIPWUQ) [55, 71]. The Ni(hfac)2 coordination complex of boaDTDA provides evidence for a second type of supramolecular exchange pathway through ligand‐to‐ligand contacts [71]. In this case, FM exchange coupling between magnetic moments of the coordinated metal and ligand is observed, as anticipated. In addition, significant intermolecular FM interactions are observed and these can be best understood in light of the McConnell I mechanism. Ni(hfac)2boaDTDA crystallizes in a π‐stacked arrangement that places the DTDA ring of one molecule in proximity to the benzoxazolyl ring of a neighboring molecule, forming a “staircase” motif in one dimension (Fig. 11). The DTDA ring has significant α spin density. The benzoxazolyl ring does not play a significant role in the spatial distribution of the SOMOs of the complex, however spin polarization of underlying filled orbitals results in β spin density on several benzoxazolyl atoms. Overlap of the α spin density of one molecule and the β spin density of a neighboring molecule results in parallel alignment of the moments of these molecules (i.e. FM coupling). Thus far, at least three possible modes of supramolecular interaction between the metal‐DTDA complexes, involving the DTDA heterocycle, have been identified: 1) pancake bonding (a.k.a. dimerization), which may be thought of as a useful supramolecular synthon [72, 73], 2) electrostatic contacts involving the Sδ+ atoms of the DTDA, which can mediate exchange coupling between the metal and radical moments of neighboring atoms, and 3) π‐stacking contacts between the thiazyl ring and the aryl R group of neighboring complexes, which can provide a pathway for exchange coupling between radical moments of these neighboring complexes. In an effort to take advantage the electrostatic Sδ+…Oδ‐ contacts as a possible magnetic exchange pathway, the Mn(hfac)2 coordination complex of the DTDA2pym biradical ligand was prepared [56]. It was proposed that the bimetallic (Mn(hfac)2)2DTDA2pym would be capable of forming Sδ+…Oδ‐ electrostatic contacts in multiple directions and thus create magnetic exchange coupling pathways in multiple dimensions. Indeed, 2‐dimensional ribbon‐like arrays, connected by Sδ+…Oδ‐ contacts, are apparent in the solid state structure of (Mn(hfac)2)2DTDA2pym (Fig. 12). The magnetic properties are consistent with FM arrangement of the magnetic moments of neighboring complexes within a “ribbon”. Furthermore, weak AF coupling between the ribbons is sufficient to stabilize an AF ordered ground state (TN = 4.5 K). This provides the first example of a magnetically ordered metalDTDA complex. [PLACE FIGURE 12 HERE]
12
Page 12 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure 12. The (Mn(hfac)2)2DTDA2pym complex crystal packs in ribbon‐like arrays via Sδ+…Oδ‐ contacts akin to those observed in the related Mn(hfac)2 complex of boaDTDA. Weak AF coupling between the “high spin ribbons” results an AF ordered ground state (TN = 4.5 K) [56]. 2.3.3 Lanthanide Complexes Most early, noteworthy examples of neutral radical ligand (NRL) complexes of lanthanide ions used nitronyl nitroxides as the NRL building blocks [60, 74]. A recent report by Long et al., suggesting that the incorporation of radical ligands can dramatically enhance the single molecule magnet properties of multinuclear Dy(III) complexes [75, 76], has played a major role in renewing interest in lanthanide‐ radical complexes. In order to make lanthanide N‐coordination complexes of DTDA‐ based ligands, it was first necessary to identify a suitable (i.e. anhydrous) Ln precursor. Ln(hfac)3(DME), was selected for this purpose (where DME = 1,2‐ dimethoxyethane, a.k.a. monoglyme) [77‐80]. To date, coordination complexes of the boaDTDA ligand with La(hfac)3, Gd(hfac)3 and Dy(hfac)3, as well as the analogous Y(hfac)3 complex, have been reported [81, 82]. The Dy complex of boaDTDA dimerizes in the solid state via the twisted cofacial pancake bond motif (Fig. 13) [82]. Each of the two Dy ions of a [Dy(hfac)3boaDTDA]2 supramolecular pair is crystallographically unique, thus, although the ligands are the same for each metal ion, the geometry of the ligand sphere is quite different, and this has significant implications in terms of magnetization dynamics. Work by Murray, Tong, Winpenny and others has definitively established that, for Dy and other Ln single molecule magnets (SMMs), the nature of the ligands and the ligand sphere geometry can dramatically alter the SMM properties [83‐85]. The implication for the [Dy(hfac)3boaDTDA]2 supramolecular pair is that the magnetization dynamics of the two Dy ions may be different from one another. Measured under zero dc field, the [Dy(hfac)3boaDTDA]2 supramolecular pair exhibits slow magnetization dynamics consistent with SMM behavior. A single relaxation mode in the frequency dependence of the ac susceptibility is observed. However, measured under a very weak applied dc field (700 Oe), two thermally activated relaxation modes of similar intensity are observed. Effectively, the applied dc field “decouples” the two weakly AF coupled Dy ion moments. This interpretation of the observed magnetic properties is supported by the investigation of the Gd and Y analogs, which are crystallographically isomorphous to the Dy species. The magnetic properties of the [Gd(hfac)3boaDTDA]2 supramolecular pair, wherein negligible magnetic anisotropy is expected, are consistent with weak AF coupling between the two Gd ions. The [Y(hfac)3boaDTDA]2 supramolecular pair is diamagnetic, as anticipated, because YIII is a diamagnetic ion and the pancake bonded boaDTDA ligand pair is also diamagnetic. More importantly, a 5% dilution of Dy in 95% Y complex exhibits two thermally activated relaxation modes under zero dc field (Fig. 13). This behavior is exactly as anticipated given the interpretation of the magnetic data for the Dy complex. At 5% dilution, there should be statistically negligible contribution from the [Dy(hfac)3boaDTDA]2 pair and any magnetic properties (other than diamagnetism) should arise primarily from an admixture of
13
Page 13 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[“Dy/Y”(hfac)3boaDTDA]2 pairs, half of which have the Dy ion in one crystallographic environment and half of which have it in the other. Thus, the results from the dilution experiment should (and do) mimic the results from “decoupling” the two Dy ions with a small dc field. One intriguing conclusion from this work is that the [Dy(hfac)3boaDTDA]2 supramolecular pair represents a possible design for a universal (CNOT) quantum computing logic gate (qugate). [PLACE FIGURE 13 HERE] Figure 13. There are two crystallographically unique Ln positions in the [Ln(hfac)3boaDTDA]2 supramolecular pair (Ln = Y, Gd, Dy). The crystal structure of the Dy complex is shown (CSD Refcode: FIGBAA) [82]. Under 0 Oe applied dc field, [Dy(hfac)3boaDTDA]2 acts as an SMM. Under 700 Oe applied dc field, two unique SMMs are observed, implying that the application of a weak field “decouples” the two unique Dy(III) ions. A dilution of 5% Dy in 95% Y complex exhibits the same properties under 0 Oe as 100% Dy complex under 700 Oe, supporting the “decoupling” hypothesis. As previously noted (vide supra), the boaDTDA ligand was designed to have a bidentate N,N‐chelation “pocket” but it also has a potential N‐coordination site on the side of the molecule with the benzoxazolyl O atom. Given the observation of monodentate N‐coordination in the cfDTDA ligand, it was anticipated that both bidentate and monodentate N‐coordination might be possible for boaDTDA, and indeed this is observed in the La(hfac)3 coordination complex of boaDTDA [81]. The resulting coordination polymer is composed of La(hfac)3boaDTDA repeat units (Fig. 14). The boaDTDA ligand bridges two La(hfac)3 fragments via bidentate N,N‐ chelation on one side and monodentate N‐coordination on the other side. More accurately, on the side of the molecule with the benzoxazolyl O atom, the N‐La distance is within that normally defined as a “bond” whereas the O‐La distance is outside that normally defined as a “bond”. This coordination complex expands the realm of possible DTDA‐based metal coordination structures, but it also provides evidence for a magnetic superexchange mechanism involving overlap between the SOMO of the DTDA and the empty 5d orbitals of the La ion. The observed FM coupling between the radical ligand moments (Fig. 14) is unlikely to arise from interactions between the 1D coordination polymers in which the boaDTDA ligands are relatively isolated from one another. Thus the coupling is most likely between ligand moments within a polymeric chain. All distances between DTDA heteroatoms of the boaDTDA ligands within a chain exceed 5 Å, so it is reasonable to conclude that empty La orbitals must be involved. The 4f orbitals are buried in the xenon core, and overlap with the ligand orbitals is expected to be minimal. The 6s orbitals, being spherical, cannot be solely responsible for FM exchange coupling, owing to symmetry. Thus, it is reasonable to conclude that the 5f orbitals must be involved, and this finding is in keeping with recent computational analysis of exchange coupling in related binuclear 3d metal‐GdIII complexes [86]. [PLACE FIGURE 14 HERE]
14
Page 14 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure 14. The [La(hfac)3boaDTDA]n coordination polymer consists of bridging boaDTDA ligands and La(hfac)3 fragments. (CSD Refcode: YCBET). FM superexchange coupling between ligand‐based magnetic moments is evident [81]. 3. Complexes of 1,3,2Dithiazyl (132DTA) Fujita and Awaga were the first to recognize that the position of the N atom in the 132DTA heterocycle makes this paramagnetic moiety a promising candidate from which to synthesize monodentate N‐coordination complexes of metal ions. In 2001, they reported the first N‐coordination complex of a thiazyl radical, coordinating 1,3,5‐trithia‐2,4,6‐triazapentalenyl (TTTA) to Cu(hfac)2 in a 1:1 ratio [87]. The resulting coordination polymer [Cu(hfac)2TTTA]n assembles such that the TTTA bridges two Cu(hfac)2 fragments via N‐coordination involving the 132DTA nitrogen atom and one of the two thiadiazole ring N atoms (Fig. 15). Since there is negligible spin density at the thiadiazole N atom site, magnetic exchange coupling is observed between the metal and radical moments of isolated “Cu(hfac)2(TTTA)” units, presumably through N‐coordination at the 132DTA N atom. As anticipated based on a simple orbital overlap model, the exchange coupling is FM. This coordination polymer provides precedent for the potential use of thiazyl radicals as paramagnetic ligands and supramolecular building blocks [24], establishing that N‐coordination is possible without disruption of the thiazyl heterocycle. [PLACE FIGURE 15 HERE] Figure 15. The [Cu(hfac)2TTTA]n coordination polymer is composed of a 1:1 ratio of TTTA radical and Cu(hfac)2; (CSD Refcode: MIJQUR) [87]. The [BBDTA+][InCl4] coordination copolymer is an ionic species that exhibits coordination between the BBDTA+ radical cation and a diamagnetic metal‐base anion (CSD Refcode: PEKXEJ) [88]. The paramagnetic susceptibility as a function of temperature for [BBDTA+][InCl4] has been reprinted (adapted) with permission from [88]. Copyright (2006) American Chemical Society. In 2006, Fujita et al. reported a coordination polymer of the benzo[1,2‐d:4,5‐ d′]bis[1,3,2]‐dithiazolium radical cation with a diamagnetic main group metal anion, [BBDTA+][InCl4] [88]. Unlike the previously reported FeCl4, GaCl4 and GaBr4 ionic solids of BBDTA+ [89‐92], wherein the cation and anion are not bonded to one another, coordination of the metal anion to two BBDTA+ radical cations creates a one‐dimensional coordination polymer (Fig. 15) that exhibits a spin‐Peierls transition at 108 K [93]. The structurally analogous [BBDTA+][InBr4] exhibits a similar spin‐Peierls transition at a higher temperature (250 K) [94]. Both coordination polymers, [Cu(hfac)2TTTA]n and [BBDTA+][InCl4], are formed via monodentate N‐coordination to a 132DTA nitrogen atom. By contrast, the 2:1 co‐crystallization of 1,3,2‐benzodithiazolyl with [Co(mnt)2] (mnt2‐ = maleonitriledithiolate), (BDTA)2[Co(mnt)2] (Fig. 16), exhibits a structural and magnetic phase transition involving the formation/rupture of an S‐coordination bond between one of the BDTA molecules and the cobalt ion of the Co(mnt)2
15
Page 15 of 48
Ac ce p
te
d
M
an
us
cr
ip t
component [95‐97]. Thus, both N‐ and S‐ coordination are viable for 132DTA heterocycles. [PLACE FIGURE 16 HERE] Figure 16. The 2:1 coordination complex (BDTA)2[Co(mnt)2] exhibits S‐ coordination to the thiazyl radical at 100 K. (CSD Refcode: ECADOC). Magnetic data reprinted with permission from [95]. Copyright (2006) American Chemical Society. 4. A Complex of 1,2,3Dithiazolyl (123DTA) To date, there is only one example in the refereed literature of a metal coordination complex of a radical ligand based on the 123DTA heterocycle [98]. The 6,7‐ dimethyl‐1,4‐dioxo‐naphtho[2,3‐d][1,2,3]dithiazolyl ligand (67Me2pDTANQ) is a dimethyl derivative of 1,2,3‐dithiazolyl‐p‐naphthoquinone [99], designed to be a better σ‐donor and to be more soluble than the parent species. There are two possible coordination sites, a bidentate N,O‐chelation pocket and monodentate O‐ coordination site on the opposite side of the molecule. Significant α spin density at heteroatoms in both coordination sites can be rationalized in terms of the two dominant resonance contributors (Fig. 17). The Mn(hfac) 2 coordination complex of 67Me2pDTANQ is a volatile trinuclear Mn(hfac)2‐Radical‐Mn(hfac)2‐Radical‐ Mn(hfac)2 species (Mn(hfac)2)3(67Me2pDTANQ)2. As anticipated, AF exchange coupling between the metal and radical moments results in an ST = 13/2 ground state for the complex. [PLACE FIGURE 17 HERE] Figure 17. The (Mn(hfac)2)3(67Me2pDTANQ)2 coordination complex exhibits both monodentate O‐coordination and bidentate N,O‐chelation to the neutral radical ligand (CSD Refcode: ACOHOR) [98]. 5. 1,2,4,6Thiatriazinyl (TTA) Complexes There are fewer examples of TTA radicals than DTDA radicals in the literature, possibly because the latter tend to be easier to synthesize. As with DTDAs, the starting point for preparing TTAs is typically an R‐CN nitrile or the R‐C(NH)NH2 amidine (which can be prepared from the nitrile) [100]. Boeré et al. are active in the design and preparation of asymmetric TTAs and have made significant contributions to the synthetic strategies for preparing TTA derivatives [101]. Like DTDA radicals, TTAs tend to be volatile enough to be sublimed, making them potential candidates for use as building blocks in the design of paramagnetic ligands. 5.1 Scoordination and π Complexes In addition to reporting the formation of η2 π‐complexes of the [Cr(Cp)(CO)2]● fragment with DTDA radicals, Boeré and co‐workers have explored the reaction of [Cr(Cp)(CO)3]2 with a series of TTA radicals [102, 103]. Interestingly,
16
Page 16 of 48
Ac ce p
te
d
M
an
us
cr
ip t
two types of coordination complexes are observed. In the case where the TTA substituents are both phenyl groups (Ph2TTA), a σ‐bond is formed between a [Cr(Cp)(CO)3]● fragment and the TTA sulfur atom. This S‐coordinated species can also be thought of as a η1 π‐complex (Fig. 18). When more electron withdrawing substituents are present on the TTA, the resulting complex is comparable to the DTDA complex formed by reaction with [Cr(Cp)(CO)3]2. For both asymmetric radicals, 3‐trifluoromethyl‐5‐phenyl‐ and 3‐trifluoromethyl‐5‐(p‐methoxyphenyl)‐ 1,2,4,6‐thiatriazinyl (CF3TTAPh and CF3TTAPhOMe respectively), a η2 π‐ complex is formed from a [Cr(Cp)(CO)2]● fragment and the TTA radical (Fig. 18). Note the loss of a CO ligand from the chromium reagent. In all cases only the exo isomer is observed and resulting complexes are all diamagnetic. [PLACE FIGURE 18 HERE] Figure 18. Line drawing of the TTA radicals that form coordination complexes with Cr(Cp)(CO)x; examples of the two types of π‐complexes, η1 and η2 (CSD Refcodes: NEVFAW and NEVFEA) [102]. 5.2 Ncoordination Ligand Design The general structure of TTA radicals is ideal for the development of N‐ coordinate paramagnetic ligands. Using the same design principles as those used for DTDAs, 2′‐pyridyl substituents at the C atom positions create a species with either two bis‐bidentate chelation sites involving the N2 and N6 position nitrogen atoms of the TTA or one tridentate chelation site involving the N4 TTA nitrogen atom (Fig. 19). If 2′‐pyrimidyl substituents were employed, all three chelation sites would be present. Furan, benzoxazole, thiophene and other similar groups are also obvious candidates for the C atom position substituents in order to develop chelating paramagnetic ligands. The idea of using 3,5‐di(pyrid‐2‐yl)‐1,2,4,6‐thiatriazinyl (py2TTA) as an N‐ coordination paramagnetic ligand (Fig. 19) was first proposed by the Preuss group in Jian Wu’s Ph.D. thesis [104]. Initial work toward this goal included the isolation and structural characterization of the closed shell 3,5‐di(pyrid‐2‐yl)‐4H‐1,2,4,6‐ thiatriazine (py2TTAH) from the dipyridyl‐imidoylamidine. In general, TTAHs are in a lower oxidation state than TTAs and can be considered the protonated closed‐shell anion of the related neutral radical, as documented for other TTA analogs (e.g., 3,5‐ diphenyl‐4‐hydro‐1,2,4,6‐thiatriazine) [105]. The thesis also reports the isolation and structural characterization of the related hydrolysis product, 3,5‐di(pyrid‐2‐yl)‐ 4H‐1,2,4,6‐thiatriazine 1‐oxide (py2TTAOH; Fig. 19), demonstrating reactivity similar to other TTA analogs [16]. Recently, Brusso et al. [106] have also expressed interest in the use of py2TTA as a paramagnetic ligand and have reported a detailed synthesis focused on the isolation and structural characterization of two unusual N‐ bridgehead‐thiadiazolium intermediates en route to the closed‐shell py2TTAH. To date, the only evidence presented in either publication for the preparation of the py2TTA radical is an EPR spectrum acquired by in situ preparation. Interestingly, the structure of py2TTAH reported by Brusso et al. is a polymorph of that reported
17
Page 17 of 48
Ac ce p
te
d
M
an
us
cr
ip t
in the thesis (Fig. 19). N‐coordination complexes of py2TTA and other possible analogs of the TTA radical have yet to be reported, however, it is clear that these types of complexes are on the horizon. Exciting results arising from some challenging work are anticipated in this field in the near future. [PLACE FIGURE 19 HERE] Figure 19. Ligand design concepts for N‐coordination using TTA as a paramagnetic building block include the py2TTA radical. The crystal structure of the related py2TTAH and py2TTAOH species as reported in the original publication [104] are shown; thermal ellipsoids drawn at 50% probability (deposited in the CSD as private communications). 6. Complexes of 1,2,5Thiadiazole (125TDA) and Related Radical Anions In a neutral oxidation state, the 125TDA heterocycle is a closed shell, 6π electron ring. However, species incorporating the 125TDA can be reduced by one electron to resonance stabilized radical anions, and ionic compounds involving 125TDA radical anions can be prepared and isolated as crystalline solids [107‐110]. As observed with thiazyl radical cations (vide supra), ionic complexes of 125TDA radical anions that incorporate a metal‐containing counterion may also exhibit structural features consistent with coordination between the thiazyl radical and the metal. Crystal structures of a solvated and an unsolvated ionic complex of the [1,2,5]thiadiazolo[3,4‐c][1,2,5]thiadiazolidyl radical anion [K(18crown6)] [bis125TDA] were reported in 2005 by Mews, Stohrer, Zibarev, et al. [111]. In the unsolvated species, the radical anion acts as a bridging ligand, forming a one‐ dimensional –metal‐radical‐ coordination polymer (Fig. 20). In the solvated species, the radical anion acts as a chelating ligand and independent metal‐radical complexes are observed (Fig. 20). Soon thereafter, the structure of the unsolvated [Na(15crown5)] [bis125TDA] was also reported to be a one‐dimensional coordination complex with the radical anion acting in a bridging capacity [112]. In all three cases, N‐coordination is observed between the radical anion and the metal cation. Upon isolation and structural characterization of the [K(THF)][BTDA] ionic complex of the 2,1,3‐benzothiadiazolidyl radical anion [113], another coordination motif was revealed. In this case, the K+ cation is in contact with both an N and the S atom of the thiazyl radical anion (Fig. 20). [PLACE FIGURE 20 HERE] Figure 20. The closed‐shell, neutral bis125TDA and BTDA thiazyl heterocycles can be isolated as potassium salts of their radical anion oxidation states. (CSD Refcodes: CAYDIQ, CAYDOW and COQTAE) [111, 113]. In 2011, Awaga et al. expanded the field of thiazyl radical anion coordination chemistry by introducing the [1,2,5]thiadiazolo[3,4‐f][1,10]phenanthroline 1,1‐ dioxide (tdapO2) [114]. This species incorporates a 125TDA ring in which the S atom is modified to be an S(=O)2 moeity, thus the S atom is in a higher formal
18
Page 18 of 48
Ac ce p
te
d
M
an
us
cr
ip t
oxidation state (6+ as opposed to 2+). Bidentate N,N‐chelation at the phenanthroline “pocket”, as well as coordination to one or both of the O atoms, is possible with this novel and interesting radical anion ligand. Isolation and structural characterization of the potassium salt [K][tdapO2] demonstrates the chelating ability of the radical anion (Fig. 21). Furthermore, this complex, in which the cation is simply a diamagnetic alkali metal, exhibits magnetic ordering at 15 K owing to supramolecular interactions between the thiazyl radical anions. A series of ionic complexes of the [tdapO2] radical anion has also recently been reported, including Cs and Rb salts. [115] Incorporation of paramagnetic metal ions is likely on the horizon and eagerly anticipated to generate new complexes with exciting materials properties. [PLACE FIGURE 21 HERE] Figure 21. The closed‐shell tdapO2 species can be isolated as the potassium salt of its radical anion oxidation state. An excerpt from one of the reported potassium structures demonstrates the chelating ability of this radical anion (CSD Refcode: NALXIJ) [114]. 7. Concluding Remarks In the past decade, significant progress has been made in the field of thiazyl radical coordination chemistry. This area, which was until recently relatively unexplored, is now beginning to flourish. New ligand designs, new complexes and new magnetic and other materials properties are being reported. There are, as yet, many unexplored avenues in this field. While DTDA‐based ligands are the most prevalent in the current literature, potential ligand designs based on the DTDA building block are by no means exhausted. Multi‐dentate derivatives mimicking terpyridine or quaterpyridine, for example, are possible. More broadly speaking, paramagnetic analogs of essentially any existing ligand with a pendant pyridyl or pyrimidyl group might be developed. Furthermore, most paramagnetic thiazyls have yet to be fully considered for ligand development. There are indications that TTA‐based ligands and coordination complexes are on the horizon. The exploration of ligands incorporating 123DTA is still nascent despite the exceptional materials properties of numerous molecular species derived from this heterocycle. Rawson et al. recently reported metal coordination of 3‐(2´‐pyridyl)‐benzo‐1,2,4‐thiadiazine [116], a closed‐shell thiazyl species that is very closely related to the [1,2,4]thiadiazinyl (TDA) radicals [18]. We can anticipate the future employment of TDA radicals as ligand building blocks, with the added benefit that these species tend to be relatively air‐stable. In all cases, the incorporation of selenium in the place of sulfur presents a further facet for ligand modification and possible tuning of materials properties. Although there are a few examples of coordination to selenazyl ligands, this area also remains largely unexplored. Looking beyond simple ligand design, the field offers the possibility to create more complex structures. For example, metal‐organic frameworks (MOFs) and porous “guest” lattices derived from DTDA or other thiazyl building blocks may be
19
Page 19 of 48
Ac ce p
te
d
M
an
us
cr
ip t
feasible. These might have electronic and magnetic properties with influence/dependence on a “host” species. Alternatively, pendant thiazyls might be used as supramolecular synthons in order to assemble large structures via pancake bonding. The observation of bistability in 132DTA species suggests that “switchable” supramolecular synthons could also be a possibility. Clearly, there are numerous untapped opportunities for imaginative ligand design using thiazyl and selenazyl building blocks. There are equally uncharted avenues of research available by modification of the metal ions. To date, most thiazyl complexes have been limited to low oxidation state (0 to +2) transition metals and to a handful of lanthanide (+3) ions. Other oxidation states, heavier (4d and 5d) transition metals, and actinides are obvious areas for new exploration. Even simple modifications may be of value. At present, a very limited number of spectator ligands (e.g., hfac) have been considered and there is a lot of opportunity to modify the properties of existing species and/or create entirely new materials simply by altering the nature of the other ligands. With such diverse possibilities as those described above, the field of thiazyl and selenazyl coordination chemistry offers a lot of intriguing opportunities. With new researchers entering the field, bringing fresh ideas, the future promises exciting and meaningful progress in this area. 8. Acknowledgements The author is grateful to Prof. R. Boeré for sharing information in a manuscript prior to its availability as a published paper and to Prof. K. Awaga for consultation regarding his contributions to the field. This review has been funded in part by the University of Guelph and the Canada Research Chair’s program (Tier II). 9. References [1] M. Goehring, Q. Rev., Chem. Soc. (1956) 437‐450. [2] M. Goehring, D. Voigt, Naturwiss. (1953) 482. [3] C.‐h. Hsu, M.M. Labes, J. Chem. Phys. 61 (1974) 4640‐4645. [4] R.L. Greene, G.B. Street, L.J. Suter, Phys. Rev. Lett. 34 (1975) 577‐579. [5] T. Chivers, Chem. Rev. 85 (1985) 341‐365. [6] R.T. Boeré, T.L. Roemmele, in: eds. J. Reedijk and K. Poeppelmeier, Comprehensive Inorganic Chemistry II: From Elements to Applications, Elsevier, Oxford, UK, 2013, vol. 1, pp. 375‐411. [7] L.D. Huestis, M.L. Walsh, N. Hahn, J. Org. Chem. 30 (1965) 2763‐2766. [8] R. Mayer, G. Domschke, S. Bleisch, Tetrahedron Lett. 42 (1978) 4003‐4006. [9] A.Y. Makarov, S.N. Kim, N.P. Gritsan, I.Y. Bagryanskaya, Y.V. Gatilov, A.V. Zibarev, Mendeleev Commun. 15 (2005) 14‐17. [10] A.J. Banister, N. Bricklebank, I. Lavender, J.M. Rawson, C.I. Gregory, B.K. Tanner, W. Clegg, M.R.J. Elsegood, F. Palacio, Angew. Chem. Int. Ed. 35 (1996) 2533‐2535. [11] A. Alberola, R.J. Less, C.M. Pask, J.M. Rawson, F. Palacio, P. Oliete, C. Paulsen, A. Yamaguchi, R.D. Farley, D.M. Murphy, Angew. Chem. Int. Ed. 42 (2003) 4782‐4785. [12] W. Fujita, K. Awaga, Science 286 (1999) 261‐262.
20
Page 20 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[13] A. Mailman, S.M. Winter, X. Yu, C.M. Robertson, W. Yong, J.S. Tse, R.A. Secco, Z. Liu, P.A. Dube, J.A. Howard, R.T. Oakley, J. Am. Chem. Soc. 134 (2012) 9886‐9889. [14] A.A. Leitch, K. Lekin, S.M. Winter, L.E. Downie, H. Tsuruda, J.S. Tse, M. Mito, S. Desgreniers, P.A. Dube, S. Zhang, Q. Liu, C. Jin, Y. Ohishi, R.T. Oakley, J. Am. Chem. Soc. 133 (2011) 6051‐6060. [15] S.K. Pal, P. Bag, A. Sarkar, X.‐L. Chi, M.E. Itkis, F.S. Tham, B. Donnadieu, R.C. Haddon, J. Am. Chem. Soc. 132 (2010) 17258‐17264. [16] A.W. Cordes, R.T. Oakley, Acta Cryst. C43 (1987) 1645‐1646. [17] N.P. Gritsan, S.N. Kim, A.Y. Makarov, E.N. Chesnokova, A.V. Zibarev, Photochem. Photobiol. Sci. 5 (2006) 95‐101. [18] J. Zienkiewicz, P. Kaszynski, V.G. Young Jr., J. Org. Chem. 69 (2004) 7525‐7536. [19] X. Yu, A. Mailman, K. Lekin, A. Assoud, C.M. Robertson, B.C. Noll, C.F. Campana, J.A. Howard, P.A. Dube, R.T. Oakley, J. Am. Chem. Soc. 134 (2012) 2264‐2275. [20] A.J. Banister, M.I. Hansford, Z.V. Hauptman, S.T. Wait, W. Clegg, J. Chem. Soc., Dalton Trans. (1989) 1705‐1713. [21] A. Caneschi, D. Gatteschi, R. Sessoli, P. Rey, Acc. Chem. Res. 22 (1989) 392‐398. [22] A.J. Banister, I. May, J.M. Rawson, J.N.B. Smith, J. Organomet. Chem. 550 (1998) 241‐253. [23] K.E. Preuss, Dalton Trans. (2007) 2357‐2369. [24] K. Awaga, T. Tanaka, T. Shirai, M. Fujimori, Y. Suzuki, H. Yoshikawa, W. Fujita, Bull. Chem. Soc. Jpn. 79 (2006) 25‐34. [25] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02, (2009) Gaussian Inc., Wallingford, CT. [26] R. Dennington, T. Keith, J. Millam, GaussView, Version 5, (2009) Semichem Inc., Shawnee Mission, KS. [27] A. Vegas, A. Pérez‐Salazar, A.J. Banister, R.G. Hey, J. Chem. Soc., Dalton Trans. (1980) 1812‐1815. [28] L.N. Markovski, O.M. Polumbrik, V.S. Talanov, Y.G. Shermolovich, Tetrahedron Lett. 23 (1982) 761‐762. [29] A.J. Banister, N.R.M. Smith, R.G. Hey, J. Chem. Soc., Perkin Trans. 1 (1983) 1181‐1186. [30] H.‐U. Höfs, J.W. Bats, R. Gleiter, G. Hartmann, R. Mews, M. Eckert‐Maksić, H. Oberhammer, G.M. Sheldrick, Chem. Ber. 118 (1985) 3781‐3804. [31] R.T. Boeré, R.T. Oakley, R.W. Reed, J. Organomet. Chem. 331 (1987) 161‐167. [32] P. Del Bel Belluz, A.W. Cordes, E.M. Kristof, P.V. Kristof, S.W. Liblong, R.T. Oakley, J. Am. Chem. Soc. 111 (1989) 9276‐9278.
21
Page 21 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[33] Y. Beldjoudi, D.A. Haynes, J.J. Hayward, W.J. Manning, D.R. Pratt, J.M. Rawson, Cryst. Eng. Comm. 15 (2013) 1107‐1113. [34] A.J. Banister, I.B. Gorrell, W. Clegg, K.A. Jørgensen, J. Chem. Soc., Dalton Trans. (1989) 2229‐2233. [35] V. Klassen, K. Preuss, K.H. Moock, R.T. Boeré, Phosphorus, Sulfur and Silicon 93‐94 (1994) 449. [36] R.T. Boeré, K.H. Moock, V. Klassen, J. Weaver, D. Lentz, H. Michael‐Schulz, Can. J. Chem. 73 (1995) 1444‐1453. [37] A.J. Banister, I.B. Gorrell, W. Clegg, K.A. Jørgensen, J. Chem. Soc., Dalton Trans. (1991) 1105‐1109. [38] A.J. Banister, I.B. Gorrell, S.E. Lawrence, C.W. Lehmann, I. May, G. Tate, A.J. Blake, J.M. Rawson, J. Chem. Soc., Chem. Commun. (1994) 1779‐1780. [39] A.J. Banister, J.A.K. Howard, I. May, J.M. Rawson, Chem. Commun. (1997) 1763‐ 1764. [40] A.J. Banister, I.B. Gorrell, J.A.K. Howard, S.E. Lawrence, C.W. Lehmann, I. May, J.M. Rawson, B.K. Tanner, C.I. Gregory, A.J. Blake, S.P. Fricker, J. Chem. Soc., Dalton Trans. (1997) 377‐384. [41] W.‐K. Wong, C. Sun, W.‐Y. Wong, D.W.J. Kwong, W.‐T. Wong, Eur. J. Inorg. Chem. (2000) 1045‐1054. [42] J.M. Rawson, A.J. Banister, I. May, Mag. Reson. Chem. 32 (1994) 487‐491. [43] J.E. Davies, R.J. Less, I. May, J.M. Rawson, New J. Chem. (1998) 763‐765. [44] N. Feeder, R.J. Less, J.M. Rawson, J.N.B. Smith, J. Chem. Soc., Dalton Trans. (1998) 4091‐4093. [45] H.F. Lau, V.W.L. Ng, L.L. Koh, G.K. Tan, L.Y. Goh, T.L. Roemmele, S.D. Seagrave, R.T. Boeré, Angew. Chem. Int. Ed. Engl. 45 (2006) 4498‐4501. [46] H.F. Lau, P.C.Y. Ang, V.W.L. Ng, S.L. Kuan, L.Y. Goh, A.S. Borisov, P. Hazendonk, T.L. Roemmele, R.T. Boeré, Inorg. Chem. 47 (2008) 632‐644. [47] N.G.R. Hearns, K.E. Preuss, J.F. Richardson, S. Bin‐Salamon, J. Am. Chem. Soc. 126 (2004) 9942‐9943. [48] W.V.F. Brooks, N. Burford, J. Passmore, M.J. Schriver, L.H. Sutcliffe, J. Chem. Soc., Chem. Commun. (1987) 69‐71. [49] A.W. Cordes, C.D. Bryan, W.M. Davis, R.H. Delaat, S.H. Glarum, J.D. Goddard, R.C. Haddon, R.G. Hicks, D.K. Kennepohl, R.T. Oakley, S.R. Scott, N.P.C. Westwood, J. Am. Chem. Soc. 115 (1993) 7232‐7239. [50] A.W. Cordes, J.D. Goddard, R.T. Oakley, N.P.C. Westwood, J. Am. Chem. Soc. 111 (1989) 6147‐6154. [51] G.G. Alange, A.J. Banister, B. Bell, P.W. Millen, J. Chem. Soc., Perkin Trans. 1 (1979) 1192‐1194. [52] M. Jennings, K.E. Preuss, J. Wu, Chem. Commun. (2006) 341‐343. [53] N.G.R. Hearns, R. Clérac, M. Jennings, K.E. Preuss, Dalton Trans. (2009) 3193‐ 3203. [54] N.G.R. Hearns, K.D. Hesp, M. Jennings, J.L. Korčok, K.E. Preuss, C.S. Smithson, Polyhedron 26 (2007) 2047‐2053. [55] E.M. Fatila, J. Goodreid, R. Clérac, M. Jennings, J. Assoud, K.E. Preuss, Chem. Commun. 46 (2010) 6569‐6571.
22
Page 22 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[56] E.M. Fatila, R. Clérac, M. Rouzières, D.V. Soldatov, M. Jennings, K.E. Preuss, J. Am. Chem. Soc. 135 (2013) 13298‐13301. [57] J. Laugier, P. Rey, C. Benelli, D. Gatteschi, C. Zanchini, J. Am. Chem. Soc. 108 (1986) 6931‐6937. [58] P.F. Richardson, R.W. Kreilick, J. Am. Chem. Soc. 99 (1977) 8183‐8184. [59] H. Oshio, M. Umeno, T. Fukushi, T. Ito, Chem. Lett. (1997) 1065‐1066. [60] C. Benelli, A. Caneschi, D. Gatteschi, J. Laugier, P. Rey, Angew. Chem. Int. Ed. Engl. 26 (1987) 913‐915. [61] N. Ishii, Y. Okamura, S. Chiba, T. Nogami, T. Ishida, J. Am. Chem. Soc. 130 (2008) 24‐25. [62] R.G. Hicks, M.T. Lemaire, L.K. Thompson, T.M. Barclay, J. Am. Chem. Soc. 122 (2000) 8077‐8078. [63] T.M. Barclay, R.G. Hicks, M.T. Lemaire, L.K. Thompson, Chem. Commun. (Cambridge) FIELD Full Journal Title:Chemical Communications (Cambridge) (2000) 2141‐2142. [64] T.M. Barclay, R.G. Hicks, M.T. Lemaire, L.K. Thompson, Inorg. Chem. 40 (2001) 5581‐5584. [65] B.D. Koivisto, R.G. Hicks, Coord. Chem. Rev. 249 (2005) 2612‐2630. [66] H.Z. Beneberu, Y.‐H. Tian, M. Kertesz, Phys. Chem. Chem. Phys. 14 (2012) 10713‐10725. [67] K.E. Preuss, Polyhedron 79 (2014) 1‐15. [68] J. Britten, N.G.R. Hearns, K.E. Preuss, J.F. Richardson, S. Bin‐Salamon, Inorg. Chem. 46 (2007) 3934‐3945. [69] N.G.R. Hearns, E.M. Fatila, R. Clérac, M. Jennings, K.E. Preuss, Inorg. Chem. 47 (2008) 10330‐10341. [70] J. Wu, D.J. MacDonald, R. Clérac, I.‐R. Jeon, M. Jennings, A.J. Lough, J. Britten, C. Robertson, P.A. Dube, K.E. Preuss, Inorg. Chem. 51 (2012) 3827‐3839. [71] E.M. Fatila, R. Clérac, M. Jennings, K.E. Preuss, Chem. Commun. 49 (2013) 9431‐9433. [72] D.A. Haynes, Cryst. Eng. Comm. 13 (2011) 4793‐4798. [73] N. Bricklebank, S. Hargreaves, S.E. Spey, Polyhedron 19 (2000) 1163‐1166. [74] C. Benelli, A. Caneschi, D. Gatteschi, L. Pardi, P. Rey, D.P. Shum, R.L. Carlin, Inorg. Chem. 28) 27‐275. [75] J.D. Rinehart, M. Fang, W.J. Evans, J.R. Long, Nature Chem. 3 (2011) 538‐542. [76] J.D. Rinehart, M. Fang, W.J. Evans, J.R. Long, J. Am. Chem. Soc. 133 (2011) 14236‐14239. [77] G. Malandrino, O. Incontro, F. Castelli, I.L. Fragalà, Chem. Mater. 8 (1996) 1292‐1297. [78] S.‐J. Kang, Y.S. Jung, I.‐H. Suh, Bull. Korean Chem. Soc. 20 (1999) 95. [79] G. Malandrino, M. Bettinelli, A. Speghini, I.L. Fragalà, Eur. J. Inorg. Chem. (2001) 1039‐1044. [80] E.M. Fatila, E.E. Hetherington, M. Jennings, A.J. Lough, K.E. Preuss, Dalton Trans. 41 (2012) 1352‐1362. [81] E.M. Fatila, R. Clérac, M. Rouzières, D.V. Soldatov, M. Jennings, K.E. Preuss, Chem. Commun. 49 (2013) 6271‐6273.
23
Page 23 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[82] E.M. Fatila, M. Rouzières, M.C. Jennings, A.J. Lough, R. Clérac, K.E. Preuss, J. Am. Chem. Soc. 135 (2013) 9596‐9599. [83] N.F. Chilton, D. Collison, E.J. McInnes, R.E. Winpenny, A. Soncini, Nature Commun. 4 (2013) 2551. [84] J.‐L. Liu, Y.‐C. Chen, Y.‐Z. Zheng, W.‐Q. Lin, L. Ungur, W. Wernsdorfer, L.F. Chibotaru, M.‐L. Tong, Chem. Sci. 4 (2013) 3310‐3316. [85] N.F. Chilton, S.K. Langley, B. Moubaraki, A. Soncini, S.R. Batten, K.S. Murray, Chem. Sci. 4 (2013) 1719‐1730. [86] E. Cremades, S. Gómez‐Coca, D. Aravena, S. Alvarez, E. Ruiz, J. Am. Chem. Soc. 134 (2012) 10532‐10542. [87] W. Fujita, K. Awaga, J. Am. Chem. Soc. 123 (2001) 3601‐3602. [88] W. Fujita, K. Awaga, R. Kondo, S. Kagoshima, J. Am. Chem. Soc. 128 (2006) 6016‐6017. [89] W. Fujita, K. Awaga, Chem. Phys. Lett. 388 (2004) 186‐189. [90] G. Wolmershäuser, G. Wortmann, M. Schnauber, J. Chem. Res. (S) (1988) 358‐ 359. [91] M. Mito, M. Fujino, H. Deguchi, S. Takagi, W. Fujita, K. Awaga, Polyhedron 24 (2005) 2501‐2504. [92] K. Shimizu, T. Gotohda, T. Matsushita, N. Wada, W. Fujita, K. Awaga, Y. Saiga, D. Hirashima, Phys. Rev. B 74 (2006). [93] M. Mito, S. Kawagoe, H. Deguchi, S. Takagi, W. Fujita, K. Awaga, R. Kondo, S. Kagoshima, J. Phys. Soc. Jpn. 78 (2009) 124705. [94] W. Fujita, K. Kikuchi, K. Awaga, Angew. Chem. Int. Ed. Engl. 47 (2008) 9480‐ 9483. [95] Y. Umezono, W. Fujita, K. Awaga, J. Am. Chem. Soc. 128 (2006) 1084‐1085. [96] M. Kepenekian, B. Le Guennic, K. Awaga, V. Robert, Phys. Chem. Chem. Phys. 11 (2009) 6066‐6071. [97] K. Awaga, K. Nomura, H. Kishida, W. Fujita, H. Yoshikawa, M.M. Matsushita, L. Hu, Y. Shuku, R. Suizu, Bull. Chem. Soc. Jpn. 87 (2014) 234‐249. [98] D.J. Sullivan, R. Clérac, M. Jennings, A.J. Lough, K.E. Preuss, Chem. Commun. 48 (2012) 10963‐10965. [99] R. Mayer, G. Domschke, S. Bleisch, A. Bartl, Z. Chem. 21 (1981) 324‐325. [100] P.J. Hayes, R.T. Oakley, A.W. Cordes, W.T. Pennington, J. Am. Chem. Soc. 107 (1985) 1346‐1351. [101] R.T. Boeré, T.L. Roemmele, X. Yu, Inorg Chem 50 (2011) 5123‐5136. [102] C.Y. Ang, R.T. Boeré, L.Y. Goh, L.L. Koh, S.L. Kuan, G. Tan, Kheng, X. Yu, Chem. Commun. (2006) 4735‐4737. [103] C.Y. Ang, S.L. Kuan, G. Tan, Kheng, L.Y. Goh, T.L. Roemmele, X. Yu, R.T. Boeré, Can. J. Chem. DOI: 10.1139/cjc‐2014‐0257 (2014). [104] J. Wu, Ph.D., University of Guelph, 2008. [105] R.T. Boeré, A.W. Cordes, P.J. Hayes, R.T. Oakley, R.W. Reed, W.T. Pennington, Inorg. Chem. 25 (1986) 2445‐2450. [106] A.A. Leitch, I. Korobkov, A. Assoud, J.L. Brusso, Chem. Commun. 50 (2014) 4934‐4936. [107] S.N. Konchenko, N.P. Gritsan, A.V. Lonchakov, I.G. Irtegova, R. Mews, V.I. Ovcharenko, U. Radius, A.V. Zibarev, Eur. J. Inorg. Chem. (2008) 3833‐3838.
24
Page 24 of 48
Ac ce p
te
d
M
an
us
cr
ip t
[108] I.S. Morgan, M. Jennings, A. Vindigni, R. Clérac, K.E. Preuss, Cryst. Growth Des. 11 (2011) 2520‐2527. [109] N.A. Semenov, N.A. Pushkarevsky, A.V. Lonchakov, A.S. Bogomyakov, E.A. Pritchina, E.A. Suturina, N.P. Gritsan, S.N. Konchenko, R. Mews, V.I. Ovcharenko, A.V. Zibarev, Inorg. Chem. 49 (2010) 7558‐7564. [110] Y. Xie, Y. Shuku, M.M. Matsushita, K. Awaga, Chem. Commun. 50 (2014) 4178‐ 4180. [111] A.Y. Makarov, I.G. Irtegova, N.V. Vasilieva, I.Y. Bagryanskaya, T. Borrmann, Y.V. Gatilov, E. Lork, R. Mews, W.D. Stohrer, A.V. Zibarev, Inorg. Chem. 44 (2005) 7194‐ 7199. [112] V.N. Ikorskii, I.G. Irtegova, E. Lork, A.Y. Makarov, R. Mews, V.I. Ovcharenko, A.V. Zibarev, Eur. J. Inorg. Chem. (2006) 3061‐3067. [113] S.N. Konchenko, N.P. Gritsan, A.V. Lonchakov, U. Radius, A.V. Zibarev, Mendeleev Commun. 19 (2009) 7‐9. [114] Y. Shuku, R. Suizu, K. Awaga, Inorg. Chem. 50 (2011) 11859‐11861. [115] Y. Shuku, R. Suizu, A. Domingo, C.J. Calzado, V. Robert, K. Awaga, Inorg. Chem. 52 (2013) 9921‐9930. [116] E.R. Clark, M.U. Anwar, B.J. Leontowicz, Y. Beldjoudi, J.J. Hayward, W.T. Chan, E.L. Gavey, M. Pilkington, E. Zysman‐Colman, J.M. Rawson, Dalton Trans. 43 (2014) 12996‐13005.
25
Page 25 of 48
Highlights: A review of thia/selenazyl radical coordination chemistry is presented.
•
S‐coordination, N‐coordination and π‐coordination complexes are possible.
•
Complexes include main group metal ions, transition metals and lanthanides.
•
Magnetic and other unique materials properties have been reported.
•
There are still many research avenues as yet unexplored.
Ac ce p
te
d
M
an
us
cr
ip t
•
26
Page 26 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract (for review)
Page 27 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 28 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 29 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 30 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 31 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 32 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 33 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 34 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 35 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 36 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 37 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 38 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 39 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 40 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 41 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 42 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 43 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 44 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 45 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 46 of 48
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 47 of 48
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 48 of 48