Synthesis, characterization and reactivity of thiolate-bridged cobalt-iron and ruthenium-iron complexes

Article abstract of DOI:10.1016/j.cclet.2021.06.070

Thiolate-bridged hetero-bimetallic complexes [Cp*M(MeCN)N₂S₂FeCl][PF₆] (2, M = Ru; 3, M = Co, Cp* = η⁵-C₅Me₅, N₂S₂ = N,N'-dimethyl-3,6-diazanonane-1,8-dithiolate) were prepared by self-assembly of dimer [N₂S₂Fe]₂ with mononuclear precursor [Cp*Ru(MeCN)₃][PF₆] or [Cp*Co(MeCN)₃][PF₆]₂ in the presence of CHCl₃ as a chloride donor. Complexes 2 and 3 exhibit obviously different redox behaviors investigated by cyclic voltammetry and spin density distributions supported by DFT calculations. Notably, iron-cobalt complex 3 possesses versatile reactivities that cannot be achieved for complex 2. In the presence of CoCp₂, complex 3 can undergo one-electron reduction to generate a stable formally Co^IIFe^II complex [Cp*CoN₂S₂FeCl] (4). Besides, the terminal chloride on the iron center in 3 can be removed by dehalogenation agent AgPF₆ or exchanged with azide to afford the corresponding complexes [Cp*Co(MeCN)N₂S₂Fe(MeCN)][PF₆]₂ (5) and [Cp*Co(MeCN)N₂S₂Fe(N₃)][PF₆] (6). In addition, complexes 2, 3 and 4 show distinct catalytic reactivity toward the disproportionation of hydrazine into ammonia. These results may be helpful to understand the vital role of the heterometal in some catalytic transformations promoted by heteromultinuclear complexes.

Full text of DOI:10.1016/j.cclet.2021.06.070

ARTICLE IN PRESS
JID: CCLET
[m5G;August 8, 2021;4:8]
Chinese Chemical Letters xxx (xxxx) xxx
Contents lists available at ScienceDirect
Chinese Chemical Letters
journal homepage: www.elsevier.com/locate/cclet
Communication
Synthesis, characterization and reactivity of thiolate-bridged
cobalt-iron and ruthenium-iron complexes
Chao Guoa, Linan Sua,∗, Dawei Yang a, Baomin Wang a, Jingping Qua,b
a State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024, China
b State Key Laboratory of Bioreactor Engineering, Shanghai Collaborative Innovation Centre for Biomanufacturing, Frontiers Science Center for
Materiobiology and Dynamic Chemistry, East China University of Science and Technology, Shanghai 200237, China
a r t i c l e i n f o
a b s t r a c t
Thiolate-bridged hetero-bimetallic complexes [Cp∗M(MeCN)N2S2FeCl][PF6] (2, M = Ru; 3, M = Co, Cp∗
= η5-C5Me5, N2S2 = N,N’-dimethyl-3,6-diazanonane-1,8-dithiolate) were prepared by self-assembly of
dimer [N2S2Fe]2 with mononuclear precursor [Cp∗Ru(MeCN)3][PF6] or [Cp∗Co(MeCN)3][PF6]2 in the pres-
ence of CHCl3 as a chloride donor. Complexes 2 and 3 exhibit obviously different redox behaviors
investigated by cyclic voltammetry and spin density distributions supported by DFT calculations. No-
tably, iron-cobalt complex 3 possesses versatile reactivities that cannot be achieved for complex 2. In
the presence of CoCp2, complex 3 can undergo one-electron reduction to generate a stable formally
CoIIFeII complex [Cp∗CoN2S2FeCl] (4). Besides, the terminal chloride on the iron center in 3 can be re-
moved by dehalogenation agent AgPF6 or exchanged with azide to afford the corresponding complexes
[Cp∗Co(MeCN)N2S2Fe(MeCN)][PF6]2 (5) and [Cp∗Co(MeCN)N2S2Fe(N3)][PF6] (6). In addition, complexes 2,
3 and 4 show distinct catalytic reactivity toward the disproportionation of hydrazine into ammonia. These
results may be helpful to understand the vital role of the heterometal in some catalytic transformations
promoted by heteromultinuclear complexes.
Article history:
Received 22 April 2021
Revised 22 June 2021
Accepted 25 June 2021
Available online xxx
Keywords:
Metallic cooperativity
Heterobinuclear complex
Hydrazine disproportionation
Metallothiolate ligand
Metal-sulfur cluster
© 2021 Published by Elsevier B.V. on behalf of Chinese Chemical Society and Institute of Materia
Medica, Chinese Academy of Medical Sciences.
The cooperativity of different transition metals is widely
adopted in metalloenzymes, such as Mo/V-dependent nitrogenase,
[NiFe]-hydrogenase and carbon monoxide dehydrogenase, which
employ heteromultinuclear metal-sulfur clusters as their active
centers [1-3]. Inspired by these natural synthetic systems, artificial
heteromultinuclear catalysts are synthesized and proved to display
unique catalytic properties. For these catalysts, different transition
metals in close proximity may activate substrates simultaneously
or consecutively, while such activation modes are hard to achieve
by mononuclear or even homodinuclear complexes [4-8]. Further-
more, incorporating another different metal into a monometallic
species is likely to modulate its stability, electrochemical proper-
ties and reactivity. Therefore, understanding the mutual influence
of different metals assumes great significance in uncovering the
mechanism of the abovementioned metalloenzymes and designing
efficient catalysts.
by Hidai et al. [15] and alkyne silylformylation by the Co/Rh cat-
alyst reported by Nakamura et al. [16], the cooperative effect of
transition metals is, however, still far from being fully understood
at a molecular level until now. From a practical standpoint, heter-
obinuclear complexes would serve as more ideal models to probe
into the influence between two different active sites in catalysts
[17-21]. However, the coordination environment of reported het-
erobinuclear complexes makes it difficult to directly evaluate the
cooperativity of metals without considering the impact of auxiliary
ligands. Therefore, heterobinuclear complexes with the same coor-
dinated sphere are specially needed.
In our previous work, bimetallic cooperativity was adopted to
synthesize a series of thiolate-bridged homo- [22-24] and hetero-
dinuclear complexes [25-29] for small molecule activation and
transformation. Recently, we reported the synthesis of a thiolate-
bridged homobinuclear complex [Cp∗Fe(MeCN)N2S2FeCl][PF6] (1)
through the assembly of [N2S2Fe]2 (N2S2 = N,N’-dimethyl-3,6-
Although many heteromultinuclear complexes have been devel-
oped for homogeneous catalysis applications [9-16], for example,
selective hydroformylation by the Pd/Co mixed catalyst reported
diazanonane-1,8-dithiolate) [30] and [Cp∗FeII(MeCN)3][PF6] (Cp∗
=
η5-C5Me5) promoted by C−Cl bond cleavage of CHCl3 [31]. As a
further extension of this strategy, [N2S2Fe]2 may also serve as a
good precursor for the synthesis of heterobinuclear complexes fea-
turing a mononuclear reaction moiety {N2S2FeCl}. Herein, we re-
port the synthesis and characterization of novel thiolate-bridged
∗
Corresponding author.
E-mail address: lnsu@dlut.edu.cn (L. Su).
https://doi.org/10.1016/j.cclet.2021.06.070
1001-8417/© 2021 Published by Elsevier B.V. on behalf of Chinese Chemical Society and Institute of Materia Medica, Chinese Academy of Medical Sciences.
Please cite this article as: C. Guo, L. Su, D. Yang et al., Synthesis, characterization and reactivity of thiolate-bridged cobalt-iron and
ruthenium-iron complexes, Chinese Chemical Letters, https://doi.org/10.1016/j.cclet.2021.06.070

ARTICLE IN PRESS
JID: CCLET
[m5G;August 8, 2021;4:8]
C. Guo, L. Su, D. Yang et al.
Chinese Chemical Letters xxx (xxxx) xxx
Fig. 1. ORTEP (ellipsoids at 50% probability) diagrams of complexes 2 (a) and 3 (b).
−
All hydrogen atoms and the PF6 anion are omitted for clarity.
Table 1
˚
Selected bond lengths [A] and bond angles [°] of complexes
1, 2 and 3.
Complex
1
2
3
M1···M2
3.115(1)
2.401(2)
2.442(1)
2.238(1)
100.61(6)
82.89(5)
83.38(5)
2.9883(7)
2.3821(14)
2.3679(13)
2.2355(14)
101.30(5)
78.19(4)
3.437(3)
Fe1−S1
2.4122(13)
2.4537(12)
2.2561(13)
82.36(4)
Scheme 1. Synthesis of complexes 1, 2 and 3.
Fe1−S2
Fe1−Cl1
S1−Fe1−S2
M1−S1−M2
M1−S2−M2
94.22(4)
iron-ruthenium and iron-cobalt complexes. Experimental results
and theoretical analysis reveal the differences in redox properties
and spin distributions between the two heterobinuclear complexes,
which may have a significant impact on their reactivity.
78.32(4)
93.21(4)
As outlined in Scheme 1, treatment of the dimeric iron
complex [N2S2Fe]2 with
2 equiv. of mononuclear ruthenium
complex [Cp∗Ru(MeCN)3][PF6] [32] in MeCN/CHCl3 (10:1)
at room temperature resulted in the dissociation and re-
combination to generate
a
thiolate-bridged iron-ruthenium
complex [Cp∗Ru(MeCN)N2S2FeCl][PF6] (2) in 62% yield as
a
blue powder. Similarly, thiolate-bridged iron-cobalt complex
[Cp∗Co(MeCN)N2S2FeCl][PF6] (3) can also be facilely synthesized
as a dark red powder by using [Cp∗Co(MeCN)3][PF6]2 [33] as the
mononuclear precursor. From simple charge balance consideration,
the formal oxidation states of the two metallic centers in these
two complexes are +2 and +3 valences. However, compared to 2,
the formation of 3 does not need additional oxidant or reductant
since the two precursors are in +2 and +3 oxidation states,
respectively.
Fig. 2. (a) Spin density analyses of 2 and 3, spin up is shown in blue, and spin
down is shown in green (Isosurface = 0.02). (b) Cyclic voltammograms of 2 (green)
and 3 (blue), measured in 0.1 mol/L nBu4NPF6 in CH2Cl2 at 100 mV/s and internally
referenced to Fc+/Fc.
The electrospray ionization high-resolution mass spectrometry
(ESI-HRMS) of 2 shows a molecular ion peak at m/z 534.0168
(calcd. 534.0169) for [2−MeCN−PF6]+. The ESI-HRMS analysis of
3 exhibits a molecular ion peak at m/z 491.0456 (calcd. 491.0456)
for [3−MeCN−PF6]+. The appropriate isotopic distributions confirm
the existence of ruthenium and cobalt in 2 and 3, respectively. The
1H NMR spectrum of 2 shows two characteristic broad peaks at δ
15.69 and 9.68 ppm, which exhibit obviously strong paramagnetic
shift. Differently, the 1H NMR spectrum of 3 shows only one rel-
atively weak broad peak at δ 0.65 ppm. We preliminarily assign
these resonances to the proton signals of the Cp∗ ligand in 2 and
3. However, further accurate assignments for these signals are dif-
ficult due to their paramagnetism. Subsequently, the magnetic sus-
ceptibility measurements of 2 and 3 in solution were conducted by
the Evans’ method [34]. The values of effective magnetic moments
(μeff) of 2 and 3 are 5.69 and 5.18 μB, respectively, which indi-
cate 2 and 3 are in an S = 5/2 and 2 ground spin states at room
temperature.
apical position. Differently, the Ru and Co cores are both in a
three-legged piano-stool coordination geometry with one MeCN
molecule and two sulfur atoms of the N2S2 ligand. Although Ru
atom has larger atomic radius than the first row Fe and Co, inter-
˚
estingly, the Ru1···Fe1 distance (2.9883(7) A) in 2 is significantly
˚
shorter than Fe1···Fe2 (3.115(1) A) in 1 and Fe1···Co1 (3.437(3)
˚
A) in 3. Additionally, the Ru1···Fe1 distance in 2 is longer than
those of other reported thiolate-bridged iron-ruthenium complexes
˚
(2.564(1) to 2.691(2) A) [35-38]. In sharp contrast, the Co1···Fe1
distance in 2 is remarkably longer than those of previously re-
ported thiolate-bridged iron-cobalt complexes (2.398(1) to 2.796(1)
˚
A) [27,28,38-40], even one possessing a similar coordination sphere
˚
(3.136 (2) A) [41]. The obvious difference of bimetallic distance
among these complexes may indicate different redox properties
and reactivity.
With the molecular structures of these bimetallic complexes as
starting points, geometry optimized models were generated from
DFT calculation at the TPSSTPSS/LanL2DZ/6–31G(d) level of theory,
which indicates 2 and 3 are in an S = 5/2 and 2 ground spin states,
respectively (Table S8 in Supporting information). The computed
key geometric parameters match those determined by X-ray crys-
tallography well, which validate the computational methodology
(Tables S9 and S10 in Supporting information). The spin density
distribution for complexes 2 and 3 is depicted in Fig. 2a, and it
Furthermore, complexes 2 and 3 were identified by X-ray crys-
tallographic characterization. The molecular structures of 2 and 3
are shown in Fig. 1 and the main bond lengths and angles are
listed in Table 1. Their overall geometric configurations resemble
their diiron analogue 1 very closely. Complexes 2 and 3 both pos-
sess a distorted square pyramid moiety with the Fe atom em-
bedding in the N2S2 plane and the chloride atom locating in the
2

ARTICLE IN PRESS
JID: CCLET
[m5G;August 8, 2021;4:8]
C. Guo, L. Su, D. Yang et al.
Chinese Chemical Letters xxx (xxxx) xxx
Fig. 3. ORTEP (ellipsoids at 50% probability) diagrams of complexes 4 (a) and 6ꢀ (b).
−
All hydrogen atoms and the BPh4 anion of 6ꢀ are omitted for clarity.
˚
˚
˚
than those of 1 (2.238(1) A), 2 (2.2355(14) A) and 3 (2.2561(13) A),
which indicates the electron-rich {Cp∗Co} moiety may increase the
lability of the chloride, thus affecting the reactivity of {N2S2FeCl}
moiety. The 1H NMR spectrum of 4 also reveals its paramagnetic
nature and shows a broad singlet at δ −2.17 ppm in the higher
field compared with its precursor 3. The μeff value of 4 is 5.87 μB,
indicating 4 is also in an S = 5/2 ground spin state at room tem-
perature.
Scheme 2. The reactivity of complex 3.
shows a significant amount of spin is located on the Fe1 (spin up)
of both the two complexes. The similar Mulliken spin populations
(3.90 and 3.81) indicate the same +2 oxidation state of Fe1 in 2
and 3. Based on these results, we assume that the spin state of Fe1
is mainly determined by its coordination configuration and less af-
fected by the other metal core. In addition, a moderate amount of
spin is located on the Ru core in 2 (0.30) and the bridging S atoms,
while the spin located on the Co core in 3 has the opposite sign
(−0.41) and there is barely no spin located on the S atoms in 2.
The redox behaviors of complexes 2 and 3 were also investi-
gated by cyclic voltammetry in dichloromethane solution (Fig. 2b).
According to the cyclic voltammograms of reported bimetallic
complexes with similar coordination spheres [42,43], we attribute
the first irreversible redox event (green) at reduction peak po-
tential Epa = −0.89 V vs. ferrocene (Fc)+/0 to the {N2S2FeCl}II/I
redox couple of 2, which shows a remarkable positive shift of
250 mV compared to that of 1 (Epa = −1.14 V) [31]. The results
imply a greater ease of reduction at the {N2S2FeCl} moiety mod-
ulated by the {Cp∗Ru} moiety. Likewise, complex 3 also under-
goes an irreversible {N2S2FeCl}II/I reduction event at Epa = −0.98 V
(blue), which is only ca. 90 mV more negative than that of 2.
The second reversible redox wave of 2 (green) at half-wave po-
tential E1/2 = −0.05 V is assigned to the {Cp∗Ru}III/II redox couple,
which shows a negative shift of 180 mV with respect to that of
In order to open the potential reaction site, we next
attempted to remove the chloride group. As illustrated in
Scheme 2, treatment of
3 with 1 equiv. of AgPF6 in CH2Cl2
at room temperature afforded a new heterobinuclear complex
[Cp∗Co(MeCN)N2S2Fe(MeCN)][PF6]2 (5). The 1H NMR spectroscopic
analysis at room temperature shows a broad paramagnetic signal
appears at δ −0.87 ppm. In the infrared (IR) spectrum of 5, a di-
agnostic weak absorption band at 2283 cm−1 is observed, which is
≡
attributed to the C N stretch vibration of the MeCN ligands. Crys-
tallographic analysis reveals the replacement of the terminal chlo-
ride by a MeCN molecule and there are two MeCN ligands sepa-
rately bound to the Co and Fe centers in a trans arrangement (Fig.
S4 in Supporting information). The Co1···Fe1 distance of 3.3545(2)
−
˚
˚
A is only ca. 0.08 A shorter than that of 3. In addition, two PF6
anions are located in the same unit cell, which confirms complex
5 is a dicationic species. Unfortunately, in the presence of AgPF6,
complex 2 cannot transform to the RuFe analogue of 5, but fast
decomposed into unknown insoluble species.
Subsequently, we examined the reactivity of complex 3 toward
ligand exchange with sodium azide (NaN3). Treatment of 3 with
NaN3 in acetonitrile at room temperature gave an iron-cobalt azido
complex [Cp∗Co(MeCN)N2S2Fe(N3)][PF6] (6) in a moderate yield,
which is different from the reaction of zero-valent iron species
with organic azide to give an imido complex [44]. The ESI-HRMS
of 6 shows an expected molecular ion peak at m/z 498.0856 (calcd.
498.0859) for [6−MeCN−PF6]+. Similarly, the 1H NMR spectro-
scopic analysis of 6 also displays a characteristic broad signal at δ
0.51 ppm, which suggests 6 should also be a paramagnetic species.
In the IR spectrum of 6, a very strong absorption band at 2069
cm−1 is attributed to the stretching vibration of azide, which is
very close to those of sulfide- or thiolate-bridged iron-containing
complexes with the azido ligand in an end-on terminally coordi-
nated fashion [45,46]. In order to obtain the single-crystals suitable
for X-ray diffraction analysis, we performed the facile counterion
exchange reaction of 6 with NaBPh4 at room temperature to afford
an analogous complex [Cp∗Co(MeCN)N2S2Fe(N3)][BPh4] (6ꢀ).
The solid-state structure of 6ꢀ was confirmed by single-crystal
1 (E1/ = 0.13 V). Differently, complex 3 shows a quasi-reversible
2
{Cp∗Co}III/II oxidation event at Ec = 0.23 V (blue), which probably
corresponds to the easy oxidative degradation of the{Cp∗Co} moi-
ety.
With the synthesis and characterization of
2 and 3 being
achieved, we next investigated the reactivity of the two heterobin-
uclear complexes. Firstly, we probed into the possibility of their
one-electron reduction as predicted by electrochemical studies.
Upon interaction of 2 in CH2Cl2 with one-electron reductant cobal-
tocene (CoCp2), insoluble species immediately formed and its poor
solubility in common solvents limited further characterization. This
experimental fact suggests the reduced product is very unstable
and cannot maintain its bimetallic framework. In sharp contrast,
one-electron reduction of 3 conducted in similar conditions ex-
hibits completely different reaction phenomenon (Scheme 2). Crys-
tallographic analysis clearly reveals the final product is a neutral
formally FeIICoII complex [Cp∗CoN2S2FeCl] (4). As shown in Fig. 3a,
the acetonitrile ligand is removed after reduction and the Co1···Fe1
X-ray diffraction analysis (Fig. 3b). The Fe–N3 bond length of
ꢀ
˚
1.993(4) A in 6 is obviously longer than those of some mono-
˚
iron azido complexes (1.859(5)–1.934(7) A) [45,47,48]. Unexpect-
edly, complex 2 cannot react with NaN3 even under heating. These
experimental results demonstrated that another different metal
could be an important factor to the reactivity of the {N2S2FeCl}
moiety.
˚
˚
distance of 3.109(2) A is ca. 0.33 A shorter than that of its precur-
˚
sor 3. The Fe1−Cl1 distance (2.2762(12) A) in 4 is slightly longer
3

ARTICLE IN PRESS
JID: CCLET
[m5G;August 8, 2021;4:8]
C. Guo, L. Su, D. Yang et al.
Chinese Chemical Letters xxx (xxxx) xxx
Table 2
of Bio-based Chemicals of Liaoning Province of China, the “111”
project of the Ministry of Education of China and the Fundamental
Research Funds for the Central Universities (No. DUT19RC(3)013).
Catalytic disproportionation of hydrazine by 2, 3 and 4.a
Cat. (5 mol%)
3 N2H4 −−−−−−−−−−> 4NH3 + N2
MeCN, r.t. 12h
Entry
Catalyst
NH3 yield (%)b
Supplementary materials
1
2
3
[Cp∗Ru(MeCN)N2S2FeCl][PF6] (2)
[Cp∗Co(MeCN)N2S2FeCl][PF6] (3)
[Cp∗CoN2S2FeCl] (4)
35
5
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.cclet.2021.06.070.
trace
a
N2H4 (0.4 mmol), catalyst (5 mol%), CH3CN (5 mL), r.t., 12 h.
The resulted NH3 was converted into NH4Cl by mixing with
b
References
HCl·Et2O. The yields of NH4Cl were quantified by integration of
+
the NH4 resonance with respect to an internal reference of fer-
[1] J.C. Fontecilla-Camps, P. Amara, C. Cavazza, et al., Nature 460 (2009) 814–822.
[2] W. Lubitz, H. Ogata, O. Rüdiger, et al., Chem. Rev. 114 (2014) 4081–4148.
[3] P.A. Lindahl, J. Inorg. Biochem. 106 (2012) 172–178.
rocene in NMR.
[4] X. Wang, Z. Wang, T.T. Zhuang, et al., Nat. Commun. 10 (2019) 5186.
[5] S. Ogo, K. Ichikawa, T. Kishima, et al., Science 339 (2013) 682–684.
[6] K. Tanifuji, Y. Ohki, Chem. Rev. 120 (2020) 5194–5251.
One step further, we also explored the catalytic hydrazine dis-
proportionation to ammonia using 2, 3 and 4 as catalysts. Treat-
ment of a CH3CN solution of 2 with 20 equiv. of hydrazine at
room temperature afforded ammonia in 35% yield (Table 2, entry
1), which is comparable to other thiolate-bridged RuFe complexes
[37]. Reaction of 3 with hydrazine was also performed under the
same conditions and ammonia was obtained in 5% yield (Table 2,
entry 2), which reveals that complex 3 can only convert hydrazine
to ammonia stoichiometrically. In addition, complex 4 exhibited no
catalytic activity towards hydrazine disproportionation and only a
trace amount of ammonia was detected (Table 2, entry 3). Given
that the initial electrons for N−N bond cleavage of hydrazine are
provided by the complex itself, we speculate that the obviously
different performance in converting the hydrazine to ammonia be-
tween the RuFe and CoFe complexes is likely attributed to the sta-
bility under oxidation, which is consistent with the results of elec-
trochemical studies.
[7] P. Buchwalter, J. Rosé, P. Braunstein, Chem. Rev. 115 (2015) 28–126.
[8] D.R. Pye, N.P. Mankad, Chem. Sci. 8 (2017) 1705–1718.
[9] L. Hao, J. Xiao, J.J. Vittal, et al., Organometallics 16 (1997) 2165–2174.
[10] S.H. Druker, M.D. Curtis, J. Am. Chem. Soc. 117 (1995) 6366–6367.
[11] V. Ritleng, M.J. Chetcuti, Chem. Rev. 107 (2007) 797–858.
[12] M. Melnik, J. Garaj, C.E. Holloway, J. Coord. Chem. 61 (2008) 3021–3065.
[13] S. Maggini, Coord. Chem. Rev. 253 (2009) 1793–1832.
[14] S. Sculfort, P. Braunstein, Chem. Soc. Rev. 40 (2011) 2741–2760.
[15] Y. Ishii, K. Miyashita, K. Kamita, et al., J. Am. Chem. Soc. 119 (1997) 6448–6449.
[16] N. Yoshikai, M. Yamanaka, I. Ojima, et al., Organometallics 25 (2006)
3867–3875.
[17] C.M. Farley, C. Uyeda, Trends Chem 1 (2019) 497–509.
[18] J. Campos, Nat. Rev. Chem. 4 (2020) 696–702.
[19] G. Liu, P. Poths, X. Zhang, et al., J. Am. Chem. Soc. 142 (2020) 7930–7936.
[20] S. Komiya, Coord. Chem. Rev. 256 (2012) 556–573.
[21] M. Clemente-León, E. Coronado, C. Martí-Gastaldo, et al., Chem. Soc. Rev. 40
(2011) 473–497.
[22] Y. Chen, Y. Zhou, P. Chen, et al., J. Am. Chem. Soc. 130 (2008) 15250–15251.
[23] Y. Chen, L. Liu, Y. Peng, et al., J. Am. Chem. Soc. 133 (2011) 1147–1149.
[24] Y. Li, Y. Li, B. Wang, et al., Nature Chem 5 (2013) 320–326.
[25] D. Yang, Y. Li, L. Su, et al., Eur. J. Inorg. Chem. 2015 (2015) 2965–2973.
[26] P. Sun, D. Yang, Y. Li, et al., Organometallics 35 (2016) 751–757.
[27] P. Tong, W. Xie, D. Yang, et al., Dalton Trans 45 (2016) 18559–18565.
[28] Y. Zhang, D. Yang, Y. Li, et al., Dalton Trans 46 (2017) 7030–7038.
[29] Q. You, D. Yang, S. Xu, et al., Inorg. Chem. Commun. 106 (2019) 27–33.
[30] K.D. Karlin, S.J. Lippard, J. Am. Chem. Soc. 98 (1976) 6951–6957.
[31] L. Su, D. Yang, Y. Zhang, et al., Chem. Commun. 54 (2018) 13119–13122.
[32] P.J. Fagan, M.D. Ward, J.C. Calabrese, J. Am. Chem. Soc. 111 (1989) 1698–1719.
[33] G. Fairhurst, C. White, J. Chem. Soc., Dalton Trans. (1979) 1524–1530.
[34] S. Venkataramani, U. Jana, M. Dommaschk, et al., Science 331 (2011) 445–448.
[35] M. Yuki, Y. Miyake, Y. Nishibayashi, Organometallics 29 (2010) 5994–6001.
[36] D. Yang, Y. Li, B. Wang, et al., Inorg. Chem. 54 (2015) 10243–10249.
[37] Y. Zhang, J. Zhao, D. Yang, et al., Inorg. Chem. Commun. 83 (2017) 66–69.
[38] Y. Zhang, D. Yang, Y. Li, et al., Catal. Sci. Technol. 9 (2019) 6492–6502.
[39] S. Cai, X.F. Hou, Y.Q. Chen, et al., Dalton Trans (2006) 3736–3741.
[40] H. Gao, J. Huang, L. Chen, et al., RSC Adv 3 (2013) 3557–3565.
[41] P. Ghosh, M. Quiroz, N. Wang, et al., Dalton Trans 46 (2017) 5617–5624.
[42] C.Y. Chiang, J. Lee, C. Dalrymple, et al., Inorg. Chem. 44 (2005) 9007–9016.
[43] S. Ding, P. Ghosh, A.M. Lunsford, et al., J. Am. Chem. Soc. 138 (2016)
12920–12927.
In conclusion, we synthesized and characterized two iron-
cobalt and iron-ruthenium complexes with the same coordina-
tion sphere as the diiron complex [Cp∗Fe(MeCN)N2S2FeCl][PF6].
The electrochemical and DFT analyses of the two heterobinuclear
complexes suggested the redox properties and spin distributions of
the {N2S2FeCl} moiety were obviously affected by another metallic
center. When reacting with reducing agent in CH2Cl2, CoFe com-
plex can keep its integral framework without decomposing, com-
pared with its FeFe and RuFe analogues. Furthermore, only CoFe
complex can accomplish the ligand exchange reaction of the la-
bile chloride with inorganic salts such as AgPF6 or NaN3. In addi-
tion, RuFe complex shows better catalytic reactivity toward the dis-
proportionation of hydrazine into ammonia. Further studies on the
design and synthesis of new heterobinuclear complexes and their
catalytic properties are underway.
[44] W. Chen, Q. Chen, Y. Ma, et al., Chin. Chem. Lett. 31 (2020) 1342–1344.
[45] Y. Zhang, T. Mei, D. Yang, et al., Dalton Trans 46 (2017) 15888–15896.
[46] E. Dumas, J. Marrot, J. Jullien, et al., Inorg. Chim. Acta 358 (2005) 70–76.
[47] C.A. Grapperhaus, B. Mienert, E. Bill, et al., Inorg. Chem. 39 (2000) 5306–5317.
[48] G. Sabenya, L. Lázaro, I. Gamba, et al., J. Am. Chem. Soc. 139 (2017) 9168–9177.
Declaration of competing interest
The authors declare no competing financial interests.
Acknowledgments
This work was supported by the National Natural Science Foun-
dation of China (Nos. 21690064, 22001031), the key laboratory
4