Skip to main content
  • Research article
  • Open access
  • Published:

Thermal decomposition of [Co(en)3][Fe(CN)6]∙ 2H2O: Topotactic dehydration process, valence and spin exchange mechanism elucidation



The Prussian blue analogues represent well-known and extensively studied group of coordination species which has many remarkable applications due to their ion-exchange, electron transfer or magnetic properties. Among them, Co-Fe Prussian blue analogues have been extensively studied due to the photoinduced magnetization. Surprisingly, their suitability as precursors for solid-state synthesis of magnetic nanoparticles is almost unexplored.

In this paper, the mechanism of thermal decomposition of [Co(en)3][Fe(CN)6] ∙∙ 2H2O (1a) is elucidated, including the topotactic dehydration, valence and spins exchange mechanisms suggestion and the formation of a mixture of CoFe2O4-Co3O4 (3:1) as final products of thermal degradation.


The course of thermal decomposition of 1a in air atmosphere up to 600°C was monitored by TG/DSC techniques, 57Fe Mössbauer and IR spectroscopy. As first, the topotactic dehydration of 1a to the hemihydrate [Co(en)3][Fe(CN)6] ∙∙ 1/2H2O (1b) occurred with preserving the single-crystal character as was confirmed by the X-ray diffraction analysis. The consequent thermal decomposition proceeded in further four stages including intermediates varying in valence and spin states of both transition metal ions in their structures, i.e. [FeII(en)2(μ-NC)CoIII(CN)4], FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] and FeIII[CoII(CN)5], which were suggested mainly from 57Fe Mössbauer, IR spectral and elemental analyses data. Thermal decomposition was completed at 400°C when superparamagnetic phases of CoFe2O4 and Co3O4 in the molar ratio of 3:1 were formed. During further temperature increase (450 and 600°C), the ongoing crystallization process gave a new ferromagnetic phase attributed to the CoFe2O4-Co3O4 nanocomposite particles. Their formation was confirmed by XRD and TEM analyses. In-field (5 K / 5 T) Mössbauer spectrum revealed canting of Fe(III) spin in almost fully inverse spinel structure of CoFe2O4.


It has been found that the thermal decomposition of [Co(en)3][Fe(CN)6] ∙∙ 2H2O in air atmosphere is a gradual multiple process accompanied by the formation of intermediates with different composition, stereochemistry, oxidation as well as spin states of both the central transition metals. The decomposition is finished above 400°C and the ongoing heating to 600°C results in the formation of CoFe2O4-Co3O4 nanocomposite particles as the final decomposition product.


Transition metal hexacyanido complexes represent an enormous and well-known group of coordination species. The Prussian blue, with the idealized formula Fe4[Fe(CN)6]3 · nH2O, is one of the oldest examples of a hexacyanido complex combining interesting structural features and magnetic properties. Therefore, many Prussian blue analogues [1], e.g. MxA[B(CN)6]z · nH2O (M is alkali metal, A and B are different divalent/trivalent transition metals, x ≤ 1, z ≤ 1, if z < 1 B-sites are fractionally occupied and B-vacancies are present) have been extensively studied due to their sorption, ion-exchange, electron transfer or magnetic properties, and in recent years, many applications using PBAs usually in form of thin films or nanoparticles have been developed [2, 3].

The reversible intercalation of alkali metal cations due to the zeolite-like structure of PBAs [4] has been employed in still active development of the selective adsorbent of 137Cs from radioactive waste waters [5] or more recently, in construction of a new active electrode material for Li-ion batteries (AxMny[Fe(CN)6] where A = K, Rb, x < 1, 1 < y < 1.5) [6]. The electrochemical coating of electrodes with thin film of PBAs or deposition with coordinated PBAs nanoparticles [2, 7, 8] gave rise a large number of amperometric and potentiometric sensors for non-electroactive ions or H2O2[3, 9], electrochromic sensors [10] or biosensors [11]. The electrochromic or thermochromic behavior of electrochemically prepared PBAs thin films was further applied in the construction of different devices including electrochromic displays and pattering devices, thermal probes or photoelectrochemical/photocatalytical solar energy convertors [12].

PBAs also represent an enormous and important group of molecular magnets [1, 1315] which play important roles in many applications mainly in molecular electronics and magnetic/electronic devices [16, 17]. During the experimental race to enhance the Curie temperatures of molecular magnets based on PBAs, the systems incorporating V(II) gave the most promising results and KV[Cr(CN)6]·2H2O was found to possess the highest Tc even above the boiling point of water (376 K) [18]. This invention of room temperature molecule-based magnets opens interesting applications as magnetic switches, oscillators or magneto-optical data storage devices. The cyanido-bridged networks, serving as a platform for the preparation of coordination nanoparticles based on Prussian blue, has also been employed for the controlling/tuning of magnetism by temperature (e.g. AI[MIIMIII(CN)6], A = alkali metal ion) [2] or light irradiation (with a general formula of A x Co y [Fe(CN)6] z  · nH2O; x, y, z are independent) [8].

An important feature of especially Co-Fe Prussian blue systems is the phenomenon of photoinduced magnetization or photoreversible enhancement of magnetization [1921] which has been observed in e.g. Rb1.8CoIII3.3CoII0.7[FeII(CN)6]3.3 · 13H2O [22, 23], Na0.34Co1.33[Fe(CN)6] · 4.5H2O [19], or K0.4Co1.4[Fe(CN)6] · 5H2O [20]. These compounds are efficient 3D-ferrimagnets at low temperatures (Tc < 30 K) due to the presence of majority of FeII–CN–CoIII diamagnetic pairs which undergo upon irradiation photoinduced electron transfer resulting in paramagnetic FeIII–CN–CoII species. Such transformation is accompanied by elongation of the Co–N bond lengths (0.15–0.20 Å) which is possible only in the case of flexible inorganic network provided by the existence of hexacyanidoferrate vacancies occupied by water molecules.

Transition metal hexacyanido complexes have been also successfully utilized as building blocks for constructing oligo- and polynuclear assemblies derived from the Prussian blue and a large number of such organic–inorganic hybrid materials containing various organic ligands, mainly N-donor linear or macrocyclic ligands, have been extensively studied [2428]. These compounds were also found to behave as molecular magnets [29] with potential applications as molecular devices or memory devices. Some of them also revealed very specific optical properties including ion pair charge-transfer (IPCT) [30] transitions corresponding to the transfer of electrons from the complex anions to the complex cations in the visible and near-ultraviolet spectral regions. This phenomenon was confirmed for ionic pair compounds involving both [Co(en)3]3+ and [M(CN)x]4– (for x = 6, M = Fe, Ru, Os; for x = 8, M = Mo, W) complex ions [31]. An irradiation of solutions containing these species into the IPCT absorption band (λirr = 405 or 510 nm) leads to a structure re-arrangement and the formation of bi- or trinuclear species. For the ionic pairs [M(NH3)5L]3+ and [Fe(CN)6]3– (M = Ru, Os; L = NH3 or H2O), an inverse direction of the electron transfer of the IPCT transition (from the cation to the anion) was found [32].

Only eleven X-ray structures of discrete dinuclear Co-Fecyanido-bridged complexes have been reported up to now, whereas about ninety hits corresponding to poly- or oligomeric Co-Fe cyanido-bridged complexes have been deposited within the Cambridge Structural Database up to now [33]. Most of the dinuclear complexes have the general formula of [L n CoIII–NC–FeII(CN)5], where Ln stands for a macrocyclic ligand, and they have been systematically studied by Bernhardt et al.[3436]. Up to date, similar number of ionic pair Co(III)-Fe(III) complexes have been structurally characterized. They always contain hexacyanidoferrate(III) anion and Co(III) complexed by ammonia [37], en[3841], bipy[4245], terpy[46], or other suitable ligands [47]. Despite Co(III)-Fe(III) hexacyanidoferrates represent very important group of compounds bearing a broad spectrum of important physico-chemical properties, surprisingly only a little attention has been paid to the monitoring of the thermal behavior of these compounds [48, 49]. Especially the fact that they can serve as suitable precursors for thermally induced solid-state synthesis of magnetic nanoparticles [50, 51] is almost unexplored.

The dehydration process of the monohydrate [Co(en)3][Fe(CN)6]∙ H2O was previously described in details by our working group [40] and recently, thermal decomposition of [Co(en)3][Fe(CN)6]∙ 2H2O in different atmospheres has been published as well [52]. In the present work, we also report the thermal degradation of [Co(en)3][Fe(CN)6]∙2H2O (1a) which was obtained upon recrystallization from water at 60°C. But furthermore, we concentrate on the progress of thermal decomposition of [Co(en)3][Fe(CN)6]∙2H2O under dynamic heating in air which includes the topotactic [53] dehydration process of the starting complex and gradual ligand liberation/decomposition resulting in the formation of CoFe2O4-Co3O4 nanocomposite particles as a final product of the thermal conversion. Based on the elemental and TG/DSC analyses, IR and 57Fe Mössbauer spectra, we suggest and discuss the unique decomposition mechanism accompanied by changes in the valence and spin states of both transition metal ions.

Results and discussion

Synthesis, X-ray structure and spectral characterization of [Co(en)3][Fe(CN)6]· 2H2O (1a)

In contrast to our previous study which described a topotactic dehydration process of [Co(en)3][Fe(CN)6]∙ H2O into [Co(en)3][Fe(CN)6] by means of single crystal X-ray analysis and Mössbauer spectroscopy [40], herein, we strive to elucidate the mechanism of the thermal decomposition of [Co(en)3][Fe(CN)6]∙ 2H2O based on a complete characterization of the decomposition intermediates. The ionic pair [Co(en)3][Fe(CN)6]∙ 2H2O complex was prepared by the straightforward reaction between [Co(en)3]3+ and [Fe(CN)6]3–, used as building blocks, in DMF [38]. The efforts to prepare the title compound by an one-pot redox reaction in water using CoCl2 6H2O instead of [Co(en)3]Cl3∙ 3H2O were unsuccessful probably due to the incomplete coordination of en and formation of Co(II)-involving PBA [23]. The synthesis following the Bok’s published procedure led to the formation of the dihydrated complex, however, two products, i.e. the dihydrated and monohydrated complexes, can be obtained after the product recrystallization from hot water in dependence on the rate of cooling of the solution. Concretely, the monohydrate is formed in the case of fast cooling of the solution heated up to 90°C, while the dihydrate is formed in the case of slow cooling of the 60°C warm solution.

As first, the molecular and crystal structures of 1a have been determined by using single crystal X-ray analysis (Figure 1, Table 1). The crystallographically independent part of the molecule consists of the [Co(en)3]3+ cation, two halves of the [Fe(CN)6]3– moiety with the iron(III) atoms lying on special positions (inversion centers) and two crystal water molecules (Figure 1). As the structure posses an inversion center, both Λδδδ (lel 3 ) and Λλλλ (ob 3 ) conformations of the [Co(en)3]3+ cations are present [54]. The Co(III) atom is coordinated by six nitrogen atoms from three chelating molecules of en in a distorted octahedral geometry. The Co–N bond lengths ranges from 1.955(2) to 1.981(2) Å (Table 2), and surprisingly, these values are significantly shorter than those determined for the same structure by Bok [1.996(3)–2.035(4) Å] [38]. For comparison, very similar values were found for [Co(en)3]I3 · H2O (1.950(4)–1.981(4) Å) [55] and for [Co(en)3]Cl3 (1.967(11) Å) [56], while they are much more distributed in [Co(en)3]3[FeCl6]Cl6·H2O (1.919(5)–2.002(5) Å) [57]. The differences may be caused by the counter-ion that significantly influences non-bonding interactions within the crystal structure and thus, a distortion in the vicinity of the central Co(III) atom can be found as a consequence of these interactions.

Figure 1
figure 1

Molecular structure of [Co( en ) 3 ][Fe(CN) 6 ] 2H 2 O (1a). The thermal ellipsoids are drawn at the 50% probability level.

Table 1 Crystal data and structure refinements for 1a and 1b
Table 2 Selected bond lengths (Å) and angles (°) for 1a and 1b

In the crystal structure, the Fe(III) atoms are coordinated by six carbon atoms originating from terminal cyanide groups. While the {Fe1C6} octahedron is almost regular (the Fe–C distances range from 1.944(3) to 1.951(3) Å), the {Fe1AC6} octahedron is slightly compressed with the shorter axial Fe1A–C2A distances (1.932(3) Å) as compared to the equatorial ones (1.941(3) and 1.946(3) Å) (Table 2). Such type of deformation of the hexacyanidoferrate(III) anions was already observed, e.g. in [(aed)2Cu2ClFe(CN)6]n · 2H2O, where the Fe–C bond lengths range from 1.931(4) to 1.953(4) Å [58]. The remaining geometric parameters determined for [Fe(CN)6]3– in 1a are comparable to those found for the same complex anions in other compounds, e.g. in [Ni(tren)]3[Fe(CN)6]2·6H2O [59] or [Ni(en)2]3[Fe(CN)6]2·2H2O [60].

The secondary structure of 1a is further stabilized by the O–HO, O–HN and N–HN hydrogen bonds together with the C–HN and N–HC non-covalent contacts connecting the complex ions and crystal water molecules (Figure 2, Table 3).

Figure 2
figure 2

Part of the crystal structure of [Co( en) 3 ][Fe(CN) 6 ] 2H 2 O (1a) showing selected non-covalent contacts (dashed lines). The CH2 groups of the ethylenediamine ligands are omitted for clarity. The thermal ellipsoids are drawn at the 50% probability level.

Table 3 Selected hydrogen bonds for 1a and 1b (Å, °)

The information on the local electronic state of the iron atom can be quantitatively obtained from 57Fe Mössbauer spectra. Concretely, the isomer shift, δ, which is related to the s-electron density at the nucleus and quadrupole splitting, Δ, related to the electric field gradient, are crucial parameters sensitive to the oxidation and spin states of iron atom as well as its coordination number. Thus, room temperature Mössbauer spectrum of 1a confirmed the presence of Fe(III) ions in low-spin state because it consists a symmetrical doublet with isomer shift δ = −0.17 mm/s and quadrupole splitting ΔEQ = 0.85 mm/s (Figure 3). The values of these parameters are typical for the hexacyanidoferrate(III) anion [61]. In measured IR spectra, typical sharp and strong absorption bands at 2117 and 2107 cm–1 corresponding to the υ(C≡N) stretching vibration were identified (Figure 4). They well document the presence of terminal coordinated cyanide groups because the vibration frequency of free CN anion (about 2080 cm–1) is blue shifted due to a coordination (removing electrons from weakly antibonding 5σ orbital during metal-carbon σ bond formation) like in K3[Fe(CN)6] which displays one sharp absorption band at 2118 cm–1[41, 62]. In comparison with K3[Fe(CN)6], there is a splitting of the absorption band that can be explained by lowering of the Oh site symmetry of the hexacyanidoferrate(III) anion. Otherwise, broad absorption bands due to the O–H (~3380 cm–1), N–H (~3234 cm–1) or C–H (~3095 cm–1) stretching as well as deformation vibrations (NH2, 1630–1578 cm–1) are present together with the pattern of fingerprint region corresponding to the three coordinated molecules of en (the full IR spectrum is available in Additional file 1: Figure S1).

Figure 3
figure 3

A representative Mössbauer spectrum of [Co( en ) 3 ][Fe(CN) 6 ] 2H 2 O (1a) at 25°C and spectra of the converted intermediates formed by heating of 1a up to 240, 300 and 350°C.

Figure 4
figure 4

Part of the room temperature IR spectrum of [Co( en ) 3 ][Fe(CN) 6 ] 2H 2 O (1a, 25°C) and its comparison with the IR spectra of the corresponding decomposition intermediates at the temperatures of 150, 230, 265 and 450°C.

The temperature dependence of the effective magnetic moment (for details see Additional file 1: Figure S2) revealed weak antiferromagnetic interactions. The effective magnetic moment of 1a at room temperature was 2.23 B.M. and it is gradually decreased to 1.84 B.M. as the temperature was reduced to 2 K. Its value was higher than expected for the spin-only value of 1.73 B.M. for S = 1/2, but similar to the value of 2.25 B.M. obtained for K3[Fe(CN)6] [63]. Fitting of the reciprocal susceptibility (see Additional file 1: Figure S3) above 50 K by Curie-Weiss law gave g = 2.67 and Θ = −19 K for S = 0 and S = 1/2 for CoIII, and FeIII, respectively. The higher values of g-factors have been previously observed for other low-spin FeIII complexes as well [64].

Thermal behavior of [Co(en)3][Fe(CN)6]∙ 2H2O (1a) – a general overview

The TG and DSC curves of 1a are displayed in Figure 5. TG curve of 1a reveals two separable steps of the thermal decomposition. The first weight loss occurs in the temperature range of 50–80°C and it is accompanied by endothermic effect with the minimum at 80°C, as can be seen on DSC curve (Figure 5). Experimental weight loss (5.6%) is in good agreement with the calculated one (5.5%) corresponding to the release of 1.5 crystal water molecules from the formula unit of 1a and thus, yielding the complex [Co(en)3][Fe(CN)6] 1/2H2O (1b), as confirmed by chemical and single crystal X-ray analyses. The hemihydrate 1b is thermally stable within the temperature interval of 72–187°C. The second step of the thermal decay is observed in the temperature range of 187–420°C, and involves continual and complete sample decomposition. Based on the course of DSC curve, it can be stated that at least three processes take place: (i) a weak exothermic process with the maximum at 208°C, (ii) an endothermic process with the minimum at 214°C and (iii) a strong exothermic effect with the maximum at 354°C. The sample weight loss corresponding to the second step was found to be 61.7%. It can be assigned to elimination of the remaining water content, all molecules of ethylenediamine as well as all CN ligands and/or oxidation of the transition metal ions yielding a spinel-type phase with the general chemical composition of Co1.5Fe1.5O4 (calculated weight loss 62.1%). In comparison with the thermal decomposition of 1a recently published in Ref. [52], the decomposition of 1a occurs at lower temperature in our case, which can be associated to different experimental conditions, and moreover, CoFe2O4 only was determined as the final decomposition product [56]. Unstable intermediates [FeII(en)2(μ-NC)CoIII(CN)5], [FeIII(H2NCH2CH3)2(μ-NC)2CoII(CN)5] and FeIII[CoII(CN)5] were identified during this complicated decomposition. The detailed description of the above-mentioned processes with identification of the intermediates by XRD method, Mössbauer and IR spectroscopy and transmission electron microscopy (TEM) will be discussed below mainly from the chemical point of view (vide infra).

Figure 5
figure 5

TG (dotted) and DSC (full) curves of complex [Co( en ) 3 ][Fe(CN) 6 ] 2H 2 O (1a). Data were recorded in the temperature interval of 20–500, and 20–400°C, respectively, and with the heating rate of 5°C/min.

Chemical, structural and spectral characterization of [Co(en)3][Fe(CN)6]∙1/2H 2 O (1b) formed in the first decomposition step

The complex [Co(en)3][Fe(CN)6] 1/2H2O (1b), as it was already mentioned, can be obtained as a stable intermediate by both dynamic or static heating of the dihydrate 1a up to 130°C (Table 4). The empirical formula of 1b was deduced from the elemental analysis results and well corresponds to the weight loss on the TG curve (Figure 5). The parameters of Mössbauer spectrum of 1b (δ = −0.15 mm/s and ΔEQ = 0.98 mm/s) exhibit small changes as compared to the corresponding values of 1a (δ = −0.17 mm/s and ΔEQ = 0.85 mm/s). Generally, the found hyperfine parameters refer to the presence of Fe(III) cations in the low-spin state in both complexes. Small increase in the quadrupole splitting value is evidently related to a small decrease in the symmetry of the iron atom surroundings as a consequence of the topotactic dehydration process. Similar findings, connecting with the variations in Mössbauer parameters depending on a number of crystal water molecules presented in the structure, were previously observed for a series of M3[Fe(CN)6]2 · nH2O complexes (M = Ni, Mn, Co, Cu, Cd, n can rise up to 14) [61]. There is only very small difference in measured IR spectra between 1a and 1b, because all above described absorption bands for 1a are also present in the spectrum of 1b (Figure 4).

Table 4 Room temperature Mössbauer parameters and the corresponding elemental analyses of complex 1a as well as the samples prepared by its dynamic heating up to 130–600°C

The single crystal character is preserved during the dehydration process of 1a to 1b what is, generally, a very rare phenomenon among coordination compounds although it is quite common for simple salts, e.g. K0.35CoO2· 0.4H2O [65], manganese(II) formate [66] or FePO4[67]. To our knowledge, the single crystal dehydration process was reported only for a series of polymeric compounds {[M(H2O)2(C4O4)(bipy2)]∙ H2O}n (M = Co, Ni, Fe) [68], and [Ni(en)3]3[Fe(CN)6]2 · 2H2O [60] up to date.

The molecular structure of the hemihydrate 1b (Figure 1, Tables 1 and 2) is analogous to that of the original dihydrate 1a, since it is composed of the same complex cations and anions. Main, but not significant, structural variances between 1a and 1b are caused by the presence of different number of crystal water molecules. Thus, the dehydration process is accompanied by small shifts only in the positions of the respective [Co(en)3]3+ and [Fe(CN)6]3– ions as can be deduced from the mean values of the CoFe distances to the six nearest neighbors which were found to be 6.32(4) and 6.22(4) Å in the crystal structures of 1a, and 1b, respectively. Both 1a and 1b crystallize in the monoclinic system in related space groups (i.e. P21/n for the dihydrate and P21/c for the hemihydrate), but with the different unit cell orientations towards the presented molecules. The relationship between two unit cell settings can be expressed by the equation (aH, bH, cH) = (010,100, 00–1) (aD, bD, cD), where, aD, bD, cD and aH, bH, cH stand for cell parameters of the hemihydrate (H), and dihydrate (D), respectively. It should be noted that the monoclinic angles in both unit cells [βD = 91.982(4)° and βH = 90.15(1)°] display only minor deviation from the right angle. As expected, the unit cell dimensions of the hemihydrate 1b are somewhat shorter in lengths as compared to the dihydrate 1a. Therefore the dehydration process from the dihydrate (1a) to the hemihydrate (1b) can be considered as topotactic because of minimal geometric changes (less than 4%) and minimal atomic and molecular movements (change of each intra-/intermolecular distances < 4.2 Å) [53].

Despite considerable similarity between both structures, also some differences can be found. The partial dehydration caused conformational change within the [Co(en)3]3+ cations. While the Λδδδ (and Δλλλ) conformation was observed in 1a, the Λλδδ (and Δδλλ) conformation was found in 1b (Figure 6). It can be assumed that these conformational variations are induced by changes in the hydrogen bonding interactions (see below) as was previously suggested in conformation analysis of [M(en)3]3+ (M = CoIII, CrIII) [54]. Moreover, together with the conformational changes within the [Co(en)3]3+ cations, the shortening in the Co–N bond lengths in the hemihydrate 1b (1.956(2)–1.968(2) Å) were found as compared to those in 1a (1.955(2)–1.981(2) Å). However, all these values are comparable to those determined for the same cation in similar systems (1.97(7) Å) [33].

Figure 6
figure 6

Molecular structure of [Co( en ) 3 ][Fe(CN) 6 ] 1/2H 2 O (1b). The thermal ellipsoids are drawn at the 50% probability level.

Two crystallographically independent centrosymmetric [Fe(CN)6]3– moieties (both iron(III) atoms occupy special positions – the inversion centers) are present in the structure of 1b, similarly as in the structure of 1a. Both Fe1C6 and Fe1AC6 octahedra in 1b are slightly deformed. The octahedron around the Fe1A atom is compressed (Table 2) with the axial Fe1A–C2A distances (1.927(3) Å) being shorter than the equatorial ones (1.947(3) and 1.943(3) Å). Similar compression was found in the vicinity of the Fe1 atom with the axial Fe1–C3 distance (1.928(3) Å) and longer equatorial distances (1.946(3) and 1.941(3) Å). All these Fe–C bond lengths as well as other interatomic parameters are from the range of typical values for [Fe(CN)6]3– anions.

Generally, it can be anticipated, that the topotactic dehydration may influence significantly an extent of hydrogen bonds and other non-covalent contacts presented within the crystal structures. Thus, as a consequence of this general statement, the number of the above-mentioned non-bonding interactions is slightly lower, and moreover, they are moderately shorter than those in the crystal structure of 1a. While in the crystal structure of the dihydrate 1a, the water molecules are connected with [Fe(CN)6]3– anions via hydrogen bonds of the type O–HN and O–HO, the O–HO hydrogen bond is absent in the crystal structure of 1b (Table 3, Figure 7).

Figure 7
figure 7

Part of the crystal structure of [Co( en ) 3 ][Fe(CN) 6 ] 1/2H 2 O (1b) showing selected non-covalent contacts (dashed lines). Hydrogen atoms of the CH2 groups of the en ligands are omitted for clarity. The thermal ellipsoids are drawn at the 50% probability level.

The magnetic properties of the hemihydrate 1b (see Additional file 1: Figure S2 and S3) are similar compared to those for the dihydrate 1a. Fitting of the temperature dependence of reciprocal susceptibility above 50 K by Curie-Weiss law gave g = 2.77 and Θ = −18 K indicating slightly weaker antiferromagnetic interaction in comparison with 1a. The effective magnetic moment of 1b at room temperature is similar to the value for 1a, but a small difference becomes visible when the temperature declines below 50 K (Additional file 1: Figure S2) where the values of μeff for 1b are slightly higher than those for 1a and thus, indicating weaker antiferromagnetic interactions. Such observation is in accordance with the topotactic dehydration process which reduces the number of intra-molecular interactions as was observed in the crystal structure of 1b. The completed CIF files for 1a and 1b can be found as Additional files 2, and 3, respectively.

Chemical, structural and spectral characterization of the intermediates formed during the second and third decomposition steps

To explain the unexpected exothermic effects in DSC curves accompanying the evolution of molecules of en at higher temperatures and generally, to understand the structure and chemical composition of intermediates formed during the appropriate stage of the thermal treatment, the complex 1a was heated dynamically in conditions of the DSC experiment up to the selected temperatures and the samples were analyzed consequently by Mössbauer and IR spectroscopy and elemental analysis. The results obtained from these measurements together with the proposed composition of conversion intermediates are summarized in Table 4. The suggested decomposition model for 1a is depicted in Scheme 1.

Scheme 1
scheme 1

Suggested scheme of thermal decomposition of [CoIII( en ) 3 ][FeIII(CN) 6 ] 2H 2 O (1a) during dynamic heating in air atmosphere. (Endo stands for endo-effects, while Exo for exo-effects).

When 1a was heated up to 240°C (sample 2), the color of sample turned from orange to green. It was demonstrated by DSC that this solid underwent an exo-effect with the maximum at 208°C followed by an endo-effect with the minimum at 214°C. The Mössbauer spectrum of 2 consists of four doublets which correspond to iron-containing species in two different oxidation states (see Figure 3 and Table 4). Three doublets with δ values ranging from 1.02 to 1.31 mm/s correspond clearly to the high spin Fe(II) atoms [69, 70]. The high values of quadrupole splitting parameters (1.95–2.53 mm/s) manifest a very low symmetry of the Fe(II) environment as expected for the coordination number 5 [69]. The presence of three broadened Fe(II) doublets varying in the values of ΔEQ is probably related to the various degrees of the deformation of the iron(II) environment. The most populated doublet (δ = 0.22 mm/s; ΔEQ = 0.60 mm/s) can be ascribed to a high spin Fe(III) environment in tetrahedral coordination [70]. The increase of the isomer shift in comparison with 1a or 1b can be attributed to the iron spin transition from low to high spin as well as to the different coordination mode of cyanide ion to Fe(III) (FeIII–CN in 1a,b, FeIII–NC in 2) [50]. The results of Mössbauer spectroscopy give an evidence for the surprising change of the spin state of all iron atoms between 130 and 240°C as all the spectral components correspond to the high-spin states. Moreover, the presence of divalent iron indicates a central atom reduction which accompanies oxidation of CN to (CN)2. The formed cyanogen can react with two molecules of en which are also liberated during this thermolytic step and one bicyclic molecule of bis2-2-imidazolinyl) [71] could be formed as a potential product of the decomposition (see Scheme 1). All the mentioned experimental results can be explained by the formation of the bridging [FeII(en)2(μ-NC)CoIII(CN)4] complex from the hemihydrate 1b (see step II in Scheme 1) what is in accordance with the preference of HS Fe(II) to be coordinated in weaker ligand field provided by N-donor atoms rather than C-donor atoms. It seems that the complete dehydration induces a drastic rearrangement of the environments of both central atoms related to the evolution of half molecule of cyanogen and one molecule of en from the formula unit of 1b, which is probably reflected on the DSC curve by the mentioned exo-effect at 208°C. Nevetheless, similar process of deamination has been described previously on other systems containing [Co(en)3]3+ and it was always accompanied by a strong endo-effect above 200°C on the DSC curve [72]. Thus, the following endo-effect at 214°C on the DSC curve could be ascribed to the liberation of en as well.

The simultaneous presence of the high spin Fe(II) and Fe(III) species in sample 2 can be related to the partial transformation of [FeII(en)2(μ-NC)CoIII(CN)4] to [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] (see step III in Scheme 1). This charge transfer process has been previously observed and confirmed in many FeII–CN–CoIII species (light-induced magnetization changes) [1923] and here, it is more pronounced at higher temperature as clearly seen from the comparison of the Fe(II)/ΣFe ratio of 0.34 found in sample 2 heated up to 240°C with that observed in sample 3 heated up to 300°C (0.18; see Table 4 and Figure 3). In this third decomposition step we assume the participation of atmospheric oxygen and gradual evolution of NOX. This assumption is in an accordance with the N/C ratio decrease between 240 (0.94) and 300°C (0.84; see Table 4). The step III is manifested in DSC curve through a long endo-effect with the minimum at 214°C or rather long exo-effect in the temperature range of 220–300°C without specified maximum (see Figure 5). It is worth to mention that both samples 2 and 3 revealed a drastic lowering of conductivity compared to hemihydrate 1b, which is in line with the presence of bridging ligands between Co and Fe central atoms in the dinuclear species.

IR spectra recorded for the intermediates in sample 2 and 3 (230 and 265°C, Figure 4) confirmed proposed thermal decomposition mechanism. The υ(C≡N) stretching vibration is significantly governed by the electronegativity, oxidation state and coordination number of the metal central atom directly bonded to the cyanide group. Thus, the formation of cyanido-bridged species is accompanied by a significant broadening of the υ(C ≡ N) bands observed in IR spectra. Furthermore, the shift of the peak at 2113 cm–1 toward higher wave numbers with increasing temperature (2113 → 2127 → 2132 cm–1 for 150 → 230 → 265°C) documents the oxidation state changes and could be assigned to υ(FeII–CN–CoIII). The intensity of this vibration band decreases with higher temperature which indicates absence of such type of coordination above 265°C. On the other hand, the shift of the peak at 2104 cm–1 toward lower wave numbers with increasing temperature (2104 → 2062 → 2066 cm–1 for 150 → 230 → 265°C) corresponds to υ(CoII–CN–FeIII) [50]. This observation confirms the proposed decomposition mechanism concerning increasing amount of [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] at higher temperature (see step III in Scheme 1 and Table 4) and it indicates that CoII–CN–FeIII is the most probable coordination bridge in [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3]. Although non-linear coordination mode of bridging cyanide in di/polynuclear complexes is known, the suggested structure of [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] requires two cyanide bridges of that type. Such coordination tension could be reduced if the formation of a structure with the cyclic tetranuclear {Fe2Co2(CN)4} core is considered. Moreover, compounds with similar core-structure have been described recently [73]. Another possibility is the formation of a polymeric structure with μ3-CN coordination mode found in the case of several Cu(I)–Zn(II) complexes.

Chemical, structural and spectral characterizations of the intermediates formed in the fourth decomposition step

Although the samples 2 and 3 differ in quantitative characteristics, mainly in the C and N contents and in the percentage of Fe(II) and Fe(III) species, their phase composition is similar from the qualitative viewpoint as evident from hyperfine parameters of Mössbauer spectra. The next principal chemical change in the chemical composition of the samples takes place after heating up to 350°C (sample 4). This fourth reaction step is reflected on the DSC curve at 335°C by a shoulder of exo-effect at 354°C and documented also by a considerable diminution of both C and N contents (see Table 4). In Mössbauer spectrum of the sample 4 (see Figure 3) heated up to 350°C, there are no indications of the divalent iron species in contrast to the room temperature spectra of samples 2 and 3. On the other hand, a new high-spin Fe(III) phase with δ = 0.46 mm/s and quadrupole splitting ΔEQ = 1.22 mm/s appears in the spectrum along with the residual doublet ascribed to [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] as its hyperfine parameters are almost unchanged compared to the spectrum of sample 3. The hyperfine parameters of the former species are in a very good accordance with those observed for iron complexes with pentacoordinated high spin Fe(III) environments [74]. Thus, this new phase with the low symmetry of iron environment, which is manifested by a high value of quadrupole splitting, can be ascribed to FeIII[CoII(CN)5]. As for the expected chemical nature of the gases evolved during the fourth decomposition step leading from [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] to FeIII[CoII(CN)5] (see step IV in Scheme 1), their experimental identification using evolved gas analysis (EGA) may be ambiguous due to the negative role of atmospheric nitrogen and oxygen. Nevertheless, previous decomposition studies of other complexes with en, e.g. [Ni(en)3](ox) [75] using EGA showed that en is decomposed to a mixture of H2, CO/CO2, N2 and CH2 = CH2 in an inert atmosphere. Thus, the decomposition of ethylamine in air may involve similar or more oxygen containing species such as H2O, CO, CO2, N2 or NOx.

Characterization of the final decomposition product – the fifth decomposition step

In accordance with TG data, decomposition process is completed in dynamic conditions at 420°C. As mentioned above, the overall weight loss corresponds well with the final formation of the M3O4 (M = Fe, Co) phase(s) which is in accordance with the absence of υ(C ≡ N) stretching vibration bands in IR spectrum (Figure 4, 450°C). Taking into account the starting molar ratio of Fe/Co = 1/1, various combinations of several oxides (CoFe2O4, Fe3O4, Fe2O3 polymorphs, Co3O4, FeCo2O4) come on force to express the phase composition of the final product. The isomer shift values in room temperature Mössbauer spectra of the final decomposition products formed after dynamic heating of 1a up to 450 and 600°C (samples 5, 6 and 7, see Table 4 and Figure 8) unambiguously exclude the presence of divalent iron (Fe3O4, FeCo2O4) in the iron oxide phase [76]. The spectra of the samples 5 and 6 heated up to 400 and 450°C display a broadened doublet as a prevailing component, which can be ascribed to superparamagnetic nanoparticles. Superparamagnetic character of the spectra disallows the identification of iron(III) oxide phase (CoFe2O4 vs. γ-Fe2O3 vs. α-Fe2O3) in all samples. In the spectrum of sample 6 heated to 450°C, the prevailing dublet is accompanied by a completely magnetically split sextet which becomes the major signal observed in the spectrum of sample 7 prepared by heating of 1a up to 600°C (see Figure 8). This reduction of the ratio between the intensity of the dublet (superparamagnetic particles) and sextet manifests the formation of larger particles (~50 nm) due to the proceeding crystallization process. The room temperature spectrum of sample 7 can be fitted by two six-line hyperfine patterns with the nearly the same isomer shift (0.29 and 0.37 mm/s) and hyperfine magnetic fields of 48.6 and 51.4. Generally the isomer shift parameters are closed to those reported for CoFe2O4[77] and γ-Fe2O3 (0.27 mm/s) [78] or hematite, α-Fe2O3 (0.38 mm/s) [78]. Two sextets differing in isomer shift and hyperfine magnetic field can be ascribed to Fe(III) ions located at the tetrahedral (Td), δ = 0.37 mm/s, Hhyp = 51.4 T, and octahedral (Oh), δ = 0.29 mm/s, Hhyp = 48.6 T, sites of the partially inverse spinel structure of CoFe2O4. In comparison with bulk CoFe2O4, (51.2 T for Td-site, 55.0 T for Oh-site) the values of Hhyp of prepared CoFe2O4 nanoparticles (42.6 T for sample 6, 48.6 and 51.4 T for sample 7) are smaller due to their lower crystallinity [79].

Figure 8
figure 8

Room temperature Mössbauer spectra of the decomposition products prepared by dynamic heating of 1a up to 450°C (sample 6) and 600°C (sample 7).

In Figure 9, the XRD spectrum of sample 7 is shown, which consists of two sets of peaks corresponding to two spinel phases. The presence of γ-Fe2O3 or α-Fe2O3 is clearly excluded because the first set of peaks corresponds to CoFe2O4 phase [77] (marked with asterisks) and the second pattern belongs to Co3O4 (labeled with short vertical lines). According to the XRD data, the molar ratio between these two phases was found to be 2.44 : 1 (CoFe2O4 : Co3O4) what is in accordance with the proposed stoichiometry of both phases.

Figure 9
figure 9

XRD pattern of the sample prepared by heating of 1a up to 600°C. (*) CoFe2O4, (|) Co3O4.

To confirm the inverse spinel phase formed during the heating of 1a up to 600°C and reveal the magnetic structure, the in-field Mössbauer spectrum of sample 7 was recorded (Figure 10, Table 5). It consists of two overlapping six-line hyperfine patterns corresponding to Fe(III) in Td- and Oh-sites of spinel lattice (partial separation of both patterns). The data were successfully fitted with two sextets having the δ = 0.38 mm/s, ΔEQ = −0.02 mm/s and Hhyp = 52.6 T corresponding to Fe(III) in Td-site and δ = 0.54 mm/s, ΔEQ = −0.08 mm/s and Hhyp = 49.0 T for Fe(III) in Oh-sites, respectively. The degree of conversion for inverse spinel, x, which identifies its formula Fe(III)xCo(II)1–x[Co(II)xFe(III)2–x]O4 (brackets indicate ions at Oh-site), can be calculated on the basis of the relative contribution of individual Td- and Oh-site sextets. The value of x was obtained from the equation x = 2 k(1 + k) where k is the ratio between Fe(III) ions in the tetrahedral and octahedral sublattices calculated from the ratio between the areas of individual sextets corresponding to Td- and Oh-sites. The obtained value of x = 0.9 indicates the formation of partially/almost fully inverse spinel structure Fe(III)0.9Co(II)0.1[Co(II)0.9Fe(III)1.1]O4. The determination of the degree of inversion at low temperature in external magnetic field is more accurate in comparison with room temperature spectra (Figure 8) because the applied magnetic field better separates the Td- and Oh-site contributions (sextets) and the low temperature reduces the vibrations of the lattice and/or the Mössbauer nucleus (significantly influence the ration between the Td- and Oh-signals).

Figure 10
figure 10

In-field (5 K / 5 T) Mössbauer spectrum of the CoFe 2 O 4 spinel phase formed by heating of 1a up to 600°C. Crosses represent the experimental data and the upper and middle full lines represent the subspectra obtained from the fitting procedure for iron atoms in T- and O-sites, respectively, and the lower full line represents their sum. External field was applied parallel to the γ ray propagation.

Table 5 Hyperfine parameters of in-field Mössbauer spectrum of sample 7 at 5 K and in external field 5 T

The intensities of 2nd and 5th lines are lower than expected for a thin sample with randomly oriented magnetic moments in the absence of magnetic field (ideally 3:2:1:1:2:3) which is characteristic for the materials in ferromagnetic and/or ferrimagnetic state and which corresponds to the partial alignment of the iron magnetic moments in the direction of the external magnetic field. If the iron magnetic moment were completely aligned, the 2nd and 5th lines would be completely vanished. Thus, the presence of the 2nd and 5th lines in the spectrum points out to the canting of Fe(III) spin in CoFe2O4 particles with respect to the direction of the external magnetic field. Based on the comparison of the areas corresponding to 2,5 and 1,6 lines, the average spin-canting angle, θ, was evaluated on the basis of the formula θ = arccos{[(4–s)/(4 + s)]1/2} [80] where s = A2,5/A3,4 and was calculated to be 30.3° (from both Td- and Oh-sites). The thickness of the spin-canted surface layer of the sample was not deduced because the shape of prepared particles was not considered as spherical.

The morphology of the product formed by heating of 1a up to 600°C (sample 7) is clearly visible on TEM pictures (Figure 11), which confirm the nanocomposite structure of the CoFe2O4–Co3O4 particles. But these nanoparticles, taken directly after the thermal decomposition, form agglomerates and are not well-dispersed and thus, their median size based on TEM analysis was not calculated.

Figure 11
figure 11

TEM micrograph of the sample prepared by heating of 1a up to 600°C demonstrating the presence of CoFe 2 O 4 -Co 3 O 4 in the form of nanocomposite particles.


The thermal decomposition of [Co(en)3][Fe(CN)6]∙ 2H2O (1a) in air atmosphere was monitored by TG/DSC techniques. Seven samples were prepared by annealing 1a at selected temperatures, which were completely characterized by elemental analyses, IR and 57Fe Mössbauer spectroscopies with the aim to elucidate the decomposition mechanism. It has been found that the topotactic dehydration of 1a led to the formation of the hemihydrate [Co(en)3][Fe(CN)6]∙ 1/2H2O (1b), which was consequently decomposed in next four steps, going through the [FeII(en)2(μ-NC)CoIII(CN)4], [FeIII(NH2CH2CH3)2(μ-NC)2CoII(CN)3] and FeIII[CoII(CN)5] intermediates, leading to CoFe2O4-Co3O4 nanocomposite particles at as the final product (confirmed by TEM). It has been shown that a relatively small change regarding the heating gradient (5°C/min in this work versus 10°C/min in Ref. 52) may substantially influence the composition and particle size of the final product. Moreover, from a general point of view, the obtained results also revealed that the thermal decomposition of such a simple salt as [Co(en)3][Fe(CN)6]∙ 2H2O involves a relatively complicated mechanism driven mainly by the reaction conditions (such as the heating gradient and air atmosphere) and is composed by several overlapping steps and resulting in the formation of interesting nanoparticles with a nanocomposite structure of two magnetically different phases. It may be anticipated that the modulation of their properties would be possible by the exact control of the temperature (gradient or prolonged heating or amine-ligand substitution) and further investigation of magnetic properties of such system would be very interesting and helpful.


Materials and reagents

K3[Fe(CN)6] (Merck) and N,N’-dimethylformamide (Lachema Brno) were purchased from commercial sources and used as received. [Co(en)3]Cl3∙ 3H2O was prepared by a slightly modified procedure described in the literature [81].


[Co(en)3][Fe(CN)6]∙ 2H2O (1a).

0.39 g (1 mmol) of [Co(en)3]Cl3∙ 3H2O and 0.32 g (1 mmol) K3[Fe(CN)6] were added into 20 ml of N,N’-dimethylformamide (DMF) with stirring and refluxed for 1 h. The yellow-orange precipitate formed in the reaction mixture and the color of the solvent changed into light-green. The solid was filtered off, and consequently dissolved in 30 ml of warm water (60°C). The solution was left to stand at room temperature. Yellow-orange crystals were obtained by slow evaporation of the solvent after several hours. They were filtered off, washed with small amount of cold water and dried on air (yield 55%). Anal. calc. for C12H28N12CoFeO2: C, 29.8; H, 5.8; N, 34.7%. Found: C, 30.2; H, 5.5; N, 34.4%. IR: 3381 bm, 3279 s, 3232 bs, 3095 bs, 2117 vs, 2107 vs, 1630 m, 1609 m, 1578 m, 1464 m, 1399w, 1334 m, 1289 w, 1156 m, 1143 m, 1051 m, 763 w, 580 m.

Physical measurements

Elemental analyses (C, H, N) were performed on a Flash EA 1112 Elemental Analyser (ThermoFinnigan). Thermogravimetric (TG) and differential scanning calorimetry (DSC) measurements of 1a were performed in static air atmosphere under dynamic heating conditions with a sample weight of 6–8 mg and heating rate 5°C/min. DSC measurements were performed in the range of 20–400°C on a DSC XP-10 analyser (THASS, GmbH) and TG curve was measured between 20–500°C on a TG XP-10 instrument (THASS, GmbH). To monitor the intermediates of the thermal decay, the samples were prepared in conditions of DSC measurement by dynamic heating up to the selected temperatures, with the heating rate of 5°C/min and consequently analyzed by suitable techniques. The transmission 57Fe Mössbauer spectra were collected at 300 K in a constant acceleration mode with a 57Co(Rh) source. The final decomposition product was also analyzed at 5 K in zero external magnetic field and at 5 K in an external magnetic field of 5 T, applied parallel to the X-ray direction using a cryomagnetic system of Oxford Instruments. The isomer shift values were calibrated against metallic α-Fe. In-field Mössbauer spectra were fitted by means of the Lorentzian line shapes using the least squares method featured in the MossWinn computer program. FT IR spectra were measured on a Nexus 670 FT-IR (Thermo Nicolet) spectrophotometer using KBr pellets in the wavenumber range of 400–4000 cm–1 at temperatures 25, 150, 230, 265 and 450°C.

The transmission electron microscopy was performed on a JEOL JEM-2010 instrument (LaB6 cathode; accelerating voltage of 200 kV; point-to-point resolution 0.19 nm) to consider the particle size and structure of the final spinel phase. A drop of high-purity distilled water, containing the ultrasonically dispersed particles, was placed onto a holey-carbon film supported by a copper-mesh TEM grid and dried in air at room temperature.

X-ray structure determinations and refinements

X-ray single crystal diffraction experiments for both 1a and 1b were carried out on a four circle κ-axis Xcalibur2 diffractometer at 100(2) K equipped with a CCD detector Sapphire2 (Oxford Diffraction). The CrysAlis software package [82] was used for data collection and reduction. The structures were solved by the SIR97 program [83] incorporated within the WinGX program package [84]. All non-hydrogen atoms of 1a and 1b were refined anisotropically by the full-matrix least-squares procedure [85], with weights: w = 1/[σ2(Fo2) + (0.0633P)2] for (1a) and w = 1/[σ2(Fo2) + (0.0605P)2 + 0.3587P] for (1b), where P = (Fo2 + 2Fc2)/3. Hydrogen atoms were found in the difference Fourier maps, nevertheless these were modeled using a riding model, except for those belonging to crystal water molecules, which parameters were fixed. The structures of 1a and 1b, depicted in Figures 1, 2, 6 and 7, were drawn using the Diamond software [86]. X-ray powder diffraction pattern of the final decomposition product was recorded using a PANalytical X’Pert MPD device (iron-filtered CoKα radiation: λ = 0.178901 nm, 40 kV and 30 mA) in Bragg-Brentano geometry, and equipped with X’Celerator detector. Powdered sample was spread on zero-background single-crystal Si slides and scanned in continuous mode (resolution of 0.017° 2 Theta, scan speed of 0.005° 2 Theta per second) within the angular range 10-120o. The acquired patterns were processed (i.e., phase analysis and Rietveld refinement) using X´ Pert HighScore Plus software (PANalytical, The Netherlands), in combination with PDF-4+ and ICSD databases (ICSD collection codes are: Co3O4 – 63165, CoFe2O4 – 98553).





2-(pyridin-2-yl)pyridine = 2,2-bipyridine


4-(pyridin-4-yl)pyridine = 4,4-bipyridine


Differential scanning calorimetry


Evolved gas analysis


Ethylenediamine = ethane-1,2-diamine








Prussian blue analogue


Transmission electron microscopy


2,6-di(pyridin-2-yl)pyridine = 2,2:6,2-terpyridine






X-ray powder diffraction.


  1. Verdaguer M, Girolami GS, Miller JS, Drillon M: Magnetic Prussian Blue Analogs. Magnetism: Molecules to Materials, Volume 5. 2005, Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA

    Google Scholar 

  2. Catala L, Volatron F, Brinzei D, Mallah T: Functional Coordination Nanoparticles. Inorg Chem. 2009, 48: 3360-3370. 10.1021/ic8012574.

    Article  CAS  Google Scholar 

  3. Ricci F, Palleschi G: Sensor and biosensor preparation, optimisation and applications of Prussian Blue modified electrodes. Biosens Bioelectron. 2005, 21: 389-407. 10.1016/j.bios.2004.12.001.

    Article  CAS  Google Scholar 

  4. Jayalakshmi M, Scholz FJ: Performance characteristics of zinc hexacyanoferrate/Prussian blue and copper hexacyanoferrate/Prussian blue solid state secondary cells. J Power Sources. 2000, 91 (2): 217-223. 10.1016/S0378-7753(00)00475-4.

    Article  CAS  Google Scholar 

  5. Hu B, Fugetsu B, Yu H, Abe Y: Prussian blue caged in spongiform adsorbents using diatomite and carbon nanotubes for elimination of cesium. J Hazard Mater. 2012, 217–218: 85-91.

    Google Scholar 

  6. Okubo M, Asakura D, Mizuno Y, Kim JD, Mizokawa T, Kudo T, Honma I: Switching Redox-Active Sites by Valence Tautomerism in Prussian Blue Analogues AxMny[Fe(CN)6].nH2O (A: K, Rb): Robust Frameworks for Reversible Li Storage. J Phys Chem Lett. 2010, 1: 2063-2071. 10.1021/jz100708b.

    Article  CAS  Google Scholar 

  7. Uemura T, Ohba M, Kitagawa S: Size and Surface Effects of Prussian Blue Nanoparticles Protected by Organic Polymers. Inorg Chem. 2004, 43: 7339-7345. 10.1021/ic0488435.

    Article  CAS  Google Scholar 

  8. Liang C, Liu P, Xu J, Wang H, Wang W, Fang J, Wang Q, Shen W, Zhao J: A Simple Method for the Synthesis of Fe-Co Prussian Blue Analogue with Novel Morphologies, Different Structures, and Dielectric Properties. Synth React Inorg, Met-Org, Nano-Met Chem. 2011, 41 (9): 1108-10.1080/15533174.2011.591354.

    Article  Google Scholar 

  9. Koncki R: Chemical Sensors and Biosensors Based on Prussian Blues. Crit Rev Anal Chem. 2002, 32: 79-96. 10.1080/10408340290765452.

    Article  CAS  Google Scholar 

  10. DeLongchamp DM, Hammond PT: Multiple-Color Electrochromism from Layer-by-Layer-Assembled Polyaniline/Prussian Blue Nanocomposite Thin Films. Chem Mater. 2004, 16: 4799-4805. 10.1021/cm0496624.

    Article  CAS  Google Scholar 

  11. Bustos L, Godinez LA: Modified Surfaces with Nano-Structured Composites of Prussian Blue and Dendrimers. New Materials for Advanced Electrochemical Applications. Int J Electrochem Sci. 2011, 6: 1-36.

    CAS  Google Scholar 

  12. de Tacconi NR, Rajeshwar K: Metal Hexacyanoferrates: Electrosynthesis, in Situ Characterization, and Applications. Chem Mater. 2003, 15: 3046-3062. 10.1021/cm0341540.

    Article  CAS  Google Scholar 

  13. Verdaguer M, Bleuzen A, Marvaud V, Vaissermann J, Seuleiman M, Desplanches C, Scuiller A, Train C, Garde R, Gelly G, Lomenech C, Rosenman I, Veillet P, Cartier C, Villain F: Molecules to build solids: high TC molecule-based magnets by design and recent revival of cyano complexes chemistry. Coord Chem Rev. 1999, 192: 1023-1047.

    Article  Google Scholar 

  14. Dunbar KR, Heintz RA: Chemistry of Transition Metal Cyanide Compounds: Modern Perspectives. Progress in Inorganic Chemistry, Volume 45. 2007, New York: Karlin KD. John Wiley, 283-391.

    Google Scholar 

  15. Herrera JM, Bachschmidt A, Villain F, Bleuzen A, Marvaud V, Wernsdorfer W, Verdaguer M: Mixed valency and magnetism in cyanometallates and Prussian blue analogues. Phil Trans R Soc A. 2008, 366: 127-138. 10.1098/rsta.2007.2145.

    Article  CAS  Google Scholar 

  16. Sato O: Optically Switchable Molecular Solids: Photoinduced Spin-Crossover, Photochromism, and Photoinduced Magnetization. Acc Chem Res. 2003, 36: 692-700. 10.1021/ar020242z.

    Article  CAS  Google Scholar 

  17. Talham DR, Meisel MW: Thin films of coordination polymer magnets. Chem Soc Rev. 2011, 40: 3356-3365. 10.1039/c1cs15015d.

    Article  CAS  Google Scholar 

  18. Holmes SM, Girolami GS: Sol–gel Synthesis of KVII[CrIII(CN)6].2H2O: A Crystalline Molecule-Based Magnet with a Magnetic Ordering Temperature above 100°C. J Am Chem Soc. 1999, 121: 5593-5594. 10.1021/ja990946c.

    Article  CAS  Google Scholar 

  19. Sato O: Photoinduced magnetization in molecular compounds. J Photochem and Photobiol C: Chem. 2004, 5: 203-223. 10.1016/j.jphotochemrev.2004.10.001.

    Article  CAS  Google Scholar 

  20. Sato O, Tao J, Zhang YZ: Control of Magnetic Properties through External Stimuli. Angew Chem Int Ed. 2007, 46: 2152-2187. 10.1002/anie.200602205.

    Article  CAS  Google Scholar 

  21. Culp JT, Park JH, Frye F, Huh YD, Meisel MW, Talham DR: Magnetism of metal cyanide networks assembled at interfaces. Coord Chem Rev. 2005, 249: 2642-2648. 10.1016/j.ccr.2005.05.011.

    Article  CAS  Google Scholar 

  22. Bleuzen A, Lomenech C, Escax V, Villain F, Varret F, Cartier dit Moulin C, Verdaguer M: Photoinduced Ferrimagnetic Systems in Prussian Blue Analogues CI x Co4[Fe(CN)6] y (CI = Alkali Cation). 1. Conditions to Observe the Phenomenon. J Am Chem Soc. 2000, 122: 6648-6652. 10.1021/ja000348u.

    Article  CAS  Google Scholar 

  23. Sato O, Einaga Y, Fujishima A, Hashimoto K: Photoinduced Long-Range Magnetic Ordering of a Cobalt-Iron Cyanide. Inorg Chem. 1999, 38: 4405-4412. 10.1021/ic980741p.

    Article  CAS  Google Scholar 

  24. Ohba M, Okawa H: Synthesis and magnetism of multi-dimensional cyanide-bridged bimetallic assemblies. Coord Chem Rev. 2000, 198: 313-328. 10.1016/S0010-8545(00)00233-2.

    Article  CAS  Google Scholar 

  25. Colacio E, Ghazi M, Stoeckli-Evans H, Lloret F, Moreno JM, Perez C: Cyano-Bridged Bimetallic Assemblies from Hexacyanometalate, [M(CN)6]3- (M = MnIIIand FeIII), and [M(N4-macrocycle)]2+(M = FeIII, NiIIand ZnII) Building Blocks. Syntheses, Multidimensional Structures, and Magnetic Properties.Inorg Chem. 2001, 40: 4876-4883. 10.1021/ic0103446.

    Article  CAS  Google Scholar 

  26. Marvaud V, Decroix C, Scuiller A, Guyard-Duhayon C, Vaissermann J, Gonnet F, Verdaguer M: Hexacyanometalate Molecular Chemistry: Heptanuclear Heterobimetallic Complexes; Control of the Ground Spin State. Chem Eur J. 2003, 9: 1677-1691. 10.1002/chem.200390192.

    Article  CAS  Google Scholar 

  27. Funck KE, Hilfiger MG, Berlinguette CP, Shatruk M, Wernsdorfer W, Dunbar KR: Trigonal-Bipyramidal Metal Cyanide Complexes: A Versatile Platform for the Systematic Assessment of the Magnetic Properties of Prussian Blue Materials. Inorg Chem. 2009, 48: 3438-3452. 10.1021/ic801990g.

    Article  CAS  Google Scholar 

  28. Newton GN, Nihei M, Oshio H: Cyanide-Bridged Molecular Squares – The Building Units of Prussian Blue. Eur J Inorg Chem. 2011, 20: 3031-3042.

    Article  Google Scholar 

  29. Černák J, Orendáč M, Potočňák I, Chomič J, Orendáčová A, Skoršepa J, Feher A: Cyanocomplexes with one-dimensional structures: preparations, crystal structures and magnetic properties. Coord Chem Rev. 2002, 224: 51-66. 10.1016/S0010-8545(01)00375-7.

    Article  Google Scholar 

  30. Billing R: Optical and photoinduced electron transfer in ion pairs of coordination compounds. Coord Chem Rev. 1997, 159: 257-270.

    Article  CAS  Google Scholar 

  31. Billing R, Vogler A: Optical and photoinduced electron transfer in tris(ethylenediamine)cobalt(III)-cyanometallate ion pairs. J Photochem Photobiol A: Chem. 1997, 103 (3): 239-247. 10.1016/S1010-6030(96)04562-5.

    Article  CAS  Google Scholar 

  32. Poulopoulou VG, Li ZW, Taube H: Comparison of the rates of substitution in [Ru(NH3)5H2O]3+ and [Os(NH3)5H2O]3+ by hexacyano complexes: substitution coupled to electron transfer. Inorg Chim Acta. 1994, 225: 173-184. 10.1016/0020-1693(94)04045-1.

    Article  CAS  Google Scholar 

  33. Allen FH, Bellard S, Brice MD, Cartwright BA, Doubleday A, Higgs H, Hummelink T, Hummelink-Peters BG, Kennard O, Motherwell WDS, Rodgers JR, Watson DG: Cambridge Structural Database System (CSDS). 1994, Cambridge, U.K

    Google Scholar 

  34. Bernhardt PV, Bozoglian F, Macpherson BP, Martinez M: Molecular mixed-valence cyanide bridged CoIII–FeII complexes. Coord Chem Rev. 2005, 249: 1902-1916. 10.1016/j.ccr.2004.11.014.

    Article  CAS  Google Scholar 

  35. Bernhardt PV, Bozoglian F, Gonzalez G, Martınez M, Macpherson BP, Sienra B: Dinuclear Cyano-Bridged CoIII-FeII Complexes as Precursors for Molecular Mixed-Valence Complexes of Higher Nuclearity. Inorg Chem. 2006, 45: 74-82. 10.1021/ic0511423.

    Article  CAS  Google Scholar 

  36. Bernhardt PV, Martınez M, Rodrıguez C: Molecular CoIII/FeII Cyano-Bridged Mixed-Valence Compounds with High Nuclearities and Diversity of CoIII Coordination Environments: Preparative and Mechanistic Aspects. Inorg Chem. 2009, 48: 4787-4797. 10.1021/ic802198s.

    Article  CAS  Google Scholar 

  37. Seitz K, Peschel P, Babel D: On the Crystal Structure of the Cyanido Complexes [Co(NH3)6][Fe(CN)6], [Co(NH3)6]2[Ni(CN)4]3 2H2O, [Cu(en)2][Ni(CN)4]. Z Anorg Allg Chem. 2001, 627: 929-934. 10.1002/1521-3749(200105)627:5<929::AID-ZAAC929>3.0.CO;2-7.

    Article  CAS  Google Scholar 

  38. Bok LD, Leipoldt JG, Basson SS: The crystal structure of tris(ethylenediamine), cobalt(III) hexacyanoferrate(III)dihydrate, Co(C2H8N2)3[Fe(CN)6]2 · H2O. Z Anorg Allg Chem. 1972, 389: 307-314. 10.1002/zaac.19723890311.

    Article  CAS  Google Scholar 

  39. Elder RC, Kennard GJ, Payne MD, Deutsch F: Synthesis and Characterization of Bis(ethylenediamine)cobalt(III) Complexes Containing Chelated Thioether Ligands. Crystal Structures of [(en)2Co(S(CH3)CH2CH2NH2)][Fe(CN)6] and [(en)2Co(S(CH2C6H5)CH2COO)](SCN)2. Inorg Chem. 1978, 17: 1296-1303. 10.1021/ic50183a039.

    Article  CAS  Google Scholar 

  40. Maľarová M, Trávníček Z, Zbořil R, Černák J: [Co(en)3][Fe(CN)6] H2O and [Co(en)3][Fe(CN)6]: A dehydration process investigated by single crystal X-ray analysis, thermal analysis and Mössbauer spectroscopy. Polyhedron. 2006, 25: 2935-2943. 10.1016/j.poly.2006.04.021.

    Article  Google Scholar 

  41. Matiková-Maľarová M, Černák J, Massa W, Varret F: Three Co(III)–Fe(II) complexes based on hexacyanoferrates: Syntheses, spectroscopic and structural characterizations. Inorg Chim Acta. 2009, 362: 443-448. 10.1016/j.ica.2008.04.038.

    Article  Google Scholar 

  42. Ohashi Y, Yanagi K, Mitsuhashi Y, Nagata K, Kaizu Y, Sasada Y, Kobayashi H: Absolute Configuration of (−)D-Tris(2,2′-bipyridine)cobalt(III). J Am Chem Soc. 1979, 101: 4739-4740. 10.1021/ja00510a056.

    Article  CAS  Google Scholar 

  43. Yanagi K, Ohashi Y, Sasada Y, Kaizu Y, Kobayashi H: Crystal structure and Absolute Configuration of (−)589-Tris(2,2′-bipyridine)cobalt(III) Hexacyanoferrate(III) Octahydrate. Bull Chem Soc Jpn. 1981, 54: 118-126. 10.1246/bcsj.54.118.

    Article  CAS  Google Scholar 

  44. Singh N, Diwan K, Drew MGB: Supramolecular assemblies of metal ions in complex bimetallic and trimetallic salts based on hexacyanoferrate(III) ion. Polyhedron. 2010, 29: 3192-3197. 10.1016/j.poly.2010.08.032.

    Article  CAS  Google Scholar 

  45. Zhuge F, Wu B, Yang J, Janiak C, Tang N, Yang XJ: Microscale hexagonal rods of a charge-assisted second-sphere coordination compound [Co(DABP)3][Fe(CN)6]. Chem Commun. 2010, 46: 1121-1123. 10.1039/b914216a.

    Article  CAS  Google Scholar 

  46. Ulas G, Brudvig GW: Redirecting Electron Transfer in Photosystem II from Water to Redox-Active Metal Complexes. J Am Chem Soc. 2011, 133: 13260-13263. 10.1021/ja2049226.

    Article  CAS  Google Scholar 

  47. Kersting B, Siebert D, Volkrner D, Kolm MJ, Janiak C: Synthesis and Characterization of Homo- and Heterodinuclear Complexes Containing the N3M(μ2-SR)3MN3 Core (M = Fe, Co, Ni). Inorg Chem. 1999, 38: 3871-3882. 10.1021/ic990087t.

    Article  CAS  Google Scholar 

  48. Mohai B: Thermolysis of cyano complexes. VII. On the thermal decomposition of hexacyanocobaltate(III); ligand exchange during thermolysis. Z Anorg Allg Chem. 1972, 392: 287-294. 10.1002/zaac.19723920313.

    Article  CAS  Google Scholar 

  49. Horváth A, Mohai B: Thermolysis of complex cyanides - XIII Structural transformations at thermal decomposition of [M(en)3][M’(CN)5NO] double complexes. J Inorg Nucl Chem. 1980, 42: 195-199. 10.1016/0022-1902(80)80239-9.

    Article  Google Scholar 

  50. Ng CW, Ding J, Gan LM: Microstructural Changes Induced by Thermal Treatment of Cobalt(II) Hexacyanoferrate(III) Compound. J Solid State Chem. 2001, 156: 400-407. 10.1006/jssc.2000.9013.

    Article  CAS  Google Scholar 

  51. Ng CW, Ding J, Shi Y, Gan LM: Structure and magnetic properties of copper(II) hexacyanoferrate(III) compound. J Phys Chem Solids. 2001, 62: 767-775. 10.1016/S0022-3697(00)00248-1.

    Article  CAS  Google Scholar 

  52. Pechenyuk SI, Domonov DP, Gosteva AN, Kadyrova GI, Kalinnikov VT: Synthesis, Properties, and Thermal Decomposition of Compounds [Co(En)3][Fe(CN)6] · 2H2O and [Co(En)3]4[Fe(CN)6]3 · 15H2O. Russ J Coord Chem. 2012, 38: 596-603. 10.1134/S1070328412090060.

    Article  CAS  Google Scholar 

  53. Kaupp G: Solid-state reactions, dynamics in molecular crystals. Curr Opin Solid State Mater Sci. 2002, 6: 131-138. 10.1016/S1359-0286(02)00041-4.

    Article  CAS  Google Scholar 

  54. Nicketic SR, Rasmussen K: Conformational Analysis of Coordination Compounds. IV. Tris(1,2-ethanediamine)- and Tris(2,3-butanediamine)cobalt(III) complexes. Acta Chem Scand. 1978, A32: 391-400.

    Article  Google Scholar 

  55. Matsuki R, Shiro M, Asahi T, Asai H: Absolute configuration of Λ-(+)589-tris(ethylenediamine)cobalt(III) triiodide monohydrate. Acta Crystallogr., Sect. E: Struct. Rep. Online. 2001, 57: m448-m450. 10.1107/S1600536801015008.

    Article  CAS  Google Scholar 

  56. Ueda T, Bernard GM, McDonald R, Wasylishen RE: Cobalt-59NMR and X-ray diffraction studies of hydrated and dehydrated (±)-tris(ethylenediamine) cobalt(III) chloride. Solid State Nucl Magn Reson. 2003, 24: 163-183. 10.1016/S0926-2040(03)00049-3.

    Article  CAS  Google Scholar 

  57. Moron MC, Palacio F, Pons J, Casabo J, Solans X, Merabet KE, Huang D, Shi X, Teo BK, Carlin RL: Bimetallic Derivatives of [M(en)3I3+ Ions (M = Cr, Co): An Approach to Intermolecular Magnetic Interactions in Molecular Magnets. Inorg Chem. 1994, 33: 746-753. 10.1021/ic00082a021.

    Article  CAS  Google Scholar 

  58. Saha MK, Lloret F, Bernal I: Inter-String Arrays of Bimetallic Assemblies with Alternative Cu2+-Cl-Cu2+ and Cu-NC-M (M = Co3+, Fe+3, Cr+3) Bridges: Syntheses, Crystal Structure, and Magnetic Properties. Inorg Chem. 2004, 43: 1969-1975. 10.1021/ic034897n.

    Article  CAS  Google Scholar 

  59. Salah El Fallah M, Rentschler E, Caneschi A, Sessoli R, Gatteschi D: A Three-Dimensional Molecular Ferrimagnet Based on Ferricyanide and [Ni(tren)]2+ Building Blocks. Angew Chem Int Ed Engl. 1996, 35: 1947-1949. 10.1002/anie.199619471.

    Article  Google Scholar 

  60. Herchel R, Tuček J, Trávníček Z, Petridis D, Zbořil R: Crystal Water Molecules as Magnetic Tuners in Molecular Metamagnets Exhibiting Antiferro-Ferro-Paramagnetic Transitions. Inorg Chem. 2011, 50: 9153-9163. 10.1021/ic201358c.

    Article  CAS  Google Scholar 

  61. Reguera E, Férnandez-Bertrán J: Effect of the water of crystallization on the Mössbauer spectra of hexacyanoferrates (II and III). Hyperfine Interact. 1994, 88: 49-58. 10.1007/BF02068701.

    Article  CAS  Google Scholar 

  62. Nakamoto I: Infrared Spectra of Inorganic and Coordination Compounds. 2002, New York: John Wiley and Sons

    Google Scholar 

  63. Landolt-Bornstein GII: Atomic and Molecular Physics, Volume 2. 1966, Verlag: Springer

    Google Scholar 

  64. Comba P: Prediction and Interpretation of EPR Spectra of Low-Spin Iron(III) Complexes with the MM-AOM Method. Inorg Chem. 1994, 33: 4511-4583.

    Google Scholar 

  65. Tang HY, Lin HY, Wang MJ, Liao MY, Liu JL, Hsu FC, Wu MK: Crystallization and Anisotropic Properties of Water-Stabilized Potassium Cobalt Oxides. Chem Mater. 2005, 17: 2162-2164. 10.1021/cm047707v.

    Article  CAS  Google Scholar 

  66. Viertelhaus M, Henke H, Anson CE, Powell AK: Solvothermal Synthesis and Structure of Anhydrous Manganese(II) Formate, and Its Topotactic Dehydration from Manganese(II) Formate Dihydrate. Eur J Inorg Chem. 2003, 12: 2283-2289.

    Article  Google Scholar 

  67. Song Y, Zavalij PY, Suzuki M, Whittingham MS: New Iron(III) Phosphate Phases: Crystal Structure and Electrochemical and Magnetic Properties. Inorg Chem. 2002, 41: 5778-5786. 10.1021/ic025688q.

    Article  CAS  Google Scholar 

  68. Greve J, Jess I, Nather C: Synthesis, crystal structures and investigations on the dehydration reaction of the new coordination polymers poly[diaqua-(μ2-squarato-O, O’)-(μ2–4,4′-bipyridine-N, N’)Me(II)] hydrate (Me = Co, Ni, Fe). J Sol State Chem. 2003, 175: 328-340. 10.1016/S0022-4596(03)00306-2.

    Article  CAS  Google Scholar 

  69. Feist M, Troyanov SI, Mehner H, Witke K, Kemnitz E: Halogeno Metallates of Transition Elements with Cations of Nitrogen-containing Heterocyclic Bases. VII Two Oxidation States and Four Different Iron Coordinations in one Compound. Synthesis, Crystal Structure, and Spectroscopic Characterization of 1,4-Dimethylpiperazinium Chloroferrate (II,III), (dmpipzH2)6[FeIICl4]2 [FeIIICl4]2[FeIICl5] [FeIIICl6]. Z Anorg Allg Chem. 1999, 625: 141-146. 10.1002/(SICI)1521-3749(199901)625:1<141::AID-ZAAC141>3.0.CO;2-W.

    Article  CAS  Google Scholar 

  70. Wyrzykowski D, Maniecki T, Pattek-Janczyk A, Stanek J, Warnke Z: Thermal analysis and spectroscopic characteristics of tetrabutylammonium tetrachloroferrate(III). Thermochim Acta. 2005, 435: 92-98. 10.1016/j.tca.2005.05.007.

    Article  CAS  Google Scholar 

  71. Roodburn HM, Fisher JR: Reaction of Cyanogen with Organic Compounds. X. Aliphatic and Aromatic Diamines. J Org Chem. 1957, 22: 895-899. 10.1021/jo01359a010.

    Article  Google Scholar 

  72. Zsakó J, Pokol G, Novák C, Várhelyi C, Dobó A, Liptay G: KINETIC ANALYIS OF TG DATA XXXV. Spectroscopic and thermal studies of some cobalt(III) chelates with ethylenediamine. J Therm Anal Calorim. 2001, 64: 843-856. 10.1023/A:1011577319016.

    Article  Google Scholar 

  73. Nihei M, Sekine Y, Suganami N, Nakazawa K, Nakao A, Nakao H, Murakami Y, Oshio H: Controlled Intramolecular Electron Transfers in Cyanide-Bridged Molecular Squares by Chemical Modifications and External Stimuli. J Am Chem Soc. 2011, 133: 3592-3600. 10.1021/ja109721w.

    Article  CAS  Google Scholar 

  74. Sakai T, Ohgo Y, Hoshino A, Ikeue T, Saigon T, Takahashi M, Nakanuta M: Electronic Structures of Five-Coordinate Iron(III) Porphyrin Complexes with Highly Ruffled Porphyrin Ring. Inorg Chem. 2004, 43: 5034-5043. 10.1021/ic049825q.

    Article  CAS  Google Scholar 

  75. Fejitha KS, Mathew S: Thermoanalytical investigations of tris(ethylenediamine)nickel(II) oxalate and sulphate complexes. J Therm Anal Calorim. 2010, 102: 931-939. 10.1007/s10973-010-0823-8.

    Article  Google Scholar 

  76. Persoons RM, Degrave E, Van Denberghe RE: Mössbauer study of Co-substituted magnetite. Hyperfine Interact. 1990, 54: 655-660. 10.1007/BF02396107.

    Article  CAS  Google Scholar 

  77. Haneda K, Morrish AH: Noncollinear magnetic structure of CoFe2O4 small particles. J Appl Phys. 1988, 63: 4258-4261. 10.1063/1.340197.

    Article  CAS  Google Scholar 

  78. Greenwood NN, Gibb TC: Mössbauer spectroscopy. 1971, New York: Barnes and Noble Inc

    Book  Google Scholar 

  79. Fedotova YA, Baev VG, Lesnikovich AI, Milevich IA, Vorobeva SA: Magnetic Properties and Local Configurations of 57Fe Atoms in CoFe2O4 Powders and CoFe2O4/PVA Nanocomposites. Phys Sol State. 2011, 53: 647-653.

    Article  Google Scholar 

  80. Ngo AT, Bonville P, Pileni MP: Nanoparticles of CoxFey_zO4: Synthesis and superparamagnetic properties. Eur Phys J. 1999, B9: 583-592.

    Article  Google Scholar 

  81. Vivier V, Aguey F, Fournier J, Lambert JF, Bedioui F, Che M: Spectroscopic and Electrochemical Study of the Adsorption of [Co(en)2Cl2]Cl on γ-Alumina: Influence of the Alumina Ligand on Co(III)/(II)Redox Potential.J Phys Chem B. 2006, 110: 900-906. 10.1021/jp058224x.

    Article  CAS  Google Scholar 

  82. Diffraction O: CrysAlis CCD. 2006, Abingdon, England: Oxford Diffraction Ltd

    Google Scholar 

  83. Altomare A, Burla MC, Camalli M, Cascarano GL, Giacovazzo C, Guagliardi A, Moliterni AGG, Polidori G, Spagna R: SIR97: a new tool for crystal structure determination and refinement. J Appl Cryst. 1999, 32: 115-119. 10.1107/S0021889898007717.

    Article  CAS  Google Scholar 

  84. Farrugia LJ: WinGX suite for small-molecule single-crystal crystallography. J Appl Cryst. 1999, 32: 837-838. 10.1107/S0021889899006020.

    Article  CAS  Google Scholar 

  85. Sheldrick GM: A short history of SHELX. Acta Crystallogr, Sect A: Found Crystallogr. 2008, 64: 112-122. 10.1107/S0108767307043930.

    Article  CAS  Google Scholar 

  86. Brandenburg K: DIAMOND. Visual Crystal Structure Information System. Version 2.1e. 2000, Bonn, Germany: Crystal Impact GbR

    Google Scholar 

Download references


The authors gratefully thank the Operational Program Research and Development for Innovations - European Regional Development Fund (CZ.1.05/2.1.00/03.0058), Operational Program Education for Competitiveness- European Social Fund (CZ.1.07/2.3.00/20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic. The authors also acknowledge the financial support from the Grant Agency of the Czech Republic (Grant No. P207/11/0841). We also thank to J. Filip for XRD experiments data and to L. Machala and J. Čuda for the fitting of Mössbauer spectra. JČ thanks grant APVV-0014-11 for partial financial support.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Zdeněk Trávníček.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

MMM synthesized all samples and made substantial contributions to the data collection and analysis. BD made substantial contributions to the survey results, data analysis/interpretation and drafting the manuscript. ZT performed single-crystal diffraction experiments and calculations, and together with RZ participated in the overall data interpretations and were involved in drafting the manuscript. BD, ZT, RZ and JČ revised the manuscript for the intellectual content. All authors read and approved the final version of the manuscript.

Electronic supplementary material


Additional file 1: Electronic Supplementary Information. The file contains IR spectrum of 1a (Figure S1) and temperature dependences of the effective magnetic moment (μeff) and reciprocal molar susceptibility (1/χmol) (Additional file 1: Figures S2 and S3). (DOC 510 KB)

Additional file 2: A CIF file for complex 1a.(CIF 18 KB)

Additional file 3: A CIF file for complex 1b.(CIF 17 KB)

Authors’ original submitted files for images

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License ( ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Trávníček, Z., Zbořil, R., Matiková-Maľarová, M. et al. Thermal decomposition of [Co(en)3][Fe(CN)6]∙ 2H2O: Topotactic dehydration process, valence and spin exchange mechanism elucidation. Chemistry Central Journal 7, 28 (2013).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: