Nanocrystalline iron oxide and sulfide by the thermal decomposition of cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O

Abstract

We reported for the first time the establishment of a simple, room-temperature synthesis protocol for the new organic-inorganic hybrid salt of cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O, which was found a convenient single-source precursor for the synthesis of nanocrystalline iron oxide or sulfide. The formation of this salt was spectrophotometrically confirmed by FT-IR and UV-Vis. In addition, SCXRD revealed that this salt had the trigonal space group R-3 with disorder of some isothiocyanate sulfur atoms. The thermal stability and the thermal decomposition products of this salt were atmosphere-dependent (air: 169 °C; α-Fe₂O₃ at 550 °C; helium: 154 °C; FeS at 800 °C). The thermal decomposition impacted the textural properties of α-Fe₂O₃ (an average crystallite size of ~ 41 nm and S_BET = ~ 4.0 m²/g) and FeS (~ 14 nm and ~ 80 m²/g, respectively). The nanoparticulate nature affected the magnetic behavior of α-Fe₂O₃ and FeS, as revealed by ac-susceptibility. They showed widen maximum at ~ 55 K due to increasing disorder effect by particle sizes. However, below 40 K, the susceptibility increased sharply, indicating a ferromagnetic ordering. In comparison, the ac-susceptibility of the salt exhibited a broad maximum at ~ 130 K with an inflection point at ~ 180 K. No transition to spin-flip was detected for all three materials.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-98046-4.

1. Introduction

During the past few years, cyclohexylammonium based-compounds have aroused researchers’ attention in the materials science field owing to cyclohexylammonium ability to act as multi-hydrogen bond donor for hydrogen bonding with numerous counter anions. As a result, such intermolecular noncovalent interactions facilitate the construction of supramolecular networks^1–21. Furthermore, cyclohexylammonium based-compounds are unusual because of the flexibility in structural arrangement and diversity in dimensional attributes with varying anions^2,3,5. This flexibility allows the insertion of transition metals in anionic complexes to form functional materials^{4,5,10,15–17,20,21}. Among these metal complexes are the homoleptic isothiocyanate metal anionic complexes such as [Co(NCS)₄]^{2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17}, [Cr(NCS)₆]^3–20, and [Ni(NCS)₆]^4–21, which have numerous applications. For instance, the organic salts of these complexes were used as catalysts, analytical reagents, and thermally sensitive materials²². On the other hand, isothiocyanatoferrate complexes were relatively well-researched, in regard to other metals, due to their applicability in biochemistry, biology, electrochemistry, and in many different applications^23,24.

Anions of thiocyanate transition metal complexes have been proven effective building blocks in the development of organic-inorganic hybrid materials²⁵. The usage of such complex anions for assembling organic-inorganic hybrid materials has become widely spread due to the ability of thiocyanate to coordinate with metal ions in various coordination modes, leading to compounds with different structures and characteristics. Thiocyanate terminally binds to metals through nitrogen or sulfur atom according to the hard soft acid base (HSAB) principle. Thus, thiocyanate usually binds to the first-row transition metals through its nitrogen terminal. In addition, thiocyanate may act as a bridging ligand and link two metal centers through 1,1-µ-NCS, 1,1-µ-SCN, or 1,3-µ-NCS configurations. According to the literature, the terminally monodentate bonding mode is more widely known than the bridging mode^25–32. Another coordination modes, but are uncommon, are the three-way bridging mode (> NCS−) or (> SCN−)^25,32. Moreover, the thiocyanate ligand can mediate a magnetic exchange which can be utilized to design new materials with significantly different magnetic properties, depending on the metal ions and the topology of the coordination network^32–37.

Herein, the synthesis and the characterization of the novel organic-inorganic hybrid salt of cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O, (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O, were reported, for the first time. The synthesis approach to this salt was performed at room temperature and needed simple work up, making it a potentially economic route. Furthermore, the thermal decomposition of this salt, as a single-source precursor for the synthesis of nanocrystalline α-phase iron oxide (α-Fe₂O₃) under air and iron sulfide (FeS) under helium atmosphere was explored. The α-Fe₂O₃is useful in various appealing applications including catalysts³⁸, gas sensors³⁹, drug delivery⁴⁰, water remediation⁴¹, magnetic devices⁴², lithium batteries⁴³, and pigments⁴⁴. Whereas FeS has a wide range of applications. It is particularly useful in soil and water remediation^45–48, high-energy density batteries⁴⁹, superconductors^50,51, and catalysts⁵².

2. Experimental

2.1. Materials

Unless otherwise noted, all reagents were used as commercially obtained: Cyclohexylamine [C₆H₁₁NH₂; > 99.0%(GC); TCI], iron(III) chloride anhydrous [FeCl₃; 99%; Riedel-de Haën AG], ethanol absolute (EtOH; GRG, Shibuya-ku, Japan; 99.9%; PETROCHEM), isopropanol [(H₃C)₂CHOH, puriss. p.a., ACS reagent, reag. ISO, reag. Ph. Eur., ≥ 99.8% (GC), Sigma-Aldrich] hydrochloric acid (HCl; 37%; certified AR for analysis; Fisher Chemical; Fisher Scientific, Waltham, MA, USA), and sodium thiocyanate (NaNCS; pure; 99%; Riedel-de Haën AG).

2.2. Synthesis

Synthesis of (C₆H₁₁NH₃Cl), 1

Cyclohexylammonium chloride, C₆H₁₁NH₃Cl, was obtained via the neutralization reaction of an aqueous solution of cyclohexylamine (110 mmol, 12.6 mL diluted with 100 mL of cold water) with hydrochloric acid (110 mmol, ~ 9.2 mL diluted with 100 ml of cold water), according to Eq. 1:

1

The HCl solution was gradually added over a period of 30 min to the cyclohexylamine solution in an ice/acetone/salt bath owing to the exothermic nature of this reaction. Cyclohexylammonium chloride (13.56 g; quantitative yield) was harvested as colorless crystals after evaporation of water at room temperature^11,17,20,21.

Synthesis of (C₆H₁₁NH₃NCS), 2

Cyclohexylammonium thiocyanate, (C₆H₁₁NH₃NCS), was prepared via the salt metathesis reaction between equimolar amounts of (1) and sodium thiocyanate in ethanol medium^11,17,20,21, as illustrated in Eq. 2:

2

Generally, a solution of sodium thiocyanate NaNCS (6.486 g; 80 mmol) in ethanol (280 mL) was added to (1) solution in ethanol (280 mL) (C₆H₁₁NH₃Cl; 10.85 g; 80 mmol) with stirring at room temperature. Subsequently, NaCl as a white precipitate was filtered off, while the filtrate was left for slow evaporation at room temperature to gain the desired product of (2).

Synthesis of Fe(NCS)₃, 3

The neutral complex of iron(III) thiocyanate was prepared in ethanol medium by metathesis process between iron(III) chloride (1.244 g; 20 mmol) and sodium thiocyanate (4.864 g; 60 mmol) at room temperature. The white precipitate of sodium chloride was filtered off, while the red filtrate, containing (3), was kept for synthesizing the desired compound in the last step, as shown in Eq. 3:

3

Synthesis of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O, 4

The preparation of cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O, (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O (4), was accomplished by a ligand addition reaction, where 60 mmol (9.496 g) of (2) was added to the ethanolic solution of (3), under continuous stirring at room temperature, according to Eq. 4:

4

Upon the evaporation of ethanol at room temperature, red crystals of (4) were harvested. These crystals were recrystallized by dissolving them in ethanol for the removal of sodium chloride impurity. For the obtainment of the highest purity of (4), it was dissolved in the minimum amount of isopropanol, filtered off, and finally was evaporated at ambient temperature.

All the above synthesis steps were performed three times to verify the purity of the product.

3. Results and discussion

3.1. Elemental analysis

Table 1 displays the calculated and experimentally found elemental contents, as weight%, of the new hybrid organic-inorganic salt, where the similarity between the calculated and the found results of each element confirmed the molecular formula of C₂₄H₄₇N₉S₆FeO_2.5 with its molecular weight of 749.92 g/mol. The elemental microanalysis results were also assured by the findings of single-crystal X-ray diffraction (SCXRD) and thermal gravimetric analysis (TGA).

Table 1.

Elemental microanalysis of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O.

Element	Wt/Wt %
	Calculated	Found
C	38.44	38.46
H	6.33	6.35
N	16.81	16.79
S	25.65	25.63
O	5.33	5.35
Fe	7.45	7.47

3.2. FT-IR spectrophotometry

The DRIFT spectrum of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O is shown in Figure S1. The weak, broad bands within the range of 3700–3350 cm⁻¹, centered at 3426 cm⁻¹, could be attributed to the asymmetric vibration of the O–H bonds (ν_OH) of the crystallization water molecules^20,21, whose presence were confirmed by elemental microanalysis, SCXRD, and TGA. The broadness of these bands pointed out the involvement of these water molecules in intermolecular hydrogen bonds with the [Fe(NCS)₆]³⁻anion and with the amine protons of the cyclohexylammonium cations^20,21, as confirmed by SCXRD investigation. Furthermore, the medium band at 1597 cm⁻¹ might be due to the bending mode of water. The asymmetric stretching vibrations of (–NH₃⁺) bonds (ν_NH) of cyclohexylammonium cation appeared as a strong, broad band at ~ 3048 cm^−117,20,21. This bathochromic shift in the asymmetric stretching of N–H might be ascribed to the positively charged ammonium nitrogen and to the hydrogen bonds between the ammonium protons and the oxygen of the water crystallization molecules and to the sulfur terminals of the [Fe(NCS)₆]³⁻anion^20,21. The medium band at 1597 cm⁻¹could be attributed to the N–H scissoring^17,20,21. On the other hand, the two weak bands at 1342 and 1265 cm⁻¹might be assigned to the N–H wagging^20,21. The very strong, sharp band at 2932 cm⁻¹ could be ascribed to the asymmetric stretching vibration of the C–H bond, whereas the strong band at 2855 cm⁻¹ would be owing to the symmetric stretching vibration of (–CH₂) bonds of the cyclohexylammonium^17,20,21. The three medium bands at 1481, 1443, and 1381 cm⁻¹ might be assigned to the C–H deformation, while the weak band at 1227 cm⁻¹ might allocate the wagging mode of this bond, the two weak bands at 1119 and 1065 cm⁻¹ might denote to its twisting mode, the very weak bands at 841, 610, and 540 cm⁻¹ might refer to its rocking mode, and the very weak band at 880 cm⁻¹ might assign its bending mode. The very weak band at 1065 cm⁻¹ and the weak band at 995 cm⁻¹ could be due to the stretching vibration of the C–N bond. The weak band at 1173 cm⁻¹ could be attributed to cyclohexylammonium ring deformation, while the very weak band at 957 cm⁻¹ could be ascribed to ring breathing. The broad, weak band, centered at 471 cm⁻¹, might be assigned to the bending of the ring C–N bond. All these bands gave clues for the existence of cyclohexylammonium cation. The very strong band at 2052 cm⁻¹ would be attributed to the C–N asymmetric stretching vibration (ν_CN) of isothiocyanate ligand, while the very weak band at 918 cm⁻¹ would be assigned for its symmetric stretching. The observed strong ν_CN at 2052 cm⁻¹ implied the bonding of the N-terminals of [NCS]⁻ligands to the Fe(III) ion center with the C–N bond order of two, i.e. C=N double bond character⁵³. The very weak bands at 841 and 786 cm⁻¹ might be due to the isothiocyanate ligand C–S stretching vibration, whereas the weak band at 470 cm⁻¹might be for its bending mode. The observation of these vibration modes is consistent with previous reports^54–56. Thus, FTIR analysis evidently provided the success of our procedure for the synthesis of the target hybrid organic-inorganic salt, consisting of cyclohexylammonium cation, hexaisothiocyanatoferrate(III) anionic complex, and crystallization water molecules.

3.3. UV-Vis spectrophotometry

Figure S2 displays a broad, strong absorbance band of (C₆H₁₁NH₃)₃[Fe(NCS)₆] with its maximum (λ_max) at 517 nm and molar extinction coefficient (ε) of ~ 7603 Lmol⁻¹cm⁻¹ in the visible region of the electromagnetic radiation. This band was characteristic for the red, homoleptic, octahedral, anionic hexaisothiocyanatoferrate(III) complex, [Fe(NCS)₆]³⁻. It was ascribed to ligand-to-metal charge-transfer (LMCT), where intramolecular redox process took place upon photon absorption, leading to an electron transfer from a thiocyanate ligand (an electron donor) to Fe(III) ion center (an electron acceptor) and the generation of LMCT excited state of iron(II) and thiocyanate radical: Fe^II–NCS^•−53,57.In addition, the high ε value of this band gave an evidence for the LMCT process⁵⁷.

3.4. TGA analysis

Figure S3 illustrates the thermogravimetric analysis of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O, where the effect of temperature on the sample weight was probed under two different atmospheres: air (Fig. S3 (a)) and helium gas (Fig. S3 (b)). The thermal decomposition was found to be stepwise, regardless of the atmosphere.

Under air atmosphere, the salt started to decompose thermally upon increasing the temperature from room temperature to lose ~ 6% of its weight at ~ 169℃. The weight loss in this step corresponded to the loss of 2.5 water molecules from the hydrated form of the salt, where the loss of water at such low temperature could indicate that they were not coordinated to the Fe(III) ion center, but rather were crystallization molecules. This assumption was consistent with the elemental microanalysis, FTIR spectrophotometry, and SCXRD analysis. The resulting anhydrous form of the salt at this stage, (C₆H₁₁NH₃)₃[Fe(NCS)₆], underwent decomposition upon increasing the temperature by sharply losing ~ 26.1% of its weight, which corresponded to the loss of two cyclohexylammonium cations, 2(C₆H₁₁NH₃)⁺, from the anhydrous form of the salt, at ~ 250 ℃, in the second step of decomposition. The third step resulted in a weight loss of ~ 13.1% at 300 ℃ due to the loss of the last cyclohexylammonium. The complex continued to decompose with rising the temperature to lose ~ 23% of its weight, corresponding to the loss of three isothiocyanate ligands from [Fe(NCS)₆]³⁻ anion at ~ 394 ℃ in the fourth step to form the neutral complex: [Fe(NCS)₃]. The last step of decomposition accounted for ~ 23% weight loss owing to the decomposition of the remaining three isothiocyanate ligands and the formation of hematite, α-Fe₂O_3, at 550 ℃, as corroborated by the PXRD analysis (Fig. 1 (a)).

Fig. 1.

PXRD patterns of (a) α-Fe₂O₃ and (b) FeS and William-Hall plots of (c) α-Fe₂O₃ and (d) FeS obtained by the thermal decomposition of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O.

On the other hand, the thermal decomposition of the complex under helium atmosphere appeared to be faster than its thermal decomposition under air atmosphere. The first step of decomposition corresponded to the loss of 2.5 crystallization water molecules (6.0 wt%) from salt molecule and the formation of its anhydrous form, (C₆H₁₁NH₃)₃[Fe(NCS)₆], at ~ 154 ℃. Subsequently, the anhydrous form of the complex lost their three counter cations of cyclohexylammonium (C₆H₁₁NH₃)⁺ (41 wt%) at ~ 246 ℃, in the second step. A loss of 8.0% in the third step could be attributed to the loss of one isothiocyanate ligands from [Fe(NCS)₆]³⁻ anion at ~ 280 ℃. The fourth step resulted in a loss of ~ 8.0% due to the decomposition of one isothiocyanate ligands to from [Fe(NCS)₄] ⁻ anion at 298 ℃. A weight loss of 23% was observed in the fifth step, which probably owing to the loss of three isothiocyanate ligands to form [Fe(NCS)]⁺ ion at 600 ℃. The loss of CN moiety could occur in the last step at 800 ℃, after which decomposition completed to form the iron(II) sulfide, as confirmed by PXRD analysis (Fig. 1(b)). The reduction of iron(III) to iron(II) in the last step of the thermal decomposition under helium would be attributed to the oxidation of one isothiocyanate ligand by donating one electron to iron(III) and the conversion of this isothiocyanate ligand to isothiocyanate radical^53,57.

3.5. Powder X-ray diffraction analysis

Figure 1 (a) depicts the powder X-ray diffraction (PXRD) pattern of α-Fe₂O₃sample, prepared under air at 600 °C, which displayed narrow sharp peaks for a possibly high purity, perfect crystalline nanostructured nanoparticles⁵⁸. The peaks located at 24.12, 33.18, 35.62, 43.56, 49.50, 54.08, 57.58, 62.42, 63.96, 71.92, 80.68 and 95.61° were characteristic XRD pattern of the hematite (JCPDS No.: 96–101–1241). All these detected peaks were indexed to the (110), (211), (10 − 1), (200), (202), (312), (332), (310), (2-1-1), (433), (411) and (400) crystallographic planes of the hexagonal (rhombohedral axes) structure of pure α-Fe₂O₃ (space group: R-3c)⁵⁹. No XRD peaks were linked to other allotropic assortments of α-Fe₂O₃ or secondary Fe-containing phases.

On the other hand, the PXRD pattern of the sample, synthesized under helium at 800 °C, Fig. 4 (b), revealed characteristic XRD peaks, positioned at 29.58, 33.26, 42.92, 52.60, 56.72, 61.99, 62.63, 70.58, 88.11 and 94.01°, of the iron sulfide (JCPDS No.: 96–153–9667), which respectively corresponded to the (100), (101), (012), (110), (013), (112), (201), (202), (121) and (015) lattice planes of the hexagonal troilite-type FeS structure (space group P 63/mmc)^60,61. This FeS crystal structure encompassed a distorted hexagonal close-packed arrangement of S ions with the Fe ions situated in the octahedral sites. In contrast to the α-Fe₂O₃, the peaks of the FeS were less sharp and broader, indicating a less crystallinity and smaller size nanocrystals⁶².

Fig. 4.

Nitrogen physisorption isotherms and BJH pore size distribution of (a) α-Fe₂O₃ and (b) FeS.

The structural properties such as crystallite size (D), dislocation density (δ), lattice spacing (d), lattice parameters (a, b, c), cell unit volume (V), and lattice strain (ε) were estimated for both α-Fe₂O₃ and FeS by using the following expressions:

5

6

7

8

9

10

The terms λ, β, and θ represent the wavelength of the Cu K_α source (1.5406 Å), the full-width at half maximum (FWHM), and Bragg’s diffraction angle, respectively. The estimated lattice data are displayed in Table 2. These data revealed larger α-Fe₂O₃ nanocrystals with less dislocation density (δ) and lattice strain (ε). Contrarily, the smaller FeS nanocrystals showed more dislocation density (δ) and lattice strain (ε) values. The estimated lattice parameters for the α-Fe₂O₃were in agreement with the values reported in literature^63,64. On the other hand, the assessed values for the FeS lattice parameters were in consistence with previously reported values^65,66.

Table 2.

Crystallite and lattice data of the prepared α-Fe₂O₃ and FeS.

Phase	D (nm)	δ (nm−2)	d(Å)	Lattice parameters(Å)	(Å3)	ε
α-Fe2O3	40.73	0.0006	2.7008	a = b = 5.0392, c = 13.7293	301.93	0.0030
FeS	13.77	0.0053	2.1074	a = b = 3.4843, c = 5.8787	53.88	0.0069

The α-Fe₂O₃is the most thermodynamically stable iron oxide⁶⁷ and its structure consists of stacking of sheets of octahedrally (six-fold) coordinated Fe³⁺ions between two close-packed stacking of oxygen ions, where each oxygen ion is attached to only two Fe ions⁶⁸. The (211) plane was the dominant over the (110), indicating that the as-grown nanoparticles were grown preferentially along the (211) direction⁶⁹. Thus, the most intense peak with Miller index of (211) denoted highly oriented epitaxial growth of the α-Fe₂O₃ nanoparticles along the c-axis to form a hexagonal wurtzite, which was consistent with the SEM surveillance and the c/aratio > 2.7 and indicated a growth along a low energy direction⁷⁰.

The grain size and microstrain developed within the nanocrystals during the nucleation and growth process were estimated from line width (FWHM) of the PXRD lines. According to the Williamson-Hall (W − H) method, the measured integral line width (β) is expressed as⁷¹:

11

where the first and second terms in turn represent the grain size and strain contributions to the line broadening, while the other symbols hold their previously given definitions. The strain (ε) and the crystallite size (D) can sequentially be obtained from the linear (βcosθ) versus (4sinθ) plot of the W − H plots Fig. 1 (c) and (d). The calculated crystallite size and strain from the W − H plot were D = 42.02, ε = −0.0002 for the α-Fe₂O₃ and D= 16.12 nm and ε = −0.0005for FeS. The negative slope of the W − H plot reflected a compressive strain which was < 1%, indicating strain free nanocrystals. Nevertheless, the crystallite size, obtained by the Scherrer formula, was smaller than that estimated by the W-H method. The discrepancy in crystallite size and strain may be associated with the strain correction factor that had been considered into the W-H approach⁷².

3.6. SCXRD structural analysis

The molecular structure of (4) is shown in Fig. 2 (a) with atom labelling. The hybrid organic-inorganic salt (C₆H₁₄N)₃[Fe(SCN)₆]·2.5H₂O consisted of one unique cyclohexylammonium cation and two crystallographically unique octahedral Fe(III) centers, where each Fe(III) was coordinated to six isothiocyanate ligands, with Fe–NCS distances of 2.048 Å for Fe1 center and 2.051 Å for Fe2 center. The coordination angle, Fe–N ≡ CS, was found to be 174.5° for Fe1 center and 170.3° for Fe2 center. These data were like those previously published for tetraalkylammonium salts (alkyl = methyl, ethyl, or n-butyl)²³. The sulfur atoms were disordered by 18% for Fe1 center and by 5% for Fe2 center. Water solvate formed hydrogen bonding with the ammonium group [H₂N–H┄OH₂] of the cation with distance of 1.860 Å, which was 0.86 Å less than the sum of their van der Waals radii⁷³. This observation indicated a strong hydrogen bonding between ammonium group hydrogen [H₂N–H┄OH₂] with water solvate. Furthermore, solvated water molecules showed strong hydrogen interaction with the sulfur terminals of the NCS^–ligands, with 2.537 Å distance, which is 0.463 Å less than the sum of their van der Waals radii⁷³. The cyclohexylammonium cation was also connected to hexaisothiocyanatoferrate(III) anions via hydrogen bonding (2.4 − 2.6 Å) besides their electrostatic interaction. The cyclohexylammonium cation was also connected to the disordered sulfur in the hexaisothiocyanatoferrate(III) anions via hydrogen bonding 3.29 Å besides their electrostatic interaction. Therefore, hydrogen bonding played key role in connecting the three different components of the salt for the creation of three-dimensional structure, as shown in Fig. 2 (b). All the crystal data relevant to the structure of (4) are given in Tables S2-S9 in the supplementary information.

Fig. 2.

(a) Thermal ellipsoid plot of the crystal structure of (4) showing the (C₆H₁₄N)₃[Fe(SCN)₆]·2.5H₂O with its two independent units. Only the asymmetric unit is labeled, and the other equivalent positions are assigned as i to vi. Thermal ellipsoids are drawn at a 50% probability level, (b) packing of (4) in a 3-D network by the assistance of hydrogen bonding among the three components of (4) molecules, and (c) encapsulated [Fe(NCS)₆]³⁻ inside (C₆H₁₄N)⁺/H₂O moieties. Color code: orange (Fe), blue (N), gray (C), yellow (S), red (O), dotted cyan and red (H-bonds). Only one cation and one water solvate are shown. Other solvent molecules, cations and disorder atoms are removed for clarity.

Cyclohexylammonium was not disordered in the crystal structure of (4). Such observation was also encountered in the crystal structure of several compounds^{2,3,5,7–9,11,12,14,16,18–20}. On the other hand, cyclohexylammonium was disordered in other crystal structures^{1,4,6,10,13,16,17,21}. The disorder of cyclohexylammonium could be attributed to lose packing in the hydrophobic region¹, short repellent C–H—C–H contacts between neighboring cyclohexylammonium cations^{4,13,15–17,21}, the residing of ammonium group in the equatorial and axial position in the cyclohexyl ring⁶, or phase transition with increasing temperature¹⁰.

The disorder in the crystal structure of (4) was owing to water crystallization molecules and to the sulfur terminals of isothiocyanate ligands coordinated to Fe(III) ion centers. This type of disorder can be considered as the softness of the material, see, for example, reference⁵³. Such kind of disorder was also observed in the crystal structure of cyclohexylammonium hexaisothiocyanatochromate(III) sesquihydrate²⁰. Both of (4) and its Cr(III) analogue had the same crystal system of trigonal and space group of R-3. The supramolecular structure in (4) showed that [Fe(NCS)₆]³⁻ was encapsulated inside (C₆H₁₄N)⁺/H₂O moieties connected via different types of intermolecular interactions, as displayed in Fig. 2 (c).

Figure 3 shows inhomogeneity because of the disorders in the atomic configuration, which were observed from the structure obtained through X-ray diffraction of a single crystal at 102 K. The standard number of atoms inside each unit cell was three times the chemical formula (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O, corresponding to three lattice points per unit cell. The positions of the main peaks in the experimental and simulated patterns were matched with the interplanar distance (d-spacing) using Bragg’s Law where λ = 1.5406 Å for Cu K_α. By analyzing the distribution of d-spacings, we could determine if the data corresponded to a single phase (homogeneous) or multiple phases (heterogeneous). If the d-spacing values were consistent and matched a single crystallographic phase, it would indicate homogeneity. Variability or clustering around multiple distinct values could suggest multiple phases.

Fig. 3.

A comparative plot of the simulated (theoretical) crystal data, obtained from single crystal X-ray data, and the experimental powder data with phase assignments (blue and green x).

The analysis of the interplanar distances (d-spacings) for the identified peaks is shown in Table 3. The results of the homogeneity analysis showed that the mean d-spacing was 5.105 Å, with a standard deviation of 3.769 Å and a coefficient of variation of 73.83%. The high coefficient of variation suggested significant variability in d-spacings, which could indicate the presence of multiple phases. If this variation was consistent with the known structural features, it might still represent a single complex phase.

Table 3.

Interplanar distance (d-Spacing) analysis.

2 Theta (°)	Relative Intensity (%)	FWHM (°)	Crystallite Size (nm)	d-spacing (Å)
7.6	53.2	0.12	663.5	11.6
11.7	11.9	0.1	798.6	7.6
23.9	100.0	0.14	580.1	3.7
24.9	10.2	0.16	508.5	3.6
43.3	48.7	0.22	388.5	2.1
43.4	34.0	0.28	305.4	2.1

3.7. Nitrogen physisorption

The nitrogen physisorption isotherms of both α-Fe₂O₃ and FeS are depicted in Fig. 4. The adsorption isotherm of α-Fe₂O₃ started at relative pressure (P/P₀) value of 0.0142, where the amount of adsorbed nitrogen increased with increasing the relative pressure until it reached its maximum at P/P₀ value of 1.0. The desorption isotherm dropped sharply with reducing the relative pressure, where the desorption isotherm superimposed on the adsorption isotherm at P/P₀ value of 0.9. The superimposing of the two isotherms continued to P/P₀ value of 0.6, where the amount of desorbed nitrogen increased with reducing the relative pressure. Decreasing the relative pressure interrupted the superimposing of the two isotherms, where the desorption process continued to release nitrogen from the pores with reducing the relative pressure without reaching zero amount of nitrogen (P/P₀value of 0.1). In general, the adsorption/desorption isotherms were similar to type-IV, which characterized mesoporous nanostructures according to the IUPAC classification⁷⁴. The H3-type hysteresis, certified from plate-like particles producing slit-shaped pores from the adsorbent with broad ranges of pore diameters, did not show any restricted adsorption at high P/P₀⁷⁵. On the other hand, the adsorption isotherm of FeS showed the ability of adsorbing nitrogen even at P/P₀ value of 0.0. The amount of adsorbed nitrogen increased with increasing the relative pressure, where the maximum was reached at P/P₀value of 1.0. The adsorption/desorption isotherms were similar to type-IV with an H3-type hysteresis loops according to IUPAC classification⁷⁶. However, the desorption was not capable of releasing all the adsorbed nitrogen, where the desorption process stopped at P/P₀value of 0.1. A hysteresis loop with a terminated desorption branch at such a low relative pressure indicated slit-like/plate-like or ink-bottle pores⁷⁷. In this context, Silvestre-Albero et al.⁷⁸reported that the low-pressure hysteresis was pretty an artifact connected with the lack of equilibrium in the adsorption isotherm and/or the dearth of appropriate outgassing, mostly found in porous solids, where narrow pore constrictions were expected and a very small deviation between the branches could be observed, leading to an opened isotherm⁷⁹.

The insets in Fig. 4 show the Barrett-Joyner-Halenda (BJH) pore size distribution of α-Fe₂O₃ and FeS. The pore size distribution of α-Fe₂O₃ covered a wide range of porosity from micro- to meso- to macro-size, which could be responsible for the low specific surface area (4.00 m²/g) and high average pore size (~ 22.7 nm), as shown in Table 4. On the other hand, the pore size distribution of FeS span only the micro- and meso-porous size, resulting in a high specific surface area of ~ 80 m²/g (Table 4), and moderate average pore size of ~ 8.4 nm. The higher specific surface area and pore volume of FeS than those of α-Fe₂O₃might be ascribed to the carbonization of cyclohexylammonium cations under the inert gas atmosphere with increasing temperature, where the cation served as a soft pore template and disallowed the collapse of the pores¹⁷. Moreover, the abundant porous structure within FeS particles could be responsible for the creation of large internal surface area and large pore volume. On the other hand, α-Fe₂O₃particles lacked such inner porous structure, and thus, they had low pore volume despite their large pore width⁸⁰.

Table 4.

N₂- physisorption results of α-Fe₂O₃ and FeS.

Sample	BET (m2/g)	Pore volume (cm3/g)	Pore width (nm)
α-Fe2O3	4.0021	0.028681	22.7113
FeS	79.9751	0.049639	8.3939

The small surface area of the α-Fe₂O₃nanoparticles may be due to the high calcination temperature, which led to particle agglomeration⁸¹, the growth of the nanocrystallites inside the pores⁸², inter-crystallite sintering, and Van der Waals attraction forces⁸³. It has been reported that during the modification process, Fe(NO₃)₃ decomposed into Fe₂O₃, blocking part of the original pore structure⁸⁴. On the other hand, the larger surface area and pore volume of FeS could be ascribed to carbonization process of the cyclohexylammonium, which operated as a pore template and prohibited the collapse of pores. However, the α-Fe₃O₄ formed via the thermal pyrolysis of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O under air, where oxygen oxidized the cyclohexylammonium, instigating the collapse of pores^17,20,21.

3.8. Morphological investigation of α-Fe₂O₃ and FeS

Figure 5 displays the scanning electron microscope (SEM) micrograph images of α-Fe₂O₃ and FeS at magnification of 20k. The SEM image of α-Fe₂O₃, Fig. 5 (a), revealed that the particles were agglomerated with the adoption of either the rod-like or hexagonal shape. The rod-like shape particles could enlarge to form the hexagonal. On the other hand, the SEM image of FeS displayed hexagonal to circle layered sheets of agglomerated particles.

Fig. 5.

SEM micrographs of (A) α-Fe₂O₃ and (B) FeS.

The surface elemental analysis of α-Fe₂O₃ and FeS was carried out by energy-dispersive X-ray spectroscopy (EDX), as shown in Fig. S4. The elemental content of the surfaces, as weight%, was very close to that of the corresponding bulks. Fig. S4 (a) shows that the surface of α-Fe₂O₃ was composed of ~ 68% Fe (in comparison to ~ 70% in the bulk) and ~ 31% O (in comparison to ~ 30% in bulk). On the other hand, Fig. S4 (b) shows the FeS surface was made of ~ 62% Fe (~ 64% bulk) and ~ 38% S (~ 37% bulk). These results indicated that the surfaces of both α-Fe₂O₃ and FeS were iron-enriched and implied that the new organic-inorganic hybrid salt was an appropriate single-source precursor for preparing iron oxide or sulfide by tuning the thermal decomposition conditions.

The TEM images confirmed the deduced morphology by SEM technique and the crystallinity of both α-Fe₂O₃ and FeS in consistence with PXRD results. Figure S5 (a) exhibited that α-Fe₂O₃ particles were hexagonal, while Figure S5 (b) showed that FeS particles were agglomerated in various shapes of hexagons, rods, and sheets. The HRTEM unambiguously showed the lattice plane fringes of α-Fe₂O₃, Figure S5 (c), and of FeS, Figure S5 (d). Table 5 displays the calculated d-spacing values from the HRTEM images and compares them with their standard PXRD corresponding values for the bulk crystals, where excellent agreement was found.

Table 5.

Comparison between the calculated d-spacing values from HRTEM and PXRD.

Sample	d-spacing calculated from HRTEM, nm	d-spacing in bulk from PXRD, nm	Miller indices (hkl) assignment
α-Fe2O3	0.376	0.368	110
	0.272	0.270	211
FeS	0.296	0.299	100
	0.262	0.265	101
	0.197	0.207	012
	0.164	0.165	111

Selected area electron diffraction (SAED) technique in TEM was used further to examine the crystallinity of our prepared α-Fe₂O₃ and FeS samples by the thermal decomposition of our novel organic-inorganic hybrid salt of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O. Figure 6 (a) shows the SAED of α-Fe₂O₃ with the corresponding Miller indices of the obtained diffraction pattern, which agreed with the its PXRD pattern (Fig. 1 (a)). On the other hand, Fig. 6 (b) displays the SAED of FeS, where some of its crystallographic planes were detected and coincided with the PXRD in Fig. 1 (b). Furthermore, the SAED of α-Fe₂O₃ had much higher intensity than the SAED of FeS, indicating that α-Fe₂O₃ had higher degree of crystallinity than FeS and confirming the PXRD results in this regard.

Fig. 6.

SAED of (a) α-Fe₂O₃ and (b) FeS.

The average particle size of α-Fe₂O₃ and FeS, based on their TEM images, was estimated from their histogram distribution, as shown in Figure S6. It was found that α-Fe₂O₃ had an average particle size of 79.2 nm (Figure S6 (a)), while it was 13.5 nm for FeS (Figure S6 (b)). These finding results of average particle size were consistent with the PXRD results, which revealed higher crystallinity of α-Fe₂O₃ than that of FeS.

3.9. Magnetic susceptibility

The bulk hematite (α-Fe₂O₃) exhibits a variety of magnetic phase transitions⁸⁵. It shows an antiferromagnetic (AFM) Neel temperature T_N ≈ 960 K and undergoes a spin-flip (Morin) transition at about T_M ≈ 260 K. In between these two phases (260–960 K), the material shows a weak ferromagnetic ordering. Below T_M, the material yet reveals another transition to spin-glass phase below T_G ≈ 50 K, as characterized by zero field cooled (ZFC) and field cooled (FC) procedures. Furthermore, below the glass transition, hematite nanoparticles display ferromagnetic-like with a sharp increase in the susceptibility with decreasing temperature. Tadic et al.⁸⁵ measured the field cooled (FC) and the ZFC susceptibilities of α-Fe₂O₃nanoparticles⁸⁵. They found that both susceptibilities exhibited a typical spin-glass behavior. The ZFC susceptibility exhibited a maximum at the blocking temperature T_B = 52 K and bifurcated with the FC susceptibility at the irreversibility temperature T_irr= 103 K. Their sample with nanoparticle sizes of ~ 10 nm did not exhibit the Morin transition down to 5 K in accord with the published data^85–87. All these magnetic transition temperatures are affected by the particle size, giving rise to the large discrepancies of the reported values, especially T_M and T_Gvalues^85–87.

On the other hand, iron sulfide (FeS) occurs in two different structures with wide range of properties. The tetragonal phase (P4/nmm) is a metastable phase and is often found with Fe slightly off stoichiometry (Fe_1.03S)⁴⁹. The other crystalline phase of FeS is the hexagonal (P63/mmc). Moreover, the off-stoichiometry leads to a variety of magnetic and physical properties.

Muon spin resonance measurements for the hexagonal phase revealed the occurrence of disordered magnetic properties below the antiferromagnetic Néel temperature of T_N = 20 K. Ferromagnetic ordering was found at lower temperature with Tc≈ 5 K⁶². The tetragonal phase of FeS showed superconducting behavior near 4 K⁶⁶.

In Fig. 7, we present the susceptibility for (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O and its products of thermal decomposition: α-Fe₂O₃ and FeS. The lowest curve in Fig. 7 revealed that the susceptibility of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O had a broad maximum ~ 130 K with an inflection point ~ 180 K. Above 200 K, it showed typical paramagnetic behavior.

Fig. 7.

The variations of the real part of the ac-susceptibility with temperature for (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O, FeS, and α-Fe₂O₃.

For α-Fe₂O₃sample, the susceptibility showed a sharp maximum at ~ 55 K in agreement with the previous published results⁸⁵, supporting the hematite hexagonal phase as revealed by PXRD results in Fig. 1 (A). Moreover, its susceptibility showed a minimum at about 40 K, then rose sharply, giving an indication to a ferromagnetic ordering at low temperature, reaching a maximum below 5 K. Similar results were observed for the prepared FeS sample. The susceptibility maximum occurred at ~ 55 K. The susceptibility also showed a minimum at ~ 40 K, then rose sharply, indicating a possible ferromagnetic ordering at low temperature, reaching a maximum below 5 K. The FeS susceptibility results also agreed with the published data of FeS hexagonal phase revealed in the PXRD pattern in Fig. 1 (B). All samples showed paramagnetic behavior above 180 K. No transition to spin-flip (the Morin phase) were observed in all samples.

It is possible that the observed wide maxima in α-Fe₂O₃ and FeS were related to the disorder effects, introduced by the nano-size particles in these materials⁸⁸. The common effect of particle sizes, stress and defects in ferromagnetic materials is to increase the degree of disorder, which widens the width of the transition temperature, changes, in some cases, the order of transition⁸⁹, and shifts the transition temperature to a lower value in spin glass materials (like Ni-Mn for example)⁹⁰.

Conclusions

In conclusion, we succeeded in establishing a four-step, room-temperature synthesis protocol for cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O with high yield. The elemental microanalysis, visible absorbance and FTIR spectrophotometric techniques as well as SCXRD confirmed the formation of this novel organic-inorganic hybrid salt.

The SCXRD analysis revealed the presence of two crystallographically distinguished, octahedrally-coordinated Fe(III) centers, varied in their Fe−NC bond lengths and angles as well as the percentage of the disorder of isothiocyanate sulfur atoms. Moreover, the solvated water molecules were disordered. The total disorder, caused by isothiocyanate sulfur atoms and water molecules was responsible for the softness of this salt and the mismatch between the theoretical (simulated) and experimental PXRD patterns of the salat. Nevertheless, the connection among cyclohexylammonium cations, hexaisothiocyanateferrate(III) anions, and solvated water molecules by hydrogen bonds created a 3-dimensional supramolecular structure.

The TGA displayed that this salt had a higher thermal stability under air than under inert atmosphere. Moreover, it revealed that this salt was a suitable single-source precursor for the formation of either nanocrystalline α-Fe₂O₃ under air or FeS under helium. The formation of the alpha-phase of iron oxide was consistent with the thermodynamic fact in that it is the most stable form of iron oxides, while the formation of FeS, under the inert atmosphere of helium and the absence of oxygen, was facilitated by a redox reaction, where an isothiocyanate ligand oxidized and iron(III) was reduced to iron(II). Furthermore, thermal decomposition had a profound influence on the textural and morphological properties of α-Fe₂O₃ and FeS.

The larger surface area and pore volume of FeS, formed by the thermal decomposition of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O under helium, could be ascribed to the carbonization process of the cyclohexylammonium, which operated as a pore template and prohibited the collapse of pores. However, the α-Fe₃O₄ formed via the thermal decomposition of (C₆H₁₁NH₃)₃[Fe(NCS)₆]0.2.5H₂O under air, where oxygen oxidized the cyclohexylammonium, instigating the collapse of pores.

Morphological studies revealed that α-Fe₂O₃ formed either a rod-like or hexagonal shape nanoparticles, while FeS exhibited hexagonal to circular layers agglomerated nanoparticles.

PXRD, electron diffraction and TEM revealed that α-Fe₂O₃ and FeS were nanocrystalline with particles in the nanometric size. However, α-Fe₂O₃ possessed higher degree of crystallinity, larger crystallite size, less dislocation density, and less lattice strain than FeS, reflecting the impacts of the thermal decomposition path and atmosphere.

The magnetic measurements revealed a paramagnetic behavior above 180 K for the salt, α-Fe₂O₃, and FeS. However, below 40 K, the susceptibility showed a sharply increased, indicating a ferromagnetic ordering at low temperature. No transition to spin-flip (the Morin phase) was observed. Moreover, the observed wide maxima in the magnetic susceptibility of α-Fe₂O₃ and FeS were related to the disorder effects, introduced by the nano-size particles in these materials. The magnetic susceptibility behavior of α-Fe₂O₃ and FeS agreed with their corresponding phases detected by PXRD analysis.

The textural and morphological diversity, emerged from nano size nature, highlighted the unique properties and potential applications of α-Fe₂O₃ and FeS. Therefore, future work would be deviated towards the investigation of the potential application of the α-Fe₂O₃ and FeS nanomaterials, derived by the thermal decomposition of cyclohexylammonium hexaisothiocyanatoferrate(III) 2.5H₂O, as adsorbents or photocatalysts for the remediation of water and the removal of organic and inorganic pollutants.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

F.S. Alqahtani: conceptualization, validation, investigation, writing – original draft, and writing – review and editing. F.M. Albaqi: investigation, validation, methodology, data curation, formal analysis, writing – original draft, writing – review and editing. R.M. Almalahi: investigation, methodology, data curation, formal analysis, writing – original draft. K.I. Anojaidi: methodology, data curation and formal analysis. R.H. Arasheed: investigation, validation, supervision, methodology, data curation, and formal analysis. M.S. Alsurayhi: methodology and data curation. B.S. Alhedaib: methodology and data curation. I.A. Albinali: supervision, methodology and data curation. A.M. Alkhalifa: methodology and data curation. E.A. Alghilan: writing – original draft, and writing – review and editing. A.I. Alromaeh: methodology. M.A. Khanfar: methodology, writing – original draft, and writing – review and editing. M.A. Aldamen: methodology, writing – original draft, and writing – review and editing. F. Obad: methodology. K.A. Ziq: methodology, writing – original draft, and writing – review and editing. K.K. Taha: investigation, writing – original draft, and writing – review and editing. A.A. Bagabas: conceptualization, validation, investigation, methodology, writing – original draft, and writing – review and editing. All authors have read and agreed to the published version of the manuscript.

Data availability

Data is provided within the manuscript and supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.

Supplementary Information

The supporting information of this manuscript is available online at.

Funding

Funding was provided by King Abdulaziz City for Science and Technology (KACST), King Fahd University of Petroleum and Minerals (KFUPM), Prince Sattam Bin Abdulaziz University (PSAU).