Ammonia and iron cointercalated iron sulfide (NH₃)Fe_0.25Fe₂S₂: hydrothermal synthesis, crystal structure, weak ferromagnetism and crossover from a negative to positive magnetoresistance

Abstract

The discovery of superconductivity in anti-PbO-type FeS has aroused a renewed interest in the intercalation compounds of FeS. Here we report a novel intercalation compound of FeS with the chemical composition of (NH₃)Fe_0.25Fe₂S₂, which is synthesized via a new hydrothermal reaction. This material crystallizes in the tetragonal space group I4/mmm, preserving the FeS tetrahedral layers with ammonia and excess iron forming planes in between. The microstructure and thermal stability of the sample were studied by scanning electron microscopy (SEM), transmission electron microscopy (TEM) and thermogravimetric analyses (TGA). These results suggest that (NH₃)Fe_0.25Fe₂S₂ is not sensitive to electron beam irradiation and is more thermally stable than the other ammonia intercalated iron selenide superconductors. Physical property measurements show that it is a ferromagnetic semiconductor. By using first-principles calculations we assess that the low-temperature ferromagnetism originates from the interlayer rather than the intralayer iron. The transport properties at low temperatures are dominated by electron-like carriers and the sign reversal and strong temperature dependence of the Hall coefficient may be caused by a multi-band effect. Most importantly, an unusual crossover from negative to positive magnetoresistance with increasing temperature was identified, which reveals relatively strong coupling between carriers and magnetic moments as well as disorder.

---

Introduction

Layered transition metal compounds continue to attract considerable attention because of their rich structural diversity and interesting physical properties such as colossal magnetoresistance (CMR) and high-temperature superconductivity.¹ Among them, the layered iron-based compounds composed of alternating FeX (X = pnictogen, chalcogen) tetrahedral layers and various spacer layers are currently extensively studied as promising high-temperature superconductors.² Like the CuO₂ plane in cuprate superconductors, the FeX layer which is made up of edge-sharing FeX₄ tetrahedra acts as conducting layer and is responsible for superconductivity. It is accepted that the geometry of FeX₄ tetrahedra together with the antiferromagnetic spin fluctuations plays important roles in the superconducting properties.³

Within the iron-based superconductors, FeSe has the simplest crystal structure, consisting of only anti-PbO-type FeSe layers and with no spacer layers.^2b The T_c of FeSe is relatively low as 8 K, but this can be greatly enhanced not only by chemical doping and external pressure but also by intercalation.⁴ The first intercalation compound of FeSe is K_xFe_2−ySe₂ (T_c ∼ 30 K), which has a similar structure to Ba_1−xK_xFe₂As₂ superconductor.⁵ A peculiar feature of K_xFe_2−ySe₂ is that holelike Fermi surface is absent near the Brillouin zone centre, which implies interband scattering may not be a dominant pairing mechanism for iron-based superconductors.⁶ However, the phase separation in this system makes it difficult to investigate the underlying physics in more detail.⁷ It is found that except K, Rb, Cs, Tl and Na, other metal or complex spacer layer is difficult to be intercalated in between the FeSe layers via high-temperature solid state reaction. Thereafter, various low-temperature synthetic routes were developed and a series of superconductors A_xB_yFe₂Se₂ (A = alkali metal, alkaline metal, rare-earth metal, B = NH₃; A = alkali metal, B = pyridine, ethylenediamine, hexamethylenediamine), A_xFe₂Se₂ and Li_1−xFe_xOHFeSe were obtained with T_c ranging from 30 to 46 K.⁸ It should be noted that most of these intercalation compounds are known only by the fact that they are superconductive, while their precise crystal structures as well as other underlying physical properties are less investigated due to the samples being thermally non-stable or air-sensitive. More new related materials with high crystallinity and stability are needed to better understand the structure and properties of this family of compounds. Besides, for A_x(NH₃)_yFe₂Se₂, the NH₃ molecule plays an important role in stabilizing the structure, and thus it is interesting to see whether there exist A-free NH₃ intercalated compounds.

Recently, superconductivity at 5 K was realized in high-quality anti-PbO-type FeS sample which is synthesized by the hydrothermal reaction of iron powder with sulfide solution.⁹ This has evoked a renewed interest in FeS as well as in the intercalation compounds of FeS.¹⁰ It is suggested that the superconductivity in FeS is multi-band and strongly anisotropic; in the normal state, FeS exhibits nonlinear Hall effects and huge positive magnetoresistance (MR), quite different from the small positive MR phenomena observed in other iron-based superconductors.¹¹ As to the intercalation compounds of FeS, efforts made previously have got limited examples such as A_xFe_2−yS₂ (A = alkali metal) and Li_1−xFe_xOHFeS, which are isostructural to A_xFe_2−ySe₂ and Li_1−xFe_xOHFeSe, respectively, and both are semiconducting.¹² Attempts to intercalate Na and CaO layer into FeS result in NaFe_1.6S₂ and Ca₂O₃Fe_2.6S₂, respectively, but these new phases do not maintain the FeS tetrahedral layers.¹³ Even now, when high-quality superconducting FeS samples are available, the family of FeS intercalates have not been enlarged. One of the major reasons is the methodology used in preparing intercalation compounds of FeSe is not applicable to the case in FeS due to that FeS precursor is not strictly dehydrated.

Here we report a new hydrothermal method to synthesize a novel FeS intercalate with chemical composition of (NH₃)Fe_0.25Fe₂S₂. It represents the first A-free NH₃ intercalated iron chalcogenide, which is highly crystalline and more thermally stable than the other ammonia intercalated iron selenide superconductors. Meanwhile, it is also a rare example of NH₃ intercalation by one-step hydrothermal method. In contrast to FeS which is a superconductor, (NH₃)Fe_0.25Fe₂S₂ is a ferromagnetic semiconductor and no superconductivity is observed down to 2 K. First-principles calculations reveal that the low-temperature ferromagnetism originates from the interlayer rather than the intralayer iron. Hall effect measurement suggests electron-type carriers at low temperatures and the sign reversal and strong temperature dependence of R_H may be caused by multi-band effect. Most interestingly, (NH₃)Fe_0.25Fe₂S₂ exhibits a novel crossover from negative to positive MR with increasing temperature which can be explained by disorder effects and spin-dependent scattering mechanism.

Experimental

Chemicals

Iron powder (Alfa Aesar, ≥99%), (NH₄)₂S aqueous solution (Sinopharm, 8%, w/w) and NaOH pellets (Sinopharm, 98%) were used as received.

Synthesis of (NH₃)Fe_0.25Fe₂S₂

Polycrystalline (NH₃)Fe_0.25Fe₂S₂ was prepared by a facile one-step hydrothermal reaction. 2 g of Fe powder and 19 mL of (NH₄)₂S aqueous solution were placed in a 25 mL Teflon-lined autoclave. 1.2 g of NaOH pellets were added as mineralizer. The autoclave was tightly closed and heated at 150 °C for 30 days. After being cooled to room temperature, the products were washed with water and ethanol and dried in vacuum. The yield is ca. 80% with respect to Fe.

Characterization

Powder X-ray diffraction (PXRD) pattern was collected at room temperature on a Bruker D8 Focus X-ray diffractometer operated at 40 kV voltage and 40 mA current using Cu Kα radiation (λ = 1.5406 Å). The 2θ range was 5–110° with a step size of 0.02°. Indexing and Rietveld refinement were performed using the DICVOL91 and FULLPROF programs,¹⁴ respectively. Elemental analysis was carried out by inductively coupled plasma-atomic emission spectrometry (ICP-AES) and CHN elemental analyzer. Scanning electron microscopy (SEM) images were taken on a Hitachi S-4800 microscope equipped with an electron microprobe analyzer for the semiquantitative elemental analysis in the energy dispersive X-ray spectroscopy (EDX) mode. Transmission electron microscopy (TEM) images and selected area electron diffraction (SAED) patterns were obtained from a JEOL JEM-2100F microscope operating at 200 kV. Thermogravimetric analyses (TGA) were carried out using a Q600SDT thermal analyzer under N₂ atmosphere in the temperature range of 30–600 °C. The samples were examined by powder X-ray diffraction immediately after the TGA experiments.

Physical properties measurements

The magnetic susceptibility was measured with a magnetic field of 10 kOe, in both zero-field-cooling and field-cooling processes. The magnetic hysteresis was recorded at 2 K and 300 K. The resistivity measurement was conducted through the standard four-wire method. The Hall resistance and magnetoresistance were measured by sweeping the magnetic field at a fixed temperature. All these measurements were performed in a physical property measurement system (PPMS) of quantum design.

First-principles calculations

First-principles calculations were performed with the Cambridge serial total energy package (CASTEP).¹⁵ The exchange correlation effects were treated by the general gradient approximation (GGA) using the Perdew–Burke–Ernzerhof (PBE) functional.¹⁶ The interactions between the ionic cores and the electrons were described by the ultrasoft pseudopotentials.¹⁷ The experimental lattice and atomic parameters obtained from the PXRD refinement were used for calculations. Structural optimization was performed, in which the lattice parameters were fixed while the internal atomic positions were relaxed until the Hellmann–Feynman forces acting on each atom were less than 0.01 eV Å⁻¹. The plane-wave energy cutoff was set to 330 eV. The k-point meshes over the total Brillouin zone were sampled by 2 × 2 × 2 Monkhorst–Pack grids.¹⁸

Results and discussion

Unlike the traditional high-temperature solid state synthesis, hydrothermal method has the advantages in synthesizing meta-stable phases due to its mild and soft reaction conditions. Recently, hydrothermal synthesis has been successfully developed to produce superconducting Li_1−xFe_xOHFeSe and semiconducting Li_1−xFe_xOHFeS, with the assistance of concentrated base (LiOH) as the mineralizer.^8i,12b Here, using (NH₄)₂S aqueous solution as the source of S and NH₃ and NaOH as the mineralizer, we synthesized a new layered iron sulfide (NH₃)Fe_0.25Fe₂S₂. The room-temperature PXRD pattern (Fig. 1(a)) of (NH₃)Fe_0.25Fe₂S₂ can be well indexed based on a tetragonal cell with lattice parameters a = 3.68795(12) Å and c = 15.1134(8) Å, and no peaks due to impurities were observed. Examination of diffraction extinction reveals that the sample crystallizes in a body-centered lattice and the probable space group is I4/mmm. A structural model adapted from A_x(NH₃)_yFe₂Se₂ superconductors^8c was used to perform the Rietveld refinement against the raw data. There are two unique Fe Wyckoff sites (Fe1 at 4d, Fe2 at 2b), one unique S Wyckoff site (4e) and one unique N Wyckoff site (2a). After refinement of the 2b site occupancy, fairly good agreement factors were obtained: R_p = 0.91%, R_wp = 1.19%, and χ² = 1.38. The final Rietveld refinement profiles were plotted in Fig. 1(a) and the final refined structural parameters were given in Table 1. According to the refinement results, the composition of the compound can be denoted as (NH₃)Fe_0.242Fe₂S₂, which is consistent with the results of elemental analysis by ICP-AES (Fe : S = 1.13/1) and CHN elemental analyzer (N: 6.63 wt%; H: 1.53 wt%). Fig. 1(b) shows the schematic crystal structure of (NH₃)Fe_0.25Fe₂S₂, which features two-dimensional FeS tetrahedral layers stacking along the c axis with NH₃ and excess Fe cointercalated between the layers. The FeS layer has the anti-PbO-type structure and is composed of FeS₄ tetrahedra connected via edge sharing. The FeS₄ tetrahedron is more distorted than that in FeS,⁹ with the Fe–S bond length being 2.227(3) Å and the two S–Fe–S angles on the same side of ab plane being 111.8(1)°. The anion height (the distance of the anion from the Fe plane) is calculated to be 1.083(5) Å, much smaller than the value (1.38 Å) for optimal superconductivity to occur in iron-based superconductors.¹⁹ The interlayer Fe, Fe2, is surrounded by four NH₃ molecules, with Fe–N distance of 2.6078(1) Å. The N–H⋯S distance is 3.633(4) Å, slightly shorter than the N–H⋯Se distance in ammonia intercalated iron selenide.^8c

Fig. 1 (a) Rietveld refinement of the PXRD pattern of (NH₃)Fe_0.25Fe₂S₂. Red small points represent the experimental values, black solid line is the calculated pattern, green solid line is the difference between the experimental and calculated values, and blue vertical bars are the Bragg positions. (b) The schematic crystal structure of (NH₃)Fe_0.25Fe₂S₂ (H atoms are not shown) viewed along [010] (left side) and [001] (right side) directions, with FeS₄ tetrahedra shown in green.

Table 1 Crystallographic data of (NH₃)Fe_0.25Fe₂S₂

Formula	(NH3)Fe0.25Fe2S2
Temperature (K)	298
Space group	I4/mmm
a (Å)	3.68795(12)
c (Å)	13.1134(8)
V (Å^3)	205.557(15)
Z	2
Radiation type	Cu Kα
Wavelength (Å)	1.5406
2θ range (deg)	5–110
Step size (deg)	0.02
Rp	0.91%
Rwp	1.19%
χ^2	1.38

---

Atomic parameters
Fe1	4d (0, 0.5, 0.25)
S1	4e (0, 0, z)
	z = 0.3326(4)
N1	2a (0, 0, 0)
Fe2	2b (0, 0, 0.5)
Bond length (Å)
Fe1–Fe1	2.6078(1) × 4
Fe1–S1	2.227(3) × 4
Bond angles (deg)
S1–Fe1–S1	108.3(3) × 4
	111.8(1) × 2

---

The SEM image in Fig. 2(a) shows that (NH₃)Fe_0.25Fe₂S₂ forms well-defined rectangular plates with average edge length of several micrometers. Unlike FeS which shows significant aggregation,⁹ the particles of (NH₃)Fe_0.25Fe₂S₂ are well separated. Efforts made to grow single crystals sizable for single crystal diffraction have failed. EDX analysis of a number of plate-like crystals reveals the presence of Fe, S and N elements, and no other elements were detected. The relative ratio of Fe, S and N elements is about 2.3/2/1, in accord with the above results. Typical TEM and high-resolution TEM images are shown in Fig. 2(b) and (c). The TEM image also illustrates the regular shape of the particles. The high-resolution TEM image shows a set of high-resolution lattice planes with the inter-planar distance of 0.26 nm, corresponding to the (110) plane of (NH₃)Fe_0.25Fe₂S₂. The SAED spot pattern (Fig. 2(d)) matches that predicted for the [001] direction of (NH₃)Fe_0.25Fe₂S₂ and can be indexed using I4/mmm unit cell. These microscopy measurements confirm the single-crystalline nature of the individual rectangular plate-like samples. It should be noted that during the SEM/TEM measurements, the sample does not undergo any evident deterioration, suggesting that (NH₃)Fe_0.25Fe₂S₂ is not sensitive to the electron beam irradiation.

Fig. 2 (a) SEM image, (b) TEM image, (c) high-resolution TEM image, and (d) SAED pattern along the [001] zone axis of (NH₃)Fe_0.25Fe₂S₂.

(NH₃)Fe_0.25Fe₂S₂ is more thermally stable than the other ammonia intercalated iron selenide superconductors. The thermal stability of (NH₃)Fe_0.25Fe₂S₂ was examined by thermogravimetric analyses in a N₂ atmosphere from 30 to 600 °C. As shown in Fig. 3, (NH₃)Fe_0.25Fe₂S₂ starts to decompose at 260 °C and continues to lose weight up to 360 °C, with a final weight loss of 8% which corresponds to the release of all the NH₃ molecules. The weight loss of 8% also excludes the possibility of NH₄⁺ species located between the Fe₂S₂ layers. For that case, the theoretical value of the weight loss would be 16.4% since the loss of NH₄⁺ species would result in simultaneous release of NH₃ and H₂S molecules, such as that occurs for K₂(NH₄)₄Zn₄Sn₅S₁₇·3H₂O.²⁰ After loss of NH₃ molecules, the sample starts to slowly increase weight, which may be attributed to the slight oxidization of the interlayer Fe. Powder X-ray diffraction performed immediately after the TGA experiments (Fig. S1†) indicated the existence of hexagonal FeS (P6₃/mmc) rather than tetragonal FeS in the residues of the sample.

Fig. 3 Thermogravimetric and differential scanning calorimetry curves of (NH₃)Fe_0.25Fe₂S₂ at a heating rate of 10 °C min⁻¹ under a N₂ atmosphere from 30 to 600 °C.

Fig. 4(a) shows the temperature-dependent magnetic susceptibility of (NH₃)Fe_0.25Fe₂S₂ measured under a magnetic field of 10 kOe. Apparently, no magnetic transition was observed in the temperature range from 300 K to 60 K, indicating the dominance of paramagnetic contribution at high temperature. The data at above 60 K can be well fitted by the Curie–Weiss law. The fitted parameters are: Curie constant C = 1.33 K emu mol⁻¹ and Curie–Weiss temperature θ = −4.3 K. The effective magnetic moment calculated from the Curie constant is μ_eff = 2.16 μ_B per Fe. This value is much smaller than that of Li_1−xFe_xOHFeS (4.98 μ_B per Fe), but is close to that of FeTe_0.95 (2.44 μ_B per Fe).²¹ The negative Curie–Weiss temperature (θ = −4.3 K) suggests antiferromagnetic nearest-neighbor interactions in the FeS layers. Since |θ| is very small, the antiferromagnetic nearest-neighbor interactions are weak. The magnetic hysteresis curve is linear at 300 K, indicating paramagnetic behavior, while that at 2 K shows magnetic hysteresis, demonstrating the existence of long-range ferromagnetic ordering. A close look at the low-temperature magnetic susceptibility reveals divergence between the zero-field-cooled (ZFC) and field-cooled (FC) magnetization below 4 K. As can be seen in the zoomed M vs. H curves described in Fig. 4(d), the coercive field is about 600 Oe, and the remanence is about 0.6 emu g⁻¹. The saturated magnetic moment at 2 K and 50 kOe is small ∼0.67 μ_B per formula.

Fig. 4 (a) Temperature dependence of the direct-current magnetic susceptibility of (NH₃)Fe_0.25Fe₂S₂ measured under a magnetic field of 10 kOe, in both zero-field-cooled (ZFC) and field-cooled (FC) processes. (b) The inversed magnetic susceptibility and a linear fitting of the data from 300 K to 60 K. (c) Magnetic hysteresis of the (NH₃)Fe_0.25Fe₂S₂ compound at 2 K and 300 K. (d) The zoomed magnetic hysteresis at 2 K and 300 K.

Since the tetrahedral FeS and FeSe layers are usually nonmagnetic, it is surprising that ferromagnetism takes place in (NH₃)Fe_0.25Fe₂S₂. Here, to figure out whether the low-temperature ferromagnetic ordering originates from the interlayer or intralayer Fe, first-principles calculations were performed. To simulate the partially occupied Fe2, we use a supercell. We tried four different magnetic ordering patterns as following: NM (both Fe1 and Fe2 are nonmagnetic); FM-1 (Fe1 couple ferromagnetically and Fe2 are nonmagnetic); FM-2 (Fe1 are nonmagnetic and Fe2 couple ferromagnetically); FM-3 (both Fe1 and Fe2 have the same spin orientation). The local spin density approximation (LSDA) calculation predicts that the FM-2 has the lowest total energy, and the FM-3 has the highest total energy, see Table 2. This suggests the low-temperature ferromagnetic ordering originates from interlayer rather than intralayer Fe, which is similar to the case in Li_1−xFe_xOHFeSe and is consistent with the fact that tetragonal FeS is nonmagnetic.^8k,9 It should be noted that, as revealed by the electronic density of states (Fig. S2†), LSDA calculation fails to reproduce the observed semiconducting behavior discussed below. This may be probably due to that the bandgap of (NH₃)Fe_0.25Fe₂S₂ is too small.

Table 2 Calculated energy for four possible magnetic structures of (NH₃)Fe_0.25Fe₂S₂. Fe1: intralayer iron; Fe2: interlayer iron. NM = nonmagnetic; FM = ferromagnetic

Type	Fe1	Fe2	Energy (eV per formula)
NM	NM	NM	−2.3
FM-1	FM	NM	−1.0
FM-2	NM	FM	−2.7
FM-3	FM	FM	0

---

Fig. 5 shows the temperature-dependent electrical resistivity, measured on a pressed pellet after annealed at 200 °C for 48 h. According to the TGA results, the annealed sample would be stable without decomposition, which is confirmed by the PXRD measured for the annealed sample (Fig. S3†). The room-temperature resistivity is 8.7 mΩ cm, close to that of FeS (7.1 mΩ cm).⁹ As the temperature decreases, the resistivity increases monotonically, indicative of semiconducting behavior. The resistivity in the temperature range of 200–300 K can be described by the thermal activation model with activation energy of 20 meV (Fig. S4†). Below 60 K, the sample becomes more insulating, meaning that weak charge-carrier localization occurs. This is most likely due to the interlayer iron which is randomly distributed and carries local moment. Similar behavior has been observed in Fe_1.18Te_0.6Se_0.4 and is attributed to the effect of the interstitial iron.²²

Fig. 5 Temperature dependence of the electrical resistivity of (NH₃)Fe_0.25Fe₂S₂.

To further reveal the underlying electronic properties, we took the Hall effect and magnetoresistance measurements. As displayed in Fig. 6(a), the Hall resistivity (ρ_H) is negative at low temperatures, suggesting that the transport properties are dominated by electron-type carriers. From this set of data, the Hall coefficient R_H = ρ_H/H is determined and shown in Fig. 6(b). It can be seen that R_H increases rapidly below 50 K. With further increasing temperature, it increases slowly and tends to change sign at above 200 K. The sign reversal and strong temperature dependence of R_H, as well as the nonlinear ρ_H at low temperatures may be caused by multi-band effect. Similar phenomena have been reported for FeS, FeSe and Li_1−xFe_xOHFeSe.^11a,23

Fig. 6 (a) Magnetic field dependence of Hall resistivity at different temperatures. (b) Temperature dependence of Hall coefficient, the data at 5 K is omitted due to the nonlinear ρ_H.

Fig. 7 shows the magnetoresistance {MR = [ρ(H) − ρ(0)]/ρ(0) = Δρ/ρ(0)} at various temperatures, measured by applying magnetic field perpendicular to the direction of excitation current. Although being polycrystalline, the sample does not show sharp drop in the resistance at low fields (<1 T), suggesting that the resistance of the sample may not be dominated by the carrier scattering at the grain boundaries.²⁴ At low temperatures, the MR is negative, reaching −1.6% at 5 K and 5 T. As the temperature increases, the magnitude of negative MR decreases gradually and becomes zero at 60 K. Interestingly, the MR turns to positive value at 180 K, and the magnitude increases with further increasing temperature. Such crossover from negative to positive MR are only known in a few materials, such as spinel Zn_0.95Cu_0.05Cr₂Se₄ through a change of conduction mechanism, Mn₂FeReO₆ from coexistence of ferrimagnetic conducting FeRe and antiferromagnetic Mn spin networks in the same phase, GdRhGe from a change of the magnetic structure and the moment amplitude, WS₂ nanoflake explained by a combination of the forward interference model and the wave function shrinkage model in the Mott VRH regime, and disordered graphene due to the monocrystalline breaking and crystallite-boundary scattering.²⁵ In the following, we will discuss the possible reasons for the unusual MR observed in (NH₃)Fe_0.25Fe₂S₂.

Fig. 7 Magnetoresitance (MR) as a function of magnetic field at different temperatures for (NH₃)Fe_0.25Fe₂S₂ sample, showing the crossover from negative to positive MR behavior.

Firstly, we propose that the negative MR at low temperatures may originate from weak localization effect and reduction of spin-dependent scattering. Weak localization is possible in iron chalcogenides. It is pointed out that the interstitial iron in Te-rich Fe_1+δSe_1−xTe_x and nonsuperconducting Fe_1+δS could induce weak localization effects, leading to negative MR in the former and semiconductivity in the latter.²⁶ For (NH₃)Fe_0.25Fe₂S₂, the Fe2 site is partially occupied and the random distribution of Fe2 could also provide possible disordered potentials and cause weak localization effects. The kink observed at around 60 K in the temperature-dependent resistivity and the small negative MR saturating at low fields agree with the characteristics of weak localization. On the other hand, the Fe2 spins are weakly ferromagnetic at low temperatures, the spin scattering from which may be reduced by the application of magnetic fields and thus give negative MR.

Let us discuss the positive MR at high temperatures. Generally, the occurrence of positive MR can be explained by the Lorentz contribution to the excess resistivity. But for our sample, the MR measured with external magnetic field parallel to the direction of excitation current is similar to that measured with external magnetic field perpendicular to the direction of excitation current, and therefore, the Lorentz contribution may not be the major reason for the positive MR. As mentioned above, the Curie–Weiss fit to the high-temperature magnetic susceptibility gives small negative Curie–Weiss temperature (θ = −4.3 K) and suggests antiferromagnetic nearest-neighbor interactions in the FeS layers. According to Yamada and Takada, small positive magnetoresistance can arise in such system due to the enhancement of spin fluctuations in one of the magnetic sublattices.²⁷ On the other hand, by calculating with random resistor network model, Parish and Littlewood proposed a classical way to explain the nonsaturating positive MR in disordered and strongly inhomogeneous semiconductor.²⁸ Since our sample can be seen as disordered and inhomogeneous, we can attribute the observed positive MR to carrier mobility fluctuations (nonuniform spatial distribution of carrier mobility) and the increase in MR with increasing temperature may be due to the enhancement of carrier mobility fluctuations at high temperatures.^25e

After discussion of the structural and physical properties, we can better understand why superconductivity is absent in the title compound. On one hand, the absence of superconductivity may be partly due to the structural distortion. For iron-based superconductors, the superconducting transition temperature attains a maximum value when the FePn₄ (Pn = pnictogen) tetrahedra form a regular shape (Pn–Fe–Pn angles being 109.5°) and meanwhile the anion height dependence of T_c shows a symmetric curve with a peak around 1.38 Å.^3a,19 In the case of (NH₃)Fe_0.25Fe₂S₂, both the S–Fe–S angle and the anion height deviate far from the above ideal values for optimal superconductivity to occur. On the other hand, the interlayer Fe, which contributes to the low-temperature ferromagnetism and leads to weak localization, may also probably depress the superconductivity, similar to the role played by excess Fe in Fe_1+δSe_1−xTe_x and nonsuperconducting Fe_1+δS.²⁶ Nevertheless, (NH₃)Fe_0.25Fe₂S₂ could act as a potential parent compound of new superconductors if appropriate chemical substitution was achieved. For example, if we substitute Se for S, superconductivity may probably be induced, just as the superconductor Li_1−xFe_xOHFeS_1−ySe_y and K_xFe_2−yS_1−zSe_z.^21a,29

Conclusions

In summary, a new intercalation compound of FeS with chemical composition of (NH₃)Fe_0.25Fe₂S₂ has been successfully synthesized via a hydrothermal method. This compound crystallizes in the tetragonal space group I4/mmm and consists of tetrahedral FeS layers stacking along the c axis with ammonia and excess iron cointercalated between the layers. It is highly crystalline and more thermally stable than the other ammonia intercalated iron selenide superconductors. Magnetic susceptibility and electrical resistivity measurements show that it is a ferromagnetic semiconductor and no superconductivity is observed down to 2 K. The transport properties at low temperatures are dominated by electron-like carriers and the sign reversal and strong temperature dependence of R_H may be caused by multi-band effect. The interlayer iron, which induces disorder and carries local moment, plays an important role in suppressing superconductivity, contributing to low-temperature ferromagnetism, resulting in a weakly localized electronic state, and more importantly, leading to a novel crossover from negative to positive MR. The intriguing behavior of MR can be explained by disorder effects and spin-dependent scattering mechanism. Our results not only enrich the understanding of the intercalation chemistry of iron chalcogenide superconductors but also will shed new light on the structure and physical properties of the intercalation compounds of them.

Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

This work was financially supported by Innovation Program and “Strategic Priority Research Program (B)” of the Chinese Academy of Sciences (Grants KJCX2-EW-W11 and XDB04040200), National Natural Science Foundation of China (Grants 91122034, 51125006, 51202279, 61376056, 21201012, and 11275012), National Program of China (Grant 2016YFB0901600), and Science and Technology Commission of Shanghai (Grant 12XD1406800).

Notes and references

1. (a) R. von Helmolt, J. Wecker, B. Holzapfel, L. Schultz and K. Samwer, Phys. Rev. Lett., 1993, 71, 2331–2333; (b) J. G. Bednorz and K. A. Müller, Z. Phys. B: Condens. Matter, 1986, 64, 189–193.
2. (a) Y. Kamihara, T. Watanabe, M. Hirano and H. Hosono, J. Am. Chem. Soc., 2008, 130, 3296–3297; (b) F. C. Hsu, J. Y. Luo, K. W. Yeh, T. K. Chen, T. W. Huang, P. M. Wu, Y. C. Lee, Y. L. Huang, Y. Y. Chu, D. C. Yan and M. K. Wu, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 14262–14264.
3. (a) C.-H. Lee, A. Iyo, H. Eisaki, H. Kito, M. Teresa Fernandez-diaz, R. Kumai, K. Miyazawa, K. Kihou, H. Matsuhata and M. Braden, J. Phys. Soc. Jpn., 2008, 77, 44–46; (b) M. D. Lumsden and A. D. Christianson, J. Phys.: Condens. Matter, 2010, 22, 203203.
4. (a) K.-W. Yeh, T.-W. Huang, Y.-l. Huang, T.-K. Chen, F.-C. Hsu, P. M. Wu, Y.-C. Lee, Y.-Y. Chu, C.-L. Chen, J.-Y. Luo, D.-C. Yan and M.-K. Wu, Europhys. Lett., 2008, 84, 37002; (b) S. Medvedev, T. McQueen, I. Troyan, T. Palasyuk, M. Eremets, R. Cava, S. Naghavi, F. Casper, V. Ksenofontov and G. Wortmann, Nat. Mater., 2009, 8, 630–633.
5. (a) J. Guo, S. Jin, G. Wang, S. Wang, K. Zhu, T. Zhou, M. He and X. Chen, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 180520; (b) M. Rotter, M. Tegel and D. Johrendt, Phys. Rev. Lett., 2008, 101, 107006.
6. T. Qian, X. P. Wang, W. C. Jin, P. Zhang, P. Richard, G. Xu, X. Dai, Z. Fang, J. G. Guo, X. L. Chen and H. Ding, Phys. Rev. Lett., 2011, 106, 187001.
7. (a) Y. Liu, Q. Xing, K. W. Dennis, R. W. McCallum and T. A. Lograsso, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 86, 144507; (b) Z.-W. Wang, Z. Wang, Y.-J. Song, C. Ma, Y. Cai, Z. Chen, H.-F. Tian, H.-X. Yang, G.-F. Chen and J.-Q. Li, J. Phys. Chem. C, 2012, 116, 17847–17852.
8. (a) T. P. Ying, X. L. Chen, G. Wang, S. F. Jin, T. T. Zhou, X. F. Lai, H. Zhang and W. Y. Wang, Sci. Rep., 2012, 2, 426; (b) T. Ying, X. Chen, G. Wang, S. Jin, X. Lai, T. Zhou, H. Zhang, S. Shen and W. Wang, J. Am. Chem. Soc., 2013, 135, 2951–2954; (c) M. Burrard-Lucas, D. G. Free, S. J. Sedlmaier, J. D. Wright, S. J. Cassidy, Y. Hara, A. J. Corkett, T. Lancaster, P. J. Baker, S. J. Blundell and S. J. Clarke, Nat. Mater., 2013, 12, 15–19; (d) J. Guo, H. Lei, F. Hayashi and H. Hosono, Nat. Commun., 2014, 5, 4756; (e) A. Krzton-Maziopa, E. V. Pomjakushina, V. Y. Pomjakushin, F. von Rohr, A. Schilling and K. Conder, J. Phys.: Condens. Matter, 2012, 24, 382202; (f) T. Hatakeda, T. Noji, T. Kawamata, M. Kato and Y. Koike, J. Phys. Soc. Jpn., 2013, 82, 123705; (g) S. Hosono, T. Noji, T. Hatakeda, T. Kawamata, M. Kato and Y. Koike, J. Phys. Soc. Jpn., 2014, 83, 113704; (h) S.-J. Shen, T.-P. Ying, G. Wang, S.-F. Jin, H. Zhang, Z.-P. Lin and X.-L. Chen, Chin. Phys. B, 2015, 24, 117406; (i) X. F. Lu, N. Z. Wang, G. H. Zhang, X. G. Luo, Z. M. Ma, B. Lei, F. Q. Huang and X. H. Chen, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 89, 020507; (j) X. F. Lu, N. Z. Wang, H. Wu, Y. P. Wu, D. Zhao, X. Z. Zeng, X. G. Luo, T. Wu, W. Bao, G. H. Zhang, F. Q. Huang, Q. Z. Huang and X. H. Chen, Nat. Mater., 2015, 14, 325–329; (k) U. Pachmayr, F. Nitsche, H. Luetkens, S. Kamusella, F. Bruckner, R. Sarkar, H. H. Klauss and D. Johrendt, Angew. Chem., Int. Ed., 2015, 54, 293–297.
9. X. Lai, H. Zhang, Y. Wang, X. Wang, X. Zhang, J. Lin and F. Huang, J. Am. Chem. Soc., 2015, 137, 10148–10151.
10. U. Pachmayr, N. Fehn and D. Johrendt, Chem. Commun., 2016, 52, 194–197.
11. (a) H. Lin, Y. Li, Q. Deng, J. Xing, J. Liu, X. Zhu, H. Yang and H.-H. Wen, Phys. Rev. B, 2016, 93, 144505; (b) C. K. H. Borg, X. Zhou, C. Eckberg, D. J. Campbell, S. R. Saha, J. Paglione and E. E. Rodriguez, Phys. Rev. B, 2016, 93, 094522.
12. (a) J. J. Ying, Z. J. Xiang, Z. Y. Li, Y. J. Yan, M. Zhang, A. F. Wang, X. G. Luo and X. H. Chen, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 054506; (b) X. Zhang, X. Lai, N. Yi, J. He, H. Chen, H. Zhang, J. Lin and F. Huang, RSC Adv., 2015, 5, 38248–38253.
13. (a) X. Lai, X. Chen, S. Jin, G. Wang, T. Zhou, T. Ying, H. Zhang, S. Shen and W. Wang, Inorg. Chem., 2013, 52, 12860–12862; (b) H. Zhang, X. Wu, D. Li, S. Jin, X. Chen, T. Zhang, Z. Lin, S. Shen, D. Yuan and X. Chen, J. Phys.: Condens. Matter, 2016, 28, 145701.
14. (a) A. Boultif and D. Louer, J. Appl. Crystallogr., 1991, 24, 987–993; (b) J. R. Carvajal, Phys. B, 1993, 192, 55–69.
15. S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. I. J. Probert, K. Refson and M. C. Payne, Z. Kristallogr., 2005, 220, 567–570.
16. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
17. D. Vanderbilt, Phys. Rev. B: Condens. Matter Mater. Phys., 1990, 41, 7892.
18. H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Solid State, 1976, 13, 5188.
19. Y. Mizuguchi, Y. Hara, K. Deguchi, S. Tsuda, T. Yamaguchi, K. Takeda, H. Kotegawa, H. Tou and Y. Takano, Supercond. Sci. Technol., 2010, 23, 054013.
20. M. J. Manos, K. Chrissafis and M. G. Kanatzidis, J. Am. Chem. Soc., 2006, 128, 8875–8883.
21. (a) U. Pachmayr and D. Johrendt, Chem. Commun., 2015, 51, 4689–4692; (b) I. Tsubokawa and S. Chiba, J. Phys. Soc. Jpn., 1959, 14, 1120.
22. T. Liu, X. Ke, B. Qian, J. Hu, D. Fobes, E. Vehstedt, H. Pham, J. Yang, M. Fang and L. Spinu, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 80, 174509.
23. (a) B. Lei, J. H. Cui, Z. J. Xiang, C. Shang, N. Z. Wang, G. J. Ye, X. G. Luo, T. Wu, Z. Sun and X. H. Chen, Phys. Rev. Lett., 2016, 116, 077002; (b) B. Lei, Z. J. Xiang, X. F. Lu, N. Z. Wang, J. R. Chang, C. Shang, A. M. Zhang, Q. M. Zhang, X. G. Luo, T. Wu, Z. Sun and X. H. Chen, Phys. Rev. B, 2016, 93, 060501.
24. H. Hwang and S.-W. Cheong, Nature, 1997, 389, 942–944.
25. (a) D. R. Parker, M. A. Green, S. T. Bramwell, A. S. Wills, J. S. Gardner and D. A. Neumann, J. Am. Chem. Soc., 2004, 126, 2710–2711; (b) A. M. Arévalo-López, G. M. McNally and J. P. Attfield, Angew. Chem., Int. Ed., 2015, 54, 12074–12077; (c) S. Gupta, K. G. Suresh and A. K. Nigam, J. Alloys Compd., 2014, 586, 600–604; (d) Y. Zhang, H. Ning, Y. Li, Y. Liu and J. Wang, Appl. Phys. Lett., 2016, 108, 153114; (e) Y.-B. Zhou, B.-H. Han, Z.-M. Liao, H.-C. Wu and D.-P. Yu, Appl. Phys. Lett., 2011, 98, 222502.
26. (a) H. Chang, J. Luo, C. Wu, F. Hsu, T. Huang, P. Wu, M. Wu and M. Wang, Supercond. Sci. Technol., 2012, 25, 035004; (b) S. Denholme, H. Okazaki, S. Demura, K. Deguchi, M. Fujioka, T. Yamaguchi, H. Takeya, M. ElMassalami, H. Fujiwara, T. Wakita, T. Yokoya and Y. Takano, Sci. Technol. Adv. Mater., 2014, 15, 055007.
27. H. Yamada and S. Takada, J. Phys. Soc. Jpn., 1973, 34, 51–57.
28. J. Hu and T. F. Rosenbaum, Nat. Mater., 2008, 7, 697–700.
29. J. Guo, X. Chen, G. Wang, S. Jin, T. Zhou and X. Lai, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 054507.

---

Footnotes

† Electronic supplementary information (ESI) available: PXRD collected immediately after the TGA experiments; electronic density of states; PXRD collected after annealed at 200 °C for 48 h; linear fitting of ln ρ vs. 1/T curve. CCDC 1480191. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra17568f

‡ Xiaofang Lai and Zhiping Lin contribute equally.

---