Template preparation of NiPS₃ polymer nanocomposites

Abstract

A template synthesis method can be used to prepare NiPS₃ nanocomposites with polyethylene oxide (PEO), polyethylenimine (PEI), polyvinyl alcohol (PVA) and polyvinylpyrrolidone (PVP). Nanocomposites of NiPS₃/PEI, NiPS₃/PVA and NiPS₃/PVP have not been previously reported. The resulting materials are characterized by powder X-ray diffraction (PXRD), thermogravimetric analysis (TGA), scanning electron microscope (SEM) and energy-dispersive X-ray spectroscopy (EDS). Reaction progress was monitored using UV-visible spectroscopy. Template synthesized NiPS₃/PEO nanocomposites show only polymer zig-zag monolayers, in contrast to the bilayers obtained using the topotactic method, and can generate denser interlayers than those obtained by topotactic methods. The crystallite sizes increase for more dilute conditions, a more polar solvent, and longer aging time. P₂S₆⁴⁻ and Ni²⁺ concentrations govern nanocomposite nucleation and growth, respectively.

---

Introduction

MPS₃ and materials containing MPS₃ layers show interesting properties for nonlinear optics,^1,2 magnetic^3,4 or photomagnetic devices,⁵ catalysis,⁶ and energy storage in secondary batteries.⁷ For the latter example, NiPS₃ shows an electrochemical reduction that proceeds in several stages at potentials from 2.7 to 1.8 V up to a Li/Ni ratio of ∼7.5. The reduction occurs quasi-reversibly over the range 0 ≤ Li/Ni ≤ 3 range, providing a high theoretical capacity as a secondary cathode.⁷

Xiao et al. reported that MoS₂ shows a significantly higher reversible capacity after an exfoliation process. Furthermore, the formation of a nanocomposite with polyethylene oxide (PEO) allows both higher capacities and improved charge/discharge cycling.⁸ Since MS₂ and MPS₃ are both layered materials with intercalation chemistries, the preparation and electrochemical characterization of PEO/NiPS₃ nanocomposites are therefore of interest.

The synthesis and structures of MPS₃ nanocomposites with different polymers has been studied by several groups. In all cases, a topotactic method has been employed, starting with a compound containing alkali metal cations between MPS₃ layers. This intercalation compound may be obtained directly at high temperature,⁹ or by introduction of alkali metal cations into the MPS₃ hostvia ion exchange¹⁰ or reduction of the M(II) cations in the host layers.⁶ The alkali metal cations likely enable nanocomposite formation by coordinating to polymer functional groups, as has been observed with many other layered hosts.¹¹

Lagadic et al. first reported the synthesis of nanocomposites of PEO and MPS₃ (M = Mn, Cd) by reacting an aqueous polymer solution with M_1-xPS₃A_2x(H₂O)_y (A = Na, K). The products obtained showed gallery expansions of ≈0.8–0.9 nm, consistent with the intercalation of bilayers of zig-zag-like PEO chains that coordinate to alkali metal ions.¹⁰ Jeevanandam et al. subsequently prepared PEO/CdPS₃ nanocomposites in a methanolic solution at 55 °C.¹² Manriquez et al. reported the preparation of PEO nanocomposites from Li_xMPS₃ (M = Ni, Fe) using a similar approach. However, in this case the authors proposed a polymer helical conformation.⁹

Manova et al. reported the intercalation of polyethylene glycol into a Na_xNiPS₃ by reaction in a methanolic aqueous solution, with a gallery expansion of ≈0.8 nm.⁶

Other polymers that are known to form nanocomposites with MPS₃ layers include linear polyethylenimine, LPEI, and polyvinylpyrrolidone, PVP. The former was reported by Oriakhi et al. to form nanocomposites with polymer monolayers (expansion of 0.4 nm) via the reaction of K_x(H₂O)_yM1-x_/2PS₃ (M = Mn, Cd) and polymer in aqueous solution;¹³ the latter was reported by Yang et al. via reaction of a potassium intercalated intermediate MPS₃ (M = Mn, Cd) and an aqueous PVP solution. The interlayer expansion in that case was ≈2.3 nm, much greater than that for PEO.¹⁴

A solvent-free, melt intercalation approach to obtain PEO/MPS₃ nanocomposites (M = Mn, Cd) was reported by Sukpirom et al. With this method, K_xM_1-x/2PS₃·δH₂O was shown to directly intercalate PEO at 125 °C. The nanocomposite basal repeat distances were increased by ≈0.8 nm, again ascribed to the presence of PEO bilayers.¹⁵

MPS₃ compounds are normally prepared by combining the elements or simple precursors at elevated temperature.^16,17 However, the synthesis of highly disordered or amorphous NiPS₃ at ambient temperature has been reported by reaction of Ni²⁺ salts with alkali metal hexathiohypodiphosphate (Na₄P₂S₆·6H₂O or Li₄P₂S₆·6H₂O) in aqueous solution.^18–20 In the following study, we describe the first template approach to MPS₃ compounds, where layered nanocomposites are obtained directly using a one-pot reaction with solution-phase precursors at ambient temperature.

Experimental

Reagents and analytical methods

Li₂S (99.9%, Cerac), P (99%, Johnson Matthey), S (99.9%, Mallinckrodt), Ni(NO₃)₂·6H₂O (Aldrich), PEO (Aldrich, M_w = 100,000), linear polyethylenimine (Alfa Aeser, M_w = 25,000), polyvinyl alcohol, 80% hydrolyzed (Aldrich, M_w = 9–10,000) and polyvinylpyrrolidone (Alfa Aeser, M_w = 8,000) were purchased and used without further purification.

Powder XRD (PXRD) patterns from 3–60° 2θ were obtained on a Rigaku Miniflex II diffractometer, using Ni–filtered Cu-Kα radiation at a scan rate of 3° min⁻¹ and step size of 0.020° 2θ. A modified Scherrer equation²¹ was used to determine crystallite size (L) for the prepared nanocomposites:

where λ = 0.15406 nm, Δ(2θ_hkl) = full width at half maximum (in radians) and θ_hkl = diffraction angle. A Si(m) powder standard was employed for these measurements. Thermogravimetric (TGA) analyses were performed under Ar flow (20 mL min⁻¹) using a Shimadzu TGA-50 analyzer, from ambient to 500 °C at 5 °C min⁻¹. An Agilent 8453 UV-visible spectrophotometer was used for UV analyses. Background subtractions were carried out for all the UV-visible measurements. Scanning electron micrographs (SEMs) were obtained using an FEI Quanta 600F FEG SEM. Energy-dispersive X-ray spectroscopy (EDS) was used as a semi-quantitative elemental analysis method.

Van der Waals volumes of C₂H₄O oligomers (C = 2–40) were calculated based on Bondi radii using a method termed Atomic and Bond Contributions of van der Waals volume (VABC).²²

Syntheses

Li₄P₂S₆ was synthesized as a white powder according to a published procedure.²³ Li₂S, P and S were combined in 1 : 1 : 2 mole ratio and sealed in a quartz tube under an inert atmosphere. The mixture was heated to 900 °C at 100 °C h⁻¹, maintained at 900 °C for 0.5 h, rapidly cooled to 450 °C and held at 450 °C for 24 h. Highly disordered or amorphous NiPS₃ was synthesized via a process previously reported by Prouzet et al.¹⁹ An aqueous solution of Ni(NO₃)₂·6H₂O (0.31 g, 5.00 mL) was added dropwise to an aqueous solution of Li₄P₂S₆ (0.10 g, 5.00 mL) and stirred vigorously for 24 h. The reddish precipitate was centrifuged for 15 min, triple washed with deionized (DI) water, and then dried under vacuum at room temperature.

To prepare the nanocomposites, Ni(NO₃)₂·6H₂O (0.31 g) and PEO (0.01–0.05 g) were dissolved in 5–500 mL of DI water and then slowly added to a solution Li₄P₂S₆ (0.1 g) dissolved in DI water (5.00–500.0 mL) and stirred vigorously for 24 h. Depending on the solvent volume, the precipitation time varied from few minutes to 12 days. After aging (1–7 days) the resulting brown precipitate was centrifuged for 15 min and washed three times with a copious amount of DI water. The solid product was dried under vacuum at 60 °C for 18 h. Where noted, some reactions substituted water with 10% ethanol as the solvent.

Samples are identified below based on starting g/g PEO/Li₄P₂S₆ ratio (R) and total solvent volume (V). For example, the product of Ni(NO₃)₂·6H₂O (0.31 g) and PEO (0.05 g) dissolved in 5 mL of DI water with 5 mL of aqueous Li₄P₂S₆ (0.1 g) solution is labeled R0.5V10.

The same procedures were used to synthesize other polymer/NiPS₃ nanocomposites with linear polyethylenimine (LPEI), polyvinyl alcohol (PVA) and polyvinylpyrrolidone (PVP).

Results and discussion

PXRD patterns (Fig. 1) show that polymer nanocomposite structures are obtained by the template method. The basal repeat distance increases from 0.64 nm in a-NiPS₃ to 1.06 nm for all the PEO nanocomposites. This expansion of 0.42 nm along the stacking direction indicates the presence of PEO monolayers incorporated between NiPS₃ layers.

Fig. 1 PXRD patterns for PEO/NiPS₃ nanocomposites obtained by the template method. a-NiPS₃ obtained by the same approach without PEO is shown for comparison. (*Al holder).

Fig. 2 shows a typical microstructure of the prepared PEO/NiPS₃ nanocomposites, in this case for sample R0.5V10. The nanocomposites show an expanded layered morphology with most particle dimensions above 10 μm. For samples with R = 0.5, i.e. with a high polymer content, SEM images also reveal the presence of surface adsorbed polymer and show charging effects during imaging, indicating the presence of non-conducting polymer on surfaces. TGA data (see below) support the presence of surface-adsorbed PEO.

Fig. 2 SEM image for a PEO/NiPS₃ nanocomposite (sample R0.5V10).

EDS data indicates the presence of Ni, P, S, C and O in the nanocomposite samples with an Ni : P  : S ratio of 1 : 1 : 2.7. This supports the formation of host NiPS₃ structure from the precursors. Some sample degradation during analysis may account for the less than nominal sulfur content.

Fig. 3 shows thermal analysis data of the obtained products. The DTG loss peak at 100–150 °C is ascribed to volatilization of intercalated water, and the loss peak around 250 °C to PEO decomposition and volatilization. From the mass losses, the relative contents of these components can be determined. For the R0.5V10 sample, some new features are observed. The water loss region occurs over a broad range starting at lower temperature, a shoulder appears on the low temperature side of the main PEO loss peak, and a high temperature loss is observed at 350 °C. The low temperature shoulder on the main PEO loss peak, and the high temperature loss, are ascribed to the presence of surface-adsorbed or bulk PEO. The separation of thermal loss events allows quantification of PEO intercalate vs. surface PEO. Water content was close to 1 mol/formula unit of NiPS₃, but this composition was highly dependent on sample history and environmental conditions (humidity and temperature).

Fig. 3 TGA of nanocomposites with different PEO compositions (* loss shoulder ascribed to surface PEO) DTG indicates the first derivative of TGA curves.

The packing fractions, i.e. fraction of volume in the expanded galleries occupied by intercalates, were derived from the known hostMPS₃ cell dimensions, the gallery expansion determined by PXRD, the product stoichiometries and intercalate volumes. The PEO monomer unit volume was estimated by calculating the C₂H₄O monomer unit volume for increasing oligomer sizes until a stable volume of 0.050 nm³ was observed. (Fig. 4) As an example, for sample R0.1V10, the packing fraction, was calculated as:

Packing fraction = [(Z * 0.34 * 0.050 nm³)/(a * b * Δd)] = 0.28 (2)

Fig. 4 Calculated C₂H₄O unit volumes from VABC.

where a = 0.5812 nm and b = 1.0070 nm are NiPS₃ unit cell parameters,¹⁶ Δd = gallery expansion = 0.42 nm, and Z = NiPS₃ units per unit cell = 4. Water content was highly variable, as noted above, and may be on particle surfaces or intercalated, and thus was not considered in these packing fractions. The calculated PEO packing fractions are shown in Table 1. Intercalate PEO content and packing fractions increase with higher PEO/Li₄P₂S₆ up to R = 0.3. At higher ratios, the additional PEO does not intercalate, although some remains as surface-adsorbed polymer.

Table 1 Compositional data from thermal analyses and calculated packing fractions for PEO/NiPS₃ nanocomposites

Sample	Reacted (PEO)/NiPS3 (mol/mol)	Product (PEO)/NiPS3 (mol/mol)			H2O/NiPS3 (mol/mol)	Packing fraction
		Total	Intercalated	Surface
R0.1V10	0.34	0.34	0.34	—	1.29	0.28
R0.2V10	0.68	0.69	0.69	—	1.21	0.56
R0.3V10	1.01	1.02	1.02	—	1.01	0.83
R0.4V10	1.35	1.03	1.03	—	1.17	0.84
R0.5V10	1.69	1.30	1.01	0.29	1.34	0.82

---

As determined from diffraction peak widths, crystallite sizes for the nanocomposites, which are the products of 0.3 g/g PEO/Li₄P₂S₆ reaction mixtures, increases from 2.6 nm to 4.2 nm upon a 20-fold decrease in the reaction concentrations as shown in Table 2. A further dilution of 5 × results in a crystallite size increase to 4.5 nm. Aggregation times were also greatly increased in the more dilute solutions.

Table 2 Effect of synthetic conditions on crystallite size

Sample	Aggregation time (days)	Solvent	Aging time (days)	L (nm)
R0.3V10	0.004	DI water	1	2.6
R0.3V200	4–5	DI water	1	4.2
R0.3V1000	11–12	DI water	1	4.5
R0.3V1000	11–12	DI water	4	4.7
R0.3V1000	11–12	DI water	7	5.2
R0.3V1000	5–6	10% Ethanol	1	4.2

---

The use of an ethanolic solvent also decreases the aggregation time with respect to the aqueous reaction mixtures, and results in a smaller crystallite size. For an R0.3V1000 sample prepared using 10% ethanol, the crystallite size decreased from 4.5 nm (in DI water) to 4.2 nm. Crystallite size increased from 4.5 to 5.2 when the aging time was increased from 1 to 7 days, suggesting that exchange between solution and precipitate continues after aggregate formation.

Nanocomposites of PEI/NiPS₃, PVA/NiPS₃ and PVP/NiPS₃ have not been previously reported, but can be prepared by this template method. Table 3 shows the structural and compositional data for these new nanocomposites, obtained as for PEO/NiPS₃.

Table 3 Structural and compositional data for new polymer/NiPS₃ nanocomposites

Nanocomposite	d (nm)	Δd (nm)	Polymer/ NiPS3 ratio (mol/mol)
PEO/NiPS3	1.06	0.42	1.0
PEI/NiPS3	1.05	0.41	1.0
PVA/NiPS3	1.06	0.42	0.9
PVP/NiPS3	2.20	1.56	0.5

---

In previous reports, PEO/MPS₃ (M = Mn, Cd) nanocomposites have shown interlayer expansions of 0.8–0.9 nm, corresponding to the presence of PEO bilayers between host sheets. For these nanocomposites the polymer/MPS₃ ratio is 1.9–2.1 which supports the zig-zag bilayer arrangement.¹⁰ Alternately, Manriquez et al. describe a single helical PEO arrangement in MPS₃ nanocomposites (M = Ni, Fe) with an interlayer expansion of ≈0.8 nm.⁹ All these compounds were synthesized using a topotactic method. The current template method results in only single PEO zig-zag polymer layers. However, these layers are more densely packed, as is indicated by the derived packing fractions shown in Table 4. Topotactic methods are kinetically limited by the slow diffusion of polymers,²⁴ which may impede development of densely packed intercalate layers. With a template method, denser polymer layers can form before the host layer structure fully develops.

Table 4 Comparison of packing fractions of known PEO/MPS₃ nanocomposites. The packing fractions were derived from compositional and structural data in the references indicated

Nanocomposite	Method of synthesis	Alkali ion	Packing fraction	Reference
PEO/NiPS3	Topotactic	Li^+	0.27	9
PEO/FePS3	Topotactic	Li^+	0.40	9
PEO/CdPS3	Topotactic	Li^+, Na^+	0.58,0.72	12
PEO/CdPS3	Topotactic	Cs^+, K^+	0.62,0.69	12
PEO/CdPS3	Topotactic	K^+	0.61	10
PEO/MnPS3	Topotactic	Na^+, K^+	0.73,0.77	10
PEO/NiPS3	Template	Li^+	0.84	this work

---

The PEI/NiPS₃ nanocomposites obtained show a similar interlayer expansion to that from the topotactic preparation (0.41 nm vs. 0.4 nm for PEI/MPS₃ (M = Mn, Cd) reported by Oriakhi et al.),¹³ corresponding to a PEI monolayer. PVA/NiPS₃ shows a similar interlayer expansion of 0.42 nm. The template synthesis of PVP/NiPS₃ gives a much smaller expansion of 1.46 nm than for the topotactic method.¹⁴

In order to evaluate the growth mechanism for these template reactions, the evolution of UV-visible absorption spectra were recorded under different reaction conditions. As shown in Fig. 5, a new absorbance at 330 nm indicates the formation of the nanocomposite. This absorbance can be ascribed to Ni coordination by sulfide in P₂S₆⁴⁻ (Ni–dithiolene complexes show a prominent absorbance in this region).^25,26

Fig. 5 UV-visible spectra of some reactants and products of the template reactions. The absorbance at 330 nm (*) is used to monitor nanocomposite formation.

In the first set of experiments, the concentrations of Ni²⁺ and PEO were varied while maintaining a constant P₂S₆⁴⁻ concentration and a constant Ni²⁺/PEO ratio of 1 : 1. The initial absorbances at 330 nm under these conditions remain unchanged (Fig. 6, compare the y-intercept of the three plots), indicating that nanocomposite nucleation is independent of PEO and Ni²⁺ concentration. These initial values are all obtained within 15 s of solution mixing.

Fig. 6 Effect of Ni²⁺ and PEO concentrations on the rate of nanocomposite formation. Curves are obtained using the indicated Ni²⁺ concentrations.

The growth in the absorbance peak (from about 1.75 to 2.25 (Fig. 6, curve for 2 mM), indicates that the rate of nucleation (as indicated by the initial values) are significantly greater than the rate of nanocomposite growth. The growth rate, however, is strongly dependent on Ni²⁺ concentration.

In contrast, the initial absorbances vary significantly for reactions using three different concentrations of P₂S₆⁴⁻ with a constant Ni²⁺ and PEO concentrations and Ni²⁺/PEO = 1. (Fig. 7) This confirms the role of P₂S₆⁴⁻ in nucleation. The growth curves are parallel under these conditions, indicating that nanocomposite growth is not significantly affected by P₂S₆⁴⁻ concentration in this range.

Fig. 7 Effect of P₄S₆⁴⁻ concentration on the rate of nanocomposite formation. Curves are obtained using the indicated P₄S₆⁴⁻ concentrations.

Conclusions

A template synthesis method can be used to synthesize NiPS₃ nanocomposites. The obtained NiPS₃/PEO nanocomposites show polymer zig-zag monolayers at all combination ratios, in contrast to PEO nanocomposites derived from a topotactic method, and the packing fractions are greater. The product crystallite size increases with higher solvent volume, use of polar solvent and higher aging time. P₂S₆⁴⁻ and Ni²⁺ concentrations govern nanocomposite nucleation and growth, respectively. Nanocomposites of NiPS₃/PEI, NiPS₃/PVA and NiPS₃/PVP are reported for the first time using this template method.

References

1. T. Coradin, R. Clement, P. G. Lacroix and K. Nakatani, Chem. Mater., 1996, 8, 2153.
2. I. Lagadic, P. G. Lacroix and R. Clement, Chem. Mater., 1997, 9, 2004.
3. X. Chen, Z. Li, H. Zhou, T. Wang, J. Qin and M. Inokuchi, Polymer, 2007, 48, 3256.
4. R. Clement, J. J. Girerd and I. Morgensternbadarau, Inorg. Chem., 1980, 19, 2852.
5. S. Benard, A. Leaustic, E. Riviere, P. Yu and R. Clement, Chem. Mater., 2001, 13, 3709.
6. E. Manova, C. Severac, A. Andreev and R. Clement, J. Catal., 1997, 169, 503.
7. Y. V. Kuzminskii, B. M. Voronin, I. M. Petrushina, N. N. Redin and G. P. Prikhodko, J. Power Sources, 1995, 55, 1.
8. J. Xiao, D. Choi, L. Cosimbescu, P. Koech, J. Liu and J. P. Lemmon, Chem. Mater., 2010, 22, 4522.
9. V. Manriquez, P. Barahona, D. Ruiz and R. E. Avila, Mater. Res. Bull., 2005, 40, 475.
10. I. Lagadic, A. Leaustic and R. Clement, J. Chem. Soc., Chem. Commun., 1992, 1396.
11. M. M. Lerner and C. O. Oriakhi, Handbook of Nanophase Materials, ed. A. N. Goldstein, Dekker, New York, N.Y, 1997, 199–219.
12. P. Jeevanandam and S. Vasudevan, Chem. Mater., 1998, 10, 1276.
13. C. O. Oriakhi, R. L. Nafshun and M. M. Lerner, Mater. Res. Bull., 1996, 31, 1513.
14. D. Yang and R. F. Frindt, J. Mater. Res., 2000, 15, 2408.
15. N. Sukpirom, C. O. Oriakhi and M. M. Lerner, Mater. Res. Bull., 2000, 35, 325.
16. G. Ouvrard, R. Brec and J. Rouxel, Mater. Res. Bull., 1985, 20, 1181.
17. R. Brec, Solid State Ionics, 1986, 22, 3.
18. P. J. S. Foot and B. A. Nevett, J. Chem. Soc., Chem. Commun., 1987, 380.
19. E. Prouzet, G. Ouvrard, R. Brec and P. Seguineau, Solid State Ionics, 1988, 31, 79.
20. H. Loboue, C. Guillot-Deudon, A. F. Popa, A. Lafond, B. Rebours, C. Pichon, T. Cseri, G. Berhault and C. Geantet, Catal. Today, 2008, 130, 63.
21. P. Fragnaud, E. Prouzet and R. Brec, J. Mater. Res., 1992, 7, 1839.
22. Y. H. Zhao, M. H. Abraham and A. M. Zissimos, J. Org. Chem., 2003, 68, 7368.
23. R. Mercier, J. P. Malugani, B. Fahys, J. Douglade and G. Robert, J. Solid State Chem., 1982, 43, 151.
24. C. O. Oriakhi, I. V. Farr and M. M. Lerner, J. Mater. Chem., 1996, 6, 103.
25. J. F. Bai, J. L. Zuo, W. L. Tan, W. Ji, Z. Shen, H. K. Fun, K. Chinnakali, I. A. Razak, X. Z. You and C. M. Che, J. Mater. Chem., 1999, 9, 2419.
26. Y. Ji, J. L. Zuo, L. X. Chen, Y. Q. Tian, Y. L. Song, Y. Z. Li and X. Z. You, J. Phys. Chem. Solids, 2005, 66, 207.

---