Microstructure and magnetism of Co-doped PbPdO₂ films with different grain sizes

Abstract

Single-phase Co-doped PbPdO₂ films with grain sizes ranging from 30 nm to 120 nm were synthesized using a sol–gel spin-coating method. The doped Co ions exhibit a mixed state of 2+ and 3+, which is caused by the existence of Pb vacancies and the low electronegativity of Co. Ferromagnetism, paramagnetism, antiferromagnetism and a charge ordering state were observed in all of the Co-doped PbPdO₂ films. Additionally, the charge ordering signal determined by the surface state of the nanograins is gradually weakened with decreasing grain size. The carrier-mediated mechanism bridged to the bound magnetic polaron model was used to explain the coexistence of the ferromagnetism, paramagnetism and antiferromagnetism. For the ferromagnetism, the tendency of the saturation magnetization to increase with temperature was found in all the Co-doped PbPdO₂ films with a nanograin structure. This abnormal magnetic behavior is believed to be one of the peculiar features of spin gapless semiconductors.

---

Introduction

PbPdO₂ with the gapless structure was first reported by Wang in 2008.¹ In contrast to other gapless semiconductors, such as Hg-based IV–VI compounds, PbPdO₂ is a type of spintronic material without the problems of being toxic and easily oxidized.² Hence, it is compatible to oxide semiconductor devices. In recent years, PbPdO₂-based magnetic semiconductors have attracted much attention from researchers.^2–7 These studies are mainly focused on introducing spin into PbPdO₂ by doping with transitional metal ions, and exploring the properties of the doped PbPdO₂, especially magnetic properties. PbPdO₂ doped with Cu, Co, Mn and Zn have been synthesized and studied.^5–7 While spin-gapless-related high-temperature ferromagnetism (FM), as predicted by theoretical calculations, was reported in a Co-doped PbPdO₂ nanograin film, only low-temperature FM at 2 K was observed in bulk PbPdO₂ doped with Cu, Co, Mn and Zn.^1,2,5–7 Apart from the doping ions and the doping level, we suggest attention is paid to the grain size, which is one of the biggest variables for these reported PbPdO₂-based semiconductors with different magnetic behaviors. As proved previously, the grain size has a great effect on the material’s magnetic properties, as observed in grain-size-related antiferromagnetism (AFM), superparamagnetism (SPM) and charge ordering (CO) states.^8–12 With a decrease in particle size, the AFM transition temperature and the SPM blocking temperature may decrease gradually and the CO transition was reported to be suppressed and finally disappear.^9–12 Currently, the study of grain-size-related magnetism in PbPdO₂-based semiconductor is rare. Under these circumstances, an investigation of the magnetism of Co-doped PbPdO₂ nanograin films with different grain sizes was performed in this paper. The films were prepared by using a sol–gel spin-coating method. Different grain sizes were achieved through adjusting the calcination time. The film’s phase structure, composition, morphology and grain size were characterized in detail. The magnetism of the films and the Co valence state were studied to clarify the magnetic origin of Co-doped PbPdO₂ films.

Results and discussion

Fig. 1 shows the XRD patterns of the Co-doped PbPdO₂ films calcined for 0.5 h, 1.5 h and 2.5 h, respectively. Besides the diffraction peaks of the sapphire substrate labeled by the black plum at about 53°, only the PbPdO₂ phase with an orthorhombic structure (JCPDS PDF # 72-2372) can be found in all three films. Impurities such as Pb-related or Pd-related oxides are not observed in these films. Considering the Pb volatility during the heat treatment, the amount of Pb that is volatilized will naturally increase with prolonging the calcination time. This indicates that the single-phase films calcined for different times can be successfully obtained through adjusting the amount of Pb(NO₃)₂.

Fig. 1 XRD patterns of the Co-doped PbPdO₂ films calcined for 0.5 h, 1.5 h and 2.5 h.

Fig. 2 shows the FE-SEM images of the surface and cross-section (inset, upper right) morphology and EDS spectra (inset, lower right) of the Co-doped PbPdO₂ films calcined for different times. Clearly, the film with a calcination time of 0.5 h, as shown in Fig. 2(a), is composed of nanograins, and is relatively dense without remarkable defects. When the calcination time increases to 1.5 h and 2.5 h, the surfaces of the films are also dense without obvious cracks, as shown in Fig. 2(b) and (c). After coating 14 times, the thicknesses of the three films are almost identical and can reach about 180 nm according to the inset cross-section graphics in Fig. 2(a)–(c). From the inset EDS spectra, the Pb, Pd, O and Co signals can be found easily. The Al signal due to the sapphire substrate can also be observed. According to the EDS analysis, the atomic ratios of Pb : Pd : Co are 0.873 : 0.898 : 0.102, 0.870 : 0.898 : 0.102 and 0.868 : 0.899 : 0.101 for the Co-doped PbPdO₂ films calcined for 0.5 h, 1.5 h and 2.5 h, respectively. This indicates that the levels of Co-doping in the films are close to each other and that many Pb vacancies, caused by the volatilization of Pb during the calcination process, exist in all Co-doped PbPdO₂ films.

Fig. 2 FE-SEM images of the surface and the cross-section (inset, upper right) morphology and EDS spectra (inset, lower right) of the Co-doped PbPdO₂ films calcined for (a) 0.5 h, (b) 1.5 h and (c) 2.5 h.

Fig. 3(a)–(c) shows the refinements of the XRD data within the 2θ range of 28.5–42.0° for the Co-doped PbPdO₂ films calcined for different times. The background curves were modeled using a function type of shifted Chebyshev, and the diffraction peak shapes were described by a pseudo-Voigt function.¹³ Obviously, the XRD data of the three films are basically consistent with the patterns indexed with an orthorhombic structure which belongs to the Imma space group. From the inset AFM images of the films’ surfaces illustrated in the upper right in Fig. 3(a)–(c), the films are composed of nanograins, and the average grain sizes are about 30 nm, 80 nm and 120 nm for the films calcined for 0.5 h, 1.5 h and 2.5 h, respectively. The grain growth can be understood by the Ostwald ripening mechanism during the calcination treatment.^14,15 Through extending the calcination time, more thermal energy and enough growth time could promote the grain growth. Based on the refinements of the main (211) peaks and Scherrer formula, the full widths at half maximum (FWHM) and the crystallization sizes for the films calcined for different times were calculated and are given in Fig. 3(d).¹⁶ For the calcination time of 0.5 h, the FWHM is 0.324° and the corresponding crystallization size is 25.19 nm. With increasing the calcination time to 2.5 h, the FWHM gradually decreases to 0.234° and the crystallization size increases to 34.92 nm. This tendency of the film’s crystallization size to increase is consistent with the grain sizes mentioned above.

Fig. 3 Refinements of the XRD data within the 2θ range of 28.5–42.0° and AFM surface images (inset, upper right) of the Co-doped PbPdO₂ films calcined for (a) 0.5 h, (b) 1.5 h and (c) 2.5 h. (d) FWHM and crystallization sizes of the Co-doped PbPdO₂ films calcined for different times.

Fig. 4(a) gives the XPS spectra of Pb in all films calcined for different times. The peaks at about 142.4 eV and 137.5 eV usually come from the excitation of Pb²⁺ 4f_5/2 and Pb²⁺ 4f_7/2, respectively. The XPS spectra of Pd in the three films are presented in Fig. 4(b). The peaks at around 342.5 eV and 337.3 eV can be attributed to Pd²⁺ 3d_3/2 and Pd²⁺ 3d_5/2, respectively. Clearly, the valence states of Pb and Pd in the Co-doped PbPdO₂ films are both near 2+.^17,18 Comparing the binding energies of Pb and Pd in the three films carefully, small binding energy deviations of about 0.02 eV in the Pb and Pd XPS spectra can be observed. This may be caused by the minute differences in the amount of Pb vacancies in the three films, or the measurement error due to the XPS spectrometer’s energy resolution of 0.45 eV. As for the Co regions shown in Fig. 4(c)–(e), the XPS curves of Co in all films were all fitted using XPSPEAK with a binding energy range of 775.0–805.0 eV. The background curves were modeled using a function type of Shirley. They all consist of Co 2p_1/2 with a binding energy of about 795.1 eV and a spin–orbit doublet with Co 2p_3/2 having a binding energy of about 779.4 eV. By performing peak fitting deconvolution, the Co 2p_3/2 spectra can be separated into two characteristic peaks at 780.9 eV and 779.3 eV which correspond to Co²⁺ and Co³⁺, respectively. Similarly, the Co 2p_1/2 spectra can be deconvoluted into two characteristic peaks at 796.4 eV and 794.6 eV which can be attributed to Co²⁺ and Co³⁺, respectively. Therefore, the valence state of Co within the Co-doped PbPdO₂ films is a mixed one of 2+ and 3+.^19,20 It is speculated that such a mixed valence state of Co ions may have some relation with the Pb vacancies, which have been proved to exist in the calcined Co-doped PbPdO₂ films using EDS analysis. Because the electronegativity of Co (1.88) is lower than those of Pd (2.20) and Pb (2.33), it is reasonable to deduce that the Co ions adjacent to Pb vacancies will provide the electrons to O ions which should have bonded with the volatilized Pb ions to achieve the charge balance. This results in the valence increase of Co ions.

Fig. 4 XPS spectra of (a) Pb and (b) Pd in the Co-doped PbPdO₂ films calcined for different times, and those of Co and their fitting curves for the films calcined for (c) 0.5 h, (d) 1.5 h and (e) 2.5 h.

The ZFC–FC M–T curves of the Co-doped PbPdO₂ films calcined for different times are displayed in Fig. 5. As the temperature decreases from 380 K to 4 K, the ZFC and FC curves of all samples gradually split, indicating that FM exists in all films. At the room temperature of 300 K, divergence between the ZFC and FC curves can also be seen clearly, and the difference ΔM between the FC and ZFC magnetizations, as shown in the insets of Fig. 5, is nonzero for all three films. This means that the films’ FM can be maintained at room temperature or even higher.^21,22 Within the low temperature region, the magnetizations of both the ZFC and FC curves for all the films increase rapidly as the temperature decreases, which indicates the existence of paramagnetism (PM).²² At a measuring temperature of about 45 K, as marked with the dotted lines in Fig. 5(a)–(c), small peaks can be observed in both the ZFC and FC curves, revealing the existence of AFM. In addition, a magnetization jump at about 175 K can also be observed in both the ZFC and FC curves of all three films. With a decrease in grain size, the jump is gradually weakened. This phenomenon may be related to CO.^12,23

Fig. 5 ZFC–FC M–T curves of the Co-doped PbPdO₂ films calcined for (a) 0.5 h, (b) 1.5 h and (c) 2.5 h. The insets show the temperature dependences of the difference ΔM between the FC and ZFC magnetizations.

Fig. 6(a)–(c) plots the M–H hysteresis loops of all three Co-doped PbPdO₂ films, measured at various temperatures. The loops, after subtracting the diamagnetism (DM) originating from the sapphire substrate, are given in Fig. 6(d–f). It is clear that FM exists in all films at a temperature range of 4–380 K. Based on the results shown in Fig. 6(d–f), the temperature dependence of the film’s saturation magnetization M_s of the FM is plotted in Fig. 7. The variation tendencies of M_s for the three films are similar, namely decreasing first and then increasing with the temperature. When the grain size is increased, this tendency of M_s to increase becomes enhanced, and the temperature at which the increase begins shifts to a lower temperature. The decrease of M_s with the temperature, caused by thermal disturbance, is normal for FM. The abnormal tendency for M_s to increase is proposed to be a peculiar feature of materials with a spin gapless band structure.⁶

Fig. 6 M–H loops without and with subtraction of the DM signal measured at 4 K, 30 K, 100 K, 200 K, 300 K and 380 K for the Co-doped PbPdO₂ films calcined for (a and d) 0.5 h, (b and e) 1.5 h and (c and f) 2.5 h.

Fig. 7 Temperature dependences of the saturation magnetization M_s of the Co-doped PbPdO₂ films calcined for 0.5 h, 1.5 h and 2.5 h.

For the magnetic origin of the Co-doped PbPdO₂ films discussed in this paper, the carrier-mediated mechanism bridged to the bound magnetic polaron (BMP) model can be adopted to explain the underlying mechanism.^21,24,25 It has been reported that Co-doped PbPdO₂ has a p-type conductivity with a hole concentration of about 10¹⁸ cm⁻³, which is usually located at the insulating regime.²⁵ The existence of Pb vacancies can increase the hole concentration. This may bring the conductivity of Co-doped PbPdO₂ into an intermediate regime in which the films’ magnetism is determined by both the BMP mechanism and the carrier-mediated interaction. For the Co ions around the Pb vacancies, they couple with each other ferromagnetically and form bound magnetic polarons. The overlap of two polarons may induce an indirect ferromagnetic carrier–ion–carrier interaction. This is the origin of the FM in Co-doped PbPdO₂ films. The PM of all films may come from free Co ions or isolated magnetic polarons that are not coupled with other polarons. Additionally, the weak AFM is likely caused by the carrier-mediated interaction between the isolated Co ions.²¹ With increasing grain size from 30 nm to 120 nm, FM, PM and AFM co-exist in the Co-doped PbPdO₂ films without obvious conversion. As for the CO signal, it may be determined by the surface state of the nanograins, and competes with FM coupling.¹² As the grain size decreases, the CO is gradually weakened. This could be the result of the enhancement of the surface disordered state effect. With decreasing grain size, the total surface area of the nanograins increases. The enhanced surface disordered state could break the surface electron state balance, subsequently resulting in the suppression of CO. Hence, it is hopeful to eliminate the CO signal by further decreasing the grain size. A detailed investigation on the relation between the grain size and CO needs more efforts in the future.

Experimental

1. Preparation of Co-doped PbPdO₂ films

In order to obtain nanograin films with different grain sizes, sol solutions were prepared first. 2.261 mmol Pd(NO₃)₂·2H₂O was dissolved in 3.5 mL nitric acid and other reagents, including 0.441 mmol Co(NO₃)₂·6H₂O, 12.647 mmol citric acid monohydrate and different amounts of Pb(NO₃)₂, were dissolved in 18 mL ethylene glycol. Then the solutions were mixed and magnetically stirred at 65 °C for 7 h to obtain a transparent sol solution. The Co-doped PbPdO₂ films were fabricated by using a spin-coating technique. Single-crystal (11̄02) R-plane sapphires were used as the substrates. Before the coating process, the substrates were cleaned with distilled water and ethanol several times. Then the sol solutions were spin-coated onto the substrates at 2000 rpm for 33 s and dried in air at 60 °C to obtain the precursor films. The films were then pre-calcined at 400 °C in air for 5 min. The procedures from the spin coating to the pre-calcination were repeated 14 times to ensure the sufficient film thickness. Finally, the films were calcined at 650 °C in air for 0.5 h, 1.5 h and 2.5 h to get different grain sizes. Here, it should be mentioned that Pb is easily volatilized during the heat treatment because of its high vapor pressure. Therefore, excessive amounts of Pb(NO₃)₂ were added into the solutions to compensate for the loss of Pb.⁶ For the films calcined for 0.5 h, 1.5 h and 2.5 h, Pb(NO₃)₂ amounts of 5.413 mmol, 6.880 mmol and 7.192 mmol were adopted in the sol solution preparation, respectively.

2. Characterization

The phase structure of Co-doped PbPdO₂ films was studied by X-ray diffraction (XRD) measurement using a Rigaku diffractometer with Cu Kα radiation (λ = 1.5406 Å). The Rietveld refinement of XRD data was performed using the GSAS + EXPGUI program. The morphology of the films was measured with a field-emission scanning electron microscope (FESEM, Hitachi SU8020). An Oxford Instrument INCA energy dispersive spectrometer (EDS) was used to determine the films’ composition. The surface morphology was analyzed using an atomic force microscope (AFM, Bruker Dimension Icon) in the contact mode. The element valence states of the films were probed using X-ray photoelectron spectroscopy (XPS) measurement, which were performed on a Thermo ESCALAB 250Xi spectrometer using Al Kα radiation as the excitation source with the energy resolution better than 0.45 eV at room temperature. The films’ surfaces were cleaned using an argon gas stream before taking the XPS measurements. The obtained binding energies were all charge-corrected with respect to the standard C 1s peak at 284.8 eV. The magnetic properties of the films were analyzed using a superconducting quantum interference device magnetometer (SQUID-VSM, Quantum Design). Zero-field-cooled (ZFC) and field-cooled (FC) magnetization versus temperature (M–T) curves were recorded at temperatures from 4 K to 380 K under an applied field of 100 Oe. The magnetic hysteresis (M–H) loops were measured at 4 K, 30 K, 100 K, 200 K, 300 K and 380 K in magnetic fields ranging from −70 to 70 kOe.

Conclusions

In conclusion, single-phase Co-doped PbPdO₂ films with different grain sizes from 30 nm to 120 nm were synthesized using a sol–gel spin-coating method. The EDS measurements indicate that many Pb vacancies exist in all films. According to the XPS analysis, the valence states of the Pb ions and Pd ions within the Co-doped PbPdO₂ films are both near 2+, while the doped Co ions have a mixed valence state between 2+ and 3+. The increase in Co valence is believed to be caused by the existence of the Pb vacancies and the low electronegativity of Co. As for the magnetic properties, FM, PM and AFM were found to coexist in all films. FM can be retained up to temperatures higher than the room temperature, and the temperature dependence of M_s for the film with a larger grain size exhibits enhanced spin-gapless-related behavior. The carrier-mediated mechanism bridged to the BMP model was used to explain the origins of these three types of magnetism. Besides these three regular types of magnetism, the CO signal was also found in all films and it gradually weakened with decreasing grain size. The enhancement of the surface disordered state effect was used to explain this phenomenon.

Acknowledgements

This work was supported by National Natural Science Foundation of China (No. 11274086), the National Science Council of Taiwan (No. NSC 100-2112-M-006-018-MY3) and the Open Research Fund of Jiangsu Provincial Key Laboratory for Nanotechnology, Nanjing University.

Notes and references

1. X. L. Wang, Phys. Rev. Lett., 2008, 100, 156404.
2. K. J. Lee, S. M. Choo, J. B. Yoon, K. M. Song, Y. Saiga, C.-Y. You, N. Hur, S. I. Lee, T. Takabatake and M. H. Jung, J. Appl. Phys., 2010, 107, 09C306.
3. T. C. Ozawa, T. Taniguchi, Y. Nagata, Y. Noro, T. Naka and A. Matsushita, J. Alloys Compd., 2005, 395, 32–35.
4. X. L. Wang, G. Peleckis, C. Zhang, H. Kimura and S. X. Dou, Adv. Mater., 2009, 21, 2196–2199.
5. K. J. Lee, S. M. Choo, Y. Saiga, T. Takabatake and M. H. Jung, J. Appl. Phys., 2011, 109, 07C316.
6. H. L. Su, S. Y. Huang, Y. F. Chiang, J. C. A. Huang, C. C. Kuo, Y. W. Du, Y. C. Wu and R. Z. Zuo, Appl. Phys. Lett., 2011, 99, 102508.
7. K. J. Lee, S. M. Choo and M. H. Jung, Appl. Phys. Lett., 2015, 106, 072406.
8. D. L. Leslie-Pelecky and R. D. Rieke, Chem. Mater., 1996, 8, 1770–1783.
9. X. G. Zheng, C. N. Xu, K. Nishikubo, K. Nishiyama, W. Higemoto, W. J. Moon, E. Tanaka and E. S. Otabe, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 014464.
10. X. H. Chen, H. T. Zhang, C. H. Wang, X. G. Luo and P. H. Li, Appl. Phys. Lett., 2002, 81, 4419.
11. V. L. Kirillov, D. A. Balaev, S. V. Semenov, K. A. Shaikhutdinov and O. N. Martyanov, Mater. Chem. Phys., 2014, 145, 75–81.
12. P. Chai, X. Y. Wang, S. Hu, X. J. Liu, Y. Liu, M. F. Lv, G. S. Li and J. Meng, J. Phys. Chem. C, 2009, 113, 15817–15823.
13. C. G. Yao, F. Z. Meng, X. J. Liu, L. Han, J. L. Meng and Q. S. Liang, Ceram. Int., 2014, 40, 13339–13346.
14. R. L. Zong, X. L. Wang, S. K. Shi and Y. F. Zhu, Phys. Chem. Chem. Phys., 2014, 16, 4236.
15. C. C. Yec and H. C. Zeng, J. Mater. Chem. A, 2014, 2, 4843–4851.
16. R. Tholkappiyan and K. Vishista, Appl. Surf. Sci., 2015, 351, 1016–1024.
17. J. Kim, S. Hwang, C. Lee and H. Kim, Met. Mater. Int., 2009, 15, 857–862.
18. A. G. M. da Silva, H. V. Fajardo, R. Balzer, L. F. D. Probst, A. S. P. Lovón, J. J. Lovón-Quintana, G. P. Valença, W. H. Schreine and P. A. Robles-Dutenhefner, J. Power Sources, 2015, 285, 460–468.
19. K. Rida, A. L. Cámara, M. A. Peña, C. L. Bolívar-Díaz and A. Martínez-Arias, Int. J. Hydrogen Energy, 2015, 40, 11267–11278.
20. J. E. Ghoul, M. Kraini, O. M. Lemine and L. E. Mir, J. Mater. Sci.: Mater. Electron., 2015, 26, 2614–2621.
21. F. L. Tang, H. L. Su, P. Y. Chuang, Y. C. Wu, J. C. A. Huang, X. L. Huang and Y. Jin, RSC Adv., 2014, 4, 49308.
22. N. Akdogan, H. Zabel, A. Nefedov, K. Westerholt, H. W. Becker, S. Gök, R. Khaibullin and L. Tagirov, J. Appl. Phys., 2009, 105, 043907.
23. Y. Liu, H. Kong and C. F. Zhu, J. Alloys Compd., 2007, 439, 33–36.
24. J. M. D. Coey, M. Venkatesan and C. B. Fitzgerald, Nat. Mater., 2005, 4, 173–179.
25. Z. L. Lu, H. S. Hsu, Y. H. Tzeng, F. M. Zhang, Y. W. Du and J. C. A. Huang, Appl. Phys. Lett., 2009, 95, 102501.

---