Cite this Article
Djabri I, Rezkallah T, Chemam F. Structural, electronic, optical, and magnetic properties of Co-doped Cu2O. Chinese Physics B, 2017, 26(2): 027102  
Permissions
Structural, electronic, optical, and magnetic properties of Co-doped Cu2O
Djabri I, Rezkallah T, Chemam F †
Laboratoire de la Physique Appliquée et Théorique, Département des Sciences de la matière, Université Laarbi tebessi, tebessa, Algérie

 

† Corresponding author. E-mail: †fchemam@gmail.com

Abstract

We investigate the magnetic properties of Co-doped Cu2O. We studied first the electronic and structural properties of Cu2O using the optimization of the lattice constant which is 4.18 Å. The calculated gap is found between 0.825 eV and 1.5 eV, these values are in good agreement with the experimental results. The Co atoms are inserted in Cu2O by means of the density functional theory (DFT) using LSDA, LSDA+U, and LSDA+MBJ approximations in the WIEN2k code, based on the supercell model by setting up 12, 24, and 48 atoms in (1 × 1 × 2), (1 × 2 × 2), and (2 × 2 × 2) supercells respectively with one or two copper atoms being replaced by cobalt atoms. The energy difference between the ferromagnetic and anti-ferromagnetic coupling of the spins located on the substitute Co has been calculated in order to obtain better insight into the magnetic exchange coupling for this particular compound. The studied compound exhibits stable integer magnetic moments of 2 µB and 4 µB when it is doped with 2 atoms of Co. Optical properties have also been worked out. The results obtained in this study demonstrate the importance of the magnetic effect in Cu2O.

PACS: 71.15.Mb;71.20.–b;75.50.Pp;78.66.–w
Keyword:Cu2O;electronic structure;magnetic properties;optical properties
1. Introduction

Recently, much effort has been put into exploring new semiconductors which can be suitable as hosts for a new DMS system because of the importance in their potential applications in spintronic devices and also for their importance in basic magnetism.

Cuprous oxide Cu2O as p-type oxide-based diluted magnetic semiconductors (DMSs) is attracting increasing attention because of its advantages, such as simple structure and composition, no toxicity and low production cost. Its energy gap is 2.1 eV,[1,2] therefore it has been used in the field of solar cell,[3–6] photocatalytic water splitting[7–9] or photo degradation of pollution,[10,11] spin transistors,[12–15] and DMS.[16] Cu2O has four different states depending on the various combinations of valence band holes and conduction band electrons.

Due to the important effect of the magnetic atom in the semiconductor, transition metals like Mn, Fe, Ni, and Co are doped into semiconducting oxide hosts such as Cu2O, TiO2, and ZnO to form diluted semi conducting material.[17] Large studies reported that the doping could improve the electronic, magnetic, and optical properties,[18,19] like Mn-, Fe-, Co-, and Ni-doped Cu2O by Sieberer,[20] Mn-doped Cu2O epitaxial films reported by Liu et al. [21,22] and Co-doped Cu2O by Antony et al. [23] Kale et al. also found that Co-doped Cu2O films could be ferromagnetic by co-doping with Al.[24] On the other hand, there are still some reports indicating that no sign of ferromagnetism is seen in Mn-doped Cu2O films.[25]

In this paper, we did a systematical study on the structural and electronic properties of Cu2O, and the main motivation of our work is to explore the possibility of inducing ferromagnetism in cuprous oxide by doping a few percent of cobalt atoms to understand and practice the spintronic or the electron spin. The other aspect is to investigate the optical properties such as reflectivity and refraction index of various compounds which is not studied as much as it should be despite the great importance in applications of optoelectronic materials.

All the properties are calculated theoretically by using the full potential linearized augmented plane wave (FL PLAW) within the density functional theory (DFT) with the local spin density approximation (LSDA), (LSDA+U) methods within density functional theory, we explicitly include the on-site Coulomb interaction term in the Hamiltonian, which describes strongly correlated systems which contain transition metal (3d electron) and LSDA+mBJ methods to use the modified Becke–Johnson potential (mBJ) to allow the calculation of band gaps with accuracy. Many useful results have been obtained and compared with other calculated or experimental ones. Our investigated results may offer theoretical data for the applications of Co-doped Cu2O in magnetic or optoelectronic sides.

2. Computational details

Cu2O crystallizes in a cubic structure with a lattice constant of a = 4.2696 Å and space group of Pn3N, its unit cell contains six atoms. The four copper atoms are situated in a face-centered cubic lattice and the two oxygen atoms are tetrahedral sites forming a body-centered cubic sub-lattice. The constants of lattice are a = b = c, and α = β = δ = 90°.[26]

Fig. 1. (color online) The 1 × 2 × 2 supercell of Cu2O in cubic structure.

All the calculations are based on the supercell model. We chose to study (1 × 1 × 2), (1 × 2 × 2), and (2 × 2 × 2) super-cells which contain 12, 24, and 48 atoms, using mesh number of 100. The calculations were performed with wien2k code,[27] which is based on the density functional theory (DFT) with the LSDA, LSDA+U, and LSDA+mBJ approaches.

For the magnetic properties two kinds of doping concentrations are created by replacing one or two copper atoms with transition metal atoms. Two special arrangements are considered with the first transition metal atom being fixed at the site (0 0 0) and the second atom being put at the first or the second neighbor site.

3. Results and discussion

The optimized lattice constant of a pure Cu2O and Codoped Cu2O is 4.18 Å, this result is in reasonable agreement with the experimental value.[28] We found almost the same in both cases because the Co radius is very close to the Cu one.

3.1. Electronic properties

In order to analyze the magnetic and the optical properties of Co-doped Cu2O, we first investigate its electronic properties based on the optimization. The energy band structure and total density of states (DOS) of pure and Co-doped Cu2O are shown in Figs. 2(a), 2(b), and 2(c). We chose to study (1 × 2 × 2) superlattice model. The top of the valence band (VB) and the bottom of the conduction band (CB) are composed of Cu 3d and O 2p states respectively.

Fig. 2. The 1 × 1 × 1 and 1 × 2 × 2 total densities of states and band structure of pure Cu2O [(a) and (b)], Co-doped Cu2O (c).

We calculated the gap energy of a pure Cu2O with LSDA and LSDA+mBJ approximations based on the density functional theory in order to find the best approximation for our study. The results are listed in Table 1.

Table 1.

Values of calculated gap energy in Cu2O obtained at the optimized lattice parameter a = 4.18 Å.

.

In all cases, both LSDA and LSDA+mBJ calculations give values in good agreement with the experimental ones. So we chose the LSDA approximation for the rest of our work.

The total densities of states of Cu4O2, Cu16O8, and Cu15Co1O8 are shown in Figs. 2(a), 2(b), and 2(c). In comparison with experimental data,[20,30,31] the total densities of states of all are made up of two groups of phonon states (low-and high-frequency states) separated by a gap. There is a distribution between −8 eV and 0 eV caused by the Co 4d states. The Co 4d and Cu 3d form a new orbital with O 2p which drives to a new distribution of energy of Co-doped Cu2O and the conduction band moves to the Fermi level.[20]

The band structure of both pure and Co-doped Cu2O present a direct band gap but for Co-doped Cu2O our results underestimate the real band gap of Cu2O. The reason for the under-estimated band gap is the exchange–correlation energy; Co is a transition metal, the spin exchange between local spin states and the sp band result in the valence and conduction bands shifting to a low energy zone, and the shift of valence band is larger than conduction band,[32] so when the interaction between 4d of cobalt and 2p of oxygen increases, the band gap decreases as shown in Table 2.

Table 2.

Values of calculated gap energy in Co-doped Cu2O obtained at the optimized lattice parameter a = 4.18 Å.

.
3.2. Magnetic properties

As pointed out in the introduction, the idea developed in this work is to dope Cu2O with impurities to change their properties. This approach can be followed to introduce magnetic elements into non magnetic semiconductors to make them magnetic. In this part we investigate the effect of the magnetic impurity in the Cu2O structure. A study of Co inserted in Cu2O with one or two copper atoms being replaced has been carried out but in this study the LSDA+U approximation was the best. The Hubbard U used in the magnetic properties of doped superlattices with value of U eff = U − J = 8.5 eV was used too.[33] The results were calculated using 4% and 8% of cobalt concentration. The compound considered in this study exhibits a stable integer magnetic moment of 2 µB,[20] the results are given in Table 3.

Table 3.

Values of total and partial magnetic moment of Co-doped Cu2O.

.

In order to estimate the stability of the ferromagnetic (FM) against an anti-ferromagnetic (AFM) alignment of the spins, the energy difference between a ferromagnetic and an anti-ferromagnetic spin arrangement has been calculated for two Co atoms in Co2Cu14O8. One Co atom was put at the origin (0 0 0), and the other one was assumed to occupy the (1/2, 1/2, 1/2) site. The ground state magnetic properties have been calculated using the total energy difference between FM and AFM ground state (ΔE = E AFM − E FM). As shown in Table 4 the AFM state is lower than that of the FM state, so that indicates that the compound is more stable with FM alignment.[20,34]

Table 4.

The magnetic coupling modes of two atoms of Co-doped Cu2O in unit Rydberg.

.
3.3. Optical properties

The optical properties are studied through the dielectric function, refractive, index, reflectivity, and energy loss function. At low energy and considering only the electric dipole approximation, the total dielectric function is:

(1)
where ε 1 is the real part of the complex dielectric function while ε 2 corresponds to the imaginary part. The imaginary part ε 2 (ω) can be calculated from the momentum matrix element between the occupied and unoccupied wave functions within the selection rules while the real part ε 1 (ω) of the complex dielectric function can be obtained using the Kramers–Kronig relationship.[35,36] The imaginary and real parts of the dielectric function ε (ω) are given respectively[36–40]
(2)
(3)

All other optical constants such as the refractive index n(ω), reflectivity R(ω), and energy loss L(ω) can be found using ε 1 (ω), and ε 2 (ω), where[36–40]

(4)
(5)
(6)

The optical constants are calculated to study the influence of cobalt impurity on the optical properties of Cu2O for (1 × 2 × 2) supercell. The calculated real and imaginary parts of the dielectric function of pure and Co-doped Cu2O are displayed in Fig. 3.

Fig. 3. Real ε 1(ω) and imaginary ε 2(ω) parts of the dielectric function for (a) pure Cu2O and (b) Co-doped Cu2O.

The curve of the imaginary part ε 2(ω) for pure Cu2O consists of two peaks: A (11.58, 3.5 eV) corresponds to the transition from Cu 3d states to the O 2p states at Γ point while the B (2, 11.85 eV) peak derives from transition from Cu 4s to O 2p states.[32] The A peak also appears in the curve of ε 2(ω) for Co-doped Cu2O but with weaker intensities A ′ (7.94, 3.44 eV) and B ′ (1.41, 11.93 eV) vanishes due to the split of energy levels after the Co doping. The valence bands shifts to the Fermi level and the transitions are forbidden at higher energy level.

In the curve of ε 1(ω), the static dielectric constants at zero frequency limit of Cu2O and Co-doped Cu2O are 9.67 and 8.22 respectively. A negative distribution appears in (4 eV) and in (4.85 eV) respectively. It indicates that both Cu2O and Co-doped Cu2O show a metallic behavior in this frequency region.

The calculated values of the energy loss function, the refractive index, and the reflectivity of Cu2O and Co-doped Cu2O are shown in Figs. 4(a), 4(b), 4(c) and Figs. 4(a′), 4(b′), 4(c′) respectively. The loss function describes a fast electron traveling in a material; the higher value in L(ω) spectra represents the characteristic associated with plasma resonance.[40] The resonant energy loss are located at about (9.80 eV) for Cu2O and about (11.76 eV) for Co-doped Cu2O. In addition the static refractive index n(0) of pure Cu2O is 2.5, while for Co-doped Cu2O it is higher (n(0) = 3.8). This increase is caused by the doping impurity.

Fig. 4. The calculated electron energy loss spectrum L(ω) [(4a) and (4a′)], refractive index [(4b) and (4b′)], and the optical reflectivity n(ω) [(4c) and (4c′)] of Cu16O8 and Cu15Co1O8.
4. Conclusion

In this study, the LSDA, LSDA+U, and LSDA+mBJ have been performed in order to analyze the electronic and magnetic properties of pure and Co-doped Cu2O supercell. The results obtained indicate that the calculated electronic structure and magnetic properties are in good agreement with the experiments ones. The band gap is equal 1.9 eV for pure Cu2O and 0.3 eV for Co-doped Cu2O which are comparable with other experimental and calculated results. For the magnetic properties the substance shows a stable ferromagnetism with a total magnetic moment of 2 µB. Also the dielectric function and optical constants are calculated to study the influence of Co impurity on the optical properties of our compound and the results agree well with the experimental data.

Acknowledgment

We are grateful to Prof. A Layadi from Setif University for his suggestions and discussions.

Reference
[1] Guangwei Y Xiaobo H Duo L Daliang S Jing L Huaijin Z Hong L Jiyang W 2010 J. Electroanal. Chem. 638 225
[2] Asar A Namdeo S Gajbhiye J 2010 Solid State Chem. 183 3100
[3] Ogwu A A Bouquerel E Ademosu O Moh S Crossan E Placido F 2005 Physica D 38 266
[4] Mittiga A Salza E Sarto F Tucci M Vasanthi R 2006 Appl. Phys. Lett. 88 163502
[5] Wei H M Gong H B Chen L Zi M Cao B Q 2012 Phys. Chem. C 116 10510
[6] Malerba C Azanza Ricardo C L D’Incau M Biccari F Scardi P Mittiga A 2012 Sol. Energy Mater. Sol. Cells 105 192
[7] Hu C C Nian J N Teng H 2008 Sol. Energy Mater. Sol. Cells 92 1071
[8] De Jongh P E Vanmaekelbergh D Kelly J 1999 Chem. Commun. 10 69
[9] Hara M Kondo T Komoda M Ikeda S Kondo J Domen K Shinohara K Tanaka A 1998 Chem. Commun. 357
[10] Zheng Z Huang B Wang Z Guo M Qin X Zhang X Wang P Dai Y 2009 Phys. Chem. C 113 14448
[11] Huang W C Lyu L M Yang Y C Huang M H 2012 Am. Chem. Soc. 134 1261
[12] Han X F Han K H Tao M 2009 Electrochem. Solid State Lett. 12 H89
[13] Wolf S A Awschalom D D Buhrman R A Daughton J M Molnár S V Roukes M L Chtchelkanova A Y Treger D M 2001 Science 294 1488
[14] Herng T S Lau S P Yu S F Yang H Y Ji X H Chen J S Yasui N Inaba H 2006 Appl. Phys. 99 086101
[15] Sato K Katayama-Yoshida H 2002 Semicond. Sci. Technol. 17 367
[16] Delatorre R G Munford M L Zandonay R Zoldan V C Pasa A A Schwarzacher W Meruvia M S Hümmelgen I A 2006 Appl. Phys. Lett. 88 3504
[17] Brandt I S Lima E Tumelero M A Acuna J J S Viegas A D C Zysler R D Pasa A A 2011 IEEE Trans. Mag. 47 2640
[18] Zhao Z Y Yi J Zhou D C 2014 Chin. Phys. B 23 017401
[19] Chieh-Cheng Ch Hsuan-Chung W 2016 Materials 9 164
[20] Sieberer M Redinger J Mohn P 2007 Phys. Rev. B 75 035203
[21] Wei M Braddon N Zhi D Midgley P A Chen S K Blamire M G MacManus-Driscoll J L 2005 Appl. Phys. Lett. 86 072514
[22] Lui Y L Harrington S Yates K A Wei M Blamire M G MacManus-Driscoll J L Liu Y C 2005 Appl. Phys. Lett. 87 222108
[23] Antony J Qiang Y Faheem M Meyer D 2007 Appl. Phys. Lett. 90 013106
[24] Kale S N Ogale S B Shinde S R Sahasrabuddhe M Kulkarni V N Greene R L Venkatesan T 2003 Appl. Phys. Lett. 82 2100
[25] Pan L Q Zhu H Fan C F Wang W G Zhang Y Xiao J Q 2005 Appl. Phys. 97 10D318
[26] Kirfel A Eichhorn K D 1990 Acta Crystal. A 46 271
[27] Blaha P Schwarz K Madsen G K H Kvasnicka D Luitz J 2001 WIEN2K, “an Augmented-Plane-Wave +Local Orbitals Program for Calculating Crystal Properties” Schwarz K Wien Technische Wien
[28] Katayama J Ito K Matsuoka M Tamaki J 2004 Appl. Electrochem. 34 687
[29] Meyer B K Polity A Reppin D Becke M Hering P Kramm B Klar P J Sander T Reindl C Heiliger C Heinemann M Müller C Ronning C 2013 Semiconductors and Semimetals 88 Elsevier Inc. 201
[30] Ghisjen J Tjeng L H Elp J Eskes H Westerink J Sawatsky G A Czyzyk M T 1988 Phys. Rev. B 38 11322
[31] Ghijsen J Tjeng L H Eskes H Sawatsky G A Johnson R L 1990 Phys. Rev. B 42 2268
[32] Jiang Z Q Yao G An X Y Fu Y J Cao L H Wu W D Wang X M 2014 Chin. Phys. B 23 057104
[33] Xinguo M Ying W Yanhui L Yongfa Zh 2013 Phys. Chem. C 117 26029
[34] Li L Cheng Y Wang W Ren Sh Yang Y Luo X Liu H 2011 Solid State Commun. 151 1583
[35] Yang Z J Guo Y D Li J Liu J C Dai W Cheng X L Yang X D 2010 Chin. Phys. B 19 077102
[36] Okoye C M I 2003 J. Phys.: Condens. Matter 15 5945
[37] Zhang X D Guo M L Liu C L Zhang L A Zhang W Y Ding Y Q Wu Q Feng X 2008 Eur. Phys. B 62 417
[38] Yao G Chen Y An X Y Jiang Z Q Cao L H Wu W D Zhao Y 2013 Chin. Phys. Lett. 30 067101
[39] Cai M Q Yin Z Zhang M S 2003 Appl. Phys. Lett. 83 2805
[40] Anderson O L 1963 Phys. Chem. Sol. 24 909