Skip to main content

Ab initio design of nanostructures for solar energy conversion: a case study on silicon nitride nanowire


Design of novel materials for efficient solar energy conversion is critical to the development of green energy technology. In this work, we present a first-principles study on the design of nanostructures for solar energy harvesting on the basis of the density functional theory. We show that the indirect band structure of bulk silicon nitride is transferred to direct bandgap in nanowire. We find that intermediate bands can be created by doping, leading to enhancement of sunlight absorption. We further show that codoping not only reduces the bandgap and introduces intermediate bands but also enhances the solubility of dopants in silicon nitride nanowires due to reduced formation energy of substitution. Importantly, the codoped nanowire is ferromagnetic, leading to the improvement of carrier mobility. The silicon nitride nanowires with direct bandgap, intermediate bands, and ferromagnetism may be applicable to solar energy harvesting.


As one of the green energy sources, solar energy has been widely investigated to replace the old forms of depletable energy, such as coal and oil, which are limited on earth and detrimental to global climate. It needs, therefore, to develop reliable technologies to efficiently convert solar energy to other usable energy forms, such as electricity and chemical energy. A few technologies have been developed to harvest solar energy, including photovoltaic cells (PV; converting solar energy to electrical energy), photoelectrochemical cells (PEC; converting solar energy to chemical energy), and solar thermal systems (converting solar energy to thermal energy). In all of these technologies, the fundamental element, materials, plays a dominant role to maximally utilize the sunlight. For photovoltaic cells, the optimum bandgap for the solar cell material is a compromise between a bandgap wide enough so that not too many electrons are wasted and yet narrow enough so that enough photons can be absorbed to create electron–hole pairs[1]. For photoelectrochemical cells, the development of an efficient photocatalyst for water splitting requires (a) narrowing its bandgap that satisfies the visible light absorption and the band edge requirement of H2/H2O and O2/H2O levels[2, 3] and (b) having high contacting surface area with the electrolyte to enhance the reaction and to increase the light absorption[4]. However, no semiconductor has a bandgap that can utilize the entire spectral distribution of sunlight. To enhance the light absorption efficiency, considerable effort has been conducted for the maximal absorption of sunlight, such as chemical doping[28], dye sensitization[911], material design[1214], defect engineering[1517], and structure engineering[1820]. However, novel cell concepts are necessary for a huge increase in the efficiency. One of the concepts, hot-carrier solar cell, is to use semiconductor nanocrystals or quantum dots to capture all of the energy of hot carriers[21, 22], where hot-carrier relaxation is only possible via slower multiphonon emission because of the quantum confinement-induced discretized band states in the nanostructures[23]. Another important concept is the intermediate band (IB) solar cell, consisting of an IB material situated between two conventional semiconductors, n- and p-types[2426], where the IB material has a band inside the bandgap. A full electron transition from the valence band to the conduction band can be completed by means of two photons with energy below the bandgap, resulting in the increase in photocurrent. The IB can arise from the quantum confinement effects in quantum dots[2729] or impurity states by doping bulk materials with a transition metal[3033]. Recently, solar cells based on nanostructures have attracted considerable attention because of possible cost reduction and efficiency improvement[3436]. Therefore, nanostructures possessing an intermediate band and quantum confinement effect may be able to enhance the efficiency and reduce the cost at the same time.

Silicon nitride (Si3N4) is a material of great technological interest in a number of applications, such as high-temperature electronics, because of its chemical inertness, high dielectric constant, large electronic gap, high resistance against radiation, and strong resistance against thermal shock[37]. Importantly, Si3N4 is a well-known antireflection coating material in the semiconductor industry to reduce the light reflection in Si-based solar cells[38]. Also, single-crystal Si3N4 nanowires on a Si substrate can be easily synthesized by chemical vapor deposition[39]. It is expected that an energy-harvesting cell based on Si3N4 nanowires and the present Si technology may make the Si3N4-based energy harvesting cell possibly produced on sustainable improved efficiency and cost reduction because of the easy integration of Si3N4 into the Si technology. In this work, we explore the electronic, magnetic, and optical properties of a Si3N4 nanowire for its possible application in solar energy conversion based on first-principles calculation. Our calculations predict that the Si3N4 nanowire is a direct-band semiconductor with reduced bandgap, and IBs can be created by doping with carbon and transition metals. We further show that anion-cation codoping can improve the solubility of a transition metal in Si3N4 and its crystallinity and enhance the magnetic moment. We further predict that the ferromagnetic Si3N4 nanowire with IB is more efficient for solar energy conversion.


To investigate the electronic, optical, and magnetic properties of the Si3N4 nanowire, first-principles calculations are carried out based on the density functional theory (DFT) and the Perdew-Burke-Ernzerhof generalized gradient approximation (PBE-GGA)[4042]. The Vienna Ab initio Simulation Package (VASP) incorporated with the projector augmented wave (PAW) scheme is used[43, 44]. The nanowire is created along the β-Si3N4 [001] direction (Figure 1). An energy cutoff of 450 eV is used for the plane wave expansion of the electronic wave function. The Monkhorst-Pack scheme is used to generate special k points with a 1 × 1 × 3 grid[45]. Good convergence is obtained with these parameters. The total energy is converged to 2.0 × 10-5 eV/atom while the Hellmann-Feynman force is smaller than 5.0 × 10-2 eV/Å in the optimized structure. A large hexagonal supercell with a shortest wall-wall distance of 10 Å in the plane perpendicular to the wire axis and four unit cells parallel to the wire axis is used to avoid interactions between the wire and its images in neighboring cells and between dopants. The nanowire has 96 nitrogen and 72 silicon atoms.

Figure 1
figure 1

Schematic drawings of the atomic structures of β-Si 3 N 4 bulk (a) and nanowire: (b) top and (c) side views. The red circle in (a) indicates the model of the nanowire in (b) and (c) created from the bulk. The yellow, blue, and white dots are Si, N, and H, respectively.

To study the optical property of the nanowire, we calculate the loss function, which is a direct probe of the collective excitation of the system under consideration. The imaginary part of the dielectric constant is calculated from:

ϵ 2 q O u , ℏω = 2 e 2 π Ω ϵ 0 k , v , c | < Ψ k c | u r | Ψ k v > | 2 δ E k c - E k v - E

where u is the vector defining the polarization of the incident electric field. This expression is similar to Fermi's golden rule for time-dependent perturbations, and ϵ2(ω) can be thought of as detailing the real transitions between occupied and unoccupied electronic states. The real part, ϵ1(ω), is obtained by the Kramers-Kronig relation. The loss function is calculated using Im(-1/ϵ(ω)) at zero momentum transfer from the macroscopic dielectric function ϵ(ω) (ϵ(ω) = ϵ1(ω) + 2(ω)) for a periodic system.

Results and discussion

β-Si3N4 bulk

The periodic unit cell of β-Si3N4 is optimized first to obtain the lattice parameters. The lattice constants of the optimized structure (a = 7.613 Å, c = 2.910 Å) within PBE-GGA are in good agreement with the experimental values (a = 7.608 Å, c = 2.909 Å)[37]. β-Si3N4 is an indirect-bandgap semiconductor with the valence band top (VBT) at one third along the Γ-A axis and the conduction band bottom (CBB) at the Γ point (Figure 2a). The calculated bandgap is 4.25 eV within PBE-GGA (Figure 2), which is consistent with the reported data (4.2 eV)[46], but less than the experimental value (5.3 eV) due to the underestimation of the bandgap by density functional theory. The analysis on the partial density of states (PDOS) reveals that the VBT states are mainly dominated by N_p electrons, while Si_p electrons mainly contribute to the CBB states (Figure 2b).

Figure 2
figure 2

Calculated band structure (a) and partial density of states (b) of β-Si 3 N 4 bulk.

β-Si3N4 nanowire

The β-Si3N4 nanowire is modeled by cutting the bulk supercell along the [001] direction (Figure 1a). The surface atoms are passivated by hydrogen atoms (totally 48 H atoms in the supercell) (Figure 1b,c). The structure of the optimized Si3N4 nanowire almost keeps unchanged, except the hydrogen atoms at the surface (Figure 3). The Si-N bond length is expanded by less than 1%. The Si-H and N-H bond lengths are 1.489 and 1.019 Å, respectively.

Figure 3
figure 3

The optimized atomic structures of β-Si 3 N 4 nanowire (a) top and (b) side views.

The calculated electronic property of the nanowire shows a notable feature in its band structure (Figure 4). The nanowire is a direct-bandgap semiconductor with both the VBT and CBB at the Γ point. Compared with those in the band structure of β-Si3N4 bulk along the Γ-A axis, the valence band top states in that of the nanowire are pushed up, especially the states at the Γ point, leading to almost flat bands from the Γ point to one third of the Γ-A axis (VBT in the band structure of β-Si3N4 bulk) (Figures 2a and4a) and the occurrence of VBT at the Γ point. The change of the VBT states is attributed to the confinement of charge carriers which results in the quasi-continuous energy bands of the bulk semiconductors becoming discretized. The VBT states are still dominated by N_p electrons (Figure 4b). The CBB remains at the Γ point with the parts of the CBB states in the nanowire pulled down. In contrast to the β-Si3N4 bulk, the p electrons from both N and Si contribute to the CBB states in the nanowire (Figure 4b,c). The up-shift of VBT and down-shift of CBB lead to the reduction of bandgap (3.92 eV). The direct band structure in the nanowire may enhance the conversion efficiency due to the improvement of carrier transportation. However, the bandgap of the nanowire is still too larger to be efficient of the sunlight absorption. It is necessary to engineer the bandgap to improve the absorption.

Figure 4
figure 4

Calculated band structure (a) and PDOS for N (b) and Si (c) of β-Si 3 N 4 nanowire.

Single-element-doped β-Si3N4 nanowire

One of the important ways to engineer the bandgap of semiconductors is to create intermediate bands within the gap by doping. Three elements, oxygen, carbon, and chromium, are studied because the contamination of oxygen and carbon easily occurs in the growth of Si3N4 and chromium is the most possible element for the creation of IB in the bandgap[31, 32]. The doping is modeled by substituting a cation atom with a chromium atom or an anion atom with an oxygen or carbon atom inside the nanowire. As an indication of the doping possibility and stability, the doping formation energy is estimated from[47]:

E f = E tot NW + doping - E tot NW - μ dopant + μ host

where Etot(NW + doping) and Etot(NW) are the total energies of the β-Si3N4 nanowire (NW) with and without doping, respectively. μdopant and μhost are the energies of dopants (O, C, Cr, CrO, or CrC) and host atoms (Si, N, or SiN), respectively. μ O = 1 2 μ O 2 , μ N = 1 2 μ N 2 , μCr = μ(CrBulk), μSi = μ(SiBulk) and μC = μ(graphite). The formation energies of single-element doping (Table 1) indicate that the Cr atoms prefer to take the cation positions (Si sites), while the O and C atoms substitute the anion positions (N sites) more easily, because of the relatively lower formation energies. The formation energy for oxygen substituting nitrogen is only 0.75 eV, indicating that the doping of oxygen is much easier than that of other elements. The calculated bond lengths for the stable doping states are 1.799 Å for the O-Si bond, 1.826 Å for the C-Si bond, and 1.845 Å for the Cr-N bond in O-, C-, and Cr-doped nanowires, respectively, which are larger than the Si-N bond length (1.741 Å). The least change in the structure of the O-doped nanowire is consistent with lower doping energy.

Table 1 Formation energy for the various doping configurations in the β-Si 3 N 4 nanowire

The calculated band structure shows that the O-doped β-Si3N4 nanowire is an n-type semiconductor with a direct bandgap of 3.89 eV at the Γ point (Figure 5a). The PDOS analysis shows that the s and p electrons are also involved in the CBB states (Figure 6a). The band structure of the O-doped nanowire is similar to that of the undoped nanowire, except that the Fermi level is within the conduction band because of the contributed electrons from oxygen, indicating that O doping cannot efficiently narrow the bandgap of the nanowire. An intermediate level with less dispersion is formed by C doping, crossing the Fermi level (Figure 5b), which is dominated by the C_p electrons (Figure 6b). The gaps between the IB and VBT or CBB are 0.96 and 2.96 eV, respectively. Three IBs are formed by Cr doping, which separate the bandgap into four regions with the gaps of 1.68, 0.92, 0.42, and 0.88 eV from the VBT to CBB, respectively (Figure 5c). The impurity states are mainly attributed to the Cr_d electrons (Figure 6c). The Cr_d electrons should form a quasi-continuous impurity band in β-Si3N4 bulk, which are discretized in the nanowire due to the quantum confinement effect. The creation of IB by doping, especially Cr doping, should improve the sunlight absorption of the nanowire.

Figure 5
figure 5

Calculated spin-unpolarized band structures of β-Si 3 N 4 nanowires doped with (a) O, (b) C, and (c) Cr.

Figure 6
figure 6

Calculated spin-unpolarized PDOS of β-Si 3 N 4 nanowires doped with (a) O, (b) C, and (c) Cr.

The calculated loss functions of the β-Si3N4 nanowires clearly demonstrate the improvement of the light absorption, especially the C- and Cr-doped nanowires (Figure 7). The intercept in the loss function is the optical gap and related to the inter-band excitation. Comparing the loss functions of β-Si3N4 bulk and nanowire (Figure 7a,b), we see that the light absorption in the nanowire is redshifted because of the reduced bandgap. For the O-doped nanowire, an absorption peak starting from 0 eV is attributed to the electron excitation near the Fermi level because the O-doped nanowire is an n-type semiconductor (Figure 5a). The loss function of the C-doped β-Si3N4 nanowire is similar to that of the undoped one, except for a weak peak at 2.1 eV (Figure 7c), which is related to the inter-band excitation from VBT to IB or from IB to CBB (Figure 5b). For the Cr-doped nanowire, the loss function clearly shows that the excitation starts at 2.3 eV. From the loss functions of the doped nanowires, we can see that the IBs within the gap improve the absorption of the sunlight because of the reduced inter-bandgap (Figure 5c).

Figure 7
figure 7

Calculated loss functions of β-Si 3 N 4 bulk (a), nanowire (b), and nanowires doped with (c) O, (d) C, and (e) Cr.

Anion-cation-codoped β-Si3N4 nanowire

The codoping is realized by simultaneously substituting a cation atom with a chromium atom and an anion atom with an oxygen or carbon atom as a pair inside the nanowire. The calculated formation energies are 1.51 and 6.06 eV for CrO and CrC codoping, respectively. The formation energy of CrO codoping is much lower than that of Cr doping because of the electrostatic attraction of the two dopants with opposite charge states[31, 32], which indicates that the Cr substitution can be greatly enhanced in the presence of O doping and results in the enhancement of solubility of Cr in the nanowire and improvement of crystallinity. However, the CrC codoping is unstable because of electrostatic repulsion between the two dopants, as indicated by the high formation energy, and will not be discussed. The Cr-N, Si-O, and Cr-O bond lengths in the CrO-codoped nanowire are 1.889, 1.703, and 1.996 Å, respectively.

The calculated electronic and optical properties of the CrO-codoped β-Si3N4 nanowire show that four IBs are observable in the bandgap (Figure 8a), which are dominated by the Cr_d electrons (Figure 8b). The space between the VBT and the first IB is about 2.1 eV. Compared with those in the Cr-doped nanowire, the Cr_d electrons in the CrO-codoped nanowire are more discretized due to the quantum confinement effect and the strong coupling between the Cr_d and O_p electrons (Figure 8b). A similar situation occurs at the conduction band bottom, where more states become degenerate. Similar to Cr-doped nanowire, CrO-codoped nanowire also shows a excitation about at 2.3 eV (Figure8c).

Figure 8
figure 8

Calculated spin-unpolarized band structure (a), PDOS (b), and loss function (c) of β-Si 3 N 4 nanowire codoped with CrO.

Magnetic properties of Cr-doped and CrO-codoped β-Si3N4 nanowires

To investigate the magnetic properties of the doped nanowires, spin-polarized calculations are performed. To study the magnetic coupling between metal dopants, two neighboring Si atoms in the nanowire are substituted by two Cr atoms (Cr2) for the Cr-doped nanowire, and two neighboring Si atoms and one N atom bonded with them are replaced by two Cr atoms and one O atom (Cr2O) for the codoped nanowire. The calculated exchange energies, defined by the energy difference between the antiferromagnetic and ferromagnetic states (Eexch = EAFM - EFM), are 135 and 293 meV for the Cr2-doped and Cr2O-codoped nanowires, respectively, indicating that the nanowires with doping are ferromagnetic. Importantly, the exchange energy of the codoped nanowire is much larger than that of the Cr-doped nanowire, leading to a much stable ferromagnetic state and higher transition Curie temperature of the codoped nanowire. The calculated magnetic moments per Cr atom are 1.85 and 2.29 μB for the Cr-doped and CrO-codoped nanowires, respectively. Clearly, the codoping enhances not only the exchange energy but also the magnetic moment.

The calculated unsymmetrical spin-up and spin-down total density of states (DOS) confirm the ferromagnetism of the Cr-doped (Figure 9a) and CrO-codoped (Figure 9b) β-Si3N4 nanowires. All of these spin-polarized electrons are within the spin-up bands. For the Cr-doped nanowire, several impurity bands are observable and the Fermi level is in the gap between two impurity bands in the spin-up band (Figure 9a). The impurity states form three bands in the spin-up band, and the Fermi level is within the second impurity band when Cr codoped with oxygen (Figure 9b), indicating stronger polarization and enhancement of the magnetic moment. The PDOS analysis shows that the impurity states are mainly attributed to the spin-polarized Cr_d electrons for the Cr-doped nanowire (Figure 10), and for the CrO-codoped nanowire, the spin-polarized Cr_d and O_p electrons dominate the impurity bands (Figure 11). The Cr_d electrons strongly hybridized with the O_p electrons in the codoped nanowire, resulting in the stabilization of parallel spin alignment. The spin-polarized states within the bandgap (Figure 9) reveal that double exchange, which is stabilized by carrier (electrons) mediation, is the dominant coupling mechanism for the ferromagnetism in the Cr-doped and CrO-codoped nanowires[4852]. That is, given the incomplete filling of bands, when the exchange splitting is bigger than the bandwidth, the band energy of the ferromagnetic state is lower than that of the antiferromagnetic state if a sufficient (usually rather small) number of holes (or electrons) exists[49]. The exchange splitting is bigger than the valence band (Figure 9), and there are a considerable number of carriers. We can therefore conclude that carrier-mediated double exchange is responsible for the observed ferromagnetism in the Cr-doped and CrO-codoped β-Si3N4 nanowires.

Figure 9
figure 9

Calculated spin-polarized total DOS of β-Si 3 N 4 nanowires doped with (a) Cr and (b) CrO.

Figure 10
figure 10

Calculated spin-polarized PDOS of β-Si 3 N 4 nanowire with Cr doping.

Figure 11
figure 11

Calculated spin-polarized PDOS of β-Si 3 N 4 nanowire with CrO codoping for (a) Cr and (b) O.

The calculated electronic structures (Figure 9) further reveal the enhancement of the sunlight absorption in the doped nanowires because of the formation of IBs within the bandgap. Importantly, the diluted magnetic semiconductors (DMSs) with IBs can have the desired optical properties and prevent radiative transition by taking the advantage of spin selection rules on IB transition[33]. The carrier mobility/lifetime is greatly enhanced because the spin degeneracy of the bands is lifted in DMSs and the unwanted recombinations are impeded by spin selection rules or by low occupancy of states involved in the allowed recombinations[53, 54]. All of these advantages lead to the enhancement of the conversion efficiency of magnetic β-Si3N4 nanowires.


In summary, a first-principles design of nanostructures is carried out to investigate their applications in solar energy harvesting. The calculated results show that the band structures of bulk materials can be engineered by reducing their size. Bulk Si3N4 is an indirect-bandgap semiconductor, while Si3N4 nanowire is a direct one. We show that the band structure of the nanowire can be further controlled by doping. Intermediate bands within its bandgap can be created by doping. The calculated optical property shows that the intermediate bands play an important role in the enhancement of visible light absorption. We also show that anion-cation codoping is easier than single-element doping because of the electrostatic attraction of the anion and cation. We further demonstrate that the ferromagnetic nanowire can be realized by codoping, where spin polarization can efficiently improve carrier mobility due to spin selection rules. The designed nanowire with a controllable band structure engineered by doping and size reduction shows efficient sunlight absorption and improved mobility and may find applications in solar energy harvesting.


  1. Moller HJ: Semiconductors for Solar Cells. Boston: Artech House; 1993.

    Google Scholar 

  2. Chen X, Mao SS: Titanium dioxide nanomaterials: synthesis, properties, modifications, and applications. Chem Rev 2007, 107: 2891–2959. 10.1021/cr0500535

    Article  Google Scholar 

  3. Lee ASH, Li K, Zhang YW, Pan H: A b initio study on the effects of dopant-defect cluster on the electronic properties of TiO2-based photocatalysts . Int J Hydro Energy 2014, 39: 2049–2055. 10.1016/j.ijhydene.2013.11.090

    Article  Google Scholar 

  4. Dou MF, Baldissera G, Persson C: ZnO–InN nanostructures with tailored photocatalytic properties for overall water-splitting. Int J Hydro Energy 2013, 38: 16727–16732. 10.1016/j.ijhydene.2013.02.071

    Article  Google Scholar 

  5. Asahi A, Morikawa T, Ohwaki T, Aoki K, Taga Y: Visible-light photocatalyst in nitrogen-doped titanium oxides. Science 2001, 293: 269–271. 10.1126/science.1061051

    Article  Google Scholar 

  6. Borgarello E, Kiwi J, Gratzel M, Pelizzetti E, Visca M: Visible light induced water cleavage in colloidal solutions of chromium-doped titanium dioxide particles. J Am Chem Soc 1982, 104: 2996–3002. 10.1021/ja00375a010

    Article  Google Scholar 

  7. Gao MQ, Xu YL, Bai Y, Xiao F: Nb, F-codoped TiO2 hollow spheres with high visible light photocatalytic activity. Nanoscale Res Lett 2013, 8: 508. 10.1186/1556-276X-8-508

    Article  Google Scholar 

  8. Tan LL, Ong WJ, Chai SP, Mohamed AR: Reduced graphene oxide-TiO2 nanocomposite as a promising visible-light-active photocatalyst for the conversion of carbon dioxide. Nanoscale Res Lett 2013, 8: 465. 10.1186/1556-276X-8-465

    Article  Google Scholar 

  9. Wang XL, Wu M, Ding J, Li ZS, Sun KN: Periodic density functional theory study on the interaction mode and mechanism of typical additives with TiO2 substrates for dye-sensitized solar cell applications. J Power Sources 2014, 246: 10–18.

    Article  Google Scholar 

  10. Lim J, Ryu SY, Kim J, Jun Y: A study of TiO2/carbon black composition as counter electrode materials for dye-sensitized solar cells. Nanoscale Res Lett 2013, 8: 227. 10.1186/1556-276X-8-227

    Article  Google Scholar 

  11. Luo Y, Li D, Meng Q: Towards optimization of materials for dye-sensitized solar cells. Adv Mater 2009, 21: 1–5.

    Google Scholar 

  12. Lim YK, Koh EWK, Zhang YW, Pan H: Ab initio design of GaN-based photocatalyst: ZnO-codoped GaN nanotubes. J Power Sources 2013, 232: 323–331.

    Article  Google Scholar 

  13. Punya A, Lambrecht WRL: Band offsets between ZnGeN2, GaN, ZnO, and ZnSnN2 and their potential impact for solar cells. Phys Rev B 2013, 88: 075302.

    Article  Google Scholar 

  14. Yang SY, Seidel J, Byrnes SJ, Shafer P, Yang CH, Rossell MD, Yu P, Chu YH, Scott JF, Ager JW, Martin LW, Ramesh R: Above-bandgap voltages from ferroelectric photovoltaic devices. Nat Nanotech 2010, 5: 143–147. 10.1038/nnano.2009.451

    Article  Google Scholar 

  15. Pan H, Guo B, Zhang ZY: Phase-dependent photocatalytic activity of TiO2: a first-principles study. J Chem Theory Comput 2009, 5: 3074–3078. 10.1021/ct9002724

    Article  Google Scholar 

  16. Nowotny MK, Sheppard LR, Bak T, Nowotny J: Defect chemistry of titanium dioxide. Application of defect engineering in processing of TiO2-based photocatalyst. J Phys Chem C 2008, 112: 5275–5300. 10.1021/jp077275m

    Article  Google Scholar 

  17. Pan H, Qiu X, Ivanov IN, Meyer HM, Wang W, Zhu WG, Paranthaman MP, Zhang ZY, Eres G, Gu B: Fabrication and characterization of brookite-rich, visible light-active TiO2 films for water splitting. Appl Catal B Environ 2009, 93: 90–95. 10.1016/j.apcatb.2009.09.016

    Article  Google Scholar 

  18. Sun B, Shi TL, Peng ZC, Sheng WJ, Jiang T, Liao GL: Controlled fabrication of Sn/TiO2 nanorods for photoelectrochemical water splitting. Nanoscale Res Lett 2013, 8: 462. 10.1186/1556-276X-8-462

    Article  Google Scholar 

  19. Pala RA, White J, Barnard E, Liu J: Design of plasmonic thin-film solar cells with broadband absorption enhancements. Adv Mater 2009, 21: 1–6.

    Article  Google Scholar 

  20. Sato H, Matsuda R, Sugimoto K, Takata MS: Photoactivation of a nanoporous crystal for on-demand guest trapping and conversion. Nat Mater 2010, 9: 661–666. 10.1038/nmat2808

    Article  Google Scholar 

  21. Tisdale WA, Williams KJ, Timp BA, Norris DJ, Aydil ES, Zhu ZY: Hot-electron transfer from semiconductor nanocrystals. Science 2010, 328: 1543–1547. 10.1126/science.1185509

    Article  Google Scholar 

  22. Alivisatos AP: Semiconductor clusters, nanocrystals, and quantum dots. Science 1996, 271: 933–937. 10.1126/science.271.5251.933

    Article  Google Scholar 

  23. Nozik AJ: Spectroscopy and hot electron relaxation dynamics in semiconductor quantum wells and quantum dots. Annu Rev Phys Chem 2001, 52: 193–231. 10.1146/annurev.physchem.52.1.193

    Article  Google Scholar 

  24. Luque A, Marti A: Increasing the efficiency of ideal solar cells by photon induced transitions at intermediate levels. Phys Rev Lett 1997, 78: 5014–5017. 10.1103/PhysRevLett.78.5014

    Article  Google Scholar 

  25. Luque A, Marti A: The intermediate band solar cell: progress toward the realization of an attractive concept. Adv Mater 2010, 22: 160–174. 10.1002/adma.200902388

    Article  Google Scholar 

  26. Wang WM, Lin AS, Phillips JD: Intermediate-band photovoltaic solar cell based on ZnTe:O. Appl Phys Lett 2009, 95: 011103. 10.1063/1.3166863

    Article  Google Scholar 

  27. Marti A, Lopez N, Antolin E, Canovas E, Luque A, Stanley CR, Farmer CD, Diaz P: Emitter degradation in quantum dot intermediate band solar cells. Appl Phys Lett 2007, 90: 233510. 10.1063/1.2747195

    Article  Google Scholar 

  28. Hubbard SM, Cress CD, Bailey CG, Raffaelle RP, Bailey SG, Wilt DM: Effect of strain compensation on quantum dot enhanced GaAs solar cells. Appl Phys Lett 2008, 92: 123512. 10.1063/1.2903699

    Article  Google Scholar 

  29. Oshima R, Takata A, Okada Y: Strain-compensated InAs/GaNAs quantum dots for use in high-efficiency solar cells. Appl Phys Lett 2008, 93: 083111. 10.1063/1.2973398

    Article  Google Scholar 

  30. Palacios P, Aguilera I, Sanchez K, Conesa JC, Wahnon P: Transition-metal-substituted indium thiospinels as novel intermediate-band materials: prediction and understanding of their electronic properties. Phys Rev Lett 2008, 101: 04640.3.

    Article  Google Scholar 

  31. Pan H, Gu B, Eres G, Zhang ZY: Ab initio study on non-compensated CrO codoping of GaN for enhanced solar energy conversion. J Chem Phys 2010, 132: 104501. 10.1063/1.3337919

    Article  Google Scholar 

  32. Zhu WG, Qiu X, Iancu V, Chen XQ, Pan H, Wang W, Dimitrijevic NM, Rajh T, III HM, Paranthaman MP, Stocks GM, Weitering H, Gu B, Eres G, Zhang ZY: Band gap narrowing of titanium oxide semiconductors by noncompensated anion-cation codoping for enhanced visible-light photoactivity. Phys Rev Lett 2009, 103: 226401.

    Article  Google Scholar 

  33. Olea J, Toledano-Luque M, Pastor D, Gonzalez-Diaz G, Martil I: Titanium doped silicon layers with very high concentration. J Appl Phys 2008, 104: 016105. 10.1063/1.2949258

    Article  Google Scholar 

  34. Tsakalakos L: Nanostructures for photovoltaics. Mater Sci Eng R 2008, 62: 75–189.

    Article  Google Scholar 

  35. Kelzenberg MD, Boettcher SW, Petykiewicz JA, Turner-Evans DB, Putnam MC, Warren EL, Spurgeon JM, Briggs RM, Lewis NS, Atwater HA: Enhanced absorption and carrier collection in Si wire arrays for photovoltaic applications. Nat Mater 2010, 9: 239–244.

    Article  Google Scholar 

  36. Chakrapani V, Thangala J, Sunkara MK: WO3 and W2N nanowire arrays for photoelectrochemical hydrogen production. Int J Hydro Energy 2009, 34: 9050–9059. 10.1016/j.ijhydene.2009.09.031

    Article  Google Scholar 

  37. Habraken FHPM, Kuiper AET: Silicon nitride and oxynitride films. Mat Sci Eng R 1994, 12: 123–175. 10.1016/0927-796X(94)90006-X

    Article  Google Scholar 

  38. El Amrani A, Menous I, Mahiou L, Tadjine R, Touati A, Lefgoum A: Silicon nitride film for solar cells. Renew Energy 2008, 33: 2289–2293. 10.1016/j.renene.2007.12.015

    Article  Google Scholar 

  39. Farjas J, Rath C, Pinyol A, Roura P, Bertran E: Si3N4 single-crystal nanowires grown from silicon micro- and nanoparticles near the threshold of passive oxidation. Appl Phys Lett 2005, 87: 192114. 10.1063/1.2130380

    Article  Google Scholar 

  40. Hohenberg P, Kohn W: Inhomogeneous electron gas. Phys Rev 1964, 136: B864-B871. 10.1103/PhysRev.136.B864

    Article  Google Scholar 

  41. Perdew JP, Burke K, Ernzerhof M: Generalized gradient approximation made simple. Phys Rev Lett 1996, 77: 3865–3868. 10.1103/PhysRevLett.77.3865

    Article  Google Scholar 

  42. Blöchl PE: Projector augmented-wave method. Phys Rev B 1994, 50: 17953–17979. 10.1103/PhysRevB.50.17953

    Article  Google Scholar 

  43. Kresse G, Furthmüller J: Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Rev B 1996, 54: 11169–11186. 10.1103/PhysRevB.54.11169

    Article  Google Scholar 

  44. Kresse G, Joubert D: From ultrasoft pseudopotentials to the projector augmented-wave method. Phys Rev B 1999, 59: 1758–1775.

    Article  Google Scholar 

  45. Monkhorst HJ, Pack J: Special points for Brillouin-zone integrations. Phys Rev B 1976, 23: 5188–5192.

    Article  Google Scholar 

  46. Liu AY, Cohen ML: Structural properties and electronic structure of low-compressibility materials: β-Si3N4 and hypothetical β-C3N4. Phys Rev B 1990, 41: 10727–10734. 10.1103/PhysRevB.41.10727

    Article  Google Scholar 

  47. Zhang SB: The microscopic origin of the doping limits in semiconductors and wide-gap materials and recent developments in overcoming these limits: a review. J Phys Condens Matter 2002, 14: R881-R903. 10.1088/0953-8984/14/34/201

    Article  Google Scholar 

  48. Pan H, Yi J, Shen L, Wu R, Yang J, Lin J, Feng YP, Ding J, Van LH, Yin JH: Room temperature dilute magnetic semiconductor in carbon-doped ZnO. Phys Rev Lett 2007, 99: 127201.

    Article  Google Scholar 

  49. Akai A: Ferromagnetism and its stability in the diluted magnetic semiconductor (In, Mn)As. Phys Rev Lett 1998, 81: 3002–3005. 10.1103/PhysRevLett.81.3002

    Article  Google Scholar 

  50. Pan H, Feng YP, Wu Q, Huang Z, Lin J: Magnetic property of carbon doped CdS: a first-principles and Monte Carlo study. Phys Rev B 2008, 77: 125211.

    Article  Google Scholar 

  51. Dalpian GM, Wei SH: Carrier-mediated stabilization of ferromagnetism in semiconductors: holes and electrons. Phys Stat Sol B 2006, 243: 2170–2187. 10.1002/pssb.200666809

    Article  Google Scholar 

  52. Pan H, Zhang YW, Shenoy V, Gao HJ: Controllable magnetic property of SiC by anion-cation codoping. Appl Phys Lett 2010, 96: 192510. 10.1063/1.3428428

    Article  Google Scholar 

  53. Olsson P, Domain C, Guillemoles JF: Ferromagnetic compounds for high efficiency photovoltaic conversion: the case of AlP:Cr. Phys Rev Lett 2009, 102: 227204.

    Article  Google Scholar 

  54. Pan H, Zhang YW: GaN-ZnO superlattice nanowires as photocatalyst for hydrogen generation: a first-principles study on electronic and magnetic properties. Nano Energ 2012, 1: 488–493. 10.1016/j.nanoen.2012.03.001

    Article  Google Scholar 

Download references


Hui Pan thanks the supports of the Science and Technology Development Fund from Macao SAR (FDCT-076/2013/A) and Multi-Year Research Grant (MYRG2014-00159-FST) and Start-up Research Grant (SRG-2013-00033-FST) from the Research & Development Office at the University of Macau. The DFT calculations were performed at the High-Performance Computing Cluster (HPCC) of the Information and Communication Technology Office (ICTO) at the University of Macau.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Hui Pan.

Additional information

Competing interests

The author declares that he has no competing interests.

Authors’ original submitted files for images

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Pan, H. Ab initio design of nanostructures for solar energy conversion: a case study on silicon nitride nanowire. Nanoscale Res Lett 9, 531 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Silicon nitride nanowire
  • Solar energy harvesting
  • Doping
  • First-principles calculation