See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/363890576

FIRST-PRINCIPLE LDA+U AND GGA+U CALCULATIONS ON STRUCTURAL AND

ELECTRONIC PROPERTIES OF WURTZITE ZnO

Article · September 2022

CITATIONS READS

0 15

10 authors, including:

Norjuliyati Hamzah Mohd Hazrie Samat

Universiti Teknologi MARA Universiti Teknologi MARA
5 PUBLICATIONS   3 CITATIONS    22 PUBLICATIONS   129 CITATIONS   

SEE PROFILE SEE PROFILE

Noor 'Aisyah Johari Ahmad Fuzairi Ahmad Faizal

Universiti Teknologi MARA Universiti Teknologi MARA
9 PUBLICATIONS   111 CITATIONS    2 PUBLICATIONS   0 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Modified Silicone-Dammar View project

battery View project

All content following this page was uploaded by Norjuliyati Hamzah on 28 September 2022.

The user has requested enhancement of the downloaded file.

Solid State Science and Technology, Vol. 30, No 1 & 2 (2022), 20-36
ISSN 0128-7389 | http://myjms.mohe.gov.my/index.php/masshp

FIRST-PRINCIPLE LDA+U AND GGA+U CALCULATIONS ON

STRUCTURAL AND ELECTRONIC PROPERTIES OF WURTZITE ZnO

N. Hamzah1, M. H. Samat2, N. A. Johari1, A. F. A. Faizal1, O. H. Hassan2,3, A. M. M.

Ali2,4, R. Zakaria2,4, N. H. Hussin5, M. Z. A. Yahya6 and M. F. M. Taib2,4
1
Centre of Foundation Studies, Universiti Teknologi MARA,
43800 Dengkil, Selangor, Malaysia
2
Ionic Materials & Devices (iMADE) Research Laboratory,
Institute of Science, Universiti Teknologi MARA,
40450 Shah Alam, Selangor, Malaysia
3
Department of Industrial Ceramics,
Faculty of Art & Design, Universiti Teknologi MARA,
40450 Shah Alam, Selangor, Malaysia
4
Faculty of Applied Sciences, Universiti Teknologi MARA,
40450 Shah Alam, Selangor, Malaysia
5
Faculty of Applied Sciences, Universiti Teknologi MARA,
26400 Jengka, Pahang, Malaysia
6
Faculty of Defence Sciences and Technology,
Universiti Pertahanan Nasional Malaysia,
57000 Kuala Lumpur, Malaysia

*Corresponding authors: [email protected]

Received: 4 January 2022; Revised: 8 August 2022; Accepted: 12 September 2022

ABSTRACT

First-principles calculations with density functional theory (DFT) were carried out to
explore the effects of Hubbard on-site Coulombic correction on the structural and
electronic properties of wurtzite zinc oxide (ZnO). For an accurate prediction of ZnO
properties, adequate Hubbard terms must be established due to changes in structural
parameters produced by the correction of hybridization between Zn 3d and O 2p states.
The calculations are based on the local density approximation (LDA) and the
generalized gradient approximation (GGA) for Perdew-Burke-Ernzerhof (PBE) and
Perdew-Burke-Ernzerhof for solids (PBEsol) were performed by applying Hubbard
corrections Ud to Zn 3d states and Up to O 2p states. The lattice parameters were closer
to the experimental data and underestimated when Hubbard corrections Ud and Up were
applied in the calculation. The combination of the correction terms Ud and Up managed
to improve the underestimated bandgap of wurtzite ZnO, which might solve the

20
 Solid State Sci. Technol.

standard DFT problems. The best Hubbard parameters Ud and Up found for LDA+U at
Ud = 6 eV and Up = 8 eV, GGA-PBE at Ud =2 eV and Up = 10 eV and GGA-PBEsol+U
at Ud = 5 eV and Up = 9 eV show a good agreement with the experimental bandgap.

Keywords: Wurtzite ZnO, first-principles, density functional theory, Hubbard

correction, electronic

INTRODUCTION

Zinc oxide (ZnO) is a compound semiconductor belonging to the II-VI groups with a
direct bandgap of 3.37 eV and a higher exciton binding energy of 60 meV at room
temperature [1,2]. ZnO has attracted much attention due to its potential application in
solar cells, photocatalyst, optoelectronic devices, semiconductor laser and light-emitting
diodes [3–7]. ZnO has a stable wurtzite crystal structure under ambient conditions.
Nevertheless, at relatively high pressure, the ZnO has observed a rock-salt crystal
structure and ZnO is stabilized by the growth of a zinc-blende crystal structure on a
cubic substrate [8].

The computational method based on density functional theory (DFT) has successfully
treated the structural, electronic, and optical properties of ZnO [9–11]. The DFT
calculations focused on local density approximation (LDA) and generalized gradient
approximation (GGA) are practical methods to simulate relatively large structures with
sufficient chemical precision. However, the standard DFT approach focused on the
exchange-correlation (XC) functionals of the LDA and GGA yields underestimated
value bandgap energy for some semiconductors [12,13]. DFT calculations report a
bandgap of ZnO is (0.65 – 0.74 eV), which is underestimated by ~70% compared with
the experimental value is (3.37 eV) [14–17]. The apparent error in the calculation of
energy bandgap failed to describe the localization of strongly correlated electronic d and
f electrons metals [18,19].

The hybrid functionals or DFT+U methods are the most commonly utilized approaches
to overcome the problems of standard DFT [20]. The Green’s function and screened
Coulomb’s potential (GW) and hybrid functionals like Perdew-Burke-Ernzerhof
(PBE0), Heyd-Scuseria-Ernzerhof (HSE06) and Becke, 3-parameters, Lee-Yang-Parr
(B3LYP) are a popular approach to be more accurate in the calculation of the energy
bandgap of ZnO [9,21,22]. However, hybrid functionals are considered to require
higher computing costs [23]. Therefore, the first-principle method based on DFT+U is a
barely suitable approach due to the less expensive computational method and low
difficulty level. Nevertheless, applying the Hubbard Ud correction only to the d state,
the bandgap of ZnO is still underestimated, even at large Ud values [24,25]. Numerous
recent theoretical studies have investigated the influence of the Hubbard parameter Ud
on the d state of Zn and Up on the p state of oxygen (O) can improve the bandgap and
position of the valence and conduction band when both Hubbard parameters Ud and Up
are used [26–29]. Therefore, the Hubbard parameters Ud and Up values need to be
rationally correct in order to obtain more accurate structural parameters and electronic
properties of ZnO.

21
 Solid State Sci. Technol.

In this work, the aim is to determine the suitable values of Hubbard parameters Ud and
Up using first-principles calculation within the DFT+U (LDA+U, GGA-PBE+U and
GGA-PBEsol+U) framework, which depends on structural (lattice parameters and
volume) and electronic (band structures and total and partial density of state) properties
of wurtzite ZnO. Different values of the Hubbard parameters Ud and Up are used to do a
series of DFT+U calculations. The results from inclusion Hubbard parameters Ud and
Up on DFT+U calculations were compared with those calculated using standard DFT
and experimental data. A similar work of inclusion Hubbard parameters Ud and Up on
DFT+U calculation of wurtzite ZnO had been carried out, but only using LDA+U
functional and different computational software [28].

COMPUTATIONAL DETAIL

First-principles calculations based on plane-wave ultrasoft pseudopotentials method

were performed for wurtzite ZnO using Cambridge Serial Total Energy Package
(CASTEP) module in Materials Studio (MS) 8.0 developed by Accelrys Sofware Inc
[30,31]. The structural and electronic properties of wurtzite ZnO were calculated using
XC functionals from LDA by Ceperley and Adler as parametrized by Perdew and
Zunger (LDA-CAPZ) and GGA of Perdew-Burke-Ernzerhof (GGA-PBE) and Perdew-
Burke-Ernzerhof for solids (GGA-PBEsol) [13,32,33]. The valence electron
configurations were treated as Zn-4s23d10 and O-2s22p4. The optimized cut-off energy
value for the plane-wave basis set of the electronic wave function was 340 eV. The
integration in the Brillouin zone was performed using 5 × 5 × 4 k-points from the
Monkhorst-Pack scheme. The geometrical optimization was performed using the total
energy of 1.0 × 10-5 eV/atom, the maximum force of 0.03 eV/Å, the maximum stress of
0.05 GPa and maximum atomic displacement of 1.0 × 10-3 Å.

Three strategies were performed by utilizing distinct XC functionals to investigate the

conceptual differences in calculation. The optimization was first performed on the unit
cell of wurtzite ZnO using a standard DFT (LDA, GGA-PBE and GGA-PBEsol). In the
second optimization, the Hubbard term Ud is applied to the 3d states of the Zn atoms
using DFT+U (LDA+U, GGA-PBE+U and GGA-PBEsol+U), which the Ud values
range from 1 eV to 10 eV. The third optimization process implemented both Hubbard
term Ud to Zn 3d states and Up to O 2p states using DFT+U (LDA+U, GGA-PBE+U
and GGA-PBEsol+U), where the Ud and Up values range from 1 eV to 10 eV. To
provide an accurate description of the calculated structural and electronic properties of
wurtzite ZnO, the DFT + Ud + Up method were adopted according to the following
formalism [26, 34];

(1)

where denotes the spin polarized on-site density matrix. The spherically averaged
Hubbard parameter U represents the increase of energy generated by placing an

22
 Solid State Sci. Technol.

additional electron at a certain position and the screened exchange energy is represented
by parameter J (1 eV). The various values in the optimized lattice parameter and
corresponding bandgap energy, Eg were investigated. The best XC functional for the
ZnO system was determined through analysis.

RESULTS AND DISCUSSION

Structural Parameters

The crystal structure of wurtzite ZnO (hexagonal, space group P63mc) is visualized in
Figure 1. The unit cell of wurtzite ZnO contains 4 atoms which consist of 2 zinc (Zn)
and 2 oxygen (O). The crystal structures of wurtzite ZnO were optimized by minimizing
the total energy and atomic forces. Optimization was carried out utilizing several
approximations in order to identify the best functional for the ZnO system. The lattice
parameters (a and c), lattice ratio (c/a) and volume (V) of optimized wurtzite ZnO
computed from standard LDA, GGA-PBE and GGA-PBEsol functionals were
compared to the experimental values were listed in Table 1. The value of lattice
parameters from standard LDA was found underestimates (1.37% – 1.82%) the values
of lattice parameters from the standard of GGA (0.10% – 1.91%). In contrast, the value
of lattice parameters GGA-PBE and GGA-PBEsol were consistent with each other,
which agrees with other findings [35–37]. The lattice parameters and volume from
GGA-PBEsol show a good agreement with the experimental values, which is less than
0.59% deviation compared to the other calculated lattice parameters and volume from
standard LDA and GGA-PBE functionals.

Figure 1: Crystal structure of unit cell for wurtzite ZnO

23
 Solid State Sci. Technol.

Table 1: The calculated lattice parameters (a and c), lattice ratio (c/a) and volume (V) of
wurtzite ZnO using standard LDA, GGA-PBE and GGA-PBEsol with the percentage
deviation from the experiment and other theoretical data

Method
Other works
Present works
[35–37] Experiment [38,39]
LDA GGA-PBEGGA-PBEsol LDA GGA-PBEGGA-PBEsol
3.1907 3.2888 3.2466
a (Å) 3.1860 3.2819 3.2710 3.2499
(‒1.82%) (+1.20%) (‒0.10%)
5.1598 5.3056 5.2317
c (Å) 5.1500 5.2943 5.2350 5.2063
(‒1.37%) (+1.91%) (+0.49%)
1.6172 1.6132 1.6115
c/a 1.6200 1.6130 1.6000 1.6020
(+0.95%) (+0.70%) (+0.59%)
3 45.492 49.699 47.755
V (Å ) (‒4.42%) (+4.42%) (+0.33%) 45.270 49.386 47.759 47.597
(+) overestimate; (‒) underestimate

The effects of inclusion Hubbard parameter Ud applied to the 3d states of the Zn on the
lattice parameters and volume of wurtzite ZnO was investigated by optimizing the
crystal structures using LDA+U, GGA-PBE+U and GGA-PBEsol+U functionals. The
lattice parameters a and c and volume gradually increased with the increment in the
Hubbard parameter Ud range from 1 eV to 10 eV from LDA+U, GGA-PBE+U and
GGA-PBEsol+U functionals as in Figure 2, in agreement with the result produced in
similar research using CASTEP software by other theoretical researchers [24,35,37].
Figure 2 (a) shows that at Ud = 2 eV, lattice parameter a from the GGA-PBEsol+U
functional value is equal to the experimental value 3.2499 Å, which represents by
horizontal dash lines [38]. However, the trend lines of lattice parameters c and volume
in Figure 2 (b) and (c) never reach the point of the experimental values, which are
5.2063 Å and 47.597 Å3 [39]. Therefore, the lattice parameters of wurtzite ZnO using
the inclusion Hubbard parameter Ud applied to the 3d states of the Zn were best
described from GGA-PBEsol+U at Ud = 2 eV for only calculation lattice parameter a
and none for calculation lattice parameter c and volume. Compared with the previous
standard DFT calculations, the percentage deviation of lattice parameters a and c and
volume of wurtzite ZnO without the inclusion Hubbard parameter Ud is still good and
below +0.49% using standard GGA-PBEsol.

The effects of inclusion Hubbard parameter Ud applied to the 3d states of the Zn and Up
applied to the 2p states of the O on the lattice parameters and volume of wurtzite ZnO
was investigated by optimizing the crystal structures using LDA+U, GGA-PBE+U and
GGA-PBEsol+U functionals. The lattice parameters a and c and volume gradually
decreased with the increment in the Hubbard parameter Up at all fixed values of Ud
range from 1 eV to 10 eV from LDA+U, GGA-PBE+U and GGA-PBEsol+U
functionals as in Figure 3, Figure 4 and Figure 5 due to stronger localization of the
electron state. Using LDA+U, GGA-PBE+U and GGA-PBEsol+U functionals, similar
trends are agreed by the other theoretical researchers using the same CASTEP software
[40]. Figures 3 (a), 4 (a) and 5 (a) show that none of the Ud and Up matches the
experimental lattice parameters a and c and volume values. The lattice parameters a and

24
 Solid State Sci. Technol.

c and the volume calculated using the LDA+U functional are underestimated and
further away from experimental values as the Hubbard parameter Up increases. A
similar trend is agreed by the other theoretical researchers using the same LDA+U
functional but different software [28]. Meanwhile, Figures 4 and Figure 5 show that the
reduction of lattice parameters a and c and the volume from GGA-PBE+U and GGA-
PBEsol+U improves the overestimation from standard GGA-PBE and GGA-PBEsol.

Figure 2: (a) Lattice parameter a, (b) lattice parameter c and (c) volume (V) of wurtzite
ZnO with respect to Hubbard parameter Ud applied to Zn for different functionals
LDA+U, GGA-PBE+U and GGA-PBEsol+U. Horizontal dashed lines represent the
experimental values

From Figures 3 (b), 4 (b) and 5 (b), it show the matches lattice parameter a value with
experimental at point Ud = 6 eV and Up = 10 eV (0.00%), lattice parameter c at point Ud
= 1 eV and Up = 9 eV (+0.02%) and volume at point Ud = 6 eV and Up = 10 eV
(+0.14%), whereas Figure 3 (c), 4 (c) and 5 (c) show the matches lattice parameter a
value with experimental at point Ud = 8 eV and Up = 3 eV (0.00%), lattice parameter c
at point Ud = 9 eV and Up = 5 eV (+0.01%) and volume at point Ud = 9 eV and Up = 4
eV ( 0.07%).

25
 Solid State Sci. Technol.

Figure 3: Lattice parameters a of wurtzite ZnO with respect to Hubbard parameter Up

applied to O with fixed Ud values for different functionals (a) LDA+U, (b) GGA-
PBE+U and (c) GGA-PBEsol+U. Horizontal dashed lines represent the experimental
Values.

26
 Solid State Sci. Technol.

Figure 4: Lattice parameters c of wurtzite ZnO with respect to Hubbard parameter Up

applied to O with fixed Ud values for different functionals (a) LDA+U, (b) GGA-
PBE+U and (c) GGA-PBEsol+U. Horizontal dashed lines represent the experimental
values.

27
 Solid State Sci. Technol.

Figure 5: Volume (V) of wurtzite ZnO with respect to Hubbard parameter Up applied to
O with fixed Ud values for different functionals (a) LDA+U, (b) GGA-PBE+U and (c)
GGA-PBEsol+U. Horizontal dashed lines represent the experimental values

28
 Solid State Sci. Technol.

The reduction of lattice parameter a with increased the Hubbard parameter Up agreed
with other theoretical researchers using the same GGA+U functional but different
software [41]. It is concluded that the structural parameters of wurtzite ZnO were most
aptly described from GGA-PBEsol+U with possible agreements of different Hubbard
Ud and Up values. The effects of inclusion Hubbard parameter Ud applied to the 3d
states of the Zn and Up applied to the 2p states of the O were underestimated the
structural properties of wurtzite ZnO.

Electronic Properties

The calculated bandgaps energy of wurtzite ZnO using standard LDA, GGA-PBE and
GGA-PBEsol functionals were found underestimated (76% – 82%) compared to the
experimental values are listed in Table 2, which is agreed well with the previous DFT
researches [40,42,43]. These problems are driven by standard LDA, GGA-PBE and
GGA-PBEsol, that failure to locate the binding energy in the 3d states electrons,
resulting in the overestimation of hybridization with the anion 2p states. Thus, a strong
p-d coupling was finally causing the reduction of the bandgaps [41].

Table 2: The bandgap (Eg) of wurtzite ZnO using standard LDA, GGA-PBE and GGA-
PBEsol and LDA+U, GGA-PBE+U and GGA-PBEsol+U with the percentage deviation
from the experiment and other theoretical data

Method Ud/eV Up/eV Eg/eV

LDA - - 0.791 (‒76.53%)
GGA-PBE - - 0.733 (‒78.25%)
GGA-PBEsol - - 0.618 (‒81.66%)
Presents works
LDA+U 6 8 3.364 (‒0.18%)
GGA-PBE+U 2 10 3.379 (+0.27%)
GGA-PBEsol+U 5 9 3.365 (‒0.15%)
LDA - - 0.770
GGA-PBE - - 0.741
Other works GGA-PBEsol - - 0.622
[26,40,42–45] LDA+U 8 7 3.310
GGA-PBE+U 10 7 3.300
GGA-PBEsol+U 10 7 3.370
Experiment [38,39] - - 3.37
(+) overestimate; (‒) underestimate

The variation bandgap values of wurtzite ZnO with the inclusion of Hubbard parameter
Ud applied to Zn for different functionals are presented in Figure 6. As Ud values
increased from 1 eV to 10 eV, the bandgap values also increased. The highest values of
bandgaps at Ud = 10 eV are 1.30 eV (LDA+U), 1.17 eV (GGA-PBE+U) and 1.11 eV

29
 Solid State Sci. Technol.

(GGA-PBEsol +U). These bandgap values are still small compared to the experimental
value (3.37 eV), which agrees with previous researches even though the inclusion of the
Hubbard Ud parameter aims to correct the localized 3d states electrons in a correlated
system but is still underestimated [28,37,46]. When only the Hubbard parameter Ud was
changed, a significant fluctuation in total energy was observed, indicating an
incomplete representation of the bandgap structure [41].

Figure 6: Dependence bandgap values of wurtzite ZnO with respect to Hubbard

parameter Ud applied to Zn for different functionals LDA+U, GGA-PBE+U and GGA-
PBEsol+U

The widespread underestimate in bandgap has brought to light the necessity of

including the 2p states in the calculations. The selection of the appropriate value of Ud
and Up are required to investigate the bandgaps properties of wurtzite ZnO. Figure 7
presented the variation bandgap values of wurtzite ZnO with the Hubbard parameter Up
applied to O with fixed Ud values for different functionals. The bandgap values
gradually increased with the increment in the Hubbard parameter Up range from 1 eV to
10 eV with fixed Ud values. The matches bandgaps value with experimental is listed in
Table 2. Thus, the bandgap from GGA-PBEsol+U functional nearly matched the
experimental by 0.15% deviation. From the previous research, the selection of Hubbard
Ud is typically in the range of 8 eV to 10 eV, whereas for Up in the range of 7 eV, using
the same CASTEP software [26,44,45]. Thus, this will prove that the inclusion of both
Hubbard parameters, Ud dan Up has successfully reproduced the correct bandgap of
wurtzite ZnO compared to the standard LDA, GGA+PBE and GGA+PBEsol. The
Hubbard parameters Ud and Up values for electronic properties are not the same as the
structural properties, but the problem is to correct the underestimated bandgap of
wurtzite ZnO from standard DFT by inclusion Hubbard parameter Ud applied to the Zn-
3d state and Up applied to the O-2p state. The structural properties using standard DFT
are still acceptable due to percentage deviation below +5.00%.

30
 Solid State Sci. Technol.

Figure 7: Dependence bandgap values of wurtzite ZnO with respect to Hubbard

parameter Up applied to O with fixed Ud values for different functionals (a) LDA+U, (b)
GGA-PBE+U and (c) GGA-PBEsol+U. Horizontal dashed lines represent the
experimental values

The calculated band structures of wurtzite ZnO using standard LDA, GGA-PBE and
GGA-PBEsol functionals and the best Hubbard parameters Ud and Up for LDA+U,
GGA-PBE+U and GGA-PBEsol+U along the highest symmetry direction of the
Brillouin zone path from G-A-H-K-G-M-L-H are presented in Figure 8. The results
imply that the wurtzite ZnO has a direct energy bandgap between the G Brillouin zone
valence and the conduction band. The bandgap energy is determined between the
valence band maximum and the conduction band minimum, in which the Fermi level is
situated at 0 eV of the energy scale. The calculated bandgaps of wurtzite ZnO from
standard LDA, GGA+PBE and GGA-PBEsol were about 0.791 eV, 0.733 eV and 0.618
eV, in good agreement with the other first-principle DFT calculations [40,47,48].
Nevertheless, the bandgap results did not affect the accuracy of comparing the related
crystal properties, such as band structure and density of state.

31
 Solid State Sci. Technol.

Figure 8: Comparison bandgap structures of wurtzite ZnO using different functionals

(a) LDA and LDA+U, (b) GGA-PBE and GGA-PBE+U and (c) GGA-PBEsol and
GGA-PBEsol+U for without Hubbard U and with the best Hubbard parameters Ud and
Up

It is well known that the underestimated bandgap can be due to the ground state DFT
results of strong Columb correlation and XC potential between excited-state electrons.
Underestimation of the bandgap from the first-principle calculations can be fixed using
Hubbard parameter Ud and Up to compare it with the experimental results. Therefore,
the effective Hubbard parameters Ud and Up in LDA+U, GGA-PBE+U and GGA-
PBEsol+U calculations tend to upper shifted the minimum conduction band to give the

32
 Solid State Sci. Technol.

larger gap between the maximum valence band and minimum conduction band. The
calculated bandgaps of wurtzite ZnO from LDA+U, GGA+PBE+U and GGA-
PBEsol+U were about 3.364 eV, 3.379 eV and 3.365 eV. All the bandgaps values are in
good agreement with the other first-principle DFT+U calculations [26,49,50]. Hence,
the bandgap from GGA-PBEsol functional nearly matched the experimental by 0.15%
deviation.

The contribution of the partial density of state (PDOS) and total density of state (TDOS)
of wurtzite ZnO obtained from standard LDA, GGA-PBE and GGA-PBEsol functionals
and the best Hubbard parameters Ud and Up for LDA+U, GGA-PBE+U and GGA-
PBEsol+U are illustrated in Figure 9. The PDOS and TDOS were used to analyze the
contribution of the electrons from s, p and d orbitals in wurtzite ZnO. The calculated
DOS of wurtzite ZnO for LDA, GGA-PBE, GGA-PBEsol, LDA+U, GGA-PBE+U and
GGA-PBEsol+U were within energy interval from ‒20 eV to 25 eV and the Fermi level
situated at 0 eV of the energy scale.

From the standard LDA, GGA-PBE and GGA-PBEsol, the first peak at around ‒17.38
eV to 17.21 eV in the valence band originates from the contribution of the Zn 4s and
O 2s states, but it is mainly controlled by O 2s state. Hence, the deep-level inner-layer
electrons belong to O 2s states. The higher energy region of the valence band situated
from ‒6 eV to 0 eV is mainly derived from Zn 3d and O 2p states due to the strong
hybridization that occurred between both states. The top region of the valence band is
primarily originated from the O 2p state. The bottom part of the conduction band above
Fermi level is contributed mainly by Zn 4s state. The bandgap values are determined by
both Zn 4s states and O 2p states. All the results of wurtzite ZnO from PDOS and
TDOS of standard LDA, GGA-PBE and GGA-PBEsol agreed with other reports using
DFT [15,48,51].

The inclusion of Hubbard parameter Ud and Up in LDA+U, GGA-PBE+U and GGA-

PBEsol+U calculations led to a downward shift of the hybridized Zn 3d and O 2p states
and tended to upper shifted the bottom conduction band to the broadening bandgap
between the highest valence band and lowest conduction band. The Zn 3d and O 2p
states at the higher energy region of the valence band moved to a relevant position
about ‒6.9 eV compared to standard LDA, GGA-PBE and GGA-PBEsol about ‒6 eV.

33
 Solid State Sci. Technol.

Figure 9: Comparison PDOS and TDOS of wurtzite ZnO using different functionals (a)
LDA and LDA+U, (b) GGA-PBE and GGA-PBE+U and (c) GGA-PBEsol and GGA-
PBEsol+U for without Hubbard U and with the best Hubbard parameters Ud and Up

CONCLUSION

In this study, the structural parameters and electronic properties of wurtzite ZnO were
evaluated by first-principles calculations based on DFT and DFT+U methods from
CASTEP software. The inclusion of Hubbard parameters Ud and Up in LDA+U, GGA-
PBE+U and GGA-PBEsol+U have marked a significant difference in the structural
parameters and electronic structure of wurtzite ZnO. The inclusion of Hubbard
parameters Ud and Up have underestimated the lattice parameters and volume of
wurtzite ZnO and improved the underestimated bandgap from standard DFT. The
calculated lattice parameters and volume were most aptly described from GGA-
PBEsol+U functional with possible agreements of different Hubbard Ud and Up values.
The best Hubbard parameters of wurtzite ZnO found for functionals LDA+U at Ud = 6
eV and Up = 8 eV, GGA-PBE+U at Ud = 2 eV and Up = 10 eV and GGA-PBEsol+U at
Ud = 5 eV and Up = 9 eV show a good agreement with the experimental bandgap. DOS
calculations were used to investigate the composition that determined the wurtzite ZnO
band structure. The O 2p state primarily dominates the top part of the valence band and
the bottom part of the conduction band is contributed mainly by the Zn 4s state.
Incorporating the Hubbard parameters Ud and Up into calculations resulted in a
downward shift of the hybridized Zn 3d and O 2p states and a slight upper shift the

34
 Solid State Sci. Technol.

bottom conduction band toward the broadening bandgap between the highest valence
band and lowest conduction band. Our results suggest that the predicted GGA-
PBEsol+U method for investigating the structural and electronic properties of wurtzite
ZnO is acceptable for solving the DFT calculations problem at a low computing cost.

ACKNOWLEDGMENT

The authors would like to acknowledge the Ministry of Higher Education (MOHE)
Malaysia for funding this research under the KEPU grant 600-RMC/KEPU 5/3
(008/2021) and Universiti Teknologi Mara (UiTM), Cawangan Selangor, Dengkil,
Selangor, Malaysia and Faculty of Applied Sciences, Universiti Teknologi MARA,
Shah Alam, Selangor, Malaysia for providing the facilities and support on this research.

REFERENCES

[1] R. Vittal and K. C. Ho (2017) Renew. Sust. Energ. Rev. 70, 920.
[2] B. Mehmood, M. I. Khan, M. Iqbal, A. Mahmood and W. Al-Masry (2021) Int. J.
Energ. Res. 45(2), 2445.
[3] J. C. Fan, K. M. Sreekanth, Z. Xie, S. L. Chang and K. V. Rao (2013) Prog.
Mater. Sci. 58(6), 874.
[4] Y. Segawa, A. Ohtomo, M. Kawasaki, H. Koinuma, Z. K. Tang, P. Yu and G. K.
L. Wong (1997) Phys. Status Solidi 202(2), 669.
[5] C. W. Litton, D. C. Reynolds and T. C. Collins (2011) Zinc Oxide Materials for
Electronic and Optoelectronic Device Applications, United Kingdom, Wiley.
[6] E. Cerrato, C. Gionco, M. C. Paganini and E. Giamello (2017) J. Phys. Condens.
Matter 29(44), 444001.
[7] J. B. Baxter, A. M. Walker, K. Van Ommering and E. S. Aydil Nanotechnology
17(11), S304.
[8] Ü. Özgür, Y. I. Alivov, C. L. A. Teke, M. A. Reshchikov, S. Doğan, V.
Avrutin, S.-J. Cho and H. Morkoç (2005) J. Appl. Phys. 98, 041301.
[9] K. Bashyal, C. K. Pyles, S. Afroosheh, A. Lamichhane and A. T. Zayak (2018) J.
Phys. Condens. Matter 30, 065501.
[10] Z. Alahmed and H. Fu (2008) Phys. Rev. B 77, 045213.
[11] Z. Li, Y. Xu, G. Gao, T. Cui, and Y. Ma (2009) Phys. Rev. B 79, 193201.
[12] J. P. Perdew (1985) Int. J. Quantum Chem. 28(S19), 497.
[13] J. P. Perdew, K. Burke and M. Ernzerhof (1996) Phys. Rev. Lett.77, 3865.
[14] A. Janotti and C. G. Van De Walle (2009) Reports Prog. Phys. 72(12), 125501.
[15] A. A. Mohamad, M. S. Hassan, M. K. Yaakob, M. F. Mohamad Taib, F. W.
Badrudin, O. H. Hassan and M. Z. A. Yahya (2017) J. King Saud Univ. - Eng.
Sci. 29(3), 278.
[16] S. Lany and A. Zunger (2009) Model. Simul. Mater. Sci. Eng. 17(8), 084002.
[17] B. Yamina, A. Abderrahim, B. Omar and B. Lazhar (2018) Der. Pharma. Chem.
10(1), 16.
[18] V. I. Anisimov, J. Zaanen and O. K. Andersen (1991) Phys. Rev. B 44, 943.
[19] V. I. Anisimov, F. Aryasetiawan and A. I. Lichtenstein (1997) J. Phys. Condens.
Matter 9(4), 767.

35
 Solid State Sci. Technol.

[20] V. Ivády, K. Szasz, A. L. Falk, P. V. Klimov, D. J. Christle, E. Janzén, I.

Abrikosov, D. D. Awschalom and A. Gali (2015) Phys. Rev. B. 92(11), 115206.
[21] J. Heyd, G. E. Scuseria and M. Ernzerhof (2003) J. Chem. Phys. 118, 8207.
[22] M. Shishkin, M. Marsman and G. Kresse (2007) Phys. Rev. Lett. 99, 246403.
[23] P. J. Hasnip, K. Refson, M. I. J. Probert, J. R. Yates, S. J. Clark and C. J. Pickard
(2014) Philos. Trans. R. Soc. A 372, 20130270.
[24] R. Farooq, T. Mahmood, A. W. Anwar and G. N. Abbasi (2016) Superlattices
Microstruct 90, 165.
[25] X. H. Zhou, Q. Hu and Y. Fu (2008) J. Appl. Phys. 104, 063703.
[26] Y. S. Lee, Y. C. Peng, J. H. Lu, Y. R. Zhu and H. C. Wu (2014) Thin Solid Films
570, 464.
[27] R. M. Sheetz (2009) Phys. Rev. B 80, 195314.
[28] E. S. Goh, J. W. Mah and T. L. Yoon (2017) Comput. Mater. Sci. 138, 111.
[29] V. N. Jafarova, G. S. Orudzhev & U. Hubbard (2021) Solid State Commun. 325,
114166.
[30] M. D. Segall, P. J. D. Lindan, M. J. Probert, C. J. Pickard, P. J. Hasnip, S. J.
Clark and M. C. Payne (2002) J. Phys.: Condens. Matter 14, 2717.
[31] J. P. Perdew and Y. Wang (1992) Phys. Rev. B 45, 13244.
[32] J. P. Perdew and A. Zunger (1981) Phys. Rev. B 23, 5048.
[33] J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A.
Constantin, X. Zhou and K. Burke (2008) Phys. Rev. Lett. 100, 136406.
[34] G. J. Shao (2009) J. Phys. Chem. C 113(16), 6800.
[35] J. Wang, T. Shen, Y. Feng and H. Liu (2020) Phys. B: Condens. Matter 576,
411720.
[36] J. Q. Wen, J. M. Zhang, Z. G. Qiu, X. Yang and Z. Q. Li (2018) Phys. B
Condens. Matter 534, 44.
[37] M. K. Yaakob, N. H. Hussin, M. F. M. Taib, T. I. T. Kudin, O. H. Hassan, A. M.
M. Ali and M. Z. A. Yahya (2014) Integr. Ferroelectr. 155, 15.
[38] K. Harun, F. Hussain, A. Purwanto, B. Sahraoui, A. Zawadzka and A. A.
Mohamad (2017) Mater. Res. Express 4(12), 122001.
[39] S. Desgreniers (1998) Phys. Rev. B 58, 14102.
[40] K. Harun, M. K. Yaakob, M. F. M. Taib, B. Sahraoui, Z. A. Ahmad and A. A.
Mohamad (2017) Mater. Res. Express 4(6), 085908.
[41] X. Ma, Y. Wu, Y. Lv and Y. Zhu (2013) J. Phys. Chem. C 117, 26029.
[42] C. Xia, F. Wang and C. Hu (2014). J. Alloys Compd. 589, 604.
[43] J. Wu, L. Shao, J. Xu and J. Hu (2015) Sci. Semicond. Process. 29, 245.
[44] K. Harun, N. Mansor and M. Kamil (2016) J. Sol-Gel Sci. Technol. 80, 56.
[45] H. C. Wu, H. H. Chen and Y. R. Zhu (2016) Materials (Basel). 9(8), 647.
[46] H. Zhang, S. Lu, W. Xu and F. Yuan (2014) Surf. Sci. 625, 30.
[47] L. Li, W. Wang, H. Liu, X. Liu, Q. Song and S. Ren (2009) J. Phys. Chem.
C 113(19), 8460.
[48] C. Zuo, J. Wen and C. Zhong (2012) J. Semicond. 33(7), 072001.
[49] L. Honglin, L. Yingbo, L. Jinzhu and Y. Ke (2014) J. Alloys Compd. 617, 102.
[50] Z. Ma, F. Ren, X. Ming, Y. Long and A. A. Volinsky (2019) Materials 12(1),
196.
[51] F. Xie, P. Yang, P. Li and L. Zhang (2012) OPTICS 285, 2660.

36

View publication stats