Visible light photoresponse properties of Ag/n-ZnO-NRs/p-Si/Al

photodiodes, the influence of optimal ZnO precursor concentration via

ed
CBD method

Fatemeh Batvandi and Hosein Eshghi*

iew
Faculty of Physics, Shahrood University of Technology, Shahrood, Iran.
* h_eshghi@shahroodut.ac.ir

fateme.batvandi1367@gmail.com

Abstract

v
re
In this study, we studied the influence of zinc nitrate hexahydrate (Zn(NO3)2.6H2O) precursor

concentrations (0.03, 0.05, and 0.07 M) in Ag/n-ZnO-NRs/p-Si/Al photodiodes on the photo-

response of the samples to the blue, green, and red light LEDs. The layers were characterised
er
by FESEM, EDX, XRD, PL, UV-Vis. reflectance spectra, and also, I-V and I-t in the dark

and under light illuminations. We found, depending on the precursor concentration: (1) the
pe
ZnO layers, while having a polycrystalline structure with a hexagonal (wurtzite) phase

mainly grown along the (002) direction, consist of nanorods (NRs) with different dimeters
ot

and lengths; (2) The optical reflectance spectra of the samples showed that the deposition of

the ZnO-NRs on the p-Si substrate acting as an antireflecting coating, and in S0.05 it reduced
tn

the visible light reflectance to less than ~ 2 %; (3) the analysis of the dark I-V data showed

the synthesised devices have rectifying behaviour; (4) the analysis of the I-t data, operated at
rin

-5 V, showed that the fabricated samples have a fast-switching characteristic (≈ 65 ms); (5)

the sample with a 0.05 M concentration, as the optimal layer, with the lowest light reflectance

feature, has the maximum light sensitivity (S=IP/ID) to various visible light illuminations (460
ep

for blue, 201 for green, and 190 for red).

Pr

Keywords: Photodiodes; Ag/n-ZnO-NRs/p-Si/Al heterostructure; CBD method;

Nanostructure; Visible light detection.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 1. Introduction

ed
In recent years, p-n heterostructure photodetectors based on nanomaterials have received

much attention [1–3]. Meanwhile, photodiodes (PDs) can detect light in different wavelength

iew
ranges, depending on the materials used. A photodiode acts as a normal diode when it is in

the dark, which provides strong electrical conduction in forward bias and very weak in

reverse bias. When a photodiode is exposed to light, a photocurrent is produced that is easily

v
detected in reverse bias. Applying a reverse bias to a photodiode leads to an increase in the

re
width of the depletion layer and also to a strengthening in the electric field at the junction,

which are very important parameters in improving the performance of the photodiode.

Among various materials, metal oxide compounds such as zinc oxide (ZnO) have attracted a
er
lot of attention due to their outstanding and unique properties. ZnO thin film with a wide

band gap energy (3.3 eV) [4,5] and high transparency (~ 80–90%) in the visible (Vis) and
pe
infrared (IR) regions, thus acting as a window layer, is a promising material in contact with

the p-Si substrate [6–13]. In this configuration, the Si substrate acts as a light-absorbing layer

[14–18]. To prepare ZnO thin film, several deposition techniques, including sol-gel [19],
ot

radio frequency (RF) sputtering [20–22], chemical bath deposition (CBD) [23–25], and spray
tn

pyrolysis [26], are used. Several research articles related to n-ZnO/p-Si heterostructures for

the fabrication of UV-PDs have been reported [27–30]. However, to the best of our

knowledge, except for a few, not much work has been done on the application of this
rin

structure in the visible range. Karthick et al. [31] reported the synthesis of ZnO nanoparticles

using the direct injection flame synthesis technique on a p-Si substrate for visible light
ep

photodetection applications. Through this investigation, they found that ZnO nanoparticles

(NPs) have a mixed morphology, including nanorods (NRs), nanoneedles, and nanoplates,
Pr

with an average particle size of 127 nm and an optical bandgap of 3.27 eV. The synthesised

ZnO-NPs were then coated on the p-Si substrate. They showed that the fabricated n-ZnO/p-Si

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 photodiode in darkness has an ideality factor of 4.10, a barrier height of 0.763 eV, and a

ed
saturation current of 6.95×10-6 A. According to this report, the fabricated device has a

photosensitivity (the ratio of the photocurrent current to the dark current) of 535% under

visible light with an intensity of 100 mW/cm2.

iew
In this investigation, first we studied the physical properties of ZnO thin layers prepared

by spin coating, as the seed layer, followed by the chemical bath deposition (CBD) method

v
using various precursor concentrations on a p-Si substrate, that led to the formation of ZnO-

re
NRs, acting as an antireflecting coating. Through this study, we compared and discussed the

influence of this quantity on the I-V rectifying behaviour of Ag/n-ZnO-NRs/p-Si/Al

heterostructure diodes in the dark and under various visible (blue, green, and red) light-
er
emitting diodes (LEDs) light illumination conditions.
pe
2. Experimental

The n-ZnO/p-Si heterostructure photodetectors were synthesised by the chemical bath

deposition (CBD) method. The preparation process consisted of three steps: In the first step,
ot

the p-Si (1–10 Ω.cm) substrates were cleaned using an ultrasonic bath containing ethanol and

deionized water for 20 min., then immersed in a solution (HF:H2O=1:4) to remove the
tn

possible thin layer of SiO2 from the surface, and dried with air flow. In the second step, the

ZnO seed layer was deposited on the Si substrate using the spin coating method. In this stage,
rin

a 0.76 M zinc acetate (Zn(CH3COO)2.2H2O) was used as the zinc salt and ethanol as the

solvent at 70 °C. The solution was then dropped on the Si substrate and dried at 300°C for 10
ep

min. This process was repeated three times, and finally it was annealed at 300 °C for 1 hour.

In the third step, the growth of ZnO-NRs on top of the seed layer/p-Si was done by the CBD
Pr

method. 50 ml of an aqueous solution of zinc nitrate hexahydrate (Zn(NO3)2.6H2O) (purity

98%) with various concentrations of 0.03, 0.05, and 0.07 M and 50 ml of an aqueous solution

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 of hexamethylene tetramine (C6H12N4) (with a purity of 99%) were mixed together at 70 °C.

ed
All the seeded Si substrates were then placed in these solutions for 3 hours at 95 °C. The

fabricated samples were labelled S0.03, S0.05, and S0.07, respectively. To prepare samples

for use as photodiodes (PDs), an Ag metallic film using a shadow mask was deposited on the

iew
ZnO layers with a thickness of 120 nm for the front contact, and for the back contact, an Al

metallic layer was deposited on the rear of the Si substrate by sputtering method.

v
It is noticeable that the main chemical reactions for the fabrication of ZnO-NRs through

re
the above process could be expected to take place as the following chemical reactions [25]:

(CH2)6N4 + 6H2O→6HCHO + 4NH3 (1)

NH3 + H2O↔NH+
4 + OH
―
er (2)
pe
Zn+2 + 2OH―↔ZnO + H2O (3)

Several characterization techniques were employed to determine the samples' physical

characteristics: for the surface and cross-section morphology of the layers a Zeiss Sigma 300-
ot

HV field emission scanning electron microscope; for the structural properties, an X-ray

diffractometer (D8 Advance Bruker) using Cu-Kα radiation (=1.5404 Å) in the 2θ range 20–
tn

80º; for the optical properties, an ultraviolet-visible-near infrared (UV-Vis-NIR)

spectrophotometer (Shimadzu UV1800) through 300–1100 nm and also photoluminescence

rin

(PL) spectroscopy (Perkin-Elmer, LS55, USA) within 300-700 nm at the excitation

wavelength of 325 nm; finally, in order to measure the current-voltage (I-V) and current-time
ep

(I-t) data of the fabricated heterojunction PDs both in the dark and under 1 W LEDs

illuminating at blue-450 nm, green-514 nm, and red-622 nm, an electrochemical analysis

system (BHP2063) was used.

Pr

3. Results and Discussions

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 3.1 Surface Morphology

ed
Fig. 1 shows the top and cross-sectional FESEM images, as well as the energy dispersive

analysis by X-ray (EDX) spectra of the ZnO layers synthesised with different precursor

iew
concentrations on the p-Si substrate, labelled as S0.03, S0.05, and S0.07, respectively. As is

obvious, all samples contain compact NRs with a hexagonal cross-section almost

perpendicular to the substrate. These images also confirm that the variations in the precursor

v
concentration have an influence on the morphology of the layers, including the length and

re
diameter of the ZnO-NRs grown on the Si substrate (Table 1). These results show that the

grown NRs in S0.05 have the largest length and diameter. An important geometrical quantity

that is considered a measure of the surface to the volume of the synthesised nanostructures is
er
called the aspect ratio. This quantity, which is defined as the ratio of length to width of NRs,

is critical for devices whose functionality depends on the surface area available for gas or
pe
light interaction. As shown in Table 1, among these layers, S0.05 has the largest average

length (1200 nm) and the aspect ratio (15). In addition to the Si element that corresponds to

the substrate, the EDX analysis, which reveals the stoichiometry of the layer, confirms the
ot

presence of Zn and O elements (Fig. 1). The atomic percentages of the studied layers are
tn

shown in the inset. According to these results, the contribution of O-atoms is higher than

50%, while it is lower for Zn-atoms. This could be considered evidence for the presence of

oxygen interstitial (Zni) and zinc vacancies (VZn) as intrinsic crystalline defects in the
rin

synthesised layers. A simple calculation shows that the ratio of atomic percentages of Zn:O in

sample S0.05 has the highest value (0.808), which is closer to 1 as the ideal elemental
ep

composition. In fact, variation in stoichiometry, and thus the number of crystalline defects,

could affect the structural and optical characteristics of the layers. As discussed in the
Pr

following sections, these properties have a great impact on the behaviour of their

corresponding photodiodes.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 3.2 Structural Properties

ed
The XRD diffractograms of the examined samples are shown in Figs. 2(a,b). Fig. 2a

shows the diffractogram patterns of the seed and S0.03 samples for a clear comparison of

iew
their constituent peaks intensity. Fig. 2b shows the diffractogram related to our studied

samples. Based on these results, each pattern contains various peaks positioned at 2θ angles

of 31.85, 34.61, 36.39, 47.69, and 63.20º, corresponding to (100), (002), (101), (102), and

v
(110), respectively (JCPDS 36-1451). While there is no secondary phase in the synthesised

re
samples, all of the orientations are related to the ZnO hexagonal (wurtzite) phase. It is notable

that based on these spectra, the synthesised layers are mainly grown along the (002) direction,

indicating that the crystallites are growing along the c-axis normal to the substrate. This
er
preferential growth direction leads to the formation of ZnO-NRs with hexagonal bases,

consistent with their FESEM images (Fig. 1).

pe
For a comprehensive investigation to evaluate the crystallite size (D), dislocation density

(δ), induced stresses (ε), lattice constants (a and c), interplanner spacing dhkl for the (hkl)
ot

planes, volume (V) of the unit cell, and Zn-O bond length (L), the following equations were

used [32]:
tn

𝐷 = 0.9𝜆 𝛽 𝑐𝑜𝑠 𝜃 (4)

rin

𝛿 = 1 𝐷2 (5)

𝛽 𝑐𝑜𝑠 𝜃
𝜀= 4
(6)
ep

2𝑑ℎ𝑘𝑙𝑠𝑖𝑛 𝜃ℎ𝑘𝑙 = 𝑛𝜆 (7)

1
( )+ 𝑙2
Pr

4 ℎ2 + ℎ𝑘 + 𝑘2
𝑑2ℎ𝑘𝑙 =3 𝑐2
(8)
𝑎2

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 V=0.866a2c (9)

ed
2
a2
+ ( ― z) c2
1
L= (10)
3 2

iew
a2 1
z= 2 +4 (11)
3c

where λ is the wavelength of X-ray radiation, β is the full width at half maximum (FWHM in

radians) of the diffracted peak, and θ is the Bragg angle. Table 2 shows the calculated

v
structural parameters. These analyses show that the structural properties of the fabricated

re
ZnO layers depend on the initial precursor concentration. As seen, sample S0.05 has the

nearest a, c, and c/a values to those of the standard expected values of 3.2498 Å, 5.2066 Å,
er
and 1.6021, respectively (JCPDS 36-1451). According to these analyses, although this

sample also has the highest unit cell volume and Zn-O bond length, S0.03 has the lowest
pe
values. In addition to these analyses, a close look into (002) diffracted peaks in S0.03, S0.05,

and S0.07 samples (the inset) reveals that they are located at slightly different 2θ angles of

34.72, 34.45, and 34.68º, respectively. The reason for these variations could be explained
ot

considering the increment in the Zn-O bond length [32]. Also using eq. (8), taking the
tn

variation of d002-spacing values into account, S0.05 has the highest value and S0.03 has the

lowest value.
rin

3.3 Optical Properties

The optical reflectance spectra of the studied samples, together with the visible LEDs
ep

(blue, green, and red) spectra, are shown in Fig. 3a. As evident, the amount of reflection

through the measured wavelength range is low (less than 9%), which means the ZnO layers

are acting as antireflecting coatings. A close look at these reflectance spectra shows that there
Pr

is no particular order among them, i.e., S0.03/S0.05 has the highest/lowest value. The

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 different amount of reflection could be due to differences in the layers’ morphology,

ed
compactness, stoichiometry, thickness, and crystallinity. As mentioned in Section 3.4.2, a

decrease in reflectivity may result in an increase in light absorption, which in turn raises the

photodiode's photocurrent.

iew
In order to determine the optical band gap (Eg) of the layers, first we evaluate the optical

absorption coefficient (αλ) of the layers using the reflectance data (Rλ) [33]:

(1 ― 𝑅𝜆)2
(12)

v
𝛼𝜆 = 𝐹(𝑅) = 2𝑅𝜆

re
Then, we used the transformed Kubelka-Munk function [34]:

[𝐹(𝑅)ℎ𝑣]𝑚 = 𝐶(ℎ𝑣 ― 𝐸𝑔) (13)

er
where C is a constant, hv is the energy of the incident photon, and m is an index associated

with the process of optical absorption. Theoretically, for direct and indirect allowed
pe
transitions, respectively, m = 2 or 0.5. By extrapolating the straight section to the energy axis,

the samples' direct optical band gaps were calculated, i.e., (𝛼ℎ𝜈)2 = 0, Fig. 3b. According to

these results, S0.05 has the lowest band gap (3.27 eV), while S0.03 has the highest value
ot

(3.31 eV). These variations could be explained by the occurrence of the quantum-confined

effect. This phenomenon mainly originates from the differences in the average grain size,
tn

which in turn is related to the volume (V) of the unit cell and the values of the lattice

constants of the synthesised layers.

rin

We studied the photoluminescence (PL) spectra of the materials to fully investigate their

optical properties (Fig. 4a). The quality of a material's crystal structure can be determined
ep

using the PL spectrum, which can also be used to identify various structural defects.

According to the basic principles of photoluminescence theory, in an ideal direct band gap

semiconductor, a peak emission related to the recombination of excited electrons in the

Pr

conduction band and holes in the valance band, close to the band gap energy of the layer, is

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 expected (here, the sharp peaks correspond to the short UV range of 380–383 nm). In reality,

ed
due to the formation of intrinsic crystalline defects that occur through the growth process of

the samples, there is a possibility of the presence of energy levels within the band gap energy,

leading to emission peak(s) related to longer wavelengths (here those that appeared in the

iew
visible range of 400–700 nm).

For a more detailed analysis, among various samples, we have selected the PL spectrum

for S0.05 with the highest UV intensity peak and widest visible peaks. Fig. 4b illustrates the

v
deconvolutions of the PL spectrum in this sample. As evident from this spectrum, it could be

re
deconvoluted by four emission peaks: a prominent peak located at 382 nm (P1-UV), a peak at

420 nm (P2-violet), and two broad peaks, one around 570 nm (P3-yellow) and the other at
er
700 nm (P4-red). The UV peak could be attributed to the direct recombination of the free

excitons; the violet peak could be ascribed to the electron transition from the bottom of the
pe
conduction band (EC) to the zinc vacancy (VZn) level [35], and/or the electron transition from

interstitial zinc (Zni) level to the top of the valence band (EV) [36], well consistent with the

EDX analysis (Table 1); and the broad visible emissions, between 500 and 700 nm, could be
ot

related to the oxygen vacancies and surface states as the intrinsic defects’ levels within the

band gap energy [37,38]. Since the intensity of the emission peaks in the visible region is
tn

affected by the number of crystalline deficiencies, the quality of the synthesised layers can be

evaluated by finding the ratio of the intensity of the UV (P1) to the visible peak (mainly P3).
rin

Therefore, the larger this ratio is in a sample, the lower the density of crystalline defects in

that sample. Here in these samples, this ratio is: 5.31, 5.54, and 4.97, respectively, with the
ep

highest in S0.05.

3.4 I-V Characteristics

Pr

3.4.1 Dark Condition

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 Fig. 5 shows the room-temperature semi-logarithmic current-voltage (I-V) plots of the

ed
Ag/ZnO/p-Si/Al heterostructure samples in dark conditions. It is clear that all samples exhibit

rectifying behaviour, which is measured by the parameter called the rectification ratio (𝑅𝑅 =

𝐼𝐹

iew
𝐼𝑅), for a given applied forward and reverse bias voltage. The estimated values for this

parameter in the examined samples at ±5 V are shown in Table 3. As seen, the RR value of

the fabricated diodes increases with increasing the ZnO precursor concentration. The amount

v
of the RR could be considered a measure of the strength of the built-in potential at the

re
junction [39].

The presence of the ZnO layer between the metal and semiconductor in this structure

modifies the voltage across the diode. Thus, the thermionic emission (TE) theory could be
er
used to describe the diode I-V relationship in these devices [40]:

[ ((
𝐼 = 𝐼𝑠 exp
𝑞 𝑉 ― 𝐼𝑅𝑠)
) ] = 𝐴𝐴∗𝑇2𝑒𝑥𝑝( ― 𝑞𝜑𝑘𝑇 )[𝑒𝑥𝑝(𝑞(𝑉𝑛𝑘𝑇
𝐵 ― 𝐼𝑅 )
𝑠
)] (14)
pe
𝑛𝑘𝑇

where Is denotes the reverse saturation current, q is the electron charge, V is the applied

voltage, Rs is the series resistance, n represents the diode ideality factor, k is the Boltzmann
ot

constant, T is the absolute temperature, A is the diode area (= 0.265 cm2), A* is the

Richardson constant (= 32 A/K2cm2 for p-Si), and φB is the zero-bias potential barrier.
tn

According to eq. (14), the series resistance causes the voltage to be lower by an amount of

IRs. To simplify this equation, we have used the low voltage range (0.1 <V< 0.2 V), thus the
rin

low applied current region. In this region, the amount of the voltage drop (IRs) is negligible,

and eq. (14) reduces to:

ep

𝑞𝑉
𝐼 = 𝐼𝑠𝑒𝑥𝑝 (𝑛𝑘𝑇) (15)

Now, the parameters Is and n could be found from the y-intersect and the slope of the fitted
Pr

q dV
line of the semi-logarithmic forward bias I-V plot, i.e., n = kT d(lnI), respectively. The results

10

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 of these analyses are shown in Table 3. The noticeable points in these results could be

ed
classified as the following: (a) Each sample under study has an ideality factor greater than 1.

This indicates that the current mechanism deviates from the ideal thermionic emission theory.

This situation could be due to the presence of a thin interfacial native layer (SiO2) between

iew
the n-ZnO/p-Si interface, the effect of barrier inhomogeneities, interface states related to the

ZnO thin film/Si substrate lattice mismatch, and secondary current processes like generation,

recombination and tunnelling currents [31,41-43]; (b) These analyses indicate that as the zinc

v
precursor concentration increases, the ideality factor (n) of the samples decreases, and in turn,

re
the RR values increase; (c) Considering the values of reverse saturation current (Is), it is

notable that, compared to other samples, S0.05 has the minimum value. This could originate

from its large barrier height (0.91 eV).

er
3.4.2 Illumination Conditions
pe
To determine the photoresponse of the studied samples, the samples were exposed to 1

W power blue, green, and red LEDs. The spectra of the light sources are shown in Fig. 3a. In

order to study the details of the optoelectronic properties of the samples, we have investigated
ot

the current voltage (I-V), Fig. 6, and the current time (I-t), Fig. 7. According to I-V data,

under light illuminations, the electrical characteristics of the studied samples in reverse bias
tn

voltage are significantly different from dark conditions (~ 2 to 3 orders of magnitude) than in

forward bias (~ 2 to 8 times). These differences could originate from: (a) an increment in the
rin

depletion layer width and also in the internal electric field at the ZnO/Si junction in reverse

bias; (b) the weak leakage dark current (~ 1 µA) that passes through the samples in reverse
ep

bias compared to that in forward bias (~ 1 mA).

In order to describe the characteristics of a photodiode more precisely, we need

specifications such as response times, including the rise time (τR), which is the time interval
Pr

for the current to increase from 10 to 90% of the maximum photocurrent when the diode is

11

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 switched on by light; the decay time (τD), which is the time interval for the current to drop

ed
from 90% of its initial value to 10% when light is off; and the light sensitivity of the device 𝑆

( = (𝐼𝐿 ― 𝐼𝐷) 𝐼𝐷 = 𝐼𝑃 𝐼𝐷) at a given voltage, where IL, ID, and IP are the light, dark, and

photo currents, respectively. All this information can be obtained using the transient

iew
photoresponse, i.e., I-t measurement data (Fig. 7). Through these measurements, the applied

voltage is -5 V, and the light source is turned “off” and “on” in 20 s intervals. As evident, all

samples show sharp changes in the current while the light source is switched from “off” to

v
“on” and vice versa. Table 4 shows the results of these analyses for τR, τD, and S in these

re
devices. Based on the detailed analysis, we found the studied samples have quick response

times (~ 65 ms) to the “on-off” switching process. In a comparative study between these
er
results, the following points are noticeable: (a) Among the studied samples, S0.05 has the

maximum photocurrent (IP) and also has the highest light sensitivity (S) to various visible
pe
light illuminations. Due to the fact that in the n-ZnO/p-Si heterostructure, the visible light

absorption mainly takes place through the Si side at the junction, the variations for IP could

originate because of the amount of optical absorbance (A) of the constituent layers, which is
ot

related to the reflectivity (R) of the layers by the equation A+R = 1. A quick look at Fig. 3a

shows S0.05 has the lowest reflectance compared to other samples and, therefore, the highest
tn

light absorbance, leading to the highest photocarrier generation. This issue is true for other

samples as well; (b) With increasing the wavelengths of the incident light (i.e., from blue to
rin

red), the amount of the corresponding photocurrent has decreased. It is known that, as the

photon flux (∅) travels through a semiconductor, the photon flux decreases due to the
ep

absorption process according to ∅(𝑥) = ∅0𝑒―𝛼𝑥, where ∅0 is the photon flux at 𝑥 = 0, and α is

the absorption coefficient, which is a function of incident light wavelength. Fig. 7d shows the
Pr

wavelength dependence of the absorption coefficient of Si [44]. It is noticeable that 63% of

the optical flux will be absorbed over a distance W = 1/α, called the penetration depth, which

12

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 is also shown in Fig. 7d. In a photodiode, only a part of the photons that are absorbed in the

ed
active region (the depletion region and that of the electrons’ diffusion length, defined as the

average distance away from the boundary of the depletion region) can contribute to the

photocurrent of the device. Since the carriers generated outside the depletion region must

iew
diffuse to the junction, resulting in considerable time delay, to minimise the diffusion effect,

i.e., the response times, the incident light should be absorbed very close to the junction,

mainly inside the depletion region [44]. According to the plot shown in Fig. 7d, the

v
penetration depths for our desired blue, green, and red lights are about 0.4, 0.9, and 2.1 µm.

re
Thus, it is expected that the photo-carriers generated by the absorption of the blue light are

mostly within the depletion region, while the photo-carriers related to the absorption of the
er
green and red lights could be partly outside this region, leading to a lower photocurrent.

4. Conclusions
pe
Using different precursor concentrations on a p-Si substrate, we first investigated the

physical properties of ZnO thin layers prepared by spin coating for the seed layer and then the

chemical bath deposition (CBD) method for the formation of ZnO-NRs. The FESEM images
ot

of the ZnO layers showed that the deposited layers consist of a continuous bottom layer

related to the seed layer, on which there are relatively long (700–1200 nm) and compact
tn

ZnO-NRs. Also, the EDX results showed that the contribution of O-atoms is higher than

50%, while it is lower for Zn-atoms. The XRD diffractograms of the samples confirmed that
rin

the synthesised ZnO layers have a polycrystalline structure with a hexagonal (wurtzite) phase

mainly grown along the (002) direction. The optical reflectance spectra of the samples
ep

showed that the deposition of the ZnO-NRs on the p-Si substrate acted as antireflecting

coatings and in S0.05 it reduced the visible light reflectance to less than ~ 2 %. The Kubelka-

Munk relation was utilised to analyse the data, and the results showed that the band gap
Pr

energy for these layers is between 3.27 and 3.31 eV. These values are in good agreement with

13

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 the wavelength of the corresponding PL peak intensity. Next, we discussed how the precursor

ed
concentration affected the I-V rectifying behaviour of Ag/n-ZnO-NRs/p-Si/Al heterostructure

photodiodes under different visible light illumination circumstances (1 W blue, green, and red

LEDs) as well as in the dark, using room-temperature I-V and I-t characteristics of the

iew
samples. All of the examined samples were found to have a rectifying characteristic, with

rectification ratios ranging from 334 to 851. The analysis of the dark I-V data, based on the

thermionic emission theory, revealed that with increasing the precursor concentration, the

v
ideality factors of the samples decreased while their rectification ratios increased. The

re
analysis of I-t data, operated at -5 V, showed that all the fabricated samples have a fast-

switching characteristic (≈ 65 ms). Due to the differences in the morphology, compactness,

er
stoichiometry, thickness, and crystallinity of the ZnO-NRs layers on the Si substrate, the

studied samples show different amounts of visible light reflectance. Based on this feature, we
pe
found that S0.05, with the lowest light reflectance, has the maximum light sensitivity

(S=IP/ID) to various visible light illuminations (460 for blue, 201 for green, and 190 for red).
ot

Declarations of interest: none

tn

This research did not receive any specific grant from funding agencies in the public, commercial,

or not-for-profit sectors.
rin

References

1. L. Peng, L. Hu, X. Fang, Energy harvesting for nanostructured self‐powered

ep

photodetectors, Adv. Funct. Mater., 24 (2014) 2591-610.

Pr

2. Z. Bai, X. Yan, X. Chen, H. Liu, Y. Shen, Y. Zhang, ZnO nanowire array ultraviolet

photodetectors with self-powered properties, Curr. Appl. Phys., 13 (2013) 165-69.

14

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 3. S.-J. Young, C.-C. Yang, L.-T. Lai, Growth of Al-, Ga-, and In-doped ZnO nanostructures

ed
via a low-temperature process and their application to field emission devices and ultraviolet

photosensors, J. Electrochem. Soc., 164 (2016) B3013.

iew
4. B. Lin, Z. Fu, Y. Jia, Green luminescent center in undoped zinc oxide films deposited on

silicon substrates, Appl. Phys. Lett., 79 (2001) 943-45.

5. Ü. Özgür, Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Doğan, V. Avrutin, S.-J.

v
Cho, H. Morkoç, A comprehensive review of ZnO materials and devices, J. Appl. Phys., 98

re
(2005) 041301.

6. Z. Zang, A. Nakamura, J. Temmyo, Single cuprous oxide films synthesized by radical

er
oxidation at low temperature for PV application, Opt. Express, 21 (2013) 11448-56.

7. C. Li, C. Han, Y. Zhang, Z. Zang, M. Wang, X. Tang, J. Du, Enhanced photoresponse of

pe
self-powered perovskite photodetector based on ZnO nanoparticles decorated CsPbBr3 films,

Sol. Energy Mater. Sol. Cells, 172 (2017) 341-46.

ot

8. Z. Zang, Efficiency enhancement of ZnO/Cu2O solar cells with well oriented and

micrometer grain sized Cu2O films, Appl. Phys. Lett., 112 (2018) 042106.
tn

9. P. Dhamodharan, C. Manoharan, M. Bououdina, Tuning the properties of ZnO thin film on

ITO substrates with ga dopant for dye sensitized solar cell applications, J. Mater. Sci., 29
rin

(2018) 12477-88.

10. Y. Waghadkar, M. Shinde, S. Rane, S. Gosavi, C. Terashima, A. Fujishima, R. Chauhan,

ep

Highly efficient dye-sensitized solar cells by TiCl4 surface modification of ZnO nano-flower

thin film, J. Solid State Electrochem., 22 (2018) 3621-30.

Pr

15

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 11. J. Cao, B. Wu, R. Chen, Y. Wu, Y. Hui, B. W. Mao, N. Zheng, Efficient, hysteresis‐free,

ed
and stable perovskite solar cells with ZnO as electron‐transport layer: Effect of surface

passivation, Adv. Mater., 30 (2018) 1705596.

iew
12. T. Marimuthu, N. Anandhan, T. Mahalingam, R. Thangamuthu, M. Mummoorthi, Effect

of P. Murex on the properties of spin coated ZnO thin films for dye sensitized solar cell

applications, J. Mater. Sci., 26 (2015) 7577-87.

v
13. R. Vittal, K.-C. Ho, Zinc oxide based dye-sensitized solar cells: A review, Renew. Sust.

re
Energ. Rev., 70 (2017) 920-35.

14. M. Rajabi, R. Dariani, F. Zahedi, Optoelectronic properties of cauliflower like ZnO–ZnO

er
nanorod/p-Si heterostructure, Solid-State Electron., 80 (2013) 33-37.

15. R. Yatskiv, J. Grym, M. Verde, Graphite/ZnO nanorods junction for ultraviolet

pe
photodetectors, Solid-State Electron., 105 (2015) 70-73.

16. J.-H. Choi, S. N. Das, K.-J. Moon, J. P. Kar, J.-M. Myoung, Fabrication and
ot

characterization of p-Si nanowires/ ZnO film heterojunction diode, Solid-State Electron., 54

(2010) 1582-85.
tn

17. P. Fallahazad, N. Naderi, M.J. Eshraghi, A. Massoudi, Combination of surface texturing

and nanostructure coating for reduction of light reflection in ZnO/Si heterojunction thin film
rin

solar cell, J. Mater. Sci., 29 (2018) 6289-96.

18. C. Betty, K. Sehra, K. Barick, S. Choudhury, Facile preparation of Silicon/ZnO thin film
ep

heterostructures and ultrasensitive toxic gas sensing at room temperature: Substrate

dependence on specificity, Anal. Chim. Acta, 1039 (2018) 82-90.

Pr

16

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 19. L. Znaidi, T. Chauveau, A. Tallaire, F. Liu, M. Rahmani, V. Bockelee, D. Vrel, P.

ed
Doppelt, Textured ZnO thin films by sol–gel process: Synthesis and characterizations, Thin

Solid Films, 617 (2016) 156-60.

iew
20. L.-Y. Chen, W.-H. Chen, J.-J. Wang, F. C.-N. Hong, Y.-K. Su, Hydrogen-doped high

conductivity ZnO films deposited by radio-frequency magnetron sputtering, Appl. Phys.

Lett., 85 (2004) 5628-30.

v
21. R. Ondo-Ndong, H. Omanda, H. Gnanga, and H. Moussambi, Effect of sputter pressure

re
on zinc oxide thin films deposed by RF magnetron sputtering, Int. J. Innov. Appl. Stud., 24

(2018) 1284-98.
er
22. G. Regmi, and S. Velumani, Impact of target power on the properties of sputtered

intrinsic zinc oxide (i-ZnO) thin films and its thickness dependence performance on CISe
pe
solar cells, Opt. Mater., 119 (2021) 111350.

23. E.C. Greyson, Y. Babayan, T.W. Odom, Directed growth of ordered arrays of
ot

small‐diameter ZnO nanowires, Adv. Mater., 16 (2004) 1348-52.

24. P. Gu, X. Zhu, D. Yang, Vertically aligned ZnO nanorods arrays grown by chemical bath
tn

deposition for ultraviolet photodetectors with high response performance, J. Alloys Compd.,

815 (2020) 152346.

rin

25. H. HajiGhasemi, H. Eshghi, Ultraviolet light sensing property in ZnO nanorod arrays

doped by various elements, Mater. Sci. Semicond. Process., 144 (2022) 106611.
ep

26. L. Chabane, N. Zebbar, M. Kechouane, M. Aida, M. Trari, Al-doped and In-doped ZnO

thin films in heterojunctions with silicon, Thin Solid Films, 605 (2016) 57-63.
Pr

17

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 27. N.S. Koc, S. Altintas, M. Gokcen, M. Dogruer, C. Altug, A. Varilci, Current-voltage

ed
characteristics of nano whisker ZnO/Si heterojunction under UV exposition, Sens. Actuators

A Phys., 342 (2022) 113618.

iew
28. P. Sahare, S. Kumar, S. Kumar, F. Singh, N-ZnO/p-Si heterojunction nanodiodes based

sensor for monitoring UV radiation., Sens. Actuators A Phys., 279 (2018) 351-60.

29. M. Monshipouri, S. Molavi, A. Mosaddegh, M. Sasar, Y. Abdi, Enhancement of

v
responsivity and sensitivity of p-silicon/n-zinc oxide-based photodetector using titanium

re
dioxide nanoparticles, IEEE Trans. Nanotechnol., 19 (2020) 744-48.

30. J.-D. Hwang, M.-C. Lin, ZnO Hole Blocking Layer Induced Highly UV Responsive P-
er
NiO/N-ZnO/N-Si Heterojunction Photodiodes, Sens. Actuators A Phys., 349 (2023) 114087.

31. K. Karthick, P. Kathirvel, R. Marnadu, S. Chakravarty, M. Shkir, Ultrafast one step direct
pe
injection flame synthesis of zinc oxide nanoparticles and fabrication of p-Si/n-ZnO

photodiode and characterization, Physica B: Condens., 612 (2021) 412971.

ot

32. A. Othman, M. A. Ali, E. Ibrahim, M. Osman, Influence of Cu doping on structural,

morphological, photoluminescence, and electrical properties of ZnO nanostructures

tn

synthesized by ice-bath assisted sonochemical method, J. Alloys Compd., 683 (2016) 399-

411.
rin

33. H. Eshghi, M.T. Goodarzi, Synthesis of CuO nanowires on Cu-foil using thermal

oxidation method, a novel annealing process, Mod. Phys. Lett. B, 30 (2016) 1650039.
ep

34. P. Jubu, F. Yam, V. Igba, K. Beh, Tauc-plot scale and extrapolation effect on bandgap

estimation from UV–Vis–NIR data–a case study of β-Ga2O3, J. Solid State Chem., 290
Pr

(2020) 121576.

18

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 35. G.H. Mhlongo, D.E. Motaung, S.S. Nkosi, H. Swart, G.F. Malgas, K.T. Hillie, B.W.

ed
Mwakikunga, Temperature-dependence on the structural, optical, and paramagnetic

properties of ZnO nanostructures, Appl. Surf. Sci., 293 (2014) 62-70.

iew
36. A. Sreedhar, J.H. Kwon, J. Yi, J.S. Kim, J.S. Gwag, Enhanced photoluminescence

properties of Cu-doped ZnO thin films deposited by simultaneous RF and DC magnetron

sputtering, Mater. Sci. Semicond. Process., 49 (2016) 8-14.

v
37. X. Zhang, Y. Liu, J. Zhang, Y. Lu, D. Shen, X. Fan, X. Kong, Structure and

re
photoluminescence of Mn-passivated nanocrystalline ZnO thin films, J. Cryst. Growth, 254

(2003) 80-85.
er
38. R. Bahramian, A. Moshaii, H. Eshghi, Effect of seeding modification of substrate on the

growth and UV detection properties of ZnO nanowires, Mater. Lett., 179 (2016) 222-25.
pe
39. S. Hamrouni, M. AlKhalifah, M. El-Bana, S. Zobaidi, S. Belgacem, Deposition and

characterization of spin-coated n-type ZnO thin film for potential window layer of solar cell,
ot

Appl. Phys. A Mater. Sci. Process., 124 (2018) 1-11.

40. E.H. Rhoderick, R.H. Williams, Metal-semiconductor contacts, Vol. 129: Clarendon
tn

press Oxford, 1988.

41. N. Tuğluoğlu, S. Karadeniz, Ş. Altındal, Effect of series resistance on the performance of

rin

silicon Schottky diode in the presence of tin oxide layer, Appl. Surf. Sci., 239 (2005) 481-89.

42. H. Card, and E. Rhoderick, Studies of Tunnel MOS Diodes I. Interface Effects in Silicon
ep

Schottky Diodes, J. Phys. D Appl. Phys., 4 (1971) 1589-1592.

43. R.T. Tung, Recent advances in Schottky barrier concepts, Mater. Sci. Eng. R Rep., 35
Pr

(2001) 1-138.

19

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 44. S.M. Sze, Semiconductor Devices: Physics and Technology, John Wiley & Sons, (2008).

ed
iew
Figure Captions
Fig. 1: The top and cross-section views of the FESEM images and EDX patterns of the

v
studied ZnO layers grown on p-Si substrate.

re
Fig. 2: (a) The XRD patterns of the seed layer and S0.03 for a comparison between their peak

intensity; (b) The corresponding patterns for the ZnO-NRs arrays of the studied samples with
er
different precursor concentrations. The inset shows a closer look of the (002) peak within the

34.0–35.0º range.
pe

Fig. 3: (a) The reflectance spectra (Rλ) of the samples with different precursor concentrations,

together with the illuminated visible light LEDs’ spectra; (b) the detailed evaluation of the
ot

band gap of the samples, based on the Kubelka-Munk method.

Fig. 4: (a) The PL spectra of the studied samples; (b) the deconvoluted PL spectrum of
tn

sample S0.05.

Fig. 5: The semi-log I-V characteristics of the studied samples in the dark and at room
rin

temperature.

Fig. 6: The semi-log I-V characteristics of (a) S0.03, (b) S0.05, and (c) S0.07.
ep

Fig. 7: (a-c) The transient response (I-t) of the studied samples in the dark and under visible

(blue, green, and red) LEDs illuminations; (d) Light absorption coefficient and penetration

depth vs. wavelength in Si [44].

Pr

20

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
iew
Figures

v
re
Fig. 1

er
pe
ot
tn
rin
ep
Pr

21

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
iew
Fig. 2

v
re
er
pe
ot
tn
rin
ep
Pr

22

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
iew
Fig. 3

v
re
er
pe
ot
tn
rin
ep
Pr

23

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot

Fig. 4
tn
rin
ep
Pr

24

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot

Fig. 5
tn
rin
ep
Pr

25

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot

Fig. 6
tn
rin
ep
Pr

26

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot
tn

Fig. 7
rin
ep
Pr

27

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot
tn

Tables
rin

Table 1: The detailed geometrical characteristics and EDX analysis data of the studied ZnO-
NRs layers.

Average diameter Average length

Sample Aspect ratio
ep

(nm) (nm)
S0.03 70 730 10.4
S0.05 80 1200 15.0
S0.07 85 700 8.2
Pr

28

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 ed
v iew
re
er
pe
ot
tn
rin

Table 2: Various crystallite parameters in the studied samples, including crystalline size (D),

micro-strains (ε), dislocation density (δ), volume of the unit cell (V), Zn-O bond length (L),

interplaner spacing (d002), and lattice parameters (a, c).

ep

2θ(002) D ε δ (×10-3) V L d(002) Lattice parameters

Sample
(degree) (nm) (×10-3) (nm-2) (Å3) (Å) (Å) a (Å) c (Å) c/a
Pr

S0.03 34.7200 25.15 1.37 1.58 46.7426 1.9651 2.5828 3.2325 5.1656 1.5980

29

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 S0.05 34.4780 24.91 1.39 1.61 47.5585 1.9769 2.6011 3.2491 5.2022 1.6011

ed
S0.07 34.6400 27.38 1.26 1.33 46.9470 1.9683 2.5880 3.2363 5.1760 1.5993

v iew
re
er
pe
ot
tn
rin

Table 3: The diode parameters for the n-ZnO/p-Si heterostructures evaluated from the dark I-

V characteristics of the studied samples.

ep

IR (µA) IF (µA)
Sample RR n Is (nA) φB (eV)
@ -5 V @ +5 V

S0.03 1.4 468 334 3.32 58.28 0.78

Pr

S0.05 2.0 1350 675 2.10 0.44 0.91

30

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 S0.07 1.5 1277 851 2.08 10.01 0.83

ed
v iew
re
er
pe
ot
tn
rin

Table 4: Performances of the fabricated Ag/n-ZnO/p-Si/Al photodetectors to various visible

ep

light illuminations.

Sample Color ID (µA) IL (µA) IP (µA) S τR (ms) τD (ms)

Pr

S0.03 Blue 1.4 480 478.5 342 63 64

31

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104
 Green 250 248.6 178 64 64

ed
Red 118 116.6 83 64 64

Blue 922 918 460 64 65

S0.05 Green 2.0 404 402 201 64 65

iew
Red 382 380 190 64 64

Blue 660 658.5 439 64 57

S0.07 Green 1.5 275 273.5 182 64 65

Red 127 125.5 84 66 64

v
re
er
pe
ot
tn
rin
ep
Pr

32

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4679104