Mechanistic exploration of charge transport and photosensitivity of metal/polymer/semiconductor (MPS) junction for sensitive light detection applications

Figure 2a–c reveals the microstructure of the prepared nanocomposite thin films at different magnifications. It can be observed that the film Ni(OAc)₂ has been formed in the form of nanoplatelets uniformly distributed in the PVA: PVP matrix. The distribution of the nanoplatelets size depicted in the inset of Fig. 2a illustrates that the mean value of the nanoplatelets size is about 79 ± 15 nm. The formation of these nanostructured Ni(OAc)₂ nanoplatelets may be supported by the interactive impacts of the PVA and PVP matrices. In addition, the halo featureless SAED pattern that is shown in Fig. 2d confirms the dominance of the amorphous nature of the prepared nanocomposite film.

On the other hand, it would be unreasonable not to inspect the surface morphology of the deposited interlayer, since it is well known that the interlayer film’s roughness has many consequences on the performance of the designed MPS device. Therefore, the surface topography of the prepared PVA: PVP@Ni(OAc)₂ thin film has been inspected as visualized in the FE-SEM micrograph shown in Fig. 3a. The observed nanoplatelets in the HR-TEM micrographs are also detected in the FE-SEM images. The micrograph reveals the smoothness and homogeneity of the surface. Additionally, the 3D FE-SEM micrograph reveals the roughness profile of the spin-coated nanocomposite film is declared in Fig. 3b. It can be observed that the film’s surface is relatively smooth with an average roughness of about 7.08 ± 2.3 nm. However, these nanoplatelets may occupy a vertical position relative to the film’s plane, which has caused them to appear clearly in some areas, resulting in a relatively high roughness.

The UV–Vis-NIR transmittance, reflectance, and absorbance spectra of the spin-coated PVA: PVP@Ni(OAc)₂ thin films are displayed as a function of photon energy as shown in Fig. 4a–c, respectively. The prepared films exhibited high NIR and visible transparency with a notable weak absorption peak at about 3.1 eV (400 nm) which is related according to the spin–orbit coupling to the electronic triplet–triplet d-d transitions ³A_2g(f) → ³T_2g(p) of Ni(OAc)₂ complex [50‐52]. Meanwhile, in the UV region, the transparency drops sharply at a fundamental edge yielding null transmittance consistently with a wide sharp absorption band which is centered at 5.49 eV. The onset of the absorption, i.e., the fundamental absorption edge and its wavelength is determined by the band-gap of the PVA: PVP@Ni(OAc)₂ composite film. The optical absorption in the UV range is usually attributed to the interaction of light with the bound electrons in the polymer composite. It is also noticed that the reflectance of the prepared film is weak overall in the spectral range as shown in Fig. 4b. This is an efficient result in increasing the light-harvesting capability of the device for more efficient optoelectronic performance especially when exploiting this film as an interlayer photosensitive device.

The absorption coefficient as a function of energy, α (E) of the prepared nanocomposite film of thickness, d, has been calculated using the following relation [53]:

Figure 4d depicts the photon energy dependence of the calculated absorption coefficient. A clear wide intense band absorption band has been observed and resolved into three components at 5.39 eV which corresponds to the electronic transition π → π^* transition of syndiotactic PVA [54]. The resolved absorption peaks at 5.58 and 6.17 eV are related to the n → π^* and n → σ^* transitions of C = O and C = C groups of PVP [55].

It is worth noting that accurate determination of the energy gap of the interlayer in the MPS junction has a pivotal role in how it interacts with the light incident on it in terms of absorption and scattering and about the energetic alignment of its molecular orbitals (i.e., HOMO and LUMO) relative to the Fermi energy of the metal and the valence band and conduction band of the semiconductor. Accordingly, various methods and approaches have been used to determine the energy gap of the prepared nanocomposite films: It is anticipated that the derivative of transmittance, dT/dλ, and reflectance, dR/dλ, with respect to wavelength will diverge and tend to infinity and will have peaks near λ_g = hc/E_g, as revealed in Fig. 5a following the present relation [53, 56, 57]:

The observed peak of energy E₂ is related to the fundamental energy gap and the second E₁ is related to the sub-bandgap developed owing to the incorporation of Ni(OAc)₂ in the electronic structure of the PVA: PVP matrix. Considering the dependence of the absorption coefficient α(E) on the photon energy, E, follows the following relation [53]:where C is a constant, and n_r(E) is the energy-dependent refractive index. Meanwhile, the transmittance of such a film of thickness, d, can be represented as follows [53]:

It can be shown that when the incident photon energy approaches the energy gap, the derivative of reflectance dR/dE remains finite at energies near E_g. This occurs because a band tail prevents singularity-type behavior from occurring. On the other hand, the derivative of transmittance dT/dE exhibits a spike toward negative infinity as illustrated in the following equation [53]:

Likewise, as depicted in Fig. 5b it is shown that when absorption tails are present, they attenuate the divergence and create distinct peaks around the gap energy. Hence, two clear transition energies are constituents with the aforementioned energies related to the fundamental and intermediate transition energy gap. Referring to the Tauc method, the photon energy dependence of the absorption coefficient is given by [53, 58]:

The natural logarithmic for both sides will yield:

By differentiating Eq. (7) with respect to photon energy E:

Therefore, the plot of d[ln(αE)]/dE versus photon energy, E, will result in the appearance of a particular peak which is related to the energy gap as illustrated in Fig. 5c. In addition, the type of the electronic transition can be estimated employing Eq. (7), by plotting the relation between ln(αE) vs ln(E-E_g), where m equals the slope of the linear fit as observed in Fig. 5d. The obtained value of m was about 0.49 (~ 0.5) leading to a direct transition.

Current–voltage (I-V) measurements under forward and reverse bias voltages are considered a powerful approach for exploring the electronic anatomy of the designed Schottky diode and providing information about the diode’s characteristics such as its barrier height, Φ_B, ideality factor, n, rectification ratio, RR, and reverse saturation current, I_s. Figure 6a and b illustrate the effective role of the spin-coated nanocomposite interlayer film of PVA: PVP@Ni(OAc)₂ on the performance of both Ag/p-Si and Ag/n-Si, respectively. Customary semilogarithmic I-V relations of the Schottky diode under ± 3 V sweep voltage range are demonstrated. The deviation of the current performance from the linearity in the high voltage region is ascribed to the role of the series resistance, R_s [59]. It can be observed that the insertion of the PVA: PVP@Ni(OAc)₂ interlayer suppressed the reverse dark current significantly by ~ 78% at -3 V and enhanced the forward current of the Ag/p-Si by ~ 30% at + 3V. Meanwhile, both forward and reverse currents of the Ag/n-Si have been suppressed upon inserting the polymeric nanocomposite interlayer, especially in the high bias regions >  ± 1 V. Furthermore, the (I-V) curves clearly show that the fabricated devices possess remarkable rectifying properties, especially when p-type Si is utilized, as it is demonstrated when calculating the rectifying ratio, RR, as depicted in Fig. 7.

Following the assumptions of Thermionic emission theory (TE) of Richardson and Dushman, the microelectronic parameters of the fabricated MS and MPS Schottky diodes (SDs) that govern the flow of current through the device can be estimated via fitting the low bias forward bias (I-V) data at (V ≥ 3k_BT/q) by TE relation, where the voltage-dependent diode’s current can be expressed as a function of temperature, T, as follows [10, 36]:where I_s is the saturation current, q is the electron charge, V_D = (V-IR_s) is the diode voltage drop, V is the applied voltage, R_s is the series resistance, k_B is Boltzmann constant, and n is the ideality factor which is a parameter that quantifies how closely a diode adheres to ideal behavior. The ideality factor can be determined by analyzing the slope of the linear part of the lnI-V plot, using the following relationship [10, 36]:

Besides, the reverse saturation current which refers to the current that flows across a diode in the absence of an applied voltage, due to the thermal generation of minority carriers, can be extracted from the intercept of the linear part of the lnI-V plot at zero bias and can be given by [10, 36]:where A is the device’s effective area, A^* is the effective Richardson constant and Φ_B0 is the zero-bias barrier height. Table 1 displays the calculated values for the electrical parameters I_s, n, and Φ_B0, R_s, utilizing p-type Si and n-type Si for both MS and MPS structures. The nonideality of MS and MPS devices in the case of p-Si (i.e., n > one) may be attributed to several factors including series resistance (R_s), heterogeneous barriers, tunneling current, the presence of interlayers, and the density of surface states [36, 39, 42, 59]. Meanwhile, in the case of n-Si, the fabricated devices almost exhibit ideal behavior with n ~ 1. The ideality deviation of MPS (p-Si) rather than other MPS of (n-Si) may be ascribed to the large Fermi level pinning in the case of p-Si rather than n-Si which affects the charge accumulation in the depletion region causing a recombination of charge carriers yielding a nonideal behavior. For p-type Si with an added polymeric interlayer, both the barrier height and ideality factor are observed to decrease. A decrease in the depletion region, caused by doping and leading to an increase in the density of charge carriers and an increase in surface states between the interfacial polymeric layer and semiconductor., is primarily responsible for the reduction observed in the ideality factor and barrier height [60‐62]. The reason why MPS-type structures exhibit a higher Φ_B0 compared to MS structures utilizing n-Si can be attributed to an increase in negative charges at the interface. These negative charges are likely generated by electron traps that are localized at the semiconductor interlayer [63, 64].

Besides, owing to the effectiveness of both parasitic series resistance, R_s, and shunt resistance, R_sh, on the electrical performance of the fabricated device, it was crucial to estimate their values. A diode with a high series resistance will experience a decrease in direct current, deteriorating its efficiency. Conversely, a diode with a high shunt resistance will experience a decrease in leakage current, increasing rectification [65]. The estimated values of both R_s and R_sh from the maximum and minimum values of junction resistance, R_j, under reverse and forward biasing cases, respectively, have been listed in Table 1. As observed both R_s and R_sh have been increased by incorporating the (Ni(OAc)₂) interlayer at the Ag/Si interface.

The modified Norde method is also a widely used approach for measuring the barrier height, Φ_B0, and the series resistance, R_s, of metal-polymer-semiconductor (MPS) structures which can be proposed as [37, 42, 66]:where F(V) is defined as the derivative of the current–voltage (I-V) curve of the Schottky barrier diode, β = q/k_BT, and the parameter γ is an integer without units, and it must be selected such that it is greater than the ideality factor n. In the Norde method, the values of Φ_B0 and R_s are obtained using Eqs. (13a) and (13b), respectively.where F(V_min) is the minimum value of the F(V)-V plot which is obtained by plotting F(V) versus voltage (V) as depicted in Fig. 8a and b for p-Si, and n-Si, respectively, while V_min and I_min are the corresponding voltage and current values, respectively. The obtained values of R_s and Φ_B0 are recorded in Table 1. The little observed differences between the estimated barrier height values from TE and modified Norde’s method may be ascribed to the voltage dependence nature of barrier height, while the significant discrepancy in series resistance values is attributed to the difference of the utilized biasing region, where Norde is applied for the whole forward bias region, while TE is applied in the linear low forward bias region only [20, 37, 67].

The Cheung and Cheung functions can be utilized as an alternative approach to determine values for R_s, n, and Φ_B0. These parameters can be calculated by applying the following relationships [65, 68]:

The curves for dV/dln(I)–I and H(I)–I for MS and MPS are presented in Fig. 9, revealing the behavior of p-type Si in (a) and (c), and n-type Si in (b) and (d). The values for R_s and n were derived from the slope and intercept, respectively, of the dV/d(lnI)-I plot using Eq. (14a) for the given structures. Likewise, the values for R_s and Φ_B0 were obtained from the slope and intercept, respectively, of the H(I)-I plot using Eq. (14a). These experimental values have been compiled and are presented in Table 1. As shown, there is a significant variance in the aforementioned parameters, highlighting the fact that their values are shaped not only by surface states, N_ss, and the interfacial layer but also by the voltage applied [69].

The mechanistic approach at which the charge carriers transport depends on the bias whether it is forward or reverse. Under forward biasing, the conduction mechanism of charge carriers within the designed devices has been evaluated via the double logarithmic relation of the current–voltage relation shown in Fig. 10a and b. As observed, each diode possesses several linear regions of different slope values yielding different conduction mechanisms. The slope of the low bias region (V_F < 0.2 V) in all the fabricated diodes ~ 1 ensures the dominance of the ohmic conduction mechanism, where net current arises from the thermally generated electron–hole (e–h) pairs regardless of the charge injection contribution. Meanwhile, in the second region (0.2 V < V_F < 1.1 V) the linear I-V relation has been converted to the power law of slope ~ 1.93 and 2.82 for Ag/p-Si (D1) and Ag/n-Si (D3), respectively. This gives rise to single dominating trap SCLC and trap charge limited current (TCLC) of exponential distribution of trap states for D1 and D3, respectively. The space charge limited current mechanism depends on the trapping and accumulation of the injected charge carriers at the junction, hence upon varying the applied bias the electric field profile charges and thereafter changes the drifting component accumulated charges [70‐72]. These trapping states may arise from intrinsic microstructure disorder and defects or due to interfacing imperfections. In the case of D1, at region (1.1 V < V_F < 3 V) the slope ~ 2.52 indicates that all the existing traps have been filled and now the conduction mechanism is a trap-free SCLC. The low slope ~ 0.13 of D3 at region (V_F > 2.5 V) is ascribed to the limited software of the employed electrometer. Upon inserting the nanocomposite film as an interlayer, the density of interface trapping states has been increased yielding continuous exponential trapping states assist SCLC over the whole bias range as observed from the estimated slopes 1.40 and 1.45 for D2 and D4, respectively [27, 28, 73]. Besides, it is well known that charge carriers in polymer nanocomposites prefer to transport via hoping between the localized trap states, thereby the charges hop between the localized trap states that are developed by the doping process or by structural defects [38].

The charge transport mechanism under the reverse bias of the designed MS and MPS can be interpreted in the light of Schottky or Pool–Frenkel emission mechanisms. Schottky emission is an electrode-limited mechanism that results when the electron overcomes the interfacial barrier after gaining enough thermal energy to be injected from/to the metal to/from the conduction band p-Si/n-Si passing by the dielectric layer. The lowering of potential barrier height at the metal/dielectric interface arises from the image force effect [70, 71]. Accordingly, the voltage dependence of the reverse current in terms of Schottky emission is given by [70, 71]:where ε_i and ε_o are the electric permittivities of insulating material and free space, respectively, while E is the electric field.

In contrast, Poole–Frenkel is a bulk-limited mechanism that depends mainly on the electrical properties of the dielectric itself. Poole–Frenkel involves the thermal emission of the trapped charge carriers (de-trapping) from the localized trapping states into the conduction band owing to the lowering of the coulombic energy barrier by the effect of the local electric field within the depletion region. Hence, the Poole–Frenkel current can be expressed as follows [70, 71, 74]:where μ is the electronic drift mobility, Φ_T is the trap barrier, and N_C is the conduction band density of states. Thereby, the relation between V ^1/2 and lnI_R has been plotted for all the fabricated MPS diodes as revealed in Fig. 11a and b. Depending on the estimated slopes of the illustrated linear fitting, the obtained values of the lowering factor for D2 were about 7.69 × 10^–5 eV. m^1/2.V^1/2 and 2.84 × 10^–5 eV. m^1/2.V^1/2 for the low and high bias regions, while the obtained values of the lowering factor for D4 were about 1.27 × 10^–4 eV. m^1/2.V^1/2 and 4.94 × 10^–5 eV. m^1/2.V^1/2 for the low and high bias regions, respectively. Theoretically, the estimated values of Poole–Frenkel and Schottky lowering factor (\(= \sqrt {q^{3} /\eta \pi \varepsilon_{i} \varepsilon_{o} }\), η = 1 and 4 for SC and PF, respectively) are about 3.65 × 10^–5 eV. m^1/2.V^1/2 and 1.82 × 10^–5 eV. m^1/2.V^1/2, respectively. These findings suggest the dominance of the Schottky emission mechanism in the low bias region (0.85 < V_R < 0) V for p-Si and (0.80 < V_R < 0) V for n-Si MPS devices, respectively. In addition to the dominance of the Poole–Frenkel mechanism in the high bias region (3.00 < V_R < 0.85) V for p-Si and (3.00 < V_R < 0.80) V for n-Si MPS devices, respectively. The discrepancy between the calculated and experimental values of lowering factors may arise from the small width of the depletion region in PVA: PVP@(NiOAc)₂ film or because of the phonon-assisted tunneling mechanism of charges through the energy barrier [74, 75].

To strengthen the significant role of the interlayered PVA: PVP@(NiOAc)₂ film on the performance of the fabricated MPS, the energetic interface states density profile with respect to the top of valance band and the minimum of conduction band for p-Si and n-Si devices, respectively, should be estimated. Hence, according to the variation of effective barrier height with respect to voltage, dΦ_e/dV, with ideality factor as a function of voltage, n (V) [75‐79]:

The effective barrier height, Φ_e, can be deduced from the estimated zero-bias barrier height, Φ_B, as follows [75‐79]:

Considering the high thickness of the embedded interlayer of thickness, d, the interface states density, N_ss, can be expressed as follows [75‐79]:

The energetic position of the interface states density with respect to the top valance band, E_ss–E_v, or with respect to the bottom of the conduction band, E_C–E_ss, for p-Si and n-Si based devices, respectively, can be expressed as follows [75‐79]:

Therefore, the energetic profiles of the estimated interface states density of the fabricated junctions have been plotted as shown in Fig. 12a and b. It can be observed that the insertion of nanocomposite film increased the interface states density significantly for both types of diodes. Furthermore, the energetic position of the interface states has been shortened. Since it can be noticed that after inserting the nanocomposite thin film into the Ag/p-Si/Al device, the distribution of interface states density has been limited to E_ss–E_v in the range (0.50–0.66) eV, while regarding the n-Si device, the distribution of interface states density has been limited to E_v–E_ss in the range (0.54–0.63) eV. It can be also observed that the insertion of the nanocomposite polymer film increased the density of interface states, especially in the energetic positions near the top of the valance band and bottom of the conduction band for p-Si and n-Si, respectively, while as far as the interface state, as their density descends. This also explains the increment of the series resistance upon inserting the polymeric nanocomposite films.

After that, the response of the prepared junction to the change in the intensity of the incident light from a halogen lamp source at room temperature was examined, as shown in Fig. 13. When the device is illuminated, its behavior can change due to photo-excitations, which generates charge carriers in the deep depletion region. These charge carriers can tunnel through the interfacial layer between the Ag electrode and the p-Si, resulting in a slight increase in forward currents compared to dark current [39], as depicted in Fig. 13. When a reverse bias voltage is applied, the depletion region of a device increases, which results in a larger photosensitive area. This, in turn, leads to the formation of more electron–hole pairs, thereby increasing the photoconductivity of the device [20, 44]. Consequently, the number of charge carriers generated is directly proportional to the intensity of the incident light, leading to an increase in the reverse bias current as shown in Fig. 13. The device under consideration exhibits exceptional sensitivity to light and delivers a consistent output in response to varying light intensities. This noteworthy attribute renders it invaluable in optoelectronic applications, particularly in the design of photodiodes and photodetectors.

In the quest to enhance the efficiency of MPS devices, various approaches have been implemented. These include fine-tuning the interfacial characteristics between the polymer and semiconductor layers, incorporating interfacial layers to augment charge separation, and modifying the surface chemistry of the layers to bolster charge transfer. Such strategies are instrumental in improving the effectiveness of charge separation and transport mechanisms, resulting in amplified photocurrent densities and superior overall performance of the device. The phenomenon of photocurrent density, J_photo, in MPS devices arises from the absorption of photons, which generates electron–hole pairs within the material. An electric field can subsequently separate these pairs of electrons and holes, producing a measurable current [79, 80]. The correlation between photocurrent density, J_photo, and incident light intensity under reverse bias is shown in Fig. 14a using the following relation [32]:where I_light and I_dark correspond to the currents with and without illumination, respectively and A is the area exposed to the light.

The remarkable stability of the reverse photocurrent density as shown in the figure, coupled with its noticeable rise in response to increasing light intensity, highlights the immense potential of this architecture in various optoelectronic applications. To gain insights into the photoconductivity mechanism, which is largely influenced by the interfacial layer, it is imperative to analyze the effect of trapping states in the materials constituting the photodetector on the dominant recombination mechanism. In this regard, plotting a graph of photocurrent density (lnJ_photo) against light intensity (lnP), as depicted in Fig. 14b, can be extremely helpful. The relationship between these two parameters can be described by the following equation [81]:where C is a constant and ζ is the exponent coefficient and indicative of the densities of uninvaded trap levels at lower concentrations and can be derived from the linear slope of the ln(J_photo) vs ln(P) curve. As shown in the figure, J_photo increases with increasing P which can be explained by the formation of electron–hole pairs. Upon analysis of the slope, the determined γ value was observed to exceed unity, which is indicative of a reduced density of unoccupied trap levels that are exponentially distributed which controls the recombination mechanism [82, 83].

The band diagram of an MPS-structured device is essential for understanding its electronic properties, such as carrier transport and photovoltaic behavior which is depicted in Fig. 15. This interfacial layer serves as a barrier that prevents the injection of charge carriers from Ag into the semiconductor. The height of this barrier is influenced by the work function of the metal and the alignment of the energy bands at the interfaces between Ag, Ni(OAc)₂, and P-Si layers. When it comes to injecting holes from the Ag into a p-Si layer, the height of the barrier must be less than the difference in energy between the Fermi level of the metal and the highest energy level in the valence band of the p-Si layer. To inject photo-generated holes into the p-Si layer in optoelectronic devices, two mechanisms can be used: direct tunneling through the polymer or carrier transport through interface states. The effectiveness of hole injection is dependent on the energy distribution of the photo-generated holes and the density of interface states [8, 84]. The presence of more interface states can facilitate hole injection by creating additional routes for carrier transport. Moreover, the narrower energy distribution of the photo-generated holes can enhance the injection efficiency by minimizing carrier scattering and recombination. Thus, optimizing the interface state density and energy distribution of photo-generated holes is critical for efficient hole injection in p-Si-based optoelectronic devices [85, 86].

When the MPS device is illuminated, photo-generated electrons (e⁻) and holes (h⁺) can be injected into the p-Si layer. The injection of these carriers depends on their concentration, mobility, and band alignment at the p-Si/insulator interface. Because p-Si has a higher concentration of holes, it is anticipated that the injection of photo-generated electrons into the p-Si will be more significant than the injection of holes. This is due to the electrons being injected into a lower concentration of carriers, allowing for more efficient transport and less recombination. On the other hand, the injection of photo-generated holes into the p-Si may be impeded by their lower mobility compared to electrons, as well as the potential barrier at the p-Si/insulator interface. This potential barrier can prevent the holes from reaching the p-Si layer and cause them to accumulate at the interface, resulting in carrier recombination. As a result, the injection of photo-generated electrons from an illuminated p-Si layer is generally expected to be higher than the injection of photo-generated holes into the p-Si layer. Nevertheless, the exact injection efficiency of these carriers will depend on various factors, such as the doping level of the p-Si, the thickness of the interfacial layer, and the intensity and wavelength of the incident light [10].

The photodetector plays an extremely crucial function as the eyes of a photoelectric system. Most photodetectors are devices that convert the absorbed light energy and generate electrical signals, that can be further processed or stored. Therefore, there will naturally be some parameters that characterize this conversion process. These parameters can be considered as quality indicators of a photodetector for particular applications. These parameters include responsivity, R, as a measure of the output signal produced by the detector in response to the incident light, detectivity, D^*, which characterizes its ability to detect a weak optical signal, linear dynamic range, LDR, is defined by the response of the detector to produce a linear output signal over a range of input optical powers, signal-to-noise ratio, SNR, represent the ratio of the detector’s output signal to the level of noise present in the signal, noise equivalent power, NEP, define the minimum detectable power of the detector, and noise current, I_N, as the current that flows through a photodetector due to random fluctuations in the detector’s internal properties, such as thermal noise, shot noise, and flicker noise. The preceding characteristic parameters of the photodetector and their corresponding values can be expressed and determined using the following relations, respectively [39, 87]:

The relationship between reverse biasing voltages at various illumination intensities and the parameters R, D^*, LDR, SNR, NEP, and I_N is presented in Fig. 16a–f. By measuring these parameters, researchers and engineers can determine the suitability of different light detectors for specific applications, as well as optimize their performance to achieve reliable and accurate measurements. Table 2 illustrates the superiority of the present MPS device relative to other many organic/Si devices.

The On–Off switching behavior is characterized by a rapid surge in current upon illumination, followed by a gradual decay once the light source is turned off. The decay can be attributed to the slow recombination of charge carriers within the semiconductor layer, resulting in a gradual decrease in current. The temporal On–Off switching behavior is highly advantageous for optoelectronic applications as it enables the detection of short-duration light pulses and the capacity to regulate the device’s response time. Moreover, the switching behavior can be modulated by adjusting the properties of the MPS layers, such as the doping concentration and thickness, or by altering the illumination conditions [88]. To study the response time of the photodetector to incident light signals, the temporal On–Off switching characteristics of the prepared Ag/PVA: PVP@ Ni(OAc)₂/p-Si/Al device were studied as shown in Fig. 17 at a bias voltage of − 3 V, a frequency of 0.05 Hz, and choppered square light signals with different intensities. Based on the obtained results, it can be concluded that the photodetector exhibited a rapid, steady, and consistent On–Off switching response at all levels of illumination. To evaluate the operational speed of the fabricated photodetector in response to a light signal, it is necessary to calculate the rise time (t_rise) and decay time (t_decay). Table 2 illustrates that as the light intensity increases from 20 to 100 mW/cm², the estimated rise time decreases from 621 to 76.4 ms, and the decay time decreases from 140 to 77.1 ms. This information reveals that the photodetector’s response time improves as the light intensity increases. This is an essential characteristic for optoelectronic devices that require a fast response time, such as in optical communication and sensing applications. According to the presented data, it can be observed that at illumination intensities below 60 mW/cm², the rise time is longer than the decay time. This discrepancy can be attributed to the limited number of photogenerated carriers that may be trapped by the trapping levels, which could be a result of the defects present in the Ni(OAc)₂ interlayer. On the other hand, at higher intensities (≥ 60 mW/cm²), the rise time is shorter than the decay time. This could be attributed to the dependence of the decay time on the discharge of junction capacitance and electron–hole recombination, while the rise time depends on the occupation of trap levels. Due to the larger effective mass and lower mobility of holes compared to electrons, the decay time takes longer than the rise time [39]. Therefore, the observed phenomenon can be explained by the difference in carrier mobility and the presence of trapping levels in the Ni(OAc)₂ interlayer.