1 Introduction

Rare-earth (RE = La, Pr, Eu) chalcogenides (X = S, Se, Te) have unique chemical, physical, electronic, optical and magnetic properties [1,2,3]. The RE chalcogenides exhibit semiconductor, superconductor, antiferromagnet, thermoelectric, and photoelectric properties. They are used in various technological applications such as IR sources and lasers, memory cells, spintronics, optical switches, electro-optic components, and electronics [1,2,3,4]. Chalcogenides are chemical compounds containing one or more chalcogen elements such as Te, Se or S, and they are covalently bonded materials. Chalcogenides can be crystalline or amorphous, metallic or semiconducting, and conductors. They are fundamentally semiconductors with a band gap typically of 1–3 eV, depending on composition. Various deposition techniques such as pulsed-laser deposition (PLD), vacuum thermal evaporation (TE), chemical vapour deposition (CVD), spin coating (SC) and sputtering techniques are used to prepare chalcogenide films [5].

Phase equilibria between Pr2S3 (Pr2Se3) and thallium chalcogenides were studied previously [6,7,8,9]. In these works was investigated to the chemical interaction of components in the Tl2Se–Pr2Se3 system and determined new semiconductor phases with controlled properties. The compound Tl2Se melts congruently at 390 °С and crystallizes in the tetragonal system (space group I4/mcm) with the unit cell parameters a = 8.02 Å, c = 7.00 Å, c/a = 0.873, and Z = 16 [10]. Pr2Se3 with space group I43d melts congruently at 1700°С and crystallizes in the cubic system with the unit cell parameter а = 8.89 Å [10]. Pr2Se3 is an insulator with σ = 10− 6 Ω−1 cm−1. Some physicochemical properties of Pr2Se3 were described in detail previously [11].

Coumarins have attracted considerable attention due to their outstanding optical properties, inherent photochemical characteristics, photostability, efficient light emission properties, reasonable stability, good solubility and easy synthesis [12,13,14]. Therefore, coumarins are widely used in different fields of chemistry, physics, biology, and medicine. Coumarin materials have also been used on the applications of electronic and photonic systems such as organic light-emitting diodes, sensors, charge-transfer agents, nonlinear optical materials and solar energy collectors.

The present work aims to study electronic and optoelectronic properties of Al/coumarin doped Pr2Se3–Tl2Se/p-Si diodes with various rates of coumarin dopant. The electrical and optical properties of the fabricated diodes were examined by the use of measured current–voltage (I–V), capacitance/conductance–voltage (C/G–V) and transient.

2 Experimental details

The Al/p-Si diodes with various rates of coumarin-doped Pr2Se3–Tl2Se interfacial layers were fabricated by using coumarin and Pr2Se3–TI2Se alloy. For the synthesis of 40% Pr2Se3–60% Tl2Se alloy, praseodymium element was first removed from the oil and cleaned. TI2Se and Pr2Se3 were synthesized from elemental thallium (Tl-00 grade), 99.98%-pure praseodymium and 99.998%-pure selenium in quartz ampoules evacuated to 0.1333 Pa. Reaching equilibrium was checked by differential thermal and microstructural analyses. Pr2Se3 was synthesized in two steps. Initially, a mixture of elemental Se and Pr was kept at 300 °C for 3 days. Then, the temperature was increased to 1150 °C, and the blend was kept at that temperature for 3 h. Thus, the compound Pr2Se3 was obtained in a powder form. The powder was compacted and annealed in a furnace at 650 °C for 450 h (solid phase reaction). Alloys in the TI2Se–Pr2Se3 system were synthesized from TI2Se and Pr2Se3 within the temperature range 600–1200 °C; the alloys were cooled at a rate of 80 eg/h.

The Pr2Se3–Tl2Se and coumarin materials were stirred (0.00, 0.05, 0.1, 0.3 wt% coumarin) at 60 °C for 10 min by using a magnetic stirrer. In order to remove the native oxide layer, p-type Si wafer was etched by HF before drop coating of materials onto the wafer. Then, the Si wafer was ultrasonically rinsed using deionized water for 10–15 min and dried under nitrogen gas. Finally, the wafer was cleaned with standard chemical cleaning procedure using methanol and acetone baths. The ohmic contacts were produced with aluminum (Al) metals by thermal deposition on back side of the Si wafer, the wafer was then annealed at 570 °C for 5 min in nitrogen atmosphere. The TI2Se–Pr2Se3 alloys with weight of coumarin dopant (0.00, 0.05, 0.1, 0.3 wt%) were covered onto the front side of the wafer by using drop casting method and dried for 10 min at 100 °C to have a solid film. Al circular top or Schottky contacts were prepared by thermal deposition on the film using a shadow mask, and the contact area of the prepared diode is 7.85 × 10−3 cm2.

Electrical and photoelectrical measurements of Al/coumarin doped Pr2Se3–Tl2Se/p-Si diodes with various coumarin-doped Pr2Se3–Tl2Se interfacial layers were performed at room temperature by the use of a Fytronix 5000-ECS Electronic Device characterization system.

3 Results and discussion

3.1 Current–voltage measurements

For forward-bias voltage values (V ≥ 3kT/q), the characteristics of I–V considering thermionic emission (TE) theory are investigated based on the following equation [15, 16],

$$I={I_0}\exp \left[ {\left( {\frac{{qV - I{R_s}}}{{nkT}}} \right)} \right]$$
(1)

where n, Rs, and I0 terms stand for the ideality factor, series resistance, and reverse saturation current occurring by the diffusion of the minority charge carriers, respectively. For the Schottky diode, the reverse saturation current is almost independent from the reverse bias voltage and is defined by,

$${I_o}=A{A^ * }{T^2}\exp \left( { - \frac{{q{\Phi _b}}}{{kT}}} \right)$$
(2)

where A, A*, and Φb are the contact area of photodiode, the effective Richardson constant of 32 Acm−2 K−2 for p-type Si semiconductor, and the zero-bias barrier height calculated following the Eq. (2), respectively. The n and I0 values were obtained from the slope and intercept of the forward bias region for ln(I) versus V plots, respectively.

Figure 1a–d shows I–V graphs of the Al/p-Si diodes at different rates of coumarin-doped Pr2Se3–Tl2Se interfacial layers under dark and different solar light powers. As illustrated in the figures, for all of the diodes, the reverse bias current increases with the increase of the light intensity. In other words, the reverse bias current affected by the illumination is higher than the one measured at dark current condition. The forward bias current does not vary by light intensity; however, it increases exponentially with voltage increase. These results confirm that the diodes exhibit a photoconducting behavior. When a diode is illuminated, photogenerated charge carriers will be occurred, and they will contribute to the dark current [17,18,19]. Also, the diodes showed a good rectifying behavior in the dark conditions. The rectification ratio (RR = IF/IR) values of the 0.00, 0.05, 0.1 and 0.3 wt% coumarin doped diodes were compared to each other at applied bias V = ± 5 V, and they were calculated as 1.57 × 105, 2.24 × 105, 9.06 × 104 and 1.07 × 105, respectively. The highest RR value was found for the diode having 0.05 wt% coumarin doping. In addition, for dark, the n and Φb values of the diodes at different rates of coumarin were found to be in the range of 2.69–4.17 and 0.71–0.76 eV, respectively. For 100 mW/cm2, their values were found to be in the range of 3.93–5.43 and 0.69–0.81 eV. The ideality factor values are greater than the unity, thus suggest the existence of the series resistance, barrier height inhomogeneities and interface states [20,21,22,23,24,25,26,27]. Furthermore, I–V plots deviate from the linearity in the forward bias region because of the series resistance effects.

Fig. 1
figure 1

I–V plots of the photodiodes with various rates of coumarin contents under dark and various illumination intensities a undoped (0.00 wt%), b 0.05 wt%, c 0.1 wt%, d 0.3 wt%

Full size image

Therefore, the series resistance (Rs) and shunt resistance (Rsh) are crucial parameters affecting the diode performance. Ohm’s law (Rj = dV/dI) was used for determining the Rs and Rsh. In very high forward bias region, the resistance of the junction reaches its maximum value and remains as a constant after a certain point, and this value corresponds to the Rs. In very high reverse bias region, the resistance of the junction shows almost constant value, which corresponds to the Rsh [28,29,30]. Figure 2a–d presents the plots of the junction resistance (Rj) versus voltage of the undoped, 0.05, 0.1 and 0.3 wt% coumarin doped photodiodes. As observed in Fig. 2a–d, the Rsh values for all the diodes increase by illumination intensity at reverse bias region. At forward bias region, the Rs value does not change almost with illumination intensity. The Rsh values for the diodes with various coumarin contents are calculated as 39.3, 24.8, 100 and 60.1 MΩ at dark and − 5 V, respectively. The Rs values are calculated as 0.25, 0.11, 1.12 and 0.56 kΩ at dark and + 5 V, respectively. As a conclusion, the ideal diode needed to be at a high shunt resistance and a low series resistance. In this study, the highest value of Rsh and the lowest value of Rs are obtained for the diodes which have 0.1 and 0.05 wt% coumarin doping rates, respectively.

Fig. 2
figure 2

Rj–V plots of the photodiodes with various rates of coumarin contents a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%

Full size image

To investigate the photoconduction mechanism of the diodes was used the following relation [31, 32]:

$${I_{ph}}=A{P^m}$$
(3)

where A is a constant, Iph stands for the photocurrent, P is the intensity of illumination and m is illumination coefficient. Using the slope of log(Iph) versus log(P) graph, the m value is obtained. Figure 3 exhibits the variation of the photocurrent by the solar light intensity. It was found that the photocurrent increased by increasing light intensity. The m values for various rates of coumarin doped photodiodes were determined from the slope of these plots and were found to be 1.19, 1.21, 1.29, and 1.30, respectively. The obtained m values indicate that the photocurrent shows a linear behavior of photoconducting. The meaning of m values, which are greater than unity, express that the photoconducting mechanism of device arise from the lower densities of the uninvaded trap levels. Beyond our results, if the value of m is between 0.5 and 1, the photoconducting mechanism of device is correspond to continuously localized states in Pr2Se3–Tl2Se/p-Si interface [33, 34]. Furthermore, the experiment confirms that the fabricated diodes can be used at different optoelectronic applications.

Fig. 3
figure 3

Log(Iph) versus Log(P) plots of the photodiodes with various rates of coumarin contents (at − 5 V)

Full size image

3.2 Transient photocurrent/photocapacitance/photoconductance-time measurements

To have a better understanding on the photoconduction mechanism of the photodiodes, the measurements of transient photocurrent were practised at different illumination intensities. The photocurrent response technique is used to evaluate the separation efficiency of photogenerated electron–hole pairs. Figure 4a–d shows time-dependent photocurrent response of the diodes at applied bias V = − 5 V. As seen in these figures, the diodes exhibit typical on/off illumination switching behavior. The photocurrent raises quickly up to a certain level since the increase of free charge carriers contributes to the current, and reaches the saturation level when the light is switched on. Ones the light is switched off, the photocurrent drops rapidly and returns to its beginning level. The observed behavior is attributed to charge carrier trapping in deep impurity levels [35,36,37,38,39,40]. As seen in Fig. 4e, the highest Iphoto/Idark photosensitivity was found to be 1327 for the diode which has 0.1 wt% coumarin doping rate at 100 mW/cm2 and − 5 V.

Fig. 4
figure 4

Transient photocurrent-time plots of the photodiodes with various rates of coumarin contents at − 5 V a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%, e 100 mW/cm2

Full size image

Figures 5a–d and 6a–d show the plots of transient photocapacitance and photoconductance measeuments as a function of time, respectively. It is observed that both the photocapacitance and photoconductance for all diodes depend strongly on illumination and increase with the increase of light intensity. When the light is switched on, the photocapacitance and photoconductance due to photogeneration of charge carriers increase until the illumination was switched off. After switching illumination off, the photocapacitance and photoconductance because of charge carriers trapping decrease and come back their initial values. As seen in Fig. 5e, the highest photocapacitance was obtained for the diode which has 0.3 wt% coumarin doping rate. As seen in Fig. 6e, the highest photoconductance was found for the diode which has 0.1 wt% coumarin doping rate. As a result, the transient responses indicate that the diodes exhibit a photo capacitive behavior.

Fig. 5
figure 5

Transient photocapacitance–time plots of the photodiodes with various rates of coumarin contents at 10 kHz a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%, e 100 mW/cm2

Full size image
Fig. 6
figure 6

Transient photoconductance–time plots of the photodiodes with various rates of coumarin contents at 10 kHz a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt% e 100 mW/cm2

Full size image

3.3 Capacitance/conductance–voltage measurements

The capacitance–voltage graphs of the diodes using various rates of coumarin dopants at different frequencies are presented in Fig. 7a–d. As shown in these figures, the values of capacitance for all diodes decreased by the increase of the frequency in the positive voltage region. However, the capacitance does not vary by the frequency in the negative voltage region. The dependence of capacitance on the frequency is attributed to the presence of interface states [41,42,43,44,45,46,47]. In addition, the capacitance is related to the ability of the interface states to follow the applied ac signal. Especially, the interface states cannot follow the ac signal, which is resulted in lower capacitance at sufficiently high frequencies. Additionally, C–V graphs exhibit a peak at all frequencies and positive voltage region. The magnitudes of these peaks decrease towards higher frequencies. Figure 8a–d shows the G–V plots of the undoped and coumarin doped diodes. Increases on the conductance are observed with increase of the frequency. This behavior indicates the presence of interface states.

Fig. 7
figure 7

C–V plots of the diodes at various frequencies a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%

Full size image
Fig. 8
figure 8

G–V plots of the diodes at various frequencies a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%

Full size image

The conductance technique, which is developed by Nicollian and Goetzberger [44, 48], was used to calculate the series resistance (Rs) of the undoped and coumarin doped photodiodes. From this technique, the Rs can be extracted from C/G–V characteristics in the strong accumulation region. The Rs value is calculated by using the following equation,

$${R_s}=\frac{{{G_{ma}}}}{{G_{{ma}}^{2}+{{\left( {\omega {C_{ma}}} \right)}^2}}}$$
(4)

where Cma and Gma stands for the measured capacitance and conductance, respectively. Figure 9a–d presents the series resistance–voltage (Rs–V) graphs of the undoped and coumarin doped photodiodes for various frequencies. It is observed that the calculated Rs values of all diodes decrease by the increase of the frequency. The behavior of Rs is attributed to the trapped charges. Beside, the Rs–V graphs exhibit a peak between 0 and 2 V, decreasing with increasing frequency. Also, the peak shifts to higher positive bias voltage by the frequency increase. The peak is observed due to the existence of interface states.

Fig. 9
figure 9

Series Rs–V plots of the diodes at various frequencies a undoped, b 0.05 wt%, c 0.1 wt%, d 0.3 wt%

Full size image

4 Conclusions

Pr2Se3–Tl2Se thin films with various coumarin (0.00, 0.05, 0.1, 0.3 wt%) contents were prepared by using drop casting technique and thus Al/coumarin doped Pr2Se3–Tl2Se/p-Si diodes were fabricated. The experimental results show that the electronic and optoelectronic properties of the diodes are changed with coumarin doping. Also, it is observed that the diodes are strongly sensitive to illumination intensity. The photoresponse of the diodes was confirmed by transient measurements. In addition, the obtained results verify that interface states and series resistance are crucial parameters which strongly affect the electrical parameters of the diodes. As a result, the fabricated photodiodes might be used as a photosensor in the applications of optoelectronic.