1 Introduction

Metal–semiconductor (MS) structure have very fast switching characteristics and very low forward-voltage drop and usually studies are focused on the increase of performance and reliability of MS type SBDs because they have both the forming fundamental piece of the device technology and important role in electronics and optoelectronics device applications [1,2,3,4,5]. Unless a specially fabricated of MS type SBDs/structures, a native insulator layer can be occurred at M/S interface, and so MS type SBD will be converted to the MIS type SBD like a capacitor which can be stored many charges and shows dielectric properties. The existence such insulator, polymer and ferroelectric interfacial layer in the MIS, MPS, MFS type structures will be strongly influenced both the performance and conduction mechanisms in these type electronic devices [6,7,8,9,10]. But, the formation of a low-dielectric insulator layer (SiO2, SnO2, Si3N4) by the traditional method cannot act passivate of the active dangling-bonds at the surface of semiconductor. For these reasons, recently, many researchers are focused on the improving the electric, dielectric and optoelectronic properties of these devices [11,12,13,14]. An interfacial polymer layer has high- surface area to volume rate, low weight, low cost, low consumption energy, good mechanical strength, the capacity charge-storage, flexibility, big-dielectric-constant, easy and cheap preparing methods including electro-spinning, sol–gel, solid–liquid phase separation and template synthesis [15,16,17,18,19,20,21]. Therefore polymers such as PVA and PVP have become more important research subject because of its high-water solubility, high-film formation capacity and, thereof its conductivity can be increased by a suitable doping meatal or metal-oxides such as Zn, Co, Ni, CuSO4, ZnO, Bi2O3, and SnO2 via the high-physical interactions between organic chains, and hydrogen-bonding between hydroxyl-groups and dopant-materials [17,18,19,20,21,22,23,24].

The values of ΦB, n, Nss, series and shunt resistances (Rs, Rsh) are used to determine the performance of MS, MIS, MPS, and MFS type structures/SBDs by using the I–V, C–V, and G/ω–V measurements in the wide range of voltage as well as frequency and temperature [9,10,11,12]. In recently, there are many studies in the literature on the characterizing and optimizing electrical parameters of polymer based SBDs by using different polymer layers on semiconductors [1,2,3,4,5,6,7, 13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]. Among these researches, Aydemir et al. [20] studied on comparative of the electrical parameters of MS and MPS structures which are performed on the n-Si, Çetinkaya et al. [21] studied on comparative of electrical properties for Au/(GP-doped PVA)/n-Si (MPS) fabricated with different rates graphene and Altındal-Yerişkin et al. [22] studied on comparative of the electrical properties for MS type SDs by using (graphene doped-PVA) polymeric interfacial layer between metal and semiconductor and they reported the RR of the MPS structure at ± 3 V was found at about 493-times higher than the MS structure. Reddy et al. [23] were also observed that the magnitude of Nss Au/polyvinylidene-fluoride/n-InP structure is ten times lower than Au/n-InP type structure due to the dangling-bonds saturation by an organic interlayer. Although all of these studies; the interfacial layer effect on the performance of MS type SBDS and conduction mechanisms and also the origin of Nss and Rs is not enough clarified yet. Additionally, CoSO4 is generally used in storage batteries and electrochemical industries because it has strong absorption in shorter wavelengths and magnetic properties [28,29,30]. There are very few studies in the literature related to CoSO4 and these are generally related to properties such as optical, thermal, thermodynamic and magnetic of CoSO4 [28,29,30,31,32,33]. However, there are hardly studies related to electrical properties of CoSO4.

Due to the above explanations, in this study, we aimed to the fabricated both the Au/n-Si and Au/(CoSO4–PVP)/n-Si structures and compare their physical characteristics in order to determine the whether or not improve the electric properties by the way of passivation active dangling bonds at the semiconductor. Therefore, I–V, C–V and G/ω–V characteristics were performed in wide range voltages considering the effects of Nss, Rs and organic interlayer. Experimental results show that the used (CoSO4–PVP) interfacial layer improves the performance of MS structure through decreasing the active dangling bonds and the leakage current, Nss, and increasing the rectifying ratio (RR), and BH. Therefore (CoSO4–PVP) interfacial layer can be successfully used instead of the conventional interfacial insulator layer and gives also them capacitor properties, which more and more storage electric charges or energy.

2 Experimental details

2.1 Synthesis of the cobalt sulfate–cobalt sulfide nanostructures and fabrication of Au/(cobalt sulfate–cobalt sulfide/PVP)/n-Si structure

Cobalt acetate hydrate (CH3CoO)2Co∙4H2O with purity greater than 99%, and sodium sulfide (Na2S) with purity greater that 99% was purchased from Loba chemie company, double distilled water was used as solvent and washing agent material. Polyvinylpyrrolidone (PVP) was also purchased from Loba chemie. All of the chemicals were used without further purifications.

The cobalt sulfate–cobalt sulfide nanostructures were synthesized by ultrasound-assisted method. In a typical preparation method, 0.10 g of cobalt acetate and 0.96 g of sodium sulfide were dissolved in 20 ml of double distilled water separately. Then these solutions were added together in a flat bottom baker and simultaneously kept under ultrasonic irradiation for 15 min in open air, using a high intensity ultrasonic processor. The result material centrifuged and washed for five times with distilled water, then the powder form of the nanostructures was dried at room temperature and used for fabrication of Au/(cobalt sulfate–cobalt sulfide/PVP)/n-Si structure. For fabrication of Au/(cobalt sulfate–cobalt sulfide/PVP)/n-Si structure, at first, n-type semiconductor Si wafer was cleaning in an ultrasonic-bath and then high-pure Au (99.999%) was grown on the back side of the n-Si wafer at about 10−6 torr. Immediately, it annealed at 550 °C at 5 min in the N2 ambient to get low-resistivity ohmic contact followed by the preparation of (cobalt sulfate–cobalt sulfide)/PVP nanostructures on n-Si wafer by utilizing spin coating method. 0.1 g of the prepared nanostructures in the powder form was dissolved in 2 ml of aqueous solution of PVP polymer (5%). A drop of the solution was coated by spin coating techniques. Finally, Au dots (7.85 × 10−3 cm2) were thermally grown on the (cobalt sulfate–cobalt sulfide/PVP). I–V and C/G–V measurements were performed by a Keithley 2400 source meter and HP4192A LF impedance analyzer with the help of an IEEE-488 AC/DC converter card.

2.2 X-ray diffraction measurement

The structure of the prepared cobalt sulfate–cobalt sulfide nanostructures is depicted in Fig. 1. The observed XRD peaks in Fig. 1 are well matched with those of mixture of cobalt sulfate and cobalt sulfide nanostructures.

Fig. 1
figure 1

XRD pattern of the synthesized cobalt sulfate–cobalt sulfide nanostructures

Full size image

2.3 FE-SEM and EDS measurement

Both the surface-morphology and elemental-analyses of the prepared samples were characterized by field emission (FE)-SEM and EDS and depicted in Figs. 2 and 3, respectively.

Fig. 2
figure 2

a, b FE-SEM images of the synthesized nanostructures in different magnifications

Full size image
Fig. 3
figure 3

EDS spectrum of the synthesized cobalt sulfate–cobalt sulfide nanostructures

Full size image

It is hard to see the real particle sizes using FE-SEM, however, an overview from the images depicts formation of polydispersity and agglomerated nanostructures with different geometries. As shown in Fig. 2a, b, there are some small nanoparticles.

The quantitative elemental analyses of the synthesized cobalt sulfate- cobalt sulfide nanostructures were carried out using the EDS technique. Figure 3 depicts typical EDS spectrum, which shows the product contains cobalt, sulfur and oxygen materials with ratio 53:8:39 in weight percent respectively.

3 Results and discussion

3.1 Forward and reverse bias I–V characteristics at room temperature

Figure 4 shows the typical semi-logarithmic I–V plots of the Au/n-Si structure with and without (CoSO4–PVP) organic interlayer at ± 3 V, and they show good saturation behavior for the reverse bias voltages and RR (= IF/IR at ± 3 V). The value of IF increases up to ~ 0.5 V almost as exponentially and then deviates from the linearity due to the effects of Rs and interlayer. In general, TE theory is used for the extraction of electrical parameters of SBDs. According to TE theory, the relationship between IF and VF (≥ 3 kT/q) can be expressed as following [9,10,11,12].

$$I = \underbrace {{AA*T^{2} \exp \left( {{{ - q\varphi_{Bo} } \mathord{\left/ {\vphantom {{ - q\varphi_{Bo} } {kT}}} \right. \kern-0pt} {kT}}} \right)}}_{Io}\left\{ {\exp \left( {\frac{{q(V - IR_{s} )}}{nkT}} \right) - 1} \right\}$$
(1a)

In Eq. (1a), A* is the Richardson constant (120 A/(cm K)2 for n-Si) and the other quantities are well-defined in the available literature [9, 11]. The reverse-saturation current (o) which is obtained from the linear part of ln(I)–V plots at zero-bias voltage. On the other and, the values of n can be obtained from the of slope of ln(IF)–V plots.

$$n = \frac{q}{kT}\left( {\frac{dV}{d(\ln (I))}} \right)$$
(1b)

After the extraction of Io values from the forward bias ln(I)–V plots, the values of zero-bias BH (ΦBo) can be calculated by using Eq. 1c.

$$\varPhi_{Bo} = \frac{kT}{q}\ln \left( {\frac{{AA*T^{2} }}{{I_{0} }}} \right)$$
(1c)

As shown in Fig. 1, the values of RR of the MPS structure at ± 3 V was found 26.77 times higher than the Au/n-Si (MS) type SBD.

Fig. 4
figure 4

The lnI–V plots of the MS and MPS structures at room temperature

Full size image

Both the series and shunt resistances (RS, RSH) are also important parameters of the SBDs which are more affecting the performance and quality of them and they can be calculated from the structure resistance (Ri) versus V plots. Thus, to examine voltage dependent of Ri for two type SBDs, the Ri–V plots calculated by using Ohm’s Law (Ri = dVi/dIi) are presented in Fig. 5. As can be clearly seen in Fig. 5, the values of Ri approach to a constant value which are corresponding the real values of RS and RSH, respectively. Thus, the obtained experimental values of n, Io, ΦB0, Rs RSH and RR for the MS and MPS structures were tabulated in Table 1.

Fig. 5
figure 5

The lnRi–Vi plots of the MS and MPS structures at room temperature

Full size image
Table 1 The values of n, Io, ΦBo, Rs RSH and RR for the MS and MPS structures extracted from the I–V characteristics at room temperature
Full size table

In order to evaluate the current-transport in more detail, the In(IF)–In(VF) plot of the MS and MPS structures were drawn and given in Fig. 6.

Fig. 6
figure 6

The In(IF)–In(VF) plots of the MS and MPS structures at room temperature

Full size image

As can be seen in Fig. 6, In(I)–In(V) plots for two type SBDs have three district linear regions in the form of I ∝ Vm and the slope of these regions are different from each other. Where, m represents the slope of each linear region. For each SBD, these regions are named as regimes I, II and III, respectively. Region I is corresponding to the low voltages and for MS and MPS type SBDs and in this region the conduction mechanism (CM) exhibits an ohmic behavior [34,35,36,37]. Regime II is corresponding to the intermediate voltages and CM is characterized by power law dependence and it can be explained by trap-charge limited current (TCLC) [37, 38]. Regime III is corresponding to the high forward-bias voltages and many electrons can escape from the traps and contributes to space-charge-limited-current (SCLC) [10, 35, 38, 39].

The surface states or interface traps (Nss/Dit) play a key role in the conduction mechanism in SBDs. Because, the existence of Nss localized between interfacial layer and semiconductor is dominate on the I–V and C/G–V characteristics in these devices. The energy-dependent profile of Nss as deduced by Card–Rhoderick can be extracted from the forward bias I–V plot considering voltage dependent BH (Φe) and n by using Eqs. 2a and 2b, respectively [10, 12, 39, 40]:

$$n(V) = \left( {\frac{q}{kT}} \right)\left( {\frac{V}{{\ln (I/I_{o} )}}} \right)$$
(2a)
$$\varphi_{e} = \varphi_{Bo} + \left( {1 - \frac{1}{n(V)}} \right).V$$
(2b)
$$N_{ss} (V) = \frac{1}{q}\left[ {\frac{{\varepsilon_{i} }}{\delta }\left( {n(V) - 1} \right) - \frac{{\varepsilon_{s} }}{{W_{D} }}} \right]$$
(2c)

In Eq. 2c; δ, ε0, WD, εi and εs are corresponding to the interfacial layer thickness, the permittivity of vacuum (= 8.5 × 10−14 F/cm), the width of depletion region, the permittivity of interfacial layer and semiconductor, respectively. The value of WD was obtained from the intercept of C−2–V plots at 1 MHz for MS and MPS type structures. The energy of surface states (Ess) with respect to edge of conduction band (Ec) for n-type semiconductor is represented as following [12]:

$$E_{C} - E_{ss} = q(\varphi_{e} - V)$$
(3)

Thus, Nss versus (Ec − Ess) plots for the MS and MPS type SBDs were extracted by using Eqs. (2a)–(2c), (3) and presented in Fig. 7. As can be seen in Fig. 7, the Nss values decrease from conduction band towards to the mid-gap of n-Si for two types SBDs. Additionally, the magnitude of Nss for MPS structure is lower than MS structure because of the dangling bonds saturation due to the use of (CoSO4–PVP) organic interlayer.

Fig. 7
figure 7

The energy-dependent profiles of Nss for MS and MPS structures at room temperature

Full size image

3.2 The forward and reverse bias C/G–V Characteristics at 1 MHz

Figures 8 and 9 show both the C–V and G/ω–V plots of the MS and MPS structures at 1 MHz, respectively. As shown from the Figs. 8 and 9, both the C–V and G/ω–V curves have inversion, depletion and accumulation regions similar to a MOS capacitor or structure. It is clear that the C–V plot for the MS structure has a distinctive peak in the depletion region, but it becomes disappeared for the Au/(CoSO4–PVP)/n-Si (MPS) type SBD (Fig. 8). Such behavior in the C–V plot for the MPS type SBD is the result of passivation of active dangling-bonds in the semiconductor crystalline thanks to the used of (CoSO4–PVP) interfacial layer. Because, un-passivated surface states (Nss) and bulk traps can be stored and released many charges under forward applied bias voltage, leading to an increase or decrease. Additionally, both the C–V and G/ω–V plots may be given some peaks due to a special density distribution of Nss and dislocations between interfacial layer and semiconductor in the forbidden energy bandgap of semiconductor [10, 12, 13, 39]. The concave curvature in the C–V plots at accumulation region is the results of Rs and interfacial layer. That is while the Nss are especially dominate in depletion and inversion regions, Rs and interfacial layer are dominate only at accumulation region. According to Nicollian and Brews [10, 39, 41, 42],

Fig. 8
figure 8

The C–V plot of the MS and MPS structures for 1 MHz at room temperature

Full size image
Fig. 9
figure 9

The G/w–V plot of the MS and MPS structures for 1 MHz at room temperature

Full size image

In order to emphasis the effect of Rs both on the C–V and G/w–V characteristics, voltage dependent profiles of Ri was extracted from the Eq. 4, proposed by Nicollian and Brews [41] and were given in Fig. 10. As shown in Fig. 10, the value of Ri becomes almost a constant at accumulation region and so it corresponds the real value of Rs for MS and MPS structures.

$$R_{s} = \, G_{m} /\left[ {G_{m}^{2} + \, \left( {\omega C_{m} } \right)^{2} } \right]$$
(4)

In Eq. 4, ω is the angular frequency, Cm and Gm values are capacitance and conductance measured at a given bias voltage, respectively. The peak behavior in the Ri–V plot for two type SBDs are results of particular density distribution of Nss between interface layer and semiconductor and their relaxation times, restructure and reordering of Nss under applied bias voltage. The effect of Rs can be neglected low at low frequencies and in inversion region contrary to high frequencies and at accumulation region.

Fig. 10
figure 10

The Ri–V plots of the MS and MPS structures for 1 MHz at room temperature

Full size image

In order to examine the changes in some main electrical parameters such as ΦB(C–V), the Fermi level (EF), donor concentration atoms (ND) and diffusion potential (VD); the C−2–V plots were also drawn and given in Fig. 11 for the MS and MPS structures for 1 MHz. As shown in Fig. 11, the C−2–V plots for each SBDs have a straight line in the wide range of voltage. Thus, the values VD, ND, EF, ΦB(C–V), and WD are calculated from the intercept and slope of these plots for two type SBDS by using following equations [9, 10, 39] and represented in Table 2.

$$\phi _{{B_{{(C - V)}} }} = V_{D} + E_{F} - \Delta \phi _{B}$$
(5)

In Eq. 5, VD (= Vo + kT/q) is the diffusion potential at zero bias, and ΔΦΒ is the image force barrier lowering. After the obtained value of ND and VD; the values of EF and WD are calculated by using following relations [39, 41].

$$E_{F} = \frac{kT}{q}\ln \left( {\frac{{N_{C} }}{{N_{D} }}} \right)$$
(6a)
$$W_{D} = \left( {\frac{{2\varepsilon_{s} V_{D} }}{{qN_{D} }}} \right)^{{\left( {{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}} \right)}}$$
(6b)

In Eq. 6a, NC is the effective density of states in the Ec band, \(m_{e}^{ * } = 0.98m_{0}\) for electrons, and h (= 6.62 × 10−34 J s) is the Planck constant, respectively.

$$N_{C} = 4.82x10^{15} T^{{{3 \mathord{\left/ {\vphantom {3 2}} \right. \kern-0pt} 2}}} \left( {{\raise0.7ex\hbox{${m_{e}^{*} }$} \!\mathord{\left/ {\vphantom {{m_{e}^{*} } {m_{0} }}}\right.\kern-0pt} \!\lower0.7ex\hbox{${m_{0} }$}}} \right)^{{{3 \mathord{\left/ {\vphantom {3 2}} \right. \kern-0pt} 2}}}$$
(7)

The values of maximum electric field (Em) at junction and image force barrier lowering (ΔΦΒ) is were also calculated by using following relation [10, 39]:

$$E_{m} = \left( {\frac{{2qN_{D} V_{0} }}{{\varepsilon_{s} \varepsilon_{0} }}} \right)^{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}}}$$
(8a)
$$\Delta \varphi_{B} = \left( {\frac{{qE_{m} }}{{4\pi \varepsilon_{s} \varepsilon {}_{0}}}} \right)^{{{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-0pt} 2}}}$$
(8b)

Thus, the obtained experimental values of V0, ND, VD, EF, Em, ΦB(C–V) and WD for the MS and MPS type structure extracted from the C–V characteristics at room temperature and were tabulated in Table 2. When compared Tables 1 and 2, the obtained value of ΦB(C–V) from reverse bias C−2–V plots is higher than the ΦBo(I–V) from forward bias ln(I)–V plots for two structures due to the nature of measured method and voltage dependent BH. But, the value of BH obtained from both the forward bias I–V data and reverse bias C–V data for MPS structure is higher than the MS structure due to the formation of (CoSO4–PVP) organic interlayer.

Fig. 11
figure 11

The C−2–V plots of the MS and MPS structure for 1 MHz at room temperature

Full size image
Table 2 The values of V0, ND, VD, EF, Em, ΦB(C–V) and WD for the MS and MPS structures extracted from the C–V characteristics at room temperature
Full size table

4 Conclusion

In order to determine the effect of (CoSO4–PVP) polymer interlayer on the electrical characteristics, the MS MPS structures were performed on the same n-Si wafer. For this purpose, electrical characteristics of them have been investigated by using the I–V, C–V, and G/ω–V measurements in a wide range of voltage and compared with each other. The values of n, ΦBo and the RR (at ± 3 V) for the MS and MPS type SBDs were obtained as 2.453, 0.732 eV, 2.01 × 103 and 2.489, 0.799, 5.37 × 104 by using the I–V measurements in dark, respectively. The value of RR for MPS is also 26.77 times higher than the MS structure. In addition, the other some experimental values of these SBDs such as ND, EF and BH were found as 15.06 × 1014 cm−3, 0.254 eV, 0.744 eV and 2.310 × 1014 cm−3, 0.303 eV, 1.010 eV from the C−2–V characteristics at 1 MHz, respectively. Obtained the value of ΦB(C–V) from reverse bias C–V data is higher than the ΦBo(I–V) from forward bias I–V data for two structures due to the nature of the measured method and voltage dependent BH. Additionally, the value of BH obtained from both the forward bias I–V data and reverse bias C–V data for MPS is higher than the MS structure due to the used of (CoSO4–PVP) organic interlayer. These results show that the use of (CoSO4–PVP) polymer interface layer at Au/n-Si interface improves the performance of the structure and so can be successfully replaced in the insulator layer in the future.