Introduction

Organic materials have attracted intensive research to study their electrical and optical characteristics for application in the electronics field due to the possibility of fabricating electronic devices with desired shapes and flexibility at low processing cost using simple methods. Configurations based on hybrid structures with organic semiconductor layers are known as organic electronic devices, e.g., solar cells, sensors, photodetectors, and transistors.1 In general, organic heterojunctions have attracted great interest due to their easy manufacture and the particular optoelectronic properties obtained, allowing smooth integration with existing technologies.2 Poly[2-methoxy-5-(2-ethylhexoxy)-p-phenylenevinylene] (MEH-PPV) is one of the most important derivatives of PPV with a π-conjugated system in the main chain.3 MEH-PPV is an emissive amorphous polymer.4 Moreover, because of the alkoxy on the phenyl group in the polymer chains, MEH-PPV can dissolve in both aromatic and nonaromatic solvents.5 To deposit organic material on p-type silicon, ultrasonically sprayed MEH-PPV was fabricated using homemade ultrasonic spray pyrolysis (USP) equipment, while silver was physically evaporated to form metallic contacts. USP is a facile approach for the synthesis of high-quality thin films for use in high-performance organic electronics, as it can prepare thin, dense, and smooth layers with fewer defects.6

In the work described herein, a metal–semiconductor–polymer (MSP) organic heterojunction structure based on MEH-PPV and p-type silicon was successfully fabricated. The ultimate aim of this work is to investigate its optical, structural, and electrical characteristics based on I–V and C–V measurements in the dark at room temperature, to determine various parameters of such heterojunction devices. The novelty of this research is the demonstration of the ability to construct such a device from organic MEH-PPV using the low-cost USP and PVD techniques.

Experimental Procedures

Starting Materials

The chemical structure of the used conducting polymer MEH-PPV is shown in Fig. 1. Chloroform as solvent was provided by Sigma Aldrich and used without further purification. Both the silver used for the electrodes and the other chemicals utilized for cleaning (distilled water, toluene, and ethanol) were provided by Kurt J. Lesker Company. The organic molecule MEH-PPV was grown on p-type silicon substrate with thickness of 406 µm to 457 µm, (111) orientation, and resistivity of 0.01 Ω cm to 0.02 Ω cm.

Fig. 1
figure 1

Chemical structure of MEH-PPV organic polymer.

Full size image

Device Fabrication

Preparation Process and Organic Layer Growth

Glass and Si substrates were washed using toluene and ethanol before deposition. A homogeneous solution was prepared from 6 mg of MEH-PPV polymer dissolved in 15 ml of chloroform, a nonaromatic solvent; the prepared solution was nearly red. The obtained solution was deposited on the silicon substrate after 2 h to 3 h of magnetic stirring at a temperature of 50°C.

The ultrasonic spray pyrolysis (USP) process was carried out on a substrate heated to temperatures between 90°C and 120°C, for 35 min at a spray rate of 0.1 ml/min. After deposition, the sample was annealed on a hot plate inside a glove box at 125°C.

The USP process was repeated several times to obtain a film of the desired thickness within the 140 nm range. A silver grid electrode as a front contact was physically evaporated at 1100°C for 40 min using a VAKSIS 1M4T PVD setup, through a shadow mask with diameter of 1 mm, to a thickness of 140 nm. Figure 2 shows a cross-sectional schematic of the organic heterojunction structure based on the MEH-PPV polymer.

Fig. 2
figure 2

Schematic cross-section of obtained Al/p-Si/MEH-PPV/Ag structure.

Full size image

Analysis and Measurement of Organic Layer

Scanning electron microscopy (SEM; Quanta) was applied to measure the thickness of the as-grown organic thin layer. The optical properties of the organic layer were analyzed by ultraviolet–visible (UV–Vis) spectrometry (Lambda 20; PerkinElmer) in the wavelength range from 400 nm to 700 nm. The crystalline structure was investigated by x-ray diffraction (XRD) analysis (D8 Advance; Bruker).

Current–voltage (I–V) and capacitance–voltage (C–V) measurements were carried out in the dark a room temperature using a Keithley 2400 source meter and Agilent 4285A impedance analyzer, respectively.

Results and Discussion

Structural Analysis

To determine the structure of the as-grown thin films, XRD analysis was carried out on the MEH-PPV polymer deposited on glass, revealing a broadened (002) peak at 16.04° (Fig. 3). The structural parameters were determined using Scherrer formula,7

Fig. 3
figure 3

XRD profile of organic MEH-PPV polymer deposited on glass substrate.

Full size image
$$ D = \frac{{0.94\lambda }}{{\beta \cos \theta }}, $$
(1)

where D is the crystal size (nm), λ = 1.54 Å is the x-ray wavelength (in nm), 2θ is the diffraction angle, and β (15.94) is the full-width at half-maximum (FWHM) measured in 2θ and expressed in radians. The average crystalline size of the thin film of MEH-PPV polymer was 11.1 nm, close to previous work that reported a grain size of 12.3 nm.8 The XRD profile of the MEH-PPV thin layer showed a broadened crystalline peak with a FWHM of 15.94°, indicating a roughly amorphous nature of the polymer.9,10

Optical Properties of the Thin Film

Figure 4 shows the optical absorbance spectrum in the wavelength range from 400 nm to 700 nm for the MEH-PPV thin film deposited on glass substrate after annealing at 125°C. The absorbance profile presents a broad peak at 511 nm. The optical bandgap was determined using the extrapolation of the absorbance plot according to the relation11

$$ E_{{{\text{onset}}}}^{{{\text{opt}}}} = \frac{hc}{{\lambda_{{{\text{onset}}}} }}, $$
(2)
Fig. 4
figure 4

Absorbance versus wavelength for the sprayed MEH-PPV polymer thin layer.

Full size image

where h is Planck’s constant and \(c\) is the speed of light in vacuum. The onset wavelength for the MEH-PPV thin film fabricated in chloroform was 611 nm, giving a \(E_{\rm g}^{\rm opt}\) value of 2.03 eV. The calculated value can be confirmed using the Beer–Lambert law as follows:12

$$ \alpha = 2.303A{/}d $$
(3)

where d is the layer thickness and A is the absorption ratio value. The bandgap value was estimated from the fundamental absorption edge of the films. For the linear plot of \(\alpha h\upsilon \) versus \(h\upsilon \) over a wide range of photon energies, the absorption coefficient can be expressed as12,13

$$ \alpha h\upsilon = B\left( {h\upsilon - E_{{\text{g}}} } \right)^{p}, $$
(4)

where \(B\) is a constant, Eg is the bandgap energy of the material, and \(p\) takes the value \(1/2\) for a direct transition, with the absorption plot showing a single slope.14 Figure 5 shows the plot of \(\left( {\alpha h\upsilon } \right)^{2}\) versus \(h\upsilon\). The intercept of the extrapolation of the straight line to the energy axis gives a \(E_{{\text{g}}}^{{{\text{opt}}}}\) value of 2.02 eV for the MEH-PPV thin film. A value of 2.21 eV is reported in literature, although it is difficult to make a direct comparison due to the various deposition methods, measurement methods and equipment, and sample preparation techniques applied.10,15

Fig. 5
figure 5

(αhυ)2 versus hυ for the sprayed MEH-PPV layer. The value of Eg is indicated.

Full size image

The UV–Vis spectrum of the MEH-PPV film exhibited a single peak at ∽ 511 nm, which can be attributed to the \(\pi - \pi^{*}\) transition of the aromatic polymer MEH-PPV.16

SEM Results

SEM images were taken at several resolutions. As depicted in Fig. 6, a homogeneous MEH-PPV film of good quality was deposited on the glass substrate, exhibiting a smooth surface and good film–substrate adhesion; no irregularities were observed, and the layer was essentially flat. However, the thickness of the sample was not constant, varying around an average of 140 nm as indicated by the arrow in Fig. 6. This variation in the thickness is due to our homemade USP system. The SEM analysis confirmed that the polymer layer had a homogeneous, smooth surface.

Fig. 6
figure 6

Cross-sectional SEM image of MEH-PPV with thickness of 123 nm and 140 nm, taken at magnification of 205,000× at 10 kV. The thickness of the organic layer is indicated by arrows.

Full size image

Current–Voltage (I–V) Characteristics of Al/p-Si/MEH-PPV/Ag Structure

To investigate the electrical behavior of the as-fabricated Al/p-Si/MEH-PPV/Ag organic heterostructure, electrical parameters including n, Rs, and \(\phi_{{\text{B}}}\) were obtained using the forward-bias I–V results, and the Cheung and Nord functions.

The experimental semilogarithmic reverse- and forward-bias I–V characteristics of the Al/p-Si/MEH-PPV/Ag organic heterostructure in the dark at room temperature are shown in Fig. 7.The current–voltage curve is linear on a semilogarithmic scale in the moderate forward-bias voltage region. According to thermionic emission theory, the current can be expressed as16

$$ I = I_{0} \exp \left[ {\frac{{q\left( {V - IR_{{\text{s}}} } \right)}}{nkT}} \right], $$
(5)
Fig. 7
figure 7

Semilogarithmic plot of current–voltage characteristics of Al/P-MEH-PPV/Ag organic device in the dark at room temperature, including the rectification ratio (RR).

Full size image

where I and V are the measured current and applied voltage of the diode, and I0 is the reverse saturation current, which can be expressed as17

$$ I_{0} = AA^{*}T^{2} \exp \left( { - \frac{{q\phi_{{\text{b}}} }}{kT}} \right), $$
(6)

where A is the effective diode area, A* is the effective Richardson constant (equal to 32 A cm−2 K−2 for p-type silicon), T is the absolute temperature in Kelvin, q is the electron charge, n is the ideality factor, and \(\phi_{{\text{b}}}\) is the zero-bias barrier height. The barrier height is the difference between the edge of the majority-carrier band of the semiconductor and the Fermi level at the interface.18 The diode equation based on thermionic emission (TE) theory indicates a linear I–V plot. The value of I0 can be extracted and inserted into Eq. (5) to calculate \(\phi_{{\text{b}}}\), which can be expressed as19

$$ \phi_{{\text{b}}} = \frac{kT}{q}\ln \left( {\frac{{AA^{*}T^{2} }}{{I_{0} }}} \right). $$
(7)

The ideality factor n is greater than unity and shows the deviation between the experimental I–V characteristics of the organic device and ideal TE theory (for which n = 1); it can be calculated using the relation20

$$ n = \frac{q}{kT}\frac{{{\text{d}}V}}{{{\text{d}}\left( {\ln I} \right)}}, $$
(8)

where k is the Boltzmann constant, T is the absolute temperature (300 K), and q is the electron charge. In the ideal case, the ideality factor is equal to 1. However, mostly n is higher than unity (n > 1) and is found to be 3.6, which may be due to the native oxide layer on electrodes21 or other effects, such as the inhomogeneous thickness of the deposited film, the organic layer effect on the structure, etc.12 It is observed that the rectification ratio (RR) of the MEH-PPV organic heterojunction is about 36.26. Cheung’s functions, given in Eq. (9), are used to evaluate \(n\), \(R_{{\text{s}}}\), and \(I_{{\text{b}}}\); these parameters are significant and influence the quality of the electrical conduction in the structure:22,23

$$ \frac{{{\text{d}}V}}{{{\text{d}}\left( {\ln I} \right)}} = IR_{\rm s} + n\left( \frac{kT}{q} \right), $$
(9)
$$ H\left( I \right) = V - \left( \frac{nkT}{q} \right)\ln \left( {\frac{I}{{AA^{*} T^{2} }}} \right), $$
(10)

Here, H(I) can be written as

$$ H\left( I \right) = IR_{{\text{s}}} + n\phi_{{\text{B}}}. $$
(11)

The experimental dV/dln I versus \(I\) plot for the studied structure is presented in Fig. 8. To define H(I) from Eq. (10), it is necessary to set the values of n and Rs from Eq. (9) using the intercept and slope of the curve fit to a straight line. The slope and y-axis intercept of the \(H(I)\) versus \( I\) plot (straight line) shown in Fig. 9 gives the values of Rs and nϕB values using Eq. (11). Having obtained the two values of Rs of 94.39 Ω and 95.70 Ω using the dV/dln I versus I and H(I) versus I plots, respectively, it can be said that Cheung’s method is a consistent approach that is in good agreement with the values. To compare the Rs values obtained using Cheung’s function with the average value of 95.04 Ω obtained using the Norde functions, a third approach is used, defined by the expression21

$$ F\left( V \right) = \frac{V}{\gamma } - \frac{kT}{q}\ln \left( {\frac{I\left( V \right)}{{AA^{*} T^{2} }}} \right), $$
(12)
Fig. 8
figure 8

Experimental dV/d(lnI) versus I plot for Al/p-Si/MEH-PPV/Ag organic MSP heterojunction diode in the dark at room temperature. The linear fit and its equation are indicated.

Full size image
Fig. 9
figure 9

Experimental H(I) versus I plot for Al/p-Si/MEH-PPV/Ag organic MSP diode in the dark at room temperature. The linear fit and its equation are displayed.

Full size image

where γ is the first integer greater than the ideality factor of the diode, and I and V are obtained from the forward-bias I–V characteristics. The barrier height of the structure is given by24

$$ \phi _{{\text{B}}} = F(V_{0} ) + \frac{{V_{0} }}{\gamma } - \frac{{kT}}{q}, $$
(13)

where F(V0) is the minimum value taken from the plot of F(V) versus V, as shown in Fig. 10, and V0 is the corresponding voltage. Moreover, from the Norde function, Rs is obtained using the relation25

$$ R_{{\text{s}}} = \frac{{kT\left( {\gamma - n} \right)}}{{qI_{{{\text{min}}}} }}, $$
(14)
Fig. 10
figure 10

Norde function versus voltage obtained by plotting the forward-bias I–V characteristics of the MEH-PPV organic device in the dark at room temperature. The minimum occurs at V0 and F(V0).

Full size image

where Imin is the minimum value of the current corresponding to the value V0. Using the extracted values of F(V0), V0, and Imin from the F(V) versus V plot, the values of Rs and \(\phi_{{\text{B}}}\) are then determined.

The values of the series resistance and barrier height obtained using Cheung’s approach and Norde method are presented in Table I. Although the values obtained using the three methods are extracted from different regions of the forward-bias I–V characteristic, the values obtained by the previous methods are in very good agreement, with a slight discrepancy that can be neglected. Indeed, the relevant results in this investigation are precise and approximately similar for the barrier height and series resistance.

Table I Experimental values of ideality factor, series resistance, and barrier height extracted by dV/dlnI, Cheung, Norde, and C–V characteristic methods
Full size table

Capacitance–Voltage Characteristics in the Dark at Room Temperature

Capacitance–voltage measurements were carried out at different experimental frequencies (from 10 kHz to 1 MHz) in the dark at room temperature. The built-in voltage Vbi and space charge density are significant parameters describing the behavior of the Al/p-Si/MEH-PPV/Ag organic heterostructure and can be determined from the measured capacitance C by plotting 1/C2 versus V, as described in this subsection. Figure 11 shows the C–V plots of the Al/p-Si/MEH-PPV/Ag structure obtained at two different frequencies of 10 kHz and 1 MHz, revealing symmetric curves that overlap below ~ 0.7 V but differ thereafter, indicating that the capacitance is a function of both voltage and frequency. Note that the C–V curve is higher at the lower frequency.

Fig. 11
figure 11

Capacitance–voltage characteristics of Al/p-Si/MEH-PPV/Ag organic device at two frequencies, 10 kHz and 1 MHz.

Full size image

To determine the barrier height and the interface state density by the capacitance–voltage method, the capacitance is represented by the classic Mott–Schottky relation, expressed as26

$$ \frac{1}{{C^{2} }} = \frac{{2\left( {V_{{\text{D}}} - V} \right)}}{{q\varepsilon_{{\text{S}}} N_{{\text{D}}} A^{2} }}, $$
(15)

where Vbi is the built-in potential, \(\varepsilon_{{\text{S}}}\) is the dielectric constant of the p-type silicon (\(\varepsilon_{{\text{S}}} = 11.7\)), and q is the elementary charge. The doping concentration \(N_{{\text{D}}}\) and diffusion potential \(V_{{\text{D}}}\) values for the diode can be calculated from the intercept and slope of Fig. 12, yielding values of 2.8 × 1013 cm−3 and 0.21 V, respectively, at a frequency of 10 kHz and 5.14 × 1020 cm−3 and 1.88 V at 1 MHz. The barrier height of the Al/p-Si/MEH-PPV/Ag heterojunction device can be expressed as27

$$ \phi_{{\text{B}}} = V_{\rm bi} + \frac{kT}{q}\ln \left( {\frac{{N_{{\text{C}}} }}{N}} \right), $$
(16)
Fig. 12
figure 12

1/C2 versus V for Al/p-Si/MEH-PPV/Ag organic device at (a) 10 kHz and (b) 1 MHz in the dark at room temperature. The linear fit and its equation are indicated.

Full size image

where N is the density of states in the conduction band at room temperature and Vbi is the intercept voltage.27 At 10 kHz and 1 MHz, the built-in voltage is found to be 0.14 V and 0.82 V, respectively; utilizing Eq. (16), the barrier heights of the heterojunction structure are found to be ~ 0.50 V and ~ 0.74 V at both frequencies, using the C2 versus V plots and Eq. (15). Figure 11 shows typical C–V curves of the organic structure at the two distinct frequencies of 10 kHz and 1 MHz, demonstrating the bias voltage and frequency dependence.28 As can be seen, the capacitance increases in parallel in the first region for both frequencies, then reaches a maximum value at each frequency; this tendency is due to the existence of various kinds of interface states.29 No impact of frequency is observed in either the deep depletion (−2 V < V < −0.5 V) or depletion (−0.5 < V 0.4 V) regimes, but the two curves separate from 0.4 V, with the capacitance becoming larger at the lower frequency in the accumulation regime (0.4 V < V < 2 V). To provide further understanding of the capacitance of the device, C−2 is plotted versus V in Fig. 12, revealing straight lines in the reverse voltage region at both 10 kHz and 1 MHz, which indicates uniform behavior of the charge density in the depletion region.

Transport Mechanism in Al/p-Si/P3HT/Ag MSP Heterojunction Diode

When the ideality factor exceeds unity (n > 1), the transport mechanism is no longer dominated by thermionic emission. To comprehend which mechanisms control the conduction in the Al/p-Si/MEH-PPV/Ag structure, the log–log forward-bias current–voltage plot is shown in Fig. 13,30 displaying power-law behavior of the current according to the expression31

$$ I\alpha V^{m}, $$
(17)
Fig. 13
figure 13

Log I versus Log V characteristic of Al/p-Si/MEH-PPV/Ag organic device in the dark at room temperature; the solid red line indicates the two different regions.

Full size image

where m changes with the injection level, likewise being identified with the distribution of trapping centers that differs with the injection level.31 In Fig. 13, two straight line sections are noted, corresponding to m values of 1.6 and 2.7 for the first (V < 0.2 V) and second (V > 0.2 V) region, respectively.

In the first region, the dominant mechanism is Ohmic at lower voltage, with the current being proportional to the voltage.32,33 In the second region, the power-law dependence demonstrates that the current is space-charge limited and that the impact of traps on the conduction current is dominated by the exponential distribution of trapping levels.34,35

Conclusions

An Al/p-Si/MEH-PPV/Ag heterojunction device was fabricated using USP and PVD techniques. Optical and morphological characterization were carried out. The XRD profile of the MEH-PPV thin film revealed its amorphous nature. SEM analysis revealed an average thickness for the deposited organic film of 140 nm. The bandgap value was confirmed to be ~ 2.03 eV by two calculations, in good agreement with literature despite the variety of methods used. Electrical device parameters such as the ideality factor \({\text{n}}\), barrier height \(\phi_{{\text{B}}}\), and average series resistance Rs of the Al/p-Si/MEH-PPV/Ag structure were obtained using different calculation methods from current–voltage and capacitance–voltage measurements in the dark at room temperature. The values of n, \(\phi_{{\text{B}}}\), and Rs are in good agreement with previous values. Detailed study on the conduction mechanism is also reported. Excellent agreement was found between the values obtained using the four methods including Cheung’s, Norde’s, and the capacitance–voltage method.

The diffusion potential is found to be 0.50 and 0.74 at two measurement frequencies of 10 kHz and 1 MHz, respectively, with doping concentrations of 2.8 × 1013 cm−3 and 5.14 × 1020 cm−3. The results of the present experimental investigation could be used to optimize devices based on MEH-PPV or other p-type organic materials for microelectronics applications.