n-Si heterojunction photodetector by arc discharge technique

n-Si heterojunction photodetector by arc discharge technique

Optik 164 (2018) 395–401 Contents lists available at ScienceDirect Optik journal homepage: www.elsevier.de/ijleo Original research article Prepara...

1MB Sizes 0 Downloads 40 Views

Optik 164 (2018) 395–401

Contents lists available at ScienceDirect

Optik journal homepage: www.elsevier.de/ijleo

Original research article

Preparation of multi-walled carbon nanotubes/n-Si heterojunction photodetector by arc discharge technique Raid A. Ismail a,∗ , Mohammed I. Mohammed b , Luma H. Mahmood c a b c

Department of Applied Science, University of Technology, Baghdad, Iraq Department of Chemical Engineering, University of Suleyman Demirel, Isparta, Turkey Department of Chemical Engineering, University of Technology, Baghdad, Iraq

a r t i c l e

i n f o

Article history: Received 7 February 2018 Accepted 13 March 2018 Keywords: MWCNTs Silicon Heterojunction Photosensitivity Drop casting

a b s t r a c t In this paper, heterojunction photodetector based on drop casting of colloidal multi-walled carbon nanotubes prepared by arc discharge technique on single crystalline silicon was demonstrated. The structural and optical of multiwalled carbon nanotubes MWCNTs was investigated by x-ray diffraction XRD, scanning electron microscopy SEM, Fourier transformation infrared spectroscopy FT-IR, and UV–vis spectroscopy. The dark current-voltage I–V characteristics revealed that MWCNTs/Si heterojunction showed a good rectification characteristics and the ideality factor was found to be around 3.1. The photocurrent to dark current ratio was 1.16 × 103 at 8 V bias applied bias and light intensity of 100 mW/cm2 . The photodetector exhibited good linearity characteristics. The photodetector responsivity was estimated as 0.32 A/W at 800 nm at 9 V reverse bias. The open circuit voltage Voc and short circuit current density Jsc of the device were measured.The photodetector has rise time of 30 ns in absence of the external field. © 2018 Elsevier GmbH. All rights reserved.

1. Introduction Carbon nanotubes (CNTs) have received great attention as a new branch of nanomaterials due to their superior physical, chemical, and electronic properties such as long free mean path, useful band gap and high carrier mobility [1,2]. CNTs have been used widely in many fields for applications as batteries [3], hydrogen storage [4], gas sensors [5] carbon nanotubes field effect transistor [6], infrared detectors, solar cells [7] and catalytic application [8]. Many methods were adopted to synthesis CNTs such as arc discharge, laser ablation, chemical vapor deposition, and spray pyrolysis [9–12]. Fabrication and characterization of hybrid photovoltaic multiwalled MWCNTs/n-type Si heterojunction was reported by many authors [13,14]. Ashkan et al. [15] fabricated high photocurrent to dark current ratio single- walled CNTs/Si Schottky contact (MSM structure) using vacuum filtration technique. Solution-processed photodetectors devices propose low cost, large device surface area, physical flexibility and convenient materials integration compared to epitaxially grown, lattice-matched and crystalline semiconductor devices [16,17]. Here, we report on preparation and characteristics study of colloidal multi-walled carbon nanotubes/silicon heterojunction photodetector by arc discharge and drop casting techniques.

∗ Corresponding author. E-mail address: [email protected] (R.A. Ismail). https://doi.org/10.1016/j.ijleo.2018.03.043 0030-4026/© 2018 Elsevier GmbH. All rights reserved.

396

R.A. Ismail et al. / Optik 164 (2018) 395–401

Fig. 1. Cross-sectional view diagram of MWCNTs/ n-Si heterojunction photodetector.

2. Experiment 2.1. Synthesis of MWCNTs Synthesis of MWCNTS was carried out with aid of homemade arc discharge system. The system consists of DC power supply to generate the arc plasma. Pure graphite movable anode cathode rods were used as electrodes. These electrodes were aligned horizontally in stainless steel chamber (250 mm diameter and 400 mm high). The anode electrode drilled with hole with diameter of 5 mm and 6 cm in depth and this hole was filled with mixture of graphite powder and iron as catalyst. Initially the chamber was pumped to the pressure less than 10−1 torr and then filled by argon gas of 99.995% purity. The anode was attached to a linear drive system controlled by computer software, which keeps the predetermined gap distance according to the desired arc voltage after the discharge is initiated via contact ignition. All experiments were done with arc current of about 50 A, discharge voltage of 20 V and the argon pressure of 500 torr inside the chamber. The arc plasma was initiated by contacting two electrodes, and the gaps between the electrodes were carefully controlled to be 1 mm to maintain stable discharge for 5 min. Two types of products were obtained, carbon nanotubes deposit at anode electrode end, and fine particles at deposited on the vessel bottom and internal surface. After that, these products were collected and kept on sample dishes for analysis later. 2.2. Purification of MWCNTs One gram of MWCNTs produced was heated at 350 ◦ C for 30 min to remove the amorphous carbon. After heat treatment, half gram of MWCNTs was dispersed into a flask containing of 20 ml of 70% sodium hypochlorite solution. The solution was then shaken in an ultrasonic bath for 20 min and was heated at 85 ◦ C in a water bath for 3 h to remove metal catalysts. After cooling, the CNTs were washed with deionised water until the pH of the solution reached approximately 7. Finally, the solution was filtered by centrifugal filtration and dried at 200 ◦ C and purified CNTs were obtained. 2.3. Characterization techniques The Structural, morphological and optical properties of MWCNTs were investigated by means of (CuK␣) XRD-6000, Shimadzu x-ray diffractometer, Fourier transformation infrared spectroscopy FTIR, Inspect S50/FEI company scanning electron microscopy SEM and UIR – 210AShimadzu UV–vis spectrophotometer. 2.4. Preparation of MWCNTs/n-Si heterojunction photodiode 0.8 mg carbon nanotubes, 30 ml ethanol (99.99%) and 0.016 mg of citric acid were mixed carefully in beaker with continuous stirring for 24 h. The (111)-oriented silicon substrates used in this study were n-type and having electrical resistivity of 1–3  cm, thermal oxide layer of 200 nm thick has been grown on silicon with aid of wet rapid thermal oxidation technique using high power halogen lamps at 1000 ◦ C for 45 s. Active area indow of 64 mm2 area was made by using HF as etchant acid. Drop casting technique was used to deposit MWCNTs layer silicon substrate. After drop casting, the samples were heated at 90 ◦ C under nitrogen to enhance the MWCNTs adhesion on the silicon substrate. To study the characteristics of photodetector, ohmic contacts have been made by deposition of silver and indium electrodes on MWCNTs and silicon substrate, respectively through special mask using thermal resistive technique. Fig. 1 illustrates the schematic diagram of MWCNTs/n-Si heterojunction photodetector structure. The dark and illuminated I–V characteristics of photodetectors were investigated; the photocurrent was estimated under white light at different intensities. The spectral responsivity of the photodetectors was measured in the range of 350–900 nm by using a monochromator. To calibrate the monochromator, Sanwa silicon power meter was used for this purpose. The rise

R.A. Ismail et al. / Optik 164 (2018) 395–401

397

Fig. 2. SEM image of MWCNTs deposited on silicon.

Fig. 3. XRD pattern of MWCNTs.

time of the photodetector was investigated by using pulsed laser diode (␭ = 640 nm, pulse width of 40 ns and average power of 1 mW) and storage osciliscope with band width of 300 MHz. 3. Results and discussion The SEM image of carbon nanotubes deposited on silicon substrate is shown in Fig. 2, it clearly seen that a large numbers of MWCNTs fibrous (random network) and some nanoparticles of catalyst have been observed. Monodisperse CNTs was observed with average diameter of 45 nm and of several micrometers in length. The XRD pattern in Fig. 3 consists of two broad peaks located at 2␪ = 26◦ and 43◦ which corresponds to (002) and (100) planes, respectively.These two diffraction peaks are belonged to graphite. The presence of strong peak at 2␪ = 26◦ can be indexed to the reflection (002) of carbon (characteristic peak of carbon material) [18]. The diffraction peak at 29◦ is indexed to activated carbon [19]. The average crystallite height Lc (C002 ) of MCNTs was determined using Scherrer equation: Lc =

k ˇcos

(1)

Where ␤ is the full width at half maximum (FWHM), ␪ is the diffraction angle, ␭ is the wavelength (1.54 Å), and k is the Scherrer constant ∼0.91. The XRD experimental data revealed that the crystallite size of synthesised MWCNTs was around 5 nm. Fig. 4 illustrates the FT-IR (transmission mode) spectrum of MWCNTs, noticeable peak at 2900 cm−1 was noticed which could be assigned to the C H asymmetric and symmetric stretching vibration [20]. The peak at 1570 cm−1 is assigned to C C vibration in aromatic rings. A strong absorption peak has been appeared at 1750 cm−1 which can be indexed to (C O) stretching vibration of carboxyl groups [21]. This band is attributed to carboxylic groups that could suggest a high degree of oxidation of CNTs. The peaks located at ∼1400 and 1200 cm−1 are assigned to oxygen functional groups. Fig. 5 shows the dark I–V characteristics of MWCNTs/Si heterojunction photodetector in the reverse and forward directions. The heterojunction showed a good rectification and the forward current obeys to tunneling-recombination model. No

398

R.A. Ismail et al. / Optik 164 (2018) 395–401

Fig. 4. FT-IR spectrum of transmission mode of MWCNTs.

Fig. 5. Room temperature dark I–V characteristics of MWCNTs/Si heterojunction. Inset is the semi-log of forward current versus voltage plot.

breakdown was noticed at low voltage. The ideality factor of photodiode was found to be 3.1. The ideality factor n of the heterojunction photodetector was calculated from forward I–V characteristics by using the following equation: n=

kT V q ln I Is

(2)

Where Is is the saturation current, k is Boltzmann’s constant, and T is operating temperature. The saturation current was found from semi-log relationship of If –V plot (inset of Fig. 5) The large value of ideality factor can be ascribed to surface defects states [22], residual surface contamination, oxidation of MWCNTs, and interfacial gap between CNTs and silicon substrate [23,24]. The reverse illuminated I–V characteristics of synthesised heterojunction photodetector at different white light intensities are given in Fig. 6. It is clearly seen that increasing light intensity resulted in increasing the photocurrent due to increasing the generation of e–h pairs. No saturation in photocurrent was noticed after increasing the light intensity up to 120 mW/cm2 indicating the good linearity characteristics (large dynamic range) of synthesised heterojunction photodetector as shown in Fig. 7. The on/off ratio (photocurrent to dark current ratio) at 100 mW/cm2 was calculated from illuminated I–V characteristics and was about 1.16 × 103 at 9 V bias. The optical transmission of MWCNTs film deposited on glass substrate is shown in Fig. 8, the average optical transmission of the film was determined and found to be around 78%. The large value of on/off ratio can be attributed to the high absorption coefficient of MWCNTs layer, large value of minority carriers diffusion length and/or the photo-induced carriers transfer from carbon nanotubes to silicon substrate [25].

R.A. Ismail et al. / Optik 164 (2018) 395–401

Fig. 6. Illuminated I–V characteristics of MWCNTs/Si heterojunction photodetector.

Fig. 7. Linearity characteristics of the photodetector.

Fig. 8. Optical transmission of colloidal MWCNTs.

Fig. 9. Variation of Voc and Jsc of the photodetector with light intensity.

399

400

R.A. Ismail et al. / Optik 164 (2018) 395–401

Fig. 10. Spectral responsivity plot of photodetector at 9 V bias. Inst is the varation of EQE with wavelength.

Fig. 11. Normalized photodetector performance as function of storage time.

The dependence of photovoltaic properties of heterojunction on light intensity is illustrated in Fig. 9. The maximum values of open circuit voltage Voc and short circuit current density Jsc at 100 mW/cm2 were 350 mV and 1.5 mA/cm2 , respectively. These values are lower than that of MWCNTs/n-Si heterojunction prepared by other methods [26,27]. We think that presence of the oxidative agent on the surface of MWCNTs is negatively affecting the photovoltaic properties of the heterojunction through formation of the trapping centers and dangling bonds. The spectral responsivity R␭ with error bars plot of MWCNTs/n-Si heterojunction photodetector at 9 V reverse bias is given in Fig. 10; the result confirmed that the photodetector has good responsivity in the region extended from visible to near infrared. Fig. 9 confirmed that there are two response regions; the first spectral operating region was 400–600 nm and the second spectral operating region 600–800 nm due to absorption edge of silicon substrate. The maximum responsivity was around 0.32 A/W at 800 nm. The maximum external quantum efficiency EQE (␩) of MWCNTs/Si photodetector was estimated and found to be 65% at 600 nm as shown in the inset of Fig. 9 which is comparable to that for MWCNTs/Si prepared by CVD method reported by Aramo et al. [28]. A semi-flat response in the spectral range of 600–800 nm was observed. The high responsivity and quantum efficiency can be strongly attributed to combination of the charges separation at MWCNT-Si interface as well as transportation and collection the charge carriers through MWCNTs network [13]. The measurement of spectral responsivity was repeated for three photodetectors fabricated in the same batch. No significant fluctuation in the responsivity was noticed indicating the good reproducibility in the detectors fabrication route. The specific detectivity D* of the photodetector has been calculated by using the following equation: D* =

R √Af In

(3)

Where A is the sensitive area, f is the bandwidth and In is the noise current. The detector exhibits D* ∼3 × 1012 W−1 Hz1/2 cm at 800 nm under 9 V reverse bias. No significant degradation in the figures of merit of the heterojunction was observed after 90 h of storing in laboratory. To investigate the stability of the photodetector, evolution of the photodetector performance in time is presented in Fig. 11. This figure demonstrates that no remarkable degradation in short circuit current density of the photodetector and responsivity at 800 nm after 90 h of ambient storage. Fig. 11 clearly shows that the value of Voc decreased to 65% of its initial value after 90 h of storing in normal environment indicating that Voc is air-sensitive parameter. Fig. 12 shows the laser pulse recorded by MWCNTs/Si photodetector without external bias, the rise time was measured from 10% to 90% of output signal and found to be around 30 ns. The fall time is longer than the rise time indicating that the

R.A. Ismail et al. / Optik 164 (2018) 395–401

401

Fig. 12. Impulse response of MWCNTs/n-Si photodetector without external bias. Inset is the photograph of impulsed response of the photodetector.

response time is non dominated by RC product but by photogenerated transit time [29]. The long tail also can be ascribed to the effect of surface trapping states and series resistance [30]. 4. Conclusions In this work, we present a simple and novel technique to fabricate high speed and high performance MWCNTs/n-Si heterojunction photodetector by drop casting of colloidal carbon nanotubes prepared by arc discharge on monocrystalline silicon substrate. The CNTs was characterized by XRD, FTIR, SEM and UV–vis spectroscopy. Heterojunction photodetector made by this technique showed good diode-like characteristics with small leakage current and large rectification ratio. The photodetector has high responsivity to white light and exhibited a large Iph /Id ratio. The responsivity has flat response in 600–800 nm region with maximum quantum efficiency of 63% at 600 nm. The experimental results confirmed that the synthesized heterojunction was air-stable. The fabricated photodetector exhibited fast response characteristics with rise time around 30 ns. On the light of obtained result the photodetectors made by this technique are promising and encouraged for short pulses and low power visible light detection application. Acknowledgment The authors would like to express grateful thanks to the Arab Science and Technology Foundation (ASTF) for funding and supporting this work. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28]

H. Alike, T. Ando, J. Phys. Soc. Jpn. 62 (1993) 1255. T. Durkop, A. Getty, E. Cobas, M. Fuhrer, Nano Lett. 4 (2004) 35. B. Landi, M. Ganter, C. Cress, R. DiLeo, R. Raffaelle, Energy Environ. Sci. 2 (2009) 638. C. Liu, Y. Chen, C. Wu, S. Xu, H. Cheng, Carbon 48 (2010) 452. S. Rajaputra, R. Mangu, P. Clore, Dali Qian, R. Andrews, V. Singh, Nanotechnology 19 (2008) 345502. H. Postma, T. Teepen, Z. Yao, M. Grifoni, C. Dekker, Science 293 (2001) 76. G. Dirk, G. Rahman, M. Prato, N. Jux, S. Qin, W. Ford, Angew. Chem. 117 (2005) 2051. V. Jourdain, C. Bichara, Carbon 58 (2013) 2. A. Hosseini, M. Allahyari, S. Besheli, Int. J. Sci. Environ. Technol. 1 (2012) 217. S. Arepalli, J. Nanosci. Nanotechnol. 4 (2004) 317. G. Che, B. Lakshmi, C. Martin, E. Fisher, Chem. Mater. 10 (1998) 260. A. Elguézabal, W. Antúnez, G. Alonso, F. Delgado, F. Espinosa, Diamond Relat. Mater. 15 (2006) 1329. P. Ong, W. Euler, I. Levitsky, Appl. Phys. Lett. 96 (2010) 033106. J. Wei, Y. Jia, Q. Shu, Z. Gu, K. Wang, D. Zhuang, G. Zhang, Z. Wang, J. Luo, A. Cao, D. Wu, Nano Lett. 8 (2007) 2317. A. Behnam, J. Johnson, Y. Choi, M. Ertosun, A. Okyay, P. Kapur, K. Saraswat, A. Ural, Appl. Phys. Lett. 92 (2008) 243116. D. Barkhouse, E. Sargent, IEEE Photonics J. 2 (2010) 267. S. McDonald, G. Konstantatos, S. Zhang, P. Cyr, E. Klem, L. Levina, E. Sargen, Nat. Mater. 4 (2005) 138. I. Stamatin, A. Morozan, A. Dumitru, V. Ciupinab, G. Prodanb, J. Niewolskic, H. Figiel, Physica E 37 (2007) 44. B. Girgis, Y. Temerk, M. Gadelrab, I. Abdullah, Carbon Sci. 8 (2007) 95. R. Yudianti, H. Onggo, J. Sudirman, Y. Saito, T. Iwata, J. Azuma, Open Mater. Sci. J. 5 (2011) 242. S. Chen, W. Shen, G. Wu, D. Chen, M. Jiang, Chem. Phys. Lett. 402 (2005) 302. F. Yakuphanoglu, Y. Caglar, M. Caglar, S. Ilican, Mater. Sci. Semicond. Process. 13 (2010) 137. L. Qin, C. Shing, IEEE Electron Dev. Lett. 32 (2011) 51. R. Ismail, A. Ali, K. Hasson, Micro Nano Lett. 7 (2012) 536. R. Ismail, R. Kadhim, W. Abdulridh, Optik 127 (2016) 8144. F. Nicola, M. Salvato, C. Cirillo, M. Crivellari, M. Boscardin, M. Scarselli, F. Nanni, I. Cacciotti, M. Crescenzi, P. Castrucci, Carbon 101 (2016) 226. P. Castrucci, C. Scilletta, S. Gobbo, M. Scarselli, L. Camilli, M. Simeoni, B. Delley, A. Continenza, M. Crescenzi, Nanotechnology 22 (2011) 115701. C. Aramo, M. Ambrosio, C. Bonavolontà, M. Boscardin, M. Crivellari, C. deLisio, V. Grossi, P. Maddalena, M. Passacantando, M. Valentino, Nucl. Instrum. Methods Phys. Res. A 845 (2017) 12. [29] R. Ismail, A. Al-Samarai, S. Mohmed, H. Ahmed, Solid-State Electron. 82 (2013) 115. [30] R. Ismail, J. Semicond. Technol. Sci. 6 (2006) 119.