Home
Search
Collections
Journals
About
Contact us
My IOPscience
Achieving high sensitivity in hybrid photodetectors based on an organic single crystal and an inorganic nanocrystal array
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 035202 (http://iopscience.iop.org/0957-4484/25/3/035202) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 144.32.128.51 This content was downloaded on 18/08/2014 at 16:13
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 035202 (5pp)
doi:10.1088/0957-4484/25/3/035202
Achieving high sensitivity in hybrid photodetectors based on an organic single crystal and an inorganic nanocrystal array Dae Sung Chung1 , Yun-Hi Kim2 and Jong-Soo Lee3 1
School of Chemical Engineering and Material Science, Chung-Ang University, Seoul, 156–756, Korea Department of Chemistry and Research Institute of Natural Science (RINS), Gyeongsang National University, Jinju 660-701, Korea 3 Department of Energy Systems Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 711-873, Korea 2
E-mail: [email protected] and [email protected] Received 13 September 2013, revised 3 November 2013 Accepted for publication 18 November 2013 Published 17 December 2013 Abstract
We demonstrate an approach to enhance the photosensitivity of an organic single-crystal photodetector by combining it with a nanocrystal (NC) array. A systematic study of the dependence of the photodetector performance on illumination wavelength and light power together with the charge carrier mobility reveals that charge separation at the single-crystal/CdSe NC interface and subsequent electron trapping in the CdSe NCs generates effective photoconductive gain by hole circulation through the high-mobility single crystal. As a result, the responsivity and specific detectivity reached values up to 50 A W−1 and 2 × 109 cm Hz1/2 W−1 , respectively. Keywords: photodetector, organic semiconductor, single crystal, nanocrystal (Some figures may appear in colour only in the online journal)
1. Introduction
that remain trapped in the semiconductor bulk [9]. In other words, the photoconductive gain is directly proportional to the charge carrier mobility of the majority carriers and the lifetime of the trapped minority carriers. Therefore, a low charge carrier mobility and poor exciton separation in organic semiconductors generally lead to low photoconductive gains and a low responsivity of the photodetector. One of the most widely accepted strategies to overcome such drawbacks of organic semiconductors is fabricating 1D single crystals or nanowires and using those as the active layer of the photodetector [6–8]. Due to the large surface-to-volume ratio of such 1D structures and their high charge carrier mobility (thanks to their defect-free nature), 1D single crystals and nanowires are regarded as suitable candidates for highly responsive photodetectors. For example, Zhang et al reported a qualitative study on the photoconductive characteristics of a single-crystal methyl-squarylium nanowire [6]. Quantitative studies on organic nanowires were also reported by Zhou et al,
Organic semiconductors have attracted enormous attention for their potential applications in low-cost and/or large-area optoelectronic devices [1], including photodetectors. They have many advantageous properties, including tunable light absorption ability, solution-processability and flexibility in structuring the photodetector device architecture [2–8]. Nonetheless, there are disadvantages that make the use of organic semiconductors less preferable than the use of conventional inorganic semiconductors in photodetector applications. Representative examples of the disadvantages are low charge carrier mobility and localized excited states, which make it difficult to generate photoconductive gain. For example, in photoconductor applications, where an active layer and two electrodes are positioned laterally, the photoconductive gain is determined by the recirculation of free charge carriers during the lifetime of the counter carriers 0957-4484/14/035202+05$33.00
1
c 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 035202
D S Chung et al
with a responsivity higher than 100 A W−1 [7]. Recently, a responsivity higher than even 1000 A W−1 was also reported by Ai et al using organic nanowires [8]. In those reports, the origin of the photoconductive gain was attributed to the long lifetime of the surface-trapped electrons. Although these reports are encouraging, we must also note that such deep surface-trapped states can also induce a very low bandwidth of the photodetector, which is inversely proportional to the lifetime of the photogenerated carriers. For example, Ai et al reported a responsivity higher than 1000 A W−1 ; however, due to the existence of deep-lying trapped states, the lifetime of the photogenerated carriers was longer than 100 s, leading to a bandwidth lower than unity [8]. Therefore, it is necessary to strategically design a photodetector so that a high responsivity may be achieved without significantly compromising the bandwidth. With these in mind, in this work, we introduced single crystals of triethylsilylethynyl anthracene end-capped with bithiophene (TESAN-BT) on an array of CdSe nanocrystals (NCs) [10]. Charge separation at the TESAN-BT/CdSe NC interface and subsequent electron trapping in the CdSe NCs generated effective photoconductive gain by hole circulation. A responsivity up to 50 A W−1 was obtained while maintaining a bandwidth of 10 Hz, leading to a specific detectivity (the figure of merit for photodetectors) of 2 × 109 jones, reported for the first time for an organic single-crystal-based photodetector.
Figure 1. (a) Chemical structure of TESAN-BT and (b) the energy
diagram of hybrid photodetector; schematic device structure of (c) pristine photodetector and (d) hybrid photodetector. In (c) and (d), a red long rectangle, corresponding to organic single crystal, was obtained from optical microscope image for the actual single crystal.
1/8 m monochromator combined with a LabView-controlled Keithley 2400 source meter. 3. Results and discussion
Figure 1(a) shows the chemical structure of TESAN-BT. In our previous report, its field effect mobility was found to be up to 3 cm2 V−1 s−1 under optimized device geometry [10]. This value of mobility is among the highest for solutionprocessable small-molecule semiconductors. Considering the energy band diagram shown in figure 1(b) and the fact that the charge carrier mobility of the TESAN-BT single crystal is far higher than that of CdSe NC treated with pyridine, we can assume that the photogenerated electrons are readily trapped by the CdSe NCs, leaving behind unpaired holes in the TESAN-BT single crystal. The free hole carriers in the single crystal then drift to the counter electrode under an applied bias and are collected there. Because the trapped electron is negatively charged, it lowers the potential for hole injection to the channel, resulting in an additional injection of holes. This hole circulating mechanism is repeated several times before it recombines with the trapped electron, generating photoconductive gain [12]. To verify the actual operating mechanism of our photodetector, similarly sized TESAN-BT single crystals were employed to fabricate conventional photodetectors without NCs and hybrid photodetectors with NCs as shown in figures 1(c) and (d). The I–V characteristics of the TESAN-BT single-crystal photodetector in the dark and under illumination by 550 nm wavelength light are shown in figure 2(a). The dark I–V curves both with and without NCs are quasi-symmetric over the bias range from −70 to 70 V and can be fitted well with the power law. The power of ∼2 indicates that the charge transport in these devices is space-charge-limited [13]. Upon light excitation (550 nm, 0.2 mW cm−2 ), the photocurrent increases significantly and reaches up to ∼0.6 µA when the NCs are employed as a charge-trapping layer and ∼0.1 µA in the case of the control device. These photocurrents correspond to responsivities of ∼20 A W−1 and ∼4 A W−1 for the two devices, respectively (figure 2(b)). To further study
2. Experimental details
To fabricate photodetectors, similarly sized TESAN-BT single crystals were deposited on both a pristine SiO2 substrate and a SiO2 substrate covered with a thin NC array. In both cases, a SiO2 substrate was cleaned carefully using piranha solution. The synthesis of the CdSe NCs was accomplished according to previously described procedures [11]. 60 mg of CdO was mixed with 3 g of trioctylphosphine oxide (TOPO), along with 0.28 g of octadecylphosphonic acid. The mixture was stirred under a N2 atmosphere and heated to 300 ◦ C. 1.5 g of trioctylphosphine (TOP) was carefully added, and again the mixture was heated to approximately 390 ◦ C. The TOP-Se (1.7 mmol ml−1 ) solution was then injected into the mixed solution. The size of the NCs was controlled by adjusting the reaction temperature and time. The synthesized NCs were carefully washed with toluene and methanol. The ligand exchange to pyridine was accomplished by mixing the NCs in anhydrous pyridine and stirring at room temperature overnight. Rod-type single crystals of TESAN-BT were grown via a simple drop-casting method on a pre-prepared substrate. The dimensions of the prepared single crystal were approximately 12 µm in width and several hundreds of µm in length. Au electrode deposition was employed to form source–drain electrodes with a channel length of 100 µm. Considering the width of the single crystal used, the active device area is 1.2 × 10−5 cm2 . The electrical characteristics of the two devices were compared by analyzing the photocurrent measured under mechanically chopped monochromatic light illumination from a 150 W xenon arc lamp dispersed with a 2
Nanotechnology 25 (2014) 035202
D S Chung et al
Figure 2. (a) I–V characteristics of pristine and hybrid photodetector, (b) dependence of responsivity on the applied voltage, (c) dependence of responsivity on the optical chopping frequency and (d) transfer curves of single-crystal transistors.
between TESAN-BT and NCs, followed by extraction, contributing to the dark current. Such a thermally generated dark current would be much lower in pristine devices, because there is no intended charge-separating interface. Finally, the photoconductive gain was found to be ∼21 000 and ∼3000 for the photodetectors with and without NCs, respectively. Therefore, the higher photocurrent of the hybrid device can be attributed to its more efficient gain-generation mechanism. Furthermore, the much higher photoconductive gain of the hybrid photodetector indicates that the CdSe NC array acted as a charge-trapping layer. To further understand the high performance of the hybrid photodetector, the light intensity dependence of the photocurrent was studied. Figure 3(a) shows the measured photocurrent (at 550 nm, 100 V) as a function of the illuminated light intensity. The photocurrent exhibits a sublinear dependence on the light intensity, which can be fitted to the power law, I ∼ P0.68 . It is known that the non-unity exponent is a result of the complex processes of electron–hole generation, trapping and recombination within the semiconductor [2]. The responsivity was found to decrease linearly with increasing light intensity in a log–log regime, as shown in figure 3(b). Under the gain-generating photodetector, the relationship between responsivity and incident light power is given by [16]
the operating mechanism of these devices, it is worthwhile to obtain information on the photoconductive gain. The photoconductive gain can be calculated from the following equation [14]: τ µV l2 where τ is the lifetime of the photogenerated carrier, µ is the charge carrier mobility, V is the applied bias and l is the channel length of the device. To obtain the lifetime of the charge carriers, the responsivity was measured as a function of the chopping modulation frequency, as shown in figure 2(c). The 3 dB bandwidth for the hybrid photodetector is lower than that of the pristine photodetector, implying longer lifetimes of the photogenerated carriers trapped in the NCs [15]. The lifetimes of the photogenerated carriers for each device were estimated from the 3 dB bandwidth to be ∼100 ms and 20 ms for the hybrid and pristine photodetectors, respectively. The charge carrier mobility was measured from the transistor analysis. Figure 2(d) shows the transfer characteristics of the two devices. A slightly higher mobility of 0.3 cm2 V−1 s−1 was obtained from the pristine device (compared to 0.22 cm2 V−1 s−1 for the hybrid), possibly due to a more favorable semiconductor/SiO2 interface. Then it is interesting to note that the dark current of a hybrid photodetector was higher than the pristine photodetector although the obtained mobility value is lower in a hybrid photodetector. One possible explanation for the higher dark current in a hybrid photodetector can be attributed to its charge-separating ability. For example, thermally generated carriers (from an impurity or surface state to the LUMO level of TESAN-BT) can be easily separated by the interface Gain =
R∼
1 1 + ( PP0 )n
where P is the incident light power and P0 is the light power at the onset of trap saturation with n as a fitting parameter. In this model, responsivity shows an almost linear 3
Nanotechnology 25 (2014) 035202
D S Chung et al
Figure 3. (a) Dependence of the photocurrent on the light intensity at the bias of 70 V and (b) dependence of the responsivity on the light
intensity at the bias of 70 V.
decrease according to the increase of incident light power at low powers and then becomes sublinear at high powers due to trap saturation effects, in good agreement with our result shown in figure 3(b). These results collectively suggest that the observed photoconductive gain is the result of the circulation of hole carriers through the TESAN-BT single crystal in response to the electrons trapped at either the NC or TESAN-BT/NC interfaces. As shown in figure 4, the responsivity of the hybrid photodetector is highly sensitive to the wavelength of illuminated light and closely follows the thin-film absorption spectra of the TESAN-BT single crystal. This result implies that light absorption by the NCs does not greatly influence the photoconductive gain generation. This could partly be due to the much lower charge carrier mobility in CdSe NC arrays treated with pyridine compared to defect-free single crystals of an organic semiconductor. Finally, the specific detectivity (D∗ ), which is widely accepted as a descriptor of the photodetector’s sensitivity, was calculated. Obtaining D∗ is particularly important because it enables a comparison among photodetectors with different device geometries. Nonetheless, there have been no efforts to measure and report the D∗ of an organic single-crystal-based photodetector. The expression for D∗ is given by [17] √ ABR ∗ D = in
Figure 4. Responsivity of hybrid photodetector and normalized
absorbance of TESAN-BT single crystal as a function of wavelength.
we are working on this issue to further improve the D∗ of organic 1D photodetectors. 4. Conclusions
Organic single-crystal photodetectors operating at room temperature under visible light are demonstrated. In particular, when a CdSe NC array is introduced as a charge-trapping layer, the photodetector exhibited a highly sensitive performance even under weak light illumination. The responsivity and specific detectivity reached values up to 50 A W−1 and 2 × 109 cm Hz1/2 W−1 , respectively. From the illumination wavelength and light power dependence of the device performance, we demonstrated that the hybrid photodetector operates by a photoconductive gain mechanism.
where A is the device area, B is the 3 dB bandwidth, R is the responsivity and in is the noise current. With a measured noise current 1.3 nA Hz−1/2 in the dark, the active area, D∗ , was calculated to be 2 × 109 cm Hz1/2 W−1 . The lower D∗ compared to other organic photodetectors is mainly due to the high noise current. There exist four major sources of noise in photodetectors: Johnson noise, shot noise, generation–recombination (G–R) noise and 1/f noise [11]. Although shot noise, related to the current flowing across potential barriers, is thought to be minimal in single-crystal photodetectors, other noise sources such as G–R noise, which is related to the statistical fluctuation in the rate of generation and recombination of carriers, can be a serious issue, especially in our photodetector where the single-crystal/NC interface plays an important role. Currently,
Acknowledgments
This research was supported by the Chung-Ang University Research Scholarship Grants in 2013 and the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (2012047047). 4
Nanotechnology 25 (2014) 035202
D S Chung et al
References
[9] Soci C, Zhang A, Bao X-Y, Kim H, Lo Y and Wang D 2010 J. Nanosci. Nanotechnol. 10 1430 [10] Yun H-J, Chung D S, Kang I, Park J W, Kim Y-H and Kwon S-K 2012 J. Mater. Chem. 22 24924 [11] Peng Z A and Peng X 2001 J. Am. Chem. Soc. 123 183 [12] Saleh B E A and Teich M C 1991 Fundamentals of Photonics (New York: Wiley) pp 654–7 [13] Chung D S, Lee D H, Yang C, Hong K, Park C E, Park J W and Kwon S-K 2008 Appl. Phys. Lett. 93 033303 [14] Chung D S and Kong H 2013 Opt. Lett. 38 2814 [15] Clifford J P, Konstantatos G, Johnston K W, Hoogland S, Levina L and Sargent E H 2009 Nature Nanotechnol. 4 40 [16] Gao J and Hegmann F A 2008 Appl. Phys. Lett. 93 223306 [17] Lee J-S, Kovalenko M V, Huang J, Chung D S and Talapin D V 2011 Nature Nanotechnol. 6 348
[1] Mishra A and B¨auerle P 2012 Angew. Chem. Int. Edn 51 2020 [2] Kind H, Yan H Q, Messer B, Law M and Yang P D 2002 Adv. Mater. 14 158 [3] Guo F, Xiao Z and Huang J 2013 Adv. Opt. Mater. 1 289 [4] Wang X, Song W, Liu B, Chen G, Chen D, Zhou C and Shen G 2013 Adv. Funct. Mater. 23 1202 [5] Tedde S F, Kern J, Sterzl T, F¨urst J, Lugli P and Hayden O 2009 Nano Lett. 9 980 [6] Zhang X, Jie J, Zhang W, Zhang C, Luo L, He Z, Zhang X, Zhang W, Lee C and Lee S 2008 Adv. Mater. 20 2427 [7] Zhou Y, Wang L, Wang J, Pei J and Cao Y 2008 Adv. Mater. 20 3745 [8] Ai N, Zhou Y, Zheng Y, Chen H, Wang J, Pei J and Cao Y 2013 Org. Electron. 14 1103
5
Search
Collections
Journals
About
Contact us
My IOPscience
Achieving high sensitivity in hybrid photodetectors based on an organic single crystal and an inorganic nanocrystal array
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 035202 (http://iopscience.iop.org/0957-4484/25/3/035202) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 144.32.128.51 This content was downloaded on 18/08/2014 at 16:13
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 035202 (5pp)
doi:10.1088/0957-4484/25/3/035202
Achieving high sensitivity in hybrid photodetectors based on an organic single crystal and an inorganic nanocrystal array Dae Sung Chung1 , Yun-Hi Kim2 and Jong-Soo Lee3 1
School of Chemical Engineering and Material Science, Chung-Ang University, Seoul, 156–756, Korea Department of Chemistry and Research Institute of Natural Science (RINS), Gyeongsang National University, Jinju 660-701, Korea 3 Department of Energy Systems Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 711-873, Korea 2
E-mail: [email protected] and [email protected] Received 13 September 2013, revised 3 November 2013 Accepted for publication 18 November 2013 Published 17 December 2013 Abstract
We demonstrate an approach to enhance the photosensitivity of an organic single-crystal photodetector by combining it with a nanocrystal (NC) array. A systematic study of the dependence of the photodetector performance on illumination wavelength and light power together with the charge carrier mobility reveals that charge separation at the single-crystal/CdSe NC interface and subsequent electron trapping in the CdSe NCs generates effective photoconductive gain by hole circulation through the high-mobility single crystal. As a result, the responsivity and specific detectivity reached values up to 50 A W−1 and 2 × 109 cm Hz1/2 W−1 , respectively. Keywords: photodetector, organic semiconductor, single crystal, nanocrystal (Some figures may appear in colour only in the online journal)
1. Introduction
that remain trapped in the semiconductor bulk [9]. In other words, the photoconductive gain is directly proportional to the charge carrier mobility of the majority carriers and the lifetime of the trapped minority carriers. Therefore, a low charge carrier mobility and poor exciton separation in organic semiconductors generally lead to low photoconductive gains and a low responsivity of the photodetector. One of the most widely accepted strategies to overcome such drawbacks of organic semiconductors is fabricating 1D single crystals or nanowires and using those as the active layer of the photodetector [6–8]. Due to the large surface-to-volume ratio of such 1D structures and their high charge carrier mobility (thanks to their defect-free nature), 1D single crystals and nanowires are regarded as suitable candidates for highly responsive photodetectors. For example, Zhang et al reported a qualitative study on the photoconductive characteristics of a single-crystal methyl-squarylium nanowire [6]. Quantitative studies on organic nanowires were also reported by Zhou et al,
Organic semiconductors have attracted enormous attention for their potential applications in low-cost and/or large-area optoelectronic devices [1], including photodetectors. They have many advantageous properties, including tunable light absorption ability, solution-processability and flexibility in structuring the photodetector device architecture [2–8]. Nonetheless, there are disadvantages that make the use of organic semiconductors less preferable than the use of conventional inorganic semiconductors in photodetector applications. Representative examples of the disadvantages are low charge carrier mobility and localized excited states, which make it difficult to generate photoconductive gain. For example, in photoconductor applications, where an active layer and two electrodes are positioned laterally, the photoconductive gain is determined by the recirculation of free charge carriers during the lifetime of the counter carriers 0957-4484/14/035202+05$33.00
1
c 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 035202
D S Chung et al
with a responsivity higher than 100 A W−1 [7]. Recently, a responsivity higher than even 1000 A W−1 was also reported by Ai et al using organic nanowires [8]. In those reports, the origin of the photoconductive gain was attributed to the long lifetime of the surface-trapped electrons. Although these reports are encouraging, we must also note that such deep surface-trapped states can also induce a very low bandwidth of the photodetector, which is inversely proportional to the lifetime of the photogenerated carriers. For example, Ai et al reported a responsivity higher than 1000 A W−1 ; however, due to the existence of deep-lying trapped states, the lifetime of the photogenerated carriers was longer than 100 s, leading to a bandwidth lower than unity [8]. Therefore, it is necessary to strategically design a photodetector so that a high responsivity may be achieved without significantly compromising the bandwidth. With these in mind, in this work, we introduced single crystals of triethylsilylethynyl anthracene end-capped with bithiophene (TESAN-BT) on an array of CdSe nanocrystals (NCs) [10]. Charge separation at the TESAN-BT/CdSe NC interface and subsequent electron trapping in the CdSe NCs generated effective photoconductive gain by hole circulation. A responsivity up to 50 A W−1 was obtained while maintaining a bandwidth of 10 Hz, leading to a specific detectivity (the figure of merit for photodetectors) of 2 × 109 jones, reported for the first time for an organic single-crystal-based photodetector.
Figure 1. (a) Chemical structure of TESAN-BT and (b) the energy
diagram of hybrid photodetector; schematic device structure of (c) pristine photodetector and (d) hybrid photodetector. In (c) and (d), a red long rectangle, corresponding to organic single crystal, was obtained from optical microscope image for the actual single crystal.
1/8 m monochromator combined with a LabView-controlled Keithley 2400 source meter. 3. Results and discussion
Figure 1(a) shows the chemical structure of TESAN-BT. In our previous report, its field effect mobility was found to be up to 3 cm2 V−1 s−1 under optimized device geometry [10]. This value of mobility is among the highest for solutionprocessable small-molecule semiconductors. Considering the energy band diagram shown in figure 1(b) and the fact that the charge carrier mobility of the TESAN-BT single crystal is far higher than that of CdSe NC treated with pyridine, we can assume that the photogenerated electrons are readily trapped by the CdSe NCs, leaving behind unpaired holes in the TESAN-BT single crystal. The free hole carriers in the single crystal then drift to the counter electrode under an applied bias and are collected there. Because the trapped electron is negatively charged, it lowers the potential for hole injection to the channel, resulting in an additional injection of holes. This hole circulating mechanism is repeated several times before it recombines with the trapped electron, generating photoconductive gain [12]. To verify the actual operating mechanism of our photodetector, similarly sized TESAN-BT single crystals were employed to fabricate conventional photodetectors without NCs and hybrid photodetectors with NCs as shown in figures 1(c) and (d). The I–V characteristics of the TESAN-BT single-crystal photodetector in the dark and under illumination by 550 nm wavelength light are shown in figure 2(a). The dark I–V curves both with and without NCs are quasi-symmetric over the bias range from −70 to 70 V and can be fitted well with the power law. The power of ∼2 indicates that the charge transport in these devices is space-charge-limited [13]. Upon light excitation (550 nm, 0.2 mW cm−2 ), the photocurrent increases significantly and reaches up to ∼0.6 µA when the NCs are employed as a charge-trapping layer and ∼0.1 µA in the case of the control device. These photocurrents correspond to responsivities of ∼20 A W−1 and ∼4 A W−1 for the two devices, respectively (figure 2(b)). To further study
2. Experimental details
To fabricate photodetectors, similarly sized TESAN-BT single crystals were deposited on both a pristine SiO2 substrate and a SiO2 substrate covered with a thin NC array. In both cases, a SiO2 substrate was cleaned carefully using piranha solution. The synthesis of the CdSe NCs was accomplished according to previously described procedures [11]. 60 mg of CdO was mixed with 3 g of trioctylphosphine oxide (TOPO), along with 0.28 g of octadecylphosphonic acid. The mixture was stirred under a N2 atmosphere and heated to 300 ◦ C. 1.5 g of trioctylphosphine (TOP) was carefully added, and again the mixture was heated to approximately 390 ◦ C. The TOP-Se (1.7 mmol ml−1 ) solution was then injected into the mixed solution. The size of the NCs was controlled by adjusting the reaction temperature and time. The synthesized NCs were carefully washed with toluene and methanol. The ligand exchange to pyridine was accomplished by mixing the NCs in anhydrous pyridine and stirring at room temperature overnight. Rod-type single crystals of TESAN-BT were grown via a simple drop-casting method on a pre-prepared substrate. The dimensions of the prepared single crystal were approximately 12 µm in width and several hundreds of µm in length. Au electrode deposition was employed to form source–drain electrodes with a channel length of 100 µm. Considering the width of the single crystal used, the active device area is 1.2 × 10−5 cm2 . The electrical characteristics of the two devices were compared by analyzing the photocurrent measured under mechanically chopped monochromatic light illumination from a 150 W xenon arc lamp dispersed with a 2
Nanotechnology 25 (2014) 035202
D S Chung et al
Figure 2. (a) I–V characteristics of pristine and hybrid photodetector, (b) dependence of responsivity on the applied voltage, (c) dependence of responsivity on the optical chopping frequency and (d) transfer curves of single-crystal transistors.
between TESAN-BT and NCs, followed by extraction, contributing to the dark current. Such a thermally generated dark current would be much lower in pristine devices, because there is no intended charge-separating interface. Finally, the photoconductive gain was found to be ∼21 000 and ∼3000 for the photodetectors with and without NCs, respectively. Therefore, the higher photocurrent of the hybrid device can be attributed to its more efficient gain-generation mechanism. Furthermore, the much higher photoconductive gain of the hybrid photodetector indicates that the CdSe NC array acted as a charge-trapping layer. To further understand the high performance of the hybrid photodetector, the light intensity dependence of the photocurrent was studied. Figure 3(a) shows the measured photocurrent (at 550 nm, 100 V) as a function of the illuminated light intensity. The photocurrent exhibits a sublinear dependence on the light intensity, which can be fitted to the power law, I ∼ P0.68 . It is known that the non-unity exponent is a result of the complex processes of electron–hole generation, trapping and recombination within the semiconductor [2]. The responsivity was found to decrease linearly with increasing light intensity in a log–log regime, as shown in figure 3(b). Under the gain-generating photodetector, the relationship between responsivity and incident light power is given by [16]
the operating mechanism of these devices, it is worthwhile to obtain information on the photoconductive gain. The photoconductive gain can be calculated from the following equation [14]: τ µV l2 where τ is the lifetime of the photogenerated carrier, µ is the charge carrier mobility, V is the applied bias and l is the channel length of the device. To obtain the lifetime of the charge carriers, the responsivity was measured as a function of the chopping modulation frequency, as shown in figure 2(c). The 3 dB bandwidth for the hybrid photodetector is lower than that of the pristine photodetector, implying longer lifetimes of the photogenerated carriers trapped in the NCs [15]. The lifetimes of the photogenerated carriers for each device were estimated from the 3 dB bandwidth to be ∼100 ms and 20 ms for the hybrid and pristine photodetectors, respectively. The charge carrier mobility was measured from the transistor analysis. Figure 2(d) shows the transfer characteristics of the two devices. A slightly higher mobility of 0.3 cm2 V−1 s−1 was obtained from the pristine device (compared to 0.22 cm2 V−1 s−1 for the hybrid), possibly due to a more favorable semiconductor/SiO2 interface. Then it is interesting to note that the dark current of a hybrid photodetector was higher than the pristine photodetector although the obtained mobility value is lower in a hybrid photodetector. One possible explanation for the higher dark current in a hybrid photodetector can be attributed to its charge-separating ability. For example, thermally generated carriers (from an impurity or surface state to the LUMO level of TESAN-BT) can be easily separated by the interface Gain =
R∼
1 1 + ( PP0 )n
where P is the incident light power and P0 is the light power at the onset of trap saturation with n as a fitting parameter. In this model, responsivity shows an almost linear 3
Nanotechnology 25 (2014) 035202
D S Chung et al
Figure 3. (a) Dependence of the photocurrent on the light intensity at the bias of 70 V and (b) dependence of the responsivity on the light
intensity at the bias of 70 V.
decrease according to the increase of incident light power at low powers and then becomes sublinear at high powers due to trap saturation effects, in good agreement with our result shown in figure 3(b). These results collectively suggest that the observed photoconductive gain is the result of the circulation of hole carriers through the TESAN-BT single crystal in response to the electrons trapped at either the NC or TESAN-BT/NC interfaces. As shown in figure 4, the responsivity of the hybrid photodetector is highly sensitive to the wavelength of illuminated light and closely follows the thin-film absorption spectra of the TESAN-BT single crystal. This result implies that light absorption by the NCs does not greatly influence the photoconductive gain generation. This could partly be due to the much lower charge carrier mobility in CdSe NC arrays treated with pyridine compared to defect-free single crystals of an organic semiconductor. Finally, the specific detectivity (D∗ ), which is widely accepted as a descriptor of the photodetector’s sensitivity, was calculated. Obtaining D∗ is particularly important because it enables a comparison among photodetectors with different device geometries. Nonetheless, there have been no efforts to measure and report the D∗ of an organic single-crystal-based photodetector. The expression for D∗ is given by [17] √ ABR ∗ D = in
Figure 4. Responsivity of hybrid photodetector and normalized
absorbance of TESAN-BT single crystal as a function of wavelength.
we are working on this issue to further improve the D∗ of organic 1D photodetectors. 4. Conclusions
Organic single-crystal photodetectors operating at room temperature under visible light are demonstrated. In particular, when a CdSe NC array is introduced as a charge-trapping layer, the photodetector exhibited a highly sensitive performance even under weak light illumination. The responsivity and specific detectivity reached values up to 50 A W−1 and 2 × 109 cm Hz1/2 W−1 , respectively. From the illumination wavelength and light power dependence of the device performance, we demonstrated that the hybrid photodetector operates by a photoconductive gain mechanism.
where A is the device area, B is the 3 dB bandwidth, R is the responsivity and in is the noise current. With a measured noise current 1.3 nA Hz−1/2 in the dark, the active area, D∗ , was calculated to be 2 × 109 cm Hz1/2 W−1 . The lower D∗ compared to other organic photodetectors is mainly due to the high noise current. There exist four major sources of noise in photodetectors: Johnson noise, shot noise, generation–recombination (G–R) noise and 1/f noise [11]. Although shot noise, related to the current flowing across potential barriers, is thought to be minimal in single-crystal photodetectors, other noise sources such as G–R noise, which is related to the statistical fluctuation in the rate of generation and recombination of carriers, can be a serious issue, especially in our photodetector where the single-crystal/NC interface plays an important role. Currently,
Acknowledgments
This research was supported by the Chung-Ang University Research Scholarship Grants in 2013 and the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (2012047047). 4
Nanotechnology 25 (2014) 035202
D S Chung et al
References
[9] Soci C, Zhang A, Bao X-Y, Kim H, Lo Y and Wang D 2010 J. Nanosci. Nanotechnol. 10 1430 [10] Yun H-J, Chung D S, Kang I, Park J W, Kim Y-H and Kwon S-K 2012 J. Mater. Chem. 22 24924 [11] Peng Z A and Peng X 2001 J. Am. Chem. Soc. 123 183 [12] Saleh B E A and Teich M C 1991 Fundamentals of Photonics (New York: Wiley) pp 654–7 [13] Chung D S, Lee D H, Yang C, Hong K, Park C E, Park J W and Kwon S-K 2008 Appl. Phys. Lett. 93 033303 [14] Chung D S and Kong H 2013 Opt. Lett. 38 2814 [15] Clifford J P, Konstantatos G, Johnston K W, Hoogland S, Levina L and Sargent E H 2009 Nature Nanotechnol. 4 40 [16] Gao J and Hegmann F A 2008 Appl. Phys. Lett. 93 223306 [17] Lee J-S, Kovalenko M V, Huang J, Chung D S and Talapin D V 2011 Nature Nanotechnol. 6 348
[1] Mishra A and B¨auerle P 2012 Angew. Chem. Int. Edn 51 2020 [2] Kind H, Yan H Q, Messer B, Law M and Yang P D 2002 Adv. Mater. 14 158 [3] Guo F, Xiao Z and Huang J 2013 Adv. Opt. Mater. 1 289 [4] Wang X, Song W, Liu B, Chen G, Chen D, Zhou C and Shen G 2013 Adv. Funct. Mater. 23 1202 [5] Tedde S F, Kern J, Sterzl T, F¨urst J, Lugli P and Hayden O 2009 Nano Lett. 9 980 [6] Zhang X, Jie J, Zhang W, Zhang C, Luo L, He Z, Zhang X, Zhang W, Lee C and Lee S 2008 Adv. Mater. 20 2427 [7] Zhou Y, Wang L, Wang J, Pei J and Cao Y 2008 Adv. Mater. 20 3745 [8] Ai N, Zhou Y, Zheng Y, Chen H, Wang J, Pei J and Cao Y 2013 Org. Electron. 14 1103
5
