energies Article
Efficient Planar Hybrid n-Si/PEDOT:PSS Solar Cells with Power Conversion Efficiency up to 13.31% Achieved by Controlling the SiOx Interlayer Chenxu Zhang 1 , Yuming Zhang 1, *, Hui Guo 1 , Qubo Jiang 2 , Peng Dong 1 and Chunfu Zhang 1, * 1
2
*
ID
Wide Bandgap Semiconductor Technology Disciplines State Key Laboratory, School of Microelectronics, Xidian University, Xi’an 710071, China;
[email protected] (C.Z.);
[email protected] (H.G.);
[email protected] (P.D.) School of Electronic Engineering and Automation, Guilin University of Electronic Technology, No. 1 Jinji Road, Guilin 541004, China;
[email protected] Correspondence:
[email protected] (Y.Z.);
[email protected] (C.Z.)
Received: 25 April 2018; Accepted: 21 May 2018; Published: 30 May 2018
Abstract: In this work, the effects of the SiOx interface layer grown by exposure in air on the performance of planar hybrid n-Si/poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) solar cells are investigated. Compared to the cell with a hydrogen-terminated Si surface, the cell with an oxygen-terminated Si surface reveals improved characteristics in power conversion efficiency, increased from 10.44% to 13.31%. By introducing the SiOx , the wettability of the Si surface can be improved, allowing an effective spread of the PEDOT:PSS solution and thus a good contact between the PEDOT:PSS film and Si. More importantly, it can change the polarity of the Si surface from a negative dipole to a positive dipole, owing to the introduction of the SiOx interface. The Si energy band will bend up and give rise to a favorable band alignment between Si and PEDOT:PSS to promote carrier separation. These results could be potentially employed to further development of this simple, low-cost heterojunction solar cell. Keywords: SiOx interface; n-Si/PEDOT:PSS solar cells; surface polarity
1. Introduction Si/organic hybrid solar cells utilizing poly(3,4-ethylenedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS) combined with an n-type silicon (n-Si) have attracted significant research interest in recent years due to their obvious advantages of a low-temperature process, remarkably low fabrication cost, and potential high efficiency [1]. In this type of hybrid solar cells, as shown in Figure 1, the PEDOT:PSS layer acts as a hole transporting path [2] when forming a heterojunction with the n-type Si substrate. Si has a strong absorption ability in a very wide spectrum range and excellent carrier transport ability. PEDOT:PSS is a water-soluble polymer which has a high conductivity, a transmission window in the visible spectral range, and excellent chemical and thermal stability. This type of Si/PEDOT:PSS hybrid solar cell combines the superior absorption property of Si in a wide spectrum range and the advantage of aqueous solution-based processes of PEDOT:PSS. Over the past few years, considerable progress towards creating highly efficient hybrid n-Si/PEDOT:PSS solar cells has been achieved by adjusting major factors such as electrical conductivity, chemical affinity, and passivation on the device surface [3–9]. Sun et al. introduced perovskite nanoparticles to the PEDOT:PSS layer to generate a positive electrical surface field [10]. Pham et al. presented adding functionalized graphene to the PEDOT:PSS solution based on Silicon nanowire (SiNW), and demonstrated an enhancement of photoelectric conversion efficiency (PCE). Wen et al. [11] fabricated an n+ amorphous silicon layer on
Energies 2018, 11, 1397; doi:10.3390/en11061397
www.mdpi.com/journal/energies
Energies 2018, 11, 1397 Energies 2018, 11, x FOR PEER REVIEW
2 of 10 2 of 10
the back of the c-Si, which could significantly improve the extraction of the photon-generated carrier and suppressofthe of hole electrons at the rear [12]. Despite significant extraction therecombination photon-generated carrier and suppress thecathode recombination of hole electrons efforts at the to improve solar cell performance, the highest power conversion (PCE) reported to date is still rear cathode [12]. Despite significant efforts to improve solarefficiency cell performance, the highest power low when compared to (PCE) the value theoretically by Shockley–Queisser for a single-junction conversion efficiency reported to date isestimated still low when compared to the value theoretically estimated device [13]. by Shockley–Queisser for a single-junction device [13].
Figure 1. Schematic structure of planar hybrid n-Si/PEDOT:PSS solar cells. Figure 1. Schematic structure of planar hybrid n-Si/PEDOT:PSS solar cells.
To further improve the performance of the hybrid n-Si/PEDOT:PSS solar cells, the interface To further improve of the hybrid n-Si/PEDOT:PSS solar cells, the interface properties between n-Sithe andperformance PEDOT:PSS are of great importance. This is because photogenerated properties between n-Si and PEDOT:PSS are of great importance. This is because photogenerated holes in n-Si are required to be effectively transmitted through the heterojunction interface [1] into holes n-Si are required tothen be effectively transmitted theaheterojunction interface [1] inton-the the in PEDOT:PSS layer and collected by the anode.through Therefore, high-quality interface between PEDOT:PSS layer and is then collected by the anode. Therefore, a high-quality interface between n-Si and Si and PEDOT:PSS critical for high-performance hybrid solar cells. However, usually, there are PEDOT:PSS is critical for high-performance hybrid solar cells. However, usually, there are numerous numerous microvoids at the Si/PEDOT:PSS interface because of the insufficient wettability of the Si microvoids at theleads Si/PEDOT:PSS interface because of the recombination. insufficient wettability the Si surface, surface, which to a bad surface coverage and severe A proper of interface layer which to a badnature surface coverage severe recombination. A proper with with leads a hydrophilic between theand PEDOT:PSS and n-Si can improve the interface wettabilitylayer of the Si a surface, which is helpful in the of ideal between the PEDOT:PSS and hydrophilic nature between theformation PEDOT:PSS andinterface n-Si cancontacts improve the wettability of the film Si surface, the Siissubstrate, a more of ideal p–n junction,contacts resulting in less charge recombination which helpful inand thethus formation ideal interface between the PEDOT:PSS film and and athe photocurrent thejunction, same time, the inserted interface may help toand passivate Si larger substrate, and thusdensity a more[14]. idealAtp–n resulting in less chargelayer recombination a larger the surface of n-Si, which could further enhance the device performance. As an effective interface photocurrent density [14]. At the same time, the inserted interface layer may help to passivate the layer between PEDOT:PSS n-Si,enhance its thickness is a critical parameter.As When the interface layerlayer is surface of n-Si, which could and further the device performance. an effective interface too thin, it is insufficient to provide good surface wettability and passivation. However, once the between PEDOT:PSS and n-Si, its thickness is a critical parameter. When the interface layer is too thin, interface layerto is provide too thick,good the photogenerated carriers difficulty when they tunnel through the it is insufficient surface wettability and have passivation. However, once the interface layer insulating layer. is too thick, the photogenerated carriers have difficulty when they tunnel through the insulating layer. Previous work has shown improved solar cell performance when using a thin atomic layer Previous work has shown improved solar cell performance when using a thin atomic layer deposition (ALD) Al2O3 layer [15,16], which was achieved based on increased build-in potential and deposition (ALD) Al2 O3 layer [15,16], which was achieved based on increased build-in potential improved charge collection by the electron blocking of Al2O3. However, the ALD instrument is and improved charge collection by the electron blocking of Al2 O3 . However, the ALD instrument expensive, and the process is complex. Other oxide layers [17,18] are also reported to provide the is expensive, and the process complex. layers are also reported methods to provide crucial interface layer for theishybrid solarOther cells. oxide However, all[17,18] the reported deposition arethe crucial interface layer for the hybrid solar cells. However, all the reported deposition methods complicated. As is well known, when Si is exposed to the air, a native oxide (SiOx) will grow easily.are complicated. As interface is well known, when Si is exposed to the air,We a native (SiO will grow easily. As an efficient layer, its thickness is also important. could oxide remove thex )unintentionally Asgrown an efficient layer, itsthe thickness is also We could the unintentionally oxide interface layer by dipping Si substrate in aimportant. dilute HF solution andremove intentionally regrow the grown oxide layer by dipping the Si substrate in a dilute HF solution and intentionally regrow SiOx layer by exposing the Si substrate to air in a simple and controllable way so that the wanted SiOxthe SiO layer by exposing the Si substrate to air in a simple and controllable way so that the wanted thickness could be obtained. However, whether this oxide layer can effectively improve the interface x SiO could be However,orwhether this layer can effectively improve the quality and enhance theobtained. device performance not is still notoxide well investigated. x thickness In quality this work, effectsthe of device the SiOperformance x layer intentionally grown though exposure to air on the interface andthe enhance or not is still not well investigated. performance of planar hybrid n-Si/PEDOT:PSS solar cells are investigated. Compared to the In this work, the effects of the SiOx layer intentionally grown though exposure tocell air with on the the bare Si surface, the cell with an oxygen-terminated Si surface reveals improved characteristics performance of planar hybrid n-Si/PEDOT:PSS solar cells are investigated. Compared to the cell in with conversion efficiency (PCE), increased from 10.44% to 13.31%. It is inferred thatcharacteristics the formation in thepower bare Si surface, the cell with an oxygen-terminated Si surface reveals improved of oxygen-terminated bonds helpsincreased to build afrom net 10.44% positivetosurface dipole, which promotes carrier power conversion efficiency (PCE), 13.31%. It is inferred that the formation separation and suppresses the Si surface recombination. Our results suggest promising strategies to of oxygen-terminated bonds helps to build a net positive surface dipole, which promotes carrier further exploit the efficiency potential for the simple, low-cost hybrid Si/organic solar cells. separation and suppresses the Si surface recombination. Our results suggest promising strategies to
further exploit the efficiency potential for the simple, low-cost hybrid Si/organic solar cells.
Energies 2018, 11, 1397
3 of 10
Energies 2018, 11, x FOR PEER REVIEW
3 of 10
2. Results and Discussion
2. Results and Discussion
Figure 2 shows the oxide thickness (dox ) of the Si substrates after exposure in air for different time Figure 2 shows the oxide thickness (dox) of the Si substrates after exposure in air for different periods. The thickness is measured by using the J. A. Woollam spectroscopic ellipsometer by fitting time periods. The thickness is measured by using the J. A. Woollam spectroscopic ellipsometer by the ellipsometric data in the wavelength range of 300–1200 nm under an angle of incidence of 75◦ . fitting the ellipsometric data in the wavelength range of 300–1200 nm under an angle of incidence of When75°. theWhen substrate is immersed in the dilute HF solution, the native oxide is easily removed the substrate is immersed in the dilute HF solution, the native oxide is easily removedand andthe Si surface becomes the hydrogen-terminated (H-terminated) surface. The wettability of the H-terminated the Si surface becomes the hydrogen-terminated (H-terminated) surface. The wettability of the HSi surface is poor, it will leadand to aitbad when the PEDOT:PSS is deposited terminated Si and surface is poor, willsurface lead to coverage a bad surface coverage when the PEDOT:PSS is later. later. Once the Si substrate exposed in the air, the to oxygen react with form the Si the andSiOx . Oncedeposited the Si substrate is exposed in theisair, the oxygen tends reacttends withtothe Si and SiOx. After the Si out substrate taken out from the we try measure the SiO Afterform the Sithe substrate is taken from is the HF solution, we HF try solution, to measure theto SiO asx soon x thickness thickness as soon as possible. However, it is inevitable that the Si substrate will come into contact as possible. However, it is inevitable that the Si substrate will come into contact with the air, and this is with the air, and thisnm is why is about 0.87 nm the SiOx0for theexposure, sample with the 0 min exposure, as the why there is about 0.87 SiOxthere for the sample with min as shown in Figure 2. For shown in Figure 2. For the first 60 min of the Si substrate exposure in the air, the SiOx thickness first 60 min of the Si substrate exposure in the air, the SiOx thickness increases nearly linearly as the increases nearly linearly as the time progresses (see the inset in Figure 2). After 360 min, the SiOx time progresses (see the inset in Figure 2). After 360 min, the SiOx thickness tends to saturate at about thickness tends to saturate at about 3.1 nm. For better clarification in the following parts, we still 3.1 nm. For better clarification in the following parts, we still name the sample withsamples 0 min exposure name the sample with 0 min exposure in air as the H-terminated sample and other as the in air asoxygen-terminated the H-terminatedsamples. sample and other samples as the oxygen-terminated samples. 3.5
2.5
2.0
Thickness (nm)
Thickness (nm)
3.0
2.0 1.5
1.0
0.5
1.0 0.5
1.5
0
15
30
45
60
Times (mins)
0
300
600
900
1200
1500
Times (mins) Figure 2. SiOx thickness for the Si substrates exposed in air for different times.
Figure 2. SiOx thickness for the Si substrates exposed in air for different times. Based on the formed SiOx layer, the planar hybrid n-Si/PEDOT:PSS solar cells are fabricated, and 2 Figure on 3a shows the current (J–V) characteristics of the devices under 100 mW/cm Based the formed SiOxdensity–voltage layer, the planar hybrid n-Si/PEDOT:PSS solar cells are fabricated, illumination (AM(air mass) 1.5G), measured from a solar simulator. The J–V characteristics of and Figure 3a shows the current density–voltage (J–V) characteristics of the devices under 100 mW/cm2 photovoltaic (PV) cells can be approximately the Shockley equation: illumination (AM(air mass) 1.5G), measureddescribed from a by solar simulator. The J–V characteristics of 𝑞(𝑉 − 𝑅 𝐽) 𝑉 −Shockley 𝑅𝐽 photovoltaic (PV) cells can be approximately described by the equation:
𝐽=𝐽
exp
−1 +
𝑛𝑘 𝑇
q (V − R s J )
𝑅
−𝐽
(1)
V − RS J
J= J0 exp − J phRsh the shunt resistance, (1) where J0 is the saturation current, Jph the photo current,−R1s the+series resistance, nk B T Rsh n the ideality factor, q the electron charge, kB the Boltzmann constant, and T the temperature. By using with ourcurrent, proposedJ explicit analytic expression method [19], the experimental data can whereEquation J0 is the(1) saturation ph the photo current, Rs the series resistance, Rsh the shunt resistance, be well rebuilt, as shown in Figure 3a, which confirmed the validity of the extracted parameters. By n the ideality factor, q the electron charge, kB the Boltzmann constant, and T the temperature. By using fitting the curves, the photovoltaic parameters of short-circuit current density (Jsc), open circuit voltage Equation (1) with our proposed explicit analytic expression method [19], the experimental data can (Voc), fill factor (FF), and PCE are extracted and summarized in Table 1. Figure 3b shows the variation be well rebuilt, as shown in Figure 3a, which confirmed the validity of the extracted parameters. of device parameters with different exposure times. For the H-terminated device, a PCE of 10.44% is By fitting the with curves, photovoltaic short-circuit current (Jsc ), open circuit achieved a Jsc the of 26.46 mA/cm2, a parameters Voc of 0.605 V,ofand an FF of 65.22%. For density the device with 15 min voltage (V ), fill factor (FF), and PCE are extracted and summarized in Table 1. Figure 3b shows oc exposure in air where the SiOx thickness is 1.12 nm, all the device parameters of Jsc, Voc, and FF are the variation of device with different exposure For The the H-terminated device, a PCE improved and parameters the optimized device performance is times. achieved. optimized device shows a of 2 2 remarkable Jsc ofwith 33.72amA/cm , Voc of 0.607 V,, and FFof of 0.605 65.01%, improves the overall 10.44% is achieved Jsc of 26.46 mA/cm a Voc V, which and an FF of 65.22%. ForPCE the to device with 15 min exposure in air where the SiOx thickness is 1.12 nm, all the device parameters of Jsc , Voc , and FF are improved and the optimized device performance is achieved. The optimized device shows
Energies 2018, 11, 1397
4 of 10
a remarkable Jsc of 33.72 mA/cm2 , Voc of 0.607 V, and FF of 65.01%, which improves the overall PCE Energies 2018, 11, x FOR PEER REVIEW 4 of 10 to 13.31%. For a much longer exposure time, the device performance begins to decrease. At 60 min 2 , V of 13.31%. For a much longer exposure time, device performance begins to of decrease. At 60 min exposure where the SiOx thickness is 1.73 nm,thethe device only shows a Jsc 26.45 mA/cm oc exposure where the SiO x thickness is 1.73 nm, the device only shows a Jsc of 26.45 mA/cm2, Voc of 0.594 0.594 V, and FF of 59.58%, resulting in a PCE of 9.36%. It is guessed that the SiOx can help enhance the V, and FF of 59.58%, resulting in a PCE of 9.36%. It is guessed that the SiOx can help enhance the uniformity of PEDOT:PSS, increase the built-in voltage (Vbi ), and suppress recombination between uniformity of PEDOT:PSS, increase the built-in voltage (Vbi), and suppress recombination between the the Si and PEDOT:PSS interface with a proper thickness, and thus improve the device performance. Si and PEDOT:PSS interface with a proper thickness, and thus improve the device performance. All the All the possible reasons will be discussed the following part. possible reasons will be discussed in the in following part. 0 min 15 min 30 min 45 min 60 min Fitting Curve
2
Current Density (mA/cm )
20
0
-20
-40 -0.2
0.0
0.2
0.4
0.6
Voltage (V)
0.620
34
0.615
32
0.610 0.605
30
0.600
28
0.595
26
0.590
72 24
FF(%)
13
64
12
60
11 10
56 3.2
PCE(%)
14
68
9 3.0 2.5
2.8
2.0
n
Voc(V)
36
2.4
1.5 1.0
2.0
0.5 1.6
Rs(Ω·cm2)
Jsc(mA/cm2)
(a)
0.0 0
10
20
30
40
50
60
Exposure time (min) (b) Figure 3. (a) J–V curves of the planar hybrid Si/PEDOT:PSS solar cells with different exposure times
Figure 3. (a) J–V curves of the planar hybrid Si/PEDOT:PSS solar cells with different exposure under 100 mW/cm2 illumination (AM 1.5 G). (b) Variation of device parameters with different times under 100 times. mW/cm2 illumination (AM 1.5 G). (b) Variation of device parameters with different exposure exposure times. Table 1. Extracted parameters of the planar hybrid Si/PEDOT:PSS solar cells at a temperature of 27 °C. ◦ Table 1. Extracted parameters ofJscthe planar solar Time Voc hybrid FF Si/PEDOT:PSS PCE Rs cells atRash temperature of 27 C.
Time (min) 0 15 30 45 60
(min) (mA/cm2) V % % Voc 0.605 FF65.22 10.44 PCE 0Jsc 26.46 2) % (mA/cm 15 33.72V 0.607 %65.01 13.31 30 28.97 0.599 64.42 11.19 26.46 0.605 65.22 10.44
33.72 28.97 28.784 26.45
0.607 0.599 0.596 0.594
65.01 64.42 60.17 59.58
13.31 11.19 10.33 9.36
n
Ω·cm2 kΩ·cm2 R0.37 Rsh 1.99 n 3.9 s 2 kΩ·cm2 1.79 4.5 Ω·cm 0.17 2.261.99 3.2 0.36 3.9 0.37
1.79 2.26 2.74 2.8
4.5 3.2 5.9 27.3
0.17 0.36 0.32 0.48
Energies 2018, 11, x FOR PEER REVIEW
Energies 2018, 11, 1397
45 60
28.784 26.45
5 of 10
0.596 0.594
60.17 59.58
10.33 9.36
2.74 2.8
5.9 27.3
0.32 0.48
5 of 10
In order to to investigate investigate the thereason reasonforforthethebetter better performance of the optimized device, In order performance of the optimized device, the the capacitance–voltage (C–V) characteristics of the devices are measured by an Agilent 1500 capacitance–voltage (C–V) characteristics of the devices are measured by an Agilent 1500 A at Aa 2–V 2 at a frequency 1 MHz. Figure 4 shows –V plot of H-terminated the H-terminated oxygen-terminated frequency of 1of MHz. Figure 4 shows the the 1/C1/C plot of the andand oxygen-terminated (15 (15 min) hybrid solar cells,and andthere thereisisaalinear linearrelationship relationship between between 1/C and the the applied minmin andand 60 60 min) hybrid solar cells, 1/C22 and applied voltage. Theintercept interceptofofthe theline lineand and x-axis corresponds tobi V Alldevices the devices the voltage. The thethe x-axis corresponds to V [20]. All the showshow the same bi [20]. same V; this means there is no difference thebuilt-in built-involtage voltagefor for the the different different Vbi ofVaround 0.8 V;0.8this means thatthat there is no difference ofofthe bi of around exposure times, which which can can be bepartially partiallyverified verifiedby byobserving observingalmost almostthe the same around 0.60 V, exposure times, same VocVaround 0.60 V, as oc as shown Table Sincethe thedifferent differentexposure exposuretimes timescannot cannotchange change the the V Vbibi, , there there should should be be other shown in in Table 1.1.Since other reasons reasons for for their their different different performances, performances, which which will willbe be investigated investigatedin inthe thefollowing followingsections. sections. 1.0
C-2p(×1014F-2)
0.8
0 min exposure 15 mins exposure 60 mins exposure
0.6
0.4
0.2
0.0 -0.2
0.0
0.2
0.4
0.6
0.8
1.0
Voltage(V) Figure 4. 4. C–V C–V measurements measurements of of the the planar planar Si/PEDOT:PSS Si/PEDOT:PSS solar exposure times. times. Figure solar cells cells with with different different exposure
We measure the contact angles of PEDOT:PSS aqueous solution drops on the H-terminated and We measure the contact angles of PEDOT:PSS aqueous solution drops on the H-terminated and oxygen-terminated substrates to investigate the wettability of the samples. As shown in Figure 5, oxygen-terminated substrates to investigate the wettability of the samples. As shown in Figure 5, although the PEDOT:PSS solution is mixed with Triton X-100 as a surfactant to help the film although the PEDOT:PSS solution is mixed with Triton X-100 as a surfactant to help the film formation, the fresh Si substrate (with 0 min exposure in air) still shows a large contact angle of 68.56°. formation, the fresh Si substrate (with 0 min exposure in air) still shows a large contact angle of With oxygen-terminated bonds instead of the H-terminated bonds, the contact angle begins to 68.56◦ . With oxygen-terminated bonds instead of the H-terminated bonds, the contact angle begins to decrease, as indicated in Table 2. When the exposure time is 60 min, the contact angle is only 27.37°. decrease, as indicated in Table 2. When the exposure time is 60 min, the contact angle is only 27.37◦ . This shows that a longer exposure time will produce thicker SiOx, and then the more H-terminated This shows that a longer exposure time will produce thicker SiOx , and then the more H-terminated bonds will be replaced by the oxygen-terminated bonds. Thus, the wettability of the Si surface is bonds will be replaced by the oxygen-terminated bonds. Thus, the wettability of the Si surface is improved. The remarkably increased wettability of the Si surface after the exposure in air promotes improved. The remarkably increased wettability of the Si surface after the exposure in air promotes easy spreading of the PEDOT:PSS solution, allowing for uniform covering of the surfaces. As a result, easy spreading of the PEDOT:PSS solution, allowing for uniform covering of the surfaces. As a result, the number of surface defects in PEDOT:PSS, such as microvoids, which are easily formed during the the number of surface defects in PEDOT:PSS, such as microvoids, which are easily formed during spin-coating process and give rise to carrier recombination, are significantly reduced. Although a the spin-coating process and give rise to carrier recombination, are significantly reduced. Although a longer exposition time shows a better wettability, it is should be noted that the main operation longer exposition time shows a better wettability, it is should be noted that the main operation principle principle of our hybrid device is the tunneling of holes through the interface oxide layer. A thicker of our hybrid device is the tunneling of holes through the interface oxide layer. A thicker SiOx normally SiOx normally prevents a tunneling process, and eventually holes are likely to be recombined with prevents a tunneling process, and eventually holes are likely to be recombined with electrons. A thicker electrons. A thicker SiOx leads to a high series resistance because of its electrically insulating nature, SiOx leads to a high series resistance because of its electrically insulating nature, which induces a which induces a decreased FF, as shown in Table 1. This could partially explain why we could obtain decreased FF, as shown in Table 1. This could partially explain why we could obtain the highest PCE the highest PCE with high values of Jsc and FF when the interlayer had a thickness of 1.12 nm, even with high values of Jsc and FF when the interlayer had a thickness of 1.12 nm, even though this sample though this sample does not show the best wettability. From this point of view, it seems that the does not show the best wettability. From this point of view, it seems that the wettability change may wettability change may be one reason for their different performance. be one reason for their different performance.
Energies 2018, 11, 1397
6 of 10
Energies 2018, 11, x FOR PEER REVIEW
6 of 10
Energies 2018, 11, x FOR PEER REVIEW
6 of 10
87.86°
27.37°
(a)
(b)
Figure 5. Contact angle images of PEDOT:PSS aqueous solution drops on the (a) H-terminated and
Figure 5. Contact angle images PEDOT:PSS aqueous solution drops on the (a) H-terminated and (b) (b) SiOx-terminated siliconof substrates. SiOx -terminated silicon substrates.87.86° 27.37° Table 2. Contact angles of PEDOT:PSS drops on the H-terminated and oxygen-terminated silicon
Table 2. substrates. Contact angles (a) of PEDOT:PSS drops on the H-terminated (b) and oxygen-terminated 0 15 30 45 60 silicon substrates. Time (min) Figure 5. Contactθ angle of PEDOT:PSS aqueous solution on the (a) H-terminated and (°) images87.86 68.56 48.51 drops39.80 27.37 (b) SiOx-terminated silicon substrates.
Time (min)
0
15
30
45
60
Table 3. RMS values of the planar n-Si/PEDOT:PSS films with different exposure times.
◦) Table 2. Contact of PEDOT:PSS drops on the H-terminated and oxygen-terminated silicon θ (angles 87.86 68.56 48.51 39.80 27.37 0 15 30 45 60 substrates. Time (min) RMS (nm) 2.64 2.57 2.57 2.55 2.29
Time (min)
0
15
30
45
60
Table 3.atomic RMS values of the planar n-Si/PEDOT:PSS filmslayers with different exposure times. θ (°) 87.86 68.56 of PEDOT:PSS 48.51 39.80 27.37 The force microscope (AFM) images coated on the Si substrates are shown in Figure 6. For all the films, the surfaces are relatively uniform with small root-mean-square (RMS) values between 2.29 nm and 2.64 nm, although the30 wettability of the samples different, as Table 3. RMS values of the n-Si/PEDOT:PSS films with different exposure times. Time (min) 0 planar 15 45 60are shown in Table 3. The samples exposed in air for 15 min, 30 min, 45 min, and 60 min after being HFTime 0 15 30 45 60 RMS(min) (nm) 2.64 2.57 2.57 2.55 2.29 etched only present slightly smaller RMS values than those of samples exposed in air for 0 min. RMS (nm) 2.64 2.57 2.57 2.55 2.29 Usually, the smooth surface would deliver fewer defects between the Si substrate and the PEDOT:PSS film to improve the efficiency of photogenerated carrier transportation. However, this should not be The atomic forceforce microscope (AFM) images of PEDOT:PSS layers coated on the Si substrates are The atomic (AFM) images of PEDOT:PSS coated onofthe Si substrates the main reason formicroscope the difference in device performance, sincelayers the differences their RMS valuesare shown in Figure 6. For all the films, the surfaces are relatively uniform with small root-mean-square (RMS) shown in Figure small. 6. For all the films, the surfaces are relatively uniform with small root-mean-square are relatively
values between 2.29between nm and2.29 2.64 the wettability of theofsamples are different, asas shown in (RMS) values nmnm, andalthough 2.64 nm, although the wettability the samples are different, shown in Table 3. The samples exposed in air for 15 min, 30 min, 45 min, and 60 min after being HFTable 3. The samples exposed in air for 15 min, 30 min, 45 min, and 60 min after being HF-etched only only present slightly smaller values than those of samples in air for 0 min. presentetched slightly smaller RMS values thanRMS those of samples exposed in airexposed for 0 min. Usually, the smooth Usually, the smooth surface would deliver fewer defects between the Si substrate and the PEDOT:PSS surface would deliver fewer defects between the Si substrate and the PEDOT:PSS film to improve the film to improve the efficiency of photogenerated carrier transportation. However, this should not be efficiency photogenerated carrier transportation. However, this not be the main reason for the theof main reason for the difference in device performance, since theshould differences of their RMS values difference in device performance, since the differences of their RMS values are relatively small. are relatively small.
（a）
（a）
（b）
（d）
（d）
（b）
（c）
（e）
（c）
（e）
Figure 6. AFM images of PEDOT:PSS films with different exposure times of (a) 0 min; (b) 15 min; (c) 30 min; (d) 45 min; and (e) 60 min.
Energies 2018, 11, x FOR PEER REVIEW
7 of 10
Energies 2018, 11, 1397 images of PEDOT:PSS films with different exposure times of (a) 0 min; (b) 15 min; (c)7 of 10 Figure 6. AFM
30 min; (d) 45 min; and (e) 60 min.
We also structures by by measuring measuring the the We also investigate investigate the the optical optical properties properties of of n-Si/PEDOT:PSS n-Si/PEDOT:PSS structures reflectance spectra. Figure 7 compares the reflectance of polished bare n-Si and n-Si/PEDOT:PSS reflectance spectra. Figure 7 compares the reflectance of polished bare n-Si and n-Si/PEDOT:PSS samples with withexposure exposuretoto different periods 0 min. to 60For min. Forsamples, all the samples, samples airair forfor different timetime periods fromfrom 0 to 60 all the with the with the addition of the PEDOT:PSS film, the reflectance is greatly decreased, which shows addition of the PEDOT:PSS film, the reflectance is greatly decreased, which shows that that the the PEDOT:PSS function being antireflectionlayer. layer.Compared Comparedto tothe the samples samples with with PEDOT:PSS filmfilm hashas thethe function of of being ananantireflection different exposure is is lowest forfor thethe H-terminated andand oxygen-terminated (15 min) different exposuretimes, times,the thereflectance reflectance lowest H-terminated oxygen-terminated (15 devices. The two samples show almost similar optical properties, and this means that the absorption min) devices. The two samples show almost similar optical properties, and this means that the difference isdifference not the main reason for their difference performance. absorption is not the main reason for theirindifference in performance. 0min 15min 30min 45min 60min bare-Si
60
Reflectance (%)
50 40 30 20 10 0
400
600
800
1000
1200
Wavelength (nm) Figure Figure 7. 7. Reflection Reflection spectra spectra of of n-Si/PEDOT:PSS n-Si/PEDOT:PSSsamples samples exposed exposed to to air air for for different different time time periods periods from from 00 to spectrum of of the the polished polished bare bare n-Si n-Si is is also also shown shown for for aa reference. reference. to 60 60 min; min; the the reflection reflection spectrum
According to the above discussion, there are no obvious differences of Vbi, surface morphologies, According to the above discussion, there are no obvious differences of Vbi , surface morphologies, and optical properties for the samples with different exposure times. The wettability change may be and optical properties for the samples with different exposure times. The wettability change may be one reason for their different performance. However, the different wettability does not lead to the one reason for their different performance. However, the different wettability does not lead to the obviously different film morphologies of PEDOT:PSS, since the differences of their RMS values are obviously different film morphologies of PEDOT:PSS, since the differences of their RMS values are relatively small. We believe that beyond the difference in wettability, there are other reasons for the relatively small. We believe that beyond the difference in wettability, there are other reasons for the different device performances. In order to interpret the performance differences for the H-terminated different device performances. In order to interpret the performance differences for the H-terminated and oxygen-terminated (15 min) devices, the possible mechanism is proposed here. Figure 8a,b shows and oxygen-terminated (15 min) devices, the possible mechanism is proposed here. Figure 8a,b the energy band diagram of the H-Si/PEDOT:PSS and oxygen-Si/PEDOT:PSS heterojunctions, shows the energy band diagram of the H-Si/PEDOT:PSS and oxygen-Si/PEDOT:PSS heterojunctions, respectively. The electron affinity of bulk Si as shown is 4.05 eV [21]. Surface properties can either respectively. The electron affinity of bulk Si as shown is 4.05 eV [21]. Surface properties can either increase or decrease the effective electron affinity (Xeff) at the Si surface relative to that in the bulk Si, increase or decrease the effective electron affinity (X ) at the Si surface relative to that in the bulk Si, depending on the polarity of the associated dipole eff [22]. It is reported that the H-Si surface would depending on the polarity of the associated dipole [22]. It is reported that the H-Si surface would pose pose a net negative surface dipole upon the formation of the covalent H-Si surface bonds [21]. This a net negative surface dipole upon the formation of the covalent H-Si surface bonds [21]. This leads leads to an increase in Xeff at the Si/PEDOT:PSS surface and thus a bending down of the Si energy to an increase in Xeff at the Si/PEDOT:PSS surface and thus a bending down of the Si energy band band near the Si/PEDOT:PSS interface, as shown in Figure 8a. As a result, a blocking internal electrical near the Si/PEDOT:PSS interface, as shown in Figure 8a. As a result, a blocking internal electrical field field exists and prevents the photoexcited holes in the Si from injecting into PEDOT:PSS, resulting in exists and prevents the photoexcited holes in the Si from injecting into PEDOT:PSS, resulting in higher higher carrier recombination at the interface. However, the oxygen-terminated Si surface is reported carrier recombination at the interface. However, the oxygen-terminated Si surface is reported to pose a to pose a net positive surface dipole, and this will lead to the drop of Xeff. Accordingly, the Si energy net positive surface dipole, and this will lead to the drop of Xeff . Accordingly, the Si energy band will band will bend up and give rise to a favorable band alignment between oxygen-Si and PEDOT:PSS bend up and give rise to a favorable band alignment between oxygen-Si and PEDOT:PSS to promote to promote carrier separation. Moreover, the SiOx layer can passivate the Si surface and help in carrier separation. Moreover, the SiOx layer can passivate the Si surface and help in suppressing the suppressing the Si surface recombination. This could be partially verified by the increased Jsc, as Si surface recombination. This could be partially verified by the increased J , as shown in Figure 3. shown in Figure 3. However, a thicker SiOx layer would become a barrier forsccharge transport in the However, a thicker SiO layer would become a barrier for charge transport in the cells, leading to a cells, leading to a higherx series resistance. This is why there is an optimized SiOx thickness. higher series resistance. This is why there is an optimized SiOx thickness.
Energies 2018, 11, 1397 Energies 2018, 11, x FOR PEER REVIEW
8 of 10 8 of 10
Ec=4.05eV Ef HOMO=5.0eV Ev=5.17eV
H-terminated Si
X
PEDOT:PSS
（ a）
Ec=4.05eV Ef HOMO=5.0eV Ev=5.17eV
SiOx-Si
PEDOT:PSS
（b）
Figure 8. The band diagram of the (a) and (b) SiO x-Si/PEDOT:PSS hybrid Figure 8. energy The energy band diagram of H-Si/PEDOT:PSS the (a) H-Si/PEDOT:PSS and (b) SiOx -Si/PEDOT:PSS heterojunctions. hybrid heterojunctions.
3. Materials andand Methods 3. Materials Methods 3.1. 3.1. FilmFilm Formation and and Device Fabrication Formation Device Fabrication TheThe devices were fabricated according to the sequences below: TheThe n-type (100) single-sided devices were fabricated according to the sequences below: n-type (100) single-sided polished silicon wafers (resistivity: 0.05–0.1 Ω·cm)Ωwith of 300 of ± 10 cutwere into cut 15 ×into polished silicon wafers (resistivity: 0.05–0.1 ·cm) thickness with thickness 300μm ± were 10 µm 15 mm the substrates. The Si substrates ultrasonically cleaned in acetone, 15 ×squares 15 mmassquares as the substrates. The Si were substrates were ultrasonically cleaned alcohol, in acetone, andalcohol, ionizedand water sequentially for 15 min.forThe native oxide onoxide the substrates were etched by ionized water sequentially 15 min. The native on the substrates were etched immersing the Si substrates in a dilute (5%)HF solution for 30 s and dried by dried N2 forby later Afteruse. by immersing the Si substrates in aHF dilute (5%) solution for 30 s and N2use. for later the After removal of the native bondsbonds on the surface. A highly the removal of theoxide nativelayer, oxidethere layer,were thereH-terminated were H-terminated onSithe Si surface. A highly conductive PEDOT:PSS solution Clevios)was wasfiltered filtered with a polyvinylidene conductive PEDOT:PSS solution(PH1000, (PH1000, Heraeus Heraeus Clevios) with a polyvinylidene fluoride fluoride membrane (0.45porosity) μm porosity) to remove agglomerations. 7% ethylene glycol 1%X-100 Triton membrane (0.45 µm to remove agglomerations. 7% ethylene glycol and 1%and Triton were X-100 were added to the PEDOT:PSS solution to improve the wettability and conductivity. All the Si added to the PEDOT:PSS solution to improve the wettability and conductivity. All the Si substrates substrates were divided fiveand groups and to exposed to air for thetime different time 0 min, were divided into fiveinto groups exposed air for the different periods of 0periods min, 15of min, 30 min, 15 min, 30 min, min,respectively. or 60 min, respectively. The PEDOT:PSS solution was then spin-coated on the 45 min, or 6045min, The PEDOT:PSS solution was then spin-coated on the H-terminated H-terminated (0 SiO min) and SiOx-terminated (exposed to min, air for3015min, min,4530min, min,or4560min, 60 min, (0 min) and (exposed to air for 15 min,orrespectively) x -terminated respectively) speed offor 2500 for 60 After the PEDOT:PSS samples substratessubstrates at a speed at of a2500 rpm 60rpm s. After thes.PEDOT:PSS deposition,deposition, the samplesthe were annealed were onata 190 hot ◦plate 15 min remove the solvent to form a uniform and onannealed a hot plate C for at 15190 min°C to for remove the to solvent to form a uniform and conductive p-type conductive p-type organic thin film. thick Ag filmevaporated was thermally evaporated form a grid organic thin film. A 150-nm thick A Ag150-nm film was thermally to form a grid astothe front contact. as the frontthe contact. Finally, eutectic In–Gatoalloy used fully cover theSirear side oftothe Si an Finally, eutectic In–Gathe alloy was used fullywas cover theto rear side of the substrate form substrate to form an ohmic contact. ohmic contact. 3.2. Device Characterization The morphology measurements of the PEDOT:PSS films were measured by atomic force microscopy (AFM, Agilent 5500). The UV–visible reflection spectra were recorded with a UV–visible reflection spectrophotometer (Perkin-Elmer Lambda 950). The thickness of the SiOx layer was
Energies 2018, 11, 1397
9 of 10
3.2. Device Characterization The morphology measurements of the PEDOT:PSS films were measured by atomic force microscopy (AFM, Agilent 5500). The UV–visible reflection spectra were recorded with a UV–visible reflection spectrophotometer (Perkin-Elmer Lambda 950). The thickness of the SiOx layer was measured by spectroscopic ellipsometry (Jawoolam M-2000). Photovoltaic parameters were measured using a Keithley 2400 source meter under simulated sunlight from a XES-70S1 solar simulator matching the AM 1.5G standard with an intensity of 100 mW/cm2 . The system was calibrated against a National Renewable Energy Laboratory (NREL)-certified reference solar cell. The capacitance–voltage characteristics were measured by an Agilent 1400 semiconductor parameter analyzer. All the measurements of the solar cells were performed under an ambient atmosphere at room temperature without any encapsulation. 4. Conclusions In summary, we demonstrated that the exposure-oxidation treatment of an H-terminated Si substrate could enhance the performance of planar hybrid Si/PEDOT:PSS solar cells. After the oxidation, the Si surface wettability could be improved, which is ascribed to the hydrophilicity of the SiOx layer, allowing an effective spread of the PEDOT:PSS and thus a good contact between the PEDOT:PSS film and Si. More importantly, it can change the polarity of the Si surface from a negative dipole to a positive dipole, owing to the formation of covalent oxygen–Si bonds. The Si energy band bends up and gives rise to a favorable band alignment between oxygen-terminated Si and PEDOT:PSS to promote carrier separation. These results could be potentially employed to further the development of this simple, low-cost heterojunction solar cell. Author Contributions: C.Z. (Chunfu Zhang) and Y.Z. conceived the idea and guided the experiment; C.Z. (Chenxu Zhang) conducted most of the device fabrication and data collection; C.Z. (Chunfu Zhang) revised the manuscript; H.G., Q.J., and P.D. helped with the device measurements. All authors read and approved the manuscript. Acknowledgments: We thank the Natural Science Foundation of China (61604119), Fundamental Research Funds for the Central Universities (Grant No. JBX171103, JB161101, JB161102), Class General Financial Grant from the China Postdoctoral Science Foundation (Grant No. 2016M602771), the Funds for University from Department of Education of Guangxi (KY2015YB109), and the Funds for the Key Laboratory of Automatic Measurement Technology and Instruments (YQ16103). Conflicts of Interest: The authors declare no conflict of interest.
References 1.
2.
3.
4. 5.
6.
Pietsch, M.; Bashouti, M.Y.; Christiansen, S. The role of hole transport in hybrid inorganic/organic silicon/poly(3,4-ethylenedioxy-thiophene):poly(styrenesulfonate) heterojunction solar cells. J. Phys. Chem. C 2013, 117, 9049–9055. [CrossRef] Jäckle, S.; Mattiza, M.; Liebhaber, M.; Brönstrup, G.; Rommel, M.; Lips, K.; Christiansen, S. Junction formation and current transport mechanisms in hybrid n-Si/PEDOT: PSS solar cells. Sci. Rep. 2015, 5, 13008. [CrossRef] [PubMed] Ouyang, J.; Xu, Q.; Chu, C.W.; Yang, Y.; Li, G.; Shinar, J. On the mechanism of conductivity enhancement in poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) film through solvent treatment. Polymer 2004, 45, 8443–8450. [CrossRef] Liu, Q.; Ono, M.; Tang, Z.; Ishikawa, R.; Ueno, K.; Shirai, H. Highly efficient crystalline silicon/Zonyl fluorosurfactant-treated organic heterojunction solar cells. Appl. Phys. Lett. 2012, 100, 183901. [CrossRef] Zhang, Y.; Liu, R.; Lee, S.T.; Sun, B. The role of a LiF layer on the performance of poly(3,4ethylenedioxythiophene):poly (styrenesulfonate)/Si organic-inorganic hybrid solar cells. Appl. Phys. Lett. 2014, 104, 083514. [CrossRef] Zhang, Y.; Zu, F.; Lee, S.T.; Liao, L.; Zhao, N.; Sun, B. Heterojunction with organic thin layers on silicon for record efficiency hybrid solar cells. Adv. Energy Mater. 2014, 4. [CrossRef]
Energies 2018, 11, 1397
7. 8. 9.
10.
11.
12.
13. 14. 15. 16. 17.
18.
19. 20.
21.
22.
10 of 10
Xia, Z.; Song, T.; Sun, J.; Lee, S.T.; Sun, B. Plasmonic enhancement in hybrid organic/Si heterojunction solar cells enabled by embedded gold nanoparticles. Appl. Phys. Lett. 2014, 105, 241110. [CrossRef] Schaadt, D.M.; Feng, B.; Yu, E.T. Enhanced semiconductor optical absorption via surface plasmon excitation in metal nanoparticles. Appl. Phys. Lett. 2005, 86, 063106. [CrossRef] Shen, X.; Xia, Z.; Chen, L.; Li, S.; Zhao, J. Optical and electrical enhancement for high performance hybrid Si/organic heterojunction solar cells using gold nanoparticles. Electrochim. Acta 2016, 222, 1387–1392. [CrossRef] Wang, Y.; Xia, Z.; Liu, L.; Xu, W.; Yuan, Z.; Zhang, Y.; Sirringhaus, H.; Lifshitz, Y.; Lee, S.; Sun, B. The Light Induced Field Effect Solar Cell Concept Perovskite Nanoparticle Coating Introduces Polarization Enhancing Silicon Cell Efficiency. Adv. Mater. 2017, 29. [CrossRef] [PubMed] Van Trinh, P.; Hong, P.N.; Thang, B.H.; Hong, N.T.; Thiet, D.V.; Chuc, N.V.; minh, P.N. Effect of Surface Morphology and Dispersion Media on the Properties of PEDOT:PSS/n-Si Hybrid Solar Cell Containing Functionalized Graphene. Adv. Mater. Sci. Eng. 2017, 2017, 2362056. [CrossRef] Wen, H.; Cai, H.; Du, Y.; Dai, X.; Sun, Y.; Ni, J.; Li, J.; Zhang, D.; Zhang, J. Improving the organic/Si heterojunction hybrid solar cell property by optimizing PEDOT: PSS film and with amorphous silicon as back surface field. Appl. Phys. A 2017, 123, 14. [CrossRef] Chen, T.G.; Huang, B.Y.; Chen, E.C.; Yu, P.; Meng, H.F. Micro-textured conductive polymer/silicon heterojunction photovoltaic devices with high efficiency. Appl. Phys. Lett. 2012, 101, 033301. [CrossRef] He, L.; Jiang, C.; Wang, H.; Lai, D.; Rusli. High efficiency planar Si/organic heterojunction hybrid solar cells. Appl. Phys. Lett. 2012, 100, 073503. [CrossRef] Junghanns, M.; Plentz, J.; Andrä, G.; Gawlik, A.; Höger, I.; Falk, F. PEDOT:PSS emitters on multicrystalline silicon thin-film absorbers for hybrid solar cells. Appl. Phys. Lett. 2015, 106, 083904. [CrossRef] Nam, Y.H.; Song, J.W.; Park, M.J.; Sami, A.; Lee, J.H. Ultrathin Al2 O3 interface achieving an 11.46% efficiency in planar n-Si/PEDOT:PSS hybrid solar cells. Nanotechnology 2017, 28, 155402. [CrossRef] [PubMed] Liu, Y.; Zhang, J.; Wu, H.; Cui, W.; Wang, R.; Ding, K.; Sun, B. Low-temperature synthesis TiOx, passivation layer for organic-silicon heterojunction solar cell with a high open-circuit voltage. Nano Energy 2017, 34, 257–263. [CrossRef] Sheng, J.; Fan, K.; Wang, D.; Han, C.; Fang, J.; Gao, P.; Ye, J. Improvement of the SiOx passivation layer for high-efficiency Si/PEDOT:PSS heterojunction solar cells. ACS Appl. Mater. Interfaces 2014, 6, 16027–16034. [CrossRef] [PubMed] Zhang, C.; Zhang, J.; Hao, Y.; Lin, Z.; Zhu, C. A simple and efficient solar cell parameter extraction method from a single current-voltage curve. J. Appl. Phys. 2011, 110, 064504. [CrossRef] Zhang, J.; Zhang, Y.; Zhang, F.; Sun, B. Electrical characterization of inorganic-organic hybrid photovoltaic devicesbased on silicon-poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate). Appl. Phys. Lett. 2013, 102, 013501. [CrossRef] Hunger, R.; Fritsche, R.; Jaeckel, B.; Jaegermann, W.; Webb, L.J.; Lewis, N.S. Chemical and electronic characterization of methyl-terminated Si(111) surfaces by high-resolution synchrotron photoelectron spectroscopy. Phys. Rev. B Condens. Matter 2005, 72, 045317. [CrossRef] Maldonado, S.; Plass, K.E.; Knapp, D.; Lewis, N.S. Electrical Properties of Junctions between Hg and Si(111) Surfaces Functionalized with Short-Chain Alkyls. J. Phys. Chem. C 2007, 111, 17690–17699. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
