2-1 Introduction
As we know, the best way to retarded the physical decay of P3HT: PCBM mixed system solar cells is to control the morphology. How to maintain an stable morphology with well distributed P3HT and PCMB and high device performance are the key points.
Several studies have reported methods for obtaining a stiff morphology.
Examples among them include using crosslinkable conjugated conducting polymers46 as the active layer and adding a component with good compatibility to P3HT and PCBM to improve the blend morphology. However, these methods involve using new or additional materials in the solar cells and thus would increase the complexity of the fabrication process.
In this study, we developed a film-forming process for the P3HT: PCBM active layer that involved drying the films at a low temperature (-5oC) and thermally annealing them afterward, in the hope of increasing the crystallinity of P3HT and uniformity of the P3HT/PCBM blend. The morphological stability was tested by accelerated tests at elevated temperatures.
17
2-2 Experimental Section
2-2-1 FabricationThe devices were fabricated in the following structure: ITO / PEDOT: PSS / P3HT: PCBM / Ca / Al, showed in Figure 2.1. First, the ITO-coated (~150 nm) glass substrates were cleaned by ultrasonic treatment in detergent, deionized water, acetone and isopropyl alcohol and by plasma for 15 minutes each step. Filtrated PEDOT: PSS by 0.20μm filter were spin-coated on the substrate with the thickness of 50 nm. The substrates were baked at 170oC for 15 minutes in air.
21mg P3HT and 21mg PCBM were dissolved in 1mL 1, 2-dichlorobenene (DCB) and stirred over 8 hours. The P3HT: PCBM mixed solvent were filtrated by a 0.20μm filter and spin-coated at 600 rpm for 30 seconds as the active layer (~200nm). The wet active layer was then subjected to the low-temperature drying process, where it was placed in sealed container at -5oC overnight for complete drying. The active layers of the control devices were dried at room temperatures instead.
The samples were annealed at 190oC for 2 minutes in nitrogen-filled glove box.
Then, 30nm of calcium and 100nm of aluminum were thermal evaporated on the top of the active layer as the cathode in vacuum (< 5x10-6 torr). The devices were encapsulated by UV glue with glass. The active area was 6.25 mm2, defined by a shadow mask.
18
Figure 2- 1 Device structure.
19
2-2-2 Measurement
The current-density-voltage (J-V) characteristics of the devices were measured in air with a Keithley 2400 source meter under simulated AM1.5G irradiation (100 mW cm−2) from a xenon-lamp-based solar simulator (Oriel 92250A-1000); the incident photon-to-electron conversion efficiency (IPCE) was measured with an Oriel® QE/IPCE measurement kit (including a 150W Xe arc light source, a CS260 monochromator, a lock-in amplifier, and a current preamplifier) with an aperture (1.77 mm2) covering the devices to eliminate extrinsic effects. TEM images of the P3HT:PCBM active layer were obtained with a JOEL JEM-1230 system using an accelerating voltage of 100 kV, with samples prepared by immersing cathode-less devices in deionized water to dissolve the PEDOT:PSS under-layer and lift off the P3HT:PCBM layer, and then collecting the floating P3HT:PCBM film with a copper grid. AFM images were obtained with a Veeco Multimode Scanning Probe Microscope operated in the tapping mode. X-ray diffraction (XRD) analysis was carried out using a PANalytical X’Pert Pro Powder Diffractometer with CuKα radiation. UV-Vis spectra were collected with a Jasco V-570 spectrometer.
20
2-3 Results and Discussion
21
2-3-1 Morphology of the Active Layer
Figure 2-2 shows the X-ray diffraction pattern of P3HT/PCBM films on ITO/PEDOT: PSS substrate. The low-temperature-dried (LT) and the room-temperature-dried (RT) samples had equal intensity from the diffraction peak at 5.4o which is assigned to the (100) plane of the P3HT crystals. After a 190oC annealing for 2 minutes, as our expectation the RT samples had an increase in intensity of the (100) peak as a result of increased P3HT crystallinity from P3HT’s re-crystallization. It was interesting that the LT samples showed much more significant intensifying of the (100) peak.Moreover, as shown in Figure 2-3, the LT sample showed additional peaks after annealing: a secondary peak (200) at 10.8o and tertiary peak (300) at 15.9o,47 indicating that the LT films formed a well-organized intraplane structure. These two figures demonstrate that through a low-temperature drying process and thermal annealing, P3HT obtained a higher crystallinity with better ordering.
The morphology of the LT and RT layers was further analyzed with AFM. Figure 2-4 shows the AFM phase images for P3HT/PCBM films in two drying processes.
Before annealing (Figure 2-4a and 2-4b), the LT and RT layers had drastically different surface morphology despite their equal P3HT crystallinity as indicated by the XRD results. After thermal annealing for 2 minutes, re-crystallization made the
P3HT crystals more clearly observable at the surface. As can be seen by comparing Figure 2-4a with Figure 2-4c, the RT films contained ample P3HT fibers that were closely interlaced, which agrees with the typical morphology observed with P3HT:PCBM blends.48 However, such fibers were absent in the LT films, where pellets-like structures of P3HT appeared instead (see Figure 2-4b and Figure 2-4d).
We attribute the distinct morphology of the LT layer to enhanced crystal nucleation of P3HT during drying at low temperature. At the low temperature, the P3HT crystal nuclei were large in number and their simultaneous growth restricted the attainable size of the eventually formed crystals. Thus, small but dense P3HT crystal formed in the entire film. This morphology of P3HT in P3HT: PCBM blends was a novel discovery in the field of polymeric solar cells.
UV-vis spectra are also consistent with the size difference of the LT and RT P3HT crystals observed with AFM. The P3HT absorption peak is at 500 nm with a shoulder at 600 nm. The vibrionic absorption structures at 552 and 602 nm correspond to the π-π* interactions of the thiophene rings and its intensity increases with the size of P3HT crystallites.49 From the wavelength shape in Figure 2-5, these two peaks were apparent in the RT films, while they were much weaker in the LT films.
Besides P3HT’s crystallization, the distribution of PCBM in active layer is also an important concern for BHJ solar cells’ performance. This can be examined
22
with TEM, as P3HT appears brighter under TEM than PCBM due to its lower density (1.10 g cm-3 vs.1.50 g cm-3).50In the TEM images in Figure 2-6a and 2-6b, the contrast (between the brighter P3HT phase and the darker PCBM phase) of the before-annealing RT film was more evident than that of LT film. This indicates that in the RT drying process, P3HT and PCBM had stronger tendency to aggregate with the RT process than with the LT process. The lower tendency to aggregate of the LT process may be because P3HT and PCBM were forced to precipitate and freeze locally.
The variation of contrast of LT films upon annealing (from Figure 2-6b to Figure 2-6d) were more unobvious than that of RT films upon annealing (from Figure 2-6a to Figure 2-6c), which means that the P3HT and PCBM in LT process was more uniform than that in RT films during an annealing process. We inferred that the uniform distribution came from the enhanced nucleation of P3HT crystals in the LT drying process. The dense crystal nuclei of P3HT may obstruct the aggregation of the PCBM and the amorphous P3HT phases. Conversely, though RT films had larger-size P3HT, loose arrangement between P3HT fibers was hard to resist the aggregation of small molecule-PCBM which facilitated to move with a tiny volume. We expected that more stable and uniform active layer by LT drying process could provide better durability in performance of solar cells.
23
4 6 8 0
2500 5000 7500
10000 LT active layer
b)
pristine
Figure 2- 2 XRD patterns of a) the LT and b) the RT active layer, pristine (open circle) and after 2 min/190 ºC annealing (straight line).
10 20
Figure 2- 3 XRD patterns of the LT (black circle) and RT (open circle) active layer after 2min/ 190oC.
24
Figure 2- 4 AFM of P3HT: PCBM film. RT films: a) pristine, c) 2min/190oC, e) stored at 65 ºC for 1368 h in vacuum, and LT films: b) pristine, d) 2min/190oC, f) stored at 65 ºC for 1368 h in vacuum. The image sizes are 3 × 3 μm for the main graphs and 0.5 × 0.5 μm for the insets. Samples for e) and f) were prepared by removing the cathode layers of the stored cells.
25
26
Figure 2- 5 UV-vis spectra of annealed RT (triangle) and LT (circle) films.
300 400 500 600 700 800
RT /annealed LT /annealed
Absorbance (a.u.)
wavelength (nm)
Figure 2- 6 TEM images of the active layer: a) RT without annealing; b) LT without annealing; c) RT with 2 min/190 ºC annealing; d) LT with 2 min/190 ºC annealing. The lengths of the scale bars are 5 μm in the main graphs, 50 nm in the insets.
27
2-3-2 Hole Mobility of the P3HT:PCBM Layer
To confirm the hole mobility of P3HT: PCBM matrix layer with compatible value in BHJ solar cells, we measure the hole mobility shown in Table 2-1.
The hole mobility of LT/annealed films had almost two order improvement from 1.4x10-5 to 1.1x10-3 cm2/Vs than the pristine one. This enhancement was contributed to the increased and well-ordered crystallinity of P3HT in LT active layer, whose interconnected P3HT network would facilitate charge transport. On the other hand, RT dried P3HT: PCBM layers with and without annealing showed near values of hole mobility in same order. Although the increased P3HT crystallinity formed in RT/annealed active layer, slightly lowered hole mobility in RT/annealed films was caused by the over-aggregation of both P3HT and PCBM, reducing the charge transporting passways and interfacial area between donor and acceptor.
Thus we confirmed that LT/annealed P3HT: PCBM layer had the compatible hole mobility with RT dried active layer in the same order of value. The effects of hole mobility to P3HT:PCBM solar cells were introduced in following part.
28
29
2-3-3 Device Performance
Current density–voltage (J–V) measurements of solar cells by RT and LT process reveals in Figure 2-7 and Table 2-2. The pristine RT solar cells were as the standard, made with the same processing condition as reported elsewhere.51
Compare to the pristine RT cells, annealed RT cells had slightly decreased in Jsc . We contributed the descended Jsc came from the over-aggregation with reducing junctions and lower hole mobility shown in Table 2-1. Though Jsc decreased after annealing, the improved Voc maintained PCE at the average about 3.32%.
In contrast to the small change of RT device, LT device had a more than 5 times improvement of the Jsc (from 1.90 to 10.95 mA/cm2) after annealing. Increased PCE from 0.33% to 4.33% was also approached by LT drying process. The significantly improvement in current density and PCE was related to the increase of crystallinity of P3HT, which could be explained below by the experiment with different annealing temperatures.
In Figure 2-8, P3HT’s crystallinity was increased with the annealing temperature.
This improvement was contributed from the characteristic of polymer crystallization.
The melt temperature (Tm) of P3HT was at 226.4oC. Under Tm, higher temperature supplies higher energy for P3HT chains moving to the surface of established P3HT nucleus, which can obtains higher crystallinity. High ordered P3HT in LT films approached a promoted hole mobility (shown in Table 2-1). Moreover, the compact
P3HT crystals could provide more junctions to dissociated excitons and unhindered passways to transport carriers, which might response to the increased Jsc (Table 2-2).
The greater efficiency of carriers transport in the LT-annealed cells also could be observed in IPCE curve. IPCE indicates the ratio of the number of photons incident on a solar cell to the number of generated charge carriers. As shown in Figure 2-9, the IPCE of LT/annealed cells showed a larger efficiency across all wavelengths than the RT/annealed cells. We contributed this high efficiency in LT/annealed films from the compact morphology and high-crystallized P3HT of LT cells’ active layer which can provide more junctions and passway for excitons’ dissociation to transport.
Besides, RT/annealed cells showed a different shape contrasted with LT/annealed cells in IPCE curve. RT/annealed cells had a relatively intensified efficiency around 550-600nm in IPCE, which was accordable to its UV-vis spectrum with a higher absorbance of peaks 552nm and 602nm, as the vibrionic absorption structure of P3HT.
On the contrary, this peaking toward 550-600nm was absent in the LT layer’s IPCE spectrum, conformed the result that the vibrionic absorption structures were weaker in the LT layer due to its small P3HT crystallites.
The enhancement of LT/annealed solar cells was proved from its high crystallinity of P3HT and compact morphology of active layer. Further, the stability of LT/annealed was examined in following part.
30
Table2- 1 Hole mobility of the RT and LT active layers, measured with hole-only devices operated in the space charge limited current regime.
Mobility
Table2- 2 Summary of device performance of ITO/PEDOT: PSS/ P3HT: PCBM/Ca/Al polymer solar cells with different fabrication conditions.
JSC VOC FF PCE
Figure 2- 7 J-V curves of the RT and LT solar cells before and after 2 min/190 ºC annealing
31
4 8
Figure 2- 8 XRD patterns of the LT active layer after annealing: 110oC/ 2min (open circle), 150 ºC/ 2min (gray circle), and 190oC/ 2min (black circle).
Table2- 3 Summary of device performance of ITO/PEDOT: PSS/ P3HT: PCBM/Ca/Al polymer solar cells by low-temperature drying process with different annealing conditions.
300 400 500 600 700 800 0
20 40 60 80 100
IPCE (%)
Wavelength (nm)
RT /annealed LT /annealed
Figure 2- 9 IPCE spectra of the RT and LT cells (annealed at 190 ºC for 2 min).
33
2-3-4 Device Stability
Another issue for BHJ solar cells is the stability. As we know that polymers tend to change their morphology with through time and temperature. This characteristic of P3HT may provide an unstable morphology in active layer with further aggregation of P3HT and PCBM, decreasing the performance of BHJ solar cells by reducing the junctions. Thus, the stability of BHJ solar cells is needed to be concerned.
The first test was whether the morphology of P3HT: PCBM active layer can against longer term of annealing. As shown in Table 2-4, after a 30minuits annealing, RT cells had a 23 % degrade of PCE from 3.23% to 2.49%. However, LT devices had a less degraded rate with 10% decay from 4.07% to 3.70%. We demonstrated the decayed-variation between RT and LT cells came from the different active layer morphology. LT cells had a compact structure of P3HT in active layer, which could efficiently prevent the further migration of PCBM during thermal annealing. On the contrary, larger but less P3HT fibers provided a loose network in RT active layer, which was hard to prohibit the migration of PCBM and obtained a worse performance of BHJ solar cells after longer annealing. The more stable BHJ cells with better performance by LT process also approached in the accelerated test with long-term storage.
Accelerated test were operating in vacuum at 65 ºC over 1300 hr. Original PCE
34
of LT and RT devices were 4.43% and 3.23%. The result is shown in Figure 2-10 and Table 2-5. After 1368 hr, LT cells maintained a 2.15% PCE in performance. From AFM figures, LT/annealed active layer (Figure 2-4f) showed a less phase separation between P3HT and PCBM. However, such morphological stability was absent in the RT layer (Figure 2-4e), as it developed pronounced phase separation upon annealing or extended heating. Also, the RT device showed a severer degraded over 80% from 3.23% to 0.45% during this storing period.
Thus, the morphological stability of LT active layer was verified. BHJ solar cells by LT process not only approached a high efficiency, but also provide a stable performance by its stiff active layer.
35
Table2- 4 Device performance of ITO/PEDOT:PSS/ P3HT:PCBM/Ca/Al polymer solar cells with different anneing temperature.
JSC
Table 2- 5 Device performance of ITO/PEDOT: PSS/ P3HT: PCBM/Ca/Al polymer solar cells in accelerated test. Annealing applied to complete, encapsulated devices stored in vacuum.
Storing condition JSC
[mA cm-2]
Figure 2- 10 Degradation of the power conversion efficiency during storage at 65 ºC in vacuum: RT/annealed (circle) and LT/annealed (square).
36
2-3-5 Models of the LT and RT Morphology
Combine with the results mentioned above, we proposed models of the LT and RT morphology in the following and illustrated it in Figure 2-11
Drying process of P3HT in active layers is constructed from two steps: P3HT’s nucleation and growth. In most used procedure for controlling P3HT: PCBM’s morphology (Figure 2-11a), prolong the drying time is effective to make P3HT’s self-arrangement well which just grows along the established P3HT nucleus. The number of P3HT nucleus is according to the concentration of wet active layer at fixed temperature. In thermal annealing step, PCBM had a great affinity to aggregate, which may compete with P3HT’s migration. The large-scale aggregation of both P3HT and PCBM may decrease the amount of junctions and carrier’s passways, bring in a lower performance and worse durability for BHJ solar cells.
37
In this study, we reduced the temperature to -5oC in drying process to lower the critical concentration of P3HT: PCBM solution in film condition. During LT drying process, P3HT was forced to precipitate (Figure 2-11b); meanwhile both P3HT and PCBM molecular froze until the DCB solvent completely vaporized. Nucleation of P3HT dominated over the LT drying process. In annealing procedure, polymer chains of P3HT could easily approach to the surface of plentiful P3HT nucleus for crystallization, supplying higher crystallinity. Furthermore, these growing P3HT had larger competence to resist the migration of small molecular PCBM. It could afford a
high uniformity and stability in P3HT: PCBM morphology even promotes the performance and durability in BHJ solar cells.
38
Figure 2- 11 Proposed model during annealing process. White region:
PCBM-rich domain; Black region: amorphous P3HT-rich domain; Gray wire:
P3HT crystal.
39
2-4 Summary
In this chapter, we develop a novel process – low-temperature drying process, for manufacturing P3HT: PCBM BHJ solar cells. This LT process effectively assists in the nucleation of P3HT in active layer. After annealing, a quite high crystallinity of P3HT with a densely interconnected network and more uniform distribution of PCBM were obtained in the active layer.
The compact morphology and highly crystallized P3HT in the active layer resulted in increased efficiency and current density through their more efficient exitons dissociation and carrier transport. Meanwhile, the more stable morphology of the LT active layer obstructed aggregation of PCBM, achieving high durability BHJ solar cells.
The findings of this study offer effective solution to the key issues of the P3HT/PCBM-based polymer solar cells: stability and efficiency.
40