Hybrid Silicon Nanowires for Solar Cell Applications Hybrid Silicon Nanowires for Solar Cell Applications

The global human population has been growing by around 1.1% per year; such growth rate will lead the humanity to cross the 10 billion-people threshold by the end of the first half of this century. Such increase is already putting a huge strain on the nonrenewable sources of energy like fossil fuel, which will run out and come to an end in few decades. Due to these social and economic trends, renewable sources of energy, such as solar cells, have attracted a huge interest as the ultimate alternative to solve humanity’s problems. Among several emerging materials, porous silicon nanowires (PSiNWs) become an active research subject nowadays in photovoltaic application mainly due to its good light trapping effect. The etched nanowires obtained by using metal-assisted chemical etching method (MACE) can reach a low reflection in the visible range. Recently, hybrid silicon nanowires/organic solar cells have been studied for low-cost Si photovoltaic devices because the Schottky junction between the Si and organic material can be formed by solution processes at low temperature. In this chapter, we will present the synthesis of SiNWs and the last progress on the fabrication of hybrid solar cells using various organic semiconductors.


Introduction
Nanoscale silicon has been intensively investigated and explored for its applications in microelectronics, photonics, and biomedical sensors [1][2][3]. Specific efforts have been concentrated in the development of new silicon nanostructures, including quantum dots, nanowires, or porous silicon (PS). Porous silicon has attracted much attention, especially in enhancing photoemission. Much research efforts have been invested to realize an optical device with porous silicon [4][5][6][7][8][9], but the inefficiency [10] and instability [11] of optical characteristic in PS still remain. In addition, silicon nanowires (SiNWs) are also ideal candidate for the study of physics of low dimensional systems. It has potential impact in realizing nanoscale interconnects and functional device elements in future nanoscale electronic and optoelectronic devices [12,13]. The field of SiNWs synthesis represents an exciting and rapidly expanding research area. Considerable efforts have been devoted to the development of versatile and controllable methods for the synthesis of SiNW. These methods can be broadly classified as: (i) bottom-up and (ii) top-down approaches. The bottom-up approach involves the construction of desirable nanostructures from the basic components, i.e., from the atomic level to the nano-or microscale wires.
Top-down approach seeks to fabricate SiNWs from high-quality single crystal silicon wafer or thin film. Silicon nanowires have also been realized using lithographically defined patterns, or spin-coating of nanospheres as etched mask [28] followed by etching of the nanowires using plasma processing technique. Metal-assisted chemical etching method (MACE) is a simple, cost-effective, and powerful semiconductor etching technique that can produce high aspect ratio semiconductor nanostructures. By combining with metal patterning lithography or nonlithographic patterning methods, accurate control of the nanowire orientation (vertical vs. slanted), size (nano vs. microscale), shape, architecture, density, length, and doping characteristics can be achieved readily at wafer scale.
The fabrication of silicon nanowires using the metal-assisted electroless etching method has also been adopted [29][30][31][32][33]. The silver (Ag) ions in an ionic solution of hydrofluoric acid (HF) and hydrogen peroxide (H 2 O 2 ) have been used to prepare the arrays of SiNWs from single crystal wafers [34,35] The effects of various process parameters such as the etchant concentration of H 2 O 2 , etching time, and postetch treatment on the morphology, and optical properties of the SiNWs have also been investigated [36]. The fabrication of nanowires using this method does not require complex sample preparation steps. Furthermore, this technique is effective, having high throughput and low cost. In this chapter, we will present the synthesis of PSiNWs and the last progress of hybrid solar cells based on these nanostructures.

Fabrication of porous silicon nanowires (PSiNWs)
Najar et al. detailed the fabrication of PSiNWs using Ag-assisted electroless etching method from an n-type Si wafer (100) with a resistivity of 0.01-0.02 ohm cm [33]. The Si wafers were cleaned using acetone followed by ethanol for 5 min in an ultrasonic bath. Next, the wafers were immersed in a piranha solution H 2 SO 4 /H 2 O 2 (3:1) for 20 min to remove the organic deposits from the surface. The cleaned wafers were transferred into HF/AgNO 3 solution with a concentration of 4.8/0.005 M for Ag-deposition, followed by rinsing with de-ionized water. Then, the treated Si samples were etched in the HF/H 2 O 2 solution for 45 min. Finally, samples were rinsed again for 10 min with HNO 3 solution to dissolve the excessive Ag nanoparticles (Ag NPs), leaving behind traces of Ag for catalyzing the etching reaction. Four samples were etched using HF concentration of 1.8, 2.8, 4.8, and 5.8 M, respectively, with a fixed H 2 O 2 concentration of 0.5 M.

PSiNWs: Physical properties
Najar et al. studied the etching morphology of the synthesis of PSiNWs [32]. Figure 1(a) shows the scanning electron microcopy image of the vertically oriented PSiNWs having an average length of 13 μm after 75 min of wet etching. The nanowires distribute uniformly on the sample. Figure 1(b) shows the top view with tips of NWs congregates together, and the diameter of the congregated bundles is of several micrometers. These congregated bundles are uniformly distributed on the entire samples and confirmed from the cross-section images. The transmission electron microscopic (TEM) image of an individual nanowire is shown in Figure 1(c), which shows that the nanowire has a diameter of ~150 nm.
A typical 2D bright-field TEM image used for volume reconstruction of the sample is presented in Figure 2(a)-(c). The three successive longitudinal slices obtained from the reconstructed volume revealed a distribution of irregular size/shape vacuum spaces and silicon frame, constituting the PSiNWs.
A typical 3D-tomography observation was conducted using a TEM, and 3D reconstructions were achieved using a simultaneous iterative reconstruction algorithm of consecutive 2D slices in Figure 2(d) [32]. The pore sizes present a distribution from 10 to 50 nm, with an estimated measurement error of 10%, and these pores go inside the nanowire showing similar structure to porous silicon. An average distance between two neighboring pores of lower than 5 nm has been observed. These mesoporous structures are expected to show strong quantum confinement effects.
It is noted that mesopores are mostly found near the top of the nanowires with an average pore size between 10 and 40 nm. AgNPs were detected at the bottom section of the nanowires by EDS analysis in STEM mode. Based on the SEM and TEM analyses, the following qualitative model for the formation of PSiNWs is proposed. In HF/H 2 O 2 solution, part of the Ag is oxidized into Ag + by H 2 O 2 [37]. The Ag + catalyzes the dissolution of Si through the formation of SiF 6

2−
. It is apparent that the transport of Ag + is isotropic while the AgNPs merely drop downward. Hence, the AgNPs facilitated the nanowire formation through vertical chemical etching, while the isotropic diffusion of Ag + promoted lateral chemical etching and pore formation as illustrated in Figure 3(a-i and a-ii) [33].
The Ag + may be chemically reduced by the n-type silicon nanowires [38] forming Ag particle that redeposited onto the sidewalls forming new Ag nucleation sites, and therefore the new localized lateral etching pathway. Furthermore, the nucleation of the AgNPs on the side walls would also reduce the Ag + concentration locally and accelerates the Ag + diffusion in the lateral direction. The laterally diffused Ag + will continue to catalyze the chemical etching of the sidewalls of Si. Since the upper Si nanowires are exposed longer to the solutions, larger Si volume dissolves at the top (and creating more mesopores) than the bottom of the nanowires (Figure 3(b-i and b-ii) and (c-i and c-ii)), being consistent with the TEM image as shown in Figure 2(a). Furthermore, increasing the HF concentration accelerates the process described in our qualitative model, with the Ag  Ag + + e − reversible reaction that continues to penetrate the nanowires laterally to catalyze the HF/H 2 O 2 chemical reaction forming mesopores. Eventually, the mesopores merged leading to a reduction in length and pore density, with the formation of conical PSiNWs as illustrated in Figure 3  In this way, the higher H 2 O 2 concentration increases the amount of out-diffused Ag + , and the increase of doping concentration increases the amount of weak defective points in the silicon lattice; both are important factors for the initiation of additional etching pathways in addition to the vertical etch, which explains the observation of the increased porosity with increasing H 2 O 2 concentration during the etching of highly doped n-Si wafers.
The heavy doping silicon wafer-induced porosification can be understood by a thermodynamic model that describes charge transport across the silicon-electrolyte interface driven by potential energy (Figure 4(a)). To et al. present a model to explain the etching mechanism [39]. In Figure 4(a), the electron transports from Si to Ag + ions, which is composed of the processes I-IV at the Si/electrolyte interface. Electrons at E F of Si wafers migrate to the interface by overcoming a potential energy barrier ∆Φ, consisting of the excitation barrier (E CB -E F , the process I) and the band edge bending in space charge layer (SCL) (E CB , S -E CB , the process II). Owing to the mismatch of E CB,S with E Ag+ , electrons are apt to relaxing to surface states (E SS ) aligned with E Ag+ (the process III) followed by the resonant migration to E Ag+ (the process IV). The increase in the doping level not only up shifts E F toward E CB to reduce ∆Φ, but also makes the SCL thinner to promote electron tunneling through the SCL (Figure 4(b)), to facilitate the process I and II. Hence, a large amount of Ag + ions nucleate on the surface and [Ag + ] (i.e., concentration of Ag + ions) is significantly reduced in the electrolyte. Low [Ag + ] can support only a portion of Ag nuclei to grow and scratch the bulk Si. The rest of the nongrown Ag nuclei will undergo a random scratching in a short period of time to porosify SiNWs and the bulky wafer (Figure 4(c)).
Since the scratching of growing Ag NPs is limited by the diffusion of Ag + ions with low [Ag + ], the random porosification tends to be faster than the NP scratching. Consequently, the electrochemical porosification tends to generate a hybrid structure composed of an array of mpSiNWson mpSi (Figure 4 (a)-(d)). On the contrary, the light and media doping increases ∆Φ to prevent Ag + ions from severely consuming. High [Ag + ] substantially supports most Ag NPs grow and scratch the wafer to create SiNWs, and effectively suppresses the porosification. This model can also account for the creation of mesoporous SiNWs via one-MACE of p++ type Si.

HF concentrations vs. morphology of PSiNWs
The modification of HF or H 2 O 2 concentrations affects the morphology of PSiNWs. In the presence of Ag catalyst, an increase in HF or H 2 O 2 concentration in electroless etching method is analogous to an increase in the current density in electrochemical-based methods [40,41].
In both cases, increasing the H 2 O 2 or HF concentration increases the oxidation rate and dissolution rate, respectively, resulting in nanostructures with varying optical properties. In The variation of NWs length and NWs diameter with increasing HF concentrations was measured and analyzed using SEM and TEM, and the results are tabulated in Table 1. The NWs length shows significant changes, while the NWs diameter varies from 80 to 210 nm for samples ( Figure 5(a), (c), and (e)). However, the aspect ratio reduces to about 9 for NWs etched with (Figure 5(g)) signifying a change in etching mechanism with faster lateral chemical etch.

Effect of HF concentration on PL
It is known that the mesoporous silicon nanostructure gives rise to strong visible emission [42]. Optical properties of the PSiNWs samples have been investigated using PL spectroscopy. After HNO 3 and HF treatments for 5 min to remove the oxide, all samples show similar PL peak wavelength at 800 nm (Figure 6(c)). However, samples without HF treatment show a significant blueshift with PL wavelength peak at 700 nm (Figure 6(a)). This blueshift and increase of PL intensity have been attributed to the presence of defect states in Si-Ox.

This has been correlated with the redshift of PL spectrum that suggests the interfacial between
Si and surface oxide, as well as the quantum confinement effect, and plays a critical role in the light emission process (i.e., the presence of nanocrystallites with size <5 nm that has been confirmed by HRTEM).
Several mechanisms have been proposed to explain the origin of strong PL emission for the indirect bandgap silicon material. These include (i) the quantum confinement effects of free excitons within the Si-nanocrystallites and (ii) SiOx/Si interface defects and/or defect states in the surface oxide, related to Si-O bond [42,44,45]. Comparing the PL spectra before and after the HF treatment, we believe that existence of oxide plays a significant role in changing the emission property of the nanowires. The quantum confinement effect dictates that the characteristic size of the Si-nanocrystallites should be less than the Bohr radius of the free exciton of bulk Si, which is around 5 nm [46]. From the TEM analysis, Si-nanocrystallites of <5 nm has been observed (Figure 7(a)). The inset of Figure 7(a) shows the calculated fast-Fourier transform (FFT) of PSiNWs prepared using 4.8 M HF concentration. The FFT revealed the spatial frequencies corresponding to {111}, {220}, and {311} d-spacing of Si crystal structure. The presence of these d-spacing implies that the PSiNW was oriented along the <110> direction. This was expected as the (001) Si wafer was used to fabricate the NWs, and etching was vertical with respect to the substrate surface. Also, the energy-filtered TEM analysis of the top region of the PSiNW, as shown in Figure 7(b), confirms the existence of Si nanocrystallites and silica amorphous structures, indicating the possible strong PL emission from the highly quantum-confined Si nanocrytallite structures. This prediction correlates positively with our previous discussion on the effect of pore density on PL enhancement. The above argument further confirms our observation that samples etched with HF concentration of 5.8 M (i.e., with low pore and Si-nanocrystallites densities) results in a decrease in PL intensity. The evolution of PL peak intensity with HF concentration is hence consistent with the physical changes in the PSiNWs developed in our etching model. However, SiO 2 alone should not give rise to the PL emission; rather, defect states would contribute to PL emission. Indeed, Najar et al. observed Si 1+ (Si 2 O) and Si 3+ (Si 2 O 3 ) using X-ray hotoelectron spectroscopy (XPS) signatures, which may originate from dangling bonds and volumetric stress that distorted the PSiNWs forming localized defect states.
Therefore, the emissions from surface/defect states in oxide may contribute to the red PL emission, in agreement with the results obtained in [47], in addition to the contribution from quantum confinement effect.

Organic-inorganic hybrid nanowire solar cells
As a platform for cost-effective crystalline Si (c-Si) solar cells, an organic-inorganic hybrid structure has been proposed to replace the conventional p−n junction. The hybrid structure is composed of a transparent conducting polymer and c-Si [48,49]. A Schottky junction at the conducting polymer/c-Si interface can easily be created using a simple spin coating process at room temperature, while a conventional Si p−n junction is formed via a high-temperature doping process [50]. In order to enhance the light-trapping efficiency, Si nanostructures such as nanopyramids [49] and nanowires [51] have been applied, leading to an increase in the short-circuit current density (J SC ).
Han-Dom Um et al. demonstrate an embedded metal electrode for highly efficient organicinorganic hybrid nanowire solar cells as shown in Figure 8 [52]. The electrode proposed here is an effective alternative to the conventional bus and finger electrode, which leads to a localized short circuit at a direct Si/metal contact and has a poor collection efficiency due to a nonoptimized electrode design. In this design, an Ag/SiO 2 electrode is embedded into a Si substrate while being positioned between Si nanowire arrays underneath poly(3,4-ethylenedi oxythiophene):poly(styrenesulfonate) (PEDOT:PSS), facilitating suppressed recombination at This power conversion efficiency is more than twice as high as that of solar cells using a conventional electrode. The microgrid electrode significantly minimizes the optical and electrical losses. This reproducibly yields a superior quantum efficiency of 99% at the main solar spectrum wavelength of 600 nm. In particular, the solar cells exhibit a significant increase in the fill factor of 78.3% compared to that of a conventional electrode (61.4%), this is because of the drastic reduction in the metal/contact resistance of the 1 μm-thick Ag electrode. Hence, the use of this embedded microgrid electrode in the construction of an ideal carrier collection path presents an opportunity in the development of highly efficient organicinorganic hybrid solar cells.
Liu et al. fabricated a hybrid silicon nanowire/polymer heterojunction solar cell combined with a polypyrrole-based supercapacitor as shown in Figure 9, to harvest solar energy and store it [53]. By efficiency enhancement of the hybrid nanowire solar cells and a dual-functional titanium film serving as a conjunct electrode of the solar cell and supercapacitor, the integrated system is able to yield a total photoelectric conversion to storage efficiency of 10.5%, which is the record value in all the integrated solar energy conversion and storage system. This system may not only serve as a buffer that diminishes the solar power fluctuations from light intensity, but also pave its way toward cost-effective high-efficiency self-charging power unit. Finally, an integrated device based on ultrathin Si substrate is demonstrated to expand its feasibility and potential application in flexible energy conversion and storage devices.
Organic-inorganic hybrid solar cells based on n-type crystalline silicon and poly(3,4ethylenedioxythiophene)-poly(styrenesulfonate) exhibited promising efficiency along with a low-cost fabrication process (Figure 10). Zhang et al. fabricated ultrathin flexible silicon substrates, with a thickness as low as tens of micrometers, which were employed to fabricate hybrid solar cells to reduce the use of silicon material [54]. To improve the light-trapping ability, Figure 8. Embedded metal electrode for highly efficient organic-inorganic hybrid nanowire solar cells, and I-V characteristic of the solar cell [52]. nanostructures were built on the thin silicon substrates by a metal-assisted chemical etching method (MACE). However, nanostructured silicon resulted in a large amount of surface-defect states, causing detrimental charge recombination. Here, the surface was smoothed by solutionprocessed chemical treatment to reduce the surface/volume ratio of nanostructured silicon. Surface-charge recombination was dramatically suppressed after surface modification with a chemical, associated with improved minority charge carrier lifetime. As a result, a power conversion efficiency of 9.1% was achieved in the flexible hybrid silicon solar cells, with a substrate thickness as low as ∼14 μm, indicating that interface engineering was essential to improve the hybrid junction quality and photovoltaic characteristics of the hybrid devices.
As shown in Figure 10(b), the leakage current of the TMAH-treated nanostructured silicon devices is much lower than that of the untreated silicon devices, suggesting that the junction quality is improved. Furthermore, the J SC of the hybrid devices based on different surface morphologies is consistent with the external quantum efficiency (EQE) spectra, as shown in Figure 10(c). The device based on ∼14-μm thick planar silicon without TMAH treatment gives a J SC of 18.4 mA cm −2 , a V OC of 0.56 V, and an FF of 0.75, achieving a PCE of 7.9%. The previously reported ultrathin planar silicon device with a thickness of 8.6 μm yields a PCE of 5.2% [55]. Here, the reduced J SC for the device based on ∼14-μm thick planar silicon is ascribed to the high light reflection of the planar silicon surfaces, which resulted in poor light absorption compared to that of a thicker device. In comparison with planar devices, nanostructured silicon solar cells exhibit a larger J SC value, indicating that the incident light could be effectively absorbed by nanostructured silicon. The absorption spectra of the planar and nanostructured silicon with or without TMAH treatment are shown in Figure 10(e). According to the absorption spectra, the nanostructured silicon exhibits an obviously better light absorption over a broad spectrum from the visible to the near-infrared range compared to that of planar silicon.  (e) Absorption spectra of silicon substrates with a planar surface, an as-prepared nanostructured silicon surface, a TMAH-treated nanostructured silicon surface, and upon Yablonovitch limit simulation [54].

Conclusion
In summary, facile silicon surface modification methods based on one or two MACE to synthesis PSiNWs were presented. The change in morphology based on HF or H 2 O 2 effect was presented. Furthermore, this change also affects the optical properties specially the photoluminescence and compensates the poor light absorption of the planar silicon substrate. Several emerging strategies result in high-performance hybrid solar cells based on PSiNWs were presented. These strategies are based on the organic materials or/and the design of the solar cells or on the grid electrode as a contact. These findings could be potentially employed to further reduce the fabrication cost of silicon solar cells by reducing the consumption of silicon materials and increasing their efficiency.