Keywords

Introduction

Visible photoluminescence (PL) from porous silicon (PS) observed at room temperature has inspired sustained research into its potential application in Si-based optoelectronic devices and its theoretical basis (Canham 1990). This property is reviewed in the handbook chapter “Photoluminescence of Porous Silicon.” Most PS layers are prepared by anodic etching on p-type Si substrates, a technique in which metal is often deposited on the rear surface of the Si substrate in order for it to be used as an ohmic back contact (see handbook chapter “Porous Silicon Formation by Anodization”). However, the requirement for a back contact electrode is a limitation of this method; for example, it is difficult to form a PS layer on a silicon-on-insulator (SOI) structure or on Si integrated circuits. A photoetching method, on the other hand, requires no electrodes and allows the formation of a visible luminescence layer on not only single-crystalline Si substrates but also SOI structures.

Photoetching Setup

An experimental setup used for the formation of PS by photoetching is shown in Fig. 1 (Xu and Adachi 2006). The sample surface is illuminated by a Xe lamp through an optical filter that blocks wavelengths shorter than 600 nm. The use of an optical filter is to block the heat rays from the Xe lamp. A laser, a W lamp, or another light source may be used instead of a Xe lamp. The use of an incoherent light source such as a Xe or W lamp enables the formation of a large and homogeneous PS layer. Typically, an n-type Si wafer is immersed in an etchant solution of HF. The addition of an oxidant (e.g., H2O2 or I2) to the HF solution results in the stable formation of PS layers in a short time period.

Fig. 1
figure 1

Experimental setup used for porous silicon formation by photoetching in an HF/oxidant solution

Full size image

n-Si/Electrolyte Interface and Photoetching Reaction

Figure 2 shows the energy band diagrams for n-Si electrodes in pure HF (pH = 2.3) and HF/oxidant solutions without and with light illumination (Xu and Adachi 2006). The electron affinity (χs) of Si is −4.05 eV. At zero pH, the redox coupling is defined as the normal hydrogen electrode with a potential of −4.5 eV with respect to vacuum. This potential shifts toward more positive values with the increase in pH (+0.059 eV/pH). Thus, the electron energy of the pure HF solution with respect to vacuum is −4.36 eV (χl). The Fermi levels (EF and EF,redox) on both sides of the n-Si/electrolyte interface are brought to the same energy level by a transfer of electrons from the Si substrate to the electrolyte (Fig. 2a).

Fig. 2
figure 2

Energy band diagram for n-Si immersed in pure HF solution (a, b) and those in HF/KIO3 solution (c, d). In (b), porous silicon (PS) is formed stably on the back side in opposition to the illuminated surface. In (d), PS is formed only on the illuminated surface

Full size image

The half reaction for the oxidizing agent KIO3 is

$$ {\mathrm{I}\mathrm{O}}_3^{-}+6{\mathrm{H}}^{+}+6{\mathrm{e}}^{-}={\mathrm{I}}^{-}+3{\mathrm{H}}_2\mathrm{O}1em \left({E}^{\mathrm{o}}=1.0850.5em \mathrm{eV}\right) $$

where e− represents the electron and Eo is the standard reduction potential with respect to the standard hydrogen electrode. The redox potential (Eabs) with respect to vacuum for the HF/KIO3 redox system is then given by Eabs = −4.5 − Eo = −5.6 eV (Fig. 2c). It is to be noted that the larger the Eo value is in the positive (negative) scale, the stronger is the oxidation (reduction) agent (Adachi and Kubota 2007; Xu and Adachi 2007; Tomioka et al. 2007).

The absorption of photons results in the generation of electron-hole pairs. The holes at the n-Si/electrolyte interface can participate in PS formation. In the case of the pure HF solution (Fig. 2b), the photoexcited holes are hard to drift toward the surface by the very small downward band bending or possibly by the almost-flat band. Thus, efficient PS formation cannot be expected in pure HF solution. When the Si wafer is dipped in the HF/oxidant solution (Fig. 2d), on the other hand, many photoexcited holes move toward the n-Si/electrolyte interface at the front surface, resulting in the formation of PS with good reproducibility (Xu and Adachi 2006, 2007; Adachi and Kubota 2007; Tomioka et al. 2007).

Reproducibility has been observed to be problematic in the formation of PS by photoetching, as with stain etching (see handbook chapter “Porous Silicon Formation by Stain Etching”). In an extreme case, no PS layer was formed on the front surface, although surprisingly PS was formed on the surface of the sample that was not exposed to illumination (i.e., on the back surface) (Andersen et al. 1995). The effectiveness of surface cleaning by sulfuric peroxide mixture (SPM) treatment or by KOH etching before PS formation has been reported in Tomioka et al. (2007) and Andersen et al. (1995).

The photo-illuminated n-Si/aqueous NH4F interface has been shown to form a hydrogenated amorphous Si overlayer which builds up progressively as photoetching proceeds with disproportionation of Si2+ species in solution (Peter et al. 1989). It is known that a galvanic cell is formed when a p-type Si is contacted with a noble metal in a HF/oxidant solution (Kobayashi and Adachi 2010). This galvanic cell leads to metal-assisted etching of p-Si, resulting in the formation of Si nanowire arrays. PS layers prepared by two routes, metal-assisted etching and laser-induced etching, have been studied by comparing surface morphologies using scanning electron microscopy (Kobayashi and Adachi 2010; Saxena et al. 2015). A PL peak at ~1.8 – 2.0 eV corresponding to red emission at room temperature was observed from such p-Si samples. The fact suggests that the PS layers can be formed not only on the laser-etched surfaces but also on the Si nanowire arrays formed by metal-assisted electroless etching. In p-Si prepared by laser etching, wider pores with some variation in pore size as compared to metal-assisted etching technique were observed because a HeNe laser having Gaussian profile of intensity was used for porosification (Saxena et al. 2015).

PS Layers Formed by Photoetching

A summary of PS formation by photoetching is presented in Tables 1 and 2 (Noguchi and Suemune 1993; Zhang et al. 1993; Cheah and Choy 1994; Andersen et al. 1995; Jones et al. 1996; Kolasinski et al. 2000; Yamamoto and Takai 2000, 2001; Mavi et al. 2001, 2006; Koker et al. 2002; Marotti et al. 2003; Zheng et al. 2005; Tomioka and Adachi 2005; Adachi and Tomioka 2005; Cho et al. 2006; Xu and Adachi 2006, 2007, 2008; Adachi and Kubota 2007; Adachi and Oi 2007; Tomioka et al. 2007; Adachi et al. 2007; Ramizy et al. 2011; Matsui and Adachi 2012).

Table 1 Photoetching for porous silicon formation in acidic solutions
Full size table
Table 2 Photoetching for porous silicon formation in alkaline electrolytes and aqueous alkali salt solutions
Full size table

To the best of our knowledge, there has not been reported any good plan-view high-resolution scanning electron microscopy images of the photosynthesized PS layers showing the morphology of their typical structures. In Xu and Adachi (2007), the atomic force microscopy images were reported to show many irregularly shaped hillocks and voids distributed randomly over the entire PS surface. The observed root-mean-squares roughnesses were a few nanometers.

Lateral patterning of PS layers has been performed using photoassisted electrochemical etching rather than pure photoetching (Baranauskas et al. 1995; Diesinger et al. 2003). Lateral modification of the porosity has also been obtained by photochemical dissolution of the anodic PS layers under illumination with a beam made of interference fringes (Ferrand et al. 2001).

A metal-insulator-semiconductor-type electroluminescent (EL) device has been fabricated from PS layers synthesized by photoetching in an HF/I2 solution (Adachi and Kubota 2008). An insulating layer was formed on the PS layer by chemical oxidation in an acidic solution. Spectral output of the EL device was in the red-yellow region peaking at 2 eV.

Photoetching of Silicon Powders

The enhanced etching of bulk silicon in hydrofluoric acid via continuous photoexcitation has been known for a long time and has been used to pattern wafers (Lim et al. 1992). Recently, the technique has received some development for nanostructuring of inexpensive silicon powders (Matsumoto et al. 2014, Lee et al. 2016) in addition to silicon wafers. Although TEM and XRD data demonstrated the presence of silicon nanoparticles (Matsumoto et al. 2014, Lee et al. 2016), it would be interesting to explore with gas adsorption analysis (see handbook chapter “Gas Adsorption Analysis of Porous Silicon”) whether significant mesoporosity can be engineered in such processed powders.

Conclusions

Photoetching enables the formation of a visible light-emitting PS layer on n-type Si wafers. The use of an incoherent light source and the addition of an oxidizing agent in the HF solution also facilitate the formation of a thicker homogeneous PS layer with good reproducibility. The thickness of the porous layer is still usually less than 1 μm. The PL and EL peak energies were observed to be in the range 1.7–2.3 eV. The photoetching technique can be applied to Si wafers with embedded circuitry, SOI wafers, and silicon powders.