Microstructure characterizations of the porous SiNWs were summarized in . SEM image shows high density vertical array of well-aligned porous SiNWs can be readily achieved with the diameters on the order of 100 nm and length on the order of 10 μm (). The TEM image clearly shows porous structure of the resulting SiNWs ().
Fig. 1 haracterization of the porous SiNWs and PtNP loaded porous SiNWs. (a) SEM image of the cross section of the as-etched porous SiNWs. (b) TEM image of a typical porous SiNW. (c) TEM image of the PtNP-pSiNW-A. (d) TEM image of the PtNP-pSiNW-B. (e) TEM image (more ...)
Incorporating selected metal nanoparticles onto semiconductor photocatalysts can enhance the photocatalytic activity because the difference in their Fermi levels can introduce a Schottky barrier between the metal and the semiconductor.16
The built-in potential within the Schottky barrier can facilitate the separation of photogenerated electron-hole pairs. Futhermore, properly selected metal nanoparticles can also function as the catalysts to facilitate certain redox reactions. To this end, we have prepared PtNPs and loaded them onto the porous SiNWs. Specifically, PtNPs with 3-4 nm diameters were s nthesized using poly(vinylpyrrolidone) (PVP) as the capping ligands29
(ESI, Fig. S2 †
). The as-synthesized PtNPs typically have negative surface charges originated from PVP ligands, and cannot be readily attached onto porous SiNWs that also have negative surface charges. To facilitate the attachment of PtNPs onto porous SiNWs with sufficient yield, the porous SiNWs were functionalized with APTMS to render positively charged surface. The PtNPs can then be loaded onto functionalized porous SiNWs through electrostatic force. Controlled densities of PtNPs on porous silicon nanowires were achieved by adding various amounts of PtNPs into a fixed volume of porous SiNW suspension. It should be noted that density of PtNPs enventually saturates and reach an adsorption/desorption equilibrium when the positive surface charges on porous SiNWs are mostly compensated by the negative charges on PtNPs. In this way, we have prepared three photocatalyst samples with variable densities of PtNPs by adding 250 μL, 500 μL and 1000 μL of PtNP solution into 2 ml of ethanol dispersion containing 3.0 mg porous SiNWs, and named the these samples as PtNP-pSiNW-A, PtNP-pSiNW-B and PtNP-pSiNW-C, respectively. TEM studies clearly show that PtNPs are well anchored on the porous SiNWs with increasing densities (). The HRTEM image further illustrates good conjugation between the PtNPs and the porous SiNWs (). Energy dispersive X-ray (EDX) spectrum also confirms the conjugation between PtNPs and porous SiNWs (ESI, Fig. S3 †
Nitrogen gas adsorption/desorption isotherm measurements were used to determine surface areas, average pore size and pore volume of the porous SiNWs. The porous SiNWs, which were prepared by immersing the wafer in an etching solution containing 0.3 M H2
and 4.8 M HF for 60 minutes were carefully scratched off from half of a 4-inch wafer with a razor blade. The mass of the porous SiNWs was determined after the samples were degassed overnight at 250 °C. A typical adsorption/desorption isotherm for 20.1 mg of the collected porous SiNWs is shown in . The standard multipoint Brunauer-Emmett-Teller (BET) analysis of the porous SiNWs yields an exceptionally high surface area of 337 m2
, which is comparable to the recently reported ptype mesoporous silicon nanowires (342 m2
) 26 and mesoporous TiO2
nanostructures (50 - 300 m2
An average pore diameter of 14.0 nm is calculated based on the Barret-Joyner-Halenda (BJH) model. The total pore volume is 1.184 cm3
/g at P/Po
= 0.97 for the porous SiNWs. The high surface area and large pore volume suggest that the semiconducting porous SiNWs will be a peculiar and promising material for energy harvesting and photocatalysis purposes.
ypical nitrogen adsorption/desorption isotherms of the porous SiNWs
Traditional metal oxide photocatlysts such as TiO2 and ZnO only absorb a small portion of the solar spectrum because of their large band gaps. Significantly, our study shows the absorption spectrum of the porous SiNWs spans across the entire UV, visible light and near IR range (). Therefore, the photocatalytic reactions based on porous silicon nanostructures could be effectively carried out under visible light or even under near IR light. To evaluate the photocatalytic activity of the porous SiNWs, two sets of experiments were carried out for the photodegradation of indigo carmine (IC) and 4-nitrophenol (4NP).
Fig. 3 hotocatalytic properties of the porous SiNWs and Pt loaded porous SiNWs. (a) Absorption spectrum of the porous SiNWs. The concentration of porous SiNWs was set at 0.1 mg/ml. (b) IC degradation catalyzed by the porous SiNWs and Pt loaded porous SiNWs. (more ...)
Under light irradiation, the electron-hole pairs are generated in the porous-SiNWs and then separated, which further react with the dissolved O2
and water to produce reactive oxygen species (ROS) such as hydroxyl radicals (OH•
), superoxide (O2 •−
), singlet oxygen (1
), and peroxide (H2
). ROS are strong oxidants and known as nonselective oxidizing agents for organic pollutants.30
They can oxidize IC into CO2
and sodium sulfate. The reactions are listed as follows:
The photodegradation of IC was carried out in a photocatalyst dispersion of 0.3 mg/ml under the irradiation of a 300 W xenon light. The change of IC concentration ([IC]) is monitored as a function of reation time. When the fresh photocatalyst sample PtNP-pSiNW-C was used in the reaction, the change of the [IC] clearly shows three distinct reaction stages: slower reaction rate at the beginning (12.2% degraded IC within 15 minutes), faster reaction rate in the middle (70.3% degraded IC within the subsequent 45 minutes), and slower reaction rate again in the end (ESI, Fig. S4 †
). The initial slower reaction rate could be attributed to the simultaneous photodegradation of IC and other absorbed organic species including PVP ligand on PtNP surfaces. After 30 minutes when PVP and other absorbed organic molecules were exhausted, the change of [IC] could be greater than that at the initial stage. When the [IC] dropped to a very low concentration, the ROS have a lower probability to be captured by the IC molecules, leading to a slower degradation rate.
To remove this initial slow reaction stage and more accurately estimate the catalytic activity of various PtNP-pSiNW photocatalysts, all catalyst samples were exposed to light irradiation for 1 hour before photodegradation of IC and 4NP. The photocatalytic activity of the four catalysts on the degradation of IC is compared in . After 60 minutes of irradiation, the percentages of degraded IC are 37.2%, 51.1%, 62.2% and 86.9% for porous SiNWs, PtNP-pSiNW-A, PtNP-pSiNW-B and PtNP-pSiNW-C, respectively. In contrast, only 4.7 % of IC molecules were degraded with the same irradiation conditions without the PtNP-pSiNW photocatalysts. Our results clearly demonstrate that porous SiNWs can function as effective photocatalysts in the visible irradiation range and that the Pt-loaded porous SiNWs are much more efficient photocatalysts than the porous SiNWs only. This catalytic enhancement by PtNPs could be attributed to their ability to facilitate electron-hole separation and to promote electron transfer process in catalytic photodegradation reaction.
It is imporant to compare the photocatalytic activity of the porous SiNW samples with more commonly used TiO2
nanoparticles. To this end, we have carried out a sesries control experiments using 5 nm anatase TiO2
(ESI, Fig. S5 †
) and P25 TiO2
nanoparticles under same conditions with same weight concentration. Under visible light, our studies show that porous SiNWs show photocatalyctic activity higher than anatase TiO2
nanoparticles but lower than P25 TiO2
nanoparticles (ESI, Fig. S6(b) †
). This difference can be largely attributed to the absorption properties of these three samples, with the P25 TiO2
nanoparticles show largest overall absoprtion, and porous SiNWs next, and anatase TiO2
nanoparticles least (ESI, Fig. S6(a) †
). Despite a band gap in the visible range (1.4-2.5 eV), the relatively low absorption of porous SiNWs may be attributed to the low intrinsic absorption coefficient of silicon compared to TiO2
In the IR regime, the porous SiNWs show best absorption and best photocatalytic activity among three photocatalysts (ESI, Fig. S6 (c) †
). In addition to the band gap and absoprtion properties, it should be noted that the relative alignment of the conduction and valence band edge vs. redox potentials may also affect the photocatalytic activity. Future studies to determine the relative band edge alignment of porous SiNWs and its impact on the photocatalytic activity may offer clues for additonal improvement. .
The photocatalytic stability of the porous SiNWs was also evaluated by recycling the porous SiNWs for multiple cycles of reactions (squares, in ). The concentration of the catalysts was controlled at 0.3 mg/mL in 100 μM IC solution. After 2.5 hours of light irradiation, the catalysts were recovered by centrifugation and redispersed in a fresh IC solution for the next cycle of test. The catalysts show stable photocatalytic behavior after 10 cycles of reactions, demonstrating the high photostability of the porous SiNWs. A similar experiment was carried out to test the life time of the PtNP-pSiNW-C sample (circles in ). The mixture was exposed under the light irradiation for 90 minutes with near complete IC degradation for each cycle. The PtNP-pSiNW-C catalyst also showed similar photocatalytic stability.
Photocatalyst is of significant interest in photodegradation of organic pollutant for waste treatment in environment. For example, ROS can oxidize 4-nitrophenol (4NP) into CO2
O. To this end, we have explored the porous SiNWs and PtNP-pSiNW-C as the photocatalysts to degrade 4NP to evaluate their photocatalytic ability in organic waste treatment (). 18.8 % and 66.0% of the 4NP were degraded after 90 minutes of light irradiation with the porous SiNWs and the PtNP-pSiNW-C, respectively. The results further demonstrate the photocatalytic ability of the porous SiNWs in degrade organic species and confirm that PtNP loading can promote photocatalytic activity. Reactions with the catalysts exposed to xenon light for 1 hour also show a catalytic behavior with an initial slow degradation rate followed by a faster and then another slow reaction rate. This could be attributed to the variable amount of 4NP molecules absorbed on the surface of the porous SiNWs, which determines the degradation rate. Previous studies on the photodegradation of 4NP catalyzed by TiO2
particles show that the amount of absorbed 4NP molecules on the catalysts is related to the pH values: a lower pH value often leads to higher 4NP molecules absorption.11
The optimal pH for 4NP absorption on TiO2
surface is 4.0. In the case of porous SiNWs, a thin layer of silicon oxide forms on the surface with abundant (SiO)3
The pKa of such silica surface is 4.5,36
similar to that of the TiO2
In porous SiNWs catalyzed photodegradation of 4NP, the starting solution has a pH value of 7.18 due to the pKa
value of 7.08 for 4NP. The initial amount of the absorbed 4NP molecules is limited by this pH value. As the reaction proceeds, CO2
are generated, which reduces the pH value of the reaction solution to 5.71 after 60 minute reaction. The lower pH value favors the adsorption of 4NP molecules on the catalysts and thus accelerates the degradation rate. The slow reaction rate at the final stage could be attributed to the lower concentration of 4NP in the reaction solution. Compared to the porous SiNWs, reactions involving the PtNP-pSiNW-C catalysts display a shorter period of the initial slower stage because the PtNPs accelerate the degradation reaction and lead to a lower pH value within a shorter time duration.