Deconvoluting Diffuse Reflectance Spectra for Retrieving

Article Cite This: J. Phys. Chem. A XXXX, XXX, XXX−XXX

pubs.acs.org/JPCA

Deconvoluting Diffuse Reflectance Spectra for Retrieving Nanostructures’ Size Details: An Easy and Efficient Approach Manushree Tanwar, Anjali Chaudhary, Devesh K. Pathak, Priyanka Yogi,# Shailendra K. Saxena,$ Pankaj R. Sagdeo, and Rajesh Kumar* Material Research Laboratory, Discipline of Physics & MEMS, Indian Institute of Technology Indore, Simrol 453552, India

Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on April 16, 2019 at 17:56:53 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: A new model has been reported here to estimate the mean size and size distribution in nanostructured materials by utilizing a simple and economic diffuse reflectance spectroscopy through spectral line-shape analysis. In the proposed model, a theoretical line shape has been derived by taking into account a size distribution function, which represents a variation in absorption coefficient as a function of size, which in turn depends on the band gap and thus on the excitation photon energy. A fitting of the experimental absorption spectra with the derived line-shape function yields the mean crystallite size and size distribution. The size and size distribution have been successfully estimated from two different silicon nanostructured samples, prepared by metal induced etching. The model has been validated by comparing the estimated values with the sizes estimated using Raman spectroscopy, which is a well-known technique. The two results are not only consistent with each other but are also found to be consistent with the electron microscopy’s results, revealing that a technique as simple and as economic as diffuse reflectance spectroscopy can be used to estimate size distribution. In addition, the proposed model can also be used to investigate the homogeneity in the size distribution in a nanostructured sample.

■

INTRODUCTION For more than three decades now, the perturbed physical and chemical properties of materials including semiconductors due to size effect (quantum confinement) have gained massive attention and inquisitiveness for the exploration of these materials in devices for different applications like field emission,1,2 batteries,3 solar cell,4 sensors,5,6 capacitors,7 etc. With new applications of nanomaterials comes the challenge to characterize them. For applications where size-dependent properties8−12 are explored it becomes very critical to have precise information about not only the size but also the size distribution (SD) if there is any. Among various methods to estimate the size of nanostructures, electron microscopy is the only direct method with a limitation to predict the size of nanostructures only in a very small area on the sample. Other indirect methods utilized for this purpose are many including Raman spectroscopy, dynamic light scattering,13 etc. Raman spectroscopy is a very versatile tool but a bit expensive whereas the dynamic light scattering needs a particular type of sample preparation. A given method has its own advantages and disadvantages; thus, it is very difficult to conclude the true picture. Thus, an alternate easy and efficient method is the need of the hour, which can predict the size and SD of nanostructures. While exploring the new ways, it is important to understand the relationship between the size (and SD) and a given observable of the experiment which is size-dependent phonon frequency shift in case of Raman spectra and correlation spectroscopy for dynamic light scattering as © XXXX American Chemical Society

examples. Dependence of band gap on size is a known phenomenon and thus its implication on the absorption and/ or reflectance (for nontransparent samples) spectra is expected. In other words, the absorption spectrum contains the information on size and SD present in the sample which is manifested in terms of the deviation from the typically observed ideal spectra from a bulk material. If this relationship is understood properly, a careful analysis of the absorption spectra may, in principle, be used to estimate the size and SD for a given nanostructures samples. The same has been explored in the current study by taking the example of silicon (Si) nanowires (NWs). Silicon when confined to one dimension (NWs) shows interesting perturbations in its optoelectronic, and several other, properties, which allows one to use it as a suitable material to be explored in distinct device applications. There are various fabrication techniques involved for nanostructure fabrication including bottom-up techniques like vapor−solid− liquid growth14 and molecular beam epitaxy15 and top-down techniques like pulsed laser ablation,16 ion beam etching,17 and metal assisted chemical etching popularly known metal induced etching.18−21 In the present study, metal induced etching has been used to fabricate SiNWs from an n- and a ptype Si wafer because metal induced etching is one of the Received: February 28, 2019 Revised: April 8, 2019

A

DOI: 10.1021/acs.jpca.9b01935 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

metal induced etching technique. The estimated size and SD has been compared with the results obtained using Raman spectroscopy, a well-established technique for this purpose. A discussion is also presented regarding the consistency between the results obtained by line-shape analysis of Raman and diffuse reflectance spectroscopy’s results. The sizes estimated using the two techniques were consistent, which means that the SiNWs samples are homogeneous from the point of view of the distribution of sizes is concerned. The proposed model establishes how the diffuse reflectance spectroscopy can be used for size analysis in a nanostructured sample. Furthermore, way to interpret the consistency or inconsistency, as may be the case, of diffuse reflectance spectroscopy’s results with Raman results has also been presented. Convergence of modifications in phonon spectrum and electronic structure has been shown as the key factor in developing the model for yielding information about SD which is connected through the confinement effect in nanostructured samples.

techniques where the SiNWs result in wires of different sizes thus is suitable for the problem identified for the current study. As mentioned above, similar to diffuse reflectance spectroscopy, Raman scattering,22−24 where the interaction of light with matter is involved, has proved to be a primordial and nondestructive technique to investigate different phases,25 chemical composition,26,27 and doping level28,29 along with reaching out and acting as a sensitive probe to inspect physical phenomena such as quantum confinement,30−32 electron− phonon interaction (Fano resonance),33,34 etc. taking place at the nano regime. The information on quantum confinement is very well put forth by Raman spectroscopy as an investigative probe. The Raman spectrum of crystalline silicon is a sharp and symmetric, Lorentzian line shape peaked at ∼520 cm−1 owing to the participation of zone center phonon,35 contrarily due to size effect, the line shape gets asymmetrically broadened with a peak shift due to breakdown of the momentum conservation rule (k = 0) for SiNWs, which in the case of crystalline silicon is by and large conserved. This perturbation in the Raman line shape due to phonon confinement was well modeled by Richter et al.36 and later modified by Campbell et al.37 by developing a universally accepted theoretical Raman line shape function elucidating the size effect due to the confinement. The theoretical fitting of the obtained spectra by the light scattering from each size manifests itself in the form of a spectrum having complete information regarding the mean size and SD present in the sample. A discrete analysis can deliver a wealth of information regarding various physical phenomena. The light−matter interaction and quantum confinement effect brings the two techniques, Raman and diffuse reflectance spectroscopy, on a common platform and thus can be correlated for the purpose of size analysis. In other words, interaction of light with matter therefore remains an insightful and intuitive probe in order to deal with the materials’ properties at the nanoscale. The confinement of electrons in the nanostructures leads to interesting perturbations in the optical properties like band gap,38−40 absorption,41 etc. Due to the size effect the reformed band gap of SiNWs changes significantly as given by eq 139 and hence absorption coefficient42 being directly related to the band gap of the material is altered for individual size. γ E(D) = Eg + m D

■

EXPERIMENTAL DETAILS Two different commercially available (Vin Karola) Si wafers (n- and p- type) with doping levels of 8.2 × 1019 cm−3 (arsenic doped; n-type; resistivity ∼0.001 Ω cm) and 1.3× 1020 cm−3 (boron doped; p-type; resistivity ∼0.001 Ω cm) have been used to fabricate the SiNWs by metal induced etching (MIE).44−46 For MIE, first Ag nanoparticles (AgNPs) were deposited on cleaned Si wafers by dipping them into a solution containing 4.8 M HF and 5 mM AgNO3 for 60 s at room temperature.The AgNPs deposited samples were then kept for etching in a solution containing 4.6 M HF and 0.5 M H2O2 for 60 min at room temperature. Etched wafers were then transferred to HNO3 to remove the AgNPs after the etching process. The samples were then dipped into HF solution to remove any oxide layer induced by nitric acid used in above step. The final sample prepared from the n-type Si wafer has been named as “sample-n” whereas the sample prepared from the p-type Si wafer has been named as “sample-p” (Table 1). A Table 1. Values of Minimum, Mean and Maximum Sizes of SiNWs Obtained from Optical and Raman Spectroscopy size estimated from optical spectroscopy (size estimated from RS) (nm)

(1)

where E(D) in electronvolts is the band gap of materials of size D in nm, Eg is the bulk band gap, and γ and m are the confinement parameters defined using effective mass theory approximation.43 If, the SiNWs fabricated using metal induced etching are to be analyzed by absorption spectroscopy, the resultant experimental spectra when fitted well with the theoretical equation will convey relevant information about the SD present in the system, because of the size effect which eventually forces the absorption to change for each size. With the in principle capability to be used for size analysis established, a way to do so needs to be understood so that scientists can use it efficiently. The main aim of this paper is to establish absorption spectroscopy as an easier and efficient technique for estimation of mean size and SD in nanostructures. This has been done by analyzing a diffuse reflectance spectral line-shape by means of the proposed line-shape function developed by incorporating the size dependence of band gap. The model has been used for consolidate size analysis of two SiNWs samples prepared by

sample

D1

D0

D2

Sample-p Sample-n

2.8 (3) 4.5 (4.3)

3 (3.5) 5.8 (6)

9.8 (10) 6.7 (7)

Supra55Zeiss SEM has been used to study the surface morphologies of these samples. The photoluminescence (PL) measurements have been carried out with a 325 nm laser excitation source using a Dong Woo Optron 80K PL system at room temperature. Raman spectra were recorded using a Horiba Jobin Yvon micro-Raman spectrometer with a 2.54 eV excitation laser with minimum power to avoid any laser induced heating. A commercial transmission electron microscope (TEM) Phillips, CM200, with a LaB6 filament as the cathode, was used at an accelerating anode voltage of 200 kV. The absorption spectra of the samples have been measured using DRS setup from Cary-60 UV−vis spectrophotometer (Agilent make). B


Article


Figure 1. Structural micrographs from (a) sample-n and sample- p in top view along with cross sectional scanning electron microscopy images (top left insets) and transmission electron microscopy images (bottom right insets).

■

RESULTS AND DISCUSSION Sample-n and sample-p prepared using metal induced etching appears to be containing SiNWs as evident from the scanning electron microscopy’s micrographs (Figure 1). The pore formation on the surface of silicon wafer is clearly visible from the top view scanning electron microscopy images (Figure 1). The detailed mechanisms involved in the pore formation and pore properties have been discussed elsewhere44,45,47,48 in detail. The chemical reaction involved during metal induced etching has been summarized in the Supporting Information. The cross sectional images in Figure 1 (top left insets) show well-aligned wirelike structures and hint at the presence of SiNWs in the sample as will be substantiated using Raman spectroscopy later on. The wirelike structures look too thick to be called “nano” wires but a closer analysis using transmission electron microscopy reveals that a few tens of nanometers thin wires are present (bottom right insets in Figure 1). However, thsese wires are thick as compared to the bohr’s excitonic radius of silicon (∼5 nm) but finer silicon nanostructures (a few nanometers) are present inside these nanowires that are small enough to show the quantum confinement effect, as has been discussed elsewhere.41 It is also improtant here to mention that all the silicon nanostructures are not of the same size and possess some distribution. Considering the fact that these microscopic techniques only reveal information about a samller area (∼0.01 μm2) of the sample, one needs to be very careful while quantifying the size and SD of these nanostructures using transmission electron microscopy. The distribution of nanostrucures estimated will depend on the technique by which it has been quantified; thus, a detailed discussion on validation of the same is required. The phenomena of the quantum confinement effect in silicon nanostructures is known to manifest itself in the form of a broadend photoluminescence49,50 observed in the visible region at room temperature which very often is taken as a validation of the quantum confienment effect in silicon nanostructures. The room temperature visible photoluminescence is a consequence of band gap broadening due to quantum confienemnt effect generating a sharp photoluminescence spectrum, if the size of all the wires present are the same throughout the sample. However, a broader photoluminescence spectrum is observed if the wires have a distrubution in their sizes. Both the samples used in the present study appear to belong to the latter case as evident

from the broadnend room temperature visible photoluminescence spectra (Figure 2). The photoluminescence emission

Figure 2. Room temperature photoluminescence spectra for sample-n and sample-p.

centered around 1.9 eV corresponds to the red emission from the samples, as can be seen in the inset of Figure 2, which shows the actual image of the sample when seen under UV illumination used for photoluminescence measurement. Though a distribution in size of SiNWs is indicated by photoluminescence spectra, to quantify the distribution in SiNWs size indicated using photoluminescence, diffuse reflectance and Raman spectrosocpies have been carried out. As explained earlier, diffuse reflectance spectroscopy can be utilized to give a thorough analysis of the SD present in the system, in principle. It carries the information about the size and SD by means of the absorption of photons and its modulation due to size-dependent variation in band structure reflected through the change in band gap and through a possible broadening in the near absorption band edges analogous to the Urbach tail.51 In other words, the SD leads to a width in the band edges with a mean band gap corresponding to the mean size. This way, an analysis of absorption spectrum line shape can be used to estimate the SD in the sample, giving information from a larger area leading to a more inclusive picture as compared to scanning and transmission electron microscopy. This proposal will enable us to estimate size and SD in the same way. Raman spectroscopy is used to estimate the same where the concept of phonon confinement inside nanostructures of different size leads to an C


Article


Figure 3. Schematic showing (a) unique phonons leading to a sharp Raman spectrum from crystalline silicon, (b) dispersed phonons leading to an asymmetric Raman spectrum from nanosilicon, (c) typical absorption spectrum from bulk silicon, and (d) deviation from ideal step-like absorption spectrum from nanomaterials.

asymmetrical Raman line shape.48 Careful analysis of the Raman line shape and its comparison with the theoretically developed line-shape function provide the estimations of mean size and SD that are very widely used for this purpose. The above-mentioned concept has been discussed using a schematic (Figure 3) that correlates the SD to given Raman or optical spectra from silicon nanostructures along with the crystalline form shown for comparison. The Raman spectrum of bulk silicon is sharp and has a symmetric Lorentzian line shape with the peak positioned at ∼520.5 cm−1 (Figure 3a), corresponding to the zone centered phonons in the whole crystal. It is important here to mention that in the crystalline silicon, all the phonons have the same frequency (corresponding to zone-centered phonons), as depicted in Figure 3a. However, when the material size is reduced to “nano” dimension, the Raman spectrum becomes asymmetric and red-shifted, as compared to its bulk counterpart due to phonon modulation under confinement induced perturbation. In silicon nanostructures, the participation of phonons is governed by the broken momentum conservation rule as discussed above, due to which several phonons with different frequencies start taking part in the Raman scattering. If the material has different nanocrystallite sizes, the light scattering from each nanowire will generate a spectrum that will have information about the sizes of the wires that participated in the scattering process. The Raman scattering from SiNWs present inside the area illuminated for collecting the Raman spectrum will result in the asymmetric line shape (Figure 3b), which will contain this information. This method is rather wellestablished and will be used in the current context for validation later on.

Another manifestation of the presence of SD is by means of modification of electronic structure leading to a change in band gap that consequently gets reflected in the absorption spectra from nanostructures. Thus, the size-dependent band gap, as shown in eq 1, plays a major role in the analysis of the absorption spectrum generated by the material. Due to the presence of uniform and large crystallite size in the bulk Si, the absorption spectrum is sharp with a band gap of 1.17 eV (Figure 3c). On the contrary, when the crystallite size is reduced to the nano regime, the effective band gap increases in accordance with eq 1, which is apparent from the schematic shown in Figure 3d. Also, the absorption spectra pertaining to each size are different from each other due to the fact that absorption is directly related to the band gap of a material. This in all signifies that if there is size distribution present in the sample, it should be manifested in the form of a monotonous continuous spectrum for SiNWs, unlike the sharp step-like spectrum, as shown in the schematic. The experimentally obtained spectrum can then be fitted with the appropriate theoretical line shape consisting of a size distribution function, as done in the present study. In order to find out the values of the minimum, mean, and maximum sized wires present in the samples, theoretical fitting of the absorption coefficient, obtained from the diffuse reflectance spectrosocpy has been carried out, which will be discussed in detail later on. Raman spectroscopy, as mentioned earlier, being a primordial tool to deal with the vivid class of materials, phenomena, and interactions, has been used in the present study to confirm the result obtained from diffuse reflectance spectroscopy. D


Article


Figure 4. Absorption spectra of sample-n and sample-p. Discrete data points show the experimentally obtained spectra whereas the solid lines correspond to the theoretical line-shape (eq 3) along with schematic of enhanced band gap of both the samples as a result of quantum confinement (insets).

been shown in Figure S1 in the Supporting Information. A closer look at the best fit curves in the spectral range 2.3−2.4 eV reveals an apparent small mismatch between the experimental and theoretical curves for sample-n (blue curve) as compared to sample-p (red curve) likely due to disorder analogous to Urbach tail. The different widths of distribution can be appreciated from the shapes of the optical spectra with sample-n yielding a sharper function as compared to sample-p, which shows a slowly varying function, as can be seen in Figure S2 in the Supporting Information, which shows the two spectra on same scale for better comparison. Raman spectroscopy, which is a widely used versatile tool53 to explain various physical phenomena and interactions taking place in the material, is used here for a specific purpose of estimation of size distribution present in the two samples so that the results obtained using diffuse reflectance spectra can be validated. As mentioned earlier, the Raman spectra from SiNWs are red-shifted and asymmetrically broadened as compared to that from crystalline silicon. The Raman spectra from sample-p, sample-n and crystalline silicon are shown in Figure 5 where the black dotted sharp and symmetric Raman

The absorption spectra from the two samples, obtained using diffuse reflectance spectrosocpy, are shown in Figure 4. The as obtained spectra have been suitably processed using the Kubelka−Munk (KM)52 model. The KM function F(R) is directly proportional to absorption coefficient (α) that can be expressed as follows: α=

A(hν − Eg )1/ n (2)

hν

where A is a constant, 1/n = 0.5 (or 2) for direct (or indirect) band gap semiconductor, and hv is the incident photon energy. The size dependence of the band gap is given by eq 1, and the absorption spectra of both samples originate from different band gaps; so to acquire a smooth fit, Eg has been replaced by the size-dependent band gap (eq 1). To evaluate the size distribution present in the samples, a log-normal size distribution function (ϕ(D)) has been incorporated for deducing the nanostructure SD present in the sample. After the above incorporation, eq 2 modifies to eq 3: α(hv) =

1 hv

∫D

D2

[hv − E(D)]1/2 ϕ(D) dD

2| l D o o o o o log( D0 ) o − where ϕ(D) = σD 2π expm with σ as the width of } 2 o o 2σ o o o o n ~ distribution and a fitting parameter, D0 is the mean SiNWs size, and D1 and D2 correspond to minimum and maximum sizes, respectively. The experimental data, represented by discrete points in Figure 4, have been fitted with eq 3 (the solid lines in Figure 4) to obtain the sizes as listed in Table 1 corresponding to the best fit. Values obtained clearly predict that sample-n contains a thicker SiNWs with a mean size of 5.8 nm whereas sample-p contains relatively thinner SiNWs of sizes in the range ∼3 nm. It is important here to mention that the sizes predicted by diffuse reflectance spectroscopy are consistent with the ones estimated roughly from the transmission electron microscopy image (Figure 1), which predicts an average size of 6 and 3.5 nm from sample-n and sample-p, respectively. Another interesting observation from the SD estimation using diffuse reflectance spectroscopy is the prediction of different distribution widths in the two samples. The distribution is rather sharp in sample-n with a distribution of 2.2 nm as compared to sample-p, which shows a broader distribution of 7 nm. The corresponding distribution functions, representing the distribution in size, for the two samples has 1

(3)

1

Figure 5. Raman spectra for sample-n and sample-p along with c-Si. Red and blue solid lines represent the theoretical Raman line shapes (eq 5) whereas the discrete data points represent the experimentally acquired data points.

line shape peaked at 520.5 cm−1 shows the latter. The asymmetric Raman line-shape, originating due to phonon confinement effect (as explained above) can be represented using eq 4, as proposed by Richter et al.36 and later modified by Campbell et al.,37 which is used to estimate the mean size and SD. More detail about origin of eq 4 and a qualitative E


Article


of the consistency between the two techniques, diffuse reflectance spectroscopy still proves to be an alternate technique for estimation of size and SD for nanomaterials.

description explaining the significance of various terms in eq 4 has been provided in the Supporting Information.35,48 It is important here to mention for completeness that eq 4 includes Fano resonance arising in heavily doped silicon due to the interaction between discrete phonons and electronic continuum available in the system due to heavy doping. Though the study has been carried out on highly doped samples, which along with quantum effect shows the Fano effect, the interpretation of size distribution is the area of interest in the present analysis. The consolidated general theoretical Fano-Raman line shape equation34,54,55 takes the following form: I(ω ,D) =

where, ε =

{

∫0

| (ε + q)2 l k 2D2 o o o expo − dnk m } 2 o 2 o o o 4 a 1 + ε n ~

1

ω − ω(k) 0.5Γ

■

CONCLUSIONS Nanowires of silicon, prepared by metal induced etching, show a distribution of size rather than a uniform size as revealed by broad photoluminescence spectra from the two samples fabricated from n- and p-type silicon wafers. The diffuse reflectance spectra from these two samples, analyzed using a newly proposed theoretical model, reveal that diffuse reflectance spectroscopy can be used to estimate the mean size and size distribution of nanostructures. The mean size and size distribution, estimated from optical spectra show consistent results with the predicted values using transmission electron microscopy. The general theoretical model has been developed for estimating the size of nanostructures by proposing a theoretical line-shape function corresponding to an absorption spectrum from the sample. The proposed equation has been devised on the basis of the concept of quantum confinement induced spectrum of band gaps from individual nanowires, thus leading to different absorption/ reflectance from each nanoparticle having different sizes. The resultant spectrum is thus the sum value of all these spectral values represented by the spectrum and the line-shape function. This has been incorporated using an appropriate size distribution function in theoretical formulation for absorption coefficient as a function of excited photon energy. These estimated values of size and size distribution are in consonance with the values estimated using well-established Raman spectroscopic framework. The consistency between Raman and diffuse reflectance spectroscopy’s results not only validates the proposed model for this purpose but also means that the size distribution is homogeneously distributed in the samples reported here. Thus, the proposed methodology here not only successfully estimates the size and size distribution but also can be utilized to understand the homogeneity of nanomaterial.

(4)

} and D, a, Γ denote the nanocrystallite

size, lattice constant, and line width (4 cm−1) of c-Si, respectively. Here “n” is the degree of quantum confinement, which has been chosen to be 2 for the case of SiNWs. The πk

ω(k) = 171400 + 100000 cos 2 is the phonon dispersion relation for Si and q is the Fano parameter, which is the measure of extent of electron−phonon interaction present in the system. The same size distribution function used to fit the absorption spectra for the two samples has been used in the theoretical Raman line-shape equation as well to yield the following final Raman line shape (eq 5):

I2(ω ,D) =

∫D

D2

1

ϕ(D) I(ω ,D) dD

(5)

The experimentally obtained Raman scattering data, represented by discrete points (Figure 5), have been fitted with eq 5 and the values of D0, D1, and D2 corresponding to best fit are listed in Table 1 (in brackets). It is interesting here to see not only that the mean size and SD estimated from Raman and diffuse reflectance spectra are the same but are consistent with the one obtained from the transmission electron microscopy image (Figure 1). Raman spectra also predict a smaller mean size and broader distribution for sample-p as compared to a relatively larger SiNWs size and smaller SD for sample-n. The consistency, with respect to the values of size and SD estimated, between the Raman and diffuse reflectance spectroscopy is not trivial but contains following important aspects. First, it means that a careful analysis of optical spectra using the proposed spectral line shape can be used for estimation of nanocrystallite size as an alternative to the Raman spectroscopy. This is important as the optical absorption technique is very economic as compared to Raman scattering; however, the latter has other advantages. Another importance comes from the comparison of the estimated values obtained by the two techniques. If the two techniques yield the same values, it means that the sample is homogeneous up to the extent of millimeter size on the sample, which is the area that gets exposed while carrying out the absorption spectra. In contrast, if the two results are not consistent, that does not mean that the model is incorrect, rather it means that the homogeneity of the sample is limited. The rationale behind this conclusion is that Raman and optical spectra examine different extents on the same sample due to different probe sizes, which is a few microns for Raman and up to a few millimeters for absorption spectroscopy. Irrespective

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.9b01935. Chemical reaction involved in silicon nanowires’ fabrication, comparison of two theoretically simulated log-normal size distribution functions, absorption spectra of both the samples, and discussion on the phonon confinement model (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Anjali Chaudhary: 0000-0002-8202-2697 Priyanka Yogi: 0000-0002-6639-014X Shailendra K. Saxena: 0000-0001-7156-3407 Pankaj R. Sagdeo: 0000-0002-2475-6676 Rajesh Kumar: 0000-0001-7977-986X Present Addresses $ National Institute for Nanotechnology, University of Alberta, Edmonton, Canada.

F


Article

The Journal of Physical Chemistry A #

Scanning Spreading Resistance Microscopy. J. Phys. Chem. C 2010, 114 (2), 760−765. (15) Fuhrmann, B.; Leipner, H. S.; Höche, H.-R.; Schubert, L.; Werner, P.; Gö sele, U. Ordered Arrays of Silicon Nanowires Produced by Nanosphere Lithography and Molecular Beam Epitaxy. Nano Lett. 2005, 5 (12), 2524−2527. (16) Fukata, N.; Mitome, M.; Bando, Y.; Seoka, M.; Matsushita, S.; Murakami, K.; Chen, J.; Sekiguchi, T. Codoping of Boron and Phosphorus in Silicon Nanowires Synthesized by Laser Ablation. Appl. Phys. Lett. 2008, 93 (20), 203106. (17) Tan, Z.; Shi, W.; Guo, C.; Zhang, Q.; Yang, L.; Wu, X.; Cheng, G.; Zheng, R. Fabrication of Ultra-Thin Silicon Nanowire Arrays Using Ion Beam Assisted Chemical Etching. Nanoscale 2015, 7 (41), 17268−17273. (18) Huang, Z.; Geyer, N.; Werner, P.; de Boor, J.; Gösele, U. MetalAssisted Chemical Etching of Silicon: A Review. Adv. Mater. 2011, 23 (2), 285−308. (19) Li, X.; Bohn, P. W. Metal-Assisted Chemical Etching in HF/ H2O2 Produces Porous Silicon. Appl. Phys. Lett. 2000, 77 (16), 2572−2574. (20) Bai, F.; Li, M.; Huang, R.; Song, D.; Jiang, B.; Li, Y. TemplateFree Fabrication of Silicon Micropillar/Nanowire Composite Structure by One-Step Etching. Nanoscale Res. Lett. 2012, 7 (1), 557. (21) Pinilla, S.; Barrio, R.; González, N.; Pérez Casero, R.; Márquez, F.; Sanz, J. M.; Morant, C. Role of Hydrogen in the Preparation of Amorphous Silicon Nanowires by Metal-Assisted Chemical Etching. J. Phys. Chem. C 2018, 122 (39), 22667−22674. (22) Raman, C. V. A New Radiation. Indian J. Phys. 1928, 2, 387− 398. (23) Raman, C. V.; Krishnan, K. S. A New Type of Secondary Radiation. Nature 1928, 121 (3048), 501−502. (24) Zhang, M.-L.; Fan, X.; Zhou, H.-W.; Shao, M.-W.; Zapien, J. A.; Wong, N.-B.; Lee, S.-T. A High-Efficiency Surface-Enhanced Raman Scattering Substrate Based on Silicon Nanowires Array Decorated with Silver Nanoparticles. J. Phys. Chem. C 2010, 114 (5), 1969− 1975. (25) Wang, R.; Yuan, P.; Han, M.; Xu, S.; Wang, T.; Wang, X. Asymmetry of Raman Scattering by Structure Variation in Space. Opt. Express 2017, 25 (15), 18378−18392. (26) Mala, S. A.; Tsybeskov, L.; Lockwood, D. J.; Wu, X.; Baribeau, J.-M. Raman Scattering in Si/SiGe Nanostructures: Revealing Chemical Composition, Strain, Intermixing, and Heat Dissipation. J. Appl. Phys. 2014, 116 (1), 014305. (27) Imamura, G.; Kawashima, T.; Fujii, M.; Nishimura, C.; Saitoh, T.; Hayashi, S. Raman Characterization of Active B-Concentration Profiles in Individual p-Type/Intrinsic and Intrinsic/p-Type Si Nanowires. J. Phys. Chem. C 2009, 113 (25), 10901−10906. (28) Cerdeira, F.; Cardona, M. Effect of Carrier Concentration on the Raman Frequencies of Si and Ge. Phys. Rev. B 1972, 5 (4), 1440− 1454. (29) Chandrasekhar, M.; Renucci, J. B.; Cardona, M. Effects of Interband Excitations on Raman Phonons in Heavily Doped n-Si. Phys. Rev. B: Condens. Matter Mater. Phys. 1978, 17 (4), 1623−1633. (30) Balkanski, M.; Jain, K. P.; Beserman, R.; Jouanne, M. Theory of Interference Distortion of Raman Scattering Line Shapes in Semiconductors. Phys. Rev. B 1975, 12 (10), 4328−4337. (31) Wu, X.; Yan, F.; Bao, X.; Li, N.; Liao, L.; Zhang, M.; Jiang, S.; Feng, D. Raman Scattering of Porous Structure Formed on C +-implanted Silicon. Appl. Phys. Lett. 1996, 68 (15), 2091−2093. (32) Gouadec, G.; Colomban, P. Raman Spectroscopy of Nanomaterials: How Spectra Relate to Disorder, Particle Size and Mechanical Properties. Prog. Cryst. Growth Charact. Mater. 2007, 53 (1), 1−56. (33) Fano, U. Effects of Configuration Interaction on Intensities and Phase Shifts. Phys. Rev. 1961, 124 (6), 1866−1878. (34) Saxena, S. K.; Borah, R.; Kumar, V.; Rai, H. M.; Late, R.; Sathe, V. G.; Kumar, A.; Sagdeo, P. R.; Kumar, R. Raman Spectroscopy for Study of Interplay between Phonon Confinement and Fano Effect in Silicon Nanowires. J. Raman Spectrosc. 47 (3), 283−288.

Institut für Festkörperphysik, Leibniz Universität Hannover, Appelstraße 2, 30167 Hannover, Germany. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

Authors acknowledge Sophisticated Instrumentation Centre (SIC), IIT Indore, for SEM measurements. Financial support from the Department of Science and Technology (DST), Govt. of India, is also acknowledged. M.T. and A.C. acknowledge IIT Indore for providing financial support through Teaching Assistantship. D.K.P. acknowledges CSIR New Delhi (Grant No. 1061651838) for providing Junior Research Fellowship. Facilities received from Department of Science and Technology (DST), Govt. of India, under the FIST Scheme with grant number SR/FST/PSI-225/2016 is also acknowledged.

(1) Esat, T.; Friedrich, N.; Tautz, F. S.; Temirov, R. A Standing Molecule as a Single-Electron Field Emitter. Nature 2018, 558 (7711), 573. (2) Chi, H.; Zhu, H.-C.; Xu, H.-J.; Shan, X.-D.; Liao, Z.-M.; Yu, D.P. One-Step Growth and Field Emission Properties of SnO2-Capped Silicon Nanowires: A Sn-Catalyzed Approach. J. Phys. Chem. C 2009, 113 (16), 6450−6453. (3) Aghajamali, M.; Xie, H.; Javadi, M.; Kalisvaart, W. P.; Buriak, J. M.; Veinot, J. G. C. Size and Surface Effects of Silicon Nanocrystals in Graphene Aerogel Composite Anodes for Lithium Ion Batteries. Chem. Mater. 2018, 30 (21), 7782−7792. (4) Tsakalakos, L.; Balch, J.; Fronheiser, J.; Korevaar, B. A.; Sulima, O.; Rand, J. Silicon Nanowire Solar Cells. Appl. Phys. Lett. 2007, 91 (23), 233117. (5) Zheng, G.; Patolsky, F.; Cui, Y.; Wang, W. U.; Lieber, C. M. Multiplexed Electrical Detection of Cancer Markers with Nanowire Sensor Arrays. Nat. Biotechnol. 2005, 23 (10), 1294−1301. (6) Vo-Dinh, T.; Dhawan, A.; Norton, S. J.; Khoury, C. G.; Wang, H.-N.; Misra, V.; Gerhold, M. D. Plasmonic Nanoparticles and Nanowires: Design, Fabrication and Application in Sensing. J. Phys. Chem. C 2010, 114 (16), 7480−7488. (7) Morel, P. H.; Haberfehlner, G.; Lafond, D.; Audoit, G.; Jousseaume, V.; Leroux, C.; Fayolle-Lecocq, M.; Baron, T.; Ernst, T. Ultra High Density Three Dimensional Capacitors Based on Si Nanowires Array Grown on a Metal Layer. Appl. Phys. Lett. 2012, 101 (8), 083110. (8) Elfström, N.; Juhasz, R.; Sychugov, I.; Engfeldt, T.; Karlström, A. E.; Linnros, J. Surface Charge Sensitivity of Silicon Nanowires: Size Dependence. Nano Lett. 2007, 7 (9), 2608−2612. (9) Lee, R.; Kwon, D. W.; Kim, S.; Kim, S.; Mo, H.-S.; Kim, D. H.; Park, B.-G. Nanowire Size Dependence on Sensitivity of Silicon Nanowire Field-Effect Transistor-Based PH Sensor. Jpn. J. Appl. Phys. 2017, 56 (12), 124001. (10) Otnes, G.; Borgströ m, M. T. Towards High Efficiency Nanowire Solar Cells. Nano Today 2017, 12, 31−45. (11) Yang, P.; Yan, R.; Fardy, M. Semiconductor Nanowire: What’s Next? Nano Lett. 2010, 10 (5), 1529−1536. (12) Brönstrup, G.; Jahr, N.; Leiterer, C.; Csáki, A.; Fritzsche, W.; Christiansen, S. Optical Properties of Individual Silicon Nanowires for Photonic Devices. ACS Nano 2010, 4 (12), 7113−7122. (13) Bryant, G.; Abeynayake, C.; Thomas, J. C. Improved Particle Size Distribution Measurements Using Multiangle Dynamic Light Scattering. 2. Refinements and Applications. Langmuir 1996, 12 (26), 6224−6228. (14) Celle, C.; Mouchet, C.; Rouvière, E.; Simonato, J.-P.; Mariolle, D.; Chevalier, N.; Brioude, A. Controlled in Situ N-Doping of Silicon Nanowires during VLS Growth and Their Characterization by G


Article


conductor Nanostructures: Analytical Study. Solid State Commun. 2016, 230, 25. (55) Yogi, P.; Mishra, S.; Saxena, S. K.; Kumar, V.; Kumar, R. Fano Scattering: Manifestation of Acoustic Phonons at the Nanoscale. J. Phys. Chem. Lett. 2016, 7 (24), 5291−5296.

(35) Kumar, R.; Sahu, G.; Saxena, S. K.; Rai, H. M.; Sagdeo, P. R. Qualitative Evolution of Asymmetric Raman Line-Shape for NanoStructures. Silicon 2014, 6 (2), 117−121. (36) Richter, H.; Wang, Z. P.; Ley, L. The One Phonon Raman Spectrum in Microcrystalline Silicon. Solid State Commun. 1981, 39 (5), 625−629. (37) Campbell, I. H.; Fauchet, P. M. The Effects of Microcrystal Size and Shape on the One Phonon Raman Spectra of Crystalline Semiconductors. Solid State Commun. 1986, 58 (10), 739−741. (38) Yorikawa, H.; Uchida, H.; Muramatsu, S. Energy Gap of Nanoscale Si Rods. J. Appl. Phys. 1996, 79 (7), 3619−3621. (39) Delerue, C.; Allan, G.; Lannoo, M. Theoretical Aspects of the Luminescence of Porous Silicon. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 48 (15), 11024−11036. (40) Needs, R. J.; Bhattacharjee, S.; Nash, K. J.; Qteish, A.; Read, A. J.; Canham, L. T. First-Principles Calculations of Band-Edge Electronic States of Silicon Quantum Wires. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50 (19), 14223−14227. (41) Luo, S.; Yu, W. B.; He, Y.; Ouyang, G. Size-Dependent Optical Absorption Modulation of Si/Ge and Ge/Si Core/Shell Nanowires with Different Cross-Sectional Geometries. Nanotechnology 2015, 26 (8), 085702. (42) Parkash, V.; Kulkarni, A. K. Optical Absorption Characteristics of Silicon Nanowires for Photovoltaic Applications. IEEE Trans. Nanotechnol. 2011, 10 (6), 1293−1297. (43) Hill, N. A.; Whaley, K. B. A Theoretical Study of Light Emission from Nanoscale Silicon. J. Electron. Mater. 1996, 25 (2), 269−285. (44) Saxena, S. K.; Kumar, V.; Rai, H. M.; Sahu, G.; Late, R.; Saxena, K.; Shukla, A. K.; Sagdeo, P. R.; Kumar, R. Study of Porous Silicon Prepared Using Metal-Induced Etching (MIE): A Comparison with Laser-Induced Etching (LIE). Silicon 2017, 9 (4), 483−488. (45) Saxena, S. K.; Sahu, G.; Kumar, V.; Sahoo, P. K.; Sagdeo, P. R.; Kumar, R. Effect of Silicon Resistivity on Its Porosification Using Metal Induced Chemical Etching: Morphology and Photoluminescence Studies. Mater. Res. Express 2015, 2 (3), 036501. (46) Saxena, S. K.; Yogi, P.; Yadav, P.; Mishra, S.; Pandey, H.; Rai, H. M.; Kumar, V.; Sagdeo, P. R.; Kumar, R. Role of Metal Nanoparticles on Porosification of Silicon by Metal Induced Etching (MIE). Superlattices Microstruct. 2016, 94, 101−107. (47) Yogi, P.; Poonia, D.; Mishra, S.; Saxena, S. K.; Roy, S.; Kumar, V.; Sagdeo, P. R.; Kumar, R. Spectral Anomaly in Raman Scattering from P-Type Silicon Nanowires. J. Phys. Chem. C 2017, 121 (9), 5372−5378. (48) Saxena, S. K.; Yogi, P.; Mishra, S.; Rai, H. M.; Mishra, V.; Warshi, M. K.; Roy, S.; Mondal, P.; Sagdeo, P. R.; Kumar, R. Amplification or Cancellation of Fano Resonance and Quantum Confinement Induced Asymmetries in Raman Line-Shapes. Phys. Chem. Chem. Phys. 2017, 19 (47), 31788−31795. (49) Canham, L. T. Luminescence Bands and Their Proposed Origins In Highly Porous Silicon. Phys. Status Solidi B 1995, 190 (1), 9−14. (50) Yorikawa, H.; Muramatsu, S. Logarithmic Normal Distribution of Particle Size from a Luminescence Line-Shape Analysis in Porous Silicon. Appl. Phys. Lett. 1997, 71 (5), 644−646. (51) Sagdeo, A.; Nagwanshi, A.; Pokhriyal, P.; Sinha, A. K.; Rajput, P.; Mishra, V.; Sagdeo, P. R. Disappearance of Dielectric Anomaly in Spite of Presence of Structural Phase Transition in Reduced BaTiO3: Effect of Defect States within the Bandgap. J. Appl. Phys. 2018, 123 (16), 161424. (52) Kubelka, P. New Contributions to the Optics of Intensely Light-Scattering Materials. J. Opt. Soc. Am. 1948, 38 (5), 448−457. (53) Hessel, C. M.; Wei, J.; Reid, D.; Fujii, H.; Downer, M. C.; Korgel, B. A. Raman Spectroscopy of Oxide-Embedded and LigandStabilized Silicon Nanocrystals. J. Phys. Chem. Lett. 2012, 3 (9), 1089−1093. (54) Yogi, P.; Saxena, S.; Mishra, S.; Mishra, V.; Rai, H. M.; Late, R.; Kumar, V.; Joshi, B.; Sagdeo, P.; Kumar, R. Interplay between Phonon Confinement and Fano Effect on Raman Line Shape for SemiH


Deconvoluting Diffuse Reflectance Spectra for Retrieving

Recommend Documents