See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/41187952

Electrochemical differential photoacoustic cell to

study in situ the growing process of porous
materials

ARTICLE in THE REVIEW OF SCIENTIFIC INSTRUMENTS · JANUARY 2010

Impact Factor: 1.61 · DOI: 10.1063/1.3271238 · Source: PubMed

CITATIONS READS

3 36

7 AUTHORS, INCLUDING:

Adriana Gutiérrez Rodríguez R. Velazquez-Hernandez

Nueva Granada Military University Universidad Tecnológica de Querétaro
7 PUBLICATIONS 10 CITATIONS 12 PUBLICATIONS 38 CITATIONS

SEE PROFILE SEE PROFILE

Alicia Del Real Mario Enrique Rodriguez-Garcia

Universidad Nacional Autónoma de México Universidad Nacional Autónoma de México
23 PUBLICATIONS 63 CITATIONS 33 PUBLICATIONS 84 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Adriana Gutiérrez Rodríguez

Retrieved on: 09 November 2015
 REVIEW OF SCIENTIFIC INSTRUMENTS 81, 013901 共2010兲

Electrochemical differential photoacoustic cell to study in situ the growing

process of porous materials
Adriana Gutiérrez,1 Jairo Giraldo,1 Rubén Velázquez-Hernández,2
Maria Luisa Mendoza-López,3 Diego G. Espinosa-Arbeláez,4 Alicia del Real,4 and
Mario E. Rodríguez-García4,a兲
1
Departamento de Física, Universidad Nacional de Colombia–Sede Bogotá, Avenida Carrera 30 No. 45-03,
Ciudad Universitaria, Bogotá, D.C., Colombia
2
División de Estudios de Posgrado, Facultad de Ingeniería, Universidad Autónoma de Querétaro,
Ciudad Universitaria, Querétaro, Qro, C.P. 76010, Mexico
3
Instituto Tecnológico de Querétaro, Av. Tecnológico S/N, Querétaro, Querétaro, C.P. 76000, Mexico
4
Departamento de Nanotecnología, Centro de Física Aplicada y Tecnología Avanzada, Universidad Nacional
Autónoma de México, Campus Juriquilla, Querétaro, Qro., A.P 1-1010, Mexico
共Received 18 June 2009; accepted 10 November 2009; published online 12 January 2010兲
In order to study in situ the growing process of porous materials, a new electrochemical differential
photoacoustic cell 共DPC兲 was developed. This system allows to obtain the thermal signals coming
from the growing process of the pores without the external noise component. The DPC is a good
system to growth porous silicon and study their growing process with reproducibility. The porous
silicon samples were obtained by using electrochemical etching of 共100兲 n-type silicon wafers with
different nominal resistivity values in the range of 1 – 25 ⍀ cm. The samples were formed in a
solution of hydrofluoric acid and ethanol having a composition ratio of 1:1 in volume with etching
voltage of 10V and an etching time of 2 min using back illumination provided by a laser beam with
a wavelength of 808 nm. The porous samples were characterized by means of Raman microscopy,
x-ray diffraction, and scanning electron microscopy. The crystallite sizes of the samples were
obtained through the analysis of the micro-Raman spectra using a phonon confinement model, and
the analysis of the x-ray diffractograms. © 2010 American Institute of Physics.
关doi:10.1063/1.3271238兴

I. INTRODUCTION reported the morphology of PS samples for different n-type

Porous silicon is a material formed by using anodic dis-
solution of single crystalline silicon in hydrofluoric acid
共HF兲 solutions.1 Recent porous silicon 共PS兲 applications in-
clude development of optical biosensors,2 optical filters for
biosensing applications,3 detection of organic solvents,4 de-
tection of organic vapors such as methanol, ethanol, acetone,
and water,5 and nanocomposites 共deposition of materials into
PS兲6 among others.
The formation of PS was first reported in the late 1950s
as a result of electropolishing of silicon.7 Since then, particu-
larly after 1990 when luminescence of PS was discovered,
extensive research has been undertaken. Some of these in-
vestigations have revealed that PS has many morphological
features with properties that are very different from those of
silicon. Also the formation process of PS is very complex
and depends on many factors such as HF concentration, sili-
con type, current density, and illumination intensity.
Although it has been done in situ studies of PS on n-type
silicon substrates,8,9 these studies did not reveal the intrica-
cies of the morphology variations and their effect on the
thermal and electrical properties of the composite system
silicon/porous silicon 共Si/PSi兲. In addition, Lehmann et al.10
a兲
Author to whom correspondence should be addressed. Electronic mail:
marioga@fata.unam.mx.
silicon substrates with doping densities between 6 ⫻ 1016 and
2 ⫻ 1019 cm−3 range; however, PS samples on n-type silicon
substrates with low doping densities and low porosities have
not been extensively studied to date.
The thermal and optical properties of n-type PS have
been studied by means of the photoacoustic 共PA兲
technique.11–13 It is well known that the PA technique is a
nondestructive method in which the optical and thermal
properties of different materials such as semiconductors,14
metals,15 and composite can be studied.16 However, in the
configurations used by Benedetto et al.,11 Bernini et al.,12
and Sheng et al.,13 it is necessary to have the microphone
calibrated as a function of the frequency or the wavelength.
The microphone response contains fluctuations from the light
source and room noises; this response is used to obtain the
physical parameters of the sample under study. Conse-
quently, it is necessary to measure the initial reference
sample signal and then measure the signal of the sample.
These measurements are taken at different times and under
varying experimental conditions. Using a differential photoa-
coustic cell 共DPC兲, it is possible to obtain both signals simul-
taneously, the sample being tested and the reference sample.
Recently Rodríguez et al.17 developed a DPC to study in situ
the process of zinc electrodeposition on steel substrate.
Further, Raman microscopy is another tool used in order
to characterize porous semiconductors. This technique al-

0034-6748/2010/81共1兲/013901/8/$30.00 81, 013901-1 © 2010 American Institute of Physics

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
013901-2 Gutiérrez et al.
lows investigators to estimate the nanocrystal size from the
wave number shift and line broadening of the spectrum using
confinement models. A recent review of the application of
this technique for that reason can be found in Irmer.18 It
has also been confirmed that studying the crystallinity
grade of the porous samples utilizing x-ray diffraction is a
suitable technique, which has already been performed many
times.19–22
The most common method used to grow PS is direct
current 共dc兲.23 Using dc anodic process, the reaction pro-
duces silicon fluoride that tends to deposit at the pore tips
creating as result that H2 bubbles are absorbed at the surface
of the sample, blocking the silicon pores and leading to a
reduction of HF inside the pores.24 As a result the bubble
formation affects the etching process reducing primarily the
current density. Hou et al.25 proposed a pulse anodic etching
process at 100 and 2000 Hz that results in more uniform PS.
According to their result the periodical behaviors of the etch-
ing cause the recovery of the HF concentration inside of the
pores and removal of the bubbles from the surface. Liu et
al.26 using ultrasonically enhanced anodic electrochemical
etching that uses acoustic cavitation produce the dissolving
of H2 bubbles that can accelerate the PS growth. In the afore-
mentioned works it is clear that the use of ac anodic etching
can reduce the formation of H2 bubbles on the Si surface
allowing the growth of PS.
In this research, a new electrochemical DPC to study
in situ the formation of porous materials is presented. The
in situ formation process of PS samples obtained by electro-
chemical etching of 共100兲 n-type silicon wafers with differ-
ent nominal resistivity values in the range of 1 – 25 ⍀ cm by
means of the DPC development was studied for the first
time. Moreover, the characterization of the PS samples was
done using Raman microscopy, x-ray diffraction, and scan-
ning electron microscopy 共SEM兲. The electrochemical etch-
ing was done under ac optical excitation at 76 Hz and dc
current.
II. EXPERIMENTAL SETUP
A. Differential photoacoustic cell
The PA technique is based on the PA effect. To observe
this effect, a gas is confined generally air, inside a camera, or
cell with the sample like one of their walls. The sample is
illuminated with light of modulated intensity and transfers
heat by means of pulses to the gas generating pressure fluc-
tuations that are registered by means of a microphone.
This frequency is selected taking into account the ther-
mal diffusivity of the sample and its thickness in order to
guaranty the thermally thick regimen. It means that any
change of the sample located at the thermal diffusion length
共␮ = 冑␣ / ␲ f兲 can be registered by the microphone. The PA
effect has also been applied for calorimetric studies.27 In
brief, if the system being studied undergoes any physico-
chemical transformation as a result of the external perturba-
tion 共heat, electrical current, etc.兲, there is absorption and
emission of heat. The gas in the chamber registers these pro-
cesses as an increment or decrement of the PA amplitude and
Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
Rev. Sci. Instrum. 81, 013901 共2010兲
FIG. 1. 共Color online兲 共a兲 Front view of the DPC without laser contact. 共b兲
Schematic diagram of the experimental setup of the DPC for the PS growth.
phase signals, due to the changes in the absorption coeffi-
cient of the sample. A detailed description of the DPC prin-
ciple has been published recently.17
In this work a new DPC was developed for the in situ
study of the formation process of porous materials and the
fabrication procedure of the samples.
The DPC is composed of polytetrafluoroethylene, PTFE,
Teflon® that serves as a vessel for the electrolytic solution.
One of the walls of the cell contains two conventional pho-
toacoustic cells 共PCs兲, one sample was used as a reference
without any electrochemical treatment, and the other sample
was the one being studied. The sample being studied and the
reference sample were obtained from the same Si substrate.
Each PC is a cylinder with diameter of 1 cm and length of
0.5 cm and contains a quartz window in which the air is
contained. There are two microphones 共electrets兲, one for
each PC. Figure 1共a兲 shows the frontal view of the DPC with
the two PCs for both samples.
A laser beam 共LASERMATE GROUP INC. LBG
8080250 A5-T兲 with a wavelength of 808 nm and power of
64 mW was used as the excitation source. The excitation
beam was modulated at 76 Hz with the internal oscillator of
the lock-in amplifier 共Stanford Research System SR830兲. By
utilizing a beam splitter the incident radiation is divided into
two beams, 60/40 respectively, each of these is then focused
on each one of the PCs. The output signals are also known as
amplitude and phase. Each one of the microphones was po-
larized using a 9 V dc battery to avoid the preamplified step
that could increase the noise and reduce the sensitivity of the
system. Polarized microphone signals were connected to a
lock-in amplifier. The experimental setup of the DPC for the
fabrication of the PS samples is shown in the Fig. 1共b兲.
To obtain the instrumental function of the DPC 共IF兲, it is
necessary to simultaneously measure both signals and con-
sider the fact that both cells are equal in size. The variations
coming from the laser and voltage changes are all involved
in this instrumental function that is time dependent 关IF共t兲兴.
Any modulated signal can be described as the real part
of an imaginary quantity; therefore, the photoacoustic signal
共PAS兲 can be described by means of Eq. 共1兲,
PAS = rSei␪S , 共1兲
where rS and ␪S are related to the amplitude and phase sig-
nals, respectively.
013901-3 Gutiérrez et al. Rev. Sci. Instrum. 81, 013901 共2010兲

The detected PAS consist of two signals; one of these is

coming from the sample, it is an ac signal 共Sac兲, and the
another is coming from the external noise 共SN兲 which can be
considered as a random signal,28

PAS = Sac · SN . 共2兲

According to Eq. 共2兲, the sample signal 共PASs兲 and the
sample reference signal 共PASref兲 can be written into Eqs. 共3兲
and 共4兲, respectively,

PASs = Sacs · SNs , 共3兲

PASref = Sacref · SNref . 共4兲

The PAS measurements of the sample and the reference
sample are carried out simultaneously and are under the
same experimental conditions; for this reason, the noise will
simultaneously affect both samples, so the noise components
SNs and SNref are equals. Finally, the instrumental function is
represented by Eq. 共5兲,

PASs
IF共t兲 = . 共5兲
PASref

Consequently, the amplitude and phase of the instrumental

function are

rs
rIF共t兲 = , 共6兲 FIG. 2. 共Color online兲 共a兲 shows a comparison between DSC signal and
rref
DPC for endothermic transition of pure Ga taken as reference and 共b兲 and
共c兲 show the PA amplitude and phase signal for both cells as a function of
␪IF共t兲 = ␪s − ␪ref . 共7兲 the time.

Figure 2共a兲 shows a comparison between differential

B. Sample preparation
scanning calorimetry 共DSC兲 of pure Ga 共99.9999%, Alpha
Aesar-USA兲 running in a MDSC TA Instruments model PS samples were obtained by using electrochemical
2920, under the same calibration conditions: constant cell etching of 共100兲 n-type silicon wafers with different nominal
with Indio in 1.0908, temperature at 156.65 ° C and, heat resistivity values in the range of 1 – 25 ⍀ cm 共Mittel,
capacity constant with sapphire at 1.1740 and PA amplitude Bromont-Ontario, Canada兲, see Table I. All the samples were
signal obtained by using a DP system in order to detect the cleaned before being placed in the cell in order to remove the
endothermic transition as a function of the temperature. It is native oxide of the surface. The cleaning process consisted of
clear that the endothermic transition can be sensed by the two solutions used in sequence. The first was composed of
PC. The microphone has a sensitivity of 1mV/mbar and 1:1:5 volumes of NH4OH 共28 wt %兲:H2O2 共4 wt %兲:H2O.
spectral response from 20 Hz to 15 KHz.31 The other solution consisted of HCl 共38 wt %兲: H2O2
Figures 2共b兲 and 2共c兲 show the PA signal amplitude and 共4 wt %兲:H2O in a 1:1:6 volume ratio. Each one of the
phase as a function of time 共time scan兲 of two n-type silicon cleaning steps was carried out for 10–20 min at 75– 85 ° C,
samples of the wafer O5 without any treatment. Although the under rapid agitation conditions. The samples were fabri-
samples have the same properties, the cells present differ- cated at room temperature in an electrolytic etching bath of
ences due to the fabrication process. 1:1 mixture of hydrofluoric acid 共HF 48 wt %兲 and ethanol
As a result, it is possible to obtain, without noise, the
PAS 共amplitude and phase兲 that arises from the electrochemi- TABLE I. Nominal resistivity and thickness values for the n-type silicon
wafers used as substrates to produce the porous silicon samples.
cal reactions and follows in situ the process of the porous
growth. The time scan for amplitude and phase using DPC Nominal resistivity Thickness
reveals the following values: for cell 1 共microphone A兲, the Si wafers 共⍀ cm兲 共␮m兲
average value of the amplitude was 0.0256⫾ 4.596e
O1 1–5 523
− 5 mV and phase was −24.761⫾ 0.0931°. In the case of cell
O2 5–10 517
2 共microphone B兲 these values were 0.0306⫾ 3.897e − 5 and
O3 10–15 530
−28.745⫾ 0.0956°. The aforementioned values under con- O4 15–20 516
stant temperature demonstrate that the DPC has a good sen- O5 20–25 502
sitivity to detect endothermic or exothermic phenomena.

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
013901-4 Gutiérrez et al.
FIG. 3. 共Color online兲 PA amplitude 关共a兲 and 共b兲兴 and phase 关共c兲 and 共d兲兴 signals for two samples 共O3 and O3D兲 and reference samples of the wafer O3 as
a function of the time; 共↑兲: applied voltage on; 共↓兲: applied voltage off. Dashed vertical lines were included to guide the eye. The applied voltage was 10 V.
共96 wt %兲 under constant voltage 共10 V兲. The etching time
was 2 min. With platinum slab as a cathode and silicon as
anode, anodization was done using the DPC as described
above, and was back illuminated with a laser 共LASERMATE
GROUP INC. LBG 8080250 A5-T兲 with a wavelength of
808 nm and 76 Hz. At this wavelength the optical absorption
coefficient of the crystalline silicon is 9.3⫻ 104 m−1.29 The
thickness of the samples was 523 ␮m 共see Table I兲, so the
optical extinction length of 1 / ␤ is lower than the sample
thickness and the sample can be considered optically opaque.
The excitation beam was modulated at 76 Hz with the inter-
nal oscillator of the lock-in amplifier. For this frequency
value the sample can be considered as thermally thick.
In order to make Ohmic contacts in the sample, gold was
sputtered on one of the corners of the back surface of each
sample 共EMS 550 Sputter coater兲, and a small wire was fixed
to the Si with indium solder 共Puratronic®兲. Both samples
were placed in their respective PC by means of silicon rubber
共Silicon RTV®兲. The current data were collected using a mul-
timeter Keithley 2000 with a GPIB universal serial bus con-
nected directly to a personal computer. After anodization, the
porous sample was retired from the cell, rinsed in ethanol,
and stored in a dryer at room temperature.
C. Raman spectra
The Raman spectra of silicon and PS samples were mea-
sured at room temperature in a commercial micro-Raman
system 共LABRAM II model of Dilor兲 equipped with a 25
mW He–Ne laser emitting at 632.8 nm and spot size of
10 ␮m. A 256⫻ 1024 pixel charge-coupled device 共CCD兲
was used as a detector, a computer-controlled XY stage with
a spatial resolution of 0.1 ␮m, and a confocal microscope
Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
Rev. Sci. Instrum. 81, 013901 共2010兲
possessing a 10⫻ objective. Raman experiments were car-
ried out for all samples the same day of the etching process
in order to avoid oxidations problems.
D. X-ray diffraction studies
The x-ray diffraction measurements of the samples were
made in a Bruker D8 advance x-ray diffractometer system,
with the Cu K␣ line 共␭ = 1.5406 Å兲 at a grazing angle of
共␪ = 2°兲 and Bragg–Brentano geometry.
E. SEM
A SEM study was carried out in a JEOL JSM-6060 LV
microscope using 20 kV electron beam. The SEM study was
done in order to show the porous that were formed on the
silicon surface of each sample produced.
III. RESULTS AND DISCUSSION
A. In situ study of the growing process
Figures 3共a兲 and 3共c兲 show the PA amplitude and
phase signals of the PS sample 共O3兲 and the
reference sample which corresponds to the substrate O3
共␳ ⬃ 10– 15 ⍀ cm兲 as a function of time. The anodization
voltage was Va = 10 V. Each one of the growing processes
was carried out twice in order to study the reproducibility of
the DPC as compared to the PS growth. Figures 3共b兲 and
3共d兲 show the PA signal for the duplicate sample 共sample
O3D兲. The arrows in these figures indicate the moment when
the voltage source was turned-on and turned-off. The con-
tinuous line in this figure shows the PA amplitude and phase
signals of the reference sample.
013901-5 Gutiérrez et al. Rev. Sci. Instrum. 81, 013901 共2010兲

FIG. 4. 共Color online兲 PA amplitude 关part 共i兲兴 and phase 关part 共ii兲兴 signals for three samples and reference samples of three porous silicon samples as a function
of the time: 共a兲 Sample O1, 共b兲 sample O4, and 共c兲 sample O5. In part 共iii兲 the current curve as a function of the time has been added. Arrows indicate the
moment of turn-on and turn-off the voltage 关共↑兲: applied voltage on; 共↓兲: applied voltage off兴, and the dashed vertical lines were included to guide the eye. The
applied voltage was 10 V.

Comparing Figs. 3共a兲 and 3共b兲, with Figs. 3共c兲 and 3共d兲,
respectively, it is noteworthy that the signals of samples O3
and O3D are similar. Consequently, DPC is a good system to
use in order to study the growing process and reproducibility
of PS.
Figure 4 shows the PA amplitude and phase signals part
共i兲 and part 共ii兲, respectively, and the current curve 关part 共iii兲兴
as a function of the time for three different PS samples; all
corresponding to the substrates seen in O1 of Fig. 4共a兲, O4 in
Fig. 4共b兲, and O5 in Fig. 4共c兲, respectively 共see Table I兲. All
the samples grown in different substrates exhibit a similar
behavior for the PA signal: an increase in the amplitude sig-
nal and a decrease in the phase signal. When the voltage
source is turned on, the signals from the reference sample do
not have temporal changes, indicating that the reference
sample does not exhibit any changes during the electro-
chemical process.
The current in all graphics of Fig. 4, part 共iii兲 shows two
different behaviors with the etching time. The first occurs at
approximately 20–30 s and is related to the porous nucle-
ation process. While the second corresponds to the remaining
time and it is related to the porous growth process. These
results concur to the stages of growth process as proposed by
Kwon et al.21 and Popkirov and Ottow.30
It is noteworthy that the PA amplitude and phase signals
during the growth process are noncontinuous demonstrating
semiperiodic shape. The main source of the acoustic signal
arises from the periodic heat flow from the solid to the sur-
rounding gas as the solid is periodically heated by the ab-
sorption of the modulated light. This signal also depends on
the thermal and optical changes of the sample and the geom-
etry of the cell.31 Due to the aforementioned factors, in this
study, the modulated frequency remains constant, and the
signal from the reference sample is taken in situ. The
changes in the PA signal are associated with changes in the
thermal and optical properties of the silicon during the elec-
trochemical etching process. The PA signals are coming from
formation of the porous layers as well as the substrate and its
ratio is a function of time because the thickness of the PS
increases. Abdallah et al.32 and Shen et al.33 found that PA
signal increases as the thickness of the PS layer is increased.
In terms of the thermal properties, it is possible that due to

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
013901-6 Gutiérrez et al. Rev. Sci. Instrum. 81, 013901 共2010兲

the creation of porous the thermal diffusivity of the surface

decreases in relation to the substrate and the heat remaining
on the surface. In terms of the optical absorption contribution
to the PA signals, it is well known that a blueshift in the band
gap of PS exists in relation to the 1.1 eV corresponding to
silicon. Shen et al.33 found that there is a direct correlation
between the shift and the PA signal. If the band gap in-
creases, the PA signal increases. On the other hand, it is also
necessary to take into account a new contribution to the PA
signal originating from the quantum confinement of carriers
located in the PS.
The PA signals in Figs. 3 and 4 can be explained as
follows: the increases in the signal are related to the porous
creation, changes in the thermal properties, and changes in
the optical absorption coefficient. The most important feature
in these figures is that after a characteristic time, the PA FIG. 5. 共Color online兲 Raman spectra of the porous silicon samples. The
amplitude signal decreases. This implies that the porous for- inset in this figure corresponds to the Raman peak of silicon.
mation is a cyclic process that is governed by the electro-
chemical behaviors.
The decreases in the PA phase signal are related to the Therefore both amplitude and phase signals return to their
electronic contribution; after the porous generation and due initial values.
to the back illumination a carrier generation exists and there According to the results obtained using the developed
is a diffusion of those carriers into the silicon substrate. The DPC this technique permitted the study of in situ growing
changes in amplitude and phase of the etched samples can be process of PS including the thermal processes that occur in it
attributed to the thermal process that take place during the for the first time.
electrochemical etching, and can be explained as follows:
initially when the voltage source is turned on there is a
change in the amplitude and phase signals of part 共i兲 and part B. Raman microscopy
共ii兲 corresponding to an increase and a decrease, respectively.
These changes are due to the holes generated in the substrate In order to study the crystallite size of the porous
by the back illumination that begin to move to the front samples, the Raman spectra were measured for each one of
surface of the substrate where they weaken the Si–Si bonds, the etched samples as well as for the silicon 共O3兲 in order to
permitting them to be attacked by fluorine ions. The changes determine the Raman peak position of the sample without
in the minority carrier density act as a thermal source that is etching. The profiles are shown in Fig. 5; the inset in this
detected by means of the PC. The dissolution process that figure correspond to the Raman peak of silicon located at
occurs next in the surface/electrolyte interface can be written 520.5 cm−1. It is noteworthy that the Raman peaks for SiO2
as34 glass 共430 cm−1兲 and crystalline quartz 共464 cm−1兲 共Ref.
37兲 do not appear, so the line broadening of the spectra in the
Si + 2HF + nh+ → SiF2 + 2H+ + 共2 − n兲e− , 共8兲 Fig. 5 are due to the pores.
The crystallite size was obtained by means of the pho-
SiF2 + 4HF → H2SiF6 + H2 , 共9兲 non confinement model proposed by Ritcher et al.36 and im-
proved by Campbell and Fauchet37 for the phonons in nano-
where h+ and e− are the exchanged hole and electron, respec-
spheres of diameter ᐉ. The Raman scattering intensity is then
tively, and n is the number of charges exchanged during the
described by a superposition of Lorentzian curves with band-
dissolution process.34 The reaction scheme was proposed by
width ⌫ centered at wave numbers ␻共q兲 of the phonon dis-
Lehman and Gösele,35 which explains the dissolution pro-
persion curve and weighed by the 兩C共qជ 0 , qជ 兲兩2 factor. The Ra-
cess. The reactions described by the Eqs. 共8兲 and 共9兲 are two
man intensity can be written as38
thermal processes that are present during all the chemical
dissolution; the first is an endothermic process while the
second is exothermic.31 These thermal processes act as two TABLE II. Crystallite sizes of the porous silicon samples produced on dif-
cyclical heat sources due to the generation of holes in a pe- ferent substrates determined by fitting Raman spectra by means of the pho-
non confinement model.
riodical process coming from the modulated back illumina-
tion, so the DPC can detect them. As a result, PA signal ⬃Crystallite size
varies as a function of the time. At 76 Hz the silicon sample Silicon substrate 共Å兲
is thermally thick and the thermal diffusion length 共␮
= 冑␣ / ␲ f, ␣ is the thermal diffusivity兲 is equal to the sample O1 350.5
O2 300.1
thickness; thus, the signal changes can be attributed to these
O3 320
thermal processes that occur on the sample surface. Finally, O4 350
at the moment that the voltage source is turned off the carrier O5 1150
diffusion process is ended and the growing process also ends.

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
013901-7 Gutiérrez et al. Rev. Sci. Instrum. 81, 013901 共2010兲

FIG. 6. X-ray diffraction patterns of

two porous silicon samples produced.
共a兲 Sample O3 and 共b兲 sample O5.

冕 1
兩C共qជ 兲兩2 Similar to the research done by Kwon et al.,21 using the

冉冊
I共␻兲 ⬀ d 3q 2,
0 ⌫
共␻ − ␻共q兲兲2 −
2 0.94␭
with
2 where ␭ = 1.5406 Å is the wavelength of the incident beam
兩C共qជ 兲兩2 = e−共qᐉ/2a兲 ,
where a is the lattice spacing for silicon.
In order to numerically simulate the Raman intensity of
different PS samples, using the phonon confinement model
described above, the phonon dispersion around q = 0 was as-
sumed to be39
␻共q兲 = 共A + xc兲 − Bq , 2
where A is the near-zone-center LO-mode frequency
共A = 520 cm−1兲 and the value of the fitting parameter B
= 120 cm for silicon. In Eq. 共12兲 the value xc that was added
corresponds to the peak shift of each spectrum relative to the
silicon single crystal 共520 cm−1兲, and the ⌫ value was taken
as the full width at half the maximum 共FWHM兲.
The fitted Raman spectra of PS samples produced are
shown in Fig. 5. The crystallite sizes are recorded in Table II.
According to these results, the structure sizes of the samples
produced are found in the range of 244–1150 Å.
C. X-ray diffraction
Figure 6 shows the x-ray diffraction patterns of two PS
samples, Fig. 6共a兲 corresponds to the porous sample grown
on substrate O3, and Fig. 6共b兲 is the x-ray diffraction pattern
of the porous sample growth on substrate O5. The diffracto-
grams show a 311 orientation with a peak at 56.12° and
56.27°, respectively, which are characteristic of nanocrystal-
lites present on the porous silicon.7,21,40,41
共10兲 Scherrer formula the silicon crystallite sizes were obtained,
D= , 共13兲
w cos共␪兲
共11兲
共Cu K␣1兲, and w is the peak width 共FWHM兲.42 The results
are summarized in Table III. The crystallite sizes are in range
of 242–1130 Å corresponding to the results found in Raman
microscopy analysis.
D. SEM images
共12兲 Figure 7 shows the SEMs of two of the n-type PS
samples. These SEM images show that the pore sizes are
lower than 1 ␮m. Therefore the crystallite sizes obtained by
the Raman microscopy analysis and the x-ray diffractograms
analysis correspond to the pores images.
IV. CONCLUSIONS
A new electrochemical system based on a DPC was de-
veloped to study in situ the electrochemical changes that take
place in the formation process of porous materials under ac
excitation and dc etching. This cell permitted the simulta-
neous growth of PS samples and allowed the in situ study of
the growth process and the physical processes that occur in
it. The changes in the PAS are associated with changes in the
thermal and optical properties of the silicon during the anod-
ization process. In addition, the porous formation is a cyclic
process which is governed by the electrochemical behaviors,
namely, the endothermic and exothermic periodic processes.
In relation to the formation of bubbles, the ac signal 共76 Hz兲
reduces the size and permanence of them on the surface but
more detailed studied are required.

TABLE III. Crystallite sizes of the porous silicon samples produced on

different substrates determined using the Scherrer formula for the x-ray
diffraction patterns.

⬃Crystallite size
Silicon substrate 共Å兲

O1 350
O2 315.77
O3 370
O4 310.66 FIG. 7. SEMs of two of the porous silicon samples for n-type 共100兲 silicon
O5 1130 electrodes anodized potentiostatically in ethanoic HF. 共a兲 Sample O1 and 共b兲
sample O4.

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp
013901-8 Gutiérrez et al. Rev. Sci. Instrum. 81, 013901 共2010兲

15
ACKNOWLEDGMENTS C. Vázquez-López, A. Calderón, M. E. Rodríguez, R. Colas, and S. Valti-
erra, J. Mater. Res. 15, 85 共2000兲.
This work was partially supported by COLCIENCIAS 16
M. E. Rodríguez, P. J. Mendoza, A. Mandelis, and L. Nicolaides, Nucl.
共Colombia兲, DINAIN-Universidad Nacional de Colombia Instrum. Methods Phys. Res. B 184, 421 共2001兲.
17
共Colombia兲, and Project PAPIIT 120809 UNAM 共México兲. M. E. Rodríguez-García, R. Velásquez-Hernández, M. L. Mendoza-López,
D. M. Hurtado-Castañeda, K. M. Brieño-Enríquez, and J. J. Pérez-Bueno,
A. Gutierrez wishes to thank COLCIENCIAS 共Colombia兲 Rev. Sci. Instrum. 78, 034904 共2007兲.
共Programa de Créditos Condonables para Doctorados Na- 18
G. Irmer, J. Raman Spectrosc. 38, 634 共2007兲.
cionales 2004兲 and Dirección Nacional de Investigación DI- 19
D. Buttard, D. Belet, G. Dolino, and T. Baumbach, J. Appl. Phys. 83, 5814
NAIN, Universidad Nacional de Colombia 共Colombia兲 for 共1998兲.
20
their support. We want to thank I. Q. Carmen Vazquez Ra- V. Chamard and G. Dolino, J. Appl. Phys. 89, 174 共2001兲.
21
D. R. Kwon, S. Ghosh, and C. Lee, Mater. Sci. Eng., B 103, 1 共2003兲.
mos for her technical support of the sample preparation. We 22
P. G. Abramof, A. F. Belot, A. Y. Ueta, and N. G. Ferreira, J. Appl. Phys.
also want to acknowledge Silvia Stroet for efforts editing of 99, 024304 共2006兲.
this paper. 23
L. T. Canham, Appl. Phys. Lett. 57, 1046 共1990兲.
24
V. Lehmann, J. Electrochem. Soc. 140, 2836 共1993兲.
1 25
X. G. Zhang, in Modern Aspects of Electrochemistry, edited by C. G. X. Hou, H. Fan, L. Xu, F. Zhang, M. Li, M. Yu, and X. Wang, Appl. Phys.
Vayenas, R. E. White, and M. E. Gamboa-Adelco 共Springer, New York, Lett. 68, 2323 共1996兲.
26
2006兲, Vol. 39, p. 65. Y. Liu, Z. H. Xiong, Y. Liu, S. H. Xu, X. B. Liu, X. M. Ding, and X. Y.
2
V. Torres-Costa, F. Agulló-Rueda, R. J. Martín-Palma, and J. M. Martínez- Hou, Solid State Commun. 127, 583 共2003兲.
Duart, Opt. Mater. 共Amsterdam, Neth.兲 27, 1084 共2005兲. 27
S. E. Braslavsky and G. E. Heibel, Chem. Rev. 92, 1381 共1992兲.
3 28
M. Arroyo-Hernández, R. J. Martín-Palma, V. Torres-Costa, and J. M. F. Reif, Fundamentals of Statistical and Thermal Physics 共McGraw Hill,
Martínez Duart, J. Non-Cryst. Solids 352, 2457 共2006兲. New York, 1965兲, p. 582.
4 29
M. Archer, M. Christophersen, and P. M. Fauchet, Sensors Actuators B D. F. Edwards, in Handbook of Optical Constants of Solids, edited by E.
106, 347 共2005兲. D. Palik 共Academic, New York, 1998兲, p. 552.
5
G. García Salgado, T. Díaz-Becerril, H. Juárez Santiesteban, and E. 30
G. S. Popkirov and S. Ottow, J. Electroanal. Chem. 429, 47 共1997兲.
Rosendo Andrés, Opt. Mater. 共Amsterdam, Neth.兲 29, 51 共2006兲. 31
A. Rosencwaig and A. Gersho, J. Appl. Phys. 47, 64 共1976兲.
6
R. Hérino, Mater. Sci. Eng., B 69, 70 共2000兲. 32
S. Abdallah, T. A. El-Brolossy, S. Negm, and H. Talaat, Eur. Phys. J.
7
X. G. Zhang, Electrochemistry of Silicon and Its Oxides 共Kluwer Aca- Spec. Top. 153, 199 共2008兲.
demic, Hornell, New York, 2001兲, p. 87. 33
8 Q. Shen, T. Takahashi, and T. Toyoda, Anal. Sci. 17, S281 共2001兲.
M. Safi, J. N. Chazalviel, M. Cherkaoui, A. Belaïdi, and O. Gorochov, 34
R. Memming and G. Schwandt, Surf. Sci. 4, 109 共1966兲.
Electrochim. Acta 47, 2573 共2002兲. 35
9 V. Lehmann and U. Gössele, Appl. Phys. Lett. 58, 856 共1991兲.
K. W. Kolasinski, J. C. Barnard, L. Koker, S. Ganguly, and R. E. Palmer, 36
Mater. Sci. Eng., B 69, 157 共2000兲. H. Richter, Z. P. Wang, and L. Ley, Solid State Commun. 39, 625 共1981兲.
37
10
V. Lehmann, R. Stengl, and A. Luigart, Mater. Sci. Eng., B 69, 11 共2000兲. I. H. Campbell and P. M. Fauchet, Solid State Commun. 58, 739 共1986兲.
38
11
G. Benedetto, L. Boarino, and R. Spagnolo, Appl. Phys. A: Mater. Sci. G. Gouadec and P. Colomban, Prog. Cryst. Growth Charact. Mater. 53, 1
Process. 64, 155 共1997兲. 共2007兲.
39
12
U. Bernini, P. Maddalena, E. Massera, and A. Ramaglia, J. Opt. A, Pure B. K. Patel, R. Mythilib, R. Vijayalaxmib, R. K. Sonic, S. N. Behera, and
Appl. Opt. 1, 210 共1999兲. S. N. Sahua, Physica B 322, 146 共2002兲.
13 40
Ch. K. Sheng, W. M. M. Mahmood Yunus, W. M. Z. W. Yunus, and Z. A. G. Cullis, L. T. Canham, and P. D. J. Calcott, J. Appl. Phys. 82, 909
Abidin Talib, and A. Kassim, Physica B 403, 2634 共2008兲. 共1997兲.
41
14
M. E. Rodriguez, J. J. Alvarado, I. Delgadillo, O. Zelaya, H. Vargas, F. K. Hong and Ch. Lee, J. Korean Phys. Soc. 42, S671 共2003兲.
42
Sánchez Sinencio, M. Tifito, and L. Baños, Phys. Status Solidi A 158, 67 B. D. Cullity, Elements of X-ray Diffraction 共Addison-Wesley, Menlo
共1996兲. Park, CA, 1978兲, p. 284.

Downloaded 12 Jan 2010 to 132.248.179.22. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/rsi/copyright.jsp