key: cord-0690715-ytovslf6
authors: Granada-Ramirez, D.A.; Arias-Cerón, J.S.; Pérez-González, M.; Luna-Arias, J.P.; Cruz-Orea, A.; Rodríguez-Fragoso, P.; Herrera-Pérez, J.L.; Gómez Herrera, M.L.; Tomás, S.A.; Vázquez-Hernández, F.; Durán-Ledezma, Angel A.; Mendoza-Alvarez, J.G.
title: Chemical synthesis and optical, structural, and surface characterization of InP - In(2)O(3) quantum dots
date: 2020-07-22
journal: Appl Surf Sci
DOI: 10.1016/j.apsusc.2020.147294
sha: 1a022b9579fbbdf76f47b96e5f43816f3d2dad21
doc_id: 690715
cord_uid: ytovslf6

InP-In(2)O(3) colloidal quantum dots (QDs) synthesized by a single-step chemical method without injection of hot precursors (one-pot) were investigated. Specifically, the effect of the tris(trimethylsilyl)phosphine, P(TMS)(3), precursor concentration on the QDs properties was studied to effectively control the size and shape of the samples with a minimum size dispersion. The effect of the P(TMS)(3) precursor concentration on the optical, structural, chemical surface, and electronic properties of InP-In(2)O(3) QDs is discussed. The absorption spectra of InP-In(2)O(3) colloids, obtained by both UV-Vis spectrophotometry and photoacoustic spectroscopy, showed a red-shift in the high-energy regime as the concentration of the P(TMS)(3) increased. In addition, these results were used to determine the band-gap energy of the InP-In(2)O(3) nanoparticles, which changed between 2.0 and 2.9 eV. This was confirmed by Photoluminescence spectroscopy, where a broad-band emission displayed from 2.0 to 2.9 eV is associated with the excitonic transition of the InP and In(2)O(3) QDs. In(2)O(3) and InP QDs with diameters ranging approximately from 8 to 10 nm and 6 to 9 nm were respectively found by HR-TEM. The formation of the InP and In(2)O(3) phases was confirmed by X-ray Photoelectron Spectroscopy.

Colloidal quantum dots (QDs) are some of the most extensively investigated nanomaterials due to their outstanding properties, which have brought them to be considered as the next-generation semiconductors. One remarkable property of QDs is the change in optical response as a function of their size [1] , thus achieving a well-defined emission peak at a specific value in the visible spectral region. As a consequence, highly luminescent QDs-based devices could be designed for novel technological applications, including monochromatic and white LEDs [2] , humidity sensors [3] , and displays [4] , among others. Other potential biomedical applications of QDs include their use in drug delivery, imaging, therapy, and as biomarkers.

Particularly, they could help face important diseases such as COVID-19, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [5] [6] .

Nowadays, III-V QDs are more preferred than typical II-VI compounds because of their lower toxicity [7] .

Moreover, these luminophores are more photostable than organic molecular compounds [8] . Over the years, many semiconductor compounds have been synthesized, including InAs, InSb, GaAs, InP, Mn:InP/ZnS, Ag:InP/ZnS, and Cu:InP/ZnS [6, [9] [10] [11] [12] . Particularly, InP has emerged as an exceptional material due to its outstanding high-performance luminescence, stability, large intrinsic extinction coefficient, large Bohr exciton radius, and wide emission range [13] . Up to now, InP QDs have been prepared by conventional techniques, including solvothermal [14] , wet chemical [3] , and one-pot [15] methods. The latter synthesis process has the advantage of producing QDs of high purity and good quality under easily controlled experimental conditions.

In addition, large-scale production of highly luminescent InP-based core-shell structures can be easily carried out [15] .

The InP luminescent properties can be enhanced by using several procedures, such as the coating of InP QDs with other semiconductors [11, 12] . For instance, our group recently reported on the structural, optical, chemical surface, and electronic properties of InP@ZnS QDs as a function of the indium myristate (IM) precursor concentration used during the synthesis process. It was found that at higher IM concentrations, the band-gap energy decreased while the QDs size increased [16] . Another alternative is the use of InP-In 2 O 3 QDs. For instance, Protière and coworkers synthesized InP/In 2 O 3 nanoparticles by adding oleylamine to the precursor materials (indium acetate, myristic acid, and octadecene) to produce the indium oxide coating. They reported that both the higher band gap energy of In 2 O 3 and the band offset in the InP/In 2 O 3 system can promote an efficient confinement of electron-hole pairs in the InP core [17] . In addition, it has been found that the In 2 O 3 shell can impede oxygen permeation to the InP core, thus suppressing the QD photooxidation [18] .

In this work, colloidal InP-In 2 O 3 QDs were synthesized by a single-step chemical process with no injection of hot precursors (one-pot). In particular, the tris(trimethylsilyl)phosphine (P(TMS) 3 ) precursor concentration was varied and its influence on the QDs properties investigated to effectively control the size and shape of the QDs.

One of the main advantages of this synthesis process, starting from a precursor solution containing indium acetate (IA), myristic acid (MA), and octadecene (ODE), is that the In 2 O 3 semiconductor material is obtained without the addition of oleylamine. Samples were characterized in detail by X-Ray Diffraction (XRD), High-Resolution Transmission Electron Microscopy (HR-TEM), Selected Area Electron Diffraction (SAED), UV-Vis-NIR Spectroscopy, Photoacoustic Spectroscopy (PAS), Photoluminescence Spectroscopy (PLS), and X-Ray Photoelectron Spectroscopy (XPS).

The chemical precursors Indium acetate (IA) (99.99%), Myristic acid (MA) (99%), Tris(trimethylsilyl)phosphine (P(TMS) 3 ) (95%), and Octadecene (ODE) (90%), as well as organic solvents, were purchased from Sigma-Aldrich (Mexico).

The colloidal QDs were synthesized through a one-pot chemical method without injection of hot precursors in a glove chamber filled with N 2 (99.995%; Infra-Mexico), following a process similar to that reported by Li et al. [19] . For the synthesis of InP-In 2 O 3 QDs, the P(TMS) 3 precursor concentration was varied in order to modify the phosphorus content in the samples. The other experimental conditions remained constant, as indicated in Table 1 . These samples were labeled as S1, S2, S3, and S4.

The synthesized colloidal InP-In 2 O 3 QDs were purified by washing them by centrifugation with a mixture of ethanol and acetone using a QDs/acetone/ethanol volume ratio of 50 / 25 / 25, respectively, in an Eppendorf 5424 centrifuge at 15,000 rpm (12,000  g) for 30 min at room temperature (RT). Samples were then dispersed in chloroform and stored at RT.

The crystallinity and structure of the samples deposited on Corning glass and silicon substrates were determined by XRD analysis with a Rigaku SmartLab diffractometer (Rigaku Americas Corporation, Woodlands, USA) For the X-ray Photoelectron Spectroscopy study, the colloidal samples were dropped on a glass substrate. Then, the QDs/glass substrate system was immediately placed into the XPS equipment and degassed at a pressure close to 1  10 -8 Torr for 48 h to avoid oxidation. The surface chemical analysis was determined with a Thermo Scientific K-Alpha System (Thermo Fisher Scientific, UK) using a monochromatic AlKα X-ray radiation source having an energy of 1,486.6 eV. The spot size was set to 400 µm. Survey spectra were collected over a range of -10 to 1,350 eV, a pass energy step of 200 eV, and a step size of 1 eV. The high resolution (HR) XPS spectra were obtained with a pass energy and a step size of 50 eV and 0.1 eV, respectively.

The core-level photoelectron lines were calibrated using the adventitious hydrocarbon signal (C 1s = 284.6 eV) [20] .

Photographs of the four types of colloidal QDs are presented in Fig. 1 . The QDs were subsequently exposed to (a) white light and (b) UV irradiation in order to observe a color change caused by the varying QD size produced by the change in the P(TMS) 3 concentration used during the synthesis.

The structural characterization of the colloidal QDs was carried out by means of HR-TEM (Fig. 2) . The images group, InP nanocrystals were demonstrated to grow randomly oriented [16] . In a similar work, Singh and coauthors synthesized small colloidal InP-based quantum dots with blue emission (sizes ranging from 2 to 5 nm), using trioctylphosphine as a source of phosphorus [21] . Additionally, Protière et al. fabricated InP-In 2 O 3 and In 2 O 3 nanoparticles with sizes of 7.2 and 10 nm, respectively [17] . Meanwhile, Li and coworkers reported two different types of InP samples, with mean diameters of 3.0 and 6.4 nm, attributing the difference in size to the experimental conditions used in the synthesis, e.g., the reaction temperature and the concentration of precursors [19] .

InP-In 2 O 3 quantum dots were deposited on glass and silicon substrates by means of a drip system to obtain the X-ray diffractograms depicted in Fig. 3 . The diffraction patterns simultaneously exhibit peaks corresponding to crystalline InP and In 2 O 3 , indicating the coexistence of these two phases. As explained before, the presence of indium oxide could be attributed to the synthesis conditions at temperatures above 280 °C, where oxygen interacts with indium myristate to produce In 2 O 3 [17] . Displayed in Fig. 3a are the X-ray diffraction patterns of InP-In 2 O 3 QDs deposited on glass substrates. The peaks at 30.4º and 51.6º are respectively assigned to the (200) and (211) In every spectrum, the peak at 26.6º corresponds to the (111) plane of InP, with an interplanar distance of 3.38 Å, in agreement with the SAED and HR-TEM analyses. Singh et al. also found that the peak position matches the zinc blende structure of InP [21] . Additionally, these authors observed that the peaks can be indexed as originated from the cubic zinc-blende structure of InP [19] . Protière et al. [17] attained a similar outcome by X-ray diffraction, showing the characteristic peaks of InP and In 2 O 3 as well (see Figs. 3a and 3b).

The optical absorption spectra of the InP QDs were measured by UV-Vis spectrophotometry and photoacoustic spectroscopy (PAS). PAS is a powerful photothermal technique used in the study of a wide range of samples ranging from opaque to highly transparent materials present in any of the three fundamental states of matter. In the particular case of solid materials, these are placed in a hermetically closed cell filled with air, commonly known as photoacoustic (PA) cell. The sample is then irradiated with a modulated monochromatic light beam in an appropriate spectral range, typically from the UV to the NIR regions. Upon light absorption, a portion of the energy transferred to the sample is converted into heat by non-radiative deexcitation processes, generating acoustic waves in the PA cell that are detected by a microphone [22] .

The band-gap energy (E g ) of the nanoparticles was determined from the equation (Ah) n = B(h -E g ), where h is the incident photon energy, B is a constant, and A can be referred to as the optical absorption coefficient,  or as the photoacoustic amplitude, PA amp , which in most important cases is proportional to  [23] . Depending on the electronic transition, n takes different values, being n = 2 for direct band-gap semiconductors, such as InP [16] . The optical absorption spectra for the InP QDs, accomplished by UV-Vis spectrophotometry, are disclosed in Fig. 4 . It is noted that the fundamental absorption edge presents a red shift as the P(TMS) 3 concentration increases, an effect associated with the QD size reduction and related quantum confinement effects [24] . For the low-energy regime, the E g values for the S1-S4 QDs were calculated as 1.99, 2.23, 2.29, and 1.89 eV, respectively, as shown in the inset (a) of Fig. 4 . On the other hand, for a high-energy regime, the band-gap energy values were 2.92, 2.48, 2.42, and 2.31 eV for the same samples (see inset (b) of Fig. 4) . These values agree with data reported by other authors [25] . For comparison, the absorption spectra for the InP samples, obtained by PAS, are illustrated in Fig. 5 . The S1 sample exhibited the fundamental absorption edge at shorter wavelengths. Accordingly, the other spectra were red-shifted, reaching a maximum for the S3 sample.

In this case, for the high-energy regime, the E g values were found to be 2.92, 2.47, 2.59, and 2.54 eV for the S1-S4 samples, respectively. For the low-energy regime, the band-gap energy was 2.18, 2.10, 2.12, and 2.16 eV for the S1-S4 samples. These results are in the same interval as those observed by UV-Vis spectrophotometry. In addition, the QDs size can be determined by using the following equation for E g : Table 2 . The band-gap energy values determined at the two energy regimes are related to In 2 O 3 -InP QDs (high-energy regime) and InP QDs (low-energy regime). The QDs sizes are between 10 and 13 nm for InP and 8 to 11 nm for InP-In 2 O 3 . These values are in good agreement with those found by other researchers [17] [18] [19] 21, 26] . In general, these data are slightly higher than those determined by HR-TEM.

Shown in Fig. 6 are the PL spectra for the InP-In 2 O 3 QDs at room temperature. As previously described, P(TMS) 3 is the phosphorus source for the QDs. The effect of the P(TMS) 3 concentration during the synthesis of nanocrystals is observed. Indeed, by performing a peak-fitting analysis with Gaussian peaks, the spectra are well-resolved in up to three components centered at 2.1 eV (low-energy band) and in the region between 2.55 and 2.95 eV (high-energy bands). The PL spectrum for the S1 sample is dominated by the emission of a highenergy band placed at around 2.9 eV, attributed to the PL emission of the In 2 O 3 QDs. For this sample, E g is estimated at 2.92 eV [26] [27] [28] [29] , an outcome caused by the oxidation of InP QDs.

A second, wide component at 2.5 eV is related to the recombination of conduction-band electrons to surface states in the QDs [30] [31] [32] [33] . Additionally, it is observed that the PL spectra for the samples S2, S3 and S4 are governed by the emission of low-energy bands centered at around 2.1 eV which we associate to the InP QDs.

These bands are asymmetric and are composed by two bands whose intensity changes when the concentration of phosphine increases during the synthesis of the InP-In 2 O 3 QDs; such change in the PL intensity of these two bands is responsible for the color observed when these samples are illuminated, as shown in Fig. 1 . Comparing the PL spectra of samples S1 and S2, we observe that the high-energy emission band due to the In 2 O 3 QDs decreases strongly compared with the PL band due to the InP QDs, which is an indication of a decrease of the oxidation process. These PL bands are associated with the recombination of conduction band electrons to states located on the InP QDs surface owing to the presence of structural defects [30] [31] [32] [33] . The PL spectra of the S1-S4 samples are compared in Fig. S3 . It is confirmed that the PL intensity progressively changes as the P(TMS) 3 concentration varies. Accordingly, the intensity of the high-energy bands for the S1 and S2 samples diminish, while the low-energy bands for the S1-S4 samples increase at higher phosphorus content. The energy position, intensity, and FWHM of the PL bands for the InP-In 2 O 3 QDs are presented in Table 3 .

The surface chemistry of the InP-In 2 O 3 quantum dots was evaluated by X-ray photoelectron spectroscopy.

Deconvolution methods for the core-level peaks were done imposing a minimal quantity of constraints in order to avoid influence, linked to presumption knowledge, on the obtained results. Such constraints, applied to the XPS doublets, include the peak area ratio, the peak-to-peak spin-orbit splitting, and the peak Full-Width at Half-Maximum (FWHM), within experimental errors [34] . These criteria are governed by quantum mechanics fundamentals. Specifically, the peak area ratio is connected to the degree of degeneracy of a quantum state,

where j is the total angular momentum operator. In turn, j is constructed by the vector addition of the orbital angular momentum, l, and the intrinsic (or spin) angular momentum, s. As a consequence, the s-subshells (with l = 0) do not exhibit spin-orbit splitting, while the p-, d-, and f-subshells, with l = 1, 2, and 3, respectively, are characterized by doublet peaks. In addition, for the p-, d-, and f-subshells, the relative intensities of the doublets fulfill 1/2, 2/3, and 3/4, respectively [35] . Depicted in Fig. 7 are the high-resolution XPS spectra of the In 3d, P 2p, and O 1s states for the S1 sample. Fig. 7a indicates that the In 3d doublet is composed of the 3d 5/2 and 3d 3/2 lines at 444.87 and 452.40 eV, respectively. The area ratio between the In 3d 3/2 and In 3d 5/2 peaks is 0.66, while the energy difference, In 3d = In 3d 3/2 -In 3d 5/2 , is 7.53 eV, corresponding to InP. This result is very similar to that reported in our previous work for InP@ZnS QDs [16] . The P 2p state shows two doublets, revealing that two different chemical environments are formed (Fig. 7b) . On the one hand, the low-intensity peak, recorded at the binding energy (BE) range 128-130 eV, is related to P 3ions of InP [36] . After peakfitting, two components are found at 128.46 eV (P 2p 3/2 ) and 129.36 eV (P 2p 1/2 ), implying P 2p = 0.9 eV and an area ratio of 0.50. On the other hand, the peak displayed at 131-135 eV is deconvolved into two components positioned at 132.91 and 133.81 eV, identified with the P 2p 3/2 and P 2p 1/2 signals, respectively. Here, the values for P 2p and area ratio are 0.9 eV and 0.51, data characteristic of InPO x (P 3+-5+ ) [37] .

In spite of the experimental conditions achieved in a globe-box filled with N 2 , it has been reported that, during the synthesis, some chemical reactions produce in situ oxidative conditions through the generation of small amounts of water [38] . This can lead to the formation of several species including In 2 O 3 , InO x , In(OH) 3 ·nH 2 O, InPO x , and oxygen from carboxylate groups [37] . To shed some light on this, and deepen in the surface chemistry of the InP QDs, we analyzed the high-resolution O 1s photoelectron line. As seen in Fig. 7c although InP native oxides (InPO x ) were not observed. This means that the InPO x species are in an amorphous phase (assuming that their content is within the detection limits of the apparatus). It has been claimed that InPO x could improve the luminescent properties of the InP QDs, while In 2 O 3 could not only enhance the InP quantum yield but also serve as an effective encapsulating shell, protecting InP against photooxidation and suppressing its photodegradation [23, 39] . However, in our case there is no evidence of the formation of core-shell nanoparticles. As shown in HR-TEM images, InP and In 2 O 3 form rather compounds. Listed in Table 4 are the XPS binding energies of the C 1s, O 1s, In 3d, and P 2p states for the S1 -S4 samples. In addition, the surface elemental composition of the In 2 O 3 -InP QDs is also included. These data indicate that the At.% ratio between

In 3d and P 2p is close to 1, except for the S1 sample, in which case a higher amount of indium is revealed.

Carbon is detected at high contents due to both the synthesis and the subsequent exposition of samples to air.

In order to extend the analysis of indium oxides, it is important to recall that when metal-oxygen bonds are formed, oxygen can receive electrons from the metal valence band, thus modifying the electronic structure of inner shells of anions and cations. This effect can be investigated by XPS, which allows the determination of a parameter called chemical bonding interaction (CBI). It has been claimed that the displacement of the valence electron density away from the nucleus of the metal ion results in the reduction of the electronic screening of the inner shells; at the same time, the electronic screening of the inner shells of oxygen increases because of the capture of valence electrons from the metal [40] . These two simultaneous effects result in a reduced BE for oxygen and an increased BE for indium [40] . In addition, the CBI parameter can be used instead of the BE value to avoid BE deviations attributed to several factors, including charging effects and calibration procedures [41] . The CBI between indium and oxygen is defined as (O-In) = O 1s -In 3d 5/2 . Furthermore, considering that every sample exhibits three oxidized environments, three CBIs must be taken into account, i.e., 85.23 eV, respectively. To the best of our knowledge, the CBI has been reported only for a few Ti-, Zn-, Mn-, Mo, and W-based compounds [40] [41] [42] [43] [44] , but for indium oxides is here reported for the first time.

Finally, a complementary analysis of the XPS valence band spectra was obtained in order to analyze the electronic structure of the InP-In 2 O 3 QDs. Presented in Fig. S4 (see the Supporting Information) are the valence band spectra for the S1, S2, S3, and S4 samples. The valence band maximum (VBM) was determined by a linear extrapolation of the low binding energy of the valence band spectrum. The S1 samples have the lowest [45] . However, in our case there is no such an apparent correspondence between the VBM energy and the QDs size.

In this work, InP-In 2 O 3 QDs were successfully synthesized using a single-step chemical method without injection of hot precursors (one-pot). As determined by HR-TEM, the influence of the P(TMS) 3 

In

The effect of ligand chain length on the optical properties of alloyed core-shell InPZnS/ZnS quantum dots

Phosphine-free synthesis of metal chalcogenide quantum dots by directly dissolving chalcogen dioxides in alkylthiol as the precursor

Enhanced luminescence efficiency of wet chemical route synthesized InP-based quantum dots by a novel method: Probing the humidity sensing properties

Stability of quantum dots, quantum dot films, and quantum dot light-emitting diodes for display applications

Understanding Antibody Testing for COVID-19

Nanobiomaterials. Nanostructured Materials for Biomedical Applications

Small-size effects on electron transfer in P3HT/InP quantum dots

Förster electronic excitation energy transfer upon adsorption of meso-tetra (3-pyridyl)porphyrin on InP@ZnS colloidal quantum dots

Controllable synthesis of dual emissive Ag:InP/ZnS quantum dots with high fluorescence quantum yield

Dual-emissive and color-tunable

Mn-doped InP/ZnS quantum dots via a growth-doping method

Enhanced tunable dual emission of Cu:InP/ZnS quantum dots enabled by introducing Ag ions

Color-tunable optical properties of cadmium-free transition metal ions doped InP/ZnS quantum dots

Synthesis of highly fluorescent InP/ZnS small-core/thick-shell tetrahedral-shaped quantum dots for blue light-emitting diodes

Solvothermal synthesis of InP quantum dots and their enhanced luminescent efficiency by post-synthetic treatments

One-pot synthesis of highly luminescent InP/ZnS nanocrystals without precursor injection

Effect of the indium myristate precursor concentration on the structural, optical, chemical surface, and electronic properties of InP quantum dots passivated with ZnS

Amine-induced growth of an In 2 O 3 shell on colloidal InP nanocrystals

Photostability enhancement of InP/ZnS quantum dots enabled by In 2 O 3 overcoating

Economic synthesis of high quality InP nanocrystals using calcium phosphide as the phosphorus precursor

Sol-gel synthesis of Ag-loaded TiO 2 -ZnO thin films with enhanced photocatalytic activity

Tapping the potential of trioctylphosphine (TOP) in the realization of highly luminescent blue-emitting colloidal indium phosphide (InP) quantum dots

Kinetic study of d-Ala induced porphyrins in mice using photoacoustic and fluorescence spectroscopies

Low-temperature photoluminescence spectra of CdO-In 2 O 3 thin films prepared by sol-gel

Cation exchange induced transformation of InP magic-sized clusters

Aminophosphines: a double role in the synthesis of colloidal indium phosphide quantum dots

Surface activation of colloidal indium phosphide nanocrystals

InP/ZnS nanocrystals: Coupling NMR and XPS for fine surface and interface description

Highly photoluminescent InP based core alloy shell QDs from air-stable precursors: excitation wavelength dependent photoluminescence quantum yield, photoluminescence decay dynamics, and single particle blinking dynamics

One-pot synthesis and characterization of InP/ZnSe semiconductor nanocrystals

Structural and optical characterization of wurtzite type ZnS

Electronic properties of In 2 O 3 surfaces

Significant increase in band gap and emission efficiency of In 2 O 3 quantum dots by size-tuning around 1 nm in supermicroporous silicas

Preparation and optical properties of highly crystalline, colloidal, and size-controlled indium oxide nanoparticles

Growth mechanism and physical properties of the type-I In 0.145 Ga 0.855 As y Sb 1-y /GaSb alloys with low As content for near infrared applications

XPS Instrumentation, in X-Ray Photoelectron Spectroscopy: An Introduction to Principles and Practices

Beneficial effects of water in the colloidal synthesis of

InP/ZnS core-shell quantum dots for optoelectronic applications

Effect of zinc incorporation on the performance of red light emitting InP core nanocrystals

InP/ZnS nanocrystals: Coupling NMR and XPS for fine surface and interface description

Photostability enhancement of InP/ZnS quantum dots enabled by In 2 O 3 overcoating

Particle size and structural control of ZnWO 4 nanocrystals via Sn 2+ doping for tunable optical and visible photocatalytic properties

Ti 2p and O 1s core levels and chemical bonding in titanium-bearing oxides

Synthesis, structural, magnetic, and electronic properties of cubic CsMnMoO 3 F 3 oxyfluoride

Surface chemistry of TiO 2 -ZnO thin films doped with Ag. Its role on the photocatalytic degradation of methylene blue

Effect of a CdSe layer on the thermoand photochromic properties of MoO 3 thin films deposited by physical vapor deposition

Electronic and surface properties of PbS nanoparticles exhibiting efficient multiple exciton generation

QDs synthesis, investigation, analysis, and writing

Supervision, writing, and reviewing. A. Cruz-Orea: Investigation. P. Rodríguez-Fragoso: Investigation. J.L. Herrera-Pérez: Investigation. M.L. Gómez Herrera: Investigation. S.A. Tomás: Investigation, writing, and reviewing. F. Vázquez-Hernández: Investigation. Angel A. Durán-Ledezma: Investigation

This research project was partially supported by CONACyT (Projects Nos. 240908 and 205733) and CINVESTAV-IPN. The authors thank the Center for Nanosciences and Micro and Nanotechnologies of the IPN. We also thank Marcela Guerrero and Angel Guillén for their technical assistance. We acknowledge Prof.Jaime Santoyo-Salazar and Prof. Daniel Bahena Uribe for enlightening discussions.