Unraveling The Origin of Enhanced Field Emission from Irradiated

Jan 26, 2016 - Material Research Group, IUAC, New Delhi 110067, India. ¶. Department of Electrical Engineering and Information Technology, Technische...
0 downloads 0 Views 4MB Size
Download PDF
Research Article www.acsami.org

Unraveling The Origin of Enhanced Field Emission from Irradiated FeCo-SiO2 Nanocomposites: A Combined Experimental and FirstPrinciples Based Study Debalaya Sarker,*,† Saswata Bhattacharya,† Raul D. Rodriguez,‡ Evgeniya Sheremet,‡ D. Kabiraj,⊥ D. K. Avasthi,⊥ Dietrich R.T. Zahn,‡ H. Schmidt,¶ P. Srivastava,† and S. Ghosh*,† †

Department of Physics, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India Institute of Physics, Technische Universität Chemnitz, Chemnitz 09107, Germany ⊥ Material Research Group, IUAC, New Delhi 110067, India ¶ Department of Electrical Engineering and Information Technology, Technische Universität Chemnitz, Chemnitz 09126, Germany ‡

ABSTRACT: This work is driven by the vision of engineering planar field emitters with ferromagnetic metal−insulator nanocomposite thin films, using swift heavy ion (SHI) irradiation method. FeCo nanoparticles inside SiO2 matrix, when subjected to SHI get elongated. Using this, we demonstrate here a planar field emitter with maximum current density of 550 μA/cm2 at an applied field of 15 V/ μm. The film, irradiated with 5 × 1013 ions/cm2 fluence (5e13) of 120 MeV Au9+ ions, shows very high electron emitting quantum efficiency in comparison to its unirradiated counterpart. Surface enhanced Raman spectroscopy analysis of unirradiated and 5e13 films further confirms that the field emission (FE) enhancement is not only due to surface protrusions but also depends on the properties of entire matrix. We find experimental evidence of enhanced valence band density of states (VB DOS) for 5e13 film from XPS, which is verified in the electronic structure of a model FeCo cluster from first-principles based calculations combining density functional theory (DFT) and molecular dynamics (MD) simulations. The MD temperature is selected from the lattice temperature profile inside nanoparticles as deduced from thermal spike model. Increasing the irradiation fluence beyond 5e13, results in reduced VB DOS and melting of surface protrusions, thus causing reduction of FE current density. We finally conclude from theoretical analysis that change in fluence alters the coordination chemistry followed by the charge distribution and spin alignment, which influence the VB DOS and concurrent FE as evident from our experiment. KEYWORDS: field emission, DFT, planar field emitter, nano particle, SERS, MD, FeCo-SiO2, elongation, DOS

■

INTRODUCTION

heavy ions (SHI) irradiation can be used as one of the tools to monitor the properties of NPs. Heavy and highly energetic (hundred of MeV) ions transfer their energy in a solid by inelastic collisions with the target atoms. This leaves the target atoms in an excited state; giving rise to a transient temperature hike, known as thermal-spike.17 Because of high thermal conductivities in metals, thermal spikes can be generated in metal NPs if and only if, they are surrounded by some insulating medium. The interface of metal−insulator restricts the heat conduction and causes melting of NPs inside the insulating matrix. Thus, thermal spikes may result in elongation or dissolution of embedded NPs,18,19 phase change,20 alteration in easy axis of magnetization,21 creation of structural defects etc. Till date substantial amount of research works have been attempted to understand the electronic and structural changes in postirradiated Au,22

Metal−insulator nanocomposites have applications in different electronic, optoelectronic, and magnetic nanodevices,1−3 such as field emitters,4,5 sensors,6 solar cells,7 laser diodes,8 magnetic storage media,9 etc. For the application of different surface structures in transmitters and displays, high temperature or high electric fields are necessary for thermionic10 or cold electron emission.11 To gain local field enhancement by applying low electric field from relatively cheap and stable planar field emitters is a challenging issue in comparison to surface structures with sharp protrusions. Recently alongside of us, a few groups have reported field emission (FE) from planar metal−insulator nanocomposites.12−14 In these heterostructures, metal nanoparticles (NPs) and insulator matrix have disparity in their electrical conductivities.15 Therefore, such structures favor electron emission from the surface under the influence of applied field instead of current in the planar path. Distribution and size of embedded or exposed metal NPs play a vital role in the FE properties of these nanocomposites.16 Swift © XXXX American Chemical Society

Received: August 27, 2015 Accepted: January 26, 2016

A

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces Pt,23 Ni,24 Co,25 etc. NPs inside SiO2 matrix. However, the interplaying roles of electronic spin and charge responsible to govern various macroscopic properties (e.g., FE, etc.) of irradiated ferromagnetic nanocomposites are still unclear. In past, FE characteristics of diamond films were tuned with low energy Pt ion implantation.26 In another study, 40 keV C ion irradiation was utilized for the improvement of FE current density from carbon nanotubes.27 But the high energy (i.e., SHI) regime of irradiation is not much explored for FE applications. We have recently observed enhancement in the FE properties of SHI irradiated Ni-SiO2 nanocomposites.16 However, this observation raised several open questions, namely, (i) the conduction mechanism of electrons through the insulating film, (ii) optimal contribution of surface roughness vs elongation of the NPs in the FE, (iii) the role of electronic charge and spin on enhanced valence band density of states (VB-DOS) after irradiation of ferromagnetic nanocomposites. In this Research Article, we aim at getting further insights, from our combined experimental and theoretical studies, into the conduction mechanism and study the role of electronic charge and spin on the emission characteristics of FeCo NPs in SiO2 matrix. Shape anisotropy of irradiated NPs is well studied in literature by Grazing incidence small-angle X-ray scattering (GISAXS).25 SHI causes elongation of the NPs, that modifies the conduction properties of the NPs along different axes. We have employed in-plane/out-of-plane I−V measurements to address the conduction mechanism of electrons through the insulating film. What role does the surface roughness play in the FE enhancement alongside the film’s conductivity needs to be clarified. Surface-enhanced Raman scattering (SERS) amplifies weak Raman signals from molecules, allowing chemical identification of small amounts of material down to the single-molecule detection level.28 Remarkable signal enhancement achieved by SERS is attributed to two factors, namely, (a) the enhancement of electric field by localized surface plasmons and (b) lightning rod effect. The lightning rod effect is a purely geometrical term and depends on the surface roughness.29 It is advantageous to study SERS on non-noble-metal substrates, (e.g., semiconductors30 and transition metals), as tuning of substrate properties such as band gap and surface morphology will affect the Raman signal.29 SERS enhancement for SiO2 with FeCo NPs without any surface treatments is expected to be very low as neither SiO2 nor Fe/Co are plasmonically active. But electric field enhancement at the surface protrusion tips of the irradiated FeCo-SiO2 nanocomposite, as in case for FE, is likely to provide a SERS enhancement due to lightning rod effect. The optimal correlation between the FE enhancement and spectral features in SERS from irradiated FeCo-SiO2 nanocomposite is established in the context of SHI induced surface roughening. The electron emission under high applied electric field depends on the electronic density near the top of the valence band. X-ray photo electron spectroscopy (XPS) measurement can give a rough idea of the same. However, to get into the origin of the FE mechanism down to atomic level of ferromagnetic FeCo NPs, the explicit role of electronic charge as well as spin must be studied. By performing first-principles based density functional theory (DFT) calculation and combining it with thermal spike model followed by molecular dynamics simulations, one can accurately address the underlying phenomena. Here, combining all these state-of-the-art

experimental and theoretical methods, we have engineered ferromagnetic planar field emitters and explored the origin of high electron quantum emission properties.

■

EXPERIMENTAL DETAILS

Fe and Co foils were cosputtered along with a SiO2 target in a high vacuum chamber by fast atom beam (FAB) sputtering technique with Ar beam for depositing FeCo-SiO2 nanogranular films on Si substrate. The atom source is mounted at an angle of 45° facing toward the sputtering target. The substrate holder is mounted on a motor for uniform deposition of the films. The area of the beam is approximately 50 mm diameter at the target, when the source is placed at a distance of 10 cm. The relative area of the quartz plate and the metal pieces exposed to the atom beam determines the amount of metal fraction (here 20%) to be doped. More technical details of experimental FAB set up can be found in ref 31. The stabilization of FeCo phase in the nanocomposite was further achieved by 2 h of annealing in H2 atmosphere in a tubular furnace at 600 °C.32,33 The as-deposited and annealed samples were then subjected to 5.0 × 1013 (5e13) and 1.0 × 1014 (1e14) ions/cm2 fluences of 120 MeV Au9+ (SHI) irradiation at the 15UD Palletron accelerator of IUAC, New Delhi, India. Following this, FE current was recorded with a Keithley multimeter (196 system DMM, resolution 1 nA) using the film as cathode in simple diode geometry kept in a high vacuum chamber. The chamber pressure was kept at 1.5 × 10−6 Torr by means of a turbo-molecular pump. The stainless steel anode was kept at a fixed distance of 200 μm from the film surface in all measurements to obtain a steady field emission. Applied electric field was varied from (1.75 ± 0.05) to (15.00 ± 0.05) V/μm with high voltage DC power supply (Aplab, H5KO2N). Surface morphology and distribution of conducting particles were studied by atomic force microscopy (AFM) and conducting atomic force microscopy (CAFM). The root-mean-square (rms) surface roughness was calculated using NANOSCOPE software from AFM images taken with Dimension Icon Model, Bruker. A bias of 1 V was applied between the tip and sample for CAFM measurements. Crosssectional TEM (XTEM) measurements were carried out to visualize the shape modification of the NPs before and after irradiation. TEM samples were prepared following conventional procedure in crosssectional geometry and images were recorded by means of FEI TITAN 80−300 microscope operating at accelerating voltage of 300 kV. Next, Kelvin probe measurements were carried out to characterize the change in the work function in different films (KP Technology, Model KP020) using gold probe as reference. The work functions were calculated from the mean relative work functions obtained experimentally, considering 5.1 eV as the work function of gold. Out-of-plane and in-plane I−V data were obtained with a KEITHLEY 2400 source meter using gold contacts made at the top of the films and the bottom of the substrate, respectively. All the I−V measurements were carried out at room temperature (300 K) to make the experimental conditions similar to FE measurements (which were also performed at room temperature). For the SERS experiment the above-mentioned samples were cleaned with a stream of water, ethanol, and acetone and then dried under nitrogen flow. Afterward, a 2 nm thick film of cobalt phthalocyanine (CoPc) was simultaneously deposited on all samples, including a reference silicon sample, by organic molecular beam deposition under high vacuum conditions. Raman spectra were acquired at several positions on the samples to ensure reproducibility and homogeneity of the results using a Raman spectrometer (LabRam HR800 Horiba JY France). A solid state laser (HeNe, λ = 632.8 nm) focused with a 100x objective (N.A. 0.9) was used for excitation. The laser power at the sample was 1 mW. Temperature dependent Raman spectroscopy experiments were performed on a CoPc microcrystal placed on a glass slide and heated by a temperature-controlled stage (Linkam) from room temperature to 573 K. GISAXS measurements were carried out at the P03“MiNaXS” beamline of the PETRA III storage ring of DESY, Hamburg, Germany. 11.48 keV X-rays were used at grazing incidence angle of 0.5°. Sample B

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces to detector distance (SDD) was 1850 ± 1 mm. A Pilatus 1 M (DECTRIS, Switzerland) was used as detector with an exposure time of 100 s. The raw data were extracted and integrated in Fit2D software. A horizontal and a vertical “cake” having the origin at the beam stop, with angular spread 10°, were selected separately for extracting the integrated intensity due to scattering from minor and major directions, respectively.33 The quantitative size distribution analysis for minor (parallel to film surface) and major (perpendicular to film surface) dimensions were carried out with Irena2 (version 2.48) software using log-normal size distribution for spheroid particles in a dilute system. XPS valence band spectra were recorded using Mg Kα source (SPECS, Company) with 1253.6 eV energy.

■

the hybrid functional level will be attempted with several hundreds of atoms consisting SiO2 matrix outside the FeCo NPs. The global minimum structure of (FeCo)8 is found from a thorough scanning of the potential energy surface [PES] using cascade genetic algorithm.37,38 We have used FHI-aims code for our DFT calculations, which is an all electron code using numerical atom centered basis set.39 The exchange and correlation functional is taken from generalized gradient approximations (GGA) as in PBE implementation40 which is duly validated with more advanced HSE06 hybrid functionals.41 The (FeCo)8 cluster is then heated to three different temperatures [viz., T = 800, 4000, and 6000 K] using molecular dynamics (MD) simulation. At each temperature the system was kept for 8 ps MD using Nose− Hoover thermostat and following that it was cooled to a low temperature [T = 300 K] via 4 ps MD simulation. Thus, a fast quenching was introduced in our model system to see the structural change when irradiated at different fluences. To draw analogy with our experiment, we have chosen 800 K for taking account of the annealing temperature for unirradiated film. We then raised the temperature up to 4000 and 6000 K as per our findings from thermal spike model for 6 and 3 nm NP respectively. We assume that this temperatures 4000 and 6000 K correspond to the 5e13 and 1e14 fluences, respectively, as the minor dimensions (6 and 3 nm) match with our GISAXS findings.

COMPUTATIONAL DETAILS

Thermal Spike. In thermal spike model, the energy lost by a heavy ion is passed on to two subsystems: namely, the target electrons and afterward to the lattice through electron−electron and electron− phonon interactions. Thus, a set of coupled partial differential equations describes the heat conduction mechanism: Ce

∂Te = ∇(Ke∇Te) + A( r ̅ , t ) − g (Te − Tl) ∂t

ρl C l

■

(1)

∂Tl = ∇(Kl∇Tl) + g (Te − Tl) ∂t

RESULT AND DISCUSSION Figure 1 shows the FE current density (J) versus applied electric field (E) for unirradiated and irradiated films 5e13 and

(2)

Ce, Cl, Ke, and Kl are the specific heats and thermal conductivities of electronic and lattice subsystems, respectively. ρl is the material density. Te and Tl are the electronic and lattice temperatures. A(r,t) ̅ is the energy transferred to the electrons from heavy ion at a time t and at a distance r from the ion’s path. Ce|SiO2 = 1 J/(cm3·K) and Ke|SiO2 = 2 J/(cm·s·K). We adapted the lattice specific heat and thermal conductivity values of Fe/Co and SiO2 from the works of Wang et al.34 and Kumar et al.24 The expression for A(r,t) ̅ is taken as described by Meftah et al.17

⎛ r2 ⎞ 1 ⎛ t⎞ 1 A( r ̅ , t ) = Se exp⎜− ⎟ exp⎜− 2 ⎟ 2 ⎝ τ ⎠ 4πσ τ ⎝ 4σ ⎠ R

(3)

2

with σ 2 = 4d + D· t , where Rd is the track of a cylinder, which absorbs 65% of incident energy, D is the thermal electron diffusivity of hot electrons, Se is the electronic energy loss, τ is electron-atom relaxation time (∼10−15 s). The electronic specific heat of metals 3 3 follows Ce = 2 kBne at high temperatures (Te > 2 TF ) below that

Figure 1. FE characteristics of unirradiated and irradiated films; insets show the schematic of FE setup.

2π

Ce =

T π2 k n e; 2 B e TF

here, TF =

ℏ2 (3π 2ne) 2mekB

is the Fermi temperature with

1e14.42 The current density of a field emitter is related to the applied field by the Fowler−Nordheim equation43

ℏ, kB, me, and ne being the Planck’s constant, Boltzmann constant, electronic mass, and electronic number density, respectively. At the melting (Tm) or vaporization (Tv) temperatures, there would be a sharp hike in Cl values because of the latent heats of melting (Lm) and vaporization (Lv) respectively. To take this into account, at temperatures between (Tm − ΔTm) and (Tm + ΔTm) we have added a peak-line function in the temperature-dependent expression of (T +ΔT ) Cl(T), such that ∫ (Tmm−ΔTmm)Cl(Tm)dT = Lm. Same was done at Tv. The electronic thermal conductivity is dependent on electronic diffusivity De (value taken from ref 35) by Ke = CeDe. After getting the tentative highest Tl values for different sized NPs, the effect of SHI bombardment was further emphasized down to atomic level by ab initio molecular dynamic simulations. DFT and MD. To further investigate the structural modifications of these ferromagnetic metal NPs due to thermal spike, MD simulation was carried out at various temperatures. Since the role of insulating matrix is to restrict the heat flow inside metal NPs to cause a sharp rise in temperature inside the same for reaching a molten state, we have confined our calculations to FeCo metal clusters only. However, silicate phases are likely to be present at the interface of NPs and SiO2 matrix.36 In the present study, we have ignored the effect of SiO2−NPs interface mixing and have first modeled a nanocluster of (FeCo)8 without the SiO2 matrix. In the future, more elaborate calculations at

J=

3/2 aE2β 2 × e−bϕ / βE ϕ

(4)

where a= =

e3 = 1.5443410−6 A·eV·V −2 and b 8πhp 2me1/2 8π × = 6.83107 × V·eV −3/2 cm−1 3 ehp

are constants, β is field enhancement factor, and ϕ is the work function of the material. On application of high bias the field electrons tunnel through the barrier between cathode and anode. This tunneling process will essentially depend on the work function of the surface. About 14% reduction in the work function is observed as one goes from unirradiated film to 5e13 irradiated film (as found from Kelvin probe measurements). This again gets increased by ∼3.5% at 1e14 fluence, which consequently reduces J. C

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces A 500 times increase in J value is observed for the 5e13 film compared to the unirradiated film, for which maximum J is 1.3 μA/cm2. Further increase in irradiation fluence in 1e14 has reduced J drastically to 75 μA/cm2. To investigate the origin of FE hike in the postirradiated film, it is important to look into the surface morphology and conducting surface particles by AFM and CAFM as shown in Figure 2. The AFM and CAFM

Figure 3. Measured out-of-plane J−V characteristics of unirradiated and 5e13 films. Insets show the in-plane I−V characteristics of the same. In addition, the top−bottom and top−top contact configurations are sketched.

Grazing incidence small-angle X-ray scattering (GISAXS) analysis, has confirmed that the embedded FeCo NPs become elongated on exposing to SHI beam.33 To crosscheck, we performed in-plane I−V measurements and found reduced inplane conductivity after irradiation (inset in Figure 3). This inhomogeneous conduction supports the elongation of metal NPs by SHI irradiation as we can directly correlate the reduced in-plane conductivity to reduced in-plane NP size and enhanced out-plane conductivity to larger out-plane NP dimension. These elongated metal NPs are therefore the main reason behind the enhanced FE characteristics observed in the irradiated film. To further investigate if the electric field enhancement from the surface microstructure is contributing in the FE enhancement, we have carried out SERS [Figure 4a]. Figure 4b shows Raman spectra of CoPc on unirradiated and 5e13 FeCo films, which are strongly enhanced w.r.t the reference Si substrate. The downshift of all modes and the bandwidth increase [Figure 4b and 4c] can result from the temperature increase due to

Figure 2. AFM and CAFM images of unirradiated (a and b), and irradiated (c and d) films, respectively. For comparison, the CAFM images (b) and (d) are shown with same color contrast evidencing the increase in current for the irradiated sample. Panels (e) and (f) are XTEM images of unirradiated and 5e13 films showing spherical and elongated NPs, respectively.

images show the changes in topography and surface conductivity in unirradiated (Figure 2a and 2b) and irradiated films (Figure 2c and 2d). An interesting observation from the CAFM data in Figure 2d, is that after irradiation only few particles are responsible for the highest current contribution that dominate the intensity scale for the SHI irradiated film. The surface protrusions enhance the local electric field at their tips in the 5e13 film, which have r.m.s roughness (RR) of 3.2 nm. On raising the fluence to 1e14, overheating causes melting of these protrusions, resulting in reduced RR (1.86 nm). The strong CAFM current signal from few particles in 5e13 film points toward the improved out-of-plane conduction path. This has direct correlation with particle elongation as discussed below. Figure 2e and f show XTEM micrographs of unirradiated and 5e13 films, respectively. The discrimination between fully exposed, partially exposed and SiO2-covered spherical NPs in unirradiated film are clearly evident from Figure 2e, while the elongation is noticed in the 5e13 film as in Figure 2f. Figure 3 shows the out-of-plane current density versus voltage plots when the top film surface is connected to positive bias, for both unirradiated and 5e13 films. Note from Figure 3 that ∼60 times more out-of-plane conduction is observed due to the elongation of embedded metal NPs in the irradiated film.

Figure 4. (a) Schematic of SERS sample cross-section. (b, c) Raman spectra of CoPc on Si in comparison with CoPc on unirradiated and 5e13 FeCo−SiO2 films. (d) Temperature dependence of Nα−Cα mode position. D

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces large electric field around FeCo NPs for 5e13 film. This local heating leads to an increased background and the intensity decrease of Raman modes due to partial decomposition of the molecular probe under a strong electric field, as previously reported.44 To verify this hypothesis and to estimate the temperature achieved at the surface, temperature dependent Raman spectroscopy experiments were performed on a CoPc needlelike microcrystal.45 We observed the increase in peak width and downshift of the Raman modes with increasing temperature [Figure 4d]. By assuming that the temperature contribution dominates the peak shifts in Figure 4b, we can estimate the increase in temperature up to 150 °C for CoPc molecules around hotspots on the 5e13 film. Assuming that CoPc is physisorbed on the film surface, in case of FeCo NPs the SERS enhancement can be attributed only to the lightning rod effect rather than to plasmonic effect. Since electric field enhancement strongly depends on the aspect ratio of the particle, we can assume that the aspect ratio was modified by Au ion irradiation. The difference between the Raman and the FE enhancements comes from the fact that SHI affects all the volume of the film, while only the surface contributes to the Raman spectra. From GISAXS analysis as in Figure 5, we have found that the minor dimensions in 5e13 and 1e14 films are ∼6 and ∼3 nm,

Figure 6. Lattice temperature profiles of (a) 6 and (b) 3 nm FeCo NPs embedded in SiO2 matrix.

Figure 5. Reduction in minor dimension size distribution from 5e13 to 1e14 film. Inset shows the 2D detector images of the same indicating the increasing shape anisotropy with irradiation fluence.

respectively.46 So from thermal spike model the lattice temperature (Tl) profiles were calculated just after the passage of ion for FeCo NPs of 6 and 3 nm diameter as shown in Figure 6a and 6b. After ∼1 ps the lattice temperature goes up to ∼4000 and 6000 K, respectively. After getting the tentative highest Tl values for different sized NPs, the effect of SHI bombardment was further emphasized down to atomic level by first-principles based MD simulations. The macro-scale structural and electrical properties are governed by the underlying electronic structure. To dig into the origin of this intensified electron occupancy, it was necessary to study the electronic structure from our theoretical simulation using a (FeCo)8 cluster. From valence shell XPS spectra of Figure 7a, we observe intensified transition metal 3d states near the Fermi level in the 5e13 film. We noticed that (FeCo)8 cluster gets elongated when subjected to rapid heating up to 6000K and subsequent quenching at 300 K [see Figure 7b]. We have then plotted the density of states (DOS) [spin up and spin down together as in Figure 7c] of each of three structures as shown in Figure 7b. The distribution of electronic states in the

Figure 7. (a) Valence band XPS spectra of unirradiated and 5e13 films: vertical arrow points toward the metal 3d states near the Fermi energy. (b) Ball and stick model (Fe, orange; Co, violet) of the structures of the (FeCo)8 clusters at different temperatures. (c) Total density of states including both spin up and spin down contributions together from DFT calculations of (FeCo)8 clusters at T = 600, 4000, and 6000 K, respectively. (d) Spin polarized projected density of states of the Co and Fe atoms contributing the first peak in the occupied level left to EFermi (as shown in panel c) at T = 4000 and 6000 K (see text for details).

occupied region near Fermi level gets improved for the structure quenched from T = 4000 K, while the same gets reduced for the structure quenched from T = 6000 K. All these findings are quite in good agreement with our experimental observations. Therefore, to have a better insight, we have E

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

embedded in SiO2 matrix. We showed that the elongation of FeCo NPs and the roughness of the FeCo nanocomposite surface enhance the FE and SERS properties of irradiated FeCo−SiO2 film. Elongated NPs inside the composite change the electrical conduction path. This reduces the overall Fowler−Nordheim type tunneling barrier faced by an outgoing electron through the FeCo-SiO2 matrix. We further explore that a modified atomic co-ordination changes the charge/spin states near Fermi level and the density of occupied states that directly influence the FE. Because of its high current density and mechanical and chemical durability, the SHI irradiated FeCoSiO2 films open new possibilities for the development of electronic displays.

investigated the Hirshfeld charge state and spin moments (parallel/antiparallel) of the atoms at three different temperatures. Following this analysis, we locate few specific atoms, where notable charge and spin redistribution take place. We then tried to understand the local interaction up to first nearestneighbor. We notice that the changes in local atomic coordination after irradiation is introduced. The structure that is quenched from T = 4000 K, has one extra Fe in the first nearest neighbor, than the one quenched from 800 K. Therefore, Fe, being electropositive, donates its electrons to Co and increases the total occupied electronic states in the DOS as in Figure 7c. As one further increases temperature to 6000 K (i.e., corresponding to the irradiation fluence in 1e14), while the first-nearest neighbor coordination remains the same, we notice significant changes in the spin state of the contributing atoms forming Fe−Co bonds. In Figure 7d, the spin polarized projected density of states of all the contributing Fe and Co atoms responsible for forming Fe−Co bonds are separately shown. It is clear that a significant portion of spins of Co and Fe atoms at T = 4000 K, is getting flipped upon increasing the temperature to 6000 K. This not only reduces the Fe−Co bond distance (because of parallel and antiparallel spin interaction) but also creates charge redistribution in the system. This leads to certain redistribution of electronic charge states and the occupation in VB near EFermi gets also reduced. Figure 8 schematically illustrates the field electron emission via tunneling in unirradiated and irradiated films to have an

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Phone: +91-26591358. Fax: +91-26582037. Author Contributions

S.G., P.S., and D.S. conceived and designed the study; D.S. performed the experiments and analyzed the data; R.D.R. and E.S. performed SERS experiment; D.S. and S.B. performed the density functional theory (DFT) calculations; H.S. provided the I−V facility; D.K.A. and D.K. provided the FAB and SHI facilities; D.S., S.B., R.D.R., and E.S. wrote the paper with input from all authors; S.G., P.S., and H.S. reviewed and edited the manuscript. All authors participated in discussions and read and approved the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the help of Mrs. Illona Skorupa, HZDR, Germany (for sample bonding to perform I−V measurements); employees associated with the XPS facility (partially funded by FIST grant of DST) at IIT Delhi and PETRA III at DESY [a member of the Helmholtz Association (HGF)], Hamburg, Germany, for providing beam time. D.S. acknowledges UGC, India, and BMWi-ZIM project, TUC, Germany (for financial support) during her research stay at TUC. We acknowledge nano research facilities (NRF) at IIT Delhi, India, for AFM/ CAFM measurements and financial support. We acknowledge the financial support from CSIR, India, for FE set up under project RP02686. R.D.R., E.S., and H.S. acknowledge funding sources, namely, DFG Research Unit FOR1713, Cluster of Excellence “Center for Advancing Electronics Dresden”, DFG project ZA146/22-1, and DFG project SCHM1663/4. This work was performed in the context of the European COST Action MP1302 Nanospectroscopy. D.S. acknowledges the advanced imaging centre at IIT Kanpur for providing XTEM facility and Mr. Ambresh Malya for his help during measurements. DS and SB acknowledge the high performance computing (HPC) facility of IIT Delhi for computational resources.

Figure 8. Schematic of electron emission from unirradiated and irradiated FeCo−SiO2 films at the position of a spherical and of an elongated FeCo particle, respectively. Elongation and new defect states near the valence band (VB) cause a narrowing of the SiO2 tunnel barrier.

insight into the overall conduction mechanism. The elongation of FeCo NPs reduces total SiO2 barrier in irradiated film. Following this, the DOS near Fermi level gets modified, which causes an overall reduction in emitter work function by 14%. The valence electrons thus easily tunnel through the vacuum between cathode and anode via Fowler−Nordheim type tunneling.43

■

■

CONCLUSION In summary, we here investigated from our combined experimental and theoretical analysis the micro and macro structural origin of enhanced FE from SHI irradiated FeCo NPs

REFERENCES

(1) Kim, S.-J.; Park, Y.-S.; Lyu, S.-H.; Lee, J.-S. Nonvolatile Nanofloating Gate Memory Devices Based on Pentacene Semiconductors and Organic Tunneling Insulator Layers. Appl. Phys. Lett. 2010, 96, 033302.

F

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (2) Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S. Electronics and Optoelectronics of Two-Dimensional Transition Metal Dichalcogenides. Nat. Nanotechnol. 2012, 7, 699−712. (3) Buravtsova, V. E.; Guschin, V. S.; Kalinin, Y. E.; Kirov, S. A.; Lebedeva, E. V.; Phonghirun, S.; Sitnikov, A. V.; Sy'rev, N. E.; Trofimenko, I. T. Magnetooptical Properties and FMR in Granular Nanocomposites (Co84Nb14Ta2)x(SiO2)100-x. Cent. Eur. J. Phys. 2004, 2, 566−578. (4) Chang, Y.-M.; Liu, M.-C.; Kao, P.-H.; Lin, C.-M.; Lee, H.-Y.; Juang, J.-Y. Field Emission in Vertically Aligned ZnO/Si-Nanopillars with Ultra Low Turn-On Field. ACS Appl. ACS Appl. Mater. Interfaces 2012, 4, 1411−1416. (5) Gu, C.; Jiang, X.; Lu, W.; Li, J.; Mantl, S. Field Electron Emission Based on Resonant Tunneling in Diamond/CoSi2/Si Quantum Well Nanostructures. Sci. Rep. 2012, 2, 699−712. (6) Li, X.; Gao, Y.; Gong, J.; Zhang, L.; Qu, L. Polyaniline/Ag Composite Nanotubes Prepared through UV Rays Irradiation via Fiber Template Approach and their NH3 Gas Sensitivity. J. Phys. Chem. C 2009, 113, 69−73. (7) Kim, A.; Won, Y.; Woo, K.; Kim, C.-H.; Moon, J. Highly Transparent Low Resistance ZnO/Ag Nanowire/ZnO Composite Electrode for Thin Film Solar Cells. ACS Nano 2013, 7, 1081−1091. (8) Li, Y.; Ma, X.; Xu, M.; Xiang, L.; Yang, D. Remarkable Decrease in Threshold for Electrically Pumped Random Ultraviolet Lasing from ZnO Film by Incorporation of Zn2TiO4 Nanoparticles. Opt. Express 2011, 19, 8662−8669. (9) Puig, J.; Hoppe, C. E.; Fasce, L. A.; Perez, C. J.; Pineiro-Redondo, Y.; Banobre-Lopez, M.; Lopez-Quintela, M. A.; Rivas, J.; Williams, R. J. J. Superparamagnetic Nanocomposites Based on the Dispersion of Oleic Acid-Stabilized Magnetite Nanoparticles in a Diglycidylether of Bisphenol A-Based Epoxy Matrix: Magnetic Hyperthermia and Shape Memory. J. Phys. Chem. C 2012, 116, 13421−13428. (10) Shakouri, A.; LaBounty, C.; Piprek, J.; Abraham, P.; Bowers, J. E. Thermionic Emission Cooling in Single Barrier Heterostructures. Appl. Phys. Lett. 1999, 74, 88−89. (11) Milne, W. I.; Teo, K. B. K.; Amaratunga, G. A. J.; Legagneux, P.; Gangloff, L.; Schnell, J.-P.; Semet, V.; Thien Binh, V.; Groening, O. Carbon Nanotubes as Field Emission Sources. J. Mater. Chem. 2004, 14, 933−943. (12) Sarker, D.; Ghosh, S.; Srivastava, P. Spectral Weight Shift in The Valence Band Density of States and Concurrent Increase in Field Emission by Hydrogenation of FeCo-SiO2 Nanocomposites. RSC Adv. 2015, 5, 63377−63381. (13) Rout, C. S.; Joshi, P. D.; Kashid, R. V.; Joag, D. S.; More, M. A.; Simbeck, A. J.; Washington, M.; Nayak, S. K.; Late, D. J. Superior Field Emission Properties of Layered WS2-RGO Nanocomposites. Sci. Rep. 2013, 3, 3282. (14) Zhang, S.; Zhang, Y.; Huang, S.; Liu, H.; Wang, P.; Tian, H. First-Principles Study of Field Emission Properties of Graphene-ZnO Nanocomposite. J. Phys. Chem. C 2010, 114, 19284−19288. (15) Tsang, W.; Stolojan, V.; Giusca, C.; Poa, C.; Sealy, B.; Silva, S.; Wong, S. Electron Field-emission Properties of Ag−SiO2 Nanocomposite Layers. 18th International Vacuum Nanoelectronics Conference; IEEE: Piscataway, NJ, 2006; pp 958−961. (16) Sarker, D.; Kumar, H.; Patra, R.; Kabiraj, D.; Avasthi, D. K.; Vayalil, S. K.; Roth, S. V.; Srivastava, P.; Ghosh, S. Enhancement in Field Emission Current Density of Ni Nanoparticles Embedded in Thin Silica Matrix by Swift Heavy Ion Irradiation. J. Appl. Phys. 2014, 115, 174304. (17) Meftah, A.; Brisard, F.; Costantini, J. M.; Dooryhee, E.; HageAli, M.; Hervieu, M.; Stoquert, J. P.; Studer, F.; Toulemonde, M. Track Formation in SiO2 Quartz and The Thermal-spike Mechanism. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 49, 12457−12463. (18) Giulian, R.; Kremer, F.; Araujo, L. L.; Sprouster, D. J.; Kluth, P.; Fichtner, P. F. P.; Byrne, A. P.; Ridgway, M. C. Shape Transformation of Sn Nanocrystals Induced by Swift Heavy-ion Irradiation and The Necessity of A Molten Ion Track. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82, 113410.

(19) Ridgway, M. C.; Giulian, R.; Sprouster, D. J.; Kluth, P.; Araujo, L. L.; Llewellyn, D. J.; Byrne, A. P.; Kremer, F.; Fichtner, P. F. P.; Rizza, G.; Amekura, H.; Toulemonde, M. Role of Thermodynamics in the Shape Transformation of Embedded Metal Nanoparticles Induced by Swift Heavy-Ion Irradiation. Phys. Rev. Lett. 2011, 106, 095505. (20) Sprouster, D. J.; Giulian, R.; Schnohr, C. S.; Araujo, L. L.; Kluth, P.; Byrne, A. P.; Foran, G. J.; Johannessen, B.; Ridgway, M. C. fcc-hcp Phase Transformation in Co Nanoparticles Induced by Swift Heavyion Irradiation. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 80, 115438. (21) D’Orléans, C.; Stoquert, J. P.; Estournès, C.; Cerruti, C.; Grob, J. J.; Guille, J. L.; Haas, F.; Muller, D.; Richard-Plouet, M. Anisotropy of Co Nanoparticles Induced by Swift Heavy Ions. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 67, 220101. (22) Rizza, G.; Dawi, E. A.; Vredenberg, A. M.; Monnet, I. Ion Engineering of Embedded Nanostructures: From Spherical to Facetted Nanoparticles. Appl. Phys. Lett. 2009, 95, 043105. (23) Giulian, R.; Araujo, L. L.; Kluth, P.; Sprouster, D. J.; Schnohr, C. S.; Byrne, A. P.; Ridgway, M. C. Swift Heavy Ion Irradiation of Pt Nanocrystals: I. Shape Transformation and Dissolution. J. Phys. D: Appl. Phys. 2011, 44, 155401. (24) Kumar, H.; Ghosh, S.; Avasthi, D. K.; Kabiraj, D.; Mucklich, A.; Zhou, S.; Schmidt, H.; Stoquert, J. P. Ion Beam Induced Shaping of Ni Nanoparticles Embedded in A SiO2Matrix. Nanoscale Res. Lett. 2011, 6, 155. (25) Sprouster, D. J.; Giulian, R.; Araujo, L. L.; Kluth, P.; Johannessen, B.; Cookson, D. J.; Ridgway, M. C. Swift Heavy-ion Irradiation-induced Shape and Structural Transformation in Cobalt Nanoparticles. J. Appl. Phys. 2011, 109, 113504. (26) Panda, K.; Sankaran, K. J.; Inami, E.; Sugimoto, Y.; Tai, N. H.; Lin, I.-N. Direct Observation and Mechanism for Enhanced Field Emission Sites in Platinum Ion Implanted/Post-annealed Ultrananocrystalline Diamond Films. Appl. Phys. Lett. 2014, 105, 163109. (27) Sun, P. C.; Deng, J. H.; Cheng, G. A.; Zheng, R. T.; Ping, Z. X. Enhanced Electron Field Emission from Carbon Nanotubes Irradiated by Energetic C Ions. J. Nanosci. Nanotechnol. 2012, 12, 6510−5. (28) Schlucker, S. Surface-Enhanced Raman Spectroscopy: Concepts and Chemical Applications. Angew. Chem., Int. Ed. 2014, 53, 4756− 4795. (29) Liao, P. F.; Wokaun, A. Lightning Rod Effect in Surface Enhanced Raman Scattering. J. Chem. Phys. 1982, 76, 751−752. (30) Lombardi, J. R.; Birke, R. L. Theory of Surface-Enhanced Raman Scattering in Semiconductors. J. Phys. Chem. C 2014, 118, 11120− 11130. (31) Kabiraj, D.; Abhilash, S.; Vanmarcke, L.; Cinausero, N.; Pivin, J.; Avasthi, D. Atom Beam Sputtering Setup for Growth of Metal Particles in Silica. Nucl. Instrum. Methods Phys. Res., Sect. B 2006, 244, 100−104. (32) Kumar, H.; Ghosh, S.; Srivastava, P.; Kabiraj, D.; Avasthi, D. K.; Olivi, L.; Aquilanti, G. Stabilization of FeCo Alloy Phase In FeCoSiO2 Nanocomposites. Adv. Adv. Mater. Lett. 2013, 4, 390−397. (33) Sarker, D.; Ghosh, S.; Kabiraj, D.; Avasthi, D. K.; Iadecola, A.; Vayalil, S. K.; Srivastava, P. Electronic Excitation Induced Structural Modification of FeCo Nanoparticles Embedded in Silica Matrix. Mater. Res. Express 2014, 1, 035017. (34) Wang, Z. G.; Dufour, C. h.; Paumier, E.; Toulemonde, M. The Se Sensetivity of Metals Under Swift Heavy Ion Irradiation. J. Phys.: Condens. Matter 1994, 6, 6733−6750. (35) Dufour, C.; Khomenkov, V.; Rizza, G.; Toulemonde, M. Ionmatter Interaction: The Three-dimensional Version of The Thermal Spike Model. Application to Nanoparticle Irradiation with Swift Heavy Ions. J. Phys. D: Appl. Phys. 2012, 45, 065302. (36) Desautels, R. D.; Rowe, M. P.; Jones, M.; Whallen, A.; van Lierop, J. Spontaneously Formed Interfacial Metal Silicates and Their Effect on the Magnetism of Superparamagnetic FeCo/SiO2 Core/ Shell Nanoparticles. Langmuir 2015, 31, 2879−2884. (37) Bhattacharya, S.; Levchenko, S. V.; Ghiringhelli, L. M.; Scheffler, M. Stability and Metastability of Clusters in a Reactive Atmosphere: Theoretical Evidence for Unexpected Stoichiometries of MgMOx. Phys. Rev. Lett. 2013, 111, 135501. G

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (38) Bhattacharya, S.; Levchenko, S. V.; Ghiringhelli, L. M.; Scheffler, M. Efficient ab initio Schemes for Finding Thermodynamically Stable and Metastable Atomic Structures: Benchmark of Cascade Genetic Algorithms. New J. Phys. 2014, 16, 123016. (39) Blum, V.; Gehrke, R.; Hanke, F.; Havu, P.; Havu, V.; Ren, X.; Reuter, K.; Scheffler, M. Ab initio Molecular Simulations with Numeric Atom-centered Orbitals. Comput. Phys. Commun. 2009, 180, 2175−2196. (40) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple [Phys. Rev. Lett. 77, 3865 (1996)]. Phys. Rev. Lett. 1997, 78, 1396−1396. (41) Heyd, J.; Scuseria, G. E.; Ernzerhof, M. Erratum: Hybrid Functionals Based on A Screened Coulomb Potential. J. Chem. Phys. 2006, 124, 219906−219906. (42) Note that there should be larger drop in effective electrical field strength between the cathode and anode for the unirradiated sample with higher electrical resistance. While drawing the J−E curves, we considered the applied field instead of the actual effective field between the cathode and anode. (43) Fowler, R. H.; Nordheim, L. Electron Emission in Intense Electric Fields. Proc. R. Soc. London, Ser. A 1928, 119, 173−181. (44) Fang, Y.; Seong, N.-H.; Dlott, D. D. Measurement of the Distribution of Site Enhancements in Surface-Enhanced Raman Scattering. Science 2008, 321, 388−392. (45) Szybowicz, M.; Ba̷la, W.; Dumecke, S.; Fabisiak, K.; Paprocki, K.; Drozdowski, M. Temperature and Orientation Study of Cobalt Phthalocyanine CoPc Thin Films Deposited on Silicon Substrate as Studied by Micro-Raman Scattering Spectroscopy. Thin Solid Films 2011, 520, 623−627. (46) Note that the minor dimension values are not the exact peak positions of Figure 5 and are calculated from statistical analysis.33

H

DOI: 10.1021/acsami.5b07937 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX