Screenplays & Play

Raman scattering and interstitial Li defects induced polarization in co-doped multiferroic Zn0.96-yCo0.04LiyO (0.00

Structural and Raman analysis confirmed a single phase wurtzite hexagonal crystalline structure of Li co-doped ZnO nanoparticles. In the Raman backscattering spectra, E Low 2 and E High 2 modes corresponded to zinc and oxygen lattice vibrations. No
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Raman scattering and interstitial Li defects inducedpolarization in co-doped multiferroic Zn 0.96-  y  -Co 0.04 Li  y  O (0.00 #  y  # 0.10) nanoparticles Saif Ullah Awan,* ab S. K. Hasanain, a M. S. Awan c and Saqlain A. Shah d Structural and Raman analysis con fi rmed a single phase wurtzite hexagonal crystalline structure of Li co-doped ZnO nanoparticles. In the Raman backscattering spectra, E Low2  and E High2  modes corresponded tozinc and oxygen lattice vibrations. No extra vibration modes of secondary or metallic phases of Co or Liwere observed. The intensity of Raman modes decreased with increasing Li content. High resolution X-rayphotoelectron spectroscopy (XPS) of Zn and Co con fi rmed the +2 oxidation state. The deconvolution ofhigh resolution XPS spectra of Li showed the presence of interstitial and substitutional Li defects. Theroom temperature polarization lay in the range of 0.155 – 0.225  m C cm  2 for di ff erent compositions. Weobserved an interesting result in the co-doped system, that even the low Li concentration sample,  e.g. ,Zn 0.94 Co 0.04 Li 0.02 O, showed a  P  – E   hysteresis loop of polarization at   0.20  m C cm  2 . There was higherpolarization for  y   ¼  0.10, due to more interstitial Li defects, and lower polarization for  y   ¼  0.06, due tomore substitutional Li defects. We suggest that polarization might have appeared and been enhanced dueto the net interaction of dipoles, formed by Li o ff -center impurities (interstitial Li defects). 1. Introduction Low dimensional Ferroelectric materials have attracted enor-mous attention in memory devices due to their potentialfeatures,  e.g. , non-volatility, non-destructive readout, low-powerconsumption, fast switching speeds, and their ability to operate without an access transistor. These features open new oppor-tunities to develop more e ffi cient and higher density memo-ries. 1 The existing ferroelectric memory devices are mainly based on perovskite materials, which are structurally compli-cated and relatively di ffi cult to synthesize. Previously, ferro-electric behavior has been reported in Pb 1   y Ge  y Te 2 andCd 1   y Zn  y Te 3 systems. Polarization induced in these systemsdue to the ionic radii of the host and dopant atoms di ff  er by 0.047 nm and 0.02 nm in Pb 1   y Ge  y Te and Cd 1   y Zn  y Te, 2,3 respectively. The required condition for the occurrence of polarization is that the substitutional atoms should be of anappreciably di ff  erent size than that of the host atoms they replace. Zinc oxide (ZnO) has a wide band gap of 3.4 eV with ahigh exciton binding energy (60 meV) and remarkable acoustic wave properties. ZnO, due to its signi  cant optical andmagnetic properties (doped systems), 4 should be having aunique worth if it also exhibits ferroelectric properties. ZnO is asophisticated substance that lacks center of symmetry inhexagonal lattice and possesses large electromechanicalcoupling, which results into strong ferroelectric, piezoelectricand pyroelectric properties. 5 ZnO has a wurtzite structure in which the oxygen atoms are arranged in a hexagonal closepacked lattice with zinc atoms occupying half of the tetrahedralsites. 6 The d-electrons of Zn hybridize with p-electrons of oxygen. Both zinc and oxygen atoms are tetrahedrally coordi-natedtoeach other and areat equivalent positions. 5,6 Therefore,ZnO is relatively an openstructure withallof the octahedralandhalf of the tetrahedral sites being empty. So, it is relatively easy to incorporate external dopants in the ZnO lattice. Although,stoichiometric ZnO is an insulator, it usually contains vacanciesfor oxygen ( i.e. , charge carriers). Furthermore, it is well knownthat external doping (impurities) as well as the intrinsic latticedefects greatly in  uence its electrical, structural, electronic,magnetic and optical properties.Inliterature,FerroelectricorderhasbeenreportedinMg, 7 Be, 8 Co, 9 Cr, 10 Cu 11 and V  12 doped ZnO systems. Moreover, the exis-tence of spontaneous polarization in Li-doped ZnO (Zn 1   y Li  y O)  lms, 5,13 nanorods 14 and ceramics 15,16 systems has also beenreported. It is generally believed that ferroelectricity in Zn 1   y Li  y Oarises due to the large di ff  erence in ionic radii of the host Zn 2+ (0.074 nm) and the dopant Li 1+ (0.060 nm) 9,16 ions. According tothese reports, Zn 1   y Li  y O mustbehavingat least8%Litogeneratea ferroelectric hysteresis loop. The ferroelectric behavior of Zn 1   y Li  y O systems motivated us to study nano-materials with a  Department of Physics, Quaid-i-Azam University, Islamabad 45320, Pakistan.; b  Department of Physics, COMSATS Institute of Information Technology (CIIT), Islamabad 44000, Pakistan c  Ibne Sina Institute of Information Technology (ISIT), Sector H-11/4, Islamabad 44000, Pakistan d   Materials Science and Engineering, University of Washington, Seattle, WA 98195, USA Cite this:  RSC Adv. , 2015,  5 , 39828Received 2nd March 2015Accepted 15th April 2015DOI: 10.1039/c5ra03691g 39828  |  RSC Adv. , 2015,  5 , 39828 – 39839 This journal is © The Royal Society of Chemistry 2015 RSC Advances PAPER  co-doping of transition metal elements,  e.g. , 4%Co, to enhancepolarization in Zn 1   y Li  y O based system. The Co concentration was   xed at 0.04 to avoid the possible complications of causing multiphase or Co segregations. 17 The nature of ferroelectricity and ferromagnetism has been observed in Li co-doped ZnO:Cothin   lms, 9,18 but any such investigation on nanoparticles hasnever been reported so far. We studied nanoparticles becauseparticles size plays a vital role in multiple physical properties of materials. Onodera  et al. 16 reported the ferroelectric response of Li 1+ and Mg  2+ co-doped in the ZnO bulk system, and observedthat co-doping Mg  2+ can improve the ferroelectric properties ( i.e. ,polarization). Similarly, co-doping of (Mn 2+ , Li +1 ) nanorods 19 and(Cu 1+ , Li +1 ) thin   lms 20 can induce and enhance the polarizationin these co-doped ZnO systems.The physical nature of the ferroelectric phenomena due tostrong interactions between the electric dipole coupling is not completely understood yet. 21 The understanding of the abovephenomena would be of great signi  cance both for funda-mental research as well as for the development of next gener-ation electronic devices. It is commonly believed that the o ff  -centered Li ions play an important role in the existence of ferroelectricity. In case of Zn 0.96   y Co 0.04 Li  y O, due to largedi ff  erences in ionic radii of host zinc and the co-dopants, Liatoms can occupy o ff  -centered positions, forming permanent electric dipoles, and thereby resulting in ferroelectric behavior.The main question here is what makes the dipoles to be fer-roelectrically ordered? Recently, Ruderman – Kittel – Kasuya – Yosida (RKKY) type indirect interaction have been reported inimpurities dipoles  via  free charge carriers, which was anattempt to understand the nature of ferroelectricity in non-perovskite oxide systems. 22 – 24 There are not so many individual materials showing bothferromagnetic and ferroelectric properties at room tempera-tures. Previously, some e ff  orts have been made on tailoring perovskite structures,  e.g. , BiFeO 3  (ref. 25) and YMnO 3  (ref. 26),for multiferroic behavior. Although these compounds areferroelectric but their magnetic ordering is essentially anti-ferromagnetic. Recently, induction of ferromagnetism in non-magnetic and non-perovskite structured materials has gainedconsiderable attention dueto the inclusion ofspin functionality in the host crystal. 27,28 However, the scarcity of single phasemultiferroic is still a challenge due to two main reasons;ferromagnetism and ferroelectricity are generally mutually exclusive, and ferromagnetism tends to be metallic, whereasferroelectricity coexists with the insulating state. 26,29 Therefore,  nding suitable multiferroic materials is an attractive andfascinating research area. We have reported the ferromagnetic behavior in these Li co-doped ZnO:Co nanoparticles 30 earlier. In this report we aredemonstrating weak polarization e ff  ects in that system. We arepresenting a detailed micro-structural analysis of a few selectedsamples to identify the single phase polycrystalline structures. We performed Raman spectroscopy for the investigation of metallic and secondary issues of Co and Li clustering. X-raysphotoelectron spectroscopic measurements were performed toinvestigate the oxidation states of ionic species. The main focus was to study the role of Li ( i.e. , interstitial or substitutional) as aco-dopant in the presence of Co to achieve polarization in ZnO.Does the increase of the electric moment ( i.e. , polarization)correlate with interstitial or substitutional Li defects? There-fore, we have performed room temperature ferroelectrichysteresis measurements. 2. Experimental details The detailed synthesis procedure of Li co-doped ZnO:Co[Zn 0.96   y Co 0.04 Li  y O (  y  ¼  0.00, 0.02, 0.04, 0.06, 0.08, 0.10)]nanoparticles has been reported previously by our group. 30 Brie   y, Zinc, Cobalt and Lithium acetates were used as starting precursors. The 0.1 M solution was prepared using 200 mlethylene glycol with required amount of chemicals (wt%). Thesolution was stirred for about 30 minutes in conical   ask andthen the suspension was heated and optimized at 180   C forabout 3 h in a re  uxing system. A    er the formation of precipi-tation of the precursors, the mixture was cooled down to roomtemperature. The as-synthesized samples were washed twice with ethanol and distilled water and centrifuged at 4000 rpm toget solid wet phase. This solid wet phase was dried in an oven at 100   C overnight. To achieve the required single phase hexag-onal (wurtzite) structure, these samples were annealed at anoptimum temperature of 600   C in a forming gas (5% H 2  + 95% Ar) for 6 h.Nanostructures were observed using FEI's Titan G260-300CTtransmission electron microscopy (TEM). The bright    eld (BF)images were acquired at around 70 000  as well as at 500 000  ,for low resolution and high resolution TEM analysis, respec-tively. Hough transform was applied to Fast Fourier Transform(FFT) of HRTEM images, to determine the  d  -spacing moreaccurately. 31 The samples were further analyzed with ElectronEnergy Loss Spectroscopy (EELS) using Gatan Inc.'s TridiemTEM energy-  lter (EF) for the identi  cation of elements andmapping. 31 Raman scattering spectroscopy was measured using diode-pumped solid state laser emitting at 532 nm with a powerof 50 mW. X-ray photoelectron spectroscopy (XPS) was per-formed on a Thermo Scienti  c ESCALAB 250-spectrometer witha focused monochromatic Al-K a  ( h n  ¼  1486.6 eV) source.Binding energies of photoelectrons correlated with aliphatichydrocarbon C-1s peak at 285 eV. Ferroelectric hysteresis curves were obtained using a commercial RT-66A ferroelectric systemat room temperature. Powder polycrystalline samples werecompacted uniaxially into circular pellets under identicalconditions for electrical measurements, so as to retain the samedensity over the entire composition range, thereby minimizing the possible variations in the inter-grain resistivity contributions. 3. Experimental observations andanalysis 3.1 Transmission electron microscopy (TEM) Low-magni  cation bright    eld transmission electron micro-scope (BF-TEM) images were acquired for Zn 0.94 Co 0.04 Li 0.02 Onanoparticles as shown in Fig. 1(a). Nanoparticles were This journal is © The Royal Society of Chemistry 2015  RSC Adv. , 2015,  5 , 39828 – 39839 |  39829 Paper RSC Advances  agglomerated as the samples were post annealed to improve thecrystallinity. Size of Zn 0.94 Co 0.04 Li 0.02 O particles was in therange 50 – 300 nm. High resolution transmission electronmicroscope (HR-TEM) image of Zn 0.94 Co 0.04 Li 0.02 O sample isshown in Fig. 1(b) showing homogeneous crystalline structurehaving no amorphous traces. The fast Fourier transform (FFT)of the HRTEM micrograph is shown in Fig. 1(c). HoughTransforms (selected area electron di ff  raction (SAED) patterns) were applied to the FFT to extract the  d  -spacing values as shownin Fig. 1(d). The variation in  d  -values is crucial to study the roleof Li dopant occupying interstitial or substitutional positions inZnO structure. Low resolution BF-TEM micrographs of Zn 0.90 Co 0.04 Li 0.06 O nanoparticles are shown in Fig. 2(a). Parti-cles exhibited size distribution of the range, 20 – 200 nm.HRTEM micrograph (Fig. 2(b)), FFT image of the HRTEM(Fig. 2(c)), and Hough transform (Fig. 2(d)) measurement wereacquired for Zn 0.90 Co 0.04 Li 0.06 O sample. The  d  -values weremeasured from Hough transform.These  d  -values were in good agreement with the X-ray di ff  raction data, as we discussed earlier. 30 From the Houghtransforms, we found that all  d  -values and planes correspondedto ZnO crystalline structure (JCPDS   le no. 36-1451). Hence, weobserved the hexagonal wurtzite (SG  P  63 mc ,  a  ¼  0.32 nm,  c  ¼ 0.52 nm) crystalline structure, which also con  rmed theabsence of any metallic or secondary phases in samples. It canbe noticed that 6% Li co-doped ZnO:Co sample has smaller d  -spacing than 2% Li co-doped sample. This observationsuggests the contraction of the unit cell which is due to thepresence of substitutional Li ions. Decrease in lattice parame-ters is expected when Li substitutes Zn, while the latticeparameter will increase when Li occupies interstitial sites (o ff  -centered positions). 27 Here we consistently observed adecrease in lattice parameters for  y ¼ 0.02 and 0.06 using TEM, which was in agreement with XRD data. 30 Energy dispersive spectroscopy (EDS) was acquired in TEMmode which showed only Zn, Co and O signals and there wereno Li peak (EDS data not shown here). Since Beryllium window did not detect Li signals, so we performed electron energy loss(EELS) spectroscopy for the con  rmation of Li in our samples. We observed the O-K, Co-L 23  and Zn-L 23  edge in full scan range(400 – 1400 eV) EELS spectrum for Zn 0.90 Co 0.04 Li 0.06 O nano-particles as is shown Fig. 3(a). In the inset of Fig. 3(a), Co-L 23 and Zn-L 23  edges are clearly visible. Fig. 3(b) shows the EELSspectra for scanning range 50 – 220 eV for Zn 0.90 Co 0.04 Li 0.06 Onanoparticles; Li-K edge and Zn-M 23  edge are quite prominent.No other element has been detected in EELS spectra. We per-formed an energy    ltered transmission electron microscopy (EF-TEM) measurement for Li mapping, Fig. 4(a) showing low resolution image of Zn 0.90 Co 0.04 Li 0.06 O sample. Li mapping wasacquired from Fig. 4(a) as shown in Fig. 4(b). This Li mapping micrograph (of selected sample) shows that Li was distributedhomogeneously and uniformly into the ZnO:Co crystal struc-ture. This reveals that no precipitates or clusters were present inZn 0.90 Co 0.04 Li 0.06 O sample. Earlier, our group has demonstratedthe single phase of Co doped ZnO nanoparticles achieved up tothe limit of 6% Co dopant. 32 From TEM analysis, it can beinferred that our co-doped samples are free from clustering orsecondary phases. A    er con  rmation of pure phase, now we aresafer to see the e ff  ect of co-dopants on the optical Raman,chemical and electrical properties of ZnO nanoparticles. Fig. 1  (a) BF-TEM micrograph at low and (b) high resolution TEMmagni fi cation (c) FFT image of the HR-TEM and (d) Hough transformfor Zn 0.94 Co 0.04 Li 0.02 O nanoparticles. Fig. 2  (a) BF-TEMmicrograph at low magni fi cation (b) HR-TEMimage(c) FFT image of the HR-TEM and (d) Hough transform for Zn 0.90 -Co 0.04 Li 0.06 O nanoparticles. 39830  |  RSC Adv. , 2015,  5 , 39828 – 39839 This journal is © The Royal Society of Chemistry 2015 RSC Advances Paper  3.2 Raman spectroscopy  Raman scattering spectroscopy is an e ff  ective and powerful nondestructive technique to study the molecular and crystal lattice vibrations, secondary phase segregation and the in  uence of impurity doping on the lattice vibrations of the host materials.It has been widely used to study the vibrational properties of ZnO bulk, nanostructures and thin  lms. 33 Raman investigationof transition metal (TM) doped ZnO has been a useful tool toconform the incorporation of TM and also to identify thehidden impurity phases present in the system. 34 The spacegroup of   C  6 y  describes the crystalline structure of wurtzitecompound with two formula units in the primitive unit cell. 35 Group theory calculations 36 suggested that the optical phononsat the  G -point of the Brillouin zone belong to the following irreducible representation:  G opt  ¼ 1A  1  + 2B 1  + 1E 1  + 2E 2 . Both A  1 and E 1  symmetries are polar and split into transverse optical(TO) and longitudinal optical (LO) phonons with di ff  erent frequencies, they can be designated as A  (TO)1  , E (TO)1  , A  (LO)1  andE (LO)1  modes. 37 Both A  1  and E 1  symmetries are polar and Ramanactive.Fig. 5, represents the room temperature un-polarized Ramanback-scattering spectra for a series of Zn 0.96   y Co 0.04 Li  y O (0.00 #  y  #  0.10) nanoparticles samples in the spectral range of 90 – Fig. 3  EELS spectra scan range 400 – 1400 eV; inset shows scan range 570 – 1200 eV (b) EELS spectra scan range 50 – 220 eV of Zn 0.90 -Co 0.04 Li 0.06 O nanoparticles. Fig.4  (a)BFmicrographand(b)EF-TEMforLimappingacquiredfromLi-K edge signal for Zn 0.90 Co 0.04 Li 0.06 O nanoparticles. This journal is © The Royal Society of Chemistry 2015  RSC Adv. , 2015,  5 , 39828 – 39839 |  39831 Paper RSC Advances  900 cm  1 . The spectrum consists of three fundamental peakslocated around 102 cm  1 , 442 cm  1 and 578 cm  1 . According toselection rules, 37 these peaks correspond to the E Low 2  , E High2  and A  (LO)1  fundamental phonon modes of hexagonal ZnO respec-tively. One of the most intensive bands in the Raman spectrumis E Low 2  mode, which ascribed to the vibrations of the Zn sub-lattices in samples. The other very intensive Raman activesharp mode is E High2  , dominantly assigned to the oxygen vibra-tions. 38,39 The Raman peaks located at about 201 cm  1 ,335 cm  1  were assigned to 2E Low 2  and E High2  – E Low 2  respectively.These bands belonged to second order phonon modes. A broadband appears in the spectral region 500 – 600 cm  1 . In thisregion, the polar A  (LO)1  and E (LO)1  modes (570 – 585 cm  1 ) are sit-uated and we also observed the presence of the disorder acti- vated 2B Low 1  silent mode at 552 cm  1 . The observation of usually forbidden multi phonons modes is possible because disorderleads to so   ening of Raman selection rule. 40 It should benoticed that the A  (LO)1  mode at 569 cm  1 and E (LO)1  modes at 586 cm  1 have relatively closed wave numbers and rise from thebackground, which srcinates from second order Raman scat-tering. Although the presence of impurities and/or defects canin  uence both of these modes, the E (LO)1  mode is more strongly a ff  ected by these e ff  ects. 38 The A  (LO)1  mode is commonly assigned to the oxygen vacancies, 41 zinc interstitials, 42 or defect complexes containing oxygen vacancies and zinc interstitial 43 inZnO system. Mode at position 667 cm  1 could be assigned to(TA + LO) multiple phonon scattering vibrations. The 2E Low 2  and2B Low 1  modesof10%Lidopedsamplesareshi   edtowardslower wave number as compared to other samples. This pronouncedasymmetry is attributed to the lattice disorder, as well as to an-harmonic phonon – phonon interactions. 38 In our spectra we seeno Co local vibration modes. We observed not a single addi-tional feature which can be explained as Co or Li related vibrations. Thus, our Raman spectra clearly indicate that allsamples exhibit the characteristics of the wurtzite structure of ZnO crystal. We have observed signi  cant changes in the intensity of Raman spectra upon co-doping Li in ZnO:Co sample. Tocompare the intensity of these samples, the multiplying factors(multiplication has been used for clear observation of intensity of Raman modes) are mentioned in Fig. 5 for each composition.The intensity of Raman modes decreased with increasing the Lico-doping concentration. This has an e ff  ect in changing thelocal environment around the host atom and thus the latticenormal symmetry. From measuring the area under the funda-mental modes, it is found that the Raman peaks becomebroader with increasing the Li content due to disorder inducedby co-doping of Li and Co species. Similar trends have beenreported for Co doped ZnO nanowires by Chang   et al. 44  Wecarefully noted that no additional lines appear and no splitting of any of the lines can be seen in any of the spectra, even at highdoping levels (  y ¼ 0.10). This indicates a good incorporation of the Li atoms into the ZnO lattice. From the analysis of Ramandata, it can be concluded that the co-doping of Co and Li ionsleadstothe decrease ofcrystalquality butitdoes notchange thehexagonal ZnO structure. 3.3 Chemical characterization To examine the prominent role of the defects that a ff  ect thepolarization e ff  ects, we have carefully considered the chemicalstate of each of the ionic species. We attempted to correlate thesubstitutional, and the interstitial role of Li atoms, simulta-neously with the presence of oxygen vacancy (defects). All X-ray photoelectron spectroscopy (XPS) spectra were deconvolutedusing XPSPEAK4.1 so    ware for Lorentzian – Gaussian   tting of asymmetric peaks by subtracting the Shirley background.High resolution Zn-2p core XPS spectra were measured toinvestigate how the chemical state of Zn is modi  ed upon co-doping of Li and Co. The measured Zn 2p 3/2  and 2p 1/2  XPSspectra of   y ¼ 0.02, 0.06 and 0.08 are displayed in Fig. 6(a – f). A single Lorentzian – Gaussian peak was nicely   tted to all Zn 2p 3/2 and 2p 1/2  core XPS spectra and no change in the peak positions(Zn 2p 3/2  and 2p 1/2 ) was noticed upon co-doping. The Zn 2p 3/2 and 2p 1/2  spectra exhibit symmetric features and single   ttedpeak rules out the possibility of existence of multiple compo-nents of Zn in these 45 samples. All spectra show a symmetricsingle peak located at 1022.20 eV and 1045.30 eV for Zn 2p 3/2 and 2p 1/2  spectra respectively. 46 From the observed spin – orbit splitting ( D  E  Zn  ¼  23.1 eV) and the binding energy positions of the two strong peaks of core level Zn 2p XPS spectra, and fromtheir line width, it has been concluded that Zn atoms are in +2oxidation state. 47 However the measured binding energy valuesof Zn-2p for Zn – O bonding are on the higher binding energy side compared to metallic Zn for which the core level Zn-2p 3/2 binding energy position is 1021.50 eV. 47  At    rst glance, the high resolution oxygen spectrum isasymmetric indicating the possibility that multi-component of oxygen species were present. The typical asymmetric O-1s peak in the surface was coherently    tted by three components,centered at 530.3 eV (O a ), 531.6 eV (O b ) and 532.8 eV (O c )respectively. This   tting for few selected samples  e.g. y  ¼  0.04and  y ¼ 0.10 has been shown in Fig. 7(a and b). The three   ttedbinding energy peaks approximate the earlier reports and weattributed the O a  peak on the low binding energy side of the Fig. 5  Room temperature Raman spectra of Zn 0.96   y  Co 0.04 Li  y  O (0.00 #  y  # 0.10) nanoparticles. 39832  |  RSC Adv. , 2015,  5 , 39828 – 39839 This journal is © The Royal Society of Chemistry 2015 RSC Advances Paper
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!