materials Review
Surface Sensitive Techniques for Advanced Characterization of Luminescent Materials Hendrik C. Swart
ID
Department of Physics, University of the Free State, P.O. Box 339, Bloemfontein ZA93002, South Africa; [email protected]; Tel.: +27-514012926 Received: 25 May 2017; Accepted: 1 August 2017; Published: 4 August 2017
Abstract: The important role of surface sensitive characterization techniques such as Auger electron spectroscopy (AES), X-ray photo electron spectroscopy (XPS), time of flight scanning ion mass spectrometry (TOF-SIMS) and High resolution transmission electron microscopy (HRTEM) for the characterization of different phosphor materials is discussed in this short review by giving selective examples from previous obtained results. AES is used to monitor surface reactions during electron bombardment and also to determine the elemental composition of the surfaces of the materials, while XPS and TOF-SIMS are used for determining the surface chemical composition and valence state of the dopants. The role of XPS to determine the presence of defects in the phosphor matrix is also stated with the different examples. The role of HRTEM in combination with Energy dispersive spectroscopy (EDS) for nanoparticle characterization is also pointed out. Keywords: AES; XPS; TOF-SIMS; HRTEM; CL degradation; valence state; phosphors
1. Introduction With the development of flat panel displays (FPDs) and light emitting diodes (LEDs), phosphor materials became an integral part of our life style. Luminescent compounds and materials also have numerous other uses such as temperature sensors, radiation dosimeters, optical probes, storage phosphor imaging, medical imaging, microscopy techniques, homeland security, detecting tools of biological structures and can possibly be used as efficient improvement of solar cells [1–3]. The emission properties, whether of a fast decay rate fluorescent material or a slow decay rate phosphorescent material, depend on the chemical composition of the host and the physical structure thereof as well as the dopants and defect concentration of the luminescent material [1]. Material synthesis conditions, binder characteristics, absorbing dyes usage, etc. also play important roles in the luminescent properties. The host, valence state and crystal field are the most important properties of phosphors to be reckoned with in the design of new phosphor materials [4–8]. The crystal field due to the local host environment in combination with the dopant ion with the correct valence/oxidation state can be used to obtain emissions from the Ultra violet (UV) to the Infra-red (IR) wavelength ranges. It remains important to be able to experimentally verify the oxidation state of the rare earth (RE) ions [9]. As an example, Ce3+ ions are good activators as well as useful sensitizers [10]. On the other hand, the possibility of Ce ions being found in the tetravalent Ce4+ oxidation state, which is non-luminescent, creates some difficulties in using these ions as luminescent centers in phosphors. Similar to Ce, the lanthanide Eu can also occur in two different valence states (Eu2+ /Eu3+ ), and, although both of these are luminescent, their emission properties are very different. Eu2+ has emission from an allowed d-f transition, which is host dependent, whereas Eu3+ has emission from forbidden f-f transitions at wavelengths that do not vary with the host material. Several analytical methods have been reported for low-level monitoring of RE ions in various sample matrices [11]. These methods include inductively coupled plasma mass spectroscopy
Materials 2017, 10, 906; doi:10.3390/ma10080906
www.mdpi.com/journal/materials
Materials 2017, 10, 906
2 of 18
(ICP-MS) [12], inductively coupled plasma atomic emission spectroscopy (ICP-AES) [13], isotope dilution mass spectrometry [14], resonance light scattering (RLS) [15], voltammetry [16], capillary electrophoresis [17], X-ray fluorescence (XRF) [18], fluorimetry [19], ion-microprobe [20] and other methodologies [21]. The stability of these phosphors under electron or photon irradiation is important for the FPD and LED market. The effects of bombarding with electrons vary from the development of charge, to quenching of the radiative decay of the activator excited states, to the dissociation of the gaseous or surface absorbed species causing the growth of oxide or carbon films on the surface (i.e., development of a surface “dead layer”) [22], creating defects in the surface and interface layers. Abrams and Holloway [23] discussed the development of the “dead layers”, defined to be a surface layer with no or weak luminescence. The molecular dissociation, surface chemical reactions and desorption of neutral and/or ionic atoms and/or molecules by electron or photon primary beams have been known since the 1950s to be important in surface and vacuum physics [24]. For example, electron-stimulated-desorption (ESD) was first observed because of its effects on the accuracy of vacuum pressure measurements using a hot filament ionization of a high vacuum gauge. ESD and many similar phenomena resulting from electron and photon interactions with solid surfaces are described in the book by Redhead et al. [25]. Duvenhage et al. [26] showed a 60% decrease in luminescence intensity upon UV light exposure of a mer-tris-8-hydroxy-quinolinato-indium (III) complex used in organic light emitting diodes (OLEDs). This was an indication that oxygen in moisture in the air caused some of the phenoxide rings in the complex to decompose and destroy the luminescent centers in the process. Surface characterization techniques play a vital role in the complete understanding of the luminescent properties and oxidation states of phosphor materials [8]. Auger electron spectroscopy (AES), X-ray photo electron spectroscopy (XPS), time of flight scanning ion mass spectrometry (TOF-SIMS) and High resolution transmission electron microscope (HRTEM) are used to characterize different phosphor materials. The important role of these techniques is illustrated in this paper by using example studies from previous published work. 2. Experimental Setup Phosphor materials are normally prepared with different synthesized methods such as chemical bath deposition (CBD) [27], sol-gel [5,7], combustion [4], sol-gel combustion [28], hydro thermal [29], solid-state reaction [30,31], etc. Surface characterization and morphology of these phosphors are important and are carried out with AES, XPS, TOF-SIMS and HRTEM. The optical characterization is carried out with photoluminescence (PL), cathodoluminescence (CL), UV-Vis, Fourier transformed infra-red (FTIR) and Raman spectroscopy. The stability and degradation of these phosphors are important properties that are studied on a regular basis, especially for applications in display technologies. 2.1. CL and AES Figure 1 shows the experimental setup for an in-house CL intensity degradation system. The same electron beam that is used to excite the Auger electrons of the phosphor surface is simultaneously been used to excite the phosphor material. Basically, the changes in intensity of the different CL peaks and Auger peak to peak heights (APPH) are monitored and compared with each other. The system consists of an AES vacuum system coupled with a CL spectrometer (Ocean Optics, Inc., Dunedin, FL, USA). The PHI (model 549) Auger spectrometer (Physical Electronics, INC. (PHI), Chanhassen, MN, USA) and Ocean Optics (S2000) spectrometer (Ocean Optics, Inc., Dunedin, FL, USA) are simultaneously used to collect the Auger and CL data, respectively. The light output is collected via a fiber optic connected to the CL spectrometer. The primary AES electron beam current can be adjusted and is typically between 6 and 12 µA. The beam size is varied from 100 to 300 µm depending on the focus of the beam and the beam voltage and beam current. Beam densities vary between 2.5 and 88 mA/cm2 . The Auger and CL data are collected in a vacuum chamber with an initial base pressure in the low 10−9 Torr range, where after the chamber maybe backfilled with O2 , N2 or CO2 up to
Materials 2017, 10, 906
3 of 18
Materials 2017, 10, 906
3 of 17
6 Torr during the degradation study. Throughout the experiment, the Auger and CL pressures 10− Auger andofCL data are recorded using the same adjustable primary electron beam of 1 to 5 keV. The data are recorded using the sameand adjustable primary electronchanges beam of during 1 to 5 keV. The decrease of decrease of the CL intensities the surface chemical prolonged electron the CL intensities the surface is chemical changes during prolonged electron of the bombardment of and the phosphors monitored continuously for periods up tobombardment 24 h at the desired phosphors is monitored continuously for periods up to 24 h at the desired different gas pressures. different gas pressures.
Figure 1. 1. The The CL/AES CL/AES setup Figure setup for for electron electron degradation degradation measurements. measurements.
2.2. XPS 2.2. XPS The surface surface chemical chemical changes changes are then determined determined with use of of XPS. XPS. The The XPS XPS data data are are The are then with the the use collected before and electronic states of collected before and and after after degradation degradationtotoevaluate evaluatethe thechemical chemicalcomposition composition and electronic states the different elements. The data are collected using the PHI 5000 Versa probe-Scanning ESCA of the different elements. The data are collected using the PHI 5000 Versa probe-Scanning ESCA microprobe (Physical (Physical Electronics, Electronics, INC. INC. (PHI), (PHI), Chanhassen, Chanhassen, MN, MN, USA). USA). Low Low energy energy Ar Ar++ ion ion gun gun and and microprobe low energy energy neutralizer neutralizerelectron electrongun gunare areused usedtoto minimize charging surface. Monochromatic low minimize charging onon thethe surface. Monochromatic Al Al Kα radiation (hν = 1486.6 eV) is used as the excitation source. A 25 W, 15 kV electron beam is used Kα radiation (hν = 1486.6 eV) is used as the excitation source. A 25 W, 15 kV electron beam is used to to excite X-ray beam of 100 µm diameter, that is used to analyze the different binding excite thethe X-ray beam of 100 µm diameter, that is used to analyze the different binding energyenergy peaks peaksenergy (pass energy 11 eV, analyzer resolution ≤0.5 eV). Multipak 8.2 software [32]to is analyze used to (pass 11 eV, analyzer resolution ≤0.5 eV). Multipak version version 8.2 software [32] is used analyze the chemical elements and their electronic states using Gaussian-Lorentz fits. The valence the chemical elements and their electronic states using Gaussian-Lorentz fits. The valence state and statepositions and site of positions of theare dopants are also confirmed site the dopants also confirmed with XPS. with XPS. 2.3. TOF-SIMS TOF-SIMS 2.3. Furthermore, TOF-SIMS TOF-SIMS measurements measurements were performed on on aa PHI PHI TRIFT TRIFT V V nanoTOF nanoTOF (Physical (Physical Furthermore, were performed Electronics, INC. (PHI), Chanhassen, MN, USA) or ION TOF-SIMS (Ion-tof GmbH, Münster, Electronics, INC. (PHI), Chanhassen, MN, USA) or ION TOF-SIMS (Ion-tof GmbH, Münster, Germany) + + + + Germany) using Au a pulsed or Bi primary ion beam, to acquire chemical images of the phosphor by using aby pulsed or BiAuprimary ion beam, to acquire chemical images of the phosphor in in both positive and the negative secondaryion ionpolarities. polarities.The Theanalytical analyticalfield-of-view field-of-view is is typically typically both thethe positive and the negative secondary 200 µm µm × × 200 256 pixel pixel × × 256 dose is is maintained maintained 200 200 µm µm with with a a 256 256 pixel pixel digital digital raster, raster, and and the the primary primary ion ion dose well within within the the static static limit limit for for each each analysis. analysis. Charge Charge compensation compensation is is achieved achieved with with aa dual-beam dual-beam well −− +) charge + (≤15 eV e and ≤10 eV Ar neutralizer. A raw data stream file is collected to allow full post(≤15 eV e and ≤10 eV Ar ) charge neutralizer. A raw data stream file is collected to allow full acquisition evaluation (i.e., retrospective analysis) of the data. post-acquisition evaluation (i.e., retrospective analysis) of the data. 2.4. HRTEM HRTEM is performed by using a JEM-2100 electron microscope (JEOL, Tokyo, Japan) at a beam voltage of 200 keV. The EDS (Energy Dispersive X-ray Spectrometer, Oxford Instruments plc, Oxfordshire, UK) mapping and chemical composition analysis is performed with an Oxford XMax
Materials 2017, 10, 906
4 of 18
2.4. HRTEM HRTEM is performed by using a JEM-2100 electron microscope (JEOL, Tokyo, Japan) at a beam voltage of 200 keV. The EDS (Energy Dispersive X-ray Spectrometer, Oxford Instruments plc, Oxfordshire, UK) mapping and chemical composition analysis is performed with an Oxford XMax 80 EDS detector (Oxford Instruments plc, Oxfordshire, UK). The ImageJ software (Laboratory for Optical and Computational Instrumentation, Wisconsin-Madison, WI, USA) is used to determine the size of the nanoparticles (NPs) from the obtained SEM/TEM micrographs. PL is done with a Cary Eclipse (Agilent, Santa Clara, CA, USA) equipped with a 150 W Xenon lamp and a 325 nm He-Cd laser. Raman spectra are obtained by using a NXR FT-Raman module microscopy (Thermo Fisher Scientific Inc, Madison, WI, USA) system with a 2.5 W Nd: YVO4 laser excitation of a 1064 nm wavelength and a high-performance liquid nitrogen-cooled germanium detector. 3. Results and Discussion 3.1. Electron Degradation—AES and XPS 3.1.1. ZnS A typical electron degradation study ends up with: (a) an APPH spectrum as a function of time; and (b) a CL intensity as function of time, as demonstrated for a ZnS phosphor exposed to an electron beam as indicated in Figure 2 [33]. Figure 2a shows the APPHs of the different elements present on the surface during the degradation study, S (152 eV), C (272 eV), O (511 eV) and Zn (994 eV), against electron dose during 2 keV, 64 mA cm−2 electron bombardment at an oxygen pressure of 1 × 10−6 Torr. Oosthuizen et al. [33] found that the C APPH immediately decreased exponentially when the surface was exposed to the electron beam. The C on the phosphor surface was present from adventitious atmospheric contamination. Simultaneous to the time when the C decreased, the CL intensity stayed more or less constant (Figure 2b), which indicated that the C on the surface acted as a protected layer for the CL degradation. The S APPH at first increased due to the removal of the C from the surface and then started to decrease exponentially while the O increased after most of the C was removed from the surface. With the increase in the O and decrease in S APPHs the CL also decreased, which showed a direct correlation between the surface reactions and the CL intensity changes. It was found by Oosthuizen et al. [33] that ZnO has formed on the surface of the ZnS under prolonged electron bombardment. They also measured the presence of SO2 gas during electron bombardment. Itoh et al. [34], using XPS, reported that ZnSO4 was formed on the surface of the ZnS phosphor during electron irradiation in a H2 O ambient. Both ZnO (−94.43 kcal/mol O2 ) and ZnSO4 (72.09 kcal/mol O2 ) have a negative heat of formation [33]. The degraded layer grew in thickness with prolonged electron irradiation time. Chen et al. [35] found a linear growth rate (Figure 3a) for the ZnO layer in a wet oxygen atmosphere and a ZnSO4 formation that decayed exponentially with time and it was postulated that this was due to the diffusion of the charge reactants through the ZnSO4 film to the reaction interfaces, as shown in Figure 3b. During the course of electron irradiation the bombarded areas of the phosphor became darkened in color and degraded in CL efficiency, as pointed out by the Secondary X-ray Imaging in Figure 4a. XPS was used to determine the chemical species in the degraded spot as well as on the undegraded area. A schematic of the degraded spot and growth of the non-luminescent layer with electron bombardment time is shown in Figure 4b. Holloway and Swart [36,37] came up with a well-known ESSCR (electron stimulated surface chemical reaction) mechanism that was applied to the degradation of several phosphors. According to this model, reactive gas molecules adsorb on the surface of the ZnS phosphor and are dissociated to reactive atomic species by the electron beam. This results in the formation of a non-luminescent ZnO/ZnSO4 layer on the surface and volatile SO2 as illustrated in Figure 5.
Materials 2017, 10, 906 Materials 2017, 906 Materials 2017, 10,10, 906 Materials 2017, 10, 906
5 of 18
APPH (arb. Units) APPH (arb. Units) APPH (arb. Units)
Relative CL intensity Relative CLCL intensity Relative intensity
5 of 5 of 1717 5 of 17
2 2) Electron dose (C/cm Electron dose (C/cm ) Electron dose (C/cm2)
2 2) Electron dose (C/cm Electron dose (C/cm ) Electron dose (C/cm2)
Figure 2.The The APPHs;and and(b) (b)CL CLintensity intensityofof ofZnO ZnOduring duringprolonged prolonged electron bombardment Figure (a)(a) APPHs; electron bombardment asasas Figure 2.2.The (a) APPHs; (b) CL intensity ZnO during prolonged electron bombardment Figure 2. of The (a) APPHs; and (b) CL intensity of ZnO during prolonged electron bombardment as function electron dose [11]. functionofofelectron electrondose dose [11]. [11]. function function of electron dose [11].
Figure ZnSO ; and(b) (b)ZnO ZnOlayer layer growth and wet oxygen atmosphere function Figure (a)(a)ZnSO ZnSO 4; ;4and growth inindry and a awet oxygen atmosphere asasfunction ofof Figure 3.3.3. (a) (b) ZnO layer growth indry dry and a wet oxygen atmosphere as function 44; and Figure 3. (a) ZnSO andan (b)illustration ZnO layer growth in dry and athe wetcharge oxygenreactants atmosphere as function of4 degradation time with of the diffusion of during the ZnSO 4 degradation time with an illustration of the diffusion of the charge reactants during the ZnSO of degradation time with an illustration of the diffusion of the charge reactants during the ZnSO 4 4 degradation time with an illustration of the diffusion of the charge reactants during the ZnSO formationasasananinset inset[35]. [35]. formation formation as an inset [35]. formation as an inset [35].
Figure4.4.(a)(a)AAschematic schematicofofthe thedegraded degradedspot spotwith withananenlargement enlargementofofone oneofofthe theparticles particlesshowing showing Figure Figure 4. (a) A schematic of the degraded spot with an enlargement ofone oneofoftime; theparticles particles showingthe Figure 4. (a) A schematic of the degraded spot with an enlargement of the showing the growth of the degraded thin film with prolonged electron bombardment and (b) secondary the growth of the degraded thin film with prolonged electron bombardment time; and (b) secondary the growth the degraded thin film prolonged withspot prolonged electron bombardment (b) secondary X-ray (SXI) a2424H Hdegraded degraded 2SiO 5:Ce. growth ofImaging the of degraded film with electron bombardment time;time; and and (b) secondary X-ray X-ray Imaging (SXI) ofofathin spot ofofY2Y SiO 5:Ce. X-ray Imaging (SXI) of a 24 H degraded spot of Y2SiO5:Ce. Imaging (SXI) of a 24 H degraded spot of Y2 SiO5 :Ce.
Materials 2017, 10, 906 Materials 2017, 10, 906
6 of 18 6 of 17
Figure Figure5. 5. Schematic Schematic illustration illustration of of the the ESSCR ESSCR mechanism. mechanism.
3.1.2. Sr Sr55(PO (PO44))33F:Eu 3.1.2. F:Eu Asmentioned, mentioned,other other phosphors showed the ESSCR mechanism to As phosphors showed thatthat the ESSCR mechanism could could also bealso usedbe to used explain explain the degradation thereof. Figure 6 illustrates the comparative PL emissions of undegraded and the degradation thereof. Figure 6 illustrates the comparative PL emissions of undegraded and degraded degraded CL emission of Srat 5(PO )3F:Eu at an oxygen pressure of [5]. 1 × This 10−6 is Torr [5]. example This is a CL emission spectra of Srspectra an4oxygen pressure of 1 × 10−6 Torr a good 5 (PO4 )3 F:Eu 2+ 3+ 2+ 2+ as well 3+ emission. good example where the CL emission consists of both as well as emission. The where the CL emission spectrum consistsspectrum of both Eu as EuEu TheEu Eu wavelength 2+ wavelength position is matrix sensitive, 3+ emission wavelength position is not 3+ Eu while the Eu position is matrix sensitive, while the Eu emission wavelength position is not affected by a change 2+ the affected by a change The Eu2+ CL as emission as the a broad bandcentred with the in the matrix. The Euin CL matrix. emission appeared a broadappeared band with maxima at maxima 430 nm 6 7 6 7 centred at 430 nm due to the 4f 5d→4f transition and is more prominent compared to the PL emission. due to the 4f 5d→4f transition and is more prominent compared to the PL emission. The prominent 2+ CL emission was obtained due to 6 65d level that can be 2+ prominent The blue that emitting 4fpopulated Eu CL emissionEu was obtained due to the blue emitting 4fthe 5d level can be using high 6 7 transitions of Eu2+ 6 7 2+ populated using such high as energy excitation such [38]. The energy excitation cathode rays [38]. Theas4fcathode 5d→4f rays transitions of 4f Eu5d→4f require assistance from require assistance from of low phonons of the the hostenergy matrix. In addition, energy low frequency phonons thefrequency host matrix. In addition, position of the the 4f6 5d must position be blue 6 5 5 of the 4f must be blue shifted to0 allow population of the D0 state by a direct non-radiative shifted to 5d allow population of the D state by a direct non-radiative relaxation process. The Eu3+ 5 D →7 Fat, 593, 7 5 7 and 578 nm due to relaxation Eu3+toemission 616 nm due to 5D0→7F1, 7F2 transition emission atprocess. 593, 616 The nm due 0 1 F2 transition and 578 nm due to D0 → F0 appeared [39] as a 5D0→7F0 appeared [39] as a mirror image of the PL spectrum. After degradation, it appears that the mirror image of the PL spectrum. After degradation, it appears that the blue emission of the Eu2+ ions 2+ bluecentered emissionatof420 thenm Euand ions was centered at 420and nmshifted and remained shifted ≈10 nm was remained prominent ≈10 nmprominent compared and to the undegraded 3+ decreased 3+ compared to the undegraded sample. The CL intensity of the red emission due to the Eu sample. The CL intensity of the red emission due to the Eu decreased drastically with an increasing 2. According to the ESSCR model, the drastically withof an0increasing in2electron dosetoofthe 0 to 450 C/cm in electron dose to 450 C/cm . According ESSCR model, the Sr–O, Sr–F and P–O bonds are Sr–O,toSr–F and P–O bonds are likely to strontium be broken into free oxygen, fluoride, strontium and likely be broken into free oxygen, fluoride, and phosphorous when irradiated with a beam phosphorous when irradiated with a beam of electron. This will be followed by a chemical reaction of electron. This will be followed by a chemical reaction resulting in new chemical compounds forming resulting in newInchemical compounds on the surface. In most cases, thetherefore new oxide layers are on the surface. most cases, the new forming oxide layers are non-luminescent, and will reduce non-luminescent, and thereforetowill reduce the CL intensity. Simultaneous to the O 2 desorption the CL intensity. Simultaneous the O desorption during CL degradation, it is most likely that P 2 during CL degradation, it is most likely that P (metallic), SrO [40] and P 2 O 5 [41] mixed layers were (metallic), SrO [40] and P2 O5 [41] mixed layers were formed on the surface according to the ESSCR formed on the surface according to thearises ESSCR mechanism. The Eu3+ emission ariseswhich from are the mechanism. The Eu3+ emission mostly from the site occupancy of the Sr2mostly metal sites 3+ site occupancy of the Sr 2 metal sites which are present on the outer surface of the crystal. This caused present on the outer surface of the crystal. This caused the decrease in the Eu CL intensity during 7 5 7 exposure. The3+5D the decrease in the exposure. Eu3+ CL intensity prolonged electron 0-7F0 (578behavior nm) and prolonged electron The 5 Dduring emission 0 - F0 (578 nm) and D0 - F2 (616 nm) Eu 5D0-7F2 (616 nm) Eu3+ emission behavior depends on the site symmetry behavior. 3+ Theemissions slight shift of depends on the site symmetry behavior. The slight shift of the 653 and 705 nm Eu CL arise 3+ CL emissions arise due to crystal ligand field splitting and asymmetrical the 653 and 705 nm Eu due to crystal ligand field splitting and asymmetrical environment of the crystal [39]. Simultaneously, environment of the crystal [39]. Eu2+ near emission from the Eu ion remains which is the Eu2+ emission arises from theSimultaneously, Eu ion which is the present to thearises defect centers, which present near to the centers,remains which remains protected. Thus,electron Eu2+ CLbeam intensity remains protected. Thus, Eu2+defect CL intensity intact during prolonged exposure. Theintact shift during prolonged electron beam exposure. The shift in the peak position, however, indicates a change in the peak position, however, indicates a change in the crystal field due to a change in the host lattice. in the crystal field due to a change in the host lattice.
Materials 2017, 10, 906 Materials 2017, 10, 906 Materials 2017, 10, 906
7 of 18 7 of 177 of 17
Figure 6. The comparative PL and CL output of the Sr5(PO4)3F:Eu as function of Coulomb dose at
Figure 6. The comparative PL and CL output of the Sr5 (PO4 )3 F:Eu as function of Coulomb dose at 1 × 10−6 Torr O2 atmosphere under electron beam excitation with the primary beam voltage and beam 1 ×Figure 10−6current Torr O under beam theasprimary voltage and 6. The PL andelectron CL output of excitation the Sr5(POwith 4)3F:Eu functionbeam of Coulomb dosebeam at of2comparative 2atmosphere keV and 10 µA, respectively [5]. current ofTorr 2 keV 10 µA, respectively [5].beam excitation with the primary beam voltage and beam 1 × 10−6 O2and atmosphere under electron 3.1.3. Yof 2SiO 5:Ce3+ current 2 keV and 10 µA, respectively [5]. 3+ 3.1.3. Y2 SiODuring 5 :Ce the degradation studies of Y2SiO5:Ce3+, an increase in the CL intensity at a wavelength of 3.1.3. 650 Y2SiO 5:Ce3+ nm was measured (Figure 7a) [42]. The XPS (Figure 8a) and CL (Figure 7a) indicated that the During the degradation studies of Y2 SiO5 :Ce3+ , an increase in the CL intensity at a wavelength electron stimulated reaction led to of theYformation of a luminescent silicon dioxide (SiO2) layer on the During degradation studies 2SiO5:Ce3+, an increase in the CL intensity at a wavelength of of 650 nm wasthe measured 7a) [42]. The The XPSfirst (Figure 8a) and CL7a) (Figure 7a) indicated surface of the Y2SiO5(Figure :Ce phosphor powder. spectrum (Figure is characteristic of thethat the 650 nmstimulated was measured (Figure 7a)the [42]. The XPSof (Figure 8a) and CL (Figure 7a) indicated thaton thethe 3+ electron reaction led to formation a luminescent silicon dioxide (SiO ) layer doublet character of the blue light emission from Ce , due to the 4f ground state splitting2[43,44]. electron stimulated reaction led to the formation of a luminescent silicon dioxide (SiO 2) layer on the the fitted resultspowder. from an XPS spectrum for the Si (Figure 2p peak in theisYcharacteristic 2SiO5 state and of the surface Figure of the8aY2shows SiO5 :Ce phosphor The first spectrum 7a) surface of 24 the Ythe 2SiO5:Ce phosphor powder. The first3+ spectrum (Figure 7a) is characteristic of the after h in SiO 2 state. The Si 2p peak shifted and changed shape from the yttrium silicate (Y2SiO5) doublet character of the blue light emission from Ce3+ , due to the 4f ground state splitting [43,44]. doublet character of thebinding blue light emission , due the 4f ground state [43,44]. chemical with 101.3 eVfrom to theCe silica (SiO2to ) chemical state 103.3 eV. splitting The increase Figure 8a showsstate the fitted resultsenergy from an XPS spectrum for the Si 2p peak in the Y SiO 5 state and after Figurein8a thethat fitted an XPSand spectrum for the Sito2pthe peak in the2peak Y2SiO 5 state and CLshows indicates theresults SiO2 is from luminescent thus contributing emission between 24 h in the SiO2700 state. The Sia2p peak shifted and changed shape from the yttrium silicate (Y2 SiO5 ) 600h and nm. SiO2 is phosphor material and thefrom electron irradiation can2SiO after 24 in the SiO 2 state. Thewide Si 2pband peakgap shifted and changed shape the beam yttrium silicate (Y 5) chemical state with binding energy 101.3 eV to the silica (SiO ) chemical state 103.3 eV. The increase break the Si–O andenergy cause intrinsic defects Skuja et2)2al.chemical [46] reported peak chemical state with bonds binding 101.3 eV to the[45]. silica (SiO statetwo 103.3 eV.intensities The increase in in CLCL indicates the SiO luminescent and thus contributing totothe emission peak between for SiO2 atthat 1.9that eV (650 nm) eV (459 nm) with a theory that the two peaks are related to intrinsic 2 is2and indicates the SiO is 2.7 luminescent and thus contributing the emission peak between600 defects involving cleavage of the bonds. Amaterial definite contribution from the SiO2 irradiation 1.9 eV defectcan to break and 700 nm. a wide band gapSi–O phosphor and the electron beam 2 isSiO 600 and 700SiO nm. 2 is a wide band gap phosphor material and the electron beam irradiation can transition from the higher 5d levels to the 4f7/2Skuja levels leads to[46] the increase in the CLpeak intensity and thebreak Si–Othe bonds and cause intrinsic defects [45]. et al. reported two intensities the Si–O bonds and cause intrinsic defects [45]. Skuja et al. [46] reported two peak intensitiesfor peak emission between 600 and 700 nm, thus also resulting in the change in color. The formation of SiO 1.9 eV nm)nm) andand 2.7 2.7 eVeV (459 nm) with a theory that twopeaks peaksare arerelated related intrinsic 2 at for SiO 2 at 1.9(650 eV (650 (459 nm) with theory that the the two toto intrinsic the luminescent SiO2 layer on the surface of the Y2a SiO 5:Ce therefore leads to the degradation of the defects involving cleavage of the Si–O bonds. A definite contribution from the SiO 1.9 eV defect defects involving cleavage of the Si–O bonds. definite contribution from the SiO 2 1.9 eV defect to to 2 blue emitting Y2SiO5:Ce phosphor powders. thethe transition from the higher 5d levels to the 4f levels leads to the increase in the CL intensity and transition from the higher 5d levels to the 4f7/2 7/2 levels leads to the increase in the CL intensity and peak emission between 600 and 700 nm, thus also resultingininthe thechange changeinincolor. color.The Theformation formationofofthe peak emission between 600 and 700 nm, thus also resulting the luminescent SiO2 layer onsurface the surface ofYthe Y 2 SiO 5 :Ce therefore leads to the degradation luminescent SiO2 layer on the of the SiO :Ce therefore leads to the degradation of of thethe blue 2 5 blue emitting Y 2 SiO 5 :Ce phosphor powders. emitting Y2 SiO5 :Ce phosphor powders.
Figure 7. CL intensity against the wavelength for the light emitted from the powders: (a) before and after 3 h and 24 h electron bombardment of Y2SiO5:Ce [42]; and (b) SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a 1 × 10−7 Torr O2 pressure. The digital images were taken before and after degradation.
Figure intensityagainst againstthe thewavelength wavelength for the (a)(a) before and Figure 7. 7. CLCL intensity the light lightemitted emittedfrom fromthe thepowders: powders: before and 2:PbS nanoparticles before and after h and electronbombardment bombardmentof ofYY22SiO55:Ce after 3 h3 and 2424 h helectron :Ce [42]; [42];and and(b) (b)SiO SiO :PbS nanoparticles before and 2 −7 − 7 Torr O 2 pressure. The digital images after degradation [47] with a 2 keV electron beam at a 1 × 10 after degradation [47] with a 2 keV electron beam at a 1 × Torr O2 pressure. The digital images were taken before andafter afterdegradation. degradation. were taken before and
Materials 2017, 10, 906
8 of 18
Materials 2017, 10, 906
8 of 17
Materials 2017, 10, 906
8 of 17
Figure 8. XPS (a) XPS fitted results theSiSi2p 2pininthe theYY2SiO SiO5:Ce after degradation [42] 24 h degradation Figure 8. (a) fitted results forforthe 2 5 :Ce after degradation [42] 24 h degradation −2 electron current density at 26.3 the (b) an oxygen pressure 10−−66Torr 2 ; and in aninoxygen pressure ofof1 1××10 Torrwith withthe the electron current density at mA·cm 26.3 mA; ·and cm−(b) Figure 8. (a) XPS fitted results for the Si 2p in the Y 2 SiO 5 :Ce after degradation [42] 24 h beam at a XPS of SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electrondegradation the XPS of SiO2 :PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a oxygen of 1 × 10−6 Torr with the electron current density at 26.3 mA·cm−2; and (b) the Torr Opressure 2 pressure. 1in×an 10−7 1 × 10−7 Torr O2 pressure. XPS of SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a
−7 Torr O2 pressure. × 102-PbS 3.1.4.1SiO
3.1.4. SiO2 -PbS
Figure 7b presents the broad CL spectra of SiO2; 0.134 mol %-PbS nanoparticles, synthesized 3.1.4. SiO7b 2-PbS Figure presents broad spectra of SiOwith mol %-PbS nanoparticles, synthesized 2 ; 0.134 with a sol-gel method,the before andCL after degradation the maximum intensity at the wavelength withofa sol-gel method, before and after degradation with the maximum intensity at the wavelength Figure 7b presents the broad CL spectra of SiO 2 ; 0.134 mol %-PbS nanoparticles, synthesized about 680 nm [47]. XPS showed that during degradation in an oxygen atmosphere the SiO2 is of with sol-gel before that and of after degradation maximum intensity atthe theSiO wavelength about 680anm showed during degradation inthe anduring oxygenelectron atmosphere is reduced reduced to[47]. SiOmethod, x XPS by the desorption oxygen from thewith surface bombardment the 2and of xabout 680 nmenriched [47]. XPS showed that during degradation in an oxygen thethat SiOsurface 2 is to SiO by the desorption ofinoxygen from surface electron bombardment and the surface was left elemental Si,the as shown induring Figure 8b. Dhlamini etatmosphere al. [47] found the reduced to SiO x by the desorption of oxygen from the surface during electron bombardment and rate of degradation of the CL intensity decreased with an increase in the oxygen pressure. The was left enriched in elemental Si, as shown in Figure 8b. Dhlamini et al. [47] found that the the rate of surface was leftintensity enriched elemental Si, as shown inthe Figure Dhlamini al. [47] SiO found that decrease in isinbest explained in terms formation of a lessetpressure. efficient x (0decrease < x
Surface Sensitive Techniques for Advanced Characterization of Luminescent Materials Hendrik C. Swart
ID
Department of Physics, University of the Free State, P.O. Box 339, Bloemfontein ZA93002, South Africa; [email protected]; Tel.: +27-514012926 Received: 25 May 2017; Accepted: 1 August 2017; Published: 4 August 2017
Abstract: The important role of surface sensitive characterization techniques such as Auger electron spectroscopy (AES), X-ray photo electron spectroscopy (XPS), time of flight scanning ion mass spectrometry (TOF-SIMS) and High resolution transmission electron microscopy (HRTEM) for the characterization of different phosphor materials is discussed in this short review by giving selective examples from previous obtained results. AES is used to monitor surface reactions during electron bombardment and also to determine the elemental composition of the surfaces of the materials, while XPS and TOF-SIMS are used for determining the surface chemical composition and valence state of the dopants. The role of XPS to determine the presence of defects in the phosphor matrix is also stated with the different examples. The role of HRTEM in combination with Energy dispersive spectroscopy (EDS) for nanoparticle characterization is also pointed out. Keywords: AES; XPS; TOF-SIMS; HRTEM; CL degradation; valence state; phosphors
1. Introduction With the development of flat panel displays (FPDs) and light emitting diodes (LEDs), phosphor materials became an integral part of our life style. Luminescent compounds and materials also have numerous other uses such as temperature sensors, radiation dosimeters, optical probes, storage phosphor imaging, medical imaging, microscopy techniques, homeland security, detecting tools of biological structures and can possibly be used as efficient improvement of solar cells [1–3]. The emission properties, whether of a fast decay rate fluorescent material or a slow decay rate phosphorescent material, depend on the chemical composition of the host and the physical structure thereof as well as the dopants and defect concentration of the luminescent material [1]. Material synthesis conditions, binder characteristics, absorbing dyes usage, etc. also play important roles in the luminescent properties. The host, valence state and crystal field are the most important properties of phosphors to be reckoned with in the design of new phosphor materials [4–8]. The crystal field due to the local host environment in combination with the dopant ion with the correct valence/oxidation state can be used to obtain emissions from the Ultra violet (UV) to the Infra-red (IR) wavelength ranges. It remains important to be able to experimentally verify the oxidation state of the rare earth (RE) ions [9]. As an example, Ce3+ ions are good activators as well as useful sensitizers [10]. On the other hand, the possibility of Ce ions being found in the tetravalent Ce4+ oxidation state, which is non-luminescent, creates some difficulties in using these ions as luminescent centers in phosphors. Similar to Ce, the lanthanide Eu can also occur in two different valence states (Eu2+ /Eu3+ ), and, although both of these are luminescent, their emission properties are very different. Eu2+ has emission from an allowed d-f transition, which is host dependent, whereas Eu3+ has emission from forbidden f-f transitions at wavelengths that do not vary with the host material. Several analytical methods have been reported for low-level monitoring of RE ions in various sample matrices [11]. These methods include inductively coupled plasma mass spectroscopy
Materials 2017, 10, 906; doi:10.3390/ma10080906
www.mdpi.com/journal/materials
Materials 2017, 10, 906
2 of 18
(ICP-MS) [12], inductively coupled plasma atomic emission spectroscopy (ICP-AES) [13], isotope dilution mass spectrometry [14], resonance light scattering (RLS) [15], voltammetry [16], capillary electrophoresis [17], X-ray fluorescence (XRF) [18], fluorimetry [19], ion-microprobe [20] and other methodologies [21]. The stability of these phosphors under electron or photon irradiation is important for the FPD and LED market. The effects of bombarding with electrons vary from the development of charge, to quenching of the radiative decay of the activator excited states, to the dissociation of the gaseous or surface absorbed species causing the growth of oxide or carbon films on the surface (i.e., development of a surface “dead layer”) [22], creating defects in the surface and interface layers. Abrams and Holloway [23] discussed the development of the “dead layers”, defined to be a surface layer with no or weak luminescence. The molecular dissociation, surface chemical reactions and desorption of neutral and/or ionic atoms and/or molecules by electron or photon primary beams have been known since the 1950s to be important in surface and vacuum physics [24]. For example, electron-stimulated-desorption (ESD) was first observed because of its effects on the accuracy of vacuum pressure measurements using a hot filament ionization of a high vacuum gauge. ESD and many similar phenomena resulting from electron and photon interactions with solid surfaces are described in the book by Redhead et al. [25]. Duvenhage et al. [26] showed a 60% decrease in luminescence intensity upon UV light exposure of a mer-tris-8-hydroxy-quinolinato-indium (III) complex used in organic light emitting diodes (OLEDs). This was an indication that oxygen in moisture in the air caused some of the phenoxide rings in the complex to decompose and destroy the luminescent centers in the process. Surface characterization techniques play a vital role in the complete understanding of the luminescent properties and oxidation states of phosphor materials [8]. Auger electron spectroscopy (AES), X-ray photo electron spectroscopy (XPS), time of flight scanning ion mass spectrometry (TOF-SIMS) and High resolution transmission electron microscope (HRTEM) are used to characterize different phosphor materials. The important role of these techniques is illustrated in this paper by using example studies from previous published work. 2. Experimental Setup Phosphor materials are normally prepared with different synthesized methods such as chemical bath deposition (CBD) [27], sol-gel [5,7], combustion [4], sol-gel combustion [28], hydro thermal [29], solid-state reaction [30,31], etc. Surface characterization and morphology of these phosphors are important and are carried out with AES, XPS, TOF-SIMS and HRTEM. The optical characterization is carried out with photoluminescence (PL), cathodoluminescence (CL), UV-Vis, Fourier transformed infra-red (FTIR) and Raman spectroscopy. The stability and degradation of these phosphors are important properties that are studied on a regular basis, especially for applications in display technologies. 2.1. CL and AES Figure 1 shows the experimental setup for an in-house CL intensity degradation system. The same electron beam that is used to excite the Auger electrons of the phosphor surface is simultaneously been used to excite the phosphor material. Basically, the changes in intensity of the different CL peaks and Auger peak to peak heights (APPH) are monitored and compared with each other. The system consists of an AES vacuum system coupled with a CL spectrometer (Ocean Optics, Inc., Dunedin, FL, USA). The PHI (model 549) Auger spectrometer (Physical Electronics, INC. (PHI), Chanhassen, MN, USA) and Ocean Optics (S2000) spectrometer (Ocean Optics, Inc., Dunedin, FL, USA) are simultaneously used to collect the Auger and CL data, respectively. The light output is collected via a fiber optic connected to the CL spectrometer. The primary AES electron beam current can be adjusted and is typically between 6 and 12 µA. The beam size is varied from 100 to 300 µm depending on the focus of the beam and the beam voltage and beam current. Beam densities vary between 2.5 and 88 mA/cm2 . The Auger and CL data are collected in a vacuum chamber with an initial base pressure in the low 10−9 Torr range, where after the chamber maybe backfilled with O2 , N2 or CO2 up to
Materials 2017, 10, 906
3 of 18
Materials 2017, 10, 906
3 of 17
6 Torr during the degradation study. Throughout the experiment, the Auger and CL pressures 10− Auger andofCL data are recorded using the same adjustable primary electron beam of 1 to 5 keV. The data are recorded using the sameand adjustable primary electronchanges beam of during 1 to 5 keV. The decrease of decrease of the CL intensities the surface chemical prolonged electron the CL intensities the surface is chemical changes during prolonged electron of the bombardment of and the phosphors monitored continuously for periods up tobombardment 24 h at the desired phosphors is monitored continuously for periods up to 24 h at the desired different gas pressures. different gas pressures.
Figure 1. 1. The The CL/AES CL/AES setup Figure setup for for electron electron degradation degradation measurements. measurements.
2.2. XPS 2.2. XPS The surface surface chemical chemical changes changes are then determined determined with use of of XPS. XPS. The The XPS XPS data data are are The are then with the the use collected before and electronic states of collected before and and after after degradation degradationtotoevaluate evaluatethe thechemical chemicalcomposition composition and electronic states the different elements. The data are collected using the PHI 5000 Versa probe-Scanning ESCA of the different elements. The data are collected using the PHI 5000 Versa probe-Scanning ESCA microprobe (Physical (Physical Electronics, Electronics, INC. INC. (PHI), (PHI), Chanhassen, Chanhassen, MN, MN, USA). USA). Low Low energy energy Ar Ar++ ion ion gun gun and and microprobe low energy energy neutralizer neutralizerelectron electrongun gunare areused usedtoto minimize charging surface. Monochromatic low minimize charging onon thethe surface. Monochromatic Al Al Kα radiation (hν = 1486.6 eV) is used as the excitation source. A 25 W, 15 kV electron beam is used Kα radiation (hν = 1486.6 eV) is used as the excitation source. A 25 W, 15 kV electron beam is used to to excite X-ray beam of 100 µm diameter, that is used to analyze the different binding excite thethe X-ray beam of 100 µm diameter, that is used to analyze the different binding energyenergy peaks peaksenergy (pass energy 11 eV, analyzer resolution ≤0.5 eV). Multipak 8.2 software [32]to is analyze used to (pass 11 eV, analyzer resolution ≤0.5 eV). Multipak version version 8.2 software [32] is used analyze the chemical elements and their electronic states using Gaussian-Lorentz fits. The valence the chemical elements and their electronic states using Gaussian-Lorentz fits. The valence state and statepositions and site of positions of theare dopants are also confirmed site the dopants also confirmed with XPS. with XPS. 2.3. TOF-SIMS TOF-SIMS 2.3. Furthermore, TOF-SIMS TOF-SIMS measurements measurements were performed on on aa PHI PHI TRIFT TRIFT V V nanoTOF nanoTOF (Physical (Physical Furthermore, were performed Electronics, INC. (PHI), Chanhassen, MN, USA) or ION TOF-SIMS (Ion-tof GmbH, Münster, Electronics, INC. (PHI), Chanhassen, MN, USA) or ION TOF-SIMS (Ion-tof GmbH, Münster, Germany) + + + + Germany) using Au a pulsed or Bi primary ion beam, to acquire chemical images of the phosphor by using aby pulsed or BiAuprimary ion beam, to acquire chemical images of the phosphor in in both positive and the negative secondaryion ionpolarities. polarities.The Theanalytical analyticalfield-of-view field-of-view is is typically typically both thethe positive and the negative secondary 200 µm µm × × 200 256 pixel pixel × × 256 dose is is maintained maintained 200 200 µm µm with with a a 256 256 pixel pixel digital digital raster, raster, and and the the primary primary ion ion dose well within within the the static static limit limit for for each each analysis. analysis. Charge Charge compensation compensation is is achieved achieved with with aa dual-beam dual-beam well −− +) charge + (≤15 eV e and ≤10 eV Ar neutralizer. A raw data stream file is collected to allow full post(≤15 eV e and ≤10 eV Ar ) charge neutralizer. A raw data stream file is collected to allow full acquisition evaluation (i.e., retrospective analysis) of the data. post-acquisition evaluation (i.e., retrospective analysis) of the data. 2.4. HRTEM HRTEM is performed by using a JEM-2100 electron microscope (JEOL, Tokyo, Japan) at a beam voltage of 200 keV. The EDS (Energy Dispersive X-ray Spectrometer, Oxford Instruments plc, Oxfordshire, UK) mapping and chemical composition analysis is performed with an Oxford XMax
Materials 2017, 10, 906
4 of 18
2.4. HRTEM HRTEM is performed by using a JEM-2100 electron microscope (JEOL, Tokyo, Japan) at a beam voltage of 200 keV. The EDS (Energy Dispersive X-ray Spectrometer, Oxford Instruments plc, Oxfordshire, UK) mapping and chemical composition analysis is performed with an Oxford XMax 80 EDS detector (Oxford Instruments plc, Oxfordshire, UK). The ImageJ software (Laboratory for Optical and Computational Instrumentation, Wisconsin-Madison, WI, USA) is used to determine the size of the nanoparticles (NPs) from the obtained SEM/TEM micrographs. PL is done with a Cary Eclipse (Agilent, Santa Clara, CA, USA) equipped with a 150 W Xenon lamp and a 325 nm He-Cd laser. Raman spectra are obtained by using a NXR FT-Raman module microscopy (Thermo Fisher Scientific Inc, Madison, WI, USA) system with a 2.5 W Nd: YVO4 laser excitation of a 1064 nm wavelength and a high-performance liquid nitrogen-cooled germanium detector. 3. Results and Discussion 3.1. Electron Degradation—AES and XPS 3.1.1. ZnS A typical electron degradation study ends up with: (a) an APPH spectrum as a function of time; and (b) a CL intensity as function of time, as demonstrated for a ZnS phosphor exposed to an electron beam as indicated in Figure 2 [33]. Figure 2a shows the APPHs of the different elements present on the surface during the degradation study, S (152 eV), C (272 eV), O (511 eV) and Zn (994 eV), against electron dose during 2 keV, 64 mA cm−2 electron bombardment at an oxygen pressure of 1 × 10−6 Torr. Oosthuizen et al. [33] found that the C APPH immediately decreased exponentially when the surface was exposed to the electron beam. The C on the phosphor surface was present from adventitious atmospheric contamination. Simultaneous to the time when the C decreased, the CL intensity stayed more or less constant (Figure 2b), which indicated that the C on the surface acted as a protected layer for the CL degradation. The S APPH at first increased due to the removal of the C from the surface and then started to decrease exponentially while the O increased after most of the C was removed from the surface. With the increase in the O and decrease in S APPHs the CL also decreased, which showed a direct correlation between the surface reactions and the CL intensity changes. It was found by Oosthuizen et al. [33] that ZnO has formed on the surface of the ZnS under prolonged electron bombardment. They also measured the presence of SO2 gas during electron bombardment. Itoh et al. [34], using XPS, reported that ZnSO4 was formed on the surface of the ZnS phosphor during electron irradiation in a H2 O ambient. Both ZnO (−94.43 kcal/mol O2 ) and ZnSO4 (72.09 kcal/mol O2 ) have a negative heat of formation [33]. The degraded layer grew in thickness with prolonged electron irradiation time. Chen et al. [35] found a linear growth rate (Figure 3a) for the ZnO layer in a wet oxygen atmosphere and a ZnSO4 formation that decayed exponentially with time and it was postulated that this was due to the diffusion of the charge reactants through the ZnSO4 film to the reaction interfaces, as shown in Figure 3b. During the course of electron irradiation the bombarded areas of the phosphor became darkened in color and degraded in CL efficiency, as pointed out by the Secondary X-ray Imaging in Figure 4a. XPS was used to determine the chemical species in the degraded spot as well as on the undegraded area. A schematic of the degraded spot and growth of the non-luminescent layer with electron bombardment time is shown in Figure 4b. Holloway and Swart [36,37] came up with a well-known ESSCR (electron stimulated surface chemical reaction) mechanism that was applied to the degradation of several phosphors. According to this model, reactive gas molecules adsorb on the surface of the ZnS phosphor and are dissociated to reactive atomic species by the electron beam. This results in the formation of a non-luminescent ZnO/ZnSO4 layer on the surface and volatile SO2 as illustrated in Figure 5.
Materials 2017, 10, 906 Materials 2017, 906 Materials 2017, 10,10, 906 Materials 2017, 10, 906
5 of 18
APPH (arb. Units) APPH (arb. Units) APPH (arb. Units)
Relative CL intensity Relative CLCL intensity Relative intensity
5 of 5 of 1717 5 of 17
2 2) Electron dose (C/cm Electron dose (C/cm ) Electron dose (C/cm2)
2 2) Electron dose (C/cm Electron dose (C/cm ) Electron dose (C/cm2)
Figure 2.The The APPHs;and and(b) (b)CL CLintensity intensityofof ofZnO ZnOduring duringprolonged prolonged electron bombardment Figure (a)(a) APPHs; electron bombardment asasas Figure 2.2.The (a) APPHs; (b) CL intensity ZnO during prolonged electron bombardment Figure 2. of The (a) APPHs; and (b) CL intensity of ZnO during prolonged electron bombardment as function electron dose [11]. functionofofelectron electrondose dose [11]. [11]. function function of electron dose [11].
Figure ZnSO ; and(b) (b)ZnO ZnOlayer layer growth and wet oxygen atmosphere function Figure (a)(a)ZnSO ZnSO 4; ;4and growth inindry and a awet oxygen atmosphere asasfunction ofof Figure 3.3.3. (a) (b) ZnO layer growth indry dry and a wet oxygen atmosphere as function 44; and Figure 3. (a) ZnSO andan (b)illustration ZnO layer growth in dry and athe wetcharge oxygenreactants atmosphere as function of4 degradation time with of the diffusion of during the ZnSO 4 degradation time with an illustration of the diffusion of the charge reactants during the ZnSO of degradation time with an illustration of the diffusion of the charge reactants during the ZnSO 4 4 degradation time with an illustration of the diffusion of the charge reactants during the ZnSO formationasasananinset inset[35]. [35]. formation formation as an inset [35]. formation as an inset [35].
Figure4.4.(a)(a)AAschematic schematicofofthe thedegraded degradedspot spotwith withananenlargement enlargementofofone oneofofthe theparticles particlesshowing showing Figure Figure 4. (a) A schematic of the degraded spot with an enlargement ofone oneofoftime; theparticles particles showingthe Figure 4. (a) A schematic of the degraded spot with an enlargement of the showing the growth of the degraded thin film with prolonged electron bombardment and (b) secondary the growth of the degraded thin film with prolonged electron bombardment time; and (b) secondary the growth the degraded thin film prolonged withspot prolonged electron bombardment (b) secondary X-ray (SXI) a2424H Hdegraded degraded 2SiO 5:Ce. growth ofImaging the of degraded film with electron bombardment time;time; and and (b) secondary X-ray X-ray Imaging (SXI) ofofathin spot ofofY2Y SiO 5:Ce. X-ray Imaging (SXI) of a 24 H degraded spot of Y2SiO5:Ce. Imaging (SXI) of a 24 H degraded spot of Y2 SiO5 :Ce.
Materials 2017, 10, 906 Materials 2017, 10, 906
6 of 18 6 of 17
Figure Figure5. 5. Schematic Schematic illustration illustration of of the the ESSCR ESSCR mechanism. mechanism.
3.1.2. Sr Sr55(PO (PO44))33F:Eu 3.1.2. F:Eu Asmentioned, mentioned,other other phosphors showed the ESSCR mechanism to As phosphors showed thatthat the ESSCR mechanism could could also bealso usedbe to used explain explain the degradation thereof. Figure 6 illustrates the comparative PL emissions of undegraded and the degradation thereof. Figure 6 illustrates the comparative PL emissions of undegraded and degraded degraded CL emission of Srat 5(PO )3F:Eu at an oxygen pressure of [5]. 1 × This 10−6 is Torr [5]. example This is a CL emission spectra of Srspectra an4oxygen pressure of 1 × 10−6 Torr a good 5 (PO4 )3 F:Eu 2+ 3+ 2+ 2+ as well 3+ emission. good example where the CL emission consists of both as well as emission. The where the CL emission spectrum consistsspectrum of both Eu as EuEu TheEu Eu wavelength 2+ wavelength position is matrix sensitive, 3+ emission wavelength position is not 3+ Eu while the Eu position is matrix sensitive, while the Eu emission wavelength position is not affected by a change 2+ the affected by a change The Eu2+ CL as emission as the a broad bandcentred with the in the matrix. The Euin CL matrix. emission appeared a broadappeared band with maxima at maxima 430 nm 6 7 6 7 centred at 430 nm due to the 4f 5d→4f transition and is more prominent compared to the PL emission. due to the 4f 5d→4f transition and is more prominent compared to the PL emission. The prominent 2+ CL emission was obtained due to 6 65d level that can be 2+ prominent The blue that emitting 4fpopulated Eu CL emissionEu was obtained due to the blue emitting 4fthe 5d level can be using high 6 7 transitions of Eu2+ 6 7 2+ populated using such high as energy excitation such [38]. The energy excitation cathode rays [38]. Theas4fcathode 5d→4f rays transitions of 4f Eu5d→4f require assistance from require assistance from of low phonons of the the hostenergy matrix. In addition, energy low frequency phonons thefrequency host matrix. In addition, position of the the 4f6 5d must position be blue 6 5 5 of the 4f must be blue shifted to0 allow population of the D0 state by a direct non-radiative shifted to 5d allow population of the D state by a direct non-radiative relaxation process. The Eu3+ 5 D →7 Fat, 593, 7 5 7 and 578 nm due to relaxation Eu3+toemission 616 nm due to 5D0→7F1, 7F2 transition emission atprocess. 593, 616 The nm due 0 1 F2 transition and 578 nm due to D0 → F0 appeared [39] as a 5D0→7F0 appeared [39] as a mirror image of the PL spectrum. After degradation, it appears that the mirror image of the PL spectrum. After degradation, it appears that the blue emission of the Eu2+ ions 2+ bluecentered emissionatof420 thenm Euand ions was centered at 420and nmshifted and remained shifted ≈10 nm was remained prominent ≈10 nmprominent compared and to the undegraded 3+ decreased 3+ compared to the undegraded sample. The CL intensity of the red emission due to the Eu sample. The CL intensity of the red emission due to the Eu decreased drastically with an increasing 2. According to the ESSCR model, the drastically withof an0increasing in2electron dosetoofthe 0 to 450 C/cm in electron dose to 450 C/cm . According ESSCR model, the Sr–O, Sr–F and P–O bonds are Sr–O,toSr–F and P–O bonds are likely to strontium be broken into free oxygen, fluoride, strontium and likely be broken into free oxygen, fluoride, and phosphorous when irradiated with a beam phosphorous when irradiated with a beam of electron. This will be followed by a chemical reaction of electron. This will be followed by a chemical reaction resulting in new chemical compounds forming resulting in newInchemical compounds on the surface. In most cases, thetherefore new oxide layers are on the surface. most cases, the new forming oxide layers are non-luminescent, and will reduce non-luminescent, and thereforetowill reduce the CL intensity. Simultaneous to the O 2 desorption the CL intensity. Simultaneous the O desorption during CL degradation, it is most likely that P 2 during CL degradation, it is most likely that P (metallic), SrO [40] and P 2 O 5 [41] mixed layers were (metallic), SrO [40] and P2 O5 [41] mixed layers were formed on the surface according to the ESSCR formed on the surface according to thearises ESSCR mechanism. The Eu3+ emission ariseswhich from are the mechanism. The Eu3+ emission mostly from the site occupancy of the Sr2mostly metal sites 3+ site occupancy of the Sr 2 metal sites which are present on the outer surface of the crystal. This caused present on the outer surface of the crystal. This caused the decrease in the Eu CL intensity during 7 5 7 exposure. The3+5D the decrease in the exposure. Eu3+ CL intensity prolonged electron 0-7F0 (578behavior nm) and prolonged electron The 5 Dduring emission 0 - F0 (578 nm) and D0 - F2 (616 nm) Eu 5D0-7F2 (616 nm) Eu3+ emission behavior depends on the site symmetry behavior. 3+ Theemissions slight shift of depends on the site symmetry behavior. The slight shift of the 653 and 705 nm Eu CL arise 3+ CL emissions arise due to crystal ligand field splitting and asymmetrical the 653 and 705 nm Eu due to crystal ligand field splitting and asymmetrical environment of the crystal [39]. Simultaneously, environment of the crystal [39]. Eu2+ near emission from the Eu ion remains which is the Eu2+ emission arises from theSimultaneously, Eu ion which is the present to thearises defect centers, which present near to the centers,remains which remains protected. Thus,electron Eu2+ CLbeam intensity remains protected. Thus, Eu2+defect CL intensity intact during prolonged exposure. Theintact shift during prolonged electron beam exposure. The shift in the peak position, however, indicates a change in the peak position, however, indicates a change in the crystal field due to a change in the host lattice. in the crystal field due to a change in the host lattice.
Materials 2017, 10, 906 Materials 2017, 10, 906 Materials 2017, 10, 906
7 of 18 7 of 177 of 17
Figure 6. The comparative PL and CL output of the Sr5(PO4)3F:Eu as function of Coulomb dose at
Figure 6. The comparative PL and CL output of the Sr5 (PO4 )3 F:Eu as function of Coulomb dose at 1 × 10−6 Torr O2 atmosphere under electron beam excitation with the primary beam voltage and beam 1 ×Figure 10−6current Torr O under beam theasprimary voltage and 6. The PL andelectron CL output of excitation the Sr5(POwith 4)3F:Eu functionbeam of Coulomb dosebeam at of2comparative 2atmosphere keV and 10 µA, respectively [5]. current ofTorr 2 keV 10 µA, respectively [5].beam excitation with the primary beam voltage and beam 1 × 10−6 O2and atmosphere under electron 3.1.3. Yof 2SiO 5:Ce3+ current 2 keV and 10 µA, respectively [5]. 3+ 3.1.3. Y2 SiODuring 5 :Ce the degradation studies of Y2SiO5:Ce3+, an increase in the CL intensity at a wavelength of 3.1.3. 650 Y2SiO 5:Ce3+ nm was measured (Figure 7a) [42]. The XPS (Figure 8a) and CL (Figure 7a) indicated that the During the degradation studies of Y2 SiO5 :Ce3+ , an increase in the CL intensity at a wavelength electron stimulated reaction led to of theYformation of a luminescent silicon dioxide (SiO2) layer on the During degradation studies 2SiO5:Ce3+, an increase in the CL intensity at a wavelength of of 650 nm wasthe measured 7a) [42]. The The XPSfirst (Figure 8a) and CL7a) (Figure 7a) indicated surface of the Y2SiO5(Figure :Ce phosphor powder. spectrum (Figure is characteristic of thethat the 650 nmstimulated was measured (Figure 7a)the [42]. The XPSof (Figure 8a) and CL (Figure 7a) indicated thaton thethe 3+ electron reaction led to formation a luminescent silicon dioxide (SiO ) layer doublet character of the blue light emission from Ce , due to the 4f ground state splitting2[43,44]. electron stimulated reaction led to the formation of a luminescent silicon dioxide (SiO 2) layer on the the fitted resultspowder. from an XPS spectrum for the Si (Figure 2p peak in theisYcharacteristic 2SiO5 state and of the surface Figure of the8aY2shows SiO5 :Ce phosphor The first spectrum 7a) surface of 24 the Ythe 2SiO5:Ce phosphor powder. The first3+ spectrum (Figure 7a) is characteristic of the after h in SiO 2 state. The Si 2p peak shifted and changed shape from the yttrium silicate (Y2SiO5) doublet character of the blue light emission from Ce3+ , due to the 4f ground state splitting [43,44]. doublet character of thebinding blue light emission , due the 4f ground state [43,44]. chemical with 101.3 eVfrom to theCe silica (SiO2to ) chemical state 103.3 eV. splitting The increase Figure 8a showsstate the fitted resultsenergy from an XPS spectrum for the Si 2p peak in the Y SiO 5 state and after Figurein8a thethat fitted an XPSand spectrum for the Sito2pthe peak in the2peak Y2SiO 5 state and CLshows indicates theresults SiO2 is from luminescent thus contributing emission between 24 h in the SiO2700 state. The Sia2p peak shifted and changed shape from the yttrium silicate (Y2 SiO5 ) 600h and nm. SiO2 is phosphor material and thefrom electron irradiation can2SiO after 24 in the SiO 2 state. Thewide Si 2pband peakgap shifted and changed shape the beam yttrium silicate (Y 5) chemical state with binding energy 101.3 eV to the silica (SiO ) chemical state 103.3 eV. The increase break the Si–O andenergy cause intrinsic defects Skuja et2)2al.chemical [46] reported peak chemical state with bonds binding 101.3 eV to the[45]. silica (SiO statetwo 103.3 eV.intensities The increase in in CLCL indicates the SiO luminescent and thus contributing totothe emission peak between for SiO2 atthat 1.9that eV (650 nm) eV (459 nm) with a theory that the two peaks are related to intrinsic 2 is2and indicates the SiO is 2.7 luminescent and thus contributing the emission peak between600 defects involving cleavage of the bonds. Amaterial definite contribution from the SiO2 irradiation 1.9 eV defectcan to break and 700 nm. a wide band gapSi–O phosphor and the electron beam 2 isSiO 600 and 700SiO nm. 2 is a wide band gap phosphor material and the electron beam irradiation can transition from the higher 5d levels to the 4f7/2Skuja levels leads to[46] the increase in the CLpeak intensity and thebreak Si–Othe bonds and cause intrinsic defects [45]. et al. reported two intensities the Si–O bonds and cause intrinsic defects [45]. Skuja et al. [46] reported two peak intensitiesfor peak emission between 600 and 700 nm, thus also resulting in the change in color. The formation of SiO 1.9 eV nm)nm) andand 2.7 2.7 eVeV (459 nm) with a theory that twopeaks peaksare arerelated related intrinsic 2 at for SiO 2 at 1.9(650 eV (650 (459 nm) with theory that the the two toto intrinsic the luminescent SiO2 layer on the surface of the Y2a SiO 5:Ce therefore leads to the degradation of the defects involving cleavage of the Si–O bonds. A definite contribution from the SiO 1.9 eV defect defects involving cleavage of the Si–O bonds. definite contribution from the SiO 2 1.9 eV defect to to 2 blue emitting Y2SiO5:Ce phosphor powders. thethe transition from the higher 5d levels to the 4f levels leads to the increase in the CL intensity and transition from the higher 5d levels to the 4f7/2 7/2 levels leads to the increase in the CL intensity and peak emission between 600 and 700 nm, thus also resultingininthe thechange changeinincolor. color.The Theformation formationofofthe peak emission between 600 and 700 nm, thus also resulting the luminescent SiO2 layer onsurface the surface ofYthe Y 2 SiO 5 :Ce therefore leads to the degradation luminescent SiO2 layer on the of the SiO :Ce therefore leads to the degradation of of thethe blue 2 5 blue emitting Y 2 SiO 5 :Ce phosphor powders. emitting Y2 SiO5 :Ce phosphor powders.
Figure 7. CL intensity against the wavelength for the light emitted from the powders: (a) before and after 3 h and 24 h electron bombardment of Y2SiO5:Ce [42]; and (b) SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a 1 × 10−7 Torr O2 pressure. The digital images were taken before and after degradation.
Figure intensityagainst againstthe thewavelength wavelength for the (a)(a) before and Figure 7. 7. CLCL intensity the light lightemitted emittedfrom fromthe thepowders: powders: before and 2:PbS nanoparticles before and after h and electronbombardment bombardmentof ofYY22SiO55:Ce after 3 h3 and 2424 h helectron :Ce [42]; [42];and and(b) (b)SiO SiO :PbS nanoparticles before and 2 −7 − 7 Torr O 2 pressure. The digital images after degradation [47] with a 2 keV electron beam at a 1 × 10 after degradation [47] with a 2 keV electron beam at a 1 × Torr O2 pressure. The digital images were taken before andafter afterdegradation. degradation. were taken before and
Materials 2017, 10, 906
8 of 18
Materials 2017, 10, 906
8 of 17
Materials 2017, 10, 906
8 of 17
Figure 8. XPS (a) XPS fitted results theSiSi2p 2pininthe theYY2SiO SiO5:Ce after degradation [42] 24 h degradation Figure 8. (a) fitted results forforthe 2 5 :Ce after degradation [42] 24 h degradation −2 electron current density at 26.3 the (b) an oxygen pressure 10−−66Torr 2 ; and in aninoxygen pressure ofof1 1××10 Torrwith withthe the electron current density at mA·cm 26.3 mA; ·and cm−(b) Figure 8. (a) XPS fitted results for the Si 2p in the Y 2 SiO 5 :Ce after degradation [42] 24 h beam at a XPS of SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electrondegradation the XPS of SiO2 :PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a oxygen of 1 × 10−6 Torr with the electron current density at 26.3 mA·cm−2; and (b) the Torr Opressure 2 pressure. 1in×an 10−7 1 × 10−7 Torr O2 pressure. XPS of SiO2:PbS nanoparticles before and after degradation [47] with a 2 keV electron beam at a
−7 Torr O2 pressure. × 102-PbS 3.1.4.1SiO
3.1.4. SiO2 -PbS
Figure 7b presents the broad CL spectra of SiO2; 0.134 mol %-PbS nanoparticles, synthesized 3.1.4. SiO7b 2-PbS Figure presents broad spectra of SiOwith mol %-PbS nanoparticles, synthesized 2 ; 0.134 with a sol-gel method,the before andCL after degradation the maximum intensity at the wavelength withofa sol-gel method, before and after degradation with the maximum intensity at the wavelength Figure 7b presents the broad CL spectra of SiO 2 ; 0.134 mol %-PbS nanoparticles, synthesized about 680 nm [47]. XPS showed that during degradation in an oxygen atmosphere the SiO2 is of with sol-gel before that and of after degradation maximum intensity atthe theSiO wavelength about 680anm showed during degradation inthe anduring oxygenelectron atmosphere is reduced reduced to[47]. SiOmethod, x XPS by the desorption oxygen from thewith surface bombardment the 2and of xabout 680 nmenriched [47]. XPS showed that during degradation in an oxygen thethat SiOsurface 2 is to SiO by the desorption ofinoxygen from surface electron bombardment and the surface was left elemental Si,the as shown induring Figure 8b. Dhlamini etatmosphere al. [47] found the reduced to SiO x by the desorption of oxygen from the surface during electron bombardment and rate of degradation of the CL intensity decreased with an increase in the oxygen pressure. The was left enriched in elemental Si, as shown in Figure 8b. Dhlamini et al. [47] found that the the rate of surface was leftintensity enriched elemental Si, as shown inthe Figure Dhlamini al. [47] SiO found that decrease in isinbest explained in terms formation of a lessetpressure. efficient x (0decrease < x
