- Email: [email protected]
Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device
Accepted Manuscript Title: Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device Authors: Qingjiao Huang, Guobo Dong, Yu Xiao, Xungang Diao PII: DOI: Reference:
S0013-4686(17)31848-0 http://dx.doi.org/10.1016/j.electacta.2017.08.177 EA 30181
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
18-6-2017 16-8-2017 29-8-2017
Please cite this article as: Qingjiao Huang, Guobo Dong, Yu Xiao, Xungang Diao, Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device, Electrochimica Actahttp://dx.doi.org/10.1016/j.electacta.2017.08.177 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device Qingjiao Huang, Guobo Dong, Yu Xiao, Xungang Diao* School of Physics and Nuclear Energy Engineering, Beijing University, Beijing 100191,China *Corresponding author. Tel.: +8601082313931 E-mail address:[email protected]
Abstract All-solid-state inorganic electrochromic devices (ECD) typically suffer from the leakage current which mainly arises from bulk defects and pinholes particularly in the ion conducting layer. The leakage current can lead to rapid self-bleaching of the ECD under open circuit, increased power consumption, and inhomogeneous coloration. Silicon nitride (Si3N4) thin films were prepared by pulsed DC reactive magnetron sputtering, and integrated into a monolithic inorganic ECD to reduce the leakage current. The device was prepared by a series of sputtering processes and the layer stacks were Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO. The optimization of Si3N4 single layer was studied by electrochemical cyclic voltammetry. The effects of leakage current on electrical and optical properties of the ECDs were studied by chronoamperometry and cyclic voltammetry. The leakage current of the device is reduced from 216.0 to 32.1 μA /cm2 with two 80nm-thick Si3N4 layers, and the open circuit memory effect is significantly improved. The optical modulation is 43 % at 550 nm by applied voltages of -2.0 V and 1.5 V. Keywords: Electrochromic device, Monolithic, Silicon nitride, Leakage current, Electrochemical property
1
1. Introduction Electrochromic devices (ECD) have been extensively investigated for many potential applications such as skylights, optical displays, smart energy-efficient windows, and rearview mirrors [1,2]. They have attracted considerable technological and commercial interests due to their unique properties of switching optical transmittance upon the insertion/extraction of small ions and electrons [3-5]. The conventional superimposed ECDs contain a five-layer structure: an ion conducting layer sandwiched by an electrochromic (EC) layer and an ion-storage layer which are individually deposited on transparent electrodes [6-8]. EC layer materials can be divided into two categories: cathodic and anodic coloration materials. Among the inorganic cathodic colored materials, WO3 is considered as the most desirable candidate for the EC layer owing to its high coloration efficiency, large dynamic range, and good cyclic reversibility [9-12]. Anodic colored NiOx is regarded as a promising material complementary to the cathodic colored WO3 [13,14]. Electrochromic device employing an inorganic electrolyte may present considerable advantages compared to ECDs having liquid or polymer gel electrolytes, especially concerning physical and electrical durability [15-17]. However, a major concern with monolithic inorganic ECDs is the inherent leakage current that mainly arises from the pinholes and structural imperfections, particularly in the ion conducting layer[18]. The leakage current usually generates negative effects on electrochromic properties of the ECDs such as a lowered dynamic range, inhomogeneous coloration, decreased ionic conductance, slower switching speed, and increased power consumption. It is reported that the yearly energy consumption can be estimated to be 4 KWh/m2 for a leakage current of 100 μA/cm2 [19,20]. Leakage current can reduced by increasing the thickness of ion conducting layer [21]. The tantalum oxide (Ta2O5) thin films of 300nm thick as electrolyte to improve the leakage current has been examined by Wang et al. [15]. Unfortunately, it is expense of degraded optical properties, increased layer deposition time and cost. However, the Si3N4 thin film of only 80nm thick as a electron blocking layer in ECDs with I-/I3- and TMTU/TMFDS2+ based redox electrolytes reduced the leakage current from 240 to 20
2
μA/cm2,which has been studied by Bogati et al. [19]. The aim of this work is to present some results of applying Si3N4 as electron blocking layers for all-solid-state inorganic electrochromic device. The Li ion conducting properties of Si3N4 thin films influenced by various sputtering conditions were investigated by cyclic voltammetry. Electro-optical characterizations were carried out to evaluate the performance of the ECDs with and without the Si3N4 layers. The effects of leakage current on the electrical and optical properties of the ECDs were studied in detail.
2. Experimental
2.1 Deposition of Si3N4 thin films The Si3N4 thin films were deposited on WO3/ITO-coated glass substrates by pulsed DC reactive magnetron sputtering in a mixture gas of argon and nitrogen. Here all these WO3 substrates were prepared at the same sputtering conditions. The target was 100 mm in diameter SiAl plate with a thickness of 6 mm. Prior to the deposition, the ITO-coated substrates were ultrasonically cleaned in anhydrous ethanol for 15 min. The distance between target and substrate was 15 cm. The base pressure was 1×10-3 Pa evacuated by a turbo molecular pump combined with a rotary pump. The SiAl target was pre-sputtered in pure Ar atmosphere for 5 min to remove surface contaminant. The gas flows of Ar (99.99 %) and N2 (99.99 %) were adjusted individually by mass flow controllers. The nitrogen partial pressure ratio was defined as PN2=P(N2)/[P(N2)+P(Ar)], where P(N2) and P(Ar) were the gas pressures of the N2 and Ar, respectively. PN2 was varied from 10 % to 40 %. The working pressure was adjusted from 0.3 to 2.0 Pa. During the deposition process, the pulsed DC power was maintained at 275 W with a pulse duty ratio of 70 %. The deposition time was controlled in order to prepare films with thickness of 40, 80, 120, 260, and 320 nm. All the samples were deposited at room temperature. 2.2 Preparation of all-solid-state inorganic electrochromic devices Inorganic all-solid-state ECDs based on complementary WO3 and NiO
3
electrochromic active electrodes were sputtered layer by layer on ITO-coated glass substrates, and the ECDs with and without Si3N4 electron blocking layers were prepared.
The
multilayer
structures
can
be
depicted
as
Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO and Glass/ITO/NiO/LiNbO3/WO3/ITO, respectively. Detailed experimental parameters of the ECDs are listed in Table 1. It is deserved to mention that the ion conducting layer was prepared by DC reactive magnetron sputtering from an oxygen-deficient LiNbO3 conductive target. The application of DC sputtering power can effectively increase the deposition rate compared with RF sputtering power, more importantly, make it possible for industrial large scale production.
Table 1 Deposition parameters of the electrochromic devices Target
Power source
Pressure (Pa)
Ar:O2:N2 (sccm)
Power (W)
Sputtering time (min)
Thickness (nm)
Ni SiAl1 LiNbO3 W ITO
DC Pulsed DC DC DC DC
3 0.5 0.8 2.2 0.3
94:6:0 40:0:10 95:5:0 27:9:0 78.4:1.6:0
236 275 152 303 180
30 6 60 10 20
320 80 130 330 250
1
Composition of the SiAl alloy target: 90 wt.% Si, 10 wt.% Al.
2.3 Characterization Film thickness was determined by surface profilometry using a Dektak instrument. The surface roughness of films was measured using a Veeco Dimension 3100 instrument employed in tapping mode. Film structures were determined by X-ray diffraction (XRD) using a Rigaku D/MAX-2500/PC diffractometer with a Cu Kα source. X-ray photoelectron spectroscopy (XPS) measurements were carried out in a VG Scientific ESCALab Mark II photoelectron spectrometer. Peak shifts due to charging were normalized by fixing the C 1s peak to 284.6 eV. The cross-sectional image of the ECD was obtained by scanning electron microscopy (SEM) using a XL30 S-FEG from FEI. The Si3N4 thin films deposited on WO3/ITO/glass substrates
4
were subjected to the cyclic voltammetry (CV) test. A three electrode cell configuration was set up containing 1M LiClO4-PC solution as the electrolyte, ITO as a working electrode, Ag/AgCl as a reference electrode, and platinum foil as a counter electrode. The measurements were carried out using a CHI660E electrochemical workstation. Cyclic voltammograms were acquired at a scan rate of 20 mV/s with sweeping from -1.0 V to 1.0 V. For the complete device, the CV cycles were performed at a scan rate of 50 mV/s between -2.0 V and 1.5 V. Step chronoamperometry (CA) cycles were conducted under -2.0 V for coloration and 1.5 V for bleaching. The duration of each step was 30 s. The optical transmittance spectra of the ECDs were measured in situ by a Hitachi U-3010 UV-Vis spectrophotometer at a wavelength of 550 nm. The transmittance spectra of the device at its original, colored, and bleached states were measured over a wavelength of 300-800 nm. 3. Results and discussion 3.1 Characterization of Si3N4 thin films
5
Fig. 1 The three-dimensional AFM images for Si3N4 films deposited at various N2 contents: (a) 10% (b) 20% (c) 30% and (d) 40%.
Fig. 1 illustrates characteristic topographies for Si3N4 films prepared at various N2 contents. The root mean square roughness (Rq) decrease first and then increase in general as the N2 contents . The values of Rq are respectively 3.17, 2.70, 2.74, and 3.29 nm, corresponding to N2 content of 10%, 20%, 30%, and 40%. The minimum value appears when the N2 content equals 20%. The as-deposited Si3N4 films are smooth which would facilitate Li ion transportation and film growth for top layers of the ECD. The Li ion conducting properties of Si3N4 thin films were evaluated by cyclic voltammetry (CV) test conducted in 1M anhydrous LiClO4-PC solution. The Si3N4 thin films prepared at various parameters were deposited on WO3/ITO/glass substrates. The considered sputtering conditions were N2 content, working pressure, and film thickness. The WO3 substrates were prepared at the same parameters, which were shown in Table 1. CV curves showed the intrinsic electrochemical behavior of WO3. By coating Si3N4 thin films prepared at different parameters, the peak current density and charge density of the CV curves changed. The charge density (area of the CV curve) reflects the optical contrast of the WO3 thin film, and the cathodic (anodic) peak current density is related with the Li ion intercalation (deintercalation) diffusion coefficient, which only depends on the Li ion conductivity of the Si3N4 thin film due to the same WO3 electrochromic active layer. In this work, the diffusion coefficients of Li ions were calculated by using Randles-Sevick equation [22,23]: 3
1
1
i p 2.72 10 n 2 D 2 C0 2 (1) 5
where D is the diffusion coefficient, C0 is the concentration of active ions in the solution, is the scan rate, n is the number of electrons and is assumed to be 1, i p is the peak current density.
6
Fig. 2 (a), (c) and (e) show the cyclic voltammograms of the Si3N4-coated WO3/ITO/glass deposited at various N2 contents, working pressures, and different film thicknesses, respectively. Only the first CV cycle of each sample is displayed. The corresponding quantification results of Li ion diffusion coefficients and charge densities are presented in (b), (d) and (f), respectively. The cyclic voltammetry measurements were carried out in 1M LiClO4-PC electrolyte at a scan rate of 20 mV/s sweeping between -1 V and 1 V.
Fig. 2(a,c,e) display the cyclic voltammograms of the Si3N4-coated WO3 thin films for the first cycle. The curves exhibit oxidation and reduction peaks of WO3 films, which can be interpreted as Li ion insertion/extraction through the Si3N4 layer. The corresponding quantification results of Li diffusion coefficient and charge density are shown in Fig. 2(b,d,f). The silicon nitride prepared by different parameters will 7
have a little influence on the diffusion of lithium ions into tungsten oxide. The minimal effect are the optimal preparation parameters. It is found that the current density and charge density are mainly influenced by the N2 content and film thickness (see Fig. 2(a,e)). While the impact of working pressure on Li ion conductivity of Si3N4 film can be neglected. The optimal value for N2 content appears at 20 % (Fig. 2(b)). As expected, the peak current density and transferred charge in WO3 films decrease gradually with the increasing thickness of Si3N4 cover layer (Fig. 2(e,f)). When the thickness of Si3N4 is less than 120 nm, the influence of Si3N4 buffer layer on electrochromic performance of WO3 can be regarded as negligible. Fig. 3(a) shows 100 CV cycles of a 100nm-thick Si3N4 covered WO3 in 1M LiClO4-PC. The poor cycling durability of bare WO3 was mainly caused by the residual water in the solution [24,25]. The corresponding charge density of each cycle is shown in Fig. 3(b). It is demonstrated that the Si3N4 cover layer can prevent WO3 from water contacting, thereby improving the cycling durability. Moreover, the Li ion transportation through Si3N4 film is stable.
Fig. 3 (a) The cyclic voltammograms of 1st and 100th cycles for bare WO3 and Si3N4-covered WO3 in 1M LiClO4-PC. (b) The corresponding caculated charge density of each cycle.
8
3.2 Characterization of complete electrochromic devices
Fig. 4 (a) and (b) respectively show the schematic diagram of the ECD without and with Si3N4 electron blocking layers
Fig. 4 shows the structure of devices with and without a silicon nitride film. The arrows indicate the directions of movement of lithium ions and electrons during the coloring process. A complementary ECD typically consist of two electrochromic active layers, an ion conductor layer, and two transparent electrodes. Up to now, WO3 is the best known and extensively used EC material due to its high coloration efficiency, large dynamic range, and good cyclic reversibility [26-28]. Corresponding to cathodic colored WO3, anodic colored NiO has been selected to be a complementary EC layer due to their similar coloration efficiency and color neutrality [29]. LiNbO3 functions as an ion conductor layer which provides Li+ ions and transportation channels between WO3 and NiO. Compared with conventional device configuration, two Si3N4 layers are integrated into the device in order to block the transportation of electrons across the LiNbO3 layer. XPS and XRD characterizations on Si3N4, WO3, LiNbO3, and NiO layers are provided in the supplementary figures. The Si 2p and N 1s binding energies were respectively found to be 102.3 eV and 398.2 eV (see Fig. S1), which correspond to Si3N4 [30,31]. The W 4f core level of the WO3 film exhibited peaks located at 35.8 eV and 37.9 eV (Fig. S2), which areattributed to the W6+ states [32]. The binding energies of Nb 3d were 207.2 eV with a splitting of 2.8 eV (Fig. S3) [33]. The Ni 2p3/2 peak of NiO displayed a lower energy at 853.9 eV, which is attributed to the Ni2+ species and higher binding energy (855.6 eV) arising from the Ni3+ oxidation state (Fig. S4) [34,35]. The XRD patterns
9
indicated that the Si3N4, WO3, and LiNbO3 were amorphous, whereas the NiO exhibited a face-centered cubic NaCl-type structure. The (111), (200), (220), (311), and (222) peaks typical to NiO (JCPDS Card No. 47-1049) were clearly identified.
Fig. 5 SEM cross-sectional image of the electrochromic device.
Two types of complete electrochromic devices (ECD) were fabricated in this work, i.e., ECD without Si3N4 buffer layers and ECD with Si3N4 buffer layers. The ECDs were based on the WO3 and NiO complementary electrochromic active layers. LiNbO3 thin films were used as the inorganic Li ion conductor. Fig. 5 exhibits a cross-sectional SEM image of the ECD having two Si3N4 buffer layers on both sides of LiNbO3. The thickness of each Si3N4 layer is approximately 80 nm. The interfaces are clearly identified indicating that all the layers are physically and chemically stable without interdiffusion in the preparation process. The smooth interface between Si3N4 and LiNbO3 is favorable for Li ion transportation. The electrochromic performances of the ECDs were characterized by step chronoamperometry (CA) and cyclic voltammetry (CV). The electrical data of 100 CA cycles (E1 = -2.0 V, E2 = 1.5 V) for the two ECDs were measured. According to O’Brien et al. [18], leakage current is defined as the steady state current at which the rate of color change becomes zero. It is obvious that the ECD having Si3N4 buffer layers exhibits lower leakage currents. This is associated with the internal leakage current, which will lead to electronic breakdown of the device under high voltage. Fig. 6 shows an amplification of the last 5 cycles of CA curves, which indicates that the ECD without buffer layers at colored state shows an increasing current with the time 10
increasing. While the ECD with Si3N4 buffer layers shows a steady current close to zero at its fully colored state. The steady state currents of the device at fully colored and bleached states are both reduced by Si3N4 buffer layers. Especially for the colored state leakage current, the value is significantly decreased from 216.0 to 32.1 μA /cm2.
Fig. 6 Step chronoamperometric curves for comparison of the devices with and without Si3N4 buffer layers.
Fig. 7 In situ transmittance variation at 550 nm plotted against time during 100 chronoamperometric cycles for the device with Si3N4 buffer layers. The applied voltages were -2.0 V for coloration and 1.5 V for bleaching. The duration was 30 s for each step.
Fig. 7 shows the in situ measurement of transmittance at 550 nm for the device with Si3N4 buffer layers during 100 CA cycles. The optical modulation is around 43 % from a colored state of 27 % to a bleached state of 70 %. The device shows a stable and reversible optical contrast. It is proved that the electrochromic performance and cycling stability of the device are not compromised by introducing the Si3N4 electron 11
blocking layers.
Fig. 8 Cyclic voltammograms of the device with and without Si3N4 buffer layers. The scan rate was 50 mV/s. The arrows refer to the sweeping directions. The inset figure shows the in situ transmittance change at 550 nm of the device during the CV cycles.
Fig. 8 shows the cyclic voltammograms (CV) of the two ECDs for the 20th cycle. The CV curves of the ECD without Si3N4 buffer layers are inclined and cross due to the poor dielectric property of the ion conductor layer. The perfect CV curves can be obtained from the device containing Si3N4 layers. When there is no electrochemical reaction, the corresponding values of current density are nearly zero. This is in accordance with the results of CA test that Si3N4 buffer layers can effectively reduce the leakage current in the full device. More importantly, the optical contrast and the response time of the device are hardly influenced (see the inset in Fig. 8). This implicates that the Li ions transportation through the Si3N4 layer is fast and stable.
Fig. 9 Optical transmittance spectra of the device containing Si3N4 buffer layers at its original, colored, and bleached states.
12
Fig. 9 shows the transmittance spectra of the ECD containing Si3N4 buffer layers at its original, colored, and bleached states. The device shows an optical modulation of around 40 % in the visible region. The optical absorption near 400 nm is mainly caused by the NiO layer and top ITO layer in the device. The device was bleached to light brown under a voltage of 1.5 V and changed to a neutral dark color under a voltage of 2.0 V.
Fig. 10 Open circuit memory test of the devices with and without Si3N4 buffer layers. The devices were both colored to a transmittance of 25 %.
Open circuit memory effect is a crucial factor in the assessment of electrochromic properties for ECDs concerning energy consumption [36,37].The device requires a low leakage current for a better open circuit memory (i.e., slower self-bleaching under open circuit), particularly with respect to the application when batteries are the main source of power for the ECDs. O’Brien et al. [18] reported that a 25 cm2 EC device has a leakage current of 1-4 mA corresponding to an open circuit memory of about 60 min before %T color may change from 8 to 20 %. In the present study, ECDs with and without Si3N4 buffer layers were subjected to a memory test. The two devices were colored to a same level (25 % at 550 nm) prior to disconnecting the external power. Fig. 10 shows the self-bleaching transmittance of the two devices. It is observed the ECD without Si3N4 buffer layers exhibits a rapid self-bleaching curve. While it takes 4000s for the ECD containing Si3N4 buffer layers, and the transmittance increases by 20 %. As discussed above, the leakage current for the device with Si3N4 buffer layers
13
is 32.1 μA/cm2. These results are in good agreement with the values reported by O’Brien et al.. The leakage current is dramatically reduced by the addition of Si3N4 layers, which leads to a better memory effect of the device. A device with a better open circuit memory can effectively reduce energy consumption in application.
4. Conclusions
Si3N4 thin films were deposited by pulsed DC reactive magnetron sputtering in a mixture of Ar and N2 atmosphere. They were used as the electron blocking layers to reduce the leakage current of electrochromic device. The Li ions conducting properties of the Si3N4 thin films were evaluated by electrochemical cyclic voltammetry. The optimal deposition conditions for Si3N4 layer were 20 % of N2 partial pressure, 275 W of sputtering power, and 0.5 Pa of working pressure. A multilayered
inorganic
all-solid-state
ECD
composed
of
Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO was fabricated by sputtering. It was observed that leakage current of the device at colored state was reduced from 216.0 to 32.1 μA /cm2 with two 80nm-thick Si3N4 buffer layers added on both sides of the ion conductor. The corresponding self-bleaching time under open circuit was extended to 4000 s as the transmittance of the colored ECD increased from 25 % to 45 %. Both chronoamperometric cycles and voltammetric cycles indicated that the Li ions transportation through the Si3N4 layer was fast and stable. The optical modulation of the ECD was 43 % at 550 nm under -2.0 V (coloration) and 1.5 V (bleaching). The switching time was within 30 s.
Acknowledgements This work has been financially supported by the National Program on Key Research Project of China (2016YFB0303901) and the Beijing Natural Science Foundation (2161001) and the Fundamental Research Funds for the Central Universities (Grant No. YWF-16-JCTD-B-03).
14
References [1] C.G. Granqvist, Handbook of Inorganic Electrochromic Materials, Elsevier, 1995. [2] R.T. Wen, G.A. Niklasson, C.G. Granqvist, Electrochromic iridium oxide films:compatibility with propionic acid, potassium hydroxide, and lithium perchlorate in propylene carbonate, Sol. Energy Mater. Sol. Cells 120 (2014) 151–156. [3] D.M. Dong, W.W. Wang, G.B. Dong, F. Zhang, Y.C. He, H. Yu, F. M. Liu, M. Wang, X.G. Diao, Electrochromic properties and performance of NiOx films and their Corresponding all-thin-film flexible devices prepared by reactive DC magnetron sputtering, Appl. Surf. Sci. 383 (2016) 49–56. [4] C.C. Liao, Lithium-driven electrochromic properties of electrodeposited nickelhydroxide electrodes, Sol. Energy Mater. Sol. Cells 99 (2012) 26–30. [5] H. Huang, J. Tian, W.K. Zhang, Y.P. Gan, X.Y. Tao, X.H. Xia, J.P. Tu, Electrochromic properties of porous NiO thin film as a counter electrode for NiO/WO3 complementary electrochromic window, Electrochim. Acta 56 (2011) 4281–4286. [6] S.V. Green, E. Pehlivan, C.G. Granqvist, G.A. Niklasson, Electrochromism in sputter deposited nickel-containing tungsten oxide films, Sol. Energy Mater. Sol. Cells 99 (2012) 339–344. [7] C.G. Granqvist, A. Azens, J. Isidorsson, M. Kharrazi, L. Kullman, T. Lindstrom, G.A. Niklasson, C.-G. Ribbing, D. Ronnow, M.Stromme Mattson, M. Veszelei, Towards the smart window: progress in electrochromics, J. Non-Cryst. Solids 218 (1997) 273–279. [8] Q. Liu, G. Dong, Y. Xiao, F. Gao, M. Wang, Q. Wang, S. Wang, H. Zuo, X. Diao, An all-thin-film inorganic electrochromic device monolithically fabricated on flexible PET/ITO substrate by magnetron sputtering, Mater. Lett. 142 (2015) 232–234. [9] D. Zhou, F. Shi, D. Xie, D.H. Wang, X.H. Xia, X.L. Wang, C.D. Gu, J.P. Tu, Bi-functional Mo-doped WO3 nanowire array electrochromism-plus electrochemical energy storage, J. Colloid Interf. Sci. 465 (2016) 112–120. [10] M. Wang, G. Fang, L. Yuan, H. Huang, Z. Sun, N. Liu, S. Xia, X. Zhao, High
15
optical switching speed and flexible electrochromic display based on WO3 nanoparticles with ZnO nanorod arrays' supported electrode, Nanotechnology 20 (2009) 185304. [11] K. Wang, H. Wu, Y. Meng, Y. Zhang, Z. Wei, Integrated energy storage and electrochromic function in one flexible device: an energy storage smart window, Energ. Environ. Sci. 5 (2012) 8384. [12] S.H. Lee, R. Deshpande, P.A. Parilla, K.M. Jones, B. To, A.H. Mahan, A.C. Dillon, Crystalline WO3 Nanoparticles for Highly Improved Electrochromic Applications, Adv. Mater. 18 (2006) 763–766. [13] S. Pereira, A. Gonçalves, N. Correia, J. Pinto, L. Pereira, R. Martins, E. Fortunato, Electrochromic behavior of NiO thin films deposited by e-beam evaporation at room temperature, Sol. Energy Mater. Sol. Cells 120 (2014) 109–115. [15] S.C. Wang, K.Y. Liu, J.L. Huang, Tantalum oxide film prepared by reactive magnetron sputtering deposition for all-solid-state electrochromic device, Thin Solid Films 520 (2011) 1454–1459 [16] R.B. Goldner, F.O. Arntz, G .Berera, T.E. Haas, G. Wei, K.K. Wong, P.C. Yu, A monolithic thin-film electrochromic window. Solid State Ionics 53-56 (1992) 617– 627. [17] K.J. Patel , G.G. Bhatt, J.R. Ray, P. Suryavanshi, C.J. Panchal, All-inorganic solid-state electrochromic devices: a review, J. Solid State Electrochem. 21 (2017) 337–347 [18] N.A. O’Brien, J. Gordon, H. Mathew, B.P. Hichwa, Electrochromic coatings– applications and manufacturing issues, Thin Solid Films 345 (1999) 312–318 [19] S. Bogati, A. Georgn, W. Graf, Sputtered Si3N4 and SiO2 electron barrier layer between a redox electrolyte and the WO3 film in electrochromic devices, Sol. Energy Mater. Sol. Cells 159 (2017) 395–404 [20] P.F. Tavares, A.R. Gaspar, A.G. Martins, F. Frontini, Evaluation of electrochromic windows impact in the energy performance of buildings in Mediterranean climates, Energ. Policy 67 (2014) 68–81.
16
[21] M.S. Burdis, B. D.Greer, D.G. J.Weir, US patent 8,004,744 B2: Electrochromic devices having improved ion conducting layers, 2011. [22] P.S. Patil, P.R .patil, S.S. Kamble, S. H. Pawar, Thickness-dependent electrochromic properties of solution thermolyzed tungsten oxide films, Sol. Energy Mater. Sol. Cells 60 (2000) 143–153. [23] R. Sivakumar, R. Gopalakrishman, M. Jayachandran, C. Sanjeeviraja, Investigation of X-ray photoelectron spectroscopic (XPS), cyclic voltammetric analyses of WO3 films and their electrochromic response in FTO/ WO3/electrolyte/FTO cells, Smart Mater. Struct. 15 (2006) 882. [24] D. Chatzikyriakou, N. Krins, B. Gilbert, P. Colson, J. Dewalque, J. Denayer, R. Cloots, C. Henrist, Mesoporous amorphous tungsten oxide electrochromic films: a Raman analysis of their good switching behavior, Electrochim. Acta 137 (2014) 75– 82 [25] G. Leftheriotis, S. Papaefthimiou, P. Yianoulis, The effect of water on the electrochromic properties of WO3 films prepared by vacuum and chemical methods, Sol. Energy Mater. Sol. Cells 83 (2004) 115–124. [26] S.K. Deb, Opportunities and challenges in science and technology of WO3 for electrochromic and related applications, Sol. Energy Mater. Sol. Cells 92 (2008) 245– 258. [27] C.G. Granqvist, Electrochromic tungsten oxide films: review of progress 1993-1998, Sol. Energy Mater. Sol. Cells 60 (2000) 201–262. [28] G.A. Niklasson, C.G. Granqvist, Electrochromics for smart windows: thin films of tungsten oxide and nickel oxide, and devices based on these, J. Mater. Chem. 17 (2007) 127–156. [29] H. Moulki, D.H. Park, B.K. Min, H. Kwon, S.J. Hwang, J.H. Choy, T. Toupance, G. Campet, A. Rougier, Improved electrochromic performances of NiO based thin films by lithium addition: From single layers to devices, Electrochimi. Acta 74 (2012) 46–49. [30] S.L. Ku, C.C. Lee, Optical and structure properties of silicon nitride thin films
17
prepared by ion assisted deposition, Opt. Mater. 32 (2010) 956–960. [31] S.U. Han, H.S. Kang, B.K. Kang, Time-dependent degradation due to negative bias temperature instability of p-MOSFET with an ultra-thin SiON gate dielectric, Microelectron. Eng. 83 (2006) 520–527. [32] I.T.S. Garcia, D.S. Corrêa, D.S.D. Moura, J.C.O. Pazinato, M.B. Pereira, N.B.D. da Costa, Multifaceted tungsten oxide films grown by thermal evaporation, Surf. Coat. Technol. 283 (2015) 177–183. [33] M. Gellert, K.I. Gries, J. Sann, E. Pfeifer, K. Volz, B. Roling, Impedance spectroscopic study of the charge transfer resistance at the interface between a LiNi0.5Mn1.5O4 high-voltage cathode film and a LiNbO3 coating film, Solid State Ionics 287 (2016) 8–12. [34] S.V. Green, C.G. Granqvist, G.A. Niklasson, Structure and optical properties of electrochromic tungsten-containing nickel oxide films, Sol. Energy Mater. Sol. Cells 126 (2014) 248–259. [35] F. Lin, D. Nordlund, T.-C. Weng, R.G. Moore, D.T. Gillaspie, A.C. Dillon, R.M. Richards, C. Engtrakul, Hole doping in Al-containing nickel oxide materials to improve electrochromic performance, ACS Appl. Mater. Inter. 5 (2013) 301–309. [36] Q.R. Liu, G.B. Dong, Y. Xiao, M.P. Delplancke-Ogletree , F. Reniers , X.G. Diao, Electrolytes-relevant cyclic durability of nickel oxide thin films as anion-storage layer in an all-solid-state complementary electrochromicdevice, Sol. Energy Mater. Sol. Cells 157 (2016) 844–852 [37] K.J. Patel, M.S. Desai, C.J. Panchal, Studies of ZrO2 electrolyte thin-film thickness on the all-solid thin-film electrochromic devices, J. Solid State Electrochem.19 (2015) 275–279.
18
S0013-4686(17)31848-0 http://dx.doi.org/10.1016/j.electacta.2017.08.177 EA 30181
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
18-6-2017 16-8-2017 29-8-2017
Please cite this article as: Qingjiao Huang, Guobo Dong, Yu Xiao, Xungang Diao, Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device, Electrochimica Actahttp://dx.doi.org/10.1016/j.electacta.2017.08.177 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Electrochemical studies of silicon nitride electron blocking layer for all-solid-state inorganic electrochromic device Qingjiao Huang, Guobo Dong, Yu Xiao, Xungang Diao* School of Physics and Nuclear Energy Engineering, Beijing University, Beijing 100191,China *Corresponding author. Tel.: +8601082313931 E-mail address:[email protected]
Abstract All-solid-state inorganic electrochromic devices (ECD) typically suffer from the leakage current which mainly arises from bulk defects and pinholes particularly in the ion conducting layer. The leakage current can lead to rapid self-bleaching of the ECD under open circuit, increased power consumption, and inhomogeneous coloration. Silicon nitride (Si3N4) thin films were prepared by pulsed DC reactive magnetron sputtering, and integrated into a monolithic inorganic ECD to reduce the leakage current. The device was prepared by a series of sputtering processes and the layer stacks were Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO. The optimization of Si3N4 single layer was studied by electrochemical cyclic voltammetry. The effects of leakage current on electrical and optical properties of the ECDs were studied by chronoamperometry and cyclic voltammetry. The leakage current of the device is reduced from 216.0 to 32.1 μA /cm2 with two 80nm-thick Si3N4 layers, and the open circuit memory effect is significantly improved. The optical modulation is 43 % at 550 nm by applied voltages of -2.0 V and 1.5 V. Keywords: Electrochromic device, Monolithic, Silicon nitride, Leakage current, Electrochemical property
1
1. Introduction Electrochromic devices (ECD) have been extensively investigated for many potential applications such as skylights, optical displays, smart energy-efficient windows, and rearview mirrors [1,2]. They have attracted considerable technological and commercial interests due to their unique properties of switching optical transmittance upon the insertion/extraction of small ions and electrons [3-5]. The conventional superimposed ECDs contain a five-layer structure: an ion conducting layer sandwiched by an electrochromic (EC) layer and an ion-storage layer which are individually deposited on transparent electrodes [6-8]. EC layer materials can be divided into two categories: cathodic and anodic coloration materials. Among the inorganic cathodic colored materials, WO3 is considered as the most desirable candidate for the EC layer owing to its high coloration efficiency, large dynamic range, and good cyclic reversibility [9-12]. Anodic colored NiOx is regarded as a promising material complementary to the cathodic colored WO3 [13,14]. Electrochromic device employing an inorganic electrolyte may present considerable advantages compared to ECDs having liquid or polymer gel electrolytes, especially concerning physical and electrical durability [15-17]. However, a major concern with monolithic inorganic ECDs is the inherent leakage current that mainly arises from the pinholes and structural imperfections, particularly in the ion conducting layer[18]. The leakage current usually generates negative effects on electrochromic properties of the ECDs such as a lowered dynamic range, inhomogeneous coloration, decreased ionic conductance, slower switching speed, and increased power consumption. It is reported that the yearly energy consumption can be estimated to be 4 KWh/m2 for a leakage current of 100 μA/cm2 [19,20]. Leakage current can reduced by increasing the thickness of ion conducting layer [21]. The tantalum oxide (Ta2O5) thin films of 300nm thick as electrolyte to improve the leakage current has been examined by Wang et al. [15]. Unfortunately, it is expense of degraded optical properties, increased layer deposition time and cost. However, the Si3N4 thin film of only 80nm thick as a electron blocking layer in ECDs with I-/I3- and TMTU/TMFDS2+ based redox electrolytes reduced the leakage current from 240 to 20
2
μA/cm2,which has been studied by Bogati et al. [19]. The aim of this work is to present some results of applying Si3N4 as electron blocking layers for all-solid-state inorganic electrochromic device. The Li ion conducting properties of Si3N4 thin films influenced by various sputtering conditions were investigated by cyclic voltammetry. Electro-optical characterizations were carried out to evaluate the performance of the ECDs with and without the Si3N4 layers. The effects of leakage current on the electrical and optical properties of the ECDs were studied in detail.
2. Experimental
2.1 Deposition of Si3N4 thin films The Si3N4 thin films were deposited on WO3/ITO-coated glass substrates by pulsed DC reactive magnetron sputtering in a mixture gas of argon and nitrogen. Here all these WO3 substrates were prepared at the same sputtering conditions. The target was 100 mm in diameter SiAl plate with a thickness of 6 mm. Prior to the deposition, the ITO-coated substrates were ultrasonically cleaned in anhydrous ethanol for 15 min. The distance between target and substrate was 15 cm. The base pressure was 1×10-3 Pa evacuated by a turbo molecular pump combined with a rotary pump. The SiAl target was pre-sputtered in pure Ar atmosphere for 5 min to remove surface contaminant. The gas flows of Ar (99.99 %) and N2 (99.99 %) were adjusted individually by mass flow controllers. The nitrogen partial pressure ratio was defined as PN2=P(N2)/[P(N2)+P(Ar)], where P(N2) and P(Ar) were the gas pressures of the N2 and Ar, respectively. PN2 was varied from 10 % to 40 %. The working pressure was adjusted from 0.3 to 2.0 Pa. During the deposition process, the pulsed DC power was maintained at 275 W with a pulse duty ratio of 70 %. The deposition time was controlled in order to prepare films with thickness of 40, 80, 120, 260, and 320 nm. All the samples were deposited at room temperature. 2.2 Preparation of all-solid-state inorganic electrochromic devices Inorganic all-solid-state ECDs based on complementary WO3 and NiO
3
electrochromic active electrodes were sputtered layer by layer on ITO-coated glass substrates, and the ECDs with and without Si3N4 electron blocking layers were prepared.
The
multilayer
structures
can
be
depicted
as
Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO and Glass/ITO/NiO/LiNbO3/WO3/ITO, respectively. Detailed experimental parameters of the ECDs are listed in Table 1. It is deserved to mention that the ion conducting layer was prepared by DC reactive magnetron sputtering from an oxygen-deficient LiNbO3 conductive target. The application of DC sputtering power can effectively increase the deposition rate compared with RF sputtering power, more importantly, make it possible for industrial large scale production.
Table 1 Deposition parameters of the electrochromic devices Target
Power source
Pressure (Pa)
Ar:O2:N2 (sccm)
Power (W)
Sputtering time (min)
Thickness (nm)
Ni SiAl1 LiNbO3 W ITO
DC Pulsed DC DC DC DC
3 0.5 0.8 2.2 0.3
94:6:0 40:0:10 95:5:0 27:9:0 78.4:1.6:0
236 275 152 303 180
30 6 60 10 20
320 80 130 330 250
1
Composition of the SiAl alloy target: 90 wt.% Si, 10 wt.% Al.
2.3 Characterization Film thickness was determined by surface profilometry using a Dektak instrument. The surface roughness of films was measured using a Veeco Dimension 3100 instrument employed in tapping mode. Film structures were determined by X-ray diffraction (XRD) using a Rigaku D/MAX-2500/PC diffractometer with a Cu Kα source. X-ray photoelectron spectroscopy (XPS) measurements were carried out in a VG Scientific ESCALab Mark II photoelectron spectrometer. Peak shifts due to charging were normalized by fixing the C 1s peak to 284.6 eV. The cross-sectional image of the ECD was obtained by scanning electron microscopy (SEM) using a XL30 S-FEG from FEI. The Si3N4 thin films deposited on WO3/ITO/glass substrates
4
were subjected to the cyclic voltammetry (CV) test. A three electrode cell configuration was set up containing 1M LiClO4-PC solution as the electrolyte, ITO as a working electrode, Ag/AgCl as a reference electrode, and platinum foil as a counter electrode. The measurements were carried out using a CHI660E electrochemical workstation. Cyclic voltammograms were acquired at a scan rate of 20 mV/s with sweeping from -1.0 V to 1.0 V. For the complete device, the CV cycles were performed at a scan rate of 50 mV/s between -2.0 V and 1.5 V. Step chronoamperometry (CA) cycles were conducted under -2.0 V for coloration and 1.5 V for bleaching. The duration of each step was 30 s. The optical transmittance spectra of the ECDs were measured in situ by a Hitachi U-3010 UV-Vis spectrophotometer at a wavelength of 550 nm. The transmittance spectra of the device at its original, colored, and bleached states were measured over a wavelength of 300-800 nm. 3. Results and discussion 3.1 Characterization of Si3N4 thin films
5
Fig. 1 The three-dimensional AFM images for Si3N4 films deposited at various N2 contents: (a) 10% (b) 20% (c) 30% and (d) 40%.
Fig. 1 illustrates characteristic topographies for Si3N4 films prepared at various N2 contents. The root mean square roughness (Rq) decrease first and then increase in general as the N2 contents . The values of Rq are respectively 3.17, 2.70, 2.74, and 3.29 nm, corresponding to N2 content of 10%, 20%, 30%, and 40%. The minimum value appears when the N2 content equals 20%. The as-deposited Si3N4 films are smooth which would facilitate Li ion transportation and film growth for top layers of the ECD. The Li ion conducting properties of Si3N4 thin films were evaluated by cyclic voltammetry (CV) test conducted in 1M anhydrous LiClO4-PC solution. The Si3N4 thin films prepared at various parameters were deposited on WO3/ITO/glass substrates. The considered sputtering conditions were N2 content, working pressure, and film thickness. The WO3 substrates were prepared at the same parameters, which were shown in Table 1. CV curves showed the intrinsic electrochemical behavior of WO3. By coating Si3N4 thin films prepared at different parameters, the peak current density and charge density of the CV curves changed. The charge density (area of the CV curve) reflects the optical contrast of the WO3 thin film, and the cathodic (anodic) peak current density is related with the Li ion intercalation (deintercalation) diffusion coefficient, which only depends on the Li ion conductivity of the Si3N4 thin film due to the same WO3 electrochromic active layer. In this work, the diffusion coefficients of Li ions were calculated by using Randles-Sevick equation [22,23]: 3
1
1
i p 2.72 10 n 2 D 2 C0 2 (1) 5
where D is the diffusion coefficient, C0 is the concentration of active ions in the solution, is the scan rate, n is the number of electrons and is assumed to be 1, i p is the peak current density.
6
Fig. 2 (a), (c) and (e) show the cyclic voltammograms of the Si3N4-coated WO3/ITO/glass deposited at various N2 contents, working pressures, and different film thicknesses, respectively. Only the first CV cycle of each sample is displayed. The corresponding quantification results of Li ion diffusion coefficients and charge densities are presented in (b), (d) and (f), respectively. The cyclic voltammetry measurements were carried out in 1M LiClO4-PC electrolyte at a scan rate of 20 mV/s sweeping between -1 V and 1 V.
Fig. 2(a,c,e) display the cyclic voltammograms of the Si3N4-coated WO3 thin films for the first cycle. The curves exhibit oxidation and reduction peaks of WO3 films, which can be interpreted as Li ion insertion/extraction through the Si3N4 layer. The corresponding quantification results of Li diffusion coefficient and charge density are shown in Fig. 2(b,d,f). The silicon nitride prepared by different parameters will 7
have a little influence on the diffusion of lithium ions into tungsten oxide. The minimal effect are the optimal preparation parameters. It is found that the current density and charge density are mainly influenced by the N2 content and film thickness (see Fig. 2(a,e)). While the impact of working pressure on Li ion conductivity of Si3N4 film can be neglected. The optimal value for N2 content appears at 20 % (Fig. 2(b)). As expected, the peak current density and transferred charge in WO3 films decrease gradually with the increasing thickness of Si3N4 cover layer (Fig. 2(e,f)). When the thickness of Si3N4 is less than 120 nm, the influence of Si3N4 buffer layer on electrochromic performance of WO3 can be regarded as negligible. Fig. 3(a) shows 100 CV cycles of a 100nm-thick Si3N4 covered WO3 in 1M LiClO4-PC. The poor cycling durability of bare WO3 was mainly caused by the residual water in the solution [24,25]. The corresponding charge density of each cycle is shown in Fig. 3(b). It is demonstrated that the Si3N4 cover layer can prevent WO3 from water contacting, thereby improving the cycling durability. Moreover, the Li ion transportation through Si3N4 film is stable.
Fig. 3 (a) The cyclic voltammograms of 1st and 100th cycles for bare WO3 and Si3N4-covered WO3 in 1M LiClO4-PC. (b) The corresponding caculated charge density of each cycle.
8
3.2 Characterization of complete electrochromic devices
Fig. 4 (a) and (b) respectively show the schematic diagram of the ECD without and with Si3N4 electron blocking layers
Fig. 4 shows the structure of devices with and without a silicon nitride film. The arrows indicate the directions of movement of lithium ions and electrons during the coloring process. A complementary ECD typically consist of two electrochromic active layers, an ion conductor layer, and two transparent electrodes. Up to now, WO3 is the best known and extensively used EC material due to its high coloration efficiency, large dynamic range, and good cyclic reversibility [26-28]. Corresponding to cathodic colored WO3, anodic colored NiO has been selected to be a complementary EC layer due to their similar coloration efficiency and color neutrality [29]. LiNbO3 functions as an ion conductor layer which provides Li+ ions and transportation channels between WO3 and NiO. Compared with conventional device configuration, two Si3N4 layers are integrated into the device in order to block the transportation of electrons across the LiNbO3 layer. XPS and XRD characterizations on Si3N4, WO3, LiNbO3, and NiO layers are provided in the supplementary figures. The Si 2p and N 1s binding energies were respectively found to be 102.3 eV and 398.2 eV (see Fig. S1), which correspond to Si3N4 [30,31]. The W 4f core level of the WO3 film exhibited peaks located at 35.8 eV and 37.9 eV (Fig. S2), which areattributed to the W6+ states [32]. The binding energies of Nb 3d were 207.2 eV with a splitting of 2.8 eV (Fig. S3) [33]. The Ni 2p3/2 peak of NiO displayed a lower energy at 853.9 eV, which is attributed to the Ni2+ species and higher binding energy (855.6 eV) arising from the Ni3+ oxidation state (Fig. S4) [34,35]. The XRD patterns
9
indicated that the Si3N4, WO3, and LiNbO3 were amorphous, whereas the NiO exhibited a face-centered cubic NaCl-type structure. The (111), (200), (220), (311), and (222) peaks typical to NiO (JCPDS Card No. 47-1049) were clearly identified.
Fig. 5 SEM cross-sectional image of the electrochromic device.
Two types of complete electrochromic devices (ECD) were fabricated in this work, i.e., ECD without Si3N4 buffer layers and ECD with Si3N4 buffer layers. The ECDs were based on the WO3 and NiO complementary electrochromic active layers. LiNbO3 thin films were used as the inorganic Li ion conductor. Fig. 5 exhibits a cross-sectional SEM image of the ECD having two Si3N4 buffer layers on both sides of LiNbO3. The thickness of each Si3N4 layer is approximately 80 nm. The interfaces are clearly identified indicating that all the layers are physically and chemically stable without interdiffusion in the preparation process. The smooth interface between Si3N4 and LiNbO3 is favorable for Li ion transportation. The electrochromic performances of the ECDs were characterized by step chronoamperometry (CA) and cyclic voltammetry (CV). The electrical data of 100 CA cycles (E1 = -2.0 V, E2 = 1.5 V) for the two ECDs were measured. According to O’Brien et al. [18], leakage current is defined as the steady state current at which the rate of color change becomes zero. It is obvious that the ECD having Si3N4 buffer layers exhibits lower leakage currents. This is associated with the internal leakage current, which will lead to electronic breakdown of the device under high voltage. Fig. 6 shows an amplification of the last 5 cycles of CA curves, which indicates that the ECD without buffer layers at colored state shows an increasing current with the time 10
increasing. While the ECD with Si3N4 buffer layers shows a steady current close to zero at its fully colored state. The steady state currents of the device at fully colored and bleached states are both reduced by Si3N4 buffer layers. Especially for the colored state leakage current, the value is significantly decreased from 216.0 to 32.1 μA /cm2.
Fig. 6 Step chronoamperometric curves for comparison of the devices with and without Si3N4 buffer layers.
Fig. 7 In situ transmittance variation at 550 nm plotted against time during 100 chronoamperometric cycles for the device with Si3N4 buffer layers. The applied voltages were -2.0 V for coloration and 1.5 V for bleaching. The duration was 30 s for each step.
Fig. 7 shows the in situ measurement of transmittance at 550 nm for the device with Si3N4 buffer layers during 100 CA cycles. The optical modulation is around 43 % from a colored state of 27 % to a bleached state of 70 %. The device shows a stable and reversible optical contrast. It is proved that the electrochromic performance and cycling stability of the device are not compromised by introducing the Si3N4 electron 11
blocking layers.
Fig. 8 Cyclic voltammograms of the device with and without Si3N4 buffer layers. The scan rate was 50 mV/s. The arrows refer to the sweeping directions. The inset figure shows the in situ transmittance change at 550 nm of the device during the CV cycles.
Fig. 8 shows the cyclic voltammograms (CV) of the two ECDs for the 20th cycle. The CV curves of the ECD without Si3N4 buffer layers are inclined and cross due to the poor dielectric property of the ion conductor layer. The perfect CV curves can be obtained from the device containing Si3N4 layers. When there is no electrochemical reaction, the corresponding values of current density are nearly zero. This is in accordance with the results of CA test that Si3N4 buffer layers can effectively reduce the leakage current in the full device. More importantly, the optical contrast and the response time of the device are hardly influenced (see the inset in Fig. 8). This implicates that the Li ions transportation through the Si3N4 layer is fast and stable.
Fig. 9 Optical transmittance spectra of the device containing Si3N4 buffer layers at its original, colored, and bleached states.
12
Fig. 9 shows the transmittance spectra of the ECD containing Si3N4 buffer layers at its original, colored, and bleached states. The device shows an optical modulation of around 40 % in the visible region. The optical absorption near 400 nm is mainly caused by the NiO layer and top ITO layer in the device. The device was bleached to light brown under a voltage of 1.5 V and changed to a neutral dark color under a voltage of 2.0 V.
Fig. 10 Open circuit memory test of the devices with and without Si3N4 buffer layers. The devices were both colored to a transmittance of 25 %.
Open circuit memory effect is a crucial factor in the assessment of electrochromic properties for ECDs concerning energy consumption [36,37].The device requires a low leakage current for a better open circuit memory (i.e., slower self-bleaching under open circuit), particularly with respect to the application when batteries are the main source of power for the ECDs. O’Brien et al. [18] reported that a 25 cm2 EC device has a leakage current of 1-4 mA corresponding to an open circuit memory of about 60 min before %T color may change from 8 to 20 %. In the present study, ECDs with and without Si3N4 buffer layers were subjected to a memory test. The two devices were colored to a same level (25 % at 550 nm) prior to disconnecting the external power. Fig. 10 shows the self-bleaching transmittance of the two devices. It is observed the ECD without Si3N4 buffer layers exhibits a rapid self-bleaching curve. While it takes 4000s for the ECD containing Si3N4 buffer layers, and the transmittance increases by 20 %. As discussed above, the leakage current for the device with Si3N4 buffer layers
13
is 32.1 μA/cm2. These results are in good agreement with the values reported by O’Brien et al.. The leakage current is dramatically reduced by the addition of Si3N4 layers, which leads to a better memory effect of the device. A device with a better open circuit memory can effectively reduce energy consumption in application.
4. Conclusions
Si3N4 thin films were deposited by pulsed DC reactive magnetron sputtering in a mixture of Ar and N2 atmosphere. They were used as the electron blocking layers to reduce the leakage current of electrochromic device. The Li ions conducting properties of the Si3N4 thin films were evaluated by electrochemical cyclic voltammetry. The optimal deposition conditions for Si3N4 layer were 20 % of N2 partial pressure, 275 W of sputtering power, and 0.5 Pa of working pressure. A multilayered
inorganic
all-solid-state
ECD
composed
of
Glass/ITO/NiO/Si3N4/LiNbO3/Si3N4/WO3/ITO was fabricated by sputtering. It was observed that leakage current of the device at colored state was reduced from 216.0 to 32.1 μA /cm2 with two 80nm-thick Si3N4 buffer layers added on both sides of the ion conductor. The corresponding self-bleaching time under open circuit was extended to 4000 s as the transmittance of the colored ECD increased from 25 % to 45 %. Both chronoamperometric cycles and voltammetric cycles indicated that the Li ions transportation through the Si3N4 layer was fast and stable. The optical modulation of the ECD was 43 % at 550 nm under -2.0 V (coloration) and 1.5 V (bleaching). The switching time was within 30 s.
Acknowledgements This work has been financially supported by the National Program on Key Research Project of China (2016YFB0303901) and the Beijing Natural Science Foundation (2161001) and the Fundamental Research Funds for the Central Universities (Grant No. YWF-16-JCTD-B-03).
14
References [1] C.G. Granqvist, Handbook of Inorganic Electrochromic Materials, Elsevier, 1995. [2] R.T. Wen, G.A. Niklasson, C.G. Granqvist, Electrochromic iridium oxide films:compatibility with propionic acid, potassium hydroxide, and lithium perchlorate in propylene carbonate, Sol. Energy Mater. Sol. Cells 120 (2014) 151–156. [3] D.M. Dong, W.W. Wang, G.B. Dong, F. Zhang, Y.C. He, H. Yu, F. M. Liu, M. Wang, X.G. Diao, Electrochromic properties and performance of NiOx films and their Corresponding all-thin-film flexible devices prepared by reactive DC magnetron sputtering, Appl. Surf. Sci. 383 (2016) 49–56. [4] C.C. Liao, Lithium-driven electrochromic properties of electrodeposited nickelhydroxide electrodes, Sol. Energy Mater. Sol. Cells 99 (2012) 26–30. [5] H. Huang, J. Tian, W.K. Zhang, Y.P. Gan, X.Y. Tao, X.H. Xia, J.P. Tu, Electrochromic properties of porous NiO thin film as a counter electrode for NiO/WO3 complementary electrochromic window, Electrochim. Acta 56 (2011) 4281–4286. [6] S.V. Green, E. Pehlivan, C.G. Granqvist, G.A. Niklasson, Electrochromism in sputter deposited nickel-containing tungsten oxide films, Sol. Energy Mater. Sol. Cells 99 (2012) 339–344. [7] C.G. Granqvist, A. Azens, J. Isidorsson, M. Kharrazi, L. Kullman, T. Lindstrom, G.A. Niklasson, C.-G. Ribbing, D. Ronnow, M.Stromme Mattson, M. Veszelei, Towards the smart window: progress in electrochromics, J. Non-Cryst. Solids 218 (1997) 273–279. [8] Q. Liu, G. Dong, Y. Xiao, F. Gao, M. Wang, Q. Wang, S. Wang, H. Zuo, X. Diao, An all-thin-film inorganic electrochromic device monolithically fabricated on flexible PET/ITO substrate by magnetron sputtering, Mater. Lett. 142 (2015) 232–234. [9] D. Zhou, F. Shi, D. Xie, D.H. Wang, X.H. Xia, X.L. Wang, C.D. Gu, J.P. Tu, Bi-functional Mo-doped WO3 nanowire array electrochromism-plus electrochemical energy storage, J. Colloid Interf. Sci. 465 (2016) 112–120. [10] M. Wang, G. Fang, L. Yuan, H. Huang, Z. Sun, N. Liu, S. Xia, X. Zhao, High
15
optical switching speed and flexible electrochromic display based on WO3 nanoparticles with ZnO nanorod arrays' supported electrode, Nanotechnology 20 (2009) 185304. [11] K. Wang, H. Wu, Y. Meng, Y. Zhang, Z. Wei, Integrated energy storage and electrochromic function in one flexible device: an energy storage smart window, Energ. Environ. Sci. 5 (2012) 8384. [12] S.H. Lee, R. Deshpande, P.A. Parilla, K.M. Jones, B. To, A.H. Mahan, A.C. Dillon, Crystalline WO3 Nanoparticles for Highly Improved Electrochromic Applications, Adv. Mater. 18 (2006) 763–766. [13] S. Pereira, A. Gonçalves, N. Correia, J. Pinto, L. Pereira, R. Martins, E. Fortunato, Electrochromic behavior of NiO thin films deposited by e-beam evaporation at room temperature, Sol. Energy Mater. Sol. Cells 120 (2014) 109–115. [15] S.C. Wang, K.Y. Liu, J.L. Huang, Tantalum oxide film prepared by reactive magnetron sputtering deposition for all-solid-state electrochromic device, Thin Solid Films 520 (2011) 1454–1459 [16] R.B. Goldner, F.O. Arntz, G .Berera, T.E. Haas, G. Wei, K.K. Wong, P.C. Yu, A monolithic thin-film electrochromic window. Solid State Ionics 53-56 (1992) 617– 627. [17] K.J. Patel , G.G. Bhatt, J.R. Ray, P. Suryavanshi, C.J. Panchal, All-inorganic solid-state electrochromic devices: a review, J. Solid State Electrochem. 21 (2017) 337–347 [18] N.A. O’Brien, J. Gordon, H. Mathew, B.P. Hichwa, Electrochromic coatings– applications and manufacturing issues, Thin Solid Films 345 (1999) 312–318 [19] S. Bogati, A. Georgn, W. Graf, Sputtered Si3N4 and SiO2 electron barrier layer between a redox electrolyte and the WO3 film in electrochromic devices, Sol. Energy Mater. Sol. Cells 159 (2017) 395–404 [20] P.F. Tavares, A.R. Gaspar, A.G. Martins, F. Frontini, Evaluation of electrochromic windows impact in the energy performance of buildings in Mediterranean climates, Energ. Policy 67 (2014) 68–81.
16
[21] M.S. Burdis, B. D.Greer, D.G. J.Weir, US patent 8,004,744 B2: Electrochromic devices having improved ion conducting layers, 2011. [22] P.S. Patil, P.R .patil, S.S. Kamble, S. H. Pawar, Thickness-dependent electrochromic properties of solution thermolyzed tungsten oxide films, Sol. Energy Mater. Sol. Cells 60 (2000) 143–153. [23] R. Sivakumar, R. Gopalakrishman, M. Jayachandran, C. Sanjeeviraja, Investigation of X-ray photoelectron spectroscopic (XPS), cyclic voltammetric analyses of WO3 films and their electrochromic response in FTO/ WO3/electrolyte/FTO cells, Smart Mater. Struct. 15 (2006) 882. [24] D. Chatzikyriakou, N. Krins, B. Gilbert, P. Colson, J. Dewalque, J. Denayer, R. Cloots, C. Henrist, Mesoporous amorphous tungsten oxide electrochromic films: a Raman analysis of their good switching behavior, Electrochim. Acta 137 (2014) 75– 82 [25] G. Leftheriotis, S. Papaefthimiou, P. Yianoulis, The effect of water on the electrochromic properties of WO3 films prepared by vacuum and chemical methods, Sol. Energy Mater. Sol. Cells 83 (2004) 115–124. [26] S.K. Deb, Opportunities and challenges in science and technology of WO3 for electrochromic and related applications, Sol. Energy Mater. Sol. Cells 92 (2008) 245– 258. [27] C.G. Granqvist, Electrochromic tungsten oxide films: review of progress 1993-1998, Sol. Energy Mater. Sol. Cells 60 (2000) 201–262. [28] G.A. Niklasson, C.G. Granqvist, Electrochromics for smart windows: thin films of tungsten oxide and nickel oxide, and devices based on these, J. Mater. Chem. 17 (2007) 127–156. [29] H. Moulki, D.H. Park, B.K. Min, H. Kwon, S.J. Hwang, J.H. Choy, T. Toupance, G. Campet, A. Rougier, Improved electrochromic performances of NiO based thin films by lithium addition: From single layers to devices, Electrochimi. Acta 74 (2012) 46–49. [30] S.L. Ku, C.C. Lee, Optical and structure properties of silicon nitride thin films
17
prepared by ion assisted deposition, Opt. Mater. 32 (2010) 956–960. [31] S.U. Han, H.S. Kang, B.K. Kang, Time-dependent degradation due to negative bias temperature instability of p-MOSFET with an ultra-thin SiON gate dielectric, Microelectron. Eng. 83 (2006) 520–527. [32] I.T.S. Garcia, D.S. Corrêa, D.S.D. Moura, J.C.O. Pazinato, M.B. Pereira, N.B.D. da Costa, Multifaceted tungsten oxide films grown by thermal evaporation, Surf. Coat. Technol. 283 (2015) 177–183. [33] M. Gellert, K.I. Gries, J. Sann, E. Pfeifer, K. Volz, B. Roling, Impedance spectroscopic study of the charge transfer resistance at the interface between a LiNi0.5Mn1.5O4 high-voltage cathode film and a LiNbO3 coating film, Solid State Ionics 287 (2016) 8–12. [34] S.V. Green, C.G. Granqvist, G.A. Niklasson, Structure and optical properties of electrochromic tungsten-containing nickel oxide films, Sol. Energy Mater. Sol. Cells 126 (2014) 248–259. [35] F. Lin, D. Nordlund, T.-C. Weng, R.G. Moore, D.T. Gillaspie, A.C. Dillon, R.M. Richards, C. Engtrakul, Hole doping in Al-containing nickel oxide materials to improve electrochromic performance, ACS Appl. Mater. Inter. 5 (2013) 301–309. [36] Q.R. Liu, G.B. Dong, Y. Xiao, M.P. Delplancke-Ogletree , F. Reniers , X.G. Diao, Electrolytes-relevant cyclic durability of nickel oxide thin films as anion-storage layer in an all-solid-state complementary electrochromicdevice, Sol. Energy Mater. Sol. Cells 157 (2016) 844–852 [37] K.J. Patel, M.S. Desai, C.J. Panchal, Studies of ZrO2 electrolyte thin-film thickness on the all-solid thin-film electrochromic devices, J. Solid State Electrochem.19 (2015) 275–279.
18