Microstructure and properties of manganese dioxide films prepared by electrodeposition

Microstructure and properties of manganese dioxide films prepared by electrodeposition

Applied Surface Science 254 (2008) 6671–6676 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/lo...

804KB Sizes 0 Downloads 27 Views

Applied Surface Science 254 (2008) 6671–6676

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Microstructure and properties of manganese dioxide films prepared by electrodeposition G. Moses Jacob, I. Zhitomirsky * Department of Materials Science and Engineering, McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4L7

A R T I C L E I N F O

A B S T R A C T

Article history: Received 26 February 2008 Accepted 11 April 2008 Available online 20 April 2008

Nanostructured manganese dioxide films were obtained by galvanostatic, pulse and reverse pulse electrodeposition from 0.01 to 0.1 M KMnO4 solutions. The deposition yield was investigated by in situ monitoring the deposit mass using a quartz crystal microbalance (QCM). Obtained films were studied by electron microscopy, X-ray diffraction analysis, energy dispersive spectroscopy, thermogravimetric and differential thermal analysis. The QCM and electron microscopy data were utilized for the investigation of deposition kinetics and film formation mechanism. It was shown that the deposition rate and film microstructure could be changed by variation of deposition conditions. The method allowed the fabrication of dense or porous films. The thickness of dense films was limited to 0.1 mm due to the insulating properties of manganese dioxide and film cracking, attributed to drying shrinkage. Porous and crack-free 1–2 mm films were obtained using galvanostatic or reverse pulse deposition from 0.02 M KMnO4 solutions. It was shown that film porosity is beneficial for the charge transfer during deposition and crack prevention in thick films. Moreover, porous nanostructured films showed good capacitive behavior for applications in electrochemical supercapacitors. The porous nanostructured films prepared in the reverse pulse regime showed higher specific capacitance (SC) compared to the SC of the galvanostatic films. The highest SC of 279 F/g in a voltage window of 1 V was obtained in 0.1 M Na2SO4 solutions at a scan rate of 2 mV/s. ß 2008 Elsevier B.V. All rights reserved.

PACS: 68.47.Gh 81.15.Pq 82.47.Uv 61.66.Fn Keywords: Manganese dioxide Porous films Electron microscopy Electrosynthesis Capacitance

1. Introduction Electrodeposition of oxides is an important technique for the surface modification of materials and fabrication of nanostructured films for electronic, catalytic, biomedical and electrochemical applications [1–7]. Oxide films can be obtained using anodic oxidation or cathodic reduction methods [1,2,8]. Electrochemical methods are especially attractive for the fabrication of manganese dioxide films. The analysis of literature indicates that manganese dioxides with various crystalline structures are important materials for lithium batteries [9], alkaline Zn/MnO2 cells [10], catalysts [11], sensors [12] and electrochemical supercapacitors [13]. The properties of MnO2 are influenced by crystalline structure, particle size, shape and surface area. Many applications of MnO2 are based on the use of porous nanostructured films. The microstructure of manganese dioxide films is especially important for the development of electrochemical supercapacitors and batteries, where small particle size, high surface area and porosity enable electrolyte access to the active material. A complicating factor

* Corresponding author. Tel.: +1 905 525 9140; fax: +1 905 528 9295. E-mail address: [email protected] (I. Zhitomirsky). 0169-4332/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2008.04.044

in the fabrication of MnO2 films by chemical methods is the lack of stable Mn (IV) precursors in aqueous solutions. However, manganese dioxide films can be prepared by electrochemical methods, using anodic oxidation of Mn2+ or cathodic reduction of Mn7+ species. The use of cationic Mn2+ species for anodic electrodeposition or anionic MnO4 (Mn7+) species for cathodic electrodeposition requires the understanding of the deposition mechanisms and kinetics of deposition, which are influenced by diffusion, electromigration, kinetics of electrochemical reactions and other factors. Many investigations have been focused on the anodic electrodeposition of manganese dioxide films [14–17]. This method leads to the formation of g-MnO2 (also called electrolytic manganese dioxide [18]). Extensive studies have shown that microstructure and properties of manganese dioxide films can be varied by the variation of bath composition and by the use of additives. Different techniques were used for the anodic deposition, including galvanostatic, potentiostatic, potentiodynamic and pulse deposition [14–17,19,20]. Electrochemical quartz crystal microbalance (QCM) has been used for the investigation of deposition kinetics and deposition mechanism [21]. The interest in the development of cathodic electrodeposition stems from the possibility of co-deposition of other materials and chemical modification of manganese dioxide films for the

G.M. Jacob, I. Zhitomirsky / Applied Surface Science 254 (2008) 6671–6676

6672

fabrication of composite films with improved properties. It is in this regard that various metals, oxides and hydroxide materials can be deposited cathodically and the problem of anodic oxidation of the metallic substrates can be avoided. The goal of this investigation was the fabrication of manganese dioxide films from KMnO4 solutions by cathodic electrodeposition. In the galvanostatic and pulse deposition experiments described below, the deposition yield was studied in situ using quartz crystal microbalance. We report the influence of deposition conditions and KMnO4 concentration on microstructure and properties of the films for applications in electrochemical supercapacitors. 2. Experimental procedures Electrodeposition was performed from the 0.01–0.1 M KMnO4 (Aldrich) aqueous solutions. The electrochemical cell for deposition included a substrate and a platinum counter electrode. The deposits were obtained on stainless steel foils (50 mm  50 mm  0.1 mm) and gold-coated quartz crystals, used as substrates. The deposition process has been monitored using a quartz crystal microbalance (QCM 922, Princeton Applied Research) controlled by a computer. The deposit mass Dm was calculated using Sauerbrey’s equation [22]: 2F02 DF ¼ pffiffiffiffiffiffiffiffiffiffiffiffi Dm A rq mq

(1)

where DF is frequency decrease of the QCM, F0 is the parent frequency of QCM (9 MHz), A is the area of gold electrode (0.2 cm2), rq is the density of the quartz (2.65 g cm3) and mq is the shear modulus of quartz (2.95  1011 dyne cm2). The deposits were scraped from the stainless steel substrates for X-ray diffraction (XRD) study, thermogravimetric analysis (TGA) and differential thermal analysis (DTA). The phase content of the deposits was determined by XRD with a diffractometer (Nicolet I2) using monochromatic Cu Ka radiation at a scanning speed of 0.5 8/min. TGA and DTA studies were carried out in air between room temperature and 1000 8C at a heating rate of 5 8C/min using a thermoanalyzer (Netzsch STA-409). The microstructure of the coatings deposited on stainless steel substrates was investigated using a JEOL JSM-7000F scanning electron microscope (SEM) equipped with energy dispersive spectroscopy (EDS). Capacitive behavior of the deposited films was studied using a potentiostat (PARSTAT 2273, Princeton Applied Research) controlled by a PowerSuite electrochemical software. The surface area of the stainless steel working electrodes was 1 cm2. A threeelectrode cell contained a working electrode, a platinum gauze counter electrode and a standard calomel reference electrode (SCE). Testing was performed in the 0.1 M Na2SO4 aqueous solutions, degassed with purified nitrogen gas. Cyclic voltammetry (CV) studies were performed within a potential range of 0–1.0 V versus SCE at scan rates of 2–100 mV/s. The SC was calculated using half the integrated area of the CV curve to obtain the charge (Q), and subsequently dividing the charge by the mass of the film (m) and the width of the potential window (DV): C¼

Q m DV

Fig. 1. Deposit mass versus deposition time for deposits obtained from the 0.02 M KMnO4 solutions at current densities of (a) 1 mA/cm2 and (b) 3 mA/cm2.

deposition yield was obtained at a higher current density (Fig. 1). These results indicate that the amount of the deposited material and film thickness can be varied. The deposition mechanism can be attributed to the diffusion and cathodic reduction of anionic MnO4 species. The reduction of MnO4 species and precipitation of manganese dioxide are in agreement with the Pourbaix diagram for Mn [23]. However, only limited information is available in the literature related to the complex chemistry of the reduction of MnO4. The kinetic pathway of reducing Mn7+ to Mn4+ depends on electrode potential, pH, concentration of MnO4 and other species in the solutions. In neutral aqueous solutions the following reaction [23] can result in the reduction of MnO4 species: MnO4  þ 2H2 O þ 3e ! MnO2 þ 4OH

(3)

The results presented in Fig. 2 indicate that the deposition rate decreased with increasing KMnO4 concentration. It is suggested that the deposition rate is governed by diffusion–electromigration kinetics in the KMnO4 solutions. When a negatively charged MnO4 ion is reduced cathodically, it has to approach the cathode by diffusion against an adverse potential gradient. In concentrated solutions, the interactions between ions can result in enhanced friction effect. It is important to note, that the decrease in diffusion

(2)

3. Results and discussion Cathodic deposits were obtained using galvanostatic and pulse deposition. Fig. 1 shows that deposit mass increased with increasing deposition time at a constant current density. Higher

Fig. 2. Deposit mass versus deposition time for deposits prepared from (a) 0.01 M, (b) 0.02 M and (c) 0.1 M KMnO4 solutions at a current density of 2 mA/cm2.

G.M. Jacob, I. Zhitomirsky / Applied Surface Science 254 (2008) 6671–6676

coefficient with increasing solution concentration has been observed in many electrochemical and non-electrochemical processes and discussed quite extensively in the literature [24,25]. In our experiments, the enhanced friction effect in concentrated solutions can be expected as a result of the adverse potential gradient and interactions of ions moving in cathodic direction due to the diffusion with ions moving in anodic direction due to the electromigration. Therefore, it is not surprising that the deposition rate decreased in concentrated solutions. An interesting finding of this work was the influence of KMnO4 concentration and deposition conditions on the microstructure and properties of manganese dioxide films. Fig. 3a and b compares the microstructures of the films prepared galvanostatically from the 0.02 M and 0.1 M KMnO4 solutions. The deposits prepared from the 0.02 M KMnO4 solutions exhibited a porous microstructure with typical pore size of about 100–150 nm. The surface structure is composed of nanowhiskers with typical length of 100 nm. In contrast, the films prepared from the 0.1 M KMnO4 solutions showed a relatively dense microstructure and exhibited cracks when film thickness exceeded 0.1 mm. It is known that electrodeposition of insulating materials can be used to produce very thin dense films or porous films [1]. The fabrication of thick films presents difficulties, related to low conductivity of MnO2 and film cracking, which results from drying shrinkage. It is suggested that the porous microstructure of MnO2 films prepared from the 0.02 M KMnO4 solutions promoted the charge transfer during electrosynthesis. Moreover, the porous microstructure is beneficial for crack prevention. It is well known that crack propagation can be prevented in porous materials by the crack-tip blunting mechanism [26,27]. Our results indicate that porous films with thickness of 1–2 mm can be produced by electrodeposition from the 0.02 M KMnO4 solutions. The porosity of MnO2 films is an important requirement for supercapacitors, batteries, catalysts and other applications. Therefore, it is necessary to investigate of the influence of KMnO4 concentration and other processing conditions on the deposition mechanism and microstructure of MnO2 films. Pulse deposition has been further utilized for the deposition of MnO2 films. In the pulse electrodeposition experiments, a series of pulses of constant current density was separated by periods of zero current. The deposition was performed from the 0.02 M KMnO4 solutions. The deposit mass versus time dependence shown in Fig. 4 indicates mass gain during the ‘‘on’’ time, when the current was applied. The sample mass remained constant during the ‘‘off’’ time, corresponding to zero current. SEM investigations revealed relatively dense microstructure of the deposited films, which exhibited cracks (Fig. 3c). This is in contrast with the results obtained at the galvanostatic regime using the solutions of the same concentration and at the same current density. The microstructure of the films prepared using pulse deposition from 0.02 M KMnO4 solutions with ‘‘on’’ time 20 s and ‘‘off’’ time of 10 s was similar to that for films prepared from the 0.1 M KMnO4 solutions (Fig. 3b). It is suggested that MnO4 concentration at the electrode surface decreased during the ‘‘on’’ time due to reaction (3) and increased during the ‘‘off’’ time due to diffusion. It is suggested that MnO4 depletion at the electrode surface is an important factor controlling the microstructure of the films. The MnO4 depletion during deposition provided higher MnO4 concentration gradient at the electrode surface, resulting in enhanced diffusion of MnO4 ions towards the cathode and higher MnO2 deposition rate (Fig. 2). The higher deposition rate can result in porous films. Deposition was also performed in a reverse pulse regime. In this approach, films were deposited at a constant cathodic current density of 1–2 mA/cm2 during 2–5 min, than current of 0.5–1 mA/ cm2 of opposite polarity and duration of 0.5–1 min was applied.

6673

Fig. 3. SEM images of the deposits prepared galvanostatically from (a) 0.02 M KMnO4 solutions, (b) 0.1 M KMnO4 solutions and (c) using pulse deposition from 0.02 M KMnO4 solutions; ‘‘on’’ time 20 s, ‘‘off’’ time 10 s. The current density was 2 mA/cm2 and total cathodic current duration was 2 min.

Fig. 5 shows mass change during the film deposition in the reverse pulse regime. The QCM data shows mass gain corresponding to the cathodic current and a small mass reduction when a current of opposite polarity was applied. Fig. 6 shows typical microstructure of a film prepared in the reverse pulse regime. SEM images at different magnifications show that the films are crack free and exhibit porosity. Porous crack-free films, prepared from the 0.02 M KMnO4 solutions at the galvanostatic conditions and in the reverse pulse regime were studied by EDS, XRD, TGA and DTA. EDS studies

6674

G.M. Jacob, I. Zhitomirsky / Applied Surface Science 254 (2008) 6671–6676

Fig. 4. (a) Current density and (b) deposit mass versus deposition time for deposits prepared from 0.02 M KMnO4 solutions using pulse deposition.

showed that the films contained K. Numerous EDS analyses showed that the K/Mn ratio in the galvanostatic films prepared at a current density of 2 mA/cm2 and duration of 2–5 min was 0.28  0.01. The K/Mn ratio for the films, prepared in the reverse pulse regime at the same cathodic current and 1 mA/cm2 anodic current density during 1 min, was found to be 0.21  0.01. Therefore, the reduction in film mass during the anodic current can be partially attributed to the removal of positively charged K+ ions from the films. However, the reduction in film mass during the anodic current can also be attributed to dissolution of Mn species. X-ray studies showed crystallinity of the films prepared at the galvanostatic and reverse pulse conditions (Fig. 7). The X-ray diffraction patterns are similar, however the films prepared in the reverse pulse regime exhibited broader peaks of lower intensity. The peak broadening can be attributed to the small particle size and poor crystallinity of the deposits. The X-ray diffraction patterns can be attributed to rancieite (JCPDS file 22-0718) or birnessite (JCPDS file 87-1497) structures. However, it is difficult to distinguish between the rancieite and birnessite phases owing to the peak broadening. The rancieite and birnessite phases have near MnO2 composition, which can be described by the formula KxMnO2+y(H2O)z.

Fig. 6. SEM images at different magnifications of the deposit prepared from the 0.02 M KMnO4 solutions in the reverse pulse regime at a cathodic current density of 2 mA/cm2 during 2 min, and anodic current density of 1 mA/cm2 during 1 min.

Fig. 5. (a) Current density and (b) deposit mass versus deposition time for deposits prepared from 0.02 M KMnO4 solutions in the reverse pulse regime.

Fig. 8 compares TGA and DTA data for the galvanostatic and reverse pulse films. The TGA data show a sharp reduction in sample weight below 200 8C and higher weight loss for the reverse pulse films. The weight loss in the temperature range of 20–400 8C and broad DTA endotherms around 100 8C can be attributed to dehydration. The weight gain of 0.3 wt.% observed in the TGA data for the reverse pulse films and corresponding exotherm in DTA data can be attributed to oxidation [28]. The weight loss at 900 8C in TGA data and corresponding DTA endotherms can be attributed to reduction and formation of Mn3O4 phase [28]. The total weight

G.M. Jacob, I. Zhitomirsky / Applied Surface Science 254 (2008) 6671–6676

Fig. 7. XRD data for the deposits prepared from the 0.02 M KMnO4 solutions (a) galvanostatically and (b) using reverse pulse deposition at a cathodic current density of 2 mA/cm2 and anodic current density of 1 mA/cm2.

loss at 1000 8C was found to be 19.1 and 23.6 wt.% for the galvanostatic and reverse pulse films, respectively. The capacitive behavior of the films was investigated in the potential range of 0–1.0 V versus SCE. Fig. 9 shows a typical CV in the 0.1 M Na2SO4 solutions. Within the potential range of 0–1.0 V versus SCE the manganese oxide electrodes exhibited capacitivelike current–potential responses, indicated by the box shape of the CVs. It is clear from Fig. 9 that there are no redox peaks in the range between 0 and 1.0 V. The reverse pulse films showed higher SC compared to the SC of galvanostatic films. The highest SC of 279 F/g was obtained at a scan rate of 2 mV/s for the reverse pulse films. The galvanostatic film of the same mass exhibited a SC of 188 F/g at the same scan rate. However, the SC decreased with increasing scan rate (Fig. 10). Similar reduction in SC was observed in other investigations [20,29] and was attributed to low electronic and ionic conductivity of the MnO2 electrodes. The mechanism of charge storage in manganese oxides is based on the adsorption of ions on the oxide surface. The pseudocapa-

Fig. 8. (a and b) TGA and (c and d) DTA data for (a and c) galvanostatic deposits and (b and d) reverse pulse deposits prepared from 0.02 M KMnO4 solutions at a cathodic current density of 2 mA/cm2 and anodic current density of 1 mA/cm2.

6675

Fig. 9. CV in the 0.1 M Na2SO4 solution at a scan rate of 20 mV/s for the 80 mg/cm2 deposit prepared using reverse pulse deposition from 0.02 M KMnO4 solution.

citance of hydrous manganese oxide is attributed to reversible redox transitions involving exchange of protons and/or cations with the electrolyte [17]: MnOa ðOHÞb þ dHþ þ de $ MnOad ðOHÞbþd

(4)

where MnOa(OH)b and MnOad(OH)b+d indicate manganese oxide at high and low oxidation states, respectively. The analysis of Eq. (4) indicates that high ionic and electronic conductivity of the active material are necessary in order to achieve a high SC. The values of SC reported in the literature [15,17,29–32] for MnO2 electrodes are usually in the range between 100 and 300 F/ g. These values are far from the theoretical SC [33] of 1370 F/g and reported experimental SC of 700 F/g for very thin films [14]. The reduction in SC with increasing film thickness was attributed to low conductivity of MnO2. The higher SC of the reverse pulse films can be attributed to different factors, such as lower K content and higher porosity. Further improvement in the electrochemical properties of manganese oxide films can be achieved by the optimization of film microstructure and composition.

Fig. 10. SC versus scan rate for the 80 mg/cm2 deposits prepared (a) galvanostatically and (b) reverse pulse deposition from 0.02 M KMnO4 solutions.

6676

G.M. Jacob, I. Zhitomirsky / Applied Surface Science 254 (2008) 6671–6676

4. Conclusions Nanostructured manganese dioxide films were prepared by cathodic reduction of KMnO4 solutions. It was shown that deposition could be performed in galvanostatic, pulse and reverse pulse deposition regimes. QCM studies showed that deposition rate increased with increasing current density and decreased with increasing KMnO4 concentration. The deposit microstructure can be varied by variation in deposition conditions. The deposition process enabled the fabrication of thin dense films, which exhibited cracks when film thickness exceeded 0.1 mm. Porous and crack-free 1–2 mm films were obtained using galvanostatic or reverse pulse deposition from 0.02 M KMnO4 solutions. It was shown that film porosity is beneficial for charge transfer during deposition and crack prevention during drying. Xray studies revealed poor crystallinity of as-prepared films. EDS, TGA and DTA studies showed that the composition of the films can be described by formula KxMnO2+y(H2O)z. Porous films showed capacitive behavior in the 0.1 M Na2SO4 solutions in a potential window of 0–1.0 V versus SCE. The films prepared in the reverse pulse regime showed higher SC compared to the SC of galvanostatic films. The SC decreased with increasing scan rate due to the low conductivity of MnO2. The highest SC of 279 F/g was obtained for the 80 mg/cm2 films at a scan rate of 2 mV/s. Obtained films can be considered as possible electrode materials for electrochemical supercapacitors. Acknowledgement The authors gratefully acknowledge the financial support of the Natural Science and Engineering Research Council of Canada.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

J.A. Switzer, Am. Ceram. Soc. Bull. 66 (1987) 1521–1524. G.H.A. Therese, P.V. Kamath, Chem. Mater. 12 (2000) 1195–1204. I. Avramova, D. Stoychev, Ts. Marinova, Appl. Surf. Sci. 253 (2006) 1365–1370. T.P. Gujar, V.R. Shinde, S.S. Kulkarni, H.M. Pathan, C.D. Lokhande, Appl. Surf. Sci. 252 (2006) 3585–3590. I. Zhitomirsky, Adv. Colloid Interf. Sci. 97 (2002) 279–317. R.S. Patil, M.D. Uplane, P.S. Patil, Appl. Surf. Sci. 252 (2006) 8050–8056. C. Gu, J. Li, J. Lian, G. Zheng, Appl. Surf. Sci. 253 (2007) 7011–7015. J.K. Barton, A.A. Vertegel, E.W. Bohannan, J.A. Switzer, Chem. Mater. 13 (2001) 952–959. L.I. Hill, R. Portal, A. Verbaere, D. Guyomard, Electrochem. Solid-State Lett. 4 (2001) A180–A183. L. Bai, D.Y. Qu, B.E. Conway, Y.H. Zhou, G. Chowdhury, W.A. Adams, J. Electrochem. Soc. 140 (1993) 884–889. D. Das, P.K. Sen, K. Das, J. Appl. Electrochem. 36 (2006) 685–690. X.-L. Luo, J.-J. Xu, W. Zhao, H.-Y. Chen, Anal. Chim. Acta 512 (2004) 57–61. J.N. Broughton, M.J. Brett, Electrochim. Acta 50 (2005) 4814–4819. S.-C. Pang, M.A. Anderson, T.W. Chapman, J. Electrochem. Soc. 147 (2000) 444–450. M.-S. Wu, P.-C.J. Chiang, Electrochem. Solid-State Lett. 7 (2004) A123–A126. S.-C. Pang, M.A. Anderson, J. Mater. Res. 15 (2000) 2096–2106. C.-C. Hu, T.-W. Tsou, J. Power Sources 115 (2003) 179–186. A.P. Malloy, G.J. Browning, S.W. Donne, J. Colloid Interf. Sci. 285 (2005) 653–664. M. Ghaemi, L. Binder, J. Power Sources 111 (2002) 248–254. K.R. Prasad, N. Miura, J. Power Sources 135 (2004) 354–360. M.P. Owen, G.A. Lawrance, S.W. Donne, Electrochim. Acta 52 (2007) 4630–4639. M.R. Deakin, D.A. Buttry, Anal. Chem. 61 (1989) 1147A–1154A. M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion Engineers, Houston, TX, USA, 1966. N.F.E.I. Nestle, R. Kimmich, J. Phys. Chem. 100 (1996) 12569–12573. L. Tang, Cem. Concr. Res. 29 (1999) 1469–1474. Z.-Y. Deng, J. She, Y. Inagaki, J.-F. Yang, T. Ohji, Y. Tanaka, J. Eur. Ceram. Soc. 24 (2004) 2055–2059. T. Nakamura, Z. Wang, ASME J. Appl. Mech. 68 (2001) 242–251. J. Moon, M. Awano, H. Takagi, Y. Fujishiro, J. Mater. Res. 14 (1999) 4594–4601. J.-K. Chang, W.-T. Tsai, J. Electrochem. Soc. 150 (2003) A1333–A1338. C.-C. Hu, C.-C. Wang, J. Electrochem. Soc. 150 (2003) A1079–A1084. S. Devaraj, N. Munichandraiah, J. Electrochem. Soc. 154 (2007) A80–A88. R.N. Reddy, R.G. Reddy, J. Power Sources 124 (2003) 330–337. S. Devaraj, N. Munichandraiah, Electrochem. Solid-State Lett. 8 (2005) A373–A377.