Effect of Film Morphology on the Li Ion Intercalation Kinetics in Anodic

Apr 16, 2014 - Electrolytic manganese dioxide (EMD) films are electrodeposited from acid MnSO4 solution, targeting equivalent film thicknesses between...
0 downloads 4 Views 2MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article

Effect of Film Morphology on the Li-ion Intercalation Kinetics in Anodic Porous Manganese Dioxide Thin-Films Ahmed Saad Etman, Aleksandar Radisic, Emara Mahmoud, Cedric Huyghebaert, and Philippe M Vereecken J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp4105008 • Publication Date (Web): 16 Apr 2014 Downloaded from http://pubs.acs.org on April 22, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Effect of Film Morphology on the Li-ion Intercalation Kinetics in Anodic Porous Manganese Dioxide Thin-Films Ahmed S. Etman1,2,3,4†, Aleksandar Radisic2, Mahmoud M. Emara3, Cedric Huyghebaert2, Philippe M. Vereecken2, 4* 1

KACST-Intel Consortium Center of Excellence in Nano-Manufacturing Applications (CENA), Riyadh, Saudi Arabia; 2imec, Kapeldreef 75, Leuven, Belgium; 3Chemistry Department, Alexandria University, Egypt; 4 Centre for Surface Chemistry and Catalysis, Faculty of Bioscience Engineering, KU Leuven, Belgium Email: [email protected]

Abstract

Electrolytic manganese dioxide (EMD) films are electrodeposited from acid MnSO4 solution, targeting equivalent film thicknesses between 25 and 100 nm onto silicon substrates coated with Pt and TiN seed layers. The structure and morphology of the deposited EMD films is characterized and tied to the observed electrochemical performance. Deposition efficiencies as large as 80% are achieved, as determined from RBS. TEM and SEM show that our EMD films are highly porous. The degree of porosity is dependent on the deposition current density and also

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 41

the deposition time. The electrochemical activity of the EMD films for lithium ion intercalation is investigated in propylene carbonate solvent with 1M LiClO4. Reversible insertion and extraction of Li+ follows after removing water from the films by annealing and subsequent electrochemical cycling. The porous EMD thin films show a Li-ion capacity up to 90% of the theoretical capacity at 0.8C rate, based on the Mn content from RBS data and assuming one Liion per MnO2 unit. The effect of film thickness and film porosity on the rate of Li+ intercalation (and thus accessible capacity) is discussed. KEYWORDS: thin film batteries, lithium-ion batteries, electrodeposition, electrolytic manganese dioxide, lithium intercalation kinetics

1. Introduction

Manganese dioxide is an excellent choice for cathode material in both aqueous and non-aqueous battery systems not only due to its good electrochemical properties, but also for its low toxicity, low cost, and relative abundance. Among all the MnO2 structures, the γ-MnO2 is the electrode material for both alkaline and lithium ion battery systems.1-3

The γ-MnO2 is an intergrowth between pyrolusite (β-MnO2) and ramsdellite-MnO2 with interstitial space of (1x1) and (2x1), respectively. Its structure can be described in terms of two types of structural defects termed De Wolff defect and microtwinning defect (Mt). De Wolff defect refers to the fraction (Pr) of pyrolusite-MnO2 structure within ramsdellite-MnO2 structure. Chabre and Pannetier introduced a method for the calculation of the Pyrolusite content (Pr).4 The level of microtwinning defect is reflected in the variation in peak widths and intensities in the powder X-ray diffraction patterns of γ-MnO2 materials. Another feature of γ-MnO2 is the

ACS Paragon Plus Environment

2

Page 3 of 41

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

presence of cation vacancies and protons in the structure as described by Ruetschi et.al..5 They reported that the chemical composition and density of EMD could be explained by assuming the presence of vacancies in γ-MnO2 only on the cationic lattice, and the existence of small fraction of the lower valent Mn (III) ions:

  

where, (

  



 

1

) refers to a cation vacancy, and x and y are the mole fractions of cation vacancies and

lower valent Mn3+ ions, respectively.1, 6 Each vacancy (vacant Mn4+ site) is coordinated with four OH- groups and each Mn3+ site is coordinated with one OH- group.

The γ-MnO2 is most commonly prepared via electrodeposition, and hence termed electrolytic manganese dioxide (EMD). Different techniques were used for the anodic deposition, including galvanostatic, potentiostatic, and pulse deposition.7 S. Sarcioux et al. showed that the structure of the EMD depends on the deposition conditions of temperature, pH and current density.8-9

Initial efforts in using manganese dioxide deposited from aqueous solutions for Li-ion batteries failed, due to the presence of a high water content in their structure. Lee et.al classified the water in MnO2 structure into three types: (a) Type 1 water, that is physisorbed water able to be reversibly removed around 100 ℃, (b) Type 2 water, which consists of surface bound hydroxyls that are irreversibly removed at 200 ℃, and (c) Type 3 water, which are bulk hydroxyl groups removed irreversibly around 300 ℃.10 Ikeda et al. reported that annealing of the MnO2 at temperatures above 300 ℃ removed the structural water and resulted in an electrochemically active material.11 In reality, the thermal treatment below 400 ℃ not only removes the structural water, but also results in the rearrangement of the ramsdellite regions in γ-MnO2 to form a

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 41

material with a mostly pyrolusite structure, commonly referred to as Heat-treated Manganese Dioxide (HEMD).12-13 The HEMD is a porous material, the degree of porosity depends on the annealing conditions and the structure of the as-deposited film.14

For battery applications, several micron thick EMD layers are made which are then removed from the substrate and grounded up to powders to be coated as electrodes in combination with conductive carbon additives on the current collector foils.9 Reversible lithium ion insertion and extraction is obtained in the voltage range between 2 and 4 V vs. Li+/Li; wherby the lithium ions are inserted predominantly into the larger (2 x 1) channels of the ramsdellite chains. The first cycle shows typically a high discharge capacity of ~250 mA.h.g-1; however in the following cycles γ-MnO2 electrodes lose capacity on electrochemical cycling. The loss of capacity upon cycling indicate that the structure of γ-MnO2 is modified upon cycling. Structural refinements of ramsdellite-MnO2 and its lithiated products have shown that the hexagonally-close-packed array of the starting material transforms on lithiation toward cubic close-packing due to the electrostatic interactions between the lithium ions and the manganese ions, which result in expanding the orthorhombic unit cell parameter.3 These changes in unit cell parameters, particularly the large increase in b are the particular reasons for losing capacity upon cycling. S. Sarcioux et al. reported that the first lithium insertion into γ-MnO2 involves irreversible transformation to form LixMnO2 phase; in the following cycles a reversible intercalation is observed.8-9 This reversible capacity was found to be dependent on the γ -MnO2 structure and is the highest for γ-MnO2 with low Pr and Mt and for those with Pr >90% and low Mt.

Recently A. Cross et al. prepared thin films of EMD of thickness ~20-100 nm onto Pt electrode using galvanostatic and potentiostatic techniques for super capacitor applications.15-16 As for the

ACS Paragon Plus Environment

4

Page 5 of 41

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

thick deposits, 7, 9 the morphology and surface area of these films were found to be dependent on the deposition conditions such as current density, deposition time, MnSO4 and H2SO4 concentrations, and the deposition temperature.

In this paper, EMD films were anodically deposited onto planar silicon substrates coated with Pt and/or TiN seed layers. As platinum is not a cost effective viable option for thin-film battery applications, also TiN was investigated as possible current collector. TiN is a good diffusion barrier for Li and is considered for 3D thin-film microbatteries fabricated on Si.17-18 However, as TiN is prone to oxidation, the anneal of the EMD films had to be done in oxygen-free ambient. For direct comparison, the films on Pt substrates were therefore also annealed in N2 atmosphere. Note that the EMD films for commercial alkaline and lithium batteries are removed from the substrate. Hence, in this case oxygen containing annealing ambient presents no obstacle and oxygen may even be an important ingredient for the activation of the films.12-13 Also for substrate bound EMD films on noble seed layers such as platinum, oxygen containing ambient is not an issue. Here, we investigate the films with an activation anneal in inert N2 atmosphere. The EMD films were physically and electrochemically characterized providing detailed information on the chemical composition, porosity and Li-ion intercalation kinetics of the films. The rate of Li-ion insertion and extraction was correlated with film thickness and porosity. For porous films, the diffusion length for Li+ is determined by (half) the pore wall thickness. It will be shown that Liintercalation kinetics becomes interface limited for EMD films with high porosity and thus thin pore walls.

2. Experimental

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 41

All electrochemical measurements as well as EMD depositions were performed in a three electrode set up, with Pt mesh counter electrode, Ag/AgCl/3M NaCl reference electrode (BASi analytical, 0.22 V vs. SHE). All potentials are referred to this Ag/AgCl electrode unless otherwise noted. The Pt coated Si substrates consisted of a stack of 85 nm PVD Pt seed layer on top of 30 nm PVD TiN barrier on a 1µm thick CVD SiO2 insulating layer. The TiN coated Si substrates consisted 70 nm PVD TiN seed layers on chemically oxidized Si. EMD films were deposited from solution of 0.3 M MnSO4·H2O (98.0-101.0%, Alfa easer) and 0.55 M H2SO4 (96%, OMG) at room temperature (22-24 °C). For comparison in electrochemical measurements, also an indifferent electrolyte solution of 0.3 M Na2SO4 (99.0%, Sigma Aldrich) and 0.55 M H2SO4 was prepared. After deposition, samples were annealed at 350 °C in N2 atmosphere (200 mbar) for 3h (+20 min. ramp up) and then allowed to cool down to room temperature.

Samples were characterized by Scanning Electron Microscope (SEM) using Philips SEM XL30 or SEM Nova 200 microscope. X-ray Diffraction (XRD) was done using Panalytical system with Cu Kα radiation. A Bragg–Brentano setup was used with a 2   scan range from   10 to   80 and a scan speed of 0.17o.s-1 and step size of 0.03o. A 3o  offset was applied to avoid the mono-crystalline Si peak. Raman spectroscopy measurements were done between 50 and 1000 cm-1 at room temperature using a green laser 532 nm, and excitation power between 0.54 3.65 mW. Chemical state analysis was performed by X-ray photoelectron spectroscopy (XPS). These measurements were performed using a monochromatized Al Kα X-ray source (1486.6 eV) with a spot size of 400 microns, and 22° exit angle. Rutherford Back Scattering (RBS) measurements were performed using He+ beam of energy 1.523 MeV, with scatter angle 170°, and tilt angle of 11°.

ACS Paragon Plus Environment

6

Page 7 of 41

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The electrochemical activity of EMD films for lithium intercalation was investigated by galvanostatic charging/discharging and cyclic voltammetry (CV) between 2-4 V using a three electrode electrochemical cell made of Teflon placed in argon filled glove box (< 1 ppm O2 and H2O levels). The EMD films were connected as the working electrode and two ribbons of Li metal were used as reference and counter electrodes. Anhydrous Propylene Carbonate (99.7%, Sigma Aldrich) solvent with 1M LiClO4 (99.99%, Sigma Aldrich) was used as Li+ electrolyte solution. The theoretical maximum capacity were calculated based on the Mn content in the films as determined with RBS and assuming one Li per MnO2 unit (i.e. 1.55 A.h.cm-3 for a MnO2 film with density of 5.02 g.cm-3). As a general procedure for screening the electrochemical behavior, the film under investigation was cycled first at scan rate 10 mV.s-1 between 2 and 4 V until a stable and reproducible voltammogram was obtained, typically more than 10 cycles. Then electrochemical measurements were done (e.g. dependence of scan rate in cyclic voltammetry or galvanostatic charging/discharging cycles). At regular intervals in between and at the end of each series of experiments, a cyclic voltammogram at 10 mV.s-1 was recorded to assess the stability of the signal and reproducibility of the data.

3. Results and Discussion 3.1. Anodic deposition of EMD layers on Pt and TiN The electrochemistry of Mn(IV)/Mn(II) on Pt and TiN was evaluated by comparison of currentpotential characteristics in sulfuric acid solutions with and without Mn2+ ions. Figure 1 shows cyclic voltammograms for Pt and TiN substrates in aqueous solutions of 0.3 M Na2SO4 + 0.55 M H2SO4 (indifferent electrolyte solution) and 0.3 M MnSO4 + 0.55 M H2SO4 (EMD bath). For Pt in the indifferent electrolyte solution, no anodic current is measured up to the onset of the oxygen evolution reaction for U>1.3 V. For the Pt substrate in the EMD bath, an exponentially

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 41

increasing anodic current is measured from U>1.05 V upon positive polarization from its opencircuit potential (OCP~0.6 V). The anodic current corresponds to the oxidation of Mn(II) to predominantly Mn(IV) with the precipitation of the hydrated MnO2 or EMD. The hysteresis when reversing the polarization direction at 2 mA.cm-2 is indeed typical for nucleation and growth processes. In the reverse scan, a reversible cross-over from anodic to cathodic electrode reactions is observed (i=0 at U=1.15 V). The cathodic current peak at 1.08 V is due to the stripping of the EMD layer. The charge under the cathodic stripping peak was -5 mC.cm-2, equivalent with a 4.5 nm MnO2 (assuming 100% density, ρ=5.02 g.cm-3). The ratio of anodic charge to cathodic charge was 1.6; i.e. a deposition efficiency of 63%. As no oxygen evolution occurs yet at these potentials (see cyclic voltammogram in indifferent electrolyte solution), the lower current efficiency is the result of incomplete precipitation of Mn(IV) (i.e. formation of the soluble species [Mn(H2O)6 ]3+). This explanation agrees well with the electrodeposition mechanism proposed by C.J. Clarke).1

The electrochemistry on TiN is quite different from that on the noble Pt. In the indifferent electrolyte solution, an anodic current is measured for U>1.05 V with a peak at 1.35 V (0.2 mA.cm-2). No oxygen evolution is observed even for potentials up to 2.0 V. At these positive potentials, TiN is electro-oxidized forming a passivating TiONx layer which blocks the oxygen evolution reaction on TiN/TiONx. The charge under the curve in Figure 1 was 6.05 mC.cm-2 and corresponds to the oxidation of about 2 nm of TiN (note: seed layer thickness is 70 nm). In the EMD bath, an additional anodic current peak for EMD deposition is observed with an onset at 1.25 V (0.15 mA.cm-2) and a maximum at 1.40 V (0.6 mA.cm-2 absolute or 0.4 mA.cm-2 with respect to TiN/TiONx in indifferent electrolyte). Hence, anodic EMD deposition on TiN coincides with TiONx formation and the porous EMD layer will be formed on top of a growing

ACS Paragon Plus Environment

8

Page 9 of 41

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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

TiONx layer. The i-U curve for TiN/TiONx/EMD joins with that of TiN/TiONx again at 1.65 V; i.e. 20 s after onset of MnO2 deposition. At this potential, the net partial EMD charge was 3.2 mC.cm-2 (corresponding to dense EMD layer of about 2.9 nm). Hence, the thickness of the EMD layer is limited by the oxidation of TiN itself. In the reverse scan (negative polarization) no cathodic current was observed in this case. Indeed, the TiONx passivation layer blocks the anodic EMD reaction as well as the cathodic stripping reaction.

The EMD layers can be easily

dissolved chemically in oxalic acid. For the colored EMD layers on Pt, immersion of the samples in solution of 0.3M oxalic acid, revealed the bright Pt surface again as the EMD layers dissolved quickly. For the layers deposited on TiN seed, a blue color remained after EMD dissolution typical for a TiONx oxide layer.

EMD was electrodeposited on Pt at a constant current density between 0.25 and 3 mA.cm-2 targeting a charge of 0.1 C.cm-2. Figure 2 shows the potential-time transients measured at 0.25, 1, and 2 mA.cm-2. As can be readily seen, the shape of the potential-time curves at low current densities is different from those at high current densities. At low current densities (≤1 mA.cm-2) an initial higher potential for nucleation and growth of EMD on Pt is seen before reaching the steady state potential after 2 to 3 s of deposition (less than 2 nm films). The steady state potentials are also plotted in Figure 1 and correspond to the potential range of EMD deposition before the onset of O2 evolution reaction (1.1 V