MWCNT composite electrodes for Li-ion batteries by using electroless tin coating

    Production of Sn-Cu/MWCNT Composite Electrodes for Li-ion Batteries by Using Electroless Tin Coating Mehmet Uysal, Tugrul Cetinkaya, Muhammet Kartal, Ahmet Alp, Hatem Akbulut PII: DOI: Reference:

S0040-6090(14)00832-3 doi: 10.1016/j.tsf.2014.08.019 TSF 33658

To appear in:

Thin Solid Films

Please cite this article as: Mehmet Uysal, Tugrul Cetinkaya, Muhammet Kartal, Ahmet Alp, Hatem Akbulut, Production of Sn-Cu/MWCNT Composite Electrodes for Li-ion Batteries by Using Electroless Tin Coating, Thin Solid Films (2014), doi: 10.1016/j.tsf.2014.08.019

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.

ACCEPTED MANUSCRIPT Production of Sn-Cu/MWCNT Composite Electrodes for Li-ion Batteries by Using

IP

T

Electroless Tin Coating

SC R

Mehmet Uysal, Tugrul Cetinkaya, Muhammet Kartal*, Ahmet Alp, Hatem Akbulut

Affiliations: Sakarya University, Engineering Faculty, Department of Metallurgical and

NU

Materials Engineering, Esentepe Campus, 54187, Sakarya, Turkey,

MA

Telephone: +90264 295 56 92, +90545 371 22 01 +90264 295 56 01

E-mail:

[email protected] , [email protected]

D

Fax:

TE

Abstract

CE P

Cycling stability of pure tin electrodes were aimed to improve by using suitable combination of copper and multiwalled carbon nanotubes (MWCNTs). For this purpose, firstly Sn-Cu

AC

composite powders were produced using an electroless process. Then, Sn-Cu/MWCNT composite electrodes were prepared with dispersing different amount of MWCNT (10 wt.%, 20 wt.%, 40 wt.%) by high energy mechanical milling method. The surface morphology of the produced Sn-Cu/MWCNT composite powders was characterized using scanning electron microscopy (SEM), and energy dispersive spectroscopy (EDS) was used to determine the elemental surface composition of the composites. X-ray diffraction (XRD) analysis was performed to investigate the structure of the Sn-Cu/MWCNT composite powders. The electrochemical performance of Sn-Cu/MWCNT composite electrodes have been investigated by charge/discharge tests and cyclic voltammetry experiments. The cells discharge capacities were determined at a constant current in voltage range between 0.02 V – 1.5 V. AC 1

ACCEPTED MANUSCRIPT Electrochemical Impedance Spectroscopy (EIS) analysis was also carried out to measure resistivity and Li-diffusion in the assembled cells. The amounts of MWCNTs were shown to

T

be a crucial factor to improve Sn-Cu /MWCNT composite anodes for cyclability and

SC R

IP

reversible capacity.

Keywords:Sn-Cu/MWCNT nanocomposite, electroless coating, electrochemical behavior,

NU

Li-ion batteries.

MA

1. Introduction

Lithium ion secondary batteries have the highest energy density in commercial batteries and

D

used for energy storage in many electric devices, such as mobile phones, laptop computers

TE

and digital cameras owing to the advantages of no memory effect, high operation voltage and superior volumetric/gravimetric energy density [1,2]. The most common anode material used

CE P

in Li-ion batteries is graphite due to its low cost, availability and durability [3]. In graphitic anodes, the Li+ insertion mechanism corresponds to the reversible, progressive intercalation of

AC

Li+ ions between graphene layers to reach a theoretical capacity of 372 mAh g−1 if LiC6 is formed, compared with a practical capacity of 350 mAh g−1[4]. Recently, there has been tremendous interest and effort to the synthesis of tin-based compounds as alternatives to graphite materials, with the aim of improving the capacity and energy density of lithium ion batteries [2-4]. However, a large specific volume changing occurs during Li insertion and extraction reactions, which causes the electrode to fail by pulverization. As a result rapid capacity fading is observed [5-8]. Therefore, the main issue on the improvement of the Sn cycle performance is how to overcome the volume change and prevent the pulverization of particles. Therefore, many studies have been focused on tin-based intermetallic alloys such as Sn–Ni [9], Sn-Cu [10] Sn–Co [11], Sn–Sb [12], etc.

2

ACCEPTED MANUSCRIPT The interface between the active material and the inactive current collector has been found crucial for electrode performances. For this reason, to improve cyclic properties of Sn alloyed

T

with these elements, which are inactive towards Li and hence function as a matrix and buffer

IP

the volume expansion. These alloy materials provided a longer cyclability than that of pure

SC R

tin, but Sn-based alloy electrodes exhibit still dramatic loss in reversible capacity after several cycles. To enhance its electrochemical performance and cycle life, several carbonaceous nanocomposite Sn-based electrodes have been proposed [13, 14]. Several carbon allotropes

NU

as; amorphous carbon, hard carbon, graphite, carbon nanofibers, CNTs and graphene were

MA

used because of their low volume changes as the negative electrodes in Li batteries. All the carbon based materials were known to be effective in terms of buffering of the stresses during the electrochemical cycling together with their specific electrochemical contribution and

TE

D

extremely high conductivities. The buffering effects are expected to be higher in the CNTs compared with layered morphologies, nanofibers and other particulate type of carbonaceous

CE P

materials due to their excellent flexibility, extremely high modulus and fiber morphology. According to the composite mechanics rules, when the stress is stored in a matrix the load

AC

bearing capacity of the hollow fiber like morphologies are more beneficial than the layered geometries [15]. Multi wall carbon nanotubes (MWCNTs), with their large internal cavity, high electrical conductivity, extensive surface area, and material flexibility should be good buffer materials to prepare superior composite anode materials [15,16]. Uysal et al. [9] fabricated Sn–Ni/MWCNT composite by pulse electrodeposition. They indicated that the Sn–Ni/MWCNT composite anode provides a longer cyclability than that of Sn–Ni alloy. Huang et al. [17] reported that Sn–Co/MWCNT composite anode prepared by reductive precipitation solution of chelating metal salts within a CNTs suspension and the specific capacity of Sn-Co/MWCNT composite anode was much better than Sn–Co anode. Similarly,

3

ACCEPTED MANUSCRIPT recently in a work published on the Sn-Sb/CNT [18], it is studied to improve the cycling performance of the Sn-based anode materials.

T

There are many methods of preparing structured nanoparticles, such as electroless plating,

IP

electrophoresis, electrodeposition and self-assembly [19, 20]. In this study, to improve the

SC R

cyclability and capacity performances of a Sn electrode; Sn-Cu composites were prepared for the first time by electroless coating of tin on copper powders. Subsequently Sn-Cu powders

NU

were reinforced with different amounts of MWCNTs by mechanical alloying to improve the interface bonding between the active material and the MWCNTs. To the best of authors’

MA

knowledge there are no reports on the negative electrodes produced by the electroless Sn coating on the ductile Cu surfaces and investigate the effect of varying amount of MWCNTs

D

on the electrochemical performances of the Sn active materials. It is aimed not only to

TE

improve the electrode conductivity but also contribute the buffering of high mechanical

CE P

stresses caused by large-volume expansion during Li alloying.

AC

2. Experimental Details

Copper powders (60 m diameters in size supplied from Alfa Aesar) and MWCNTs (supplied by Arry Nano with the diameter of 50–60 nm and length of 10 m) were used in this experimental study for producing Sn-Cu/MWCNT nanocomposite electrodes. At the first step, the surface of pure copper powders were coated with tin by electroless deposition process. Before the Sn deposition process, the surfaces of pure copper powders were pretreated to obtain catalytic activity and then cleaned with acetone to remove any contaminants on the surfaces. Later, the surfaces of copper powders were micro-etched to provide sufficient bonding between Cu and Sn deposits. Following the micro-etching, copper powders were filtered, washed with distilled water several times and the pretreatment of 4

ACCEPTED MANUSCRIPT copper powders were completed after drying of activated powders for 10 h in a vacuum oven at 60 oC. After pretreatment process, the surfaces of Cu powders were coated with Sn by an

T

electroless process. The basic composition of the bath, and the plating conditions are shown in

IP

Table 1.All solutions were prepared with de-ionized water and reagent grade chemicals.

SC R

Plating process was carried out at 70 oC temperature and plating time kept constant at 10 min for all samples. The pH value of plating bath was controlled continuously during plating between 12-13 by using NaOH as a buffering agent. After the plating process, tin coated

NU

copper powders were washed up with distilled water and then dried at 60 oC in a vacuum oven

MA

for12 h.

At the second step, the surfaces of the MWCNTs were prepared, since the nanocomposites

D

based on the Sn-Cu powders reinforced with varying amount of MWCNTs. In this

TE

investigation, surface treatment of MWCNTs was carried out using a solution of nitric acid/sulphuric acid mixture to improve the dispersion of the CNTs into Sn-Cu matrix. The

CE P

acid treatment not only removes most of the metal catalyst but also produces carboxyl, aldehyde, and other oxygen containing functional groups on the surface of the MWCNTs and

AC

help in uniform dispersion of CNTs [21]. In the present investigation, MWCNTs were added into the solution of 2:1 nitric acid to sulphuric acid ratio in a glass bottle. The suspension was later filtered and neutralized with sodium hydroxide pellets. The MWCNTs were then subjected to repeated process of rinsing by distilled water until a pH value of 7 was obtained. The suspension was then filtered and baked dry at 80 oC for 2 h to remove any residues. The third step was dispersing MWCNTs into Sn-Cu powders. Three different nanocomposite compositions were prepared; Sn-Cu/MWCNT composites with amounts of 10 wt. %, 20 wt. %, and 40 wt. % MWCNTs. To compare the effect of Sn-Cu/MWCNT composites, Sn-Cu electrodes were also prepared at same conditions. MWCNTs and the Sn-Cu powders were placed in 250 ml stainless steel mixing jars for the fabrication of Sn-Cu/MWCNT 5

ACCEPTED MANUSCRIPT nanocomposite electrodes. Ball milling was carried out at 400 rpm in a Fritsch P7 planetary ball mill with a ball-to-powder weight ratio of 10:1 using stainless steel milling balls having

T

diameters of 10 mm. The methanol were added as a process control agent (PCA) in order to

IP

minimize cold welding of the copper particles and also to prevent powders sticking to the

SC R

balls and the jar walls.

The surface morphology of the Sn-Cu and Sn-Cu/MWCNT nanocomposites were

NU

characterized by SEM (JEOL 6060LV) equipped with EDS. Possible growth planes and the crystallographic relationship of Sn-Cu/MWCNT composites were performed by XRD method

MA

using a Rigaku D/MAX 2000 X-ray diffractometer.

The Sn-Cu/MWCNT nanocomposites were tested as anode active materials in Li-ion battery

D

CR2016 cells. To prepare the electrodes, 75 wt. % Sn-Cu/MWCNT nanocomposite powders

TE

and 10 wt.% carbon black were mixed with 15 wt.% PVDF binder dissolved in a N-methyl-2-

CE P

pyrrolidinone (NMP) solution. The resulting slurry was cast on a copper foil, pasted using a doctor blade, and dried at 120 °C in a vacuum oven for 12 h. The sample on the copper foil was then cut using a cutter disc. For comparison, tin coated copper electrodes were also

AC

prepared using the same conditions. Coin type CR2016 test cells were assembled in argon filled glove box, the prepared electrodes were used as working electrode, Li foil used as counter electrode, 1M LiPF6 dissolved in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 in volume) as the electrolyte. The working and counter electrodes were separated with polypropylene (PP) separator. Charge–discharge characteristics of the electrodes were tested between 0.02 V and 1.5 V at a constant current of 150 mAh g-1 (C/5) based on tin and carbon nanotube weight by MTI Battery Tester.

6

ACCEPTED MANUSCRIPT

T

3. Result and Discussion

IP

The bath solutions were prepared with NaH2PO2 as a reducing agent and sodium citrate

SC R

(Na3C6H5O7) served as the complexing agent for the reduction process [22]. To investigate the tin deposition on the surface of the copper powders, SEM studies were performed. The

NU

SEM micrographs in Fig. 1a and 1b show the microstructure of the uncoated copper powders and Sn coated copper powders, respectively. The morphology of the irregularly sized pure

MA

copper powder is also illustrated in Fig. 1a. Fig. 1b shows the SEM images of Sn coated copper powders with an average particle size of 10-30 μm. As can be seen in Fig 1b, copper

D

particle surfaces were successfully coated with a continuous tin layer. Moreover, a relatively

TE

continuous uniform and dense tin layer is observed on the surface of the copper powders. Especially, Sn–Cu/MWCNT composite electrode was developed cycling performance and

CE P

capacity of Li-ion batteries. Copper can not only be helpful to buffer the stresses caused by

AC

volume changes of Sn electrode during electrochemical reaction [23, 24].

Distribution of Sn and Cu atoms was revealed by X-ray color mapping carried out with a SEM-EDS, as shown in Fig. 2.

The EDS elemental mapping in Fig. 2 represents the

homogeneous distribution of Sn and Cu atoms in the composite Sn-Cu powder. The elemental mapping results indicates that Sn-coated copper particles are uniformly distributed on the copper surface (Fig. 2).

The coated particles were mounted into a polymeric resin and polished in order to observe the thickness and the continuity of the coating layer. Fig. 3 exhibits the SEM cross-section images 7

ACCEPTED MANUSCRIPT of Sn coated copper particles. The cross-section images show that the coating layer continuously surrounds the copper particles with a reasonably homogenous thickness, which

T

is approximately 2-3 m.

IP

Fig. 4 shows SEM micrographs of the produced Sn-Cu/MWCNT composite electrodes having

SC R

different weight fraction of the MWCNTs. As seen in Fig 4, SEM images of produced SnCu/MWCNT composite electrodes display clearly visible MWCNTs when carbon nanotube

NU

content is increased in composite structure. Increasing carbon nanotube content results in obtaining homogenously distributed pores between the MWCNTs. Increasing MWCNTs in

MA

the Sn-Cu/MWCNT composite leads to create a network like structure, which facilitates the infiltration and rapid transport of electrolyte [25, 26]. The homogenous distribution of the

D

MWCNT in the matrix and dispersion ability of carbon nanotubes is significant because the

TE

MWCNT are expected to bear the stresses during charge/discharge process. Because of the

CE P

high-volume increase in Sn-Cu composite electrodes, the electrodes are pulverized after a certain number of electrochemical cycles. As MWCNTs are very strong and flexible, they can carry the load, which is transferred from the Sn matrix materials. Therefore, it can be

AC

concluded that MWCNTs can also act as stress buffering components [9,17,23,27]. A uniform distribution of MWCNTs in the Sn-Cu matrix support to reduce the diffusion path for Li-ion and the sufficient conductivity provided by the MWCNTs network, which expected to enhance the electrochemical performance[28]. The obtained XRD patterns for the Sn-Cu/MWCNT structures are shown in Fig. 5. After coating, typical reflection peaks of tin were observed at 2θ values of 26.5°, 30.5°,31.9°, 43.8°, 55.3°, 62.5°,64.5°,72.3°, and 79.4° [29,30]. Therefore, the XRD patterns of the SiCu/MWCNT nanocomposite evidently confirms that Sn existed on the surface of the copper powders. It is clearly seen from XRD patterns of Sn-Cu/MWCNT nanocomposite electrodes that carbon reflection peak intensity at 2θ values of 26.3 increases when MWCNT content is 8

ACCEPTED MANUSCRIPT increased in composite structure. It is evident from the XRD analysis that introducing MWCNTs into the copper matrix does not show significant effect on the change of

T

preferential crystal growth of Sn.

IP

To investigate the electrochemical properties of nanocomposite electrodes, the CV curves of

SC R

tin coated copper and tin coated Cu/MWCNT composite electrodes were scanned at 0.5mVs−1 between 0.02 and 1.5V. The initial four scanning cycles are shown in Fig. 6. It is apparent that

NU

the integral peak area of Sn–Cu/40MWCNT composite is larger than that of Sn-Cu, indicating that the electrocatalytic activity of Sn–Cu/40MWCNT composite is the best. From Fig. 6a, it

MA

can be seen that, during the first scanning cycle, there are two reduction peaks and two oxidation peaks. The first reduction peak in the potential range 0.8–1.2 V is ascribed to the

D

formation of solid electrolyte interface (SEI) film on the surface of electrode [31], which

TE

disappears in subsequent cycles. During the first discharge process, the trapping of the lithium ions in the defected areas resulted in and pulverized tin particles [13, 17, 32]. The second and

CE P

third reduction peaks at 0.61 V and 0.35 V originating from trace pure Sn alloying with lithium. The second cathodic peak below 0.3 V can be assigned to the formation of Li4.4Sn

AC

from Cu6Sn5, according well with above discharge curves. From Fig. 6b it can be seen that the shape of curves are nearly overlapping, indicating that the Sn-Cu/MWCNT composite electrode shows better cycling performance. In the anodic curve, two peaks at 0.71 and 0.93 V are visible, which shifted to 0.73 and 0.96V in the second cycle, 0.76 and 0.99V in 4th cycle, and they are ascribed to the de-alloying process. In addition, the CV curves of the 3rd and 4nd cycle are almost overlapped, indicating the good electrochemical reversibility of the SnCu/40MWCNT composite electrode starting from the 2nd cycle. Fig. 7 shows the charge–discharge potential profile of Sn-Cu alloy , Sn-Cu/10MWCNT, SnCu/20MWCNT and Sn-Cu/40MWCNT type composite electrodes in the coin cells between 0.02 and 1.5 V vs. Li/Li+. The Sn-Cu anode shows a relatively highest discharge capacity of 9

ACCEPTED MANUSCRIPT 750 mAh g-1 and first charge capacity is only 447 mAh g-1 with a coulombic efficiency of 59.5 % (Fig.7a). Sn-Cu/10MWCNT type composite electrode demonstrates an initial

T

discharge capacity of 721 mAh g-1 and first charge capacity of mAh g-1 with a coulombic

IP

efficiency of 75 % (Fig. 7b). Sn-Cu/20MWCNT type composite electrode indicates an initial

SC R

discharge capacity of 690 mAh g-1 and first charge capacity of 544 mAh g-1 with a coulombic efficiency of 78.4 % (Fig. 7c). Sn-Cu/40MWCNT type composite electrode has initial discharge capacity of 690 mAh g-1 and first charge capacity of 545 mAh g-1 with a coulombic

NU

efficiency of 79.6 % (Fig. 7d). Consequently, as increased MWCNT content in composite

MA

structure, first initial discharge capacity of composite electrodes decreased. Moreover, first coulombic efficiency of composite electrodes increased. The capacity contribution of the MWCNTs when used as active and buffering component is not so clear. It is expected that the

TE

D

surfaces of the MWCNTs were partially coated with Sn active element since they pretreated with a solution of nitric acid/sulphuric acid mixture to improve the dispersion of the CNTs

CE P

into Sn-Cu matrix. Pure Sn electrode suffers severely from its poor cycle performance due to its abrupt volume expansion (up to approximately 300 %) during the charge/discharge

AC

process. This morphological change results in the cracking and crumbling of the electrode with the concomitant loss of electrical contact between interparticles, particles and the current collector, which results in poor cycling performance [33,34]. The ductile Cu and the MWCNTs with their excellent flexibility prevent cracking and loosing electrical contact. In Sn based electrodes, it was claimed that Sn atoms aggregated in certain voltage range during cycling, and the coexistence between the Sn aggregates and LixSn alloy phases caused fracturing, and eventual capacity fade upon cycling [35]. Since the MWCNTs prevented Sn agglomerate, ion and loosing electrical contact we have not measured irreversible capacity in our Sn-Cu/MWCNTs. These results show that as the MWCNT content in the Sn-Cu matrix is increased, the charge and discharge capacities of the composite electrodes increase, most

10

ACCEPTED MANUSCRIPT likely due to an improvement in the electronic contact of the Sn powders with the current collector, which prevents rapid pulverization of the electrodes during the electrochemical

T

reactions.

IP

Cycling performance and coulombic efficiency of Sn-Cu alloy electrode and produced Sn-

SC R

Cu/MWCNT composite electrodes at a current density 150 mA/g (C/5) are shown in Fig. 8. The cycle performance of the Sn-Cu electrode is compared with that observed for the Sn-

NU

Cu/MWCNT composite electrode. In the case of the Sn-Cu electrode, the columbic efficiency and cycle performance are the poorest among those for the three types of composite

MA

electrodes. Due to the mechanical fracture and electrical contact loss induced by volume expansion of Sn-Cu upon cycling, the specific capacity of Sn-Cu decreased continuously and

D

became less than 411 mAhg-1 after the 15th cycle, shown in Fig. 8. However, Sn-Cu/MWCNT

TE

composite electrodes show better cyclability and when MWCNT content is increased in composite structure, cycling stabilities of composite electrodes are increased. Furthermore,

CE P

when capacity retention of composite electrodes investigated, Sn-Cu/10MWCNT, SnCu/20MWCNT and Sn-Cu/40MWCNT composite electrodes show 42 %, 58 % and 67 %

AC

capacity retention after 30 cycles, respectively and the discharge capacity was obtained as 439 mAh/g even after 30 cycles in Sn-Cu/40MWCNT composite electrode. The use of tin coated copper powders can also contribute to a reduction of the intrinsically large volume change during the alloying and de-alloying. The enhanced capability is attributed to the high conductivity of MWCNTs, which makes a percolating network throughout the electrodes [17, 25, 36]. MWCNTs play an important role as a buffering agent to prevent the agglomeration of active materials as an electron conductor of one dimensional structure to promote charge transfer. Furthermore , the copper in Sn-Cu/MWCNT composite film is an inactive, which provides good binding properties and can also withstand the stresses attributed to alloying and de-alloying of tin with lithium due to its good electrical conductivity, mechanical strength and 11

ACCEPTED MANUSCRIPT chemical inertness [15, 27, 37]. As result of, Sn–Cu/40MWCNT composite film electrodes show better charge/discharge capacity and good cyclability and this makes them a good

T

candidate material for Li-ion rechargeable batteries.

IP

To investigate the effect of the MWCNT amount on the electrical resistance of the Sn-

SC R

Cu/MWCNT composite electrodes, EIS was employed to characterize the impedance properties of Sn-Cu and Sn-Cu/MWCNT electrodes before the electrochemical cyclic test on

NU

fresh electrode, and the results are presented in Fig. 9. The diameter of the EIS spectra represents the charge transfer resistance of the electrodes [38]. The diameter of the Sn-Cu

MA

electrodes is greater than that of the Sn-Cu/MWCNT composite electrodes. Moreover, when the MWCNT amount is increased in the Sn-Cu matrix, the diameter of the EIS spectra

D

dramatically decreases. This result indicates that when the MWCNT amount is increased, the

TE

charge transfer resistance of the electrode decreases due to the increasing electronic contact of the tin powders and the current collector, which was attributed to improving the conductivity

CE P

of the electrode. Moreover, as can be seen in Fig 9 increasing the MWCNT content in the SnCu matrix results in increasing the slope of the linear curve at low frequency values. This

AC

result occurs because the lithium-ion diffusion rate is increased when the conductive buffer inactive copper surfaces are coated with an active tin layer. This result indicates that the electronic conductivity of the Sn-Cu/40MWCNT composite electrode was improved due to the good electrical conductivity of the MWCNTs in the composite materials, where they served as both an active material and a conductor in the anode composite [25,26,37]. Moreover, MWCNTs prevent pulverization of electrodes leads to increasing capacity retention of composite electrodes. The EIS spectra of the produced electrodes were well fitted on an equivalent circuit (Fig. 10). On this circuit, Re is the electrolyte’s resistance, Rsei is the resistance of any film formed on the cathode surface (first HF semicircle), Rct is the charge transfer resistance of the electrode’s 12

ACCEPTED MANUSCRIPT reaction with lithium ions (middle frequency semicircle), and Wdif is the resistance of the lithium ion diffusion to the electrode (low frequency semi-circle) [35, 38, 39]. The Rct of the

T

electrodes was investigated after fitting the samples to the equivalent circuit and as seen in

IP

Table 2, the charge transfer resistances for the Sn-Cu, Sn-Cu/10MWCNT, Sn-Cu/20MWCNT

SC R

and Sn-Cu/40MWCNT after cycling were measured as 92.15 Ω, 76.23 Ω, 50.15 Ω and 33.22 Ω, respectively. Hence, when the MWCNT amount is increased, the diameter of the EIS spectra dramatically decreases. This result indicates that increasing the MWCNT amount

NU

result in increasing amount the MWCNT in the tin-copper matrix and thus, the charge transfer

MA

resistance of the electrode decreases due to the increasing electronic contact of the Sn-Cu matrix and the current collector, which was attributed to improving the conductivity of the

D

composite.

TE

To further investigate the electrode kinetic process, AC impedance spectra (Fig. 9a) were recorded to calculate the lithium-ion diffusion coefficient in electrode by using the Formula as

AC

CE P

shown in Eq. (1) [40]:

Eq. (1)

where D Li is the diffusion coefficient of lithium-ion, R is the gas constant (8.314), n is the number of transferred electrons, F is Faraday constant, A is the reaction area of electrode, σ is the coefficient of Warburg impedance, which can be obtained from the slope of straight line of Zὶ against ω1/2 , CLi is the lithium-ion concentration in the electrode. The calculated lithium diffusion

coefficients

for

the

Sn-Cu,

Sn-Cu/10MWCNT,

Sn-Cu/20MWCNT,

Sn-

Cu/40MWCNT electrodes are 1,85.10-12 cm2*s-1, 1,66.10-11 cm2*s-1, 1,79.10-10 cm2*s-1 and 2,25.10-10 cm2*s-1, respectively. In comparison, the diffusion coefficient for the Sn-Cu 13

ACCEPTED MANUSCRIPT electrode shows a lower value of 1,85.10-12 cm2*s-1. However, Sn-Cu/40MWCNT shows the highest lithium diffusion coefficient, which should be responsible for the good rate-capability,

T

as shown in Fig. 8. The Li+ diffusion coefficient in the Sn-Cu/40MWCNT nanocomposite is

IP

significantly improved compared to the Sn-Cu electrode. This enhancement could be

SC R

attributed to the Sn-Cu structure and conductive MWCNT, which serves as an efficient diffusion channel for Li+. Therefore, Sn–Cu/40MWCNT composite electrodes are able to a

NU

good candidate material for Li-ion rechargeable batteries. 4. Conclusions

MA

The surfaces of copper powder were successfully coated with tin by electroless coating.

D

MWCNT reinforced Sn-Cu matrix were successfully produced by powder technology method

TE

When MWCNT content is increased in composite structure, first initial discharge capacity of

increased.

CE P

composite electrodes increased. Moreover, first coulombic efficiency of composite electrodes

MWCNTs prevented pulverization of electrodes during cycling, and lead to increase capacity

AC

retention of nanocomposite electrodes. Increasing carbon nanotube content in composite structure caused decreasing electronic contact resistance between current collector and active material. The electrochemical performance of the Sn-Cu/MWCNT composite electrode has been compared with those of Sn-Cu alloy electrodes. Among the four, the Sn-Cu/40MWCNT composite electrode delivered the best performance. This must be attributed to the conductive nature of Cu substrate and the buffering role of the MWCNTs. Presumably, the void space in the MWCNTs network resulted bearing of the stresses caused from the volume expansion of Sn particles on the Cu powders to minimize the electrode pulverization. As a result, the 14

ACCEPTED MANUSCRIPT electrical conductive network is maintained even after a volume change in the Sn particles. The double components contribution of the Sn active material seems an encouraging way to

IP

T

extend the Sn electrodes to the commercial applications.

SC R

Acknowledgements

This work is supported by the Scientific and Technological Research Council of Turkey

NU

(TUBITAK) under the contract number 109M464–Improving the Capacity of Li-Ion Batteries by Using New Semi-conducting Metal Oxide Based Anodes. The authors thank the

MA

TUBITAK MAG workers for their financial support.

TE

D

References

(2012) 536–542

CE P

[1] R. Ma, Z. Lua, S. Yang, L. Xi, C. Wang, H.E. Wang, C.Y. Chung, J. Solid State Chem196

[2] M.J. Wang, C.F. Li, W.J.Lai, S.K. Yen Thin Solid Films 520 (2012) 6744–6751

AC

[3] M. Shafiei, A. T. Alpas, J.Power Sources 196 (2011) 7771–7778 [4] U. Kasavajjula, C. Wang, A.J. Applebyb, J Power Sources 163 (2007) 1003–1039 [5] G. Liu, X. Shen, K. Ui, L. Wang, N. Kumagai, J Power Sources 217 (2012) 108-113 [6] J. Chen, Materials, 6, (2013)156-183 [7] M. Alaf, D. Gultekin, H. Akbulut, Applied Surface Science 275 (2014) 244-251 [8] G. Liu, X. Shen, K. Ui, L. Wang, N. Kumagai, J. Power Sources 217 (2012) 108-113 [9] M. Uysal, H. Gul , A. Alp, H. Akbulut, Int. J.Hydrogen Energy, in press (2014) 1-8

15

ACCEPTED MANUSCRIPT [10] D. H. Nam, R. H. Kim, D. W. Han, H. S. Kwon, ElectrochimaActa 66 (2012) 126–132 [11] M. Valvo, U. Lafont, L. Simonin, E.M. Kelder, J Power Sources 174 (2007) 428–434

IP

T

[12] Z. Wang, W. Tian, X. Li, J. Alloy Compds 439 (2007) 350–354

SC R

[13] J.W. Park, J.Y. Eom, H.S. Kwon, Electrochemistry Communications 11 (2009) 596–598 [14] J. Yue, X. Zhao, D. Xia, Electrochemistry Communications 18 (2012) 44–47

NU

[15] Y.Wang, M.Wu, Z.Jiao, J. Y. Lee, Chem. Mater. 2009 21 (14) 3210–3215

MA

[16] G. Du, C. Zhong, P. Zhang, Z. Guo, Z. Chen, H. Liu, Electrochimica Acta 55 (2010) 2582–2586

D

[17] L. Huang, J. S. Cai, Y. He, F.S. Ke, S.G. Sun, Electrochemistry Communications 11

TE

(2009) 950–953

CE P

[18] W. X. Chen, J. Y. Lee, Z. Liu, Electrochemistry Communications 4 (2002) 260–265 [19] Y.Q. Kang, M.S. Cao, J. Yuan, L. Zhang, B. Wen, X.Y. Fang, J. Alloys Compds 495

AC

(2010) 254–259

[20] M. Hamadanian, V. Jabbari, A. Gravand, M. Asad, Surf. Coat Tech. 206 (2012) 4531– 4538

[21] X. H. Chen, C. S. Chen, H. N. Xiao, H.B. Liu, L. P. Zhou, S. L. Li, G. Zhang, Tribology International 39 (2006) 22–28 [22] G. Z. Zou, C. M. Sheng, L. Zhang, J. Li, H. Xu, Y. J. Chen, Surf. & Coat. Tech. 201 (2006) 108–112 [23] R.M. Gnanamuthu, S. Mohan, C. W. Lee, Mater.Lett 84 (2012) 101–103 [24] L. Xue, Z. Fu, Y. Yao, T. Huang, A. Yu, Electrochimica Acta 55 (2010) 7310–7314 16

ACCEPTED MANUSCRIPT [25] T. Cetinkaya, M.O. Guler, H. Akbulut, Microelectronic Engineering 108 (2013) 169–176 [26] H. Zhang, H. Song, X. Chen, J. Zhou, H. Zhang, Electrochimica Acta 59 (2012) 160–167

IP

T

[27] Y.R. Jhan, J.G. Duh, S.-Y. Tsai, Diam. Relat. Mater. 20 (2011) 413–417

SC R

[28] L. Congxiang, F. Yu, L. Hong, Y. Yi, T. B. Kang, T. Edwin, Z. Qing, Carbon 2013, 63 54–60

[29] C. Zanella , S. Xing, F. Deflorian, Surface & Coatings Technology 236 (2013) 394–399

NU

[30] W. Pu, X. He, J. Ren, C. Wan, C. Jiang, Electrochimica Acta 50 (2005) 4140–4145

MA

[31] L. Shen, X. Guo, X. Fang, Z. Wang, L. Chen, J. Power Sources 213 (2012) 229–232 [32] L. Shuzhao, Z. Xuefeng, L. Peichao, Y. Weishen, W. Haihui, J Solid State Chem 2011;

D

184: 1400–1404.

TE

[33] M. Uysal, T. Cetinkaya, A. Alp, H. Akbulut, Applied Surface Science 290 (2014) 6– 12

CE P

[34] M. Alaf, H. Akbulut, Journal of Power Sources 247 (2014) 692-702 [35] T. Cetinkaya, M. Uysal, M. O. Guler, H. Akbulut, A.Alp, Powder Technology 253

AC

(2014) 63–69

[36] Z. Jin-Liang, Z. Ya-Pu Composites: Part B 43 (2012) 76–82 [37] L. C. Ming, Z. R. Yuan, L. W. Shan, Z.L. Zhi, H. S. Jun, R. M. Min, X. J. Xia, Trans. Nonferrous Met. Soc. China 17(2007) 934-936 [38] Z.Y. Zeng, J.P. Tu, X.H. Huang, X.L. Wang, J.Y. Xiang, Thin Solid Films 517 (2009) 4767–4771 [39] Z. P. Guo, Z.W. Zhao, H. K. Liu, S. X. Dou, Carbon 2005; 43: 1392–1399 [40] J. Wang, H. Zhao, X. Liu, J. Wang, C. Wang, Electrochimica Acta 56 (2011) 6441–6447

17

ACCEPTED MANUSCRIPT Table Captions Table 1. Overview of electroless coating parameters for preparation of Sn-Cu coatings.

T

Table 2. Kinetic parameters of Sn-Cu and Sn-Cu/MWCNT composite electrodes after 30

SC R

IP

cycles.

NU

Figure Captions

Fig.1. SEM images of (a) uncoated copper powders, (b) Sn coated Cu powders.

MA

Fig.2. SEM image of Sn-Cu powder; a) SEM image, b) X-ray color mapping for copper c) X-

D

ray color mapping for Sn and d) EDS analysis of Sn-Cu powders.

TE

Fig.3. SEM cross-section images of the Sn coated copper powders.

CE P

Fig.4. SEM images of the Sn-Cu/MWCNT composite electrodes prepared at different amount of MWCNTs; a) 10 wt. % MWCNT, b) 20 wt. % MWCNT, c) 40 wt. % MWCNT. Fig.5. XRD patterns of the Sn-Cu/MWCNT nanocomposite electrodes produced at various

AC

amount of MWCNT.

Fig.6. Cyclic voltammetry curves of a) Sn-Cu b) Sn-Cu/40MWCNT composite electrodes. Fig.7. Charge–discharge curves of the electrodes; a) Sn-Cu , b) Sn-Cu/10MWCNTs, c) SnCu/20MWCNTs and d) Sn-Cu/40MWCNTs. Fig.8. Cyclic test data of Sn-Cu alloy and Sn-Cu/MWCNT composite electrodes including different amounts of MWCNT. Fig.9. Impedance response of the cells containing the Sn-Cu and Sn-Cu/MWCNT composite electrodes (a) before charge–discharge and (b) after 30 cycles. 18

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig.10. The equivalent circuit used to model the impedance spectra in Fig. 9.

19

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Table 1

20

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Table 2

21

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

Figure 1

22

AC

Figure 2

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

23

AC

Figure 3

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

24

CE P AC

Figure 4

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

25

AC

Figure 5

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

26

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

Figure 6

27

AC

Figure 7

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

28

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 8

29

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

Figure 9

30

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 10

31

ACCEPTED MANUSCRIPT Highlights

T

► Sn-Cu/MWCNT anodes were produced by high energy mechanical milling and electroless method.

IP

► MWCNT increases electronic contact between current collector and active material.

AC

CE P

TE

D

MA

NU

SC R

► Discharge capacity of Sn-Cu/40MWCNT composite found 439 mAh/g even after 30 cycles

32