The electrode characteristics and modified surface properties of V0.9Ti0.1 alloy sintered with Ni powder

Journal of

ALLOYS AND COMPOUNDS ELSEVIER

Journal of Alloys and Compounds 231 (1995) 650-654

The electrode characteristics and modified surface properties of Vo.9Tio.1 alloy sintered with Ni powder Dong-Myung Kim*, Ki-Young Lee, Jai-Young Lee Department of Materials Science and Engineering, Korea Advanced Institute of Science and Technology, Kusong-dong 373-1, Yusung-gu, Taejon, South Korea

Abstract

The Vo9Tio.1 alloy absorbs a large amount of hydrogen, but it has never been used for the anode material for a Ni-MH battery due to the poor discharge behavior in KOH electrolyte. V0.9Ti0.1 alloy powder has been sintered with Ni powder to provide a catalytic effect on the absorption/desorption of hydrogen in KOH electrolyte. The optimal sintering condition was 5 min at 900°C and the amounts of Ni powder were varied to investigate the effects of surface modification on the discharge characteristics. All electrodes sintered with Ni powder were fully activated within 10 cycles in KOH electrolyte. The discharge capacities of the electrodes showed a maximum behavior with Ni content of the sintered alloy. The optimal amount of Ni powder showing the highest discharge capacity of 302 mA h g 1 was found to be 25 wt.%. From the results of the various analysis such as scanning electron microscopy (energy dispersive spectroscopy) and X-ray, it is found that VNi 3 is formed during sintering. The V N i 3 phase formed on the surface of Vo.9Ti0. l particles provides the optimum discharge capacity for the electrode. In accordance with the Brewer-Engel theory, VNi 3 phases are found to be highly electrocatalytic, which is reflected in an increase of the exchange current density and a decrease of the discharging overpotential. In order to develop an anode material using metal hydride which has a large hydrogen storage capacity but is inactive for electrochemical hydrogenation, it has been suggested as a new process to modify the surface by sintering with Ni powder.

Keywords: Surface; Modification; Sintering; Electrode; Nickel

1. Introduction The nickel-metal hydride (Ni-MH) batteries have received attention in recent years because of several advantages over nickel-cadmium batteries; e.g. higher energy density, longer cycle life, the capability of overcharging and overdischarging, and the absence of the poisonous heavy metal [1-3]. A lot of works [4-7] have been carried out to develop anode materials using AB2-type and A B : t y p e M H alloys. In the alloy design, these M H alloys should contain nickel as a catalyst. In general, at the discharge current of 50 m A g-l, the discharge capacity of ABs-type and A B 2type M H alloys are about 250 m A h g-1 and 350 m A h g - l , respectively due to the limited number of nickelcontaining alloy systems. It has not been so far attempted to utilize a M H alloy which does not contain nickel. In this work, a new process has been introduced as a guideline to * Corresponding author. 0925-8388/95/$09.50 © 1995 Elsevier Science S.A. All rights reserved SSDI 0 9 2 5 - 8 3 8 8 ( 9 5 ) 0 1 7 4 6 - 1

develop an anode material using a M H alloy which has a large hydrogen storage capacity but is inactive for electrochemical hydrogenation in K O H solution. V0.9Ti01 alloy is a very attractive material for the anode of a N i - M H secondary battery because of its large hydrogen storage capacity [8]. However, the alloy has not been used as an anode material because of its poor discharge behavior in K O H electrolyte. The V0.9Tio.1 alloy powder and Ni powder have been sintered to modify the alloy surface by providing a catalytic nickel layer on the surface. The discharge characteristics of V0.gTi0n alloy has been investigated and the dischargeability has been drastically improved by the surface modification.

2. Experimental details

2.1. Preparation of the alloys and electrodes W0.9Tio.1 alloys were prepared by arc melting in an

D.-M. Kim et al. / Journal o f Alloys and Compounds 231 (1995) 650-654

argon atmosphere. The alloys were turned over and remelted several times to obtain a homogeneous structure. Then the alloys were mechanically pulverized to prepare powder below 400 mesh (less than 37 ~m). The crystal structures of the alloys were confirmed as BCC structure by X-ray analysis. V0.9Ti0.1 alloy powders (37/zm) and Ni powders (10/.~m) were mixed together in the weight ratios of 1:1 (100 wt.% Ni per alloy), 1:0.5 (50 wt.% Ni per alloy), 1:0.25 (25 wt.% Niperalloy),l:0.125(X2.5wt.%Niperalloy)and pressed to pellets with a diameter of 10 mm and a thickness of about 1 mm at the compacting pressure of 10 tons cm -2. These pel][ets were sintered in a vacuum of about 10 -2 to 10 -3 Torr at 900°C for 5 min. The pellet was charged into the quartz tube which was connected to the vacuum system, and the quartz tube was slid into the furnace which had been heated up to 900°C. Shortly after 5 min of sintering, the quartz tube was removed from the furnace and quenched in the water bath. In order to identify the crystal structures of the sintered electrodes, X-ray analysis was performed. The hydrogen storage perfo:rmance was measured by determining pressure-composition-temperature (PCT) curves using a Sievert's-type apparatus. The surface morphology of the sintered electrodes was examined by scanning electron microscopy (SEM) and their chemical composition was characterized by energy dispersive spectroscopy (EDS) analysis,

1.o.

30°C

651

,-.--,~-.'-7¢-~--,'-~Is~'_ ~-~ / . /

0.8.

0.8 O-3o.,.

/

~_.!~.&//~./C, NNi~ in)

0.2.

o.o

_ .-.-o-o----o--o

0

~

o

.

lO0~.%Ni(Notsintered) o

o

~ ~ Cycle number

~

1'0

Fig. 1. A c t i v a t i o n b e h a v i o r s of W0.9Ti0.1 - x wt.% Ni (x = 25, 50, 100)

alloys.

in Fig. 1. The electrode mixed with Ni powder and pressed without sintering was not activated even after the tenth cycle. However, all of the electrodes sintered with Ni powder were fully activated within the tenth cycle. Furthermore, the activation behavior was improved by increasing the nickel content.

2.2. Electrochemical measurements 3.2. Discharging efficiency A half-cell was constructed using platinum wire as a counter electrode and a mercury/mercury oxide (Hg/ HgO) electrode as a reference electrode in the 30 wt.% K O H electrolyte. The charge-discharge was controlled by the potential of the working electrode, i.e. sintered pellet, with respect to a reference electrode with an automatic galvanostatic charge-discharge apparatus. Cycle tests were conducted at 30°C by repeatedly charging and discharging at 50 mA g-1 (V0.9Ti0.l alloy) for 12 h with a resting time of 30 s. The cutoff voltage w a s - 0.75 V vs. Hg/HgO. The discharge capacity was expressed in milliampere hours (mA h) per gram of V0.gTi0.1 alloy, It was shown that the kinetics of the charge-transfer reaction, i.e. exchange current density, was able to be characterized electrochemically using the linear polarization method at an open-circuit potential (OCP) between - 1 0 mV and +10 mV.

3. Results and discussion 3.1. Activation of the sintered electrodes The activation behavior of the electrodes is shown

The pressure-composition desorption isotherms measured at 30°C for the sintered Wo.9Ti0.1- x wt.% Ni (x = 0, 25, 50, 100) are shown in Fig. 2(a). In the V0.9Ti0.1alloy which contains no Ni, the hydrogen absorption capacity (H/Malioy: Malloy = average atomic weight of the alloy) and the effective hydrogen content (A[H/M]) are about 1.8 and 0.9, respectively. However, a clear decrease in the maximum hydrogen absorption capacity and the effective hydrogen content have been observed in the W0.9Ti0.1 - x wt.% Ni (x = 25, 50, 100) alloy with increasing Ni content. The theoretical discharge capacity per unit weight of the alloy was calculated as described in Ref. [4]. The theoretical discharge capacities and the discharge characteristics of W0.9Ti0.1 - x wt.% Ni (x = 12.5, 25, 50, 100) after activation cycles are shown in Figs. 2(b) and 2(c) and Fig. 4. In Fig. 2(b), the absolute value of the discharge potential was increased with increasing Ni content, and the discharge capacities of the electrodes showed a maximum value with respect to the Ni content, as shown in Fig. 2(c). The optimal amount of Ni powder showing the highest discharge capacity of 302 mA h g-1 has been found to be 25 wt.%. The discharge efficiencies of the W0.9Ti0.1 alloy

D.-M. Kim et al. / Journal o f Alloys and Compounds 231 (1995) 650-654

652

(a)

1007

0.

0,1 ~

o.o,

O) 400t

80

0.6 0.8 1.0 1.2 1.4 1.6 1.8 .

.

.

.

.

.

.

.

.

.

.

.

.

.......

1

2oo

• (h)

Nl~S

•'P 0.90" k""',r"

....

i ..........i

/

50%2$

i ~

I-4o

I2°

"

/.--o--Th~tlc,Idlechmrge capacit(Cm/[ y Dlachergecapaclty(C) I.0 ()~20DIscharge4' effiOciencY6' 8'0 0" 1{)0 /

%

'

,I ~

--~-->~0.800'85"

- ~ %

~. , 0.75

30°C

NI wt.%

" sb ~ 6 o 1~o 2 6 o 2~o

a6o 3so

Capacity (mAhlg) Fig. 2. Thermodynamic and electrochemical properties of V o . g T i o j - x wt.% Ni. (a) Desorption P C T curves for Vo.gTio. 1 - x wt.% Ni (x = 0, 25, 50, 100) alloys sintered at 900°C for 5 min. (b) Discharge curves for VogTio ~ - x wt.% Ni (x = 12.5, 25, 50, 100) after activation cycling. (c) Theoretical discharge capacity, discharge capacity and discharge efficiency of Vo.gTioA alloy with Ni content.

electrodes are also shown in Fig. 2(c). The discharge efficiency is the percentage of the discharge capacity (measured at 50 mA g-l) to the theoretical capacity (Cth). The discharge efficiency of Vo.9Ti0.1 electrodes was improved with increasing Ni.

3.3. X-ray and SEM(EDS) analysis The X-ray diffraction patterns of Vo.9Tio.1 alloy sintered with 100 wt.% Ni powder are shown in Fig. 3(a). The W0.9Ti0.1 alloy and Ni powder peaks represent

(b) (a) •

VMTiM

,,

• Ni

ii C

•

5

rain

•

~

20 m ~

~v~ IO0

~ ,

~o

~o

~__

so

~o

Two theta

;o

80

~

.

.

!

i

~,,

eo

~ 40 o

TI

20

~ o ,,

PrO Distance

(l~m)

.

,

.

t~)

Fig. 3. Characterization of the V N i 3 phase in V o . 9 T i o j - x wt.% Ni alloy sintered at 900°C for 5 min. (a) X-ray diffraction patterns of Vo 9Tioj - 100 wt.% Ni and VNi 3. (b) SEM image of Vo 9Tio 1 - 25 wt.% Ni. (c) EDS analysis of Vo.gTio.x - 25 wt.% Ni.

D.-M. Kim et al. / Journal of Alloys and Compounds 231 (1995) 650-654

the typical BCC and FCC structures, respectively, and there appears four new peaks representing the V N i 3 phase. As the sintering time increases, the amount of VNi 3 phase increased. As this phase cannot absorb hydrogen because of too high a nickel content, the formation of the V N i 3 phase will decrease the theoretical discharge capacities of the electrodes, T h e surface morphology of the V0.9Ti0.1 alloy sintered at 900°C for 5 rain with 25 wt.% Ni powder is shown in Fig. 3(b). The second phase is observed between V0 9Ti0.1 and a Ni phase with the thickness of 2-3 /zm. The second phase has been proved to be VNi 3 from energy dispersive spectroscopy (EDS) analysis as shown in Fig. 3(c). In the second phase the ratio of Ni to V is approximately three. At the same sintering condition (at 900°C for 5 rain), as the amount of Ni powder increases, the electrochemical efficiencies are improved by V N i 3 and Ni phase, whilst the hydrogen storage capacity decreases. As a result, the discharge capacity shows a maximum value with respect to Ni content. It is well known that the V N i 3 hardly absorbed hydrogen, so the theoretical discharge capacity of the electrode is decreased by the formation of VNi 3. However, as it has an excellent catalytic activity [9] on the absorption/desorption of hydrogen in K O H electrolyte, the electrochemical efficiencies are improved by the V N i 3 and Ni phases,

653

during activation cycling by continuous micro-cracking. So, in this experiment, the values of the exchange current densities are normalized by the weight of V0.9Ti0.1 alloys, i.e. current g-i of V0.9Ti0.1 alloy. The exchange current densities of V0.9Ti0.1 - x wt.% Ni (x = 0, 12.5, 25, 50, 100) are shown in Fig. 4(a). The exchange current density increased with the increase of Ni content.

3.5. Discharge overpotential In order to identify the electrochemical charactefistics of the VNi 3 phase, the discharge overpotential of the electrodes has been measured. The discharge overpotential can be written in the following form [4,10,11]: Discharge over potential = IV,eal- Veql

(1)

Veq---0.9409- 0.0301 IOgPn2 (VS. H g / H g O , at 30 °C, pH--14)

(2)

The values of Veq are obtained by the potential calculated from the pressure at the middle of the plateau region in each PCT curve by the Nernst equation (Eq. 2). The values of V r e a l a r e directly obtained by the potential at the middle of each discharge curve. The discharge overpotentials of V0.9Ti0.1 - x wt.% Ni (x = 25, 50, 100) are shown in Fig. 4(b). As the amount of Ni powder increases, the discharge overpotential decreases. However, it is not clear whether this decrease of discharge overpotential was due to the V N i 3 phase or due to the sintered Ni phase. In order to clarify which phase is more dominant, the amount of the W N i 3 phase has been changed keeping the total weight of Ni powder constant. The

3.4. Exchange current density The catalytic activity of the V N i 3 phase has been clarified in detail by measuring the exchange current density of the electrodes. In general, the exchange current density is calculated on the basis of the surface area of electrodes. However, it is hard to measure the surface areas of the electrodes because they increased (a)

(b) 10

1.0.

30°C

-o.e2

30°C

9 0.8

.-0.84.

x

0

o.e.

o o "

~ -o.ee -r*

s

_--~ ._

~

~.

=." o.,-

~ -o.~.

=

-~

>

~ "N 0.2-

o

~ . -0.90.

. .........

Veq

oo-

-o.9:

d

.6

=-'"'~

~

,=

g

R 5

"

io

"

;o

'

6'o

"

8~

"1~o

2b

"

4b

NI Wt%

Fig. 4. (a) Exchange current densities of Vo.gTio. ~ - x

overpotentials of

Vo.gZio. 1 -

"

6'0

"

8'0

"

I~o

NI wt.%

wt.% Ni (x = 0, 12.5, 25, 50, 100) alloys sintered at 900°G for 5 mira (b) Discharge

X wt.% Ni (x = 25, 50, 100) alloys sintered at 900°C for 5 min.

D.-M. Kim et al. / Journal of Alloys and Compounds 231 (1995) 650-654

654

(a) 100

(b)

3ooc

30Oc

o.9510. "r' 0.902 E

:~ " > o.ss-

1.

"~

oso-

•

"~

0.1.

'

min

o" '

0.75-

o 0.01

'

s'o 1 ~ o

1~o

'

2~o

2so

Capacity (mAlt/g) 0.6

0.8

1.0

1.2

1.4

HIM

Fig. 5. Thermodynamic and electrochemical properties of Vo.9Ti0.1 - 100 wt.% Ni alloy sintered at 900°C for 5 min and 25 min. (a) Desorption PCT curves. (b) Discharge curves.

P C T curves and discharge curves of the V0.9Ti0.1 alloy sintered with 100 wt.% Ni powder at 900°C for 5 min and 25 min are shown in Figs. 5(a) and 5(b), respectively. The values of Veq are obtained from these P C T curves. The discharge overpotential of V0.9Ti0.1 - 100 wt.% alloy sintered for 5 and 25 min are 0.072 V and 0.057 V, respectively. As the sintering time increases, the discharge overpotential decreases. So, the decrease of the discharge overpotential is thought to be attributed to the increased amount of the VNi 3 phase.

4. Condusions

analysis, the second phase was observed and it has been found to be V N i 3. As VNi 3 hardly absorbs hydrogen, the theoretical capacity of the electrode is decreased by the formation of the VNi 3 phase. The electrochemical catalytic effects of V N i 3 a r e found to be more dominant than those of the sintered Ni phase.

References

[1] H.F. Bittner and C.C. Badcock, J. Electrochem. Soc., 130 (1983) 193C. [2] J.J.G. Willems and K.H.J. Bushow, J. Less-Common Met., 129 (1987) 13.

By surface modification, V0.9Ti0 1 alloy which has not

[3] M.A. Fetcenko, S.Venkatesanand S.R. Ovshinsky, Proc. Int.

previously been used for anode material owing to the poor discharge characteristics, has been successfully hydrogenated in KOH solution. In this new surface modification process, W0.9Ti0.1 alloy powder has been sintered with Ni powder at 900°C for 5 min changing

96, [4] J.J.G.Willems,Philips J. Res., 39(1)(1984). [5] S. Wakao, H. Sawa and J. Furukawa, J. Less-Common Met., 172-174 (1991) 1219. [6] T. Sakai, H.Yoshinaga, H. Miyamura, N. Kuriyamaand H.

the weight of Ni powder. The electrodes have been activated within ten cycles in KOH electrode. As the amount of nickel powder increased, the electrochemical efficiencies were improved, whilst the hydrogen storage capacity decreased. As a result the discharge

[7] S.R. Kim and J.Y. Lee, J. Alloys Comp., 210 (1994) 109-113. [8] J.F. Lynch, A.J. Maeland and G.G. Libowitz,Z. Phys. Chem., 145 (1985) 51. [9] M.M. Jaksic, Int. J. Hydrogen Energy, 12(11) (1987) 727. [10] C. Iwakura, Y. Kajiya, H. Yoneyama, T. Sakai, K. Oguro and

capacity showed a maximum value of 302 mA h g-1 at 25 wt.% of Ni content. From X-ray and SEM(EDS)

[11]

Symp. on Metal-Hydrogen System, Upssala, June 1992, PII, p.

Ishikawa, J. Alloys Comp., 180 (1992) 37-54.

H. Ishikawa, J. Electrochem. Soc., 136 (1989) 1351.

S.R. Kim, Ph.D. Thesis, KAIST, Taejon, South Korea, 1993.