electrochromic properties of tungsten- molybdenum …gaia.lbl.gov/btech/papers/39078.pdf · 234 a....

Solar Energy Materials and Solar Cells 28 (1992) 233-247 North-Holland

Solar Energy Materials and Solar Ceils

Electrochromic properties of tungsten- molybdenum oxide electrodes

A. Pennisi, F. Simone Dipartimento di Fisica dell'Universitft di Catania, Gruppo Nazionale di Struttura della Materia del C.N.R.-Catania, Corso Italia, 57, 95129 Catania, Italy

and

C.M. Lamper t

Lawrence Berkeley Laboratory, Energy and Environment Division, Building Technologies Program, MS 62-203, University of California, 1 Cyclotron Road, Berkeley, CA 94720, USA

Received 6 June 1992; in revised form 6 August 1992

Tungsten oxide and molybdenum doped tungsten oxide electrochromic films have been electrochemically prepared from a metal peroxide bath and experimentally tested. They were characterized by electrical and optical methods, using cyclic voltammograms and calculating the charge capacity as a function of the number of deposition cycles.

The color of the electrochromic WO 3 film is strongly altered by doping it with molybdenum. Electrochromic W-Mo oxide films were fabricated by adding different percentages of molybdenum to tungsten oxide, ranging from 0.5% to 7% weight of deposition solution. Normal optical transmittance was also measured and evidence is given of different behavior of the switching for oxidation and reduction cycle, so that coloration occurs in a shorter time than the bleaching. The time constant of the process tends to increase as a function of the molybdenum oxide doping concentration.

1. Introduction

Electrochromic materials are very promising to realize smart windows, displays and for large-area applications [1-3]. Experimental research over the past several years has been devoted to the perfection of the electrochromic and associated electrodes including the optimization of the switching time and the chromatic properties of the electrodes in solid state electrochromic devices.

A powerful and experimented technique, in this sense, is the insertion of dopant ions in the crystalline lattice of the electrochromic material, in order to modify its physico-chemical properties. One of the most studied and promising material

Correspondence to: A. Pennisi, Dipartimento di Fisica deli'Universith di Catania, Gruppo Nazionale di Struttura della Materia del C.N.R.-Catania, Corso Italia, 57, 95129 Catania, Italy.

0927-0248/92/$05.00 © 1992 - Elsevier Science Publishers B.V. All rights reserved

2 3 4 A. Pennisi et al. / Tungsten-molybdenum oxide electrodes

appears to be tungsten oxide (WO 3) [4-7]. The possibility to improve the dynamical answer of the transition by doping, has been widely discussed in literature. The use of lithium as dopant [8,9] seems to improve the infrared reflectance, while the use of molybdenum [10-12] permits to modify the chromatic properties.

The electrochemical deposition technique, useful for the preparation of large and uniform electrochromic layers, appears very adapt for the addition of dopant elements.

2. Experimental

An investigation was made on the electrical and optical characteristics of electrochromic (EC) tungsten oxide and molybdenum doped tungsten oxide films. The films were electrochemically prepared from a metal peroxide bath.

Tungsten oxide EC films were deposited electrochemically from 0.1 M solution of tungsten peroxide obtained by preliminar dissolution of 3 g of tungsten metallic powder in 200 cm 3 of acqueous solution containing 2% H20 2 [13-15]. The substrates were indium tin oxide (ITO) glass electrodes. A cyclic triangle waveform potential ranging from - 400 up to - 700 mV versus a standard calomel electrode (SCE) with a scanning rate of 100 mV/s was used per deposition. The ITO conducting glasses (2 x 4 cm) were furnished by Donnelly Corporation Holland (MI, USA) and had a sheet resistance of 10 l l / [3 . The voltammetric characteristic during deposition was recorded. In fig. 1 the n-cyclic I - V pattern is shown for tungsten oxide films, deposited respectively with n = 5, 20, 50 and 100 cycles. As shown in fig. 1, the exchanged charge, that can be calculated by the loop area, reaches its maximum when deposition utilizes a scan number between 50 and 100.

- 0 . 6

v

-b2 ¢D

<.9

- - 0 . 6

-1 .1~ '

' l [ -- 5 c y c l e s

r ' ' ' ' r ' " ' ' I . . . . I I t '

2 0 c y c l e s

- 0 . 7 - 0 , 4 Potential (V vs. SCE)

Fig . 1. V o l t a m m e t r y o f E C t u n g s t e n o x i d e d e p o s i t i o n p r o c e s s f r o m a p e r o x i d e s o l u t i o n f o r a d i f f e r e n t

n u m b e r o f cyc l e s . T h e a c t i v e e l e c t r o d e a r e a is 5 c m 2.

_ 5 0 c y c l e s

, I . . . . f . . . . I . . . . I , , -0.7 -0.4

A. Pennisi et al. / Tungsten-molybdenum oxide electrodes 235

0

v ..~ - 5

5 L,

CD

- 5

" - ' ' '1 . . . . I . . . . I . . . . I . . . . I . . . . i 50 cycles

~Q-3.5

150 c y c l e s

AQ=4. I

¥,,,I .... I .... I .... I .... I ....

- 0 . 5 0 0 . 5

.... I~=:.t .... I .... I .... r "_

6Q-8.8 _

200 c y c l e s

. . . . I . . . . I . . . . I . . . . I . . . . I, , , -r

- 0 . 5 0 0 . 5

P o t e n t i a l (V vs . SCE)

Fig. 2. Voltammetry in 0.1 M H2SO 4 of tungsten oxide electrodes deposited with a different number of cycles. The scan rate is 50 m V / s , The charge capacity, AQ (in m C / c m 2) for each curve is indicated.

The electrode properties were measured in 0.1 M H 2 S O 4 , using a calomel reference and a platinum wire counter electrode. Typical cyclic voltammograms are shown in fig. 2. A cyclic triangle potential ranging from 0.6 to -0 .6 V and a scan rate of 50 mV/s was used. During each cycle the EC layer switches from the colored to the bleached phase and vice versa. Biasing potentials were generated by a PAR 173/276 potentiostat, driven by a IEEE-488 Data Acquisition System. Both current and voltage were monitored during the deposition and recorded by a Linseis LY 1600 X - Y recorder. The curve in fig. 2 for 100 deposition cycles exhibits the highest charge capacity (8.8 mC/crn2). The calculated charge capacity for each curve is indicated. The active electrode area is 2.5 cm x 2 crn, because a portion of ITO glass substrate, used for electrical contact, is uncoated.

Normal optical transmittance, over the 350 to 1100 nm wavelength range, was measured in situ in 0.1 M H 2 S O 4 solution. Optical measurements were made using a Perkin-Elmer 'Lambda 2' UV/VIS spectrophotometer equipped with a special quartz cell (1 × 5 x 5 cm, path length 1 cm). This cell allows the insertion of the 2 × 4 cm EC films normally to the light beam and also a platinum wire counterelectrode and a micro SCE reference electrode. The EC layer was biased at 600 mV to obtain the bleached state (fig. 3), and at -600 mV for the colored one (fig. 4). The curves show the optical transmittance for samples prepared using a different number of scans for the deposition. The bleached phase is not influenced significantly to 150 deposition cycles. Increased absorbance is evident only for sample (d) at 200 cycles, for wavelengths greater than 800 nm. The small variations observed in the interference fringes are due to differences in the film thickness. The position of the interference fringes indicated the thickness of the films. By recording the position of maxima as a function of wavelength, we obtained a typical value of about 2000 ~k for 100 deposition cycles. Large differences of the interference maxima are noticed in the optical transmittance for the same elec-

236 A. Pennisi et aL / Tungsten-molybdenum oxide electrodes

120

100

80

6o

~ 4 0 ,t..,

20

I . . . . I . . . . I . . . . I a - 50 c y c l e s b tO0 "

. ~ c - 150 " -

d - 2 0 0 " b

o I , , , , { . . . . I . . . . I , 4 0 0 6 0 0 8 0 0 10o0

W a V e l e n g t h ( n m )

Fig . 3. N o r m a l t r a n s m i t t a n c e o f E C t u n g s t e n ox ide f i lms o f v a r i o u s t h i cknes s in t he b l e a c h e d s ta te . T h e

f i lms a r e b i a s e d a t 600 m V ve r sus S C E a n d a re p l a c e d in a mic roce l l c o n t a i n i n g 0.1 M H 2 S O 4

e lec t ro ly te .

trodes in the co lored phase ( see fig. 4), they are related to the effect ive thickness o f the optical ly active e l ec trochromic layer.

A n ion backscatter ing spectrometry ( IBS) analysis, us ing 2 M e V H e ÷ ions, was carried on e l ec trochromic e lec trodes in order to verify thickness [16]. A typical IBS spectrum is s h o w n in fig. 5. T h e arrows indicate the energy o f ions backscat tered by a toms o f different spec ies (W and Sn). T h e width o f the box at higher energy

120 ~, [ . . . . [ . . . . [ . . . . I t [ a - 50 c y c l e s

b - 100 " i00[--- c - 150 " --{

a o d 60

4 0 b

2O

o , . . . . 400 600 800 I000

W a v e l e n g t h ( n m )

Fig. 4. N o r m a l t r a n s m i t t a n c e o f E C t u n g s t e n ox ide fi lms o f v a r i o u s t h i cknes s in the c o l o r e d s ta te . T h e

f i lms a r e b i a s e d a t - 6 0 0 m V ve r sus S C E a n d a r e p l a c e d in a mic roce l l c o n t a i n i n g 0.1 M H 2 S O 4

e lec t ro ly te .

A. Pennisi et aL / Tungsten-molybdenum oxide electrodes 237

Energy (MeV) 1.0 1.2 1.4 1.6 1.8 2.0

3000 l t I t I i

Sn W

2500

2000

1500 0

1000

500"

J 0 I t., 300 400 500 600 700

Channe l

Fig. 5. He + ions backscattering spect rum of a W O 3 EC layer deposited on ITO substrate. The arrows indicate the energy of ions backscattered by surfacial a toms of W and Sn. The width of the two boxes are related to the thickness of the W O 3 and ITO layers. The W O 3 lay,er was found to have 2350 ,A

thickness, and the SnO 2 layer was found to have 4000 A thickness.

represents the ion energy loss in the WO 3 layer, strongly dependent on the thickness of the film. The value of the thickness was found from IBS to be about 2350 ,A for 100 deposition cycles. The width of second box (at lower energies) is proportional to the thickness of the ITO layer (about 4000 .A). Signals coming from indium and tin cannot be resolved due to the similar mass of the two elements, so that the energy of backscattered ions is nearly the same. No substantial difference can be observed in the spectra obtained from tungsten oxide samples doped with molybdenum, at least up to a Mo concentration value of 7%.

As a consequence of the voltammetry charge capacity and of the optical transmittance, we chose 100 cycles for film deposition. A light blue coloration was observed during the film growth. The electrodes were rinsed successively in distilled water and thermally treated for 1 h at 130°C.

3 . E l e c t r o c h r o m i c p r o p e r t i e s o f W - M o a l l o y s

To alter the color of electrochromic W O 3 films we doped them with molybdenum. Tungsten-molybdenum oxide films were fabricated by adding to the solution previously described for WO 3 deposition, different quantities of a 0.1 M molybdenum oxide solution obtained dissolving 1 g molybdenum metallic powder in 10 ml of 35% H20 2. The molybdenum concentration we report thereafter is coincident with the percentage in weight of molybdenum dissolved in the electrolytical solution. The parameters for the electrochemical deposition were the same as for


-2

E "6

to -6

-8

I i I 0.7 OA

Voltage (V vs SCE}

Fig. 6. Voltammetry of an EC WO 3 film doped with 0.5% Mo oxide, deposited from a peroxide solution. T h e scan rate is 100 mV/s . T h e arrows indicate the evolution in time of the deposition cycle .

the undoped tungsten oxide films. We noted that with an increased percentage ( > 10%) of molybdenum in the film, weaker adhesion to the substrate occurred, in fact an electrical or thermal shock can be responsible of film peeling.

Figs. 6 and 7 show the voltammetry of the electrochromic W-Mo oxide alloy deposition process, respectively for a 0.5% and 7% Mo oxide concentration. The arrows indicate the evolution in time of the voltammetry pattern. A decrease in exchanged charge per cycle is evident when the Mo concentration is increased.

-2

-8

I 07

I

Vol tage (V vs SCE)

I

Fig. 7. Voltammetry of an EC WO 3 film doped with 6% Mo oxide deposition processed from a peroxide solution. T h e scan rate is 100 mV/s . T h e arrows indicate the evo lut ion in time of the

deposition cycle.

A. Pennisi et a L / Tungsten-molybdenum oxide electrodes 239

'~ -0.'75

d - " 6 ~,

- 1 . 5 , I . . . . I . . . . I . . . . l , ,

-0.7 -0.6 -0.5 -0.4 P o t e n t i a l (V vs. SeE)

Fig. 8. I - V pattern of the 100th deposition cycle for samples containing different percentages of Mo oxide.

This means there is a different amount of deposited layer per cycle. In fig. 8 this effect is clearly sketched. The pattern of the last deposition cycle (the 100th) is shown for samples containing a different percentage of molybdenum.

Fig. 9 shows the charge-discharge relationship for two different EC electrodes in 0.1 M H E S O 4. The applied potential was a d c triangle potential shown in the inset. The charge and discharge branches of the solid line curve, obtained for an undoped sample, show that the current goes to the zero value within the time

. . . . I . . . . I . . . . I . . . . t . . . . I . . . .

5 a - l ~ o O ~

b - M o 2 ~

, . o 0 ~ - - - -

r . )

- 5 - - \ ~ T i m e ( s e e ) 2 4

0 25 50 T ime (see)

Fig. 9. Charge-discharge relationship for two different W-Mo oxide EC films in sulfuric acid. The applied potential was a triangle potential, shown in the inset. The scan rate is 50 mV/s and the active

area is 5 cm 2.

240 A. Pennisi et al. / Tungsten-molybdenum oxide electrodes

120

I00

80 r~ .~ 60

~ 4o

20

I . . . . f . . . . r . . . . r

c - " 3 ~

d-" 4% I e - ' " 7%

0 f . . . . I . . . . J . . . . J , 400 600 800 1000

Wavelength (rim) Fig. 10. Normal transmittance for EC tungsten oxide films containing various percentages (indicated) of Mo oxide, in the bleached state. The films are biased at 600 mV versus SCE in a cell containing 0.1 M

H 2 S O 4.

interval of the sweep. The dashed curve, obtained with a sample doped with 2% molybdenum, shows two slightly different behaviors for oxidation and reduction processes. The oxidation currents reach the zero value in a longer time and the amplitude of the oxidation peak reduces with increasing molybdenum concentration.

Transmittance data in the bleached state are shown in fig. 10 for several samples prepared with different Mo concentrations (between 0% and 7%). The transmittance in the oxidized form of the electrochromic film is not substantially altered by the Mo presence, because no significant variations of transmittance are noted up to 1100 nm. The same data for the colored state, reported in fig. 11, show a different behavior. In the visible region a larger percentage of Mo shows a decrease of transmittance. This tendency is reversed in the infrared region where the films labelled with e, f and g exhibit a lower absorbance. The dip close to 1000 nm is due to sulfuric acid absorption. A flatter curve shape is observed for the samples with greater doping. This behavior shows another significant effect, the change in film color in the reduced phase. While the undoped WO 3 switches from transparent to blue, by doping it with Mo oxide, the coloration shifts to grey (see figs. 12 and 13). Also this effect is shown in fig. 14 where the transmittance, in the .colored state, is reported as a function of Mo concentration at various wavelengths (compare the decreasing curve obtained at 400 nm, with the increasing one obtained at 1100 nm).

Photopic (Tp) and solar (T s) transmittance in the two EC states of the films were calculated and plotted in fig. 15 as a function of the dopant Mo concentration. No relevant variation in the values of Tp and T~ in the bleached state is observed when Mo doping concentration ranges from 0% up to 8%. The most significant effect is


100 I . . . . I . . . . I . . . . I '

a - }1o 0 % b - 0.5 g

8 0 - / c - 1 g p / b d - 2 ~o

e - " 4 70 c f - " 6

¢J 60 g - 7 Fo

.~ " d

~ g cn 40 f e -

[--, 20

0 I . . . . I . . . . I . . . . I 400 600 800 1000

W a v e l e n g t h ( r i m )

Fig. 11. N o r m a l t r a n s m i t t a n c e for E C tungs t en oxide f i lms con t a in ing var ious p e r c e n t a g e s ( ind ica t ed ) of

Mo oxide, in the co lo red s ta te . The f i lms are b i a sed at - 600 m V versus SCE in a cell con ta in ing 0.1 M

H2SO 4.

Fig. 12. Two E C films, in the b l e a c h e d s ta te , in H 2 S O 4 bath . The s a m p l e on the lef t ( m a r k e d wi th a dot) is an u n d o p e d W O 3 fi lm on I T O glass, the s a m p l e on the r ight ( m a r k e d wi th two dots) re fers to a

W O 3 f i lm con ta in ing 6 % M o oxide.


WO=

Fig. 13. Two E C films, in the co lored state , in H 2 S O 4 bath. The b lue s amp le m a r k e d wi th a dot refers

to u n d o p e d W O 3 fi lm on I T O glass, whi le the grey s ample m a r k e d wi th two dots, refers to a W O 3 film

con ta in ing 6% M o oxide.

100~ . . . . [ . . . . I . . . .

+ - 400 n m o - 600 "

E 80"~ o - 900 " i , x - 1 1 0 0 "

g E'~ C -

0 . . . . I . . . . I . . . . 0 2.5 5 7.5

Mo c o n c e n t r a t i o n (%)

Fig. 14. N o r m a l t ransmi t t iv i ty of W - M o oxide films, in the co lo red s ta te , versus the p e r c e n t a g e of d o p a n t M o concen t ra t ion . The var ious curves a re for d i f fe ren t wave lengths .


i 00 . . . . I . . . . I . . . . I

..-. 80' ~ - ~ - . . . ~ " ~ ~ ~-"-'"-"~ + - --''Ip~bleached)

60 ~ D- T s ( b l e a c h e d )

~0 : o - Tp(co lored)

x _ , i s ( co lo red )

0 . . . . . I . . . . I . . . . I 0 2.5 5 7.5

MO c o n c e n t r a t i o n (%)

Fig. 15. Ca lcu la ted pho top ic (To) and solar (T~) t r ansmi t t ance , for the b leached and colored condi t ion of the E C W - M o oxide films, as a func t ion of dopan t Mo concent ra t ion .

a substantial decrease (about 30%) of the photopic and solar transmittance, in the colored state, followed by an almost flat region, by the addition of a low percentage of molybdenum (less than 2%).

4. Discussion of the EC swiching rate

The EC W-Mo oxide samples were successively tested under cyclic potential conditions, in order to observe the transmittance behavior in the two states. In fig. 16 we report the transmittance versus time for samples containing different percentages of molybdenum dopant. The curve marked with a refers to the undoped sample, while the curves b, c and d refer to samples containing 0.5%, 3% and 6% of Mo respectively. The applied voltage has a trapezoidal waveform, with a period of 200 s, ranging from 600 mV to - 600 mV, as shown in lower part. This type of waveform is used to avoid an excessive current during the EC transition which otherwise tends to degrade the active layer. The time constants of the process tend to increase as a function of Mo oxide concentration and, as a consequence, it is possible to detect a less marked optical transition at a fixed biasing waveform. In general, the switching time for oxidation is longer than the one for reduction, so coloration occurs in a shorter time than bleaching.

In order to explain this behavior the switching process has to be analyzed from a microscopic point of view. In fact, while the coloring is mainly governed by the properties of the interface between the EC material and the electrolyte that injects protons, the bleaching is governed by the proton and electron injection. During this latter phase, electrons move from the EC layer to the electrolyte, while protons diffuse towards the ITO. The flow of the current is limited only by the


1 ° ° ~ . : _ ~ 1 ' ' ' ) f ' ' ' k l ' ' ' j I ' ' ' ',1 ' '

50 L ~ " ~ / ~ ./ :~, -

~ ioo ~ - - ~ ~ / ~. / Mo 0

50 -~

1 0 0 ~ ~ _

100

50!__-- b lo6% -

,---. 0 >~ o . 6 ~ .~-o6

0 100 200 300 400 500 T i m e (sec)

Fig. 16. Normal transmittance versus time for W-Mo oxide EC layers containing different percentages of Mo oxide. The molybdenum content of each samples is indicated. The applied potential has a

trapezoidal form, with a period of 200 s, as shown in the lower plot.

presence of a proton space charge layer in the region adjacent to the electrolyte interface.

The gradient of proton concentration determines the behavior in time of the current flow. It is possible to calculate the diffusion constant of protons into the EC alloy. We utilized the MacArthur model [17] that permits the determination of the diffusion coefficient for protons moving in the EC lattice, following analysis of the current transient under a potential step. The model is similar to that used for one dimensional heat transfer. The experimentally measured quantity is the diffusion current i a as a function of time, that can be expressed in an exponential form as:

neFAD 2C° n o ( D'rr2(2n + l)2t i a = ~ ] e x p - , (1)

where n is an integer, n e is the number of positive charges on the diffusion species, F is the Faraday constant, A the area of the electrode, D the diffusion coefficient, t the time from the start of the diffusion process, C O the concentration of tungsten sites in the WO 3 lattice and h the diffusion length. The values of D and h may be found by noting that for sufficiently long times only the term with n = 0 in the summation will be significant. By extrapolating to zero time, the current intercept in a semilog plot is proportional to the ratio D/h , while the ratio between the intercept and the slope of a straight line fitting the ia(t) curve, is proportional to h.


Following this method we plotted on a semilog frame the oxidation current flowing in a three-electrode cell versus time, under a step potential going abruptly from -600 to 200 inV. This abrupt potential step leads to a rapid peeling of the EC layer and the film comes off from the ITO substrate after only a few steps, because of the large currents in the cell. Analyzing this curve by the method previously described, we obtained for the proton diffusion coefficients of the oxidized EC phases values ranging from 8 × 10 -7 cm2/s for the undoped sample to 1 × 10 -7 cm2/s for a sample doped with 6% of Mo oxide. The values of Dox (proton diffusion coefficient during the oxidizing reaction) decrease with the increasing of Mo percentage in the sample.

During the coloration process, the ITO layer and the electrolyte inject respectively electrons and protons into the WO 3 layer. Because WO 3 and counterelectrode are different materials, there will be an EMF between them proportional to their chemical potential difference. As the film colors, its chemical potential changes, to make the film more positive with respect to the counterelectrode. This EMF is directed opposite to the applied voltage and the current is limited by many different mechanisms, among which the presence of a barrier at the contacts injecting respectively electrons and protons is very important.

For times large compared to the characteristic time for proton diffusion, the expression giving the current exhibits the commonly observed dependence on the inverse square root of time, which is typical in the presence of a barrier. However the diffusion of protons in the film can significantly alter the coloring speed only at times short if compared with the diffusion time. In this case the diffusion

t . . . . . . . . . . . .

~"" . . . . . reduction

- 3 o ;

oxidation _ 6 0 [E: . . . . I . . . . I . . . . I . . . . I . . . .

0 4 8 12 16 2 0

Time ( s )

Fig. 17. Reduction and oxidation current densities (dotted curves) for a WO 3 sample doped with 1% Mo oxide. The applied potential is a step going from 200 mV to -600 mV for the reduction (coloration) and from -600 mV to 200 mV for the oxidation (bleaching). The solid curves show respectively the behavior of 1 / J 2 and log(J) versus time. These curves are utilized to calculate the

diffusion coefficient of the protons into the EC layer.

246 A. Pennisi et al. / Tungsten-molybdenum o.ride electrodes

60

30 g-. E Q

< 0

v

L . . . . I . . . . I . . . . I . . . . I ' ' ' 5

. . . . . . -

-30 ~

-6Q , I , , o 4

oxidation ~-

8 12 16 Time (s)

G" v

Q ,-.a

Fig. 18. The same as in fig. 17 for a WO3 sample doped with 5% Mo oxide.

coefficient can be estimated by diffusion current i a through the expression [18]

i d = e p x s A ~ / D - / a r t , (2)

where p is the tungsten atom density, e the electron charge and x s is a constant of the order of 0.5.

We utilized expression (2) to estimate the diffusion coefficient for the reduction phase, analyzing the current transient induced by the application of a potential step going from -600 mV to 200 mV. The obtained values of D are comparable with those calculated for the oxidation phase. All values are in agreement with ones reported in the literature [18,19].

In figs. 17 and 18, reduction and oxidation current densities J (dotted curves) are reported for two samples doped respectively with 1% and 5% Mo oxide. The linear dependence of 1 / J z with time gives evidence that the behavior of the reduction current is well described by eq. (2). Similarly the linear dependence at large times of log(J) with time gives evidence that the diffusive model described by eq. (1) is working well for the oxidation state.

5. Conclusions

In summary, the transition-metal oxide films we prepared by electrochemical deposition exhibit good electrochromic properties. Adding of molybdenum, at least to a percentage of 2%, gives better film dynamical optical responce, because it increases the switching range between the transmittivity values. In fact, while the photopic transmittance of the bleached state remains unaltered, its value changes from Tp = 0.65 to Tp = 0.38, in the colored phase, for two samples containing 0% and 2% molybdenum respectively.


A larger dopant concentration leads to a longer transition time for both processes. However the reduction process is always faster if compared with the oxidation one. Correspondingly there is evidence of a longer "memory" of each phase and of a decrease in the infrared absorbance.

The possibility to alter gradually the color of the EC film by controlling the molybdenum content, gives new perspectives in the development of EC devices. A clear evidence is given in the paper of a strongly different behavior of normal transmittance in the colored state for differently doped films. In fact, while the undoped samples exhibit only a large peak at 400 nm, responsible of the blue coloration, the increase of doping produces a flatter spectrum at visible and infrared region, which gives the grey coloration.

Acknowledgements

The authors are grateful to Professor E. Rimini and Dr. M.G. Grimaldi who gave their kind assistance for the backscattering measurements and to Mr. V. Connelli for technical assistance.

References

[1] C.M. Lampert, Sol. Energy Mater. 11 (1984) 1. [2] C.M. Lampert, ISES Solar World Congress, Kobe, Japan, Sep. 1989. [3] S.E. Selkowitz and C.M. Lampert, in: Large Area Chromogenics: Materials and Devices for

Transmittance Control, Eds. C.M. Lampert and C.G. Granqvist (Optical Engineering Press, Bellingham, WA, 1989).

[4] R.B. Goldner, K. Wong, G. Foley, P. Norton, L. Wamboldt, G. Seward, T. Haas and R. Chapman, Proc. SPIE 692 (1986) 39.

[5] K. Bange, U. Martens, A. Nemetz and A. Temmink, in: Electrochromic Materials, Eds. M.K. Carpenter and D.A. Corrigan (The Electrochemical Society, 1990) p. 334.

[6] O. Bohnke, C. Bohnke and G. Robert, Solid State Ionics 6 (1982) 121. [7] S.K. Deb, Proc. SPIE 692 (1986) 19. [8] G. Seward, R.B. Goldner, K. Wong, T. Haas, G. Foley, R. Chapman and S. Schuiz, Proc. SPIE 823

(1987). [9] R.B. Goldner, G. Seward, K. Wong, G. Berera, T. Hass and P. Norton, Proc. SPIE 823 (1987).

[10] UH. Dao, A. Guerfi and M.T. Nguyen, in: Electrochromic Materials, Eds. M.K. Carpenter and D.A. Corrigan (The Electrochemical Society, 1990) p. 30.

[11] L.H. Dao, M.T. Nguyen, in: Electrochromic Materials, Eds. M.K. Carpenter and D.A. Corrigan (The Electrochemical Society, 1990) p. 246.

[12] C.M. Lampert, Y-P. Ma, A. Pennisi and F. Simone, Lawrence Berkeley Laboratory, report September 1990.

[13] K. Yamanaka, Jpn. J. Appl. Phys. 26 (1987) 1884. [14] M.T. Nguyen and UH. Dao, J. Electrochem. Soc. 136 (1989) 2131. [15] A. Guerfi and L.H. Dao, J. Electrochem. Soc. 136 (1989) 2435. [16] G.J.W. Mayer and E. Rimini, in: Ion Beam Handbook for Material Analysis (Academic Press, New

York, 1977). [17] D.M. MacArthur, J. Electrochem. Soc. 117 (1970) 729. [18] B.W. Faughnan and R.S. Crandail, in: Display devices, Ed. J.l. Pankove (Springer, Berlin, 1980). [19] B.W. Faughnan, R.S. Crandall and M.H. Lampert, Appl. Phys. Lett. 27 (1975) 275.

electrochromic properties of tungsten- molybdenum …gaia.lbl.gov/btech/papers/39078.pdf · 234 a....

Documents