Electrochromic devices: A comparison of several systems

Solar Energy Materials and Solar Cells 39 (1995) 213-222

ELSEVIER

Electrochromic devices: A comparison of several systems. C. Arbizzani, M. Mastragostino and A. Zanelli Unitersity of Bologna, Department of Chemistry "G. Ciamician ", via F. Selmi 2, 40126 Bologna, Italy

Abstract The great variety of colour contrast achieved with electronically conducting polymers !s the most significant characteristic of this class of ion-insertion organic materials. Tile performance of selected electrochromic devices based on electronically conducting pOlymers, with organic or inorganic materials as optically passive or electrochromic in compl~mentary mode counter-electrodes, is reported and the advantages and drawbacks of each system discussed.

1. Introduction

The electrochromism of electronically conducting polymers (ECPs) is a w~ll known phenomenon and the great variety of colour achieved with ECPs is the most significant characteristic of this class of ion-insertion organic materials with respect to that of ion-insertion inorganic ones [1-4]. The doping process of ECPs modifies the electronic band structure of t~e polymer by producing in the gap new electronic states that cause the colc~ur changes; doping shifts the absorption towards the low energies and the col~iur contrast is related to the magnitude of the energy gap (Eg). ECPs with Eg grealer than 3 eV are transparent or slightly eoloured in the undoped form and gener,"lly highly absorbent in the visible region in the doped form. By contrast, ECPs with small Eg (1.5-2 eV) are highly absorbent in the undoped form, the colour be:ng related to the wavelength of the maximum absorption, but after doping they ire almost transparent in the visible region. Polymers with tailored Eg (i.e. tailo ed electroehromic properties) can be synthesized by using suitably designed start ng molecules. There are various strategies for tuning the eolour contrast of thiophene-ba~ed polymers. We synthesized poly(alkylthiophenes) with very different electroehro~nic properties starting from isomeric oligomers of different length, whether symme~c (equivalent a-o~' positions) or unsymmetric (non equivalent a - a positions) 0167-8191/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved ¢~Dt 0027-0248(95)00057-7

214

C. Arbizzani et al. / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

molecules, with alkyl groups of different length [5-8]. The conformational differences deriving from the steric hindrance of the alkyl groups, which are arranged so as to originate different patterns in the resulting chains, play the main role in tailoring the effective conjugation length of the polymers, to which Eg is related. We also tested a method to modulate the conjugation length via copolymerization of 3-alkyl and 3,4-dialkylthiophenes and synthesized polymers with very different optical properties [9], although these composites present some drawbacks as electrochromic materials. Polymers with an Eg lower than that of the poly(alkylthiophenes) can be synthesized by starting from alkoxy-disubstituted thiophenes because of the two electron-donating oxygen atoms adjacent to the thiophene unit. Poly(3,4-alkylenedioxythiophene) has been recently synthesized [10-12] and J.C. Gustafsson et al. [13] have developed a successful electrochromic device by using this polymer, an ion-insertion inorganic material as optically passive counter-electrode and a polymer electrolyte. The polymer is dark purple-blue in the undoped form and upon doping is highly transparent. Reducing the polymer's Eg involves increasing the quinoid character of the 7r-conjugated system at the expense of its aromatic character: poly(isothianaphtene) can be considered the forerunner of the family based on condensed rings. The starting monomer is a thiophene unit with a benzene ring built on in 3,4-positions, and the polymer is dark in the undoped and transparent in the oxidised form. Given its low Eg, it can be p- and n-doped, and an electrochromic device with both poly(isothianaphtene) electrodes has been recently proposed by M. Onoda et al. [14]. Poly(dithieno[3,4-b:3',4'-d]thiophene) (pDTr) [15,16], even if it is synthesized from a monomer marked by a high degree of aromaticity, has a small Eg and its contrast colour is like that of poly(isothianaphtene). In this polymer, too, charge injection of both signs is possible, although performance tests indicate that it is not as promising an electrochromic material as it is an electrochemical one [17,18]. A very recent strategy for band gap reduction involves the synthesis of bridged polymers. J.P. Ferraris et al. [19,20] introduce an electron-withdrawing group at sp 2 carbon bridging the 3,3' positions of a 2,2'-bithiophene (the guideline is the non-aromaticity). H. Brisset et al. [21] graft a dioxolane moiety at sp 3 carbon bridging the 3,3' positions of a 2,2'-bithiophene and strongly suggest that molecule stiffening plays a determining role in the decrease of Eg. The present paper outlines the main strategies designed to synthesize polymers with tailored electrochromic properties and reports the comparative performance of selected electrochromic devices based on ECPs with counter-electrodes that are optically passive or electrochromic in complementary mode, organic or inorganic materials. The advantages and disadvantages of each system are also discussed. 2. Experimental The polymers were electrosynthesized on indium tin oxide (ITO, AET Zirst, France) or tin oxide (TO, Nippon Sheets, Japan) transparent conductive glasses.

C. Arbizzani et aL / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

215

Table 1 Wavelengths of the maximum absorption (A,,_ ,r*) and colours of the doped and undoped forms of poly(alkylthiophenes) from different starting molecules, R = methyl Starting molecule

/R

~

~

s

~R

~ R

R

R = hexyl

A,,_ ,7.

Colour

(nm)

Undoped

Doped (nm)

A,,_ ,,.

Undoped

Doped

530

purple

paleblue

520

red

paleblue

505

red

blue

455

orange red

paleblue

500 red blue (pristine 400-500 nm)

Colour

470 red pale(pristine 390-470 nm) blue

415

yellow

blue- 390 violet

palegreen

paleblue

445

orangered

blue

palegreen

paleblue

R

390

Each electrosynthesis condition as well the syntheses of the starting molecules and the preparation of the polymer electrolytes used in the laminated solid-state devices are reported in the references. The spectroelectrochemical characterization of the poly(alkylthiophene) elec~ trodes were performed in degassed propylene carbonate (PC) -1M LiCIO4; all the 60-

2O

o

- -

lOth

....

lO,O00th

~

switch

I

Switch

g

time (s) Fig. 1. Optical response to 10th and 10000th potential steps (8 s period) of poly(3,3'-dimethyl-2,2'-bithiophene) between 0 and 1 V ts SeE, recorded at 415 nm.

216

C. Arbizzani et al. / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

measurements were carried out with a standard electrochemical equipment, a 3091 Nicolet oscilloscope and a Lambda 19 Perkin Elmer spectrophotometer. All the chemicals were reagent-grade products purified before use. The potentials in 3-electrode mode experiments were related to the satured calomel electrode (SCE); the potential in the 2-electrode mode experiments performed on the devices is that between the working- and the counter-electrode. 3. Results and discussion

3.1. Poly(alkylthiophenes) It is possible to attain a very fine gradation of colour contrast with poly(alkylthiophenes) [7]. The wavelength of the maximum absorption of the undoped form varies in the range of 150 nm with colour changes from transparent pale yellow for poly(3,3'-methyl-2,2'-bithiophene) or transparent pale green for poly(3,3'-hexyl-2,2'-bithiophene) to dark purple of poly(3-methylthiophene) from the monomer. Table 1 shows the wavelengths of the 7r - 7r* transition (A~_ ~.) and the colours of the polymers undoped and doped forms. Differences in the optical properties of 3-methlythiophene-based polymers make it possible to design a variable light transmission electroehromic device with two regiochemically and conformationally different polymers, poly(3-methylthiophene) and poly(3,3'-methyl-2,2'-bithiophene), operating in complementary mode in the visible region [22]. We investigated how differences of regiochemistry and conformation between poly(3-methylthiophene) and poly(3,3'-methyl-2,2'-bithiophene), which lead to polymers with very different intrinsic optical properties, affect the electrochromic

Table 2 Comparison between the electrochromic performance of poly(3,3'-dimethyl-2,2'-bithiophene) and poly(3-methylthiophene)

Undoped a,,_ 1,, (nm) switching voltage (V vs SCE) T% at A,~_,~,

415

optical memory over 8 h

Undoped

1

12

57

- 0.5

]

6

54

4 11 1.0

Doped

530

0

switching time (s) T% at A~. ~, after 1.104 switches

Doped

4 55 0.86

9 1.0

54 0.93

C. Arbizzani et al. / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

217

performance of these polymers. Table 2 compares the h~._,r' of the undoped form, the switching voltage values between the doped and undoped states and the related transmittance values (T%) at h~_,~., switching time, cycle-life (evaluated by comparing the T% at h,~_~. recorded before and after 104 switches) and optical memory (expressed as the ratio and the inverse ratio, respectively, between the transmitted light at h~._,r * 8 hours after the removal of the driving potential in the doped and undoped forms, respectively, and that immediately after the removal) of poly(3,3'-methyl-2,2'-bithiophene) and of poly(3-methylthiophene), the latter being recognized as a very promising electrochromic material [23]. Despite a very tilted structure due to the steric hindrance of the head-to-head linked methyl groups, poly(3,3'-methyl-2,2'-bithiophene), meets the most important requisites for use in electrochromic devices; Figure 1 shows the transmittance changes during the 10th and the 104th potential step between 0 and 1 V us SCE. 3.Z Solid-state electrochromic detices.

The electrochromic performance of three solid-state electrochromic devices, developed in the last few years at our laboratory, based on ECPs with counter-electrodes that are organic or inorganic materials optically passive or electrochromic in complementary mode, were compared. The first system consisted of a poly(3-methylthiophene) working-electrode, an optically passive ITO counter-electrode (at which occurs an electrochemical process without significant colour changes that is presumably due to lithium ion insertion up to 7.5 mC cm-2), and a polymer electrolyte. The latter a polymer network including PC as plasticizer, was prepared via a cross-linking reaction of the terminal amino groups of a copolymer of polyethers with a di-epoxy compound [22]. By switching the applied voltage between -0.3 and 2.5 V, the total electrochemical process is .-./ ~,,o~ )n + ITO + ny LiCIO4 [.( purple

•

)+Y(CIO4")y]n+ LinylTO

transparent pale blue

and the colour change from purple to transparent pale blue is related to the electrochromism of poly(3-methylthiophene). The second device is based on poly(pyrroledodecylsulfate), a WO 3 counter-dec, trode electrochromic in complementary mode, and a polymer electrolyte based on poly(ethylene oxide-co-epychloridrina) [24]. By switching the applied voltage between -0.8 and 0.4 V, the overall electrochemical process is ~ ' ~ n + WO3 + ny LiCIO4< . . . . . > ~

~

)+Y(CIO4-)y]n + LinyWO3

transparent pale yellow dark blue and the colour change is from transparent pale yellow to dark blue.

218

C. Arbizzani et aL / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

The third system was assembled by facing off a doped poly(3-methylthiophene) and an undoped poly(3,3'-methyl-2,2'-bithiophene) counter-electrode. For the solid state configuration we used a polyacrylonitrile-based gel electrolyte [25]. By switching the applied voltage between 0.9 and - 0 . 9 V, the resulting electrochemical process is

transparent pale green

dark violet

and the colour switches between transparent pale green and dark violet. 100-

ITO

a)

80-

:

60-

i I,'"

40-

g

20= 0 200

80-

400

6bo

8~0

' 1000

12'O0

' 1400

1600 -1

400

61~0

8(~0

r 1000

12'O0

1400

1600

400

800

6oo

~ooo

12oo

14oo

16oo

b)

60-

i

40-

I

200 200

"]-~ 401 c)

200

(nm) Fig. 2. Optical spectra of an ITO electrode (dotted line) and of the three selected devices in the transparent (dashed line) and dark (solid line) states. (a) poly(3-methylthiophene)//ITO device at 2.5 V (----) and - 0.3 V (--);(b) poly(pyrroledodecylsulfate)//WO 3 device at - 0.8 V (----) and 0.4 V ( - - ) ; (c) poly(3-methylthiophene)//poly(3,3'-dimethyl-2,2'-bithiophene) at 0.9 V (----) and - 0 . 9 V ( - - ) .

C. Arbizzani et aL / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

219

40-

T~ 302010-

o) O-

lb

6

20

~o

~'o 1st

40-

T~ 302010-

b) 0

~ ' 1'o

1'5

2'0

40-

T~s 302010-

c) O-

time (s) Fig. 3. Optical response to potential step of: (a) poly(3-methylthiophene)//ITO device between - 0 . 3 and 2.5 V recoreded at 500 nm; (b) pol~(pyrroledodecylsulfate)//WO 3 device between - 0 . 8 and 0.4 V recorded at 700 nm; (c) poly(3-methylthiophene)//poly(3,3'-dimethyl-2,2'-bithiophene) device between 0.9 and - 0 . 9 V recorded at 560 nm.

Figure 2 shows the optical spectra in the transparent and dark states of th e three devices (the absorption of the electrolytes and of the ITO glasses supporting electroehromic materials have been subtracted). The extent of light modulation in the entire visible region of these devices is good and the electrochromic efficiency (e.e.) at ;t,~_~, is 0.16, 0.08 and 0.25 mC-1 cm 2, respectively, and at 560 nm (i.e. the wavelength of maximum sensitivity of the human eye) the e.e. is 0.15, 0.10 an~i 0.25 mC -~ cm 2, respectively. High e.e. is an advantage for large area device, especially for energy saving windows. Figure 3 shows the electroehromie response time to repeated potential switches of the three devices. The response time of device 3, in which both electrodes a~e ECPs, is shorter than in the devices with inorganic ion-insertion counter-electrode materials, as the doping-undoping process of conjugated polymers is generally fa~t.

220

C. Arbizzani et aL / Solar Energy Materials and Solar Cells 39 (1995) 213- 222 2.0

,8 1.5

1.0 '/

•

/I

0.5

0.0

,

"

560

1 O0

,.

I ~t,

960

1 ~'00

~7'00

2100

2500

2100

2500

2.0

b)

s

E

1,5

E

?

/

,.

1.0

E 0.5

!

,'. -. ,....,

./ ~

,

:

.

¢

~ ~ - C,, .............,

0.0 100

5(~0

960

1300 '

1700 '

2.0 c)

1.0

0.5

o.o1O0 .

I

~ ,,,

!

!i'-,,,

"

,, -

/

560

, .,-..,

v

~

~

g00

' 1300-

1700

.............. 21100 25b0

~k ( n m )

Fig. 4. Solar irradiance (dotted line) and transmitted solar energy flux by each device in the bleached (dashed line) and coloured (solid line) states of (a) poly(3-methyltiophene)//ITO, (b) poly(pyrroledodecylsulfate)//WO 3 and (c) poly(3-methylthiophene) / / poly(3,3'-dimethyl-2,2'-bithiophene) devices.

The stability to repeated switches of these devices is good, as is the optical memory. Reactions to impurities, for example to oxygen, can lead to an imbalance of the electrochromic charge at one or both electrodes and, hence, to an alteration of the device's performance. Unlike the first two devices, the ECP-based one has the advantage of being stable with respect to oxygen. Smart windows for energy saving in buildings is the most important application of electrochromic devices and Figure 4 shows how ECP-based devices regulate the solar energy flux. The dotted line is the solar irradiance [26,27] and the dashed and solid lines are the solar energy flux transmitted by each device in the bleached and coloured state, respectively. Here, too, the input of the ITO and of the electrolyte,

C. Arbizzani et al. / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

221

which both reduce transmittance regardless of the applied potential, has b e e n subtracted. Figure 5 shows the partitioning of the transmitted energy flux in t h e three spectral regions. The first column of the histogram gives the percentage o f the solar energy in the three regions. The others are the percentages of t h e transmitted energy in the ultra-violet, visible and near-infrared regions by the t h r e e devices in the purple/pale-blue, pale-yellow/dark-blue, and pale-green/dark-violet colouring, respectively. The first device seems a poor candidate for the development of smart windows because it does not regulate visible and infrared radiation simultaneously. By contrast, the second device can modulate at the same time t h e visible and infrared radiations. The third device regulates only the visible light, the infrared remaining almost constant because one polymer electrode is always in the doped state (it could be of some interest in tropical climates).

4. Conclusions

ECPs are very attractive electrochromic materials for the great variety of colour contrast they produce. Yet, for smart window technology, the most promising system may be that with an electrochromic ECP and an ion-insertion inorganic electrochromic material operating in complementary mode.

100-

d

c,i

"100 -> ~ d o I

m

d

ol

~

700-1800

50

nm

"100

E c o

400--?00 nm

"1000 320--400 nm

Fig. 5. Percentage of the solar energy (first column) and percentages of the energy transmitted by (*) poly(3-methyltiophene)//ITO, (b) poly(pyrroledodecylsulfate)//WO3 and (c) poly(3-methyithiophen~) / / poly(3,3'-dimethyl-2,2'-bithiophene) devices in the ultra-violet, visible and near-infrared spectral regions.

222

C. Arbizzani et al. / Solar Energy Materials and Solar Cells 39 (1995) 213- 222

Acknowledgements W e w o u l d like to t h a n k Dr. G. Barbarella, I . C o . C E A . , C N R , O z z a n o (Bologna), Italy, who p r e p a r e d the a l k y l t h i o p h e n e starting materials, a n d Dr. A.M. Rocco B e r h e n s , U n i v e r s i t y of C a m p i n a s , Brasil, for h e r c o n t r i b u t i o n to this study. The r e s e a r c h was f u n d e d by a g r a n t of C N R , P r o g e t t o Finalizzato Materiali Speciali per Tecnologie A vanzate,

References [1] M.A. Habib, in: O.J. Murphy et al. (Eds.), Electrochemistry in Transition, Plenum Press, New York (1992) 51. [2] M. Mastragostino, in: B. Scrosati (Ed.), Applications of Electroactive Polymers, Chapman & Hall, London (1993) 223. [3] K. Hyodo, Electrochim. Acta 39 (1994) 265. [4] C.G. Granqvist (Ed.), Materials Science for Solar Energy, Pergamon Press, Oxford (1991). [5] M. Mastragostino, C. Arbizzani, A. Bongini, G. Barbarella and M. Zambianchi, Electrochim. Acta 38 (1993) 135. [6] C. Arbizzani, G. Barbarella, A. Bongini, M. Mastragostino and M. Zambianchi, Synth. Metals 52 (1992) 329. [7] C. Arbizzani, A. Bongini, M. Mastragostino, A. Zanelli, G. Barbarella and M. Zambianchi, Adv. Mat., 7 (1995) 571. [8] C. Arbizzani, M. Mastragostino and A. Zanelli, to be published. [9] M. Catellani, C. Arbizzani, M. Mastragostino and A. Zanelli, Synth. Metals, 69 (1995) 373. [10] M. Dietrich, J. Heinze, G. Heywang and F. Jonas, J. Electroanal. Chem. 369 (1994) 87. [ll] Q. Pei, G. Zuccarello, M. Ahlskog and O. Inganas, Polymer 35 (1994) 1347. [12] G. Heywang and F. Jonas, Adv. Mat. 4 (1992) 116. [13] J.C. Gustafsson, B. Liedberg and O. lnganas, Solid St. lonics 69 (1994) 145. [14] M. Onoda, H. Nakayama, S. Morita and K. Yoshino, J. Electrochem. Soc. 141 (1994) 338. [15] A. Bolognesi, M. Catellani, S. Destri, R. Zamboni and C. Taliani, J. Chem. Soc., Chem. Comm. (1988) 246. [16] C. Quattrocchi, R. Lazzaroni, J.L. Br6das, R. Zamboni and C. Taliani, Synth. Metals 55-57 (1993) 4399; Macromolecules 26 (1993) 1260. [17] C. Arbizzani, M. Mastragostino and L. Meneghello, Electrochim. Acta, 40 (1995). [18] C. Arbizzani, M. Catellani, M. Mastragostino and C. Mingazzini, Electrochim. Acta, 40 (1995). [19] J.P. Ferraris and T.L. Lambert, J. Chem. Soc., Chem. Comm. (1991) 1268. [20] T.L. Lambert and J.P. Ferraris, ibid (1991) 752. [21] H. Brisset, C. Thobie-Gauthier, A. Gorgues, M. Jubault and J. Roncali, ibid (1994) 1305. [22] C. Arbizzani, M, Mastragostino, L. Meneghello, X. Andrieu and T. Vic6do, Mat. Res. Soc. Symp. Proc. 293 (1993) 169. [23] A. Corradini, A. Marinangeli and M. Mastragostino, Electrochim. Acta 35 (1990) 1757. [24] M-A. De Paoli, M. Mastragostino, A.M. Rocco, to be published. [25] C. Arbizzani, M. Mastragostino, L. Meneghello, M. Morselli and A. Zanelli, Appl. Electrochem. 25 (1995). [26] B.P. Jelle, G. Hagen, J. Electrochem. Soc. 140 (1993) 3560. [27] Solar-Hydrogen Energy System, T. Ohta, (Ed.) Pergamon Press, Oxford (1979).