Optical properties of polyaniline

W. S. Huang*
IBM East Fishkill, Hopewell Junction, NY 12533, USA

and A. G. MacDiarmid
University of Pennsylvania, Philadelphia, PA 19104, USA
(Received 20 April 1992; revised 20 August 1992)

Optical absorption has been used to study polyaniline at various levels of oxidation and protonation. The
conversion of leucoemeraldine base to its salt accounts for the blue shift of the n-n* absorption from 3.94
to 4.17 eV. The emeraldine base has an absorption peak at 2.1 eV due to the localized benzenoid HOMO
(highest occupied molecular orbital) to quinoid LUMO (lowest unoccupied molecular orbital) excitation.
Protonation of the emeraldine base causes a lattice distortion of polyaniline to form a polaronic lattice
structure, which accounts for the shift of the 2.1 eV absorption to 1.5 eV. Further protonation of emeraldine
salt causes the breakdown of the polaronic lattice to a confined bipolaron lattice. When the acid strength
of the solution is very high (96% H2SO4), most of the amine nitrogens are protonated. Characteristic
absorptions of isolated quinoid imine units ( ~ 2.5 eV) and benzene units ( ~ 4.8 eV) have been found. The
evolution of band structure corresponding to the optical-absorption change during the protonation of
polyaniline at each specific oxidation state is studied by gradually changing the acid strength of the solution
to which the polymer is subjected. We have also studied the optical-absorption change at one fixed acid
strength by slowly oxidizing leucoemeraldine to emeraldine then to pernigraniline by electrochemical and
chemical means. Detailed information has been obtained on the band structure changes in relation to lattice
structure interconversions during the transitions between various oxidation and protonation states.

(Keywords: polyaniline; conducting polymer; optical; emeraldiae; oxidation; protonation)

INTRODUCTION data obtained from electrochemical studies and to be

The structure ofpolyaniline has only recently been clearly consistent with the structure proposed previously.
elucidatedl_4.Thenon.conductingformofthispolymer The effects of changing the oxidation level and the
protonation level on the electronic structure of
y ~ polyaniline have been among the most important issues ;
r~N~N~_~/-~N=/'~N~__ ~ the resulting spectral changes have been extensively
L k ~',~.~' %__.4/ 1,, \%_4' ~ ),_,.J , investigated in recent years, and a number of papers have
been published 1'3'6-16. Our early studies on this subject
can, in principle, exist in a continuum of oxidation states appear to be unknown to workers in the field~ 7. For
ranging from completely reduced leucoemeraldine example, we showed that polyaniline could be dissolved
(y = 1) to completely oxidized pernigraniline (y = 0). in concentrated sulfuric acid, and the polymer did not
The conductivity of this material is a function of both show any degradation, as evidenced by its spectrum, if
the degree of oxidation and the degree of protonation 2'5. the dissolution process was maintained in the cold
The emeraldine (y = 0.5) oxidation state of polyaniline condition ~a. A similar approach was later adopted by
exhibits the highest level of conductivity upon proton- other workers. In this paper, we will provide a broad
ation, reaching a value of 5 S c m - 1. Stafstrom et al. 6 picture of the electronic structure of polyaniline in various
proposed a polaronic lattice structure for this highly conditions. We believe that this paper contains a lot of
conducting state, which agrees with theoretical band- new information that has not been seen in the open
structure calculations, optical properties and Pauli literature. Although similar work has been done before,
susceptibility, the unique nature of the work presented here can help
The structural change of polyaniline in corresponding to increase the understanding of this subject.
to the change of pH and redox potential has been
illustrated in the literature before 2-4 using electro- EXPERIMENTAL
chemical methods. In this paper, we present the optical
properties of polyaniline at different oxidation and n-re* transition inreducedpolyaniline(leucoemeraldine)
protonation states, in order to gain more understanding Spectra of leucoemeraldine at various degrees of
of the chemical reactions that occur and the materials protonation. A green, free-standing film of emeraldine
that are formed during the variation of redox potential hydroperchlorate 19
and pH. The results appear to agree very well with the H H H -- H
I I I I

* To whom correspondence should be addressed cto,~ clo~

0032-3861/93/091833-13
© 1993 Butterworth-HeinemannLtd. POLYMER, 1993, Volume 34, Number 9 1833
Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

(10 mm x 30 mm x ~0.01 mm) was placed on a quartz measure, the cell was re-treated with sodium dithionite
plate ( S m m x 5 0 m m x 1.5mm) so that it covered solution and then washed with HCl solution of pH 2.0, as
approximately one-half of the slide. The film was then described above. The cell was finally filled with the
wetted with 1 N aqueous HC1 so that it adhered to the solution and a spectrum was recorded. This procedure
quartz surface. The film was then permitted to partly dry was repeated with solutions of pH 1.0, - 0.2 ( 1 N HC1),
in the air for ~ 1 h. A very thin layer of 'five minute' - 1 . 0 5 (3 N HCI) and - 2 . 1 2 (6 N HC1). (The ' p H
epoxy cement was placed around the edge of the film so values' given for solutions more acidic than pH 1.0
that the edges were firmly attached to the quartz. The are actually Hammett acidity functions (H o) which reflect
main body of the film adhered firmly and smoothly to the proton donating ability of highly acidic solutions
the quartz when it was wet. The assembly was then placed more accurately than pH 2°'21. In dilute solutions the
in a quartz cell (4 mm x 10 mm x 60 mm). Hammett function and pH are identical.)
A sodium dithionite ( N a 2 S 2 0 4 ) s o l u t i o n (pH g 6 ) , A series of spectra using a different piece of
prepared by dissolving ~ 1 g of NazS204 in ~ 20 ml of free-standing film were recorded using aqueous solutions
distilled water, was then added to the cell so that it of H2SO4 at pH2.0, 1.0, 0.13 ( 1 . 0 N H 2 S O 4 ) , - 1 . 1 2
completely covered the top of the film. Within a few (5.0 N H2SO4) and - 2 . 0 6 (9.0 N HzSO4). The second
seconds the film changed from dark transparent green spectrum recorded at each concentration was identical
to a very pale transparent yellow, indicating that it had to the first spectrum recorded at that concentration. The
been reduced to the leucoemeraldine oxidation state: spectrum of the leucoemeraldine base was recorded at
pH 13.0 (0.1 N N a O H ) using the same piece of film used
H H H __ H for the studies in H2SO 4 and employing the same
(O ( 3 - O ' ( 3 )
• . ÷ No~S~O4*4H2O procedure described above.

c,o; cto;~ Spectrum of leucoemeraldine during air oxidation. A

leucoemeraldine film in 1.0 x 10 -2 N H2SO 4 on a quartz
H H H H plate (synthesized using sodium dithionite as described
~ ' - 4 / - - ~ - 4 f - - ' ~ -N~ - 4 f - ' ~N
' f ' ~ - ' _ ' ~NN , NozS04*H~S0,*~HCtO, above) in the spectroscopic cell was exposed to
x laboratory air for ~ 1 0 h by removing the Parafilm
protective wrapping from the top of the cell. The
From our earlier studies we have shown that the leucoemeraldine film became light green during this time,
emeraldine hydrochloride is completely deprotonated at indicating that some oxidation had occurred. The
pH 6-7, hence it is assumed that the leucoemeraldine spectrum was then measured using the procedure
base is produced by this treatment. The sodium dithionite described above.
solution was removed by tilting the cell and placing a
piece of absorbent tissue paper at the entrance to the Change in spectrum of leucoemeraldine during electro-
cell. The tissue immediately absorbed the solution and chemical oxidation
emptied the cell. Electrochemical potential versus electric charge study.
A solution of HC1 with pH 3.0 was then placed in the Emeraldine hydrotetrafluoroborate film was electro-
cell and was removed using tissue paper as described chemically deposited on a piece of Pt foil (1 cm 2) using
above. This washing procedure was repeated twice more a constant potential method 22 at 0.75 V for ~ 3 h. After
and the HCI solution with pH 3.0 was then placed in the rinsing the film with 0.1 M Pb(BF4) 2 + 0.5 M H B F 4
cell for optical studies. The top of the cell was covered solution for ~ 1-2 min the assembly was transferred
tightly with several layers of semiadhesive 'Parafilm' in to a small vial containing 20 ml of 0.1 M Pb(BF4)2 +
order to exclude air and prevent oxidation of the leuco- 0.5 M HBF4 solution. Two pieces of Pb ribbon (6 mm
emeraldine to the emeraldine. If the total time taken for x 70 mm) were placed in the solution as counter and
the removal of the sodium dithionite and for the washing reference electrodes. The Pb ribbon surface was scraped
etc. did not exceed ~ 1 rain, air oxidation of the with a knife before being placed in the solution.
leucoemeraldine film was negligible, as could be seen by The emeraldine hydrotetrafluoroborate film was
the absence of any green tint caused by the formation of reduced at 0.2V (versus Pb) for ~ 2 - 3 h to the
the emeraldine salt. corresponding leucoemeraldine. A constant current of
Spectra were obtained at room temperature using a 10 #A cm -2 was then applied between the working
Beckman DU-8 u.v.-visible spectrophotometer. In all electrode (Pt foil containing polyaniline film on the
studies, a spectrum of the aqueous solution was first surface ) and the counter electrode. The distance between
obtained using the same cell containing a piece of blank the two electrodes was ~ 1 cm. The potential was
quartz slide in order to confirm that there was no recorded versus the Pb reference electrode so as to
absorption due to the cell or its contents. Spectra were controllably oxidize the slightly protonated leuco-
obtained at room temperature by first storing in the emeraldine base to a maximum potential of 1.0 V (versus
computer of the spectrophotometer the spectrum of air Pb), i.e. to the approximate oxidation state of nigraniline.
(no cell present) followed by the recording of the This oxidation process took ~ 1 7 h . The potential was
spectrum during ~ 15 min. continuously recorded and is given in Figure 1.
The spectrum recorded on the chart paper had the air
blank automatically subtracted. The spectrum was then Spectra of electrochemically oxidized polyaniline.
checked by recording it again in order to ensure that no The top of a piece of conducting glass (8 mm x 50 mm x
immediate decomposition of the film had occurred during 1.5 mm) (with indium oxide coating) was connected to
the recording of the first spectrum. After the spectrum an alligator clip. The contact between the clip and the
had been obtained, the HCI solution was removed by conducting glass was secured with a coating layer of
tissue paper as described above and, as a precautionary Electrodag. A piece of Pt foil (12.5 mm x 12.5 mm) was

1834 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline W S Huang and A G MacDiarmid
Pt

~2 .
~ --Five minute epoxy
~O.e - • " " " .;.
Teflon tope

0.6

~0.4 Conducting gloss

#_ i~i
0 I~ ~ Polyoniline
tl
I I I I I I t !; O l M Pb(BF4) 2
0 OI 02 03 014 05 06 017
Electric chorge 0 (C) h~ iii ~015 M HBF4

Figure 1 Plot of potential (V) versus electric charge (Q) for the slow .................... ". . . . . . . . " ..........

electrochemical oxidation of leucoemeraldine at 10 #A cm -2 in 0.1 M

Pb(BF4) 2 + 0.5 M H B F 4 solution Figure 2 Electrochemical cell used in the step-wise electrochemical
oxidation of leucoemeraldine in 0.1 M Pb(BF4)2 + 0.5 M HBF4
solution
used as a counter electrode. The conducting glass was
used as the working electrode. The electrolyte solution
was0.1 M Pb(BF4)2 + 0.5 M HBF4. The two electrodes A constant potential of 0.50V versus Pb was then
were ~ 1 cm apart with the conducting side of the glass applied between the polyaniline electrode and the Pb
facing the Pt foil. The film of emeraldine hydrofluoro- electrode. After 3 h the cell was disconnected from the
borate was synthesized on the conducting glass using the power source and was allowed to rest for 1 h. The voltage
constant applied potential method (0.75 V). The time had dropped to 0.46 V versus Pb. The spectrum of this
used for growing the film was ~ 180 rain. The glass material was recorded as that used for the first spectrum.
electrode was then rinsed with 0.1 M Pb(BF4)2 + Additional spectra were taken after a 3 - 4 h constant
0.5 M HBF4 solution for ~ 2 - 3 min. An electrochemical potential oxidation of the polyaniline followed by a stand
cell was constructed as shown in Figure 2. A Pt wire was period of ~ 1 h. The potentials (versus Pb) applied and
wound tightly around the indented part of the conducting the potentials (versus Pb) that were recorded after the
glass. A piece of Pb ribbon was scraped with a knife to 1 h stand period (given in parentheses) were 0.60 V
clean the surface, then the upper part was surrounded (0.57 V), 0.90 V (0.86 V) and 1.00 V (0.93 V). A constant
by Teflon tape to prevent its contacting the Pt wire or potential oxidation during ~ 16 h was carried out at
the conducting glass. 0.70 V versus Pb (0.70 V versus Pb) and at 0.80 V versus
The conducting glass and Pb ribbon were inserted into Pb (0.80 V versus Pb). All spectra except the first three
a rectangular cell (4 mm x 10 mm x 60 mm). A solution were recorded both with a Cary 14 and DU-8 u.v.-visible
containing 0.1 M Pb(BF4)2 + 0 . 5 M H B F 4 was added spectrophotometer to check that the spectra were
to the cell to a level such that the polyaniline was identical. The spectra given in Figure 3 are those
completely covered. Some Teflon tape was inserted into recorded on the Cary 14 spectrophotometer.
the entrance of the cell to seal the cell and to fix the Change in spectrum during protonic acid doping o f
electrodes firmly in place, then 'five minute' epoxy emeraldine base
cement was placed over the tape. The Teflon tape on the
top of the cell prevented the epoxy cement from flowing On conducting glass substrate. A film of emeraldine
into the cell. All of the above manipulations were carried hydrotetrafluoroborate (8 mm x 30 mm) was electro-
out in an argon-filled glove bag to prevent air from chemically deposited on a piece of conducting glass
entering the cell. The polymer was reduced at a constant (8 mm x 50 mm ; In/Sn oxide coating) as described
applied potential of 0.20 V versus Pb during ,-~ 16 h. above; the time for polymerization was ,-~3 h. The final
The first spectrum was taken immediately after potential to which the film was subjected was
disconnecting the cell from the power source, using a ~0.50-0.70 V versus SCE. The resulting assembly was
DU-8 u.v.-visible Beckman spectrophotometer which then rinsed briefly for ~ 1 min in distilled water and
covers the spectral range 850-350 nm. The spectrum was placed in 500 ml of distilled water for ~ 16 h. No attempt
was made to exclude air from any of these procedures.
recorded in the same way as described before, except that
a similar cell containing a piece of conducting glass and This treatment should result in complete deprotonation
0.1 M Pb(BF4) 2 + 0.5 M HBF4 solution was used as a of the emeraldine salt to yield the corresponding
blank. Two hours after the spectrum had been recorded emeraldine base, namely:
it was observed that the potential of the polyaniline had
increased to 0.40V versus Pb. This might have been
caused by a trace amount of oxygen remaining in the
(._~._~I__~lN _~+~_~_lN-
H H H H
/
• ..
solution, which would be expected to oxidize the film. A BF; BF,~
spectrum was taken of this film using a Cary 14 H H
spectrophotometer to extend the spectral range between
350 nm (3.54 eV) and 1250 nm ( ,-~ 1.0 eV). (-~ I- - ~ I--~-- N= ~ N-)x*(2x)HBF4

POLYMER, 1993, Volume 34, N u m b e r 9 1835

 Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

as described for the leucoemeraldine. A portion of the

solution was then diluted to half the original
concentration by adding an equal amount of 80% acetic
acid to the solution prepared from the solid emeraldine
base. Repeating the same procedure four times yielded
=~ concentrations of 1/2, 1/4, 1/8 and 1/16 of the original
concentration. Spectra were taken of each of these
solutions and 80% acetic acid solution.
/

/// " ~,,~ Comparison of spectra of emeraldine salt as asolid and

8 HI 1;~7'~~~~'\\ in solution. A piece of free-standing emeraldine •2HC104
¢ 1.0 ' / ," attached to a quartz plate was prepared electrochemically.
,. The film was allowed to dry in air, and the spectrum was
,~ recorded in air. The spectrum of a solution of chemically
made emeraldine base in 80% aqueous acetic acid was
! ~ obtained as described above.

i~
'\ ' Spectrum of emeraldine base in concentrated sulfuric
05 acid. A 20 mg sample of chemically made emeraldine
hydrochloride was put in 200 ml of 96% H2SO4. The
solution was then stirred until no solid could be seen.
A The colour of the solution was brown-black. Then 30 ml
portions of this solution in 96% HzSO4 were diluted by
t ~ f ~ adding 3.66, 7.14, 11.04, 15.45 and 20.49ml of H 2 0
M.o 1.5 z.o 2.5 3.0 315 respectively to make 90%, 85%, 80%, 75% and 70%
Photon eneroy(eV) concentrated H2SO 4. These solutions were prepared in
Figure 3 Absorption spectra of different oxidation states of an ice bath to keep them cool. Then the spectra were
polyaniline. The voltages given are the open circuit potentials recorded as described for the leucoemeraldine in quartz
characteristic of polyaniline in different oxidation states: A, 0.20 V, cells (no blank in reference beam).
B, 0.40V; C, 0.46V; D, 0.57V; E, 0.70V; F, 0.80V; G, 0.86V;
H, 0.93V Slow oxidation of emeraldine base in 80% acetic
acid. Emeraldine base was dissolved in acetic acid. A
The piece of conducting glass on which the film capillary was used to transfer a very tiny amount of
adhered firmly was placed in the same cell as described ammonium persulfate aqueous solution to the quartz cell.
for the leucoemeraldine and distilled water (pH ~ 5 - 6 ) The solution was stirred using the capillary tube. A
was added; the spectrum was recorded as described for spectrum was then recorded. This procedure was
the leucoemeraldine. The conducting glass/film assembly repeated until the solution became purple.
was placed in 300 ml of 1.0 x 10 -4 N HCI for ,-, 16 h. A
spectrum was then recorded. The film was washed and
the cell was refilled with HCI solutions having molarities RESULTS AND D I S C U S S I O N
of 2.0 x 1 0 - 4 ( 3 h),4.0 x 10-4(4.5 h),6.0 x 1 0 - 4 ( 2 h ) ,
8.0 x 10 -4 (16 h) and 1.0 x 10 -3 (2 h). The figures in n-n* transition in reducedpolyaniline (leucoemeraldine)
parentheses correspond to the equilibrium times in acid The spectral range obtainable with the Beckman DU-8
before spectra were recorded, u.v.-visible spectrophotometer is ~ 1.5 to ,-, 5.6 eV
(850-220 nm ). The spectra were identical in the pH range
On quartz substrate. A green, free-standing film of 2-13, both in HC1 and H2SO 4. However, on changing
emeraldine hydroperchlorate was attached to a quartz the normality of the H2SO 4 solution from 10-2 to 9.0 N,
substrate as described for the leucoemeraldine. The film a small but significant blue shift from 315 nm to 298 nm
on quartz was first rinsed with 0.1 M K O H and was then was observed, as can be seen in Figure 4. This shift was
placed in 0.1 M K O H for 30min. The film was reversible. On changing the normality back to 1 0 - 2 N
transferred to a quartz cell (4 mm x 10 mm x 60 mm) the peak maximum was 315 nm. The extent of
which contained 0.1 M K O H . The spectrum was protonation is not known at any pH.
recorded as described for the leucoemeraldine on a When the leucoemeraldine film in 1 0 - 2 N H 2 S O 4
Perkin-Elmer Hitachi model 330 spectrophotometer. The was exposed to air for ~ 10 h, during which time it turned
solution was removed. The film was washed with water to a light green colour, its spectrum changed from that
twice and then with 1 M HC1 twice. Finally, after placing given in Figure 4 to that given in Figure 5. The main
fresh 1 M HCI in the cell, the spectra were recorded, effects noted were that the maximum at 315 nm in Figure
4 shifted slightly to 327 nm in Figure 5 and that a new
Emeraldine salt peak appeared at 424 nm in Fiaure 5. Also in Fioure 5 a
Spectrum ofemeraldine salt in aqueous solutions of 80% new absorption appeared to be growing at a value greater
acetic acid. The emeraldine base form of polyaniline than 850 nm.
powder 2 was put in 80% acetic acid. The solution was Since polyaniline contains benzene rings, the dis-
stirred for 10 min then the powder and solution were cussion will be built up from the benzene rc-r~* transition.
filtered through a funnel containing a filter paper. The Benzene exhibits two absorption bands around 180 nm
filtered green solution was then transferred to a quartz and 200 nm and a weak absorption band around 260 nm
cell ( 4 m m x 10mm x 6 0 m m ) and spectra were taken ( ~ 4 . 8 eV). All three bands are associated with the

1836 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

E
¢3

g
o

1.5 20 2.5 3.0 40 5.0 (eV) 1.5 2.0 2.5 5.0 4,0 5.0 (eV)

i
I I t , l
I I '
I t
II i
IJ I I
I I ' , i
II t
I i
II I
Ii '
I I
800 600 400 200 (nm)
800 600 400 200 (nm)
Wovelength Wovelength
Figure 5 Absorption spectrum of the polymer obtained by oxidation
Figure 4 The absorption spectra of A, leucoemeraldine base and of leucoemeraldine base by air in 10-2 N H2SO4
B, leucoemeraldine salt
from VEH calculations and X-ray photoelectron
spectroscopy data, the first two or three bands are
electron system of benzene. The two intense bands may contributed from n-n* transitions in these polyaromatic
be ascribed to transitions to dipolar excited states, while systems (PPP, PPS 3~, ppya2, etc.). In the case of
the weak 260nm band is ascribed as the forbidden polyphenylene sulfide, Bredas et al. 31 pointed out that
transition to a homopolar excited state. The 260 nm the lowest band was mainly a contribution from 3p=
( ~4.8 eV) band possesses fine structure and is intense sulfur and 2p= carbon orbitals (where z is the direction
in substituted benzenes, perpendicular to the plane containing the sulfur atoms).
Substitution of benzene with polar groups containing This band has a bonding character between the p=
unshared electrons (auxochromes, like OH or NH 2), orbitals on sulfur atoms and sulfur connected to carbon
shifts the absorption bands to longer wavelengths and atoms. The second band is flat and gives rise to a very
also intensifies them. The effect of auxochromes on the high peak in the density of states due to the very small
benzene bands is explained in terms of the interaction of contributions coming from the sulfur atoms as well as
unshared electrons with the benzene nucleus. Braude from the carbon atoms bonded to sulfur. Similarly, the
calls this interaction ~-p conjugation2a. The phenyl two highest occupied bands for leucoemeraldine were
derivatives of the nitrogen show progressive red shifts in composed of delocalized n orbitals from the rings with a
the 260 nm band, which shifts to longer values with strong admixture of nitrogen p= orbitals based on
an increase in the number of phenyl groups. For example, Boudreaux's VEH results 27.
aniline, diphenylamine and triphenylamine exhibit bands Based on the above discussion, the ~-~* transition at
at 280, 285 and 297 nm, respectively. Such a variation 3.94 eV of leucoemeraldine base is the transition
in a series of compounds is a measure of the resonance associated with n electrons of the benzene ring delocalized
interaction of the phenyl groups with the central atom 24. on nitrogen atoms. Possible resonance forms are :
U.v. spectra of some oligoanilines have been studied
by Honzl and Tlustakova 2s. In the case of polyaniline, (_~ ~ _~_! ) (~_~= ~ _~! )
Zach 26 investigated the band structure of an infinite ~ - ~ N - ,
linear polyaniline using the HCO approximation.
According to his calculation )]'max of polyaniline was Thus, we would expect that protonation of the lone pair
327 nm (E = 3.79 eV). Most of the band gap studies of electrons on the nitrogen would hinder this conjugation
conducting polymers were carried out by using valence and would shift the absorption band to a value greater
effect Hamiltonian (VEH) calculations. By assuming the than 3.94 eV. This effect is similar to the protonation of
torsional angle of 30 ° between adjacent phenyl rings, aniline monomer. It has been found that the anilinium ion
Boudreaux et al. 27 obtained a band gap of ~ 3.8 eV for shows very similar u.v. spectra to benzene 33, because the
leucoemeraldine. When a torsional angle of 0 ° was rc electrons are localized in the benzene ring. The 280 nm
adopted, a band gap of ~ 3.6 eV on the polymer was (4.43 eV) band in aniline shifts to 260 nm (4.77 eV) in
obtained. These theoretical results are fairly close to the the C6HsNH ~" ion 33.
experimental findings. Recently, Kobayashi et al. 2s Huang et al. 2 have shown that leucoemeraldine base
reported the ~-n* transition at 305 nm (4.07 eV) for the starts to become protonated at pH ~< -0.2, i.e.
reduced polyaniline (leucoemeraldine). Our study shows > 1 N HC1. The electrochemical data indicate that
that the leucoemeraldine base form (pH/> 2) has an somewhere between pH0.3 and p H - 0 . 2 extensive
absorption maximum at 315 nm( --~3.94 eV ) (Figure 4). protonation is expected to occur, namely:
In general, well reduced forms of polyaromatic systems H H
have the first absorption band assigned to the ~-rc* ('-(fl-~NY - - ~ ' ~ NY- ~ *(Z~)HA~(-~"~N.-~---~Nz~
, ,
transition. For example, 3.2eV for polypyrrole29 and \~..x_~/.. /, \ k~/ ~1 ~J,
2.6 eV (band edge 2.0 eV) for polythiophene3°. Actually, x A-

POLYMER, 1993, Volume 34, Number 9 1837

Optical properties of polyaniline." W. S. Huang and A. G. MacDiarmid

oxidized for the following spectral studies. A plot of the

~o - ~ O electric charge released by leucoemeraldine as a function
of increasing applied potential is given in Figure 1. This
- Q- V curve shows the potentials with an inflection point
at ~0.7 V versus Pb. At a potential greater than ~ 1.0 V
0.8- a constant potential was obtained which continued
indefinitely. This may be caused by the polyaniline being
- chemically reduced by some species in the electrolyte as
- rapidly as it was oxidized electrochemically. By plotting
06 - ~ _ the data in Figure 1 as V versus dQ/dV, the curve in
o_ - o . . ~ Figure 6 is obtained. A maximum at ~0.5 V versus
,~ - o . ~ P b / P b 2+ and a minimum at ,-,0.70 V versus P b / P b 2+
>" -- J are clearly observed. This plot is equivalent to a cyclic
" o.a ~...._.._--o~ voltammogram.
As the polyaniline was oxidized electrochemically from
0.20 to 0.93 V in an electrolyte consisting of 0.1 M
Pb(BF4) 2 +0.5 M HBF4, several changes in the spectrum
o.2 - were observed (see Figure 3). These were as follows.
1. The maximum at 1.45 eV in curve A increased in
- intensity and shifted to 1.78 eV (curve H).
o I I I ~ I I I I I I r I v t I I 2. Two peaks were observed on this absorption at low
o 0.5 to =.5 oxidation potential (curve B) and high oxidation
dO/dV (C V") potential (curve H).
3. The minimum at 1.7eV (curve B) changed to
Fignre 6 Graphof potential, V, versusdQ/dV obtained by replotting 2.6 eV in curve H.
the data given in Figure 1 4. A shoulder at ~ 2.70 eV (curve B) developed rapidly
into a well defined peak at 2.75 eV in curve C. On
further oxidation this grew and shifted to a maximum
Thus very little protonation is expected at pH 2. Hence,
the shift in the ~-n* transition in leucoemeraldine base at 2.90 eV (curve E). On further oxidation it decreased
from 315 nm (3.94 eV) at pH 2 to 298 nm (4.17 eV) in slightly in intensity and shifted to 3.1 eV (curve G).
9 N H2SO4 is obviously associated with protonation of 5. As oxidation proceeded the intensity of curve A at
the leucoemeraldine base as given by the equation above. 3.54 eV (the end of the spectrum ) decreased, as can be
Although the extent of protonation is not known, the seen by the values of curves B, C, D, E, F, G and H
electrochemistry seems to suggest that at least half of the at 3.54 eV. The maximum was not obtained due to
protonation has been achieved when the acidity is greater the high absorption of conducting glass at the deep
u.v. range.
than 1 N HCI. This correlates well to the equal amount
of electron transfer and proton transfer during the The spectrum of completely reduced polyaniline has
oxidation of leucoemeraldine to emeraldine (half been discussed in the previous section. The two new
oxidation of the polymer). On the other hand, no major absorption peaks (at 1.5 eV and 2.7 eV) appearing
absorption peak at 260 nm, which is the absorption of between the n - n * transition seems to suggest the creation
benzene or the anilinium ion, indicates that full of common polaronic and bipolaronic band structures
protonation of the polymer may not have been achieved during the p-doping process, similar to that of
even in 9 N H2SO 4. polythiophenes and polypyrroles. In these polymers, two
Leucoemeraldine is not stable in air. After 1 day in new within-gap defect bands are created during the
10 -2 N H2SO4, the colour changed from transparent oxidation process. However, the band structure calcula-
yellow to light green-yellow. The spectrum in Figure 5 tion by Stafstrom et al. using VEH gives a totally different
shows that the leucoemeraldine base form is oxidized, conclusion 6, and predicts that only one defect band exists
Thus the peak in non-protonated leucoemeraldine base within the gap states during the oxidation of polyaniline.
at 315nm (3.94eV)(Figure 4 ) h a s shifted to 327nm A polaronic band structure was proposed for the
(3.79eV). At the same time a new peak at 425 nm half-oxidized polyaniline (emeraldine salt) in this
(2.92 eV), characteristic of polyaniline oxidized to an calculation, which is consistent with magnetic studies 3.
oxidation state slightly lower than emeraldine (Figure 5), A schematic band structure based on theoretical
has appeared, calculations'6 27 ' 3 4 is summarized in Figure 7. The
reduced state of polyaniline (leucoemeraldine) has a
Change in spectrum of leucoemeraldine during n - n * transition between the highest occupied band (band
electrochemical oxidation a) and lowest unoccupied band (band x) at 3.9 eV as
The step-wise electrochemical oxidation of leuco- described in the previous section. When the polymer is
emeraldine was carried out at a constant pH of 0.3, i.e. oxidized to form emeraldine salt, half of the electrons in
0.5 N H +. As shown in the above discussion, leuco- the valence band (band a) are removed. Due to the
emeraldine base is only slightly protonated at this pH destabilization of this highest occupied band, the band
value, whereas emeraldine base is about completely moves up slightly to form the within-gap defect band
protonated TM at pH 0.3. (polaron band). The lowest unoccupied band (band x)
The study of electrochemical potential (V) versus is so flat that the stabilization of the upper polaron defect
electric charge (Q) was performed in order to determine band (band x') does not shift the band down too much.
the potential range for which the polyaniline could be Therefore, the bands x and x' are nearly degenerate. The

1838 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline." W. S. Huang and A. G. MacDiarmid

x (CB) x, x' x

5.9 eV

o (CB)

=~ l l a(vB) l o 2.2eV
(Poloron bond)
'" 1.5 eV
I I 0,v ,
tl0

Leucoemeroldine bose Emeroldine solt Pernigroniline bose

Figure 7 Schematic energy level diagrams ofleucoemeraldine base, emeraldine salt and pernigraniline (CB, conduction band ; VB, valence band )

observed absorption at 1.5 eV is due to the transition is a significant deprotonation of the polymer as described
from band b to band a (polaron band), and the in a previous paper 2. The continued growth of the
absorption at 2.7 eV is the transition from band c to adsorption maximum at 1.5 eV with blue shift to 1.78 eV
band a. When the polymer is further oxidized to (from spectrum E to spectrum H) means that the non-
pernigraniline (fully oxidized polyaniline), band a is protonated quinone-imine segments formed during the
completely empty, forming the lowest unoccupied band. partial deprotonation of polyaniline still contribute to
The absorption at 2.2 eV corresponds to the transition the overall absorption in this range. However, these
from band b to band a. The value of 2.2 eV is based on quinone-imine segments do not contribute any absorp-
the recent calculation by Santos and Bredas 34. Based on tion to the range near the absorption maximum at
their choice of parameters, the calculation shows that the 2.7-3.0 eV, and hence cause the decrease of intensity at
valence band (VB) and the conduction band (CB) this absorption peak. This can be easily understood by
globally extend over 2.9 eV and are separated by a gap looking at the emeraldine base and pernigraniline base.
of 1.4 eV. With this data, it is expected that the absorption Both of them contain non-protonated quinone-imine
maximum of the spectrum is around 2.2 eV. The segments. Emeraldine base has an absorption maximum
experimental result is around 2.35eV 35. For simplicity, at 2.1 eV and pernigraniline base has an absorption
only the absorption maximum is illustrated in Figure 7, maximum at 2.3 eV. The absorption in this range will
not the band gap transition (band edge). Also, the levels contribute to the higher energy part of the absorption
used do not represent the real energy bands, maximum at 1.5-1.78 eV, as shown in spectra E to H.
Spectrum A in Figure 3 represents a fully reduced On the other hand, neither emeraldine base nor
polyaniline (leucoemeraldine) at an applied potential of pernigraniline base has an absorption maximum near
0.2 V versus Pb/Pb 2 +. No absorption within the band 2.7-3.0 eV. If there is no separated domain of dimerized
gap is found. Spectrum E corresponds approximately to quinoid segments in the lattice structure, a single peak
the emeraldine salt oxidation state which is half oxidized at around 1.5-1.78 eV is expected. Therefore, the double
as indicated at the minimum of the Q - V curve peak at 1.78eV on spectrum H may be due to the
( V = 0.7 V ) in Figure 6. The growth of the peaks at 1.5 eV existence of an inhomogeneous distribution of deproton-
and 2.7 eV with blue shift during oxidation from 0.2 V ated segments in the polymer chain.
(spectrum A) to 0.7 V (spectrum E) is consistent with
the shift of energy band as illustrated in Figure 7. The Change in spectrum during protonic acid doping of
depletion of electrons in band a (polaron band ) to form emeraldine base
empty energy states results in an increase of probability In the late 1960s and early 1970s, Jozefowicz and
of electron transitions from bands b and c to band a, co-workers 36'37 published a series of papers on
hence the absorption peaks grow with the increase of polyaniline. One of their major areas of study was the
applied potential. In this case, the higher the potential conductivity and pH relationship; at that time, they did
applied to the polymer, the greater the number of not know that polyaniline undergoes insulator to metal
electrons that will be removed from band a. The shift of transition during acid treatment. Recent literature clearly
the polaron band to higher energy due to the shows that this acid treatment is true protonic acid
destabilization of this defect band accounts for the blue doping on a conducting polymer. In this section, the
shift of the absorption maxima in these spectra. When change of band structure during the acid and base treatment
the polymer is oxidized further to nigraniline from using u.v.-visible-near i.r. spectra will be discussed.
spectrum E (at 0.7 V) to spectrum H (at 0.93 V), there As can be seen from Figure 8, the spectra at pH 6.0

POLYMER, 1993, Volume 34, Number 9 1839

Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

Extrapolation of the slope down to the baseline gives a

value of ~ 1.4 eV. Between the two absorption maxima,
A there is a minimum at ~2.7 eV; there is also a second
minimum at ~ 4.8 eV. On the other hand, the emeraldine
hydrochloride in 1 N HCI solution shows one absorption
maximum at ~ 1.4 eV and another maximum at ~ 3.0 eV.
B The absorption peak at ~ 1.4 eV tails down to the i.r.
regime of the spectrum reaching a value of 0.5 eV at the
cut-off of the spectrum. Between these two absorption
I / ~ / - - ' , c maxima there is a minimum at ~ 2.3 eV. A second small
e / minimum appears at ~4.8 eV.
/¢ ~ Protonic acid doping of polyaniline has been extensively
~" o investigated. The polymer goes through an insulator
o / /~ ~ / (tr < 10-1° S cm-1)to conductor (~10 S cm-1) transition
/ during protonation of the emeraldine base form. It has
been proved that the conducting state (emeraldine salt)
/ E is metallic with a significant amount of Pauli magnetic
o ~ ~ / susceptibility 3. The doping process involves the con-
version of alternating benzenoid amine nitrogen units
F (reduced unit) and quinoid imine nitrogen units
(oxidized unit) to a semiquinone nitrogen cation type
G polaronic lattice as shown in Figure 10. The metallic
emeraldine salt form shows a finite density of states at the
Fermi energy level 6.
Studies have recently been carried out 3a~° on
oligomers of polyaniline in order to understand the
I0 1'5 2'0 2'5 3'o ~.5 40 structure of the emeraldine oxidation state of polyaniline.
Energy (eV)
For example, diphenyl-p-phenylenediamine

Figure8 The change in the absorption spectrum of emeraldine base O i _ ~ O

during its protonation: A, pH6, 16h; B, 10-4M HCI, 24h; C,
2 x 10-4 MHCI, 3 h;D, 4 x 10-4M HC1,4.5h;E, 6 x 10-4 MHCI,
2h; F, 8 x 10-4 M HC1, 16h; G, 10-3M HCI, 2h
is a dimer of aniline with a phenyl ring as a cap on the
end, with an absorption peak 38'41'42 around 4.1 eV. This
compound is completely oxidized to diphenyl-p-
phenylenediimine

,
~"
/ which has an absorption peak around 2.8 eV in EPA
solvent (a mixture of ether, isopentane and ethanol, in
8:3:5 ratio). A simple .correlation between emeraldine
base and this compound can be achieved by assuming
that the benzenoid amine units cause the 3.9eV

1.0 1.5 2.10 310 5/,0 (eV) H \

01i5 I I I I II I I i [ I ..

Figure 9 Absorption spectra of A, emeraldine base and B, emeraldine

hydrochloride
H
(A) and pH 4.0 (1.0 x 10 -~ N HC1) (B) are essentially !
identical. As the acidity was increased from pH 3.7 1 protonofion
(2.0 x 10-4 N HC1) to pH3.0 (1.0 x 10-3 N HC1), i.e. re,) H+
spectra B to G inclusive, the maximum at -,~2.1 eV in B
decreased monotonically to ~ 1.6 eV in G. At the same H H
time the minimum in B decreased from ~ 3.1 to ~ 2.8 eV I I" ~ \ /
in D. In E a new absorption appeared at --~3,0 eV which 0~,~N+.,~ [~N+
became progressively more well defined in spectra F and
,.co eo ~
G. N" ~ ' " Ct- " " " ~ N / ' ' ' ~ Ct-
As shown in Figure 9, the emeraldine base in 0.1 M I I ~ ,
aqueous KOH exhibits an absorption maximum at H H
2.1 eV and another absorption maximum at ~ 3.9 eV. Figure 10 Change of structure from emeraldinebase to emeraldine
The absorption peak at ,~2.1 eV tails down to -~ 1.2 eV. salt during protonation

1840 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline : W. S. Huang and A. G. MacDiarmid

_ _ rr* concentration, the reaction is only an acid/base reaction

and the polymer retains the same oxidation state.
The shift of the peak at 2.1 to 1.5eV during
protonation means that the peaks at 1.5 and 2.1 eV have
the same origin. In the theoretical calculation for
_ _ 7rq polyaniline by Sjogren and Stafstrom49, the 2.1 eV peak
3.9 ev I in the spectrum of emeraldine base is shown to have the

"'
[
~ A 8
~
2.1 eV

7rb
same origin as the 2.3 eV peak in the spectrum of
pernigraniline. As shown in Figure 7, band a is the
half-filled polaronic band in emeraldine salt. At the same
time, band a is the empty conduction band of
pernigraniline. The shift of peak position (from 2.1 to
1.5 eV) starts at pH 4, which is consistent with the onset
of conductivity during protonic acid doping of
polyaniline as reported before. The sharp change in
---~b spectrum between pH 4 and 3 was also observed recently
by Wan 5°. However, in this study the peak at 2.1 eV does
~ c not shift slowly to 1.5 eV; instead the peak at 2.1 eV
decreases while the peak at 1.5 eV increases simul-
Emeroldine bose taneously. Similar results to that of Wan have been
Figure 11 Energyleveldiagramof emeraldinebase reported by other authors x4. However, Weiss et al. 51
showed that the peak at 1.5 eV shifted systematically
absorption similar to that (4.1 eV) of the diamine about 260 nm (approximately 70 nm per pH unit) over
(described above) and the quinoid imine units cause the the pH range 3-7 on the spectra of their polyaniline/
2.1 eV absorption similar to that (2.8 eV) of the diimine, dinitrocellulose composite film. We believe that the
This simplified correlation is not far from the theoretical formation of two peaks is simply an indication of the
calculation. The absorption at 2.1 eV for the emeraldine formation of protonated and non-protonated domains
base has been attributed by Stafstrom et al. 43 to the l in the polymer at such a low protonation level. If the
excitation from HOMO (highest occupied molecular polymer chain is not in a confined condition and is
orbital, rib) of the three-ring benzenoid part of the system allowed to equilibrate with the surroundings, a single
to LUMO (lowest unoccupied molecular orbital, rrq) of peak in the spectrum is expected.
the localized quinoid ring and the two surrounding imine Protonation of polyaniline results in an increase in the
nitrogens (see Figure 11). The excitation of the 3.9 eV amount of charge on the polymer backbone, thus
peak was attributed to two different transitions. One is increasing the conductivity. However, it is clear that the
the z-rr* transition, similar to that of leucoemeraldine, dramatic 10-fold increase in conductivity from insulator
and the other one is a transition from low-lying orbitals to conductor is due to lattice distortion of the polymer
to the ~rq orbital, chain, which then increases the mobility of the charges
All the theoretical calculations on the emeraldine base in the polymer, and not simply due to the increase in the
form of polyaniline were based on our proposed amount of charge on the chain. The half-filled polaronic
structure, which contains alternating benzenoid amine band model explains the observation of metallic
nitrogen units and quinoid imine nitrogen units. This has behaviour in the emeraldine salt form of polyaniline.
been indirectly proved by electrochemistry24 ' , chemical Emeraldine salt will not be completely protonated until
analysis44 and e.s.r, studies45'46. Recently, n.m.r, the pH of the solution reaches 0. When the pH of the
studies47'48 have provided the strongest evidence for this solution is 3, it is expected that the polymer will be
proposed structure, protonated to only a very small extent. The shift of the
In order to understand the change of band structure peak at 2.1 eV to 1.5 eV between pH 3 and 4 indicates
during protonation of the imine nitrogens of emeraldine that the lattice distortion of the polymer backbone occurs
base, protonation experiments were carried out at even with a small amount of protonation.
different pH values. It was found that the band structure
underwent a great change between pH 4 and 3. Therefore, Emeraldine salt
one experiment was performed under conditions very As shown in Figure 12, the spectrum of a solution of
close to equilibrium at six different pH values between chemically made emeraldine base in 80% acetic acid is
4 and 3. The emeraldine salt film was then deprotonated almost identical to that of the free-standing electro-
in distilled water for more than 16 h and was equilibrated chemically made emeraldine-2HCIO 4 film. Hence the
at pH 4 overnight, spectra are not affected by the nature of perchlorate or
The curves in Figure 8 clearly show the tendency of acetate anion.
the p e a k t o s h i f t from ~2.1 to ~ l . 5 e V w h e n t h e HCI Figure 13 gives a series of spectra at different
concentration increases from 10 -4 to 10 -3 M HC1. At concentrations of emeraldine base in 80% acetic acid.
the same time, the peak at ,-~3.0 eV starts to grow during The intensity of each peak decreases when the
this change. The observation of this peak is consistent concentration decreases. The intensity of each peak is
with the deprotonation of the polymer when the potential roughly proportional to the concentration. In general,
ofpolyaniline versus Pb/Pb 2+ is above 0.7 V as discussed peaks A, B and C more or less obey Beer's Law, i.e.
in the previous section. This indicates that deprotonation A = abc, where~A is absorbance, a is molar absorptivity,
can be achieved either by decreasing the proton b is sample cell length and c is concentration.
concentration in the solution or by increasing the applied Mohilner et al. s2 suggested in 1962 that anodically
oxidizing potential. If we increase or decrease the proton formed polyaniline is principally emeraldine. They

POLYMER, 1993, Volume 34, Number 9 1841

Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

A of electrochemically grown film. Identical spectra were

obtained as shown in Figure 12. Since the chemically
made powder shows the same optical properties as
electrochemically grown film, all discussions based
-~ on electrochemically grown film are also valid for
chemically made powder. Furthermore, these solutions
should exhibit significant electronic conductivity as well
as ionic conductivity. The identical electrochemical
~S
o
behaviour of chemically made and electrochemically
~ made polymer has been demonstrated previously 2. Both
~ ~ materials exhibit reversibility in both oxidation/
"~ reduction and acid/base reactions.
~ In the previous section we pointed out that the high
conductivity of emeraldine salt was attributed to the low
absorption edge extended through the i.r. regime of the
band at ~1.5 eV, which is consistent with metallic
behaviour. This raises the question of whether any
1.5 2.0 2.5 3.0 4.0 5.0 (eV) absorption band is attributed to the interchain
I I I 1 I I
I ' ' ~ I a , , I , i , I interaction. To answer this we can either apply high
800 600 400 200 (am) pressure to a piece of film or dilute the polymer solution
Wavelength to see the change in the spectrum.
Figure 12 Absorption spectra of A, emeraldine base in 80% acetic
Yakushi et al.29 could not obtain direct experimental
acid and B, electrochemically grown emeraldine hydroperchlorate film evidence for an interchain transition, because high
pressure did not significantly affect the transmittance
data of polypyrrole. On the other hand, Wellinghoff
et al.53, studying polycarbazole, found in their model
compound 9,9-dimethyl-3,3'-dicarbazolyl tetrafluoro-
A borate
cH~
N÷

g
" that there is an intermolecular charge transfer band
c at 1900nm, because the intensity of the 1900nm
transition increased continuously relative to the other
B band absorbance at 8500 nm with increasing solution
,~ concentration.
The observed absorption spectra of the emeraldine salt
i~ (Figure 13) do not show any interchain effects. This is
proved by diluting the solution of polyaniline in 80%
acetic acid. If there is any absorption peak due to
interchain interaction, when the polyaniline concen-
tration decreases the peak will disappear or decrease
:..--------------'----w~6
80 % acetic acid dramatically. The experimental results demonstrate that
the absorption bands do not deviate significantly from
~5 20 25 3o 40 5o (eVl Beer's law. Although no significant interchain interaction
I I I I I I was found in the emeraldine salt, the experimental results
' I ~ ' ' I ~ J ~ I J , I of Wan 5° on emeraldine base in N-methylpyrrolidone
800 600 400 200 (nm)
Wavelength ( N M P ) solution show a red shift of the peak at 630 nm
when the concentration of the polymer solution is high.
Figure 13 Absorption spectra of emeraldine base in 80% acetic acid This suggests that there is some interchain interaction
at relative concentrations of 1, 1/2, 1/4, 1/8 and 1/16. Peaks A, B and which decreases the excitation energy needed to transfer
c are discussed in the text the electrons from the benzenoid rings to the quinoid
rings in the polymer chain. Wan thinks this shift may be
reported examples of similar behaviour between electro- due to an unpaired electron on the quinoid segment of
chemically synthesized and chemically made material, the polyemeraldine chain. Stafstrom et al.43 have shown
Both materials were reported to be amorphous and that any charge removed from the benzenoid amine
neither melted up to 300°C. In addition, it was found segments to form positive polarons or any charge added
that the electrochemically made emeraldine base to the quinoid imine segments to form negative polarons
dissolved in cold 80% acetic acid to give the same green will cause a transition lower than the 2.0 eV absorption.
colour as chemically made emeraldine base dissolved in Therefore, any charge separation in the same chain or
cold 80% acetic acid. Both were reported to have similar charge transition from one chain to the other will cause
i.r. spectra. No visible spectra were reported, a red shift of this 2.0 eV absorption peak.
We obtained the solution spectrum of emeraldine base As shown in Figure 14, the spectrum of emeraldine
dissolved in 80% acetic acid and compared it to that hydrochloride in 70% H 2 S O 4 (H o = - 5 . 6 5 ) i s identical

1842 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

~/
tetramers at their half oxidation state. Therefore, we
A believe that this transition is due to significant
protonation which breaks the conjugation into small
segments, thus the lattice structure changes from
polaronic to bipolaronic.
~ / / ~ ~ B When further protons are added to the amine nitrogen,
a peak at ~2.5 eV starts to grow and the peak at 1.5 eV
g
>,
starts to decrease. This could indicate that the quinoid
segment now has localized positive charges, when all the
amine nitrogens are protonated.

H H H H
g N

/ .......j E ,, . y

j
F (£3- ,"
H
"+ -(3- ,";,)
H

The peak at ~ 2 . 5 e V may be due to the electron

transition from the benzenoid rings to the quinoid rings,
.5 20 2.5 3.0 4.0 5.0 (eV) similar to the case of emeraldine base or pernigraniline.
[ I ' ~ , I I . .I t I I t It I I The fine structure at ~ 4 . 8 e V is an indication of the
800 600 400 200 (nm) existence of isolated benzene rings due to the extensive
W0velength protonation of amine nitrogens.
Figure 14 Absorption spectra ofemeraldine hydrochloridein H2SO, Figure 15 is a qualitative study of the slow oxidation
solutions of various acid strengths (Ho refers to the Hammett acidity ofemeraldine salt to pernigraniline. The peak at ~ 1.6 eV
function): A, 96% H2SO4 (Ho=-8.98); B, 90% H2SO, shifts to higher energy during the treatment with
(Ho = -8.27); C, 85% H2SO4 (H0 = -7.66); D, 80% H2SO,
(Ho=-6.97); E, 75% H2SO4 (Ho = -6.30); F, 70% H2SO, ammonium persulfate. Two peaks at ,-,3.0 and ~ 3 . 6 e V
(Ho =-5.65) decrease in intensity. Another small peak appears at

to that of emeraldine base in 80% aqueous acetic acid

as observed in Figure 13. When the acidity is increased,
the peak at ,-, 3.0 eV decreases in intensity. This peak
disappears between the concentration of 75% and 80%
H 2 S O , . Above 80% H 2 8 0 4 , a new peak at ,-,2.5 eV
starts to grow and becomes very intense. On the other
hand, the peak at ,-, 1.6 eV decreases in intensity. The
colour of the solution changes from green to blue to
violet to brown when the concentration increases from
70% to 96%. When concentrated H2SO , was added to
the more dilute solutions, spectra characteristic of the
more concentrated solutions were obtained. Hence the E=
effects are reversible. ,,
Since 12 M HC1 has an acidity function of only
H o = - 4 . 4 1 compared to H o = - 5 . 6 5 for 70% H2SO4,
the acidic strength is still not high enough to cause
significant protonation on the polyaniline in 12 M HCI. o
We would not expect any change of spectra in 12 M HCI. ~
The spectra shown in Figure 14 have a transition <
.a

between 75% and 80%, where the peak at ,,,3.0eV

disappears. We believe this could be related to the
transition from polarons to bipolarons due to significant
protonation.
H H H H

(-0'0- -0-'-0' )
l xH*
.5 2.0 2.5 5.0 4.0 5.0 (eV)

I
\ _k~/ 7+-'~,~7- '~- ~ / = N -'%=/-N - J x 800 600 400 2 0 0 (nm)
H
Wovelength
It has been found both experimentally 39 and theoreti- Figure 15 Absorption spectra of emeraldine base in 80% acetic acid
cally 49 that the bipolaron lattice is preferred in the during treatment with ammonium persulfate

POLYMER, 1993, Volume 34, Number 9 1843

 Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid

,,,4.5 eV (278 nm) and becomes more intense during interchain interaction. However, the study ofemeraldine
treatment with (NH4)2S208. base in N M P by Wan seems to indicate some interchain
Recently Sun et al. 3s synthesized a pernigraniline form interaction in the high concentration range of polymer
of polyaniline that has two absorption maxima, at 528 nm solution.
( ,~2.35 eV) and 320 n m ( ~3.88 eV). The absorption at
528 nm is the transition through the Peierls gap of the REFERENCES
dimerized pernigraniline structure. The absorption at 1 Salaneck,W. R., Lundstrom, I. Huang, W. S. and MacDiarmid,
320 nm is probably the transition from the low-lying A.G. Synth. Met. 1986, 13, 291
orbitals to the conduction band. Due to the big 2 Huang, W. S., Humphrey, B. D. and MacDiarmid, A. G.
absorption peak at 278 nm, the peak at 320 nm is not J. Chem. Soc., Faraday Trans. 1 1986, 82, 2385
clearly shown in Figure 15. 3 Epstein,A. J., Ginder, J. M., Zuo, F., Bigelow, R. W., Woo,
H.-S., Tanner, D. B., Richter, A. F., Huang, W. S. and
MacDiarmid, A. G. Synth. Met. 1987, 18, 303
4 Huang,W. S., MacDiarmid, A. G. and Epstein, A. J. J. Chem.
Soc., Chem. Commun. 1987, 1784
CONCLUSION 5 Paul,E. W., Ricco, A. J. and Wrighton, M. S. J. Phys. Chem.
1985, 89, 1441
Our experimental findings have provided detailed 6 Stafstrom,S., Bredas, J. L., Epstein, A. J., Woo, H.-S., Tanner,
information on the electronic structures of polyaniline at D.B., Huang, W. S. and MacDiarmid, A. G. Phys. Rev. Lett.
various oxidation and protonation stages. We have 1987, 59, 1464
investigated the interconversion of band structures using 7 Cushman,R. J., McManus, P. M. and Yang, S. C. J. Electroanal.
Chem. 1987, 219, 335
optical-absorption data in two ways : (1) changing the 8 Genies,E. M. and Lapkowski, M. J. Electroanal. Chem. 1987,
degree of protonation by changing the acid strength of 220, 67
the polymer surroundings; and (2) changing the degree of 9 Genies,E. M. and Lapkowski, M. Synth. Met. 1987, 21, 117
oxidation by changing an applied electrochemical 10 Cushman,R. J. Makromol. Chem. Rapid Commun. 1987, 8, 69
potential or treating the polymer with different amounts 11 Sariciftci,N. S. and Kuzmany, H. Synth. Met. 1987, 21, 157
12 Li, Y., Yan, B., Yang, J., Cao, Y. and Qian, R. Synth. Met.
of oxidizer. 1988, 25, 79
When the p H is above 2, very little protonation 13 Neugebauer,H., Neckel, A., Sariciftci, N. S. and Kuzmany, H.
occurs on leucoemeraldine base. It has a n-~t* Synth. Met. 1989,29, E185
absorption peak at 3.94 eV. When p H < 2, the polymer 14 Glarum, S.H.andMarshall, J.H.J.Phys. Chem. 1988,92, 4210
15 Cao, Y.,Li, S.,Xue, Z. andGuo, D.Synth. Met. 1986,16,305
is protonated. In 9 N H2SO4 solution, the absorption 16 Cao, Y.,Smith, P.andHeeger, A.J. Synth. Met. 1989,32,263
peak shifts to 4.17 eV. This is an indication of reduced 17 Huang,W. S. PhD Dissertation, University of Pennsylvania,
conjugation length due to the protonation on the amine 1986
nitrogens, which then divides the polymer chain into 18 Huang,W. S. PhD Dissertation, University of Pennsylvania,
1986, pp. 286-297
small conjugated segments. 19 Huang,W. S. PhD Dissertation, University of Pennsylvania,
The oxidation ofleucoemeraldine in 0.1 M Pb(BF4) 2 + 1986, pp. 205-215, 313-318
0.5 M HBF4 solution causes two absorption bands to 20 Rochester,C. H. 'Acidity Functions', AcademicPress, London,
rise within the n - n * absorption band. It is believed that 1970, p. 39
a polaronic lattice structure has been formed in this 21 Paul, M. A. and Long, F. A. Chem. Rev. 1957, 57, 1
22 MacDiarmid,A. G., Chiang, J.-C., Halpern, M., Huang, W. S.,
oxidation process. The 1.5 eV absorption peak is due to Mu, S.-L., Somasiri, N. L. D., Wu, W. and Yaniger, S. I. Mol.
the transition from the valence band to the polaronic Cryst. Liq. Cryst. 1985, 121, 173
defect band. The 2.7 eV absorption peak is due to the 23 Bowden,K. and Braude, E. A. J. Chem. Soc. 1952, 1068
transition from the lower-lying band to the polaron band. 24 Matsen,F. A. et al. J. Am. Chem. Soc. 1950, 72, 5243, 5248,
5250, 5252, 5256, 5260
Both peaks show blue shift during the oxidation of the 25 Honzl, J.andTlustakova, M.J. Polym. Sci.,PartC1968,22,451
polymer from 0.2 to 0.7 V versus Pb. Above 0.7 V, the 26 Zach,P. J. Polym. Sci., Polym. Phys. Edn 1972, 10, 2379
oxidation state exceeds the level of emeraldine and a 27 Boudreaux,D. S., Chance, R. R., Wolf, J. F., Shacklette, L. W.,
significant deprotonation causes the peak near 2.9 eV to Bredas, J. L., Themans, B., Andre, J. M. and Silbey, R. J, Chem.
decrease in intensity. When the polymer is oxidized to Phys. 1986, 85(8), 4584
28 Kobayashi, T., Yoneyama, H. and Tamura, H. J. Electroanal.
pernigraniline, the dimerized structure contains a Peierls Chem. 1984, 177, 281
gap of 2.35 eV. 29 Yakushi,Y., Lauchlan, J. L., Clarke, T. C. and Street, G. B.
Emeraldine base has an absorption at 2.1 eV which is J. Chem. Phys. 1983, 79, 4774
due to the electronic excitation from benzenoid rings (rib) 30 Chung,T.-C., Kaufman, J. H., Heeger,A. J. and Wudl, F. Phys.
Rev. B 1984, 30, 702
to quinoid rings (nq). The peak at 3.9 eV was attributed 31 Bredas,J.L, Chance, R. R.,Silbey, R.,Nicolas, G. andDurand,
to two different transitions : one is the n - n * transition a.J. Chem. Phys. 1982, 77, 371
similar to that of leucoemeraldine; the other is a 32 Ford, W. K., Duke, C. B. and Salaneck, W. R. J. Chem. Phys.
transition from low-lying orbitals to the nq orbital. When 1982, 77, 5030
the p H changes from 4 to 3, the absorption at 2.1 eV 33 Kortum, G. Z. Physik. Chem. 1939, i!42, 39
34 Santos, M. C. and Bredas, J. L. Phys. Rev. Lett. 1989, 62, 2499
shifts to 1.5 eV, due to lattice distortion of polyaniline 35 Sun, Y., MacDiarmid, A. G. and Epstein, A. J. J. Chem. Soc.,
tO a polaronic structure by protonation of the imine Chem. Commun. 1990, 529
nitrogens. When the acid strength increases to above 36 Doriomedoff, M., Hautiere-Christofini, F., De SurviUe, R.,
Ho = --5.65 (70% H2SO4), significant protonation Jozefowicz, M., Yu, L.-T. and Buvet, R. J. Chim. Phys.
Physicochim. Bio. 1971, 68(7-8), 1055
s e e m s to break down the polaronic s t r u c t u r e to a c o n - 37 De Surville, R., Jozefowicz, M., Yu, L.-T., Perichon, J. and
fined bipolaronic structure. In highly acidic solution Buvet, R. Electrochim. Acta 1968, 13, 1451
(Ho = - 8 . 9 8 ) , the existence of isolated quinoid imine 38 McManus,P. M., Yang, S. C. and Cushman, R. J. J. Chem.
nitrogen units ( ~2.5 eV) and benzene units ( ~4.8 eV) Soc., Chem. Commun. 1985, 1556
seems to be obvious. 39 Shacklette,L. W., Wolf, J. F., Gould, S. and Baughman, R. H.
J. Chem. Phys. 1988, 88(6), 3955
The optical-absorption study of different concentrations 40 Lu, F. L., Wudl, F., Nowak, M. and Heeger,A. J. J. Am. Chem.
of emeraldine salt in 80% acetic acid does not show any Soc. 1986, 108, 8311

1844 POLYMER, 1993, Volume 34, Number 9

 Optical properties of polyaniline: W. S. Huang and A. G. MacDiarmid
41 Linschitz, H., Rennert, J. and Korn, T. M. J. Am. Chem. Soc. 195
1954, 76, 5839 48 Richter, A. F., Ray, A., Ramanathan, K. V., Manohar, S. K.,
42 Linschitz, H., Ottolenghi, M. and Bensasson, R. J. Am. Chem. Furst, G. T., Opella, S. J., MacDiarmid, A. G. and Epstein,
Soc. 1967, 89, 4592 A.J. Synth. Met. 1989, 29, E243
43 Stafstrom, S., Sjogren, B. and Bredas, J. L. Synth. Met. 1989, 49 Sjogren, B. and Stafstrom, S. J. Chem. Phys. 1988, 88(6), 3840
29, E219 50 Wan, M. J. Polym. Sci., Part A 1992, 30, 543
44 Chiang, J.-C. and MacDiarmid, A. G. Synth. Met. 1986,13, 193 51 Weiss, H., Pfefferkorn, O., Kotora, G. and Humphrey, B. D.
45 Genies, E. M., Syed, A. A. and Tsintavis, C. Mol. Cryst. Liq. J. Electrochem. Soc. 1989, 136, 3711
Cryst. 1985, 121, 181 52 Mohilner, D. M., Adams, R. N. and Argersinger, W. J. J. Am.
46 Travers, J. P., Chroboczek, J., Devreux, F., Genoud, F., Chem. Soc. 1962, 84, 3618
Nechtschein, M., Syed, A., Genies, E. M. and Tsintavis, C. MoL 53 Wellinghoff, S. T., Deng, Z., Kedrowski, T. J. and Dick, S. A.
Cryst. Liq. Cryst. 1985, 121, 195 Mol. Cryst. Liq. Cryst. 1984, 106, 289
47 Hgiwara, T., Yamaura, M. and Iwata, K. Synth. Met. 1988, 26,

POLYMER, 1993, Volume 34, Number 9 1845