J . Chem. SOC., Faraday Trans. 1, 1986, 82, 2323-2332 The Electrochemical Reduction of Polyacetylene with Selected Reducing Agents Richard B. Kaner,? Simon J. Porter$ and Alan G. MacDiarmid* Department of Chemistry, University of Pennsylvania, Philadelphia, Pennsylvania 19104, U.S.A. The Coulombic efficiency, stability, constant-current discharge characteris- tics, energy density and the relation of cell potential to degree of reduction of a partly reduced polyacetylene cathode, [Na,t(CH)Y--], ( y d 0. lo), in a cell of the type NalNaPF,(tetrahydrofuran)l[Naj(CH)Y-1, have been investi- gated. By comparison with data obtained with a (Lit(CH)Q-], electrode, thermodynamic properties, such as the relation of cell potential to degree of reduction at diffusion equilibrium, appear to be intrinsic t o the reduced polyacetylene and independent of the countercation, whereas kinetic prop- erties, such as the cell potential during constant-current electrochemical reduction, vary with the countercation.The electrochemical reduction of polyacetylene with the incorporation of potassium countercations has been accomplished in a KIKClO,(tetrahydrofuran)l[K,+(CH)Y-], cell using a complexing agent to solubilize the KC10, in the tetrahydrofuran. A convenient electrochemical method for incorporating organic countercations into polyacetylene is discussed, together with selected properties of a [(Bu,N)i(CH)y-], electrode. In a previous paper we have reported the properties of polyacetylene, (CH),, reduced electrochemically with the incorporation of lithium countercations.l In this paper, we investigate the electrochemical reduction of polyacetylene with the incorporation of a variety of different countercations to examine which of the observed properties are intrinsic to the reduced polyacetylene and which vary with the counterion.Experiment a1 Polyacetylene film was synthesized as described previously.2 All electrochemical experi- ments were performed using cells constructed in a purified argon dry box and then sealed on a vacuum line as described previous1y.l Lump sodium or potassium metals (J. T. Baker Chemical Co.) were scraped with a knife and pressed into 189 nickel grid (Delker Corp.) to serve as electrodes. The electrolytes consisted of 1 .O mol dm-3 NaPF, (Aldrich Chemical Co.) in tetrahydrofuran (THF), 1 .O mol dm-3 KClO, (Fisher Scientific Co.) with 1 .O mol dm-3 dicyclohexano[ 181- crown-6 (Alfa Ventron Corp.) in THF and 1.0 mol dm-3 Bu,NClO, (Eastman Kodak Co.) in THF.Anhydrous NaPF,, KClO, and Bu,NClO, were heated at 120 "C under dynamic vacuum for 48 h prior to use. Reduction of the (CH), electrodes (cell discharging studies) was performed with a Princeton Applied Research potentiostat/galvanostat model I73 at constant currents. Reoxidation to convert the reduced polyacetylene back to neutral (CH), (cell charging studies) were carried out by oxidizing the [M,+(CH)u-], (ha = Na, K or Bu,N) first at t Permanent address : Department of Chemistry and Biochemistry, University of California, Los Angeles, $ Permanent address : University Chemical Laboratory, Canterbury, Kent, CT2 7NH.California 90024, U.S.A. 23232324 Electrochemical Reduction of Poli~acetylene a constant current and then at a constant applied potential. The amount of charge passed during reduction or reoxidation was recorded using a PAR coulometer, model 179. Voltages, resistances and currents were measured with a Keithley 177 microvolt digital multimeter. Voltage-charge, voltage-time and current-time curves were recorded with a Houston Instruments Corp. Omnigraphics xy recorder. All potentials are given with respect to either a lithium, sodium or potassium reference electrode which was also used as the counter electrode. It .must be stressed that the electrochemical characteristics of (CH), electrodes are extremely sensitive to the method of cell construction, presence of impurities (especially water and oxygen), rate of reduction, etc.The results given in this report were obtained by following the described procedures exactly. Results and Discussion Relation between the Degree of Reduction, y , in [(CHY,-], and Cell Potential A cell was constructed from a sodium anode and a (CH), cathode, both immersed in an electrolyte consisting of 1 .O mol dm-3 NaPF, in THF. The open-circuit voltage, Voc, of such a cell falls in the range 1.5-2.8 V. The reason for this variation has been discussed previously. When the electrodes were connected by an external wire a spontaneous electrochemical reaction occurred in which the sodium was oxidized and the (CH), was reduced according to the following equations : anode reaction: xyNa + xyNa+ + xye- (1) cathode reaction : (CH),+xye- -+ [(CH)Y-], (2) giving the overall net reaction: xyNa + (CH), -+ [Nai(CH)Y-], (3) where y < 0.10.The reaction given by eqn ( 3 ) is the discharge reaction of a voltaic cell, which in its charged state consists of parent, neutral (CH), and metallic sodium. It is completely analogous to the electrochemical reaction which occurs between polyacetylene and 1ithium.l To obtain the relationship between the V,, and percent reduction given in fig. 1, two different cells were reduced using a constant current method. Each data point given in fig. 1 involved: (a) a constant current reduction of the (CH), at a 100 A kg-l of (CH), rate, equivalent to 5 mol% (CH), reduction per hour; followed by (b) a 48 h stand period to promote equilibration of the Na+ ions within the ca.200 A diameter polyacetylene fibrils, followed by (c) a constant current reoxidation of the polyacetylene at a 100 A kg-l of (CH), rate, equivalent to 5 mol% (CH), oxidation per hour, until a cell potential of 2.0 V was reached, at which point the cell was held at a constant potential of 2.0 V for 16 h to promote complete reoxidation of the [Naj(CH)Y-],. The data points given in fig. 1 show the relationship between Voc, 48 h and the percent reduction of the (CH),. The arrows in fig. 1 represent the increase in Yo, of each cell on standing for 48 h, while the circle and square represent the final VOc, 48 h values for the NalNaPF,(THF)I(CH), cells containing 2.7 mg (0.8 cm2) and 3.0 mg (0.9 cm2) of (CH),, respectively.After 48 h, no further diffusion equilibration within the (CH), fibrils as measured by changes in the open circuit voltage, could be observed. In order to see if the data plotted in fig. 1 for the equilibrium cell potential of [Naj(CH)Y-], us. Na are intrinsic to the reduced polyacetylene, the empirical equation used to relate the equilibrium potential of [Lii(CH)g-], us. Li [ref. (l)] was drawn in fig. 1 with a correction for the difference in electrochemical reduction potential between Na and Li. This difference is calculated by taking Ked for Na (2.71 1 V us. the standard hydrogen electrode, SHE) and subtracting it from Ged for Li (3.045 V us. SHE) to giveI .5 I .o 0.5 voc I I 1 I I I I I 2325 0 1 2 3 4 5 6 7 8 9 1 0 reduction (%) Fig.1. Relationship between the open-circuit voltage, V,,,, 4R h and the percent reduction of (CH), in an NalNaPFG(THF)IINai(CH)u-], cell. The empirical relationship Voc, 4R = I. 17 - (0.13 + 0.02q) In q, where q (q = 1OOy) is the percent reduction was used to draw the curve. The two different cells used in this study employed the following amounts of polyacetylene: 0, 3.0 mg (0.9 cm2); and 0, 2.7 mg (0.8 cm2). A g e d = 0.334 V. This difference was then subtracted from the empirical relationship between Voc (us. Li) at approximate diffusion equilibrium and the percent reduction of [Li$(CH)g-], to give Voc = (1.50 - 0.334) - (0.13 + 0.02q) In q = 1.17-(0.13+0.02q) lnq (4) where q (q = 1OOy) is the percent reduction. A fairly good agreement can be seen in fig.1 between the curve drawn based on eqn (4) and the experimental data points for reduction levels between 1.0 and 10.0 mol%. Thus the thermodynamic function of V,, us. percent reduction at apparent diffusion equilibrium for [M,f(CH)y-], (us. M) is essentially identical for all values of y < 0.10, for M = Li or Na. Therefore, the potential of the [Na$(CH)g-], electrode under these conditions reflects only the degree of reduction of the polyacetylene and is essentially independent of the countercation. The 48 h stand period is a condition needed to promote diffusion equilibration of dopant cations within the (CH), fibrils as is demonstrated in fig. 2. The upper solid line in fig. 2 is based on the best fit through the [Na$(CH)y-], us. Na data points from two different cells as given in fig.1. The middle dashed line was obtained by Shacklette et aZ.3 by measuring immediate Voc values of an Nal(CH), cell after stepping the voltage in small increments and allowing the current to decay to 10 pA cm+. This is theoretically equivalent to a constant current discharge at 10 pA cm-2 [2 A kg-l of (CH),]. The lower dotted line was obtained by discharging a Nal(CH), cell at a constant current of 0.135 mA [50 A kg-l or 175 pA cm-2 of (CH),], as described in a later section (see fig. 5, later). Note the similarity of the dotted and dashed lines, both of which exhibit a marked plateau effect which is much more pronounced than that observed in the solid line. This strongly suggests significant continuing equilibration of dopant ions during the 48 h stand period.The change in potential is not due to loss of charge by the reduced polyacetylene, since, as shown in a later section, the Coulombic recovery obtained on electrochemical oxidation of the reduced material back to neutral (CH), was ca. 100%. Baughman et aZ. have reported that alkali-metal doping of partially oriented (CH), involves ' staging' analogous to that observed when graphite is intercalated with alkali2326 - - Electrochemical Reduction of Polyacetylene 0.41 metals4* We suggest that alkali-metal doping of (CH), may involve the formation of the kinetically favoured structure exhibiting staging. Once a deformation has been introduced into the (CH), lattice by the insertion of a metal ion, further insertion of metal ions is favoured at this defect site to give a staging effect.If solvent is then removed, diffusion of metal ions to give a more homogeneous distribution of dopant ions is inhibited, resulting in the observed staging. If, however, solvent is present it promotes attainment of a more homogeneous thermodynamically favourable distribution of dopant, although some staging, but less pronounced, is still evident after the 48 h equilibration period. It would be of interest to ascertain whether analogous X-ray studies of polyacetylene, of the type previously reported by Baughman et al., if carried out on electrochemically reduced material as described here with a 48 h stand period, would lead to a more homogeneous distribution of dopant ions throughout the polymer. Diffusion of Na+ Ions in Polyacetylene Fibrils The change in Voc (us.Na) during the 48 h stand period after completion of each constant current reduction step was monitored periodically. Four typical curves showing the change in V,, with time are shown in fig. 3 for polyacetylene reduced to 1.7,4.0, 5.8 and 7.0 mol%. The increase in cell potential during the 48 h stand period is consistent with a decrease in the degree of reduction on the outside of the [(CH)Y-], fibrils as the counter Na+ ions diffuse toward the interior of the fibril together with their attendant negative charge on the polyacetylene. Exactly the opposite effect is observed after a partial electrochemical oxidation of the [(CH)Y-Iz to a less-reduced state. In this case, the Voc falls on standing as Na+ ions, which now have a greater concentration in the interior of the [(CH)Y-], fibrils, diffuse toward the surface of the (CH), fibrils.As discussed previously,' the diffusion constant, D, for the diffusion of ions within the (CH), fibrils and the time constant, z, can be obtained from the V,, us. time data given in fig. 3. Graphs of In [( - &)/( 5 - vf)] against time, t, for the essentially linearR. B. Kaner, S. J . Porter and A . G. MacDiarmid 2327 0.4 1 &-0- 0 0 0 0- 0-0 - - - - t parts of the Voc us. time curves in fig. 3 give a straight line with a slope of z-l. Here, 5 and V, are the initial and final open-circuit voltages, respectively, and V, is the open-circuit voltage at any time t. For 4.0 mol% reduced (CH),, a z value of 11.3 h was found, while for 7.0 mol% reduced (CH),, a z value of 14.9 h was obtained.Using = (2.405)2 D for diffusion into or out of a cylinder, the diffusion constant, D, can be calculated from the z values obtained.6 Assuming a radius of 100 A for the polyacetylene fibril^,^ the T value of 11.3 h, obtained for 4.0 mol% reduced (CH),, gives a D value of 4.2 x cm2 s-l and the z value of 14.9 h, obtained for 7.0 mol% reduced (CH),, gives a D value of 3.2 x 10-ls cm2 s-l. The values obtained for z and D are for intrafibrillar diffusion of Na+ ions from the outside to the inside of the 200 A diameter fibrils under a concentration gradient only. It should be stressed that these diffusion studies involve no external applied electric potential and hence involve no diffusion of ions within the electrolyte between the (CH), fibrils. These values may be compared to the corresponding z values obtained for the diffusion of Li+ ions in [Li&,3(CH)-0.03], (10.9 h) and in [Li,'.0,(CH)-o-06], (12.4 h).l It should be noted that the diffusion of the M+ ions discussed above involves only a redistribution of these ions within the (CH), fibrils and does not necessitate any migration of (PF,)- ions.The diffusion of NaS cations into or out of the polyacetylene fibrils will be 99% complete after a time equal to 4z. Since the V,, values used in the Voc us. percent reduction curve (fig. 1) were taken after 48 h, a time close to 42, these values may be considered to be equilibrium values. Coulombic Efficiency and Stability The two Na(NaPF,(THF)((CH), cells employed in the last section were also used to determine the reversibility of the reaction given by eqn (3).After each reduction of the polyacetylene followed by the 48 h stand period, the polyacetylene was reoxidized, first at a constant current until 2.0 V (us. Na) was reached and then at a constant applied potential of 2.0 V (us. Na) for 16 h to remove residual dopant ions. The total amount2328 0.5 - - 0 I I I I 1 1 I Electrochemical Reduction of Polyacetylene Y Y ( 6 ) 1 A n X- 2.0 t h-x-x-x - voc 1.0 1.51 time/ d a y s Fig. 4. V,, us. stand time, demonstrating the stability in potential (vs. Na) of (a) 7.0 mol% reduced polyacetylene and (b) neutral polyacetylene in an NalNaPF,(THF)I(CH), cell. of charge, Q (in, total) obtained in each oxidation process was recorded and was used in the relation [Q(out, total)/Q(in, total)] x 100 to calculate the Coulombic efficiency, Q,,,.Up to reduction levels of ca. 10 mol%, the Coulombic efficiency associated with each data point was ca. 100% (e.g. at 1.1 mol% reduction, Qeff = 101.6%, at 5.8 mol% reduction, Qeff = 100.4% and at 9.5 mol% reduction, Qeff = 100.8%). Since the reduction and reoxidation of polyacetylene is completely reversible up to ca. 10 mol%, one would expect the polymer to be stable on standing in this electrolyte. This is indeed the case, as is demonstrated in fig. 4. A 7.0 mol% reduced polyacetylene electrode in a Na) 1 .O dm3 mol-l NaPF6(THF)I[Na$~07(CH)-o.07], cell maintained a constant potential of 0.64 V (us. Na) during a 40-day period, as shown in the lower curve of fig.4. The slight rise in potential during the first two days is consistent with diffusion of the Na+ ions from the surface of the fibrils to their interior. After the 40-day stand period, oxidation to 2.0 V [neutral (CH),] gave a Coulombic efficiency of 100.2%. The cell potential then remained constant at 1.65 V for the following 40-day period, as shown in the upper curve of fig. 4. The slight decrease in potential during the first two days is consistent with diffusion of residual Na+ ions from the interior of the fibrils to their surface. These studies demonstrate the very great stability of polyacetylene in both its neutral and reduced forms in an appropriate electrolyte such as 1.0 dm3 mol-1 N aP F 6 (TH F) . In a further attempt to determine the maximum level of polyacetylene reduction which is stable in the NaPF,(THF) electrolyte, the following chemical reduction experiment was carried out.Two pieces of (CH),, one of which was encased in a Pt mesh, were placed in a 0.5 dm3 mol-1 sodium naphthalide solution. A spontaneous reaction took place as given by (CH), + xyNa+Nphth.- + [Na$(CH)y-], + xyNphth. (6) The reaction was allowed to proceed for 48 h to promote approximate diffusion equilibrium of the dopant ions within the polyacetylene fibrils. To evaluate the reduction level of the [Nai(CH)y-], produced, one sample was placed in methanol for 24 h to remove all the Na+ dopant ions in the form of sodium methoxide. The NaOMe-MeOH solution was then titrated with 0.1 dm3 mol-1 HC1. The results indicated a 17 mol% reduction level, i.e.the formation of [Na&7(CH)-0-17],. The second piece of washed and dried reduced polyacetylene was used an an electrode. Its potential us. Na, 0.246 V, wasR. B. Kaner, S . J . Porter and A . G. MacDiarmid 2329 0.5 time/min Fig. 5. Cell potential, V,, duringconstant-current discharge of polyacetylene reduced to 10.0 moly;, i.e. [Na~~,o(CH)-o.lo],, at (a) 0.135 mA, (b) 0.27 mA, (c) 0.54 mA and (d) 1.08 mA in an NalNaPF,(THF)I(CH), cell employing 2.7 mg (0.8 cm2) of (CH),. Table 1. Discharge characteristics of an NalNaPF,(THF)I[Na'(CH)Y-], cell discharge curvea. . constant discharge/mA applied current/A kg-l discharge time/h final cell potential/V overall Coulombic efficiency (%) average cell potential/V energy densityb/W h kg-l average power densityb/W kg-l energy efficiency (%) 0.135 50 ca.4.0 101.0 115.4 28.1 76.6 0.23 0.66 0.27 100 ca. 2.0 100.4 113.2 55.1 74.2 0.22 0.65 0.54 200 ca. 1.0 c 0.20 0.63 101.1 109.7 106.7 70.8 1.08 400 'a. 0.5 101.5 105.7 205.6 66.5 0.18 0.6 1 a Discharge curves (a), (b), (c) and (d) are given in fig. 5. employed and on the weight of Na consumed in the discharge reaction. Based only on the weight of (CH), measured immediately in the drybox after immersion in a 1.0 dm3 mol-l NaPF,(THF) electrolyte. This potential corresponds to an equilibrium reduction level of ca. 12 mol%, i.e. [Na:.12(CH)-0.12],, based on eqn (4). A sealed rectangular glass NalNaPF,(THF)I[Na$.l,(CH)-0.12], cell was then constructed using this same sample of reduced polyacetylene. By the time the cell construction was finished (ca.30 min), the cell potential was 0.4 V, corresponding to an equilibrium reduction level of ca. 10 mol% . This was confirmed by applying a constant potential of 2.0 V to the cell to oxidize the reduced polyacetylene to neutral (CH),. The amount of charge passed (1.85 C) corresponded to a reduction level of 10.3 mol%. The above results demonstrate that a higher level of reduction of polyacetylene can be attained in a sodium naphthalide-THF solution than in a NaPF,(THF) solution. In the presence of the latter solution the [Nat.,7(CH)-o.17], is oxidized rapidly to [Na$.lo(CH)-o-lo],, presumably with the concomitant reduction of the PF; ions. Hence, the stability of the reduced polyacetylene in the THF electrolyte is limited by the nature of the solute.A recent report* has shown that reduction levels in THF of ca. 18% can be obtained when alkali-metal alkyl borate salts, M+(BR,)-, in THF are used as the elec tr ol yte, Constant-current Discharge (Reduction) Studies The constant-current discharge characteristics of an NalNaPF,(THF)I(CH), cell em- ploying 2.7 mg (0.8 cm2) of (CH), are shown in fig. 5. In each of the four studies the (CH), was reduced to a 10.0 mol% reduction level. The results are given in table 1. Inte-23 30 Electrochemical Reduction of Polyacetylene grating the area, (V,Q), under a discharge curve and dividing by the charge (Q) invo!ved, gives the average cell potential during discharge. The ratio of the area under the discharge curve to that under the charge curve (not shown) gives a value for the energy efficiency, Eeff.After each constant-current discharge (reduction) the cell was charged to 2.0 V to oxidize the [Na~,,o(CH)-o.lO], to neutral (CH),. A constant current, identical to that used during reduction, was employed. When a potential of 2.0 V was reached, oxidation was completed at a constant potential of 2.0 V for 16 h. The constant-current oxidation step removed ca. 90% of the Na+ countercations and the constant potential step removed the remaining ca. 10%. The overall Coulombic efficiency in each case was close to 100% as given in table 1. A theoretical energy density for the reduction of polyacetylene to a 10.0 mol% level with Na+ countercations can be calculated from the area under the equilibrium curve in fig.1 . The value of 143.9 W h kg-’ obtained is based only on the weight of (CH), employed and the weight of sodium consumed in the reduction process. A comparison of this theoretical value to the experimental value of 115.4 W h kg-l obtained above during the 0.135 mA constant-current reduction shows that 80.2% of the theoretical energy capacity can be utilized at this ca. 4 h reduction rate. The polarization of the NalNaPF,(THF)J(CH), cell, i.e. change in voltage with increasing magnitude of the applied constant current, is remarkably low. The average cell potential during reduction changed from only 0.66 to 0.61 V when the current was increased eight-fold, i.e. from 0.135 to 1.08 mA. Similarly, the energy density decreased only very slightly from 115.4 to 105.7 W h kg-’.Electrochemical Reduction of Polyacetylene with the Incorporation of K + Countercations At the time the present studies were being performed, the electrochemical synthesis of [K,+(CH)y-], could not be carried out readily in THF since potassium salts such as KClO, are essentially insoluble in this solvent. This problem was overcome by the use of the crown ether complexing agent, dicyclohexano[ 181-crown-6, which rendered KClO, soluble in THF. However, recent studies have demonstrated that certain potassium tetra-alkyl borate salts are soluble in THF and can be used in the electrochemical reduction of (CH)2.8~ A KIKClO,(THF)I(CH), cell, employing 5.6 mg (1.6 cm2) of (CH), and ca. 0.5 cm3 ofan electrolyteconsisting of 1 .O dm3 mol-1 KClO, and 1 .O dm3 mol-l dicyclohexano[ 181- crown-6(THF), was constructed.The cell was discharged at a constant current of 0.056 mA [lo A kg-l of (CH),] for ca. 6 h to a (CH), reduction level of 3.2 mol%, i.e. [K,’~.,3,(CH)-o~032]x, based on the amount of charge passed. The current used corresponded to a reduction rate of ca. 0.5 molx h-l. The cell potential during discharge was recorded as a function of time and is given as curve ( d ) in fig. 6. A relatively small current was used, since at higher currents, e.g. at 100 A kg-’ (0.56 mA), the cell potential fell rapidly (below 0.5 V at ca. 1 mol% reduction), indicating a diffusion-limited reaction. This may be due to low conductivity of the electrolyte, slow diffusion of the large complexed K+ ions into the (CH), fibrils or slow dissociation of the K+ ion from the crown ether complex.This is in contrast to LiILiClO,(THF)I(CH), cells, Li~LiC10,-dicyclohexano[ 181- crown-6(THF)I(CH), cells and NalNaPF,(THF)I(CH), cells, all of which can be discharged at high current densities [at least 400 A kg-l of (CH),] with only a small decrease (< 0.1 V) in cell potential. Electrochemical Reduction of (CH), with the Incorporation of Organic Countercations If the electrolyte in which (CH), is reduced electrochemically contains an organic cation instead of an alkali-metal cation, then the organic cation will act as the ‘dopant’ cation.R. B. Kaner, S. J. Porter and A . G. MacDiarmid 233 1 1.5 In i I I I 1 I I I I I 1 0 I 2 3 4 5 6 7 8 9 10 average reduction (%) Fig. 6. Cell potential, V,, during reduction (discharge) as a function of the average percent reduction of polyacetylene in the following cel!s : (a) LilBu,NClO,(THF)I(CH), ; (b) LilLiClO,(THF)I(CH), ; (c) NalNaPF,(THF)I(CH), ; and ( d ) KIKClO,-dicyclohexano[ 1 81-crown- 6(THF)I(CH),.Cells (a)-(c) were reduced (discharged) at a rate of 100 A kg-l of (CH),; cell (d) was reduced at a rate of 10 A kg-'. For example, if tetra-n-butylammonium cations, (Bu,N)+, are present in a LilBu,NClO,(THF)!(CH), cell, then on electrochemical reduction of the polyacetylene, these countercations will be incorporated into the reduced (CH), to give [(Bu,N)t(CH)v-],, as given in eqn (8), assuming that diffusion of Li+ ions, liberated from the Li anode during reduction, to the (CII), cathode is prevented by means of, for example, a semipermeable membrane : (8) It should be noted that the oxidation of the reduced polyacetylene formed in this reaction is not the reverse of that given by eqn (8).Instead, as the [(Bu,N),+(CH)y-], is electrochemically oxidized back to neutral (CH),, unstable ' (Bu,N)O' is formed at the Li anode, as given by eqn (9): (9) (CH), + xy(Bu,N)+ + xye- + [(Bu,N),+(CH)v-I,. [(Bu,N);(CH)y-], + (CH), + xy' (Bu,N)O '. Spontaneous decomposition of the (Bu,N)O occurs to produce Bu,N, butane and but- 1 -ene.l0 Therefore, a cell such as Li(Bu,NClO,(THF)((CH),, employing organic cations, is not infinitely rechargeable per se. Instead, on each charge cycle some of the tetrabutylammonium ions are decomposed and replaced in solution by Li+ ions. As this cell is continually cycled the concentration of LiClO, in the THF increases. Eventually, on continued cycling, this cell will become equivalent to a LiILiClO,(THF)1(CH), cell.An Li(Bu,NClO,(THF)[(CH), cell, employing 4.9 mg (1.4 cm2) of (CH),, was con- structed. The cell was discharged (reduced) at a constant current of 0.49 mA [ 100 A kg- of (CH),], for ca. 2 h to a (CH), reduction level of ca. 10 molod , i.e. [ ( B U ~ N ) ~ ~ ~ ~ ( C H ) - ~ ~ ~ ~ The cell potential during reduction was recorded as a function of time and is given as curve (a) in fig. 6. The [(Bu,N)c(CH)u-], can be oxidized to give (CH), with ca. 100% Coulombic efficiency up to a reduction level of ca. 6.5 mol% if oxidation is carried out immediately after the reduction cycle. However, on standing in the electrolyte, the [(Bu,N)$(CH)y-], is somewhat unstable.For example, a [ ( B U ~ N ) ~ , , ~ , ( C H ) - ~ . ~ ~ ~ ] ~ elec- trode with a Voc, 24 h of 1.19 V (us. Li) increased to 1.25 V during a 10-day stand period. From the amount ofchargeinvolved in its oxidation back to neutral (CH), it was found that the reduction level had decreased to 3.8 mol%. The slight decrease in the reduction level may be due to reaction of the [(Bu,N)t(CH).v-], with degradation products formed from reducing (Bu,N)+ during the previous recycling studies. The reaction which produced [(Bu,N),+(CH)v-],, in an Li(Bu,NClO,(THF)I(CH), cell, can be generalized to incorporate other organic cations into (CM),. For example, if2332 Electrochemical Reduction of Polyacetylene tetraphenylphosphonium perchlorate, Ph,PC10,, is used in place of Bu,NC10,, [(Ph,P)t(CH)y-], is formed.ll The discharge curves of the Li(Bu,NClO,(THF)I(CH!, cell [fig.6(a)] and the K IKClO,(THF)I(CH), cell [fig. 6 (d)] may be compared with typical constant-current dis- charge curves to a 10.0 mol% (CH), reduction level for an LiJLiClO,QTHF)J(CH), cell’ and for an NaJNaPF,(THF)I(CH), cell at rates of 100 A kg-l [fig.6(b) and (c), respectively]. Note that the general shape of reduction curves (a), (b) and (c) is comparable. There is very little difference between the reduction curves of an LilBu,NClO,(THF)I(CH), cell [curve (a)] and an LilLiClO,(THF)I(CH), cell [curve (b)]. However, the reduction curve of an NalNaPF,(THF)I(CH), cell [curve (c)] possesses a larger ‘plateau’ region, extending to a (CH), reduction level of ca.5 mol%. The reduction curve of a K~KC10,-dicyclohexano[ 18]-crown-6(THF)I(CH), cell [curve (d)] exhibits a lower voltage, presumably owing to diffusion effects discussed previously. These data demonstrate that the kinetic properties of [Mi(CH)y-], electrodes vary with the countercation. The thermodynamic equilibrium potential of reduced polyacetylene, [(CH)g-],, as a function of y (y 6 0.10) is essentially independent of the nature of the countercation, M+, at least when M is Li or Na. Electrochemical properties of polyacetylene which depend on kinetic factors, such as cell potential during reduction are, however, dependent on the nature of countercation. This study was supported by the U.S. Department of Energy, contract no. DE-AC02-81-ER10832 and the S.E.R.C. References 1 R. B. Kaner and A. G. MacDiarmid, J . Chem. Soc., Faraday Trans. 1, 1984, 80, 2109. 2 H. Shirakawa and S. Ikeda, Polym. J., 1971,2, 231 ; H. Shirakawa, T. Ito and S. Ikeda, Die Makromoi. Chem., 1978, 179, 1565; H. Shirakawa, T. Ito and S. Ikeda, Polym. J., 1973,4, 460. 3 L. W. Shacklette, R. L. Elsenbaumer and R. H. Baughman, J. Phys. (Paris), 1983, 44, C3-559. 4 R. H. Baughman, N. S. Murthy and G. G. Miller, J. Chem. Phys., 1983, 79, 515. 5 R. H. Baughman, N. S. Murthy, G. G. Miller, L. W. Shacklette and R. M. Metzger, J . Phys. (Paris), 6 W. Jost, Diffusion (Academic Press, New York, 1960), p. 45. 7 A. G. MacDiarmid and A. J. Heeger, Synth. Met., 1979/80, 1, 101. 8 L. W. Shacklette, J. E. Toth, N. S. Murthy and R. H. Baughman, J . Electrochem. SOC., 1985,132, 1529. 9 T. R. Jow, L. W. Shacklette and M. Maxfield, The Electrochem. Soc. Extended Abstracts, Vol. 1984-2, 1983, 44, C3-53. No. 620, p. 902, New Orleans, Louisiana, Oct. 7-12 (1984). 10 J. E. Dubois, A. Monvernay and P. C. Lacaze, Electrochim. Acta, 1970, 15, 315. 11 D. MacInnes, Jr and A. G. MacDiarmid, 1981, unpublished results. Paper 5/1249; Receiued 22nd July, 1985