Hydrous manganese oxidecarbon nanotube composite electrodes for electrochemical capacitors
ORIGINAL PAPER
Y.K.Zhou?B.L.He?F.B.Zhang?H.L.Li
Hydrous manganese oxide/carbon nanotube composite electrodes for electrochemical capacitors
Received:28June2003/Accepted:29September2003/Published online:9January2004
óSpringer-Verlag2004
Abstract A novel type of composite electrode based on hydrous manganese oxide and a single-walled carbon nanotube has been prepared and used in electrochemical capacitors.Cyclic voltammetry,galvanostatic charging/ discharging tests and electrochemical impedance mea-surements were applied to investigate the performance of the composite electrodes with di?erent ratios of hy-drous manganese oxide and single-walled carbon nano-tube.For comparison,the performance of pure hydrous manganese oxide and pure carbon nanotubes was also studied.In this way,the composite electrode with a6:4 ratio of hydrous manganese oxide to carbon nanotube was found to be the most promising active material for an electrochemical capacitor,which shows both good capacitance and power characteristics.
Keywords Carbon nanotube?Composite electrode?Electrochemical capacitor?Hydrous manganese oxide Introduction
Electrochemical(EC)capacitors have attracted much attention owing to their high power capability and long cycle life.The EC capacitor can be used for power enhancement and cycle life improvement of primary power sources such as batteries and fuel cells.It has been demonstrated that the hybrid power source,which pla-ces the EC capacitor and battery side-by-side,can sig-ni?cantly improve the pulse power performance of batteries[1,2,3,4].The possible applications for hybrid power sources include space communications,digital cellular phones,and electric and hybrid vehicles.In or-der to achieve maximum performance in such applica-tions,EC capacitors with high power and energy densities are critical[4,5].
The amorphous hydrous ruthenium oxide(a-RuO2?n H2O)is known as the most promising material for high power and energy density EC capacitors be-cause of its high speci?c capacitance and excellent cyclability[6].However,the high cost of the material itself and the environmental unfriendliness of electro-lytes such as acidic media lessen its commercial attrac-tion.Recently,amorphous manganese oxide(a-MnO2?n H2O)has received attention as a new candidate for solving these problems[7,8].
As a-MnO2?n H2O shows high resistivity and the equivalent series resistance(ESR)of a-MnO2?n H2O electrode is very large,carbon should be added as a conducting agent to increase the electrical conductivity [9].It is well known that carbon nanotubes have obvious advantages over active carbon,for example,low resis-tivity,high stability,a highly accessibly surface area and a narrow distribution of pore sizes[10].Hence,in this work,we added carbon nanotubes to a-MnO2?n H2O to form a-MnO2?n H2O/carbon nanotube composite elec-trodes to observe their performance.In addition,the performance of a pure carbon nanotube electrode and the role of the carbon nanotubes in the composite elec-trode are discussed.
Experimental
a-MnO2?n H2O was synthesized by oxidation of aqueous MnSO4 with KMnO4solution at pH%10.6,as described previously by Shan et al.[11].Carbon nanotubes were prepared and puri?ed as de-scribed by Li et al.[12].TEM and HRTEM were conducted at 200kV using a JEM-2010(JEOL)instrument.
The composite electrode materials were made by mixing a-MnO2?n H2O and carbon nanotubes in a ball mill.The mixing ratio of a-MnO2?n H2O to carbon nanotubes was8:2,6:4,4:6and2:8. For comparison,pure a-MnO2?n H2O and carbon nanotubes were also used as active materials.In addition,composite electrodes composed of a-MnO2?n H2O and graphite with di?erent weight
J Solid State Eletrochem(2004)8:482–487
DOI10.1007/s10008-003-0468-7
Y.K.Zhou?B.L.He?F.B.Zhang?H.L.Li(&) College of Chemistry and Chemical Engineering, Lanzhou University,730000Lanzhou,
People’s Republic of China
E-mail:lihl@https://www.wendangku.net/doc/6d13881661.html,
Tel.:+86-931-8912517
Fax:+86-931-8912582
fractions (8:2,6:4,4:6and 2:8,respectively)were also prepared as comparison.
The electrode was formed by mixing 80wt%active material,15wt%acetylene black and 5wt%PTFE as binder before rolling into a thin sheet of uniform thickness.Pellets cut out of the sheet were pressed with a hand oil press onto a Ni-foam current collector (1·1cm 2).
The electrode performance was measured in a beaker-type electrochemical cell equipped with the working electrode,a plati-num counter electrode and a standard calomel reference electrode (SCE).The electrolyte was a 0.5M Na 2SO 4aqueous solution and the geometric surface area of the working electrode was 1cm 2.Cyclic voltammetry scans were recorded from )0.2to 0.8V at a scan rate of 5mV/s using a CHI660A electrochemical workstation (Covarda).Charge/discharge cycle tests were performed using a LAND Celltest (Wuhan,China)at di?erent constant current densities,with a cuto?voltage of )0.2to 0.8V.Impedance mea-surements were performed on a CHI660A electrochemical work-station (Covarda)in the frequency range from 100,000to 0.01Hz,and the a.c.modulation was controlled at 5mV.
Results and discussion
A TEM image of the carbon nanotubes used here is shown in Fig.1.From it the abundant rope-like nano-tube networks are observed,and the support material and other contaminants could hardly be found in the sample.Shown in the inset is a typical high-magni?ca-tion TEM image,from which high-quality individual single-walled nanotubes with a diameter of around 1.3nm is clearly observed.
Figure 2shows the cyclic voltammograms (CVs)of di?erent composite electrodes between )0.2and 0.8V versus SCE,taken at a sweep rate of 5mV/s.Curves a,b,c,d,e and f correspond to pure carbon nanotubes,8:2,6:4,4:6,2:8composite electrodes and pure a -MnO 2?n H 2O,respectively.All these curves show no peaks,which indicates that the electrode capacitor is
charged and discharged at a constant rate over the complete cycle.Moreover,these CVs show a mirror image with respect to the zero-current line and a rapid current response on voltage reversal at each end poten-tial,namely,the rectangular-like and symmetric I –E responses of ideal capacitive behavior are clearly ob-served.Figure 2shows,under such conditions,that the carbon nanotube electrode has minimal speci?c dis-charge and charge capacitances,and the speci?c capac-itance increases as the a -MnO 2?n H 2O content increases.This is because the carbon nanotubes have only electric double layer capacitance,while a larger faradaic pseudocapacitance occurs for a -MnO 2?n H 2O.The com-posite electrode d shows the largest speci?c capacitance,which is even higher than that of f and e.This may due to the added carbon nanotubes reducing the electronic resistivity of the electrode and increasing the overall electrode porosity.With the help of the added carbon nanotubes,the accessible speci?c capacitance increases as the available active material increases and therefore the speci?c capacitance increases rapidly.The composite d shows the appropriate ratio of a -MnO 2?n H 2O to car-bon nanotubes for which the speci?c capacitance reaches the maximum.However,accurate capacitance values should be obtained from the constant charge/discharge tests.
Figure 3shows the constant current charge/discharge curves for di?erent working electrodes.The cuto?voltages for charging/discharging were )0.2and 0.8V with a 5mA/cm 2current density.For all these curves,a good linear variation of potential versus time was ob-served,which is another typical characteristic of ideal capacitor behavior.The measured speci?c discharge capacitance values for a,b,c,d,e and f are 38.0,65.0,116.1,160.7,141.5and 127.2F/g,respectively.
The
Fig.1TEM image of carbon nanotubes (shown in the inset is a typical HRTEM image)
483
carbon nanotubes initially have a minimal capacitance
value of 38.0F/g;then,as faradaic pseudocapacitance occurs,the capacitance increase as the a -MnO 2?n H 2O content increases.Interestingly,the composite electrode d shows the largest capacitance,higher than the pure a -MnO 2?n H 2O and the others.Here the result is agreement with that of the CV tests,and the role of the added carbon nanotubes is proved.
To investigate the performance of the composite electrode,all of these electrodes were charged/dis-charged at di?erent current densities.Figure 4shows the speci?c capacitance as a function of charge/discharge current density for all these electrodes.The special capacitance was obtained from Eq.1:C ?
I D t D V
e1T
The pure carbon nanotubes show almost unchanged speci?c capacitance of about 40F/g when the current density is altered.This is because the carbon nanotubes exhibit almost only electric double layer capacitance,and have low resistivity and a highly accessible surface
area;also the polarization is very small and the revers-ibility of the electrode is very good when the current density increases.The contrary case is f,the pure a -MnO 2?n H 2O electrode.This shows the largest speci?c capacitance,385.4F/g when the current density is 1mA/cm 2,but as the current density increases the spe-ci?c capacitance decreases rapidly,owing to the high resistivity and the strong polarization of this electrode.For the other composite electrodes,as the a -MnO 2?n-H 2O content alters,the speci?c capacitance decreases to some extent as the current density increases.The com-posite electrode d shows a larger speci?c capacitance even when the current density is bigger.This indicates that d has the best performance with the proper ratio of carbon nanotubes to a -MnO 2?n H 2O.
The power characteristics of di?erent kinds of com-posite electrodes and pure electrodes were also investi-gated.Figure5shows the normalized capacitance of di?erent electrode materials.Normalized capacitance was obtained using Eq.2:
Normalized capacitance?
C
Ce1mA=cm2T
e2T
The carbon nanotubes showed the best power char-acteristics between1mA/cm2and10mA/cm2current density,while the a-MnO2?n H2O showed the worst. From Fig.3it is shown that the speci?c power can be improved by adding carbon nanotubes to a-MnO2?n H2O to form a composite electrode.To investigate the e?ect of the additive type,a-MnO2?n H2O/graphite composite electrodes were also prepared and tested.The speci?c capacitance values of di?erent kinds of a-MnO2?n H2O/ carbon nanotube and a-MnO2?nH2O/graphite compos-ites are summarized in Table1.The speci?c capacitances for the a-MnO2?n H2O/carbon nanotube composite electrodes are consistently higher than those of the corresponding ones with the same weight fraction of additive.The results especially show that a-MnO2?n H2O/ carbon nanotube composites containing lower amounts of a-MnO2?n H2O exceed the performance of a-MnO2?n H2O/graphite composites with larger amounts, i.e.the speci?c capacitance of the former with a ratio of 4:6is higher than the latter with8:2.This means that the active sites of a-MnO2?n H2O may be increased with less carbon nanotube than graphite and therefore the speci?c capacitance is greatly increased.At higher rates,this point becomes more obvious.It implies that the elec-tronically conducting network and the nature of the contact between the conductive component and the MnO2phase may be an important consideration for achieving higher speci?c capability;from this point of view,carbon nanotubes are much superior to graphite for use as a conductive component.In fact,the carbon nanotubes have a particular hollow tube structure with low resistivity and highly accessible surface area com-pared to graphite,the proper addition of which may enormously improve the contact between the a-MnO2?n H2O particles,and the surface of the a-MnO2?n H2O particles far from the current collector can work as an active site for a faradiac reaction rapidly and therefore a better performance may be obtained.
Electrochemical impedance spectroscopy,as a pow-erful technique for the investigation of the capacitive behavior of electrochemical cells,has been also used to check the ability of the composite materials to store electrical energy.Figure6presents Nyquist plots of the composite electrodes and pure electrodes.All the impedance spectroscopy is almost similar in form, composed of one semicircle followed by a linear part at the low-frequency end.At very high frequencies,the measured resistance is composed of the following terms: the ionic resistance of electrolyte,the intrinsic resistance of the active material,and the contact resistance at the active material/current collector interface.We can see that although the intrinsic resistance of the material is di?erent,the very high frequency resistance is roughly the same,about2.2W.At high frequencies,Fig.6shows the presence of a semicircle,which has been described as a pseudo-charge transfer resistance and is associated with the porous structure of the electrode[13].The semicircles a and b are almost the same,and c and d only increase a little;however,for e and f,there is a rapid increase.For double layer charging only,semicircles in the high-frequency range of the impedance of porous electrodes were also discussed by the presence of oc-cluded pores[14].Here we can see that carbon nanotu-bes and the carbon nanotube composite electrodes contain smaller occluded pores in the electrode and therefore smaller pseudo-charge transfer resistance; when the a-MnO2?n H2O content further increases,the occluded pore amount and pseudo-charge transfer resistance increase rapidly.This may due to the added carbon nanotubes having a low resistance;also it may change the porous structure of the electrode and facili-tate the charge transfer.
Table1The speci?c capacitance values(F/g)for di?erent kinds of a-MnO2?n H2O/carbon nanotube and a-MnO2?n H2O/graphite composites
Current density(mA/cm2)125810 Pure MnO2385.46178.56127.28109.98102.06 MnO2–SWNT(8:2)279.18174.96141.59120.78115.92 MnO2–SWNT(6:4)219.57189.89160.76142.60134.63 MnO2–SWNT(4:6)147.42130.01116.15112.19109.31 MnO2–SWNT(2:8)76.3969.1365.0763.4362.43 Pure SWNT41.4538.1637.6236.3634.92 MnO2–graphite(8:2)100.7290.4579.2568.7664.55 MnO2–graphite(6:4)97.5773.2860.3651.5249.55 MnO2–graphite(4:6)80.6355.7849.5141.7240.49 MnO2–graphite(2:8)49.4240.3634.9131.2629.45 Pure graphite 1.59 1.25 1.24 1.2
0.68
At low frequencies,the impedance plot should theo-retically be a vertical line,which is parallel to the imaginary axis.The carbon nanotubes show almost ideal
capacitive behavior;however,the others show some
departure from that expected,which can also be ex-plained by the electrode surface inhomogeneity and a ‘‘constant phase element’’occurs[15],which results in the straight-line behavior of the Nyquist impedance with a slope angle smaller than p/2for these electrodes,with the exception of the pure carbon nanotubes.This means that the porous electrodes have di?erent pores with di?erent dimensions and for the composite electrodes there exists some inhomogeneity in the surface,which leads to the low-frequency impedance behavior being https://www.wendangku.net/doc/6d13881661.html,pared to the total impedance plots,we think that the d composite electrode has a smaller resistance and better capacitive behavior;it should be the most promising for use in electrochemical capacitors. Conclusions
To achieve both high speci?c capacitance and high power simultaneously,adding a proper proportion of carbon nanotubes to hydrous manganese oxide to form a composite electrode may be an e?ective method. Comparing di?erent ratios,the composite of6:4of a-MnO2?n H2O to carbon nanotubes was the most suitable for electrode materials in electrochemical capacitors. The reason may be due to the added carbon nanotubes lowering the electric resistivity and improving the contact between the a-MnO2?n H2O particles,and the surface of the a-MnO2?n H2O particles can work as an active site for rapid faradiac reactions.References
1.Burke AF(1994)In:Nedell R(ed)Proceedings of the36th
power sources conference,Cherry Hill,NJ.IEEE,Piscataway, NJ,p6
https://www.wendangku.net/doc/6d13881661.html,ler JR(1995)In:Proceedings of the5th international sem-
inar on double layer capacitors and similar energy storage de-vices,Florida Educational Seminars Inc,Boca Raton,Fla,Dec 4–6
3.Isaacson MJ,Kraemer BJ,Laramore TJ,Mayer ST,Josephs
LC(1996)Proceedings of the37th power sources conference, Cherry Hill,NJ.IEEE,Piscataway,p112
4.Jow TR,Zheng JP,Ding SP(1997)Proceedings of the7th
international seminar on double layer capacitors and similar energy storage devices,Florida Educational Seminars Inc, Deer?eld Beach,Fla,Dec8–10
5.Zheng JP(1999)Electrochem Solid-State Lett2:359
6.Zheng JP,Jow TR(1995)J Electrochem Soc142:L6
7.Zheng JP,Cygan PJ,Jow TR(1995)J Electrochem Soc
142:2699
8.Lee HY,Goodenough JB(1999)J Solid State Chem144:220
9.Lee HY,Kim SW,Lee HY(2001)Electrochem Solid State Lett
4:A19
10.Niu CM,Sichel EK,Hoch R,Moy D,Tennent H(1997)Appl
Phys Lett70:1480
11.Shan X,Dong GJ,Jing XY,Zhang ML(2001)Chin J Inorg
Chem17:669
12.Li XH,Zhang J,Li QW,Li HL,Liu ZF(2003)Carbon41:598
13.Gamby J,Taberna PL,Simon P,Fauvarque JF,Chesneau M
(2001)J Power Sources101:109
14.Keiser H,Beccu K,Gutjahr MA,(1976)Electrochim Acta
21:539
15.Brug GJ,Van den Eeden ALG,Sluyters-Rehbach M,Sluyters
JH(1984)J Electroanal Chem176:275
487