REPORTS

compatibility with organic electrolytes, and short

cycle life have limited the practical application
Carbon-Based Supercapacitors of these pseudo-capacitive materials. Carbon
nanotubes (CNTs), especially single-walled CNTs
Produced by Activation of Graphene (SWNTs), have an ideal limit SSA of 1300 m2/g
(10), can have high electrical conductance along
the tubes, and demonstrate good performance in
Yanwu Zhu,1 Shanthi Murali,1 Meryl D. Stoller,1 K. J. Ganesh,1 Weiwei Cai,1 Paulo J. Ferreira,1 organic electrolyte (11). However, the high cost
Adam Pirkle,2 Robert M. Wallace,2 Katie A. Cychosz,3 Matthias Thommes,3 Dong Su,4 for mass production of high-quality SWNTs is a
Eric A. Stach,4 Rodney S. Ruoff1* challenge for the commercialization of SWNT-
based supercapacitors.
Supercapacitors, also called ultracapacitors or electrochemical capacitors, store electrical charge on Graphene has a theoretical SSA of 2630 m2/g
high-surface-area conducting materials. Their widespread use is limited by their low energy storage and a very high intrinsic electrical conductivity in
density and relatively high effective series resistance. Using chemical activation of exfoliated plane as well as high mechanical strength and
graphite oxide, we synthesized a porous carbon with a Brunauer-Emmett-Teller surface area of up chemical stability (12). Graphene-based material
to 3100 square meters per gram, a high electrical conductivity, and a low oxygen and hydrogen derived from graphite oxide (GO) is now being
content. This sp2-bonded carbon has a continuous three-dimensional network of highly curved, manufactured in ton quantities at low cost (13).
atom-thick walls that form primarily 0.6- to 5-nanometer-width pores. Two-electrode We have previously demonstrated supercapaci-
supercapacitor cells constructed with this carbon yielded high values of gravimetric capacitance tors that were based on reduced graphene oxide
and energy density with organic and ionic liquid electrolytes. The processes used to make this with capacitance values of approximately 130 and

Downloaded from http://science.sciencemag.org/ on November 22, 2017

carbon are readily scalable to industrial levels. 100 F/g in aqueous KOH and organic electro-
lytes, respectively (14). With a low equivalent
upercapacitors store energy by forming a capacitance via pseudo-capacitive redox reactions. series resistance (ESR), the supercapacitor per-

S double layer of electrolyte ions on the sur-

face of conductive electrodes. Superca-
pacitors are not limited by the electrochemical
Although capacitances of up to 1300 F/g (such as
with MnO2) have been reported in aqueous elec-
trolytes (9), the low electrical conductance, poor
formance did not show much degradation with
an increase in the scan rate. Various graphene-
based materials derived from GO have reported
charge transfer kinetics of batteries and thus can
operate at very high charge and discharge rates
and can have lifetimes of over a million cycles
(1). However, the energy stored in supercapaci-
tors is currently an order of magnitude lower than
that of batteries, which limits their adoption to
those applications that require high cycle life
and power density. The energy density of exist-
ing state-of-the-art supercapacitor devices, which
are mainly based on porous activated carbon (AC),
is about 4 to 5 watt-hour (Wh)/kg, whereas that
of lead acid batteries is in the range of 26 to
34 Wh/kg (2). A typical AC material, with a
Brunauer-Emmett-Teller (BET) specific surface
area (SSA) in the range of 1000 to 2000 m2/g and
a pore size distribution in the range of 2 to 5 nm,
has a gravimetric capacitance of 100 to 120 F/g in
organic electrolytes (3). Research has thus been
focused on increasing energy density without sac-
rificing cycle life or high power density (4). An
increased capacitance in the organic electrolyte
tetraethylammonium tetrafluoroborate (TEA BF4)
in acetonitrile (AN) by using carbide-derived car-
bons (CDCs) has been reported (5). Metal ox-
ides such as RuO2 or MnO2 (6), MoO3 (7), and
electronically conducting polymers (8), or their
composites, have been used to increase specific
Fig. 1. (A) Schematic showing the microwave exfoliation/reduction of GO and the following chemical
activation of MEGO with KOH that creates pores while retaining high electrical conductivity. (B) Low-
1
Department of Mechanical Engineering and Materials magnification SEM image of a 3D a-MEGO piece. (C) High-resolution SEM image of a different sample
Science and Engineering Program, University of Texas at region that demonstrates the porous morphology. (D) ADF-STEM image of the same area as (C), acquired
Austin, One University Station C2200, Austin, TX 78712, USA.
2
simultaneously. As seen, a-MEGO contains micro- and mesopores with a distribution of sizes between ~1
Department of Materials Science and Engineering, University and ~10 nm. (E) High-resolution phase contrast electron micrograph of the thin edge of an a-MEGO
of Texas at Dallas, 800 West Campbell Road, Richardson, TX
chunk, taken at 80 kV. There is a variation in focus across the image because of the sloped nature of the
75080, USA. 3Quantachrome Instruments, 1900 Corporate
Drive, Boynton Beach, FL 33426, USA. 4Center for Functional sample and changes in sample thickness. The image shows the presence of a dense network of nanometer-
Nanomaterials, Brookhaven National Laboratory, Upton, NY scale pores surrounded by highly curved, predominantly single-layer carbon. (F) Exit wave reconstructed
11973, USA. HR-TEM image from the edge of a-MEGO. The in-plane carbon atoms are clearly resolved, and a variety of
*To whom correspondence should be addressed. E-mail: n-membered carbon rings can be seen. Substantial curvature of the single-carbon sheets is visible, with
r.ruoff@mail.utexas.edu the in-plane crystallinity being preserved.

www.sciencemag.org SCIENCE VOL 332 24 JUNE 2011 1537

 REPORTS
high-end capacitance values of ~200 F/g in aque- mission electron microscopy (ADF-STEM) (Fig. Characterization of a sample of a-MEGO by
ous electrolytes (15, 16), ~120 F/g in organic elec- 1D), and high-resolution TEM (HR-TEM) (Fig. means of synchrotron powder x-ray diffraction
trolytes (16, 17), and ~75 F/g in an ionic liquid 1, E and F) images of the microstructure. These (XRD) and x-ray photoelectron spectroscopy
(18). Recently, supercapacitors using oriented images [and additional images (fig. S2 and movie (XPS) is shown in Fig. 2, A and B. Compar-
graphene grown on nickel by means of chemical S1)] clearly indicate that the activation process ison with MEGO indicates that the (002) peak
vapor deposition were reported (19) that demon- etches the MEGO and has generated a three- of a-MEGO has a markedly reduced intensity
strated efficient filtering of 120 Hz current with a dimensional (3D) distribution of what are referred and is dramatically broadened. These results are
resistance capacitance (RC) time constant of less to as meso- and micropores in the porous mate- consistent with the observations from HR-TEM,
than 0.2 ms, but at the cost of effective energy rials literature. The activation with KOH yields which indicate that a-MEGO is composed of pre-
storage because of the very low density of the a continuous 3D network of pores of extremely dominantly single-carbon sheets: Thus, a strong
electrode material. small size, ranging from ~1 nm to ~10 nm. Thus, it decrease in the (002) peak would be expected.
To date, the reported SSA values of carbon appears that the chemical activation is not merely A large increase in the low-angle scatter from
materials derived from GO have been well below digesting the MEGO but also dramatically re- a-MEGO versus MEGO is also noted, which is
2630 m2/g. Here, we report a simple activation structuring it. (The 3D nature of these very small consistent with the presence of a high density of
with KOH of microwave exfoliated GO (MEGO) pores makes a statistically accurate quantitative pores. In the XPS C1s spectrum of MEGO shown
and thermally exfoliated GO (TEGO) to achieve analysis with EM of the distribution of pore sizes in Fig. 2B, the tail between 286 and 290 eV is due
SSA values up to 3100 m2/g. As described in difficult because not all pores are visible in a to C-O groups and energy loss “shake-up” fea-
(20), we prepared MEGO powders by irradiating given image.) The spherical aberration–corrected tures (26). These oxygen-containing groups were
GO in a microwave oven. The as-made MEGO HR-TEM image presented as Fig. 1E (taken at strongly suppressed after activation, with two new
powder was then placed in KOH solution, fol- 80 kV so as to prevent electron beam damage) peaks appearing between 292 and 296 eV in the
lowed by filtration and drying, to form a series of further corroborates a dense pore structure with a a-MEGO sample that were assigned as K2p peaks.

Downloaded from http://science.sciencemag.org/ on November 22, 2017

MEGO/KOH mixtures for chemical activation. continuous three-dimensional network of highly The K2p peaks (<2 atomic % as determined from
Each MEGO/KOH mixture was put in a tube curved, predominantly atom-thick walls. An exit- XPS) in a-MEGO are due to potassium residue,
furnace under flowing argon at a pressure of about wave reconstructed image (Fig. 1F) taken with primarily as K2CO3 with a small amount of KOH.
400 torr and heated at 800°C for 1 hour. A sche- the Transmission Electron Aberration-Corrected Quantification of the amount of sp2-bonding can
matic of this activation process is shown in Fig. Microscope (TEAM) instrument (spherical and be determined by measuring the ratio between p*
1A. Chemical activation has been extensively chromatic aberration correction, at 80 kV) clear- bonding and p* + d* bonding by using electron
used to obtain porous ACs (21). KOH activation ly resolves the individual carbon atoms in the energy loss spectroscopy (EELS) (27). A com-
has been used on CNTs (22), carbon nanofibers structure. This image shows that a-MEGO is parison of the carbon K near-edge structure for
(23), and polyacrylonitrile-carbon nanotube com- composed of n-membered rings in plane, with n a-MEGO and graphite of equivalent thickness is
posites (24), and improved porosity and enhanced varying between 5 and 8. Additionally, it is clear presented in Fig. 2C. With the assumption that
supercapacitor performance were reported. It is that even as the walls bend through high de- the sp2 bonding in the graphite reference spectra
suggested that the activation of carbon with KOH grees of curvature the in-plane crystallinity is is 100%, the a-MEGO was found to have 98%
(25) proceeds as 6KOH + C ↔ 2K + 3H2 + preserved. (T2%) sp2 bonding. Complementary measurements
2K2CO3, followed by decomposition of K2CO3 Electron paramagnetic resonance (EPR) data were also made by XPS, reaching similar con-
and/or reaction of K/K2CO3/CO2 with carbon. (fig. S3) indicate an unpaired spin count at the clusions (further details of the fitting procedures
The activation with KOH generated nano- parts-per-million level. This—in hand with the can be found in fig. S4). Micro Raman spectros-
scale pores in the product carbon. The SSA of the very low H and O content in the a-MEGO indi- copy (fig. S5A) and Fourier transform infrared
activated MEGO (a-MEGO) could be readily cated by elemental analysis—shows that a-MEGO spectroscopy (fig. S5B) are provided for com-
controlled by the ratio of KOH versus MEGO (fig. has a small fraction of edge atoms. (If a large pleteness but did not supply additional insights
S1). The scanning electron microscopy (SEM) im- fraction of the C were edge atoms, they would about a-MEGO beyond the other methods of
age in Fig. 1B shows the morphology of a typical either be functionalized or would be present as analysis discussed here.
a-MEGO piece. Figure 1 shows high-resolution “dangling bonds” that would register an EPR State-of-the-art surface and pore-size char-
SEM (Fig. 1C), annular dark field scanning trans- signal.) acterization of the a-MEGO was performed by

Fig. 2. Characterization of the a-MEGO material (SSA ~ 2520 m2/g) with MEGO reduction of the (002) in a-MEGO indicates that there are essentially no inter-
or graphite as a control. (A) Synchrotron powder XRD pattern (plotted as Cu Ka). layer correlations of the carbon sheets. (B) XPS C1s spectra, with K2p region
Full width at half maximum of the (100) peak is ~2 degrees for both samples, in the inset. (C) EELS spectra from a-MEGO and graphite. Quantification of the
indicating average in-plane crystal size of ~5 nm with the Scherrer equation; the near-edge structure indicates that the a-MEGO has 98% (T2%) sp2 bonding.

1538 24 JUNE 2011 VOL 332 SCIENCE www.sciencemag.org

 REPORTS
coupling high-resolution nitrogen (77.4 K) and sumption also with regard to the SEM/STEM/TEM to 5 mHz yields the Nyquist plot shown in Fig.
argon (87.3 K) adsorption/desorption experi- results. The obtained pore-size/volume distribu- 4C. The plot features a vertical curve, indicating a
ments with advanced methods based on density tion indicates that this carbon sample is distinctive nearly ideal capacitive behavior of the cell. From
functional theory (DFT) (28). In addition, CO2 because of the existence of well-defined micro- the magnified data in the high-frequency range
adsorption at 273.2 K has been performed to as- and mesopores. The presence of ultramicropores (Fig. 4C, inset), a transition between the RC semi-
sess the ultramicropores (pores of width <1 nm). is seen from the CO2 data, and the analysis of the circle and the migration of electrolyte was observed
These isotherms, as shown in Fig. 3A, reveal the nitrogen adsorption data reveals the presence of at a frequency of about 382 Hz, corresponding to
details of the low-pressure region in which mi- micropores in the ~1-nm size range as well as a resistance of 2.45 ohms. The diffusion of elec-
cropore filling occurs, as well as the linear plot of narrow mesopores centered around 4 nm in size. trolyte ions stopped at about 3 Hz, and thereafter
the argon and nitrogen isotherms that reveal pore The latter is in good agreement with the bimodal the whole capacitance was reached (33). The volt-
condensation and type H2 hysteresis [according distribution of pore sizes observed with high- age drop at the beginning of discharge curves was
to the International Union of Pure and Applied resolution EM images. A quenched solid DFT used to estimate the internal resistance. An ESR
Chemistry (IUPAC) classification] (29) for the (QSDFT), which quantitatively accounts for the of 3.2 ohms was obtained from a-MEGO in the
a-MEGO sample, which is indicative of an in- surface roughness (31), also has been used to ob- BMIM BF4/AN electrolyte. Based on a series RC
terconnected pore system exhibiting constrictions tain the pore-size distribution of a-MEGO from model, the capacitance from the FRA data as a
(29, 30). For comparison, nitrogen adsorption the nitrogen and argon adsorption data, as shown function of frequency is shown in Fig. 4D. The
on a MEGO control sample is also shown in fig. in fig. S7. The pore size distribution curves ob- capacitance decreases sharply at about 4 Hz and
S6, and it is evident that the pore volume of tained from argon and nitrogen agree very well. remains 0.035 F at 10 Hz. Supercapacitor per-
MEGO is much smaller than that of a-MEGO; Using best-practice methods for determining formance of the a-MEGO (SSA ~ 3100 m2/g) in
the small overall pore volume should be caused an electrode material’s performance for superca- the TEA BF4/AN electrolyte was also measured
by the platelet-like structure in MEGO—likely just pacitors (32), we constructed and measured the and yielded a specific capacitance of above

Downloaded from http://science.sciencemag.org/ on November 22, 2017

the gap between platelets (20). In contrast, the pores performance of two-electrode symmetrical super- 150 F/g as obtained from the discharge curve
in a-MEGO have a well-defined micro-mesopore capacitor cells on the basis of a-MEGO (SSA ~ with a constant current of 0.8 A/g, with an ESR
size distribution as shown in Fig. 3B, with a 2400 m2/g) and 1-butyl-3-methyl-imidazolium of 4.6 ohms (fig. S8). One other carbon (KOH-
huge increase in pore volume (up to 2.14 cm3/g) tetrafluoroborate (BMIM BF4)/AN electrolyte, activated CDCs) using the same AN-based elec-
relative to MEGO. The a-MEGO had a high ni- as shown in Fig. 4. The cyclic voltammetry (CV) trolyte with comparable specific capacitance
trogen BET SSA of ~3100 m2/g (calculated in testing (Fig. 4A) shows rectangular curves from values has been reported (34). a-MEGO has the
the linear relative pressure range from 0.1 to 0.3). 0 to 3.5 V over a wide range of voltage scan highest gravimetric capacitance in organic elec-
Within this context, in a strict sense the BET rates. The galvanostatic charge/discharge curves trolyte reported to date for any carbons derived
method is not applicable to microporous solids, at three current densities are shown in Fig. 4B. The from graphene-based materials.
and hence the obtained BET surface area should specific capacitance was calculated from the dis- Values for energy and power density were esti-
be considered as an apparent or equivalent area charge curves with values of 165, 166, and 166 F/g mated on the basis of the supercapacitor mea-
only. Figure 3B displays the results of the cu- obtained at current densities of 1.4, 2.8, and 5.7 A/g, surements in the BMIM BF4/AN electrolyte. Using
mulative pore volume and pore size analysis from respectively. The corresponding volumetric ca- the specific capacitance value of 166 F/g (from the
CO2 and nitrogen adsorption by applying a hybrid pacitance is ~60 F/cm3. The voltage drop at the discharge curve with a constant current of 5.7 A/g)
nonlocal DFT (NLDFT) kernel, assuming a slit initiation of the discharge is 0.034 V (for the cur- and working voltage of 3.5 V, the energy density is
pore geometry for the micropores and a cylindrical- rent density of 1.4 A/g), suggesting a very low ~70 Wh/kg for the a-MEGO in the cell. Based on
pore geometry for the mesopores, which appears ESR in the test cell. A frequency response anal- a weight ratio of 30% for the active electrode mate-
(although oversimplified) to be a reasonable as- ysis (FRA) of the frequency range from 500 kHz rial in a packaged supercapacitor device—typical

Fig. 3. Gas adsorption/desorption analysis of an a-MEGO sample (SSA ~ volume and (inset) pore-size distribution for N2 (calculated by using a
3100 m2/g). (A) High-resolution, low-pressure N2 (77.4 K) and Ar (87.3 K) slit/cylindrical NLDFT model) and CO2 (calculated by using a slit pore NLDFT
isotherms. (Inset) The CO2 (273.2 K) isotherm. (B) Cumulative pore model).

www.sciencemag.org SCIENCE VOL 332 24 JUNE 2011 1539

 REPORTS
for large-scale AC-based supercapacitors—a imidazolium bis(trifluormethylsulfonyl)imide almost entirely of single sheets of sp2-bonded
practical energy density of above 20 Wh/kg for (EMIM TFSI) as electrolyte (fig. S10), the mea- carbon. This suggests that a large fraction of
a packaged device is expected. This is four times sured gravimetric capacitance of a-MEGO (SSA ~ “negative curvature carbon” (37–39) could be
higher than existing AC-based supercapacitors, 3100 m2/g) at 3.5 V and a current density of present.
two times higher than that reported for carbon- 0.7 A/g is 200 F/g, with an ESR of 8.6 ohms. The excellent performance obtained for var-
oxide hybrid electrochemical devices (2), and However, the curves in fig. S10 are not as ideal as ious electrolytes opens the possibility to engineer
nearly equal to the energy density of lead acid bat- those from a-MEGO in either BMIM BF4/AN supercapacitor electrodes based on this form of
teries. At the same current density (5.7 A/g), the or TEA BF4/AN electrolyte. Other carbons (such carbon in order to target a wide range of applica-
power density is also very high at ~250 kW/kg, as CDCs) using the EMIM TSFI electrolyte have tions, such as high energy, high power, or low
as estimated by using the voltage drop and ESR been reported with comparable performance val- cost. Unlike other carbon materials, no special
obtained from the discharge curve. For a pack- ues, although the measurements were performed substrates or transfer procedures are required for
aged cell, the power density of ~75 kW/kg is one at an elevated temperature (60°C) (36). synthesis. For supercapacitor manufacturing, this
order higher than the values from commercial The high powder conductivity of ~500 S/m, a material can be treated the same as current com-
carbon supercapacitors that have energy den- C/O atomic ratio of up to ~35, the very low H con- mercial ACs. Electrodes used for testing were of
sity values of 4 to 5 Wh/kg (2). This material is tent, and the essential absence of dangling bonds the same thickness used in commercial cells, and
also very stable. After 10,000 constant current in the a-MEGO suggest that it has a high content testing was performed using commercial collec-
charge/discharge cycles at a current density of of sp2-bonded carbon and very few edge atoms tors, separators, binders, and electrolytes. As pre-
2.5 A/g in neat BMIM BF4 electrolyte (fig. S9), for the samples with SSA of above 2500 m2/g. viously stated, the processes used to synthesize
97% of its capacitance was retained. We reasoned This along with the SEM, TEM, STEM, EELS, this carbon electrode material are readily scal-
that this carbon might perform even better with EPR, XPS, XRD, and adsorption isotherm data able to industrial levels. For example, we have
smaller-diameter ions (35). With neat ethyl-methyl- thus support a highly porous carbon comprised demonstrated that this simple activation process

Downloaded from http://science.sciencemag.org/ on November 22, 2017

Fig. 4. Supercapacitor performance of a-MEGO (SSA ~ 2400 m2/g) in the pedance. (Inset) The data at high-frequency ranges, with frequency values
BMIM BF4/AN electrolyte. (A) CV curves for different scan rates. Rectangular corresponding to the transition of the curves marked. (D) Frequency re-
shapes indicate the capacitive behavior. (B) Galvanostatic charge/discharge sponse of the gravimetric capacitance of the a-MEGO supercapacitor. Ca-
curves of a-MEGO–based supercapacitor under different constant currents. pacitances of 35 and ~8.8 mF remain at the frequencies of 10 and 100 Hz,
(C) Nyquist plot, showing the imaginary part versus the real part of im- respectively.

1540 24 JUNE 2011 VOL 332 SCIENCE www.sciencemag.org

 REPORTS
also can be applied to TEGO (figs. S11 and S12), 18. S. R. C. Vivekchand, C. S. Rout, K. S. Subrahmanyam, 38. H. Terrones, A. L. Mackay, Prog. Cryst. Growth Charact.
which is already being manufactured in ton quan- A. Govindaraj, C. N. R. Rao, J. Chem. Sci. 120, 9 (2008). Mater. 34, 25 (1997).
19. J. R. Miller, R. A. Outlaw, B. C. Holloway, Science 329, 39. E. Barborini et al., Appl. Phys. Lett. 81, 3359 (2002).
tities (13). By use of this type of simple activation 1637 (2010). Acknowledgments: We appreciate funding support from
process already commercially demonstrated for 20. Y. Zhu et al., Carbon 48, 2118 (2010). NSF under award DMR-0907324, the U.S. Department
ACs, scaled a-MEGO and a-TEGO production 21. H. Marsh, F. Rodriguez-Reinoso, Activated Carbon. of Energy (DOE) under award DE-SC001951, and the
for advanced energy/power electrochemical elec- (Elsevier, London, 2006). Institute for Advanced Technology. The research by
22. E. Raymundo-Pinero et al., Carbon 43, 786 (2005). E.A. and D.S. has been carried out at the Center for
trical energy storage devices may be realized in a 23. V. Barranco et al., J. Phys. Chem. C 114, 10302 (2010). Functional Nanomaterials, Brookhaven National Laboratory,
short period. 24. T. Liu, T. V. Sreekumar, S. Kumar, R. H. Hauge, which is supported by the DOE, Office of Basic Energy
R. E. Smalley, Carbon 41, 2440 (2003). Sciences, under contract DE-AC02-98CH10886. We thank
25. M. A. Lillo-Rodenas, D. Cazorla-Amoros, A. Linares-Solano, P. Stephens for XRD data collection, which is supported by
References and Notes Carbon 41, 267 (2003). DOE under contract DE-AC02-98CH10886, and P. Ercius
1. J. R. Miller, P. Simon, Science 321, 651 (2008). 26. J. A. Leiro, M. H. Heinonen, T. Laiho, I. G. Batirev, for data collection on the TEAM instrument, supported
2. A. Burke, Electrochim. Acta 53, 1083 (2007). J. Electron Spectrosc. Relat. Phenom. 128, 205 (2003). by DOE contract DE-AC02-05CH11231. We thank
3. J. A. Fernandez et al., J. Power Sources 175, 675 27. S. D. Berger, D. R. McKenzie, P. J. Martin, Philos. Mag. M. Nilges for help with EPR. We thank J. Potts for
(2008). Lett. 57, 285 (1988). providing graphite oxide samples. We appreciate use
4. P. Simon, Y. Gogotsi, Nat. Mater. 7, 845 (2008). 28. P. I. Ravikovitch, A. Vishnyakov, R. Russo, A. V. Neimark, of equipment in K. Johnston’s lab. R.M.W. and A.P.
5. J. Chmiola et al., Science 313, 1760 (2006). Langmuir 16, 2311 (2000). acknowledge the partial support of the GRC-NRI SWAN
6. N. L. Wu, Mater. Chem. Phys. 75, 6 (2002). 29. K. S. W. Sing et al., Pure Appl. Chem. 57, 603 (1985). Center for the XPS data collection and analysis. R.S.R.,
7. T. Brezesinski, J. Wang, S. H. Tolbert, B. Dunn, 30. J. C. Groen, L. A. A. Peffer, J. Perez-Ramirez, Microporous Y.Z., M.D.S., and S.M. have filed a U.S. patent application
Nat. Mater. 9, 146 (2010). Mesoporous Mater. 60, 1 (2003). (application no. PCT/US2011/036164) regarding work
8. A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, 31. A. V. Neimark, Y. Lin, P. I. Ravikovitch, M. Thommes, in this paper.
J. P. Ferraris, J. Power Sources 47, 89 (1994). Carbon 47, 1617 (2009).
9. M. Toupin, T. Brousse, D. Belanger, Chem. Mater. 16, 32. M. D. Stoller, R. S. Ruoff, Energy Environ. Sci. 3, 1294
Supporting Online Material

Downloaded from http://science.sciencemag.org/ on November 22, 2017

3184 (2004). (2010).
www.sciencemag.org/cgi/content/full/science.1200770/DC1
10. A. Izadi-Najafabadi et al., Adv. Mater. 22, E235 (2010). 33. P. L. Taberna, P. Simon, J. F. Fauvarque, J. Electrochem.
SOM Text
11. D. N. Futaba et al., Nat. Mater. 5, 987 (2006). Soc. 150, A292 (2003).
Figs. S1 to S12
12. Y. Zhu et al., Adv. Mater. 22, 3906 (2010). 34. C. Portet, M. A. Lillo-Ródenas, A. Linares-Solano,
References
13. M. Segal, Nat. Nanotechnol. 4, 612 (2009). Y. Gogotsi, Phys. Chem. Chem. Phys. 11, 4943 (2009).
14. M. D. Stoller, S. J. Park, Y. W. Zhu, J. H. An, R. S. Ruoff, Movie S1
35. C. O. Ania, J. Pernak, F. Stefaniak, E. Raymundo-Pinero,
Nano Lett. 8, 3498 (2008). F. Beguin, Carbon 44, 3126 (2006).
15. Y. Wang et al., J. Phys. Chem. C 113, 13103 (2009). 36. C. Largeot et al., J. Am. Chem. Soc. 130, 2730 (2008). 22 November 2010; accepted 2 May 2011
16. W. Lv et al., ACS Nano 3, 3730 (2009). 37. S. J. Townsend, T. J. Lenosky, D. A. Muller, C. S. Nichols, Published online 12 May 2011;
17. Y. Zhu et al., ACS Nano 4, 1227 (2010). V. Elser V, Phys. Rev. Lett. 69, 921 (1992). 10.1126/science.1200770

Disorder-Enhanced Transport in The electronic structure of atomic QCs has

been shown to have multifractal eigenstates
(15, 16), which may or may not be normalizable
Photonic Quasicrystals (thus, localized), depending on the critical expo-
nent associated with the given state. The transport
Liad Levi,* Mikael Rechtsman,* Barak Freedman, Tal Schwartz, Ofer Manela, Mordechai Segev† properties of QCs are directly related to the crit-
ical nature of their eigenstates, in particular, in the
presence of disorder (17). QCs have been shown
Quasicrystals are aperiodic structures with rotational symmetries forbidden to conventional to exhibit counterintuitive transport properties,
periodic crystals; examples of quasicrystals can be found in aluminum alloys, polymers, and including extremely low conductivity that in-
even ancient Islamic art. Here, we present direct experimental observation of disorder-enhanced creases with both temperature (inverse Matheisen
wave transport in quasicrystals, which contrasts directly with the characteristic suppression of rule) and spatial disorder arising from structural
transport by disorder. Our experiments are carried out in photonic quasicrystals, where we find defects (14). Both of these effects have been
that increasing disorder leads to enhanced expansion of the beam propagating through the attributed (16, 18) to hopping between critical
medium. By further increasing the disorder, we observe that the beam progresses through a states of different spatial extents near the Fermi
regime of diffusive-like transport until it finally transitions to Anderson localization and the energy (due to inelastic electron-phonon scatter-
suppression of transport. We study this fundamental phenomenon and elucidate its origins by ing for the former and elastic scattering from
relating it to the basic properties of quasicrystalline media in the presence of disorder. structural defects for the latter). This increase in
transport with disorder is directly opposite to the
nderson localization (1), a fundamental quasicrystals. Quasicrystals (QCs) (11, 12) con- characteristic behavior of crystals, wherein trans-

A concept in solid-state physics, describes

how introducing disorder can transform
a conducting crystal into an insulator. This pre-
stitute an intermediate phase between fully pe-
riodic and fully disordered media: They do not
have a unit cell and do not exhibit translation
port is reduced with increasing disorder.
Thus far, experiments on transport in atomic
QCs were carried out by the study of macroscop-
diction and subsequent experiments have shown symmetry; nevertheless, they possess noncrystal- ic conductivity. However, conductivity experiments
that, generally, disorder works to arrest trans- lographic rotational symmetry and long-range are problematic for addressing some basic ques-
port in periodic systems containing disorder order and display Bragg diffraction. Although tions on QCs. First, the mechanisms proposed to
(2–5), as well as in fully random potentials (6–10). many of the properties of QCs are now well un- explain the unusual transport in QCs assume non-
However, some systems still pose fundamen- derstood, some fundamental questions remain. Per- interacting electrons; however, conductivity mea-
tal challenges to this concept—most notably, haps one of the most intriguing questions related surements inevitably incorporate electron-electron
to QCs has to do with transport. Opposite to crys- interactions. Second, conductivity measurements
tals containing disorder, which exhibit Anderson do not allow direct observation of wave packets,
Department of Physics and Solid State Institute, Technion,
Haifa 32000, Israel. localization, it has been suggested that disorder which could be a key property in unraveling the
*These authors contributed equally to this work.
can enhance transport in QCs (13, 14). Indirect mechanisms underlying transport. With the recent
†To whom correspondence should be addressed. E-mail: experiments have indicated that in some regimes, progress in photonic lattices (19), manifesting anal-
msegev@tx.technion.ac.il increasing disorder could enhance transport (14). ogies between light propagating in a waveguide

www.sciencemag.org SCIENCE VOL 332 24 JUNE 2011 1541

Carbon-Based Supercapacitors Produced by Activation of Graphene
Yanwu Zhu, Shanthi Murali, Meryl D. Stoller, K. J. Ganesh, Weiwei Cai, Paulo J. Ferreira, Adam Pirkle, Robert M. Wallace,
Katie A. Cychosz, Matthias Thommes, Dong Su, Eric A. Stach and Rodney S. Ruoff

Science 332 (6037), 1537-1541.

DOI: 10.1126/science.1200770originally published online May 12, 2011

Downloaded from http://science.sciencemag.org/ on November 22, 2017

ARTICLE TOOLS http://science.sciencemag.org/content/332/6037/1537

SUPPLEMENTARY http://science.sciencemag.org/content/suppl/2011/05/11/science.1200770.DC1
MATERIALS

REFERENCES This article cites 38 articles, 4 of which you can access for free
http://science.sciencemag.org/content/332/6037/1537#BIBL

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Use of this article is subject to the Terms of Service

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights reserved; exclusive
licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. The title
Science is a registered trademark of AAAS.