tron and hole at room temperature and gen-
eration of free charge carriers.
The large exciton binding energies in
π-conjugated materials are related not only
to their low dielectric constants (which offer
less screening between charges and lead to
stronger interactions) but also to the presence
of strong electron-vibration and electron-electron interactions (5). Because excitons
are neutral species unable to carry a current,
the efficiency of organic solar cells depends
critically on charge-separation processes at
heterojunctions between an electron donor
(D; typically a π-conjugated polymer or molecule) and acceptor (A; typically a fullerene
derivative), such as those displayed in the figure. These D-A heterojunctions are required
to produce a driving force to dissociate excitons into spatially separated charges and are
generally considered as an important factor
limiting efficiency compared to inorganic
p-n junction solar cells [the efficiency of
crystalline silicon solar cells is on the order
of 25% (2)]. Understanding the energy-har-vesting mechanisms in organic solar cells,
thus requires the ability to characterize the
elementary charge-generating processes as a
function of materials choice and in the presence of multiple nano- and mesoscale morphological variations.
From an electronic-structure standpoint, when an exciton appears at the D-A
interface, the exciton state can evolve into
a charge-transfer (CT) state and eventually
into a charge-separated state (5). A CT state
is a D-A interfacial state for which a hole
on a donor molecule or polymer segment
is located next to an electron on an acceptor (fullerene) molecule. In the CT state, the
electron and hole are still electrostatically
bound to one another; as a result, there is limited electronic polarization of the surrounding molecules because the CT exciton is neutral. For the electron and hole to separate,
they have to overcome their Coulomb attraction; this is facilitated by an increased electronic polarization of the surrounding molecules, which stabilizes the separated charges.
Gélinas et al. developed ultrafast spectroscopic tools that resolve the electron-hole
separation in the femtosecond regime. They
exploit the signature of the electric field that
is generated between the electron and the hole
as they separate. This field alters the molecular orbital energies of the surrounding molecules and thus their optical transition energies,
i.e., it leads to an electro-absorption (EA) signal (6). Gélinas et al. could measure the EA
signals with <30-fs precision and quantify the
electrostatic energy stored in the electric field.
In this way, they could characterize the initial
steps of electron-hole separation as they occur
in the tens of femtoseconds regime (4). The
authors measured two model systems, consist-
ing of a small molecule–fullerene blend (illus-
trated in the figure) and a polymer-fullerene
blend, and varied the D-A compositions. In
the blends with compositions leading to high
power conversion efficiencies, there appears a
clear EA signal indicative of long-range sep-
aration of electrons and holes. For the poly-
mer-fullerene blend, the electrostatic energy
between electron and hole reaches up to ~0.2
eV (about eight times the thermal energy at
room temperature) within 40 fs of excita-
tion, at which time the electron has left the
hole some 4 to 5 nm behind and the charges
can move apart freely. Efficient separation
is accomplished without requiring excess
energy beyond that needed to overcome Cou-
lomb attraction, which is in line with recent
results from Vandewal et al. (7).
Although providing a much needed char-
acterization of the charge-separation process
as it occurs at D-A interfaces, the work of
Gélinas et al. opens up many intriguing ques-
tions, particularly with regard to the nature
of the interfacial morphology and elec-
tronic structure that enables ultrafast charge
separation. The authors suggest that rela-
tively ordered domains of fullerenes at the
D-A interface allow the electrons to access
delocalized “bandlike” states. It remains to
be seen what happens in the regions where
donors and acceptors are mixed and disor-
dered, which are often invoked as a key com-
ponent of efficient solar cells (8). Also, most
of the energetic analyses to date have focused
on enthalpy considerations. Reaching a com-
plete picture will require including the role of
entropy (9), in addition to obtaining an accu-
rate description of the polarization effects.
1. C. W. Tang, Appl. Phys. Lett. 48, 183 (1986).
2. See www.nrel.gov/ncpv/images/efficiency_chart.jpg.
3. R. A. J. Janssen, J. Nelson, Adv. Mater. 25, 1847 (2013).
4. S. Gélinas et al., Science 343, 512 (2014); 10.1126/
5. J. L. Brédas et al., Acc. Chem. Res. 42, 1691 (2009).
6. L. Sebastian et al., Chem. Phys. 61, 125 (1981).
7. K. Vandewal et al., Nat. Mater. 13, 63 (2014).
8. P. Westacott et al. Energy, Environ. Sci. 6, 2756 (2013).
9. B. A. Gregg, J. Phys. Chem. Lett. 2, 3013 (2011).
Methane on the Rise—Again
Euan G. Nisbet,1 Edward J. Dlugokencky,2 Philippe Bousquet3
Atmospheric concentrations of the greenhouse gas methane are rising, but the reasons remain
Roughly one-fifth of the increase in radiative forc- ing by human-linked
greenhouse gases since 1750 is due
to methane. The past three decades
have seen prolonged periods of
increasing atmospheric methane, but the growth rate slowed
in the 1990s (1), and from 1999
to 2006, the methane burden (that
is, the total amount of methane in
the air) was nearly constant. Yet
strong growth resumed in 2007.
The reasons for these observed changes
remain poorly understood because of lim-
ited knowledge of what controls the global
methane budget (2).
Estimates of methane emis-
sions vary widely. Global esti-
mates derived from process
studies of sources (termed “bot-
tom-up”) are generally much
larger than those from direct
observation of the air (“top-
down”) (2). Local industrial
emissions may be far under-
estimated (3). The renewed
rise in the methane burden
prompts urgent questions about
the causes, but globally, in situ
monitoring to track atmospheric methane is
very limited outside the major nations.
Methane sources and sinks vary with lat-
itude. Overall, about two-thirds of the emis-
sions are caused by human activities; the
remaining third is from natural sources. At
polar latitudes, methane sources include wet-
lands, natural gas wells and pipelines, thaw-
ing permafrost, and methane hydrate associ-
ated with decaying offshore permafrost. In
1Department of Earth Sciences, Royal Holloway, University of
London, Egham TW20 0EX, UK. 2National Oceanic and Atmo-
spheric Administration, Earth System Research Laboratory,
Boulder, CO 80305, USA. 3Laboratoire des Sciences du Climat
et de l’Environnement, CEA-CNRS-UVSQ, Saclay, 91191, Gif-
sur-Yvette, France. E-mail: firstname.lastname@example.org C R E