comment

The journey of conducting polymers from

discovery to application
Shirakawa, MacDiarmid and Heeger received the 2000 Nobel Prize in Chemistry for the discovery of conducting
polymers. Here we summarize the impact of (semi)conducting polymers on fundamental research, synthetic
accessibility at scale, industrial applicability and the future.

Xugang Guo and Antonio Facchetti

T
he ground-breaking accomplishments by exposing it to halogen vapours, as with a very low intrinsic carrier density
of Shirakawa, MacDiarmid and done by the three Nobel Laureates — is (semiconductor) and enabling coupled
Heeger in 1977 have fundamentally to reduce the energy gap and enhance the electronic/ionic transport for exploration
changed our view of organic polymers, carrier density to achieve a different degree in electronic circuits, light emission, clean
from insulating ‘plastics’ to electrically of semiconducting (or even metallic) energy production, and bioelectronics, as
(semi)conducting functional materials1. behaviour (Fig. 1b) and tuning the material’s described in the following sections.
Twenty-three years later, the Nobel Prize in colour (Fig. 1c). Organic chemists also
Chemistry was awarded to the trio for their realized that the use of (hetero)aromatic Applications and commercialization
discovery. Starting from the serendipitous rings enabled the design of undoped efforts
synthesis by Shirakawa’s group of a bright polymers with tailorable intrinsic (semi) CP applications depend on their processing
and silvery polymer, trans-polyacetylene conducting properties. Thus, starting from characteristics, the doping (charge density)
(Fig. 1a), collaborative efforts with the discovery in the 1980s of important level, redox properties, and whether the
MacDiarmid’s and Heeger’s teams realized CPs such as polyaniline and polypyrrole charge transport is purely electronic or
the great potential of this material, (Fig. 1d), the field exploded in the 1990s mixed ionic/electronic (Fig. 2)2,4. The first
showing that its electrical conductivity with the realization of several structures CP use was in areas requiring heavily doped/
can be increased by ten million times on via different synthetic strategies (Fig. 1e). highly conducting materials. Importantly,
exposure to halogens vapours. Since these Equally important for the properties and it was observed that objects placed in the
pioneering studies the field of conducting application of CPs is the way the isolated reaction mixture for the electrochemical
polymers (CPs) took off, with joint efforts polymeric chains interact with each other in synthesis of polyaniline or polypyrrole
from chemists, materials scientists and condensed phase, forming supramolecular became preferentially coated with a highly
physicists at universities and in industry. assemblies that determine the processability, conducting polymer film; this opened
Note, with CP we will here refer to any opto-electronic characteristics, and film the way to explore using CPs as electrode
π-conjugated polymer — polymer having morphology. Unlike inorganic (semi) materials in batteries (Fig. 2a) and in
a backbone with alternating single and conductors, where the atoms are held antistatic/anticorrosion coating layers.
double (or triple) covalent bonds — that by strong covalent bonds, CPs consist of However, commercial success in these
can transport charges, independently of discrete macromolecules that are bound areas has varied greatly: while their use as
their intrinsic conductivities (conductor together via weaker supramolecular π–π, antistatic/electromagnetic shielding has
or semiconductor) and charge transport van der Waals and dipolar interactions. found extensive commercialization, with
characteristics. These forces are collectively strong the CP market reaching ~US$900 million
Before the discovery of polyacetylene, enough in first-generation CPs to prevent in 2018, commercial batteries based on
the CP prototype was the inorganic material their processing with solution-based CPs are unlikely to materialize. Although
poly(sulfur nitride) (SN)x (Fig. 1a), methodologies. However, molecular recent studies suggested CP-based lithium
a material exhibiting intrinsic (thus without engineering by side-chain modification and ion batteries with performance and stability
doping) metallic conductivity where use of particular heteroatoms has unlocked comparable/surpassing those of inorganics5,
electrons in a partially filled conduction these processing techniques, giving CPs a CP cost is a roadblock considering the large
band move freely at long (ballistic) paramount advantage over ‘hard’ inorganic quantity of electro-active CP required for
distances2. In (SN)x all bonds have equal analogues3. Synthesis efforts in this field their fabrication (>100–1,000 times larger
length, meaning that the bond length have been driven both by the need to than the need for other opto-electronic
alternation (BLA) along the chain is zero, understand charge transport mechanisms devices). To become competitive with
a key feature to achieve high conductivity. and structure–property relationships in inorganics, the CP cost should be ≲US$50
However, it has been suggested that these solids, as well as by more practical kg–1, a real challenge for even the cheapest
the structure of polyacetylene is more goals — such as replacing heavy metal wires CP scaled to date (see the section ‘Synthetic
energetically stable when the chain has a and lead batteries in airplanes and cars with accessability and industrial aspects’).
BLA ≠ 0, which opens an energy gap (Eg) lighter materials, developing new functional More recent applications are in organic
between the top of the valence band and coatings to dissipate static electricity and solar cells (OSCs; Fig. 2b) for photovoltaic
bottom of the conduction band (Fig. 1b), shield electromagnetic radiation, and modules, organic thin-film transistors
resulting in insulating behaviour. The realizing new displays. Efforts have also (OTFTs; Fig. 2c) for displays and circuits,
effect of doping a polymer — for instance, been directed to the development of CPs and organic light-emitting diodes (OLEDs;
922 Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials
 comment

a b c
S N S N S N S N S N S N
N S N S N S N S N S N S Insulator Undoped (p-)Doped 0.3
semiconductor semiconductor
Poly(sulfur nitride) (SN)x ‘Metallic’
Same bond length (BLA = 0) conductor
CB Undoped PDTzTI

CB 0.2

Absorbance
trans-Polyacetylene (PA) CB
CB
Different bond length (BLA ≠ 0)

Energy
Doped PDTzTI
Lower BLA Eg + +
0.1
Br2 + +
VB
n p-Doping – – n
Br Br VB VB
PA Br-doped PA VB
0
No charge carriers Charge carriers 500 1,000 1,500 2,000 2,500
Insulator Conductor
Eg, BLA decreasing Wavelength (nm)

Electrical conductivity increasing

d 2DT
OCH3 C14H29 O N
C6H13 S S S
O O n S
NH n S
n N n n S n
H n n S S N
S 2EH O C9H19 C9H19 O
S n C14H29 2DT
SO3H 2DT = 2-decyltetradecyl
2EH = 2-ethyhexyl
PEDOT PSS
Polyaniline Polypyrrole PEDOT:PSS P3HT MEH-PPV PF PBTTT DPP-T-T2
(PANI) (PPy)
2OD
O 7 O N O
2HD S
2OD O N O n
O N O O O S 0.1
O O S N S
n N
S 0.9
N N S S S
N S S O N O
N N 2OD = 2-octyldodecyl n 2OD
N
n
n O N O 2HD = 2-hexyldecyl O N O
O N 2OD = 2-octyldodecyl
2OD O
2HD
PTAA BBL P(NDI2OD-T2) or N2200 PDTzTl O P-90
7

e Chemical
Electrochemical Oxidative Metal-catalysed
C14H29
Fe2(SO4)3 C6H13
–n e– O O S Pd2(dba)3
Na2S2O8 FeCI3 S
PPy PEDOT:PSS P3HT Br Br + ME3Sn SnMe3 PBTTT + Me3SnBr
N –n H +
S S P(o-Tol)3
H S PSS S
Bias Regioirregular C14H29 Stille polycondensation Toxic

Fig. 1 | Structures, synthesis and properties of CPs. a, Chemical structure of poly(sulfur nitride) and trans-polyacetylene as well as doping of the latter
polymer with an elemental halogen (for example, Br2). b, Schematic evolution of the band structure going from an insulator to a semiconductor to a
conductor. Note, the effect of doping is to reduce the bond length alternation (BLA), decrease bandgap (Eg), increase charge carrier density, and enhance
electrical conductivity. c, Colour and optical property variation on doping of a CP (PDTzTI). Insets show the photographs of films of polymer PDTzTI coated
on glass before and after doping. d, Chemical structure of representative CPs. PEDOT:PSS, poly(3,4-ethylenedioxythiophene):polystyrene sulfonate; P3HT,
poly(3-hexylthiophene); MEH-PPV, poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene]; PF, poly(9,9-dinonyl-9H-fluorene-2,7-diyl); PBTTT, po
ly(2,5-bis(3-tetradecylthiophen-2-yl)thieno[3,2-b]thiophene); DPP-T-T2, poly{[2,5-bis(2-decyltetradecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione]-alt-
[2,2’:5’,2’’:5’’,2’’’-quaterthiophene]}; PTAA, poly(triaryl amine); BBL, poly(benzimidazobenzophenanthroline); P(NDI2OD-T2), poly{[N,N’-bis(2-octyldodecyl)
naphthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,5’-(2,2’-bithiophene)}; PDTzTI, poly(N,N’-bis(2-hexyldecyl)-2,2′-bithiazolothienyl-4,4′,10,10′-tetra
carboxydiimide); P-90, poly{[N,N’-bis(2-octyldodecyl)naphthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,5’-(2,2’-bithiophene)-ran-[N,N’-bis(7-glycol)
naphthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,5’-(2,2’-bithiophene)}. Blue structures highlight (hetero)aromatic building blocks used for the
realization of more complex CPs than PA (trans-polyacetylene). e, Example of CP synthesis. Electrochemical methods use a bias to oxidize or reduce the
monomers to form polymers. Main chemical methods include oxidative polymerization, where the reaction employs an oxidizing reagent, and metal-catalysed
polycondensations, where monomers carrying proper functional groups are coupled to produce a polymer chain using a metal complex as the catalyst. For the
Stille polycondensation the functional groups are a halogen (for example, Br) and a trialkylstannyl (for example, Me3Sn).

Fig. 2d) for display and lighting. For OSC doped or undoped CP film is needed. with less success, to replace the conducting
modules, which are expected to be a For instance, PEDOT:PSS (Fig. 1d), transparent electrode. In contrast, the layer
~US$150 million market in 2022 (of the formulated in different ways to tune producing electricity (bulk heterojunction
~US$200 billion photovoltaic market), CPs work-function, conductivity and rheological photoactive layer) requires two semicon­
can be potentially used in several of the properties, has been widely investigated as ductors that each transport one type of
device components where either a heavily an interfacial hole transport layer, as well as, charge and have a very low carrier density.

Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials 923

comment

a Li-ion battery cell b BHJ solar cell c Thin-film transistor d Light-emitting diode
+
+
+ VD Cathode –
Cathode Anode
–
e Source Drain Electron-transport layer
+ Electrolyte Hole-transporting layer
Li Semiconductor
charge ID hv Semiconductor V
V Separator + BHJ blend layer VG + + + + + emitter
– –
– – – –
Li+ Dielectric Hole-transport layer
discharge Electron-transporting layer
e– – Gate Transparent anode +
Anode Cathode
– Substrate Substrate
Substrate Substrate
–
Function: energy storage Function: light-to-electricity conversion Function: amplify/switch electronic Function: light emission
signal and power
Major parameter: energy density (Wh kg–1) Major parameter: power conversion Major parameter: quantum
efficiency (%) Major parameter: charge mobility efficiency (%)
(cm2 V–1 s–1)

Battery Flexible solar cell module TFT array OLED display

Fig. 2 | Applications of conducting polymers. Schematics of representative opto-electronic and energy devices. a, Li-ion battery. When the battery is charged,
lithium ions are generated on the positive electrode of the cell, and then move to the negative electrode through the electrolyte. Similarly, when the battery
is discharged, the lithium ions embedded in the carbon layer of the negative electrode leave and move back to the positive electrode. The energy density
represents the energy that can be derived per unit mass or volume of the cell. b, Bulk heterojunction (BHJ) OSC. Light adsorbed in the BHJ photoactive layer
generates excitons (bound hole–electron pairs), which split at the interface of the acceptor(red)/donor(blue) domains forming free electrons and holes.
These free charges drift via the separate and bicontinuous donor and acceptor domains and are eventually collected at the cathode and anode electrodes,
respectively, producing electricity. The power conversion efficiency of a solar cell is determined as the fraction of incident (solar) power that is converted to
electricity. c, Thin-film transistor (TFT) with a bottom-gate top-contact architecture. When a gate bias is applied, the TFT device turns on, triggering charge
carrier accumulation at the semiconductor–dielectric interface and thus an increased electrical current. The field-effect mobility describes how fast the
electrical charges move in the channel upon an electric field. d, OLED. Light is produced by recombination of charge carriers (holes and electrons) injected
from the anode and cathode electrical contacts, respectively, into the emitter. The external quantum efficiency represents the ratio of the number of emitted
photons to the number of electrons passing through the device. Note, CP-based battery and OLED commercialization lag behind those of OTFT and OSC.
Credit for bottom image in d: Cigdem Simsek/Alamy Stock Photo.

Here, the most successful CP has been P3HT transport layer in emerging perovskite solar of mobility (~0.5 cm2 V–1 s–1), but not in
as the hole transporter and a molecular cells, respectively; however, the commercial terms of cost since a-Si TFT backplane
fullerene as the electron transporter impact of this young technology remains to technology on glass for liquid crystal
counterpart. OSC modules based on be established6. displays is extremely inexpensive (a few tens
both PEDOT:PSS and P3HT were first OTFTs require high charge carrier of US dollars per square metre). For OLED
reported by Konarka. This company ceased mobilities to achieve high on-current, but applications, OTFTs should compete with
operations but left an impressive amount of low intrinsic charge densities to maximize poly-Si (mobility > 10 cm2 V–1 s–1), whose
data that leads us to believe that CPs are a the on-current/off-current ratio. Crystalline performance and resolution at scale are at
viable option for OSC commercial success. small molecules with a high/mid bandgap present out of reach for CPs. Thus, CP-based
Compared to batteries, the amount and thus have been initially preferred to CP in OTFTs OTFTs could address markets where added
the CP cost impact on the OSC module is far applications — with the notable exception functionalities such as compatibility with
lower, the technology competes in markets of the seminal OTFT demonstration at inexpensive flexible substrates and impact
where silicon modules cannot, and recently Mitsubishi Electric Corporation using resistance are advantageous. For instance,
developed non-fullerene acceptors have a polythiophene. OTFTs can be used as OTFT circuits on plastic foils can drive
enabled performance unthinkable only few switching elements in liquid-crystal and bendable electrochromic/electrophoretic
years ago. Thus, efforts are continuing with OLED displays. A few CPs with a low displays, whose market is predicted to reach
several companies (pre-)commercializing charge density and field-effect mobilities ~US$50 billion by 2025. While a-Si TFTs
modules such as Armor/Opvius, Epishine, that surpass those of the inorganic can be implemented on plastic substrates,
Kolon Korea, Moresco Japan and Sunew benchmark, amorphous silicon (a-Si), this platform fails key mechanical stress
Brazil. Finally, note that CPs such as have been discovered, as exemplified by protocols as demonstrated recently by E
PEDOT:PSS (or PTAA) as well as N2200 PBTTT, DPP-T-T2 and N2200 (Fig. 1d). Ink/Flexterra (Touch Taiwan Conference
are also explored as the hole and electron These CPs can compete with a-Si in terms 2018). In contrast, the mechanical and

924 Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials

 comment

electronic properties of CP give us future, justify higher fabrication and and Rieke8 pioneered the synthesis of
confidence that, on further optimization materials costs (that is why vapour-phase regioregular P3HT with a head-to-tail
in terms of performance uniformity over processing of some small molecules, which regioregularity >98% using 2-bromo-5-
large area and compatibility with common can reach costs as high as ~ US$200,000 kg–1, (bromomagnesio)-3-alkylthiophene (US
industrial manufacturing processes, OTFT can be tolerated). Technical specifications in patent 6,166,172) and 2-bromo-3-alkyl-
could have a real marketing opportunity in lighting are more relaxed, but material and 5-(bromozincio)thiophene (US patent
flexible electronics. Indeed, companies such processing cost must be kept very low to be 5,756,653), respectively, with nickel catalysts.
as E Ink, AUO, Flexenable, Plastic Logic, commercially viable, which is a substantial P3HT has been scaled with costs ranging of
pi-Crystal and Flexterra are continuing challenge for CPs. US$1,000–2,000 kg–1 for regioirregular P3HT
efforts in this direction. Another area to about US$10,000 kg–1 for regioregular
where CP OTFTs have been explored is Synthetic accessibility and industrial P3HT (regioregularity >97%).
in printing complementary circuitries to aspects There are two additional semiconducting
drive simple pressure/gas/temperature CPs can be synthesized by several polymers worth mentioning. The first is the
sensors and radiofrequency identification electrochemical (bias-induced) and chemical hole-transporter PBTTT9 developed by Merck
tags, a ~US$200 billion market in 2019. methodologies such as oxidative/reductive (US patent 7,714,098). The second is Polyera/
Complementary circuits require both polymerization and polycondensation BASF now Flexterra’s electron-transporter
electron- and hole-transporting OTFTs, reactions, which (based on literature reports P(NDI2OD-T2)10 (or N2200; US patent
which can be realized either with a single and our experience) are scalable provided 9,219,233). Both polymers have been
ambipolar CP (showing high/balanced that the production of the monomer units/ synthesized at kilogram scales using Stille
mobility for both charge carrier types) or reagents used in the synthesis is scalable coupling protocols involving dibrominated
two different unipolar CP semiconductors (Fig. 1e). The overall CP cost at scale and distannylated building blocks with a
that must be process-compatible and depends on several factors, including the palladium catalyst. These synthetic protocols
show comparable electronic performance. cost of the monomer(s) and reagents, are more complex than those for the other
The latter strategy has been pioneered by the polymerization procedure, product polymers, requiring preparation of two
Organic ID and Thin-Film Electronics but isolation and purification steps, waste monomers for each polymer and their
both companies failed to commercialize it management, and location of production. extensive purification, resulting in estimated
mostly because of the poor performance of Without reservations, PEDOT:PSS is the costs of US$60,000–80,000 kg–1.
the semiconductors prior 2015 and lack of most successful CP and has been produced Beyond synthesis scale-up and related
supply reliability. Current CPs are far better at >100 tonnes per year successfully. costs, the success of CPs depends also
performing and could meet specifications, Bayer first patented PEDOT (US patent on their compatibility with industrial
but the fundamental issue of this technology 4,959,430) and then PEDOT:PSS (US manufacturing processes. The electronic/
is the absence of reliable manufacturing patent 5,300,575) synthesis, with the semiconductor industry typically uses
tools for printing or print-transfer, processes latter compound’s synthesis involving solvent-free or water-based materials; thus,
that would dramatically reduce costs and the chemical oxidation of EDOT with CPs must be provided in formulations
enable broad applicability in the so called Fe2(SO4)3/Na2S2O8 in the presence of PSS. based on solvents that are halogen-free
internet-of-everything area. Since no Currently, this polymer can be purchased and have a low toxicity level. Furthermore,
established company has invested into a for US$300–600 kg–1. Other scalable the boiling and the flash points of the
facility of this type, we do not foresee CP conducting polymers are polypyrrole, formulated product must be below a certain
commercialization for these applications yet. synthesized by electropolymerization value to ensure compatibility with electrical
CPs have been explored in OLEDs, of pyrrole or by chemical oxidation in equipment used for their dispensation.
whose display market in smartphones solution (GB2134125), and polyaniline, Solution viscosity and particle content must
(Samsung, Apple) and televisions (LG) usually synthesized in solution starting also remain constant over a period of at
will grow above US$300 billion by 2025. from aniline, an acid (HCl) and an oxidizer least six months, preferably after storing in
However, current commercial products are (ammonium persulfate) and collected as ambient conditions, for process fidelity. For
based on small-molecule materials. Several a solid (GB1399259). The costs of these instance, formulations of non-conducting
CP-based prototypes — using, for instance, polymers are estimated below US$5,000 kg–1 polymer or polymer precursors meeting
PEDOT or PTAA as the hole-injecting and US$500 kg–1, respectively. these requirements and broadly used in the
interlayer and poly(phenylene vinylene) Among undoped CPs, P3HT is by far the display industry are photoresists, which
or polyfluorene as the electroluminescent most studied with the earliest claim back to sell for less than US$40 kg–1. In contrast,
component — have been fabricated. Among 1982 by a patent from the Tokyo Institute specifications and cost expectations of
the companies pursuing CP-based OLEDs Tech (US patent 4,521,589). The synthesis formulated CPs have strongly limited the
is Sumitomo Chemical, which developed was first specifically described in 1986 by transfer of the record-performing CPs from
materials and manufacturing technology oxidative polymerization of 3-hexythiophene research to development stage and are at
for low cost (printed) and high-efficiency with FeCl3, which is scalable. However, it the base of the delay that some technologies
large-area OLED lighting. However, it is has been shown that the solubilizing hexyl have experienced in using CPs.
very important to stress that performance, chain arrangement along the polymer
manufacturing complexity and cost backbone (regioregularity) is an important Future
requirements for the materials for OLED aspect governing charge transport, and this We believe that the field of CPs will continue
display and lighting applications diverge approach produces a regioirregular polymer to grow via three interconnected activities:
and vary to a great extent. The former as a result of the unsymmetric nature of new molecular design and chemistries,
requires high-performance/ultra-stable 3-hexylthiophene. Allied Corporation’s patent fundamental studies refining materials
characteristics, multiple well-defined colour (US patent 4,711,742) first pointed to claims understanding and discovering new
coordinates and rigorously patterned related to the side-chain arrangement without phenomena, and exploration of new or
emitting areas that, at least in the near describing synthetic details. McCullough7 revisited fields of use for commercialization.

Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials 925

comment

a b Catenated CP c
1. Metallation
N2200 Twisted catenated D–A CP H Ar1 H M Ar1 M
n 2. Purification
Ladder CP
X Ar2 X DARP Conventional X Ar2 X

n
2MX
+ 2HX Ar1 Ar2
PDTzTI Planar catenated A–A CP Two-dimensional CP n

H
H Aldol Ar2
Ar1 Ar1 n
H condensation
H
BBL
Planar ladder A–A CP

O 2H2O
Ar2
n
O

d e f
G
W G
VG
MSP L Electrolyte
Pentane
CP S D
Liquid 1
Monomer S D
Cu2+aq.
EMG
Liquid 2 VD ID
ECG
On state (VG = 0 V) Off state (VG < 0 V)
Monomer 2D-CP – – + –
– +
AP
– –
+ LFP
– –+
– + ++ +–
++–
+ ++ + – –+ +
Polymerization at the –+
– –+
liquid–liquid interface – –+

Fig. 3 | Ongoing fundamental studies and exploration of applications. a, Optimized geometries of molecular models for the benchmark n-type polymers
N2200 (D–A type), PDTzTI (A–A type) and BBL (A–A type) comprising imide acceptor (A) units and donor (D) or acceptor co-units in a ratio of 1:1. Going
from severely twisted but easily solution-processable N2200 to planar ladder BBL several properties improve such as dopability and conductivity; however,
BBL is poorly processable in conventional solvents. PDTzTI achieves planarity and good processability. b, Schematic structure going from a catenated (top)
to a ladder-type (centre) and to a two-dimensional (bottom) CP. Yellow and blue hexagons represent generic (hetero)aromatic, possibly fused units. c,
Comparison between conventional transition-metal catalysed (for example, Stille) direct (hetero)arylation polycondensation (DARP) and aldol condensation
reactions. In Pd-catalysed Stille reactions M is a trialkylstannyl group producing toxic by-products versus more treatable HX and H2O for the other reactions.
d, Schematic of a surface-confined polymerization reactor for two-dimensional conducting polymer (2D-CP). The monomers were introduced from the
edge of the reactor and directly delivered onto the interface of orthogonal solvents where the reaction occurs. e, Schematic representation of an organic
electrochemical transistor (S, source; D, drain; G, gate contacts) operating in depletion mode where applying a gate voltage changes the doping and
conductivity of the CP (inset), turning the device off. VG, gate voltage; VD, drain voltage; ID, source-drain current. f, Optical micrograph displaying the top
view of an OECT (top centre) and an OECT array conforming to the surface of a human hand (bottom left). Representative electrophysiological signals
(ECG, electrocardiography; EMG, electromyography; LFP, local field potential; AP, action potential) acquired by OECTs (bottom right). Figure adapted with
permission from: d, ref. 19, AAAS; e, ref. 4, AAAS; Springer Nature Ltd; f, ref. 28, Springer Nature Ltd.

Regarding the first point, the design of edge (bottommost) is reduced below ~–4 delocalization of the n-polaronic structure
electron-conducting polymers represents eV, efficient or stable electron transport in the backbone, moving from typical
a clear opportunity for expansion. Indeed, occurs in OTFTs10. However, despite great (hetero)arene catenation of N2200, to
it is striking that the vast majority of CPs, progress11–13, n-doped CPs do not yet meet catenated but planarized PDTzTl and
and all commercially available heavily the conductivity and ambient stability of ladder-type polymers (BBL), to even more
doped CPs, are hole transporters given the best p-doped polymers14–16. Various rigidified two-dimensional CPs (2D-CPs)
that, fundamentally, electrons are as mobile design and n-doping strategies, including proposed more recently (Fig. 3a,b). The
as holes in organic solids. This is due to all-CP blends15, are being explored. For challenges for those more rigidified CPs are
the greater extent of electron traps versus instance, given the awareness of N2200 synthesis scalability and reliable solution
hole traps formed in these materials and torsion in the backbone, and taking processability.
to the oxidizing role of our atmosphere, inspiration from the donor–donor backbone For both established and new CPs,
which further contributes to trapping in p-type PEDOT’s structure, planarized the development of efficient synthetic
electrons. Chemists attempted to address n-type CPs with an acceptor–acceptor methodologies is critical18. By definition,
this point by molecular engineering the backbone, such as PDTzTl, are being polymers do not have the structural
conducting band, and it was experimentally developed (Fig. 3a)17. Conductivity can identity of small molecules and, in most
found that when the conduction band also be further promoted by favouring cases, their properties are controlled

926 Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials

 comment

by the chain length and chain length polymers already discussed. An ideal substituents (P-90; Fig. 1d) demonstrating
distribution. Furthermore, regioregularity, n-dopant should efficiently enhance CP impressive OECT electron transport even
a small amount of crosslinking and bond electron-density/conductivity and achieve a in water29. More opportunities are also
connectivity defects may greatly affect intra- stable n-doped polymer22. Of the two major opened by the electromechanical properties
or inter-chain packing and thus charge n-doping methodologies, direct electron of CPs, whose volume changes substantially
transport. In addition, synthetic approaches transfer requires the n-dopant to have a with the insertion of ions and solvent.
avoiding toxic reagents and by-products, high-lying valence band that compromises This suggests the use of CPs for actuators
such as those used in Stille reactions, will stability particularly in ambient conditions, in surgical devices and in applications in
enhance sustainability and have a real cost whereas hydride transfer (which involves a cell biology30. The development of these
advantage. For instance, N2200 synthesis, C–H bond cleavage step in a dopant before bio-oriented applications of CPs presents
which can be carried out at scale, produces electron transfer to the CP) can potentially important fundamental and practical
more than two moles of highly toxic lead to better stability and doping challenges to scientists and engineers to
stannylated waste per mole of monomer; efficiency with further optimization of the fully leverage the potential of CPs for new
this has considerable environmental impact dopant molecules and the bond cleavage commercialization.
and also adds ~30% to the production costs process. Here lessons can be learned from From the discovery and seminal studies
for waste management. Direct (hetero) C–H activation and bond cleavage in by the Nobel Laureates to present-day
arylation polycondensation via C–H organic synthesis where catalysis plays activity, the CP field has undergone a
activation, together with aldol condensation a key role; thus, developing catalysts dramatic expansion. Pioneering studies
reactions that produce, for instance, HBr specifically designed for promoting this taught us the fundamentals of CP properties,
and H2O as by-products, respectively, are reaction by lowering the activation energy further research broadened their structural
far greener approaches (Fig. 3c). Regarding could substantially increase n-doping diversity and field of use, and nowadays
realizing two-dimensional CPs, the recent efficiencies. researchers aim at both finding new
breakthrough using surface-confined Most of the recent CP applications fundamental insights and demonstrate
polymerizations at liquid–liquid and air (or are centred around their electronic applications with commercial viability.
vacuum)–solid interfaces is a stepping-stone conductivities. However, doped samples Multiple established companies have
for additional discoveries (Fig. 3d)19. exhibit mixed ionic/electronic transport invested heavily, and more were created,
Additional progress will originate from that, in combination with the superior with the prospect of commercializing CPs
studies, both at single-polymer-chain and CP mechanical flexibility/stretchability as materials and/or implementing them
supramolecular levels, on charge transport versus hard inorganics, makes them into opto-electronic devices. Although
physics20. Questions requiring further studies appealing to bioelectronic applications applications in batteries, lighting and
are how chain conformations/motion affects such as skin-attachable or -implantable (printed) OTFTs are struggling (as a
polaron structure and charge transport; devices able to record electrophysiological combination of insufficient performance–
and how doping and ion incorporation signals, stimulate the electrical activity cost benefits for the envisioned technology
affects CP electronic structure and charge of the nervous system and promote drug and absence of suitable infrastructures for
transport characteristics, and ultimately the delivery on demand23. Researchers are their production), we believe CP-based
maximum achievable carrier mobility and making progress in understanding and products such as (printed) solar modules
conductivity. Theoretical simulations can making use of the coupled transport and (flexible) electrophoretic displays
provide quantitative data regarding transport characteristics of these materials, learning will gain traction thanks to performance
processes in polymers over multiple length from purely electronic or ionic transport comparable to or surpassing current
scales, particularly for more complex systems and developing models able to commercial (a-Si) platforms, proven
blends and doped systems. In addition, capture the intimate electronic–ionic stability, and clear advantages for
transport studies comparing polymer films interactions in these mixed conducting mechanical robustness and new design
with isotropic and aligned morphology materials4. Further improvements of the CP concepts for products. CPs will certainly
demonstrated that polaronic structures mechanical properties to better match those continue to be an active area in materials
localize on CP domains aligned along the of the biological tissues are also needed science, branching out to other disciplines
charge transport direction21, suggesting to improve electronic communication and industry. ❐
processing design strategies to fabricate and long-term operation of bioelectronics
better performing opto-electronic devices. devices24. To this end, recent progress Xugang Guo 1 ✉ and
Doping has been instrumental in includes blending CPs with elastomers and Antonio Facchetti 2,3 ✉
understanding charge transport in controlling their chain conformation25. The 1
Department of Materials Science and Engineering,
semiconductors by tuning/controlling the majority of device demonstrators, which Southern University of Science and Technology
charge carrier density, Fermi level and use an organic electrochemical transistor (SUSTech), Shenzhen, Guangdong, China. 2Flexterra
electrical conductivity. The achievement (OECT; Fig. 3e) architecture as its building Corporation, Skokie, IL, USA. 3Department of
of insulator-to-metal transition in block, mostly rely on p-type P3HT and Chemistry and the Materials Research Center,
doped polyacetylene was carried out on PEDOT:PSS26,27. Modification of the latter Northwestern University, Evanston, IL, USA.
elemental vapour exposure in vacuum CP by blending with a reductant to achieve ✉e-mail: [email protected];
to, for instance, Br2 and I2 for p-doping enhancement-mode OECTs has recently [email protected]
and Na for n-doping, which are very enabled unprecedented electrophysiological
effective but impractical. Molecular signal monitoring (Fig. 3f)28. At present, a Published online: 20 August 2020
doping is a more promising manufacturing limited number of CPs has been specifically https://doi.org/10.1038/s41563-020-0778-5
approach; however, in this case progress in designed for OECTs, and also in this case
n-doping lags behind molecular p-doping, n-type CPs are less developed — with the References
1. Shirakawa, H., MacDiarmid, A. & Heeger, A. Chem. Commun.
reflecting the intrinsic challenges in notable exception of a N2000-like backbone 1–4 (2003).
developing efficient electron-transporting functionalized with hydrophilic glycolated 2. Swager, T. M. Macromolecules 50, 4867–4886 (2017).

Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials 927

comment

3. Bronstein, H., Nielsen, C. B., Schroeder, B. C. & McCulloch, I. 16. Tang, C. G. et al. Nature 573, 519–525 (2019). 29. Paterson, A. F. et al. Nat. Commun. 11, 3004 (2020).
Nat. Rev. Chem. 4, 66–77 (2020). 17. Shi, Y. et al. Adv. Mater. 30, 1705745 (2018). 30. Melling, D., Martinez, J. G. & Jager, E. W. H. Adv. Mater. 31,
4. Rivnay, J. et al. Nat. Rev. Mater. 3, 17086 (2018). 18. Onwubiko, A. et al. Nat. Commun. 9, 416 (2018). 1808210 (2019).
5. Jia, X., Ge, Y., Shao, L., Wang, C. & Wallace, G. G. ACS Sustain. 19. Zhong, Y. et al. Science 366, 1379–1384 (2019).
Chem. Eng. 7, 14321–14340 (2019). 20. Fratini, S., Nikolka, M., Salleo, A., Schweicher, G. & Sirringhaus, H.
6. Zhang, W. et al. Adv. Sci. 5, 1800159 (2018). Nat. Mater. 19, 491–502 (2020). Acknowledgements
7. McCullough, R. D. & Lowe, R. D. Chem. Commun. 70–72 (1992). 21. Xiao, M. et al. Adv. Mater. 32, 2000063 (2020). The authors are grateful to Shenzhen Science
8. Chen, T. A. & Rieke, R. D. J. Am. Chem. Soc. 114, 22. Lin, X. et al. Nat. Mater. 16, 1209–1215 (2017). and Technology Innovation Commission
10087–10088 (1992). 23. Gerasimov, J. Y. et al. Adv. Sci. 6, 1801339 (2019). (KQTD20140630110339343), AFOSR (FA9550-18-1-0320),
9. McCulloch, I. et al. Nat. Mater. 5, 328–333 (2006). 24. Fahlman, M. et al. Nat. Rev. Mater. 4, 627–650 (2019). and ONR (N00014-20-1-2116).
10. Yan, H. et al. Nature 457, 679–686 (2009). 25. Kathy Liu, K., Tran, H., Feig, V. R. & Bao, Z. MRS Bull. 45,
11. Kiefer, D. et al. Nat. Mater. 18, 149–155 (2019). 96–102 (2020).
12. Zhao, X. et al. Adv. Mater. 29, 1606928 (2017). 26. Dijk, G., Rutz, A. L. & Malliaras, G. G. Adv. Mater. Technol. 5, Competing interests
13. Lu, Y. et al. Angew. Chem. Int. Ed. 58, 11390–11394 (2019). 1900662 (2020). A.F. is the founder and chief technology officer of Flexterra
14. Yamashita, Y. et al. Nature 572, 634–638 (2019). 27. Macchia, E. et al. Nat. Commun. 9, 3223 (2018). Corporation and has been a scientific advisor of Raynergy
15. Xu, K. et al. Nat. Mater. 19, 738–744 (2020). 28. Cea, C. et al. Nat. Mater. 19, 679–686 (2020). Teck. X.G. declares no competing interests.

928 Nature Materials | VOL 19 | September 2020 | 922–928 | www.nature.com/naturematerials