An All-Organic Proton Battery

Article abstract of DOI:10.1021/jacs.7b00159

Rechargeable batteries that use organic matter as the capacity-carrying material have previously been considered a technology for the future. Earlier batteries in which both the anode and cathode consisted of organic material required significant amounts

Full text of DOI:10.1021/jacs.7b00159

Subscriber access provided by Fudan University
Article
An All-Organic Proton Battery
Rikard Emanuelsson, Mia Sterby, Maria Strømme, and Martin Sjödin
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b00159 • Publication Date (Web): 15 Mar 2017
Downloaded from http://pubs.acs.org on March 15, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 23
Journal of the American Chemical Society
1
2
3
4
5
An AllꢀOrganic Proton Battery
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Rikard Emanuelsson, Mia Sterby, Maria Strømme, and Martin Sjödin*
Nanotechnology and Functional Materials, Department of Engineering Sciences, The
Ångström Laboratory, Uppsala University, Box 534, SE-751 21 Uppsala, Sweden
* Martin.Sjodin@angstrom.uu.se
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 23
1
2
Abstract
3
4
5
6
7
8
Rechargeable batteries that use organic matter as the capacityꢀcarrying material have
previously been considered a technology for the future. Earlier batteries in which both the
anode and cathode consisted of organic material required significant amounts of conductive
additives and were often based on metalꢀion electrolytes containing Li+ or Na+. However, we
have used conducting poly(3,4ꢀethylenedioxythiophene) (PEDOT), functionalized with
anthraquinone (PEDOTꢀAQ) or benzonquinone (PEDOTꢀBQ) pendant groups as the negative
and positive electrode materials, respectively, to make an allꢀorganic proton battery devoid of
metals. The electrolyte consists of a proton donor and acceptor slurry containing substituted
pyridinium triflates and the corresponding pyridine base. This slurry allows the 2eꢀ/2H+
quinone/hydroquinone redox reactions while suppressing proton reduction in the battery cell.
By using strong (acidic) proton donors, the formal potential of the quinone redox reactions is
tuned into the potential region in which the PEDOT backbone is conductive, thus eliminating
the need for conducting additives. In this allꢀorganic proton battery cell, PEDOTꢀAQ and
PEDOTꢀBQ deliver 103 and 120 mAh gꢀ1 which correspond to 78 % and 75 %, respectively,
of the theoretical specific capacity of the materials at an average cell potential of 0.5 V. We
show that PEDOTꢀBQ determines the cycling stability of the device whilst PEDOTꢀAQ
provides excellent reversibility for at least 1000 cycles. This proofꢀofꢀconcept shows the
feasibility of assembling allꢀorganic proton batteries which require no conductive additives,
and also reveals where the challenges and opportunities lie on the path to producing plastic
batteries.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 23
Journal of the American Chemical Society
1
2
Introduction
3
4
5
6
7
8
Assembling batteries in which the capacityꢀcarrying electrode material consists of organic
redoxꢀactive molecules is an attractive route towards clean, sustainable, inexpensive
electrochemical energy storage.1ꢀ2 Unlike metalꢀcontaining batteries that presently power
most of everyday life, organic batteries could be produced from renewable sources, and could
provide competitive electrochemical characteristics while being safer and easier to recycle at
endꢀofꢀlife.3ꢀ4 The above benefits, together with the vast potential for tailoring properties
through organic synthesis, are fueling research on organic matterꢀbased batteries.5 However,
the huge success of and continuous progress in metalꢀion battery technology make this a
formidable challenge.6 It is unlikely that this will be an eitherꢀor proposition, as energy
storage is needed on multiple scales ranging from minute to small electronic devices, up to
storage facilities the size of warehouses; of these, some can be stiff while others need to be
flexible and even wearable.7ꢀ9 So far, the primary approach to organic battery research has
been to replace the inorganic metal oxides used for the cathode in metalꢀion batteries with
redoxꢀactive organic molecules, which are often polymeric to avoid dissolution, while
keeping the metalꢀcontaining anode and electrolyte.10ꢀ11 Quinones and other carbonyl
derivatives and sulphur and organic radicals are primarily used as the redoxꢀactive organic
compounds.12 Of these, quinones are particularly attractive because of their high specific
capacity, fast 2eꢀ redox reactions and structural diversity.13 Using a metal as the anode
maximizes the operating window of the device. However, it also requires the electrolyte to be
tailored for metalꢀion cycling reactions to avoid side reactions at the anode,14 and thus it is not
necessarily optimal for the organic cathode. Allꢀorganic batteries, where the capacityꢀcarrying
materials for both the anode and the cathode are organic redoxꢀactive molecules, are much
rarer. Nishide pioneered this field by using polymeric organic radicals based on Nꢀoxides and
phenoxyls to assemble dualꢀionꢀtype batteries in which both cations and anions are needed to
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 23
1
2
3
4
5
6
7
8
9
chargeꢀcompensate the redox reactions at the electrodes.15 His group also assembled bipolar
(i.e. the same material in the cathode and anode) allꢀorganic batteries using organic radicals;
they produced dualꢀion batteries as well as conventional singleꢀion batteries in which one ion,
tetrabutylammonium, moves between the electrodes in a rockingꢀchair motion to compensate
the charge.16 Yang and coꢀworkers have constructed bipolar dualꢀion batteries based on
polyparaphenylene,17 as well as allꢀorganic Naꢀion batteries with polytriphenylamine and
poly(anthraquinonyl sulphide) as the positive and negative electrodes, respectively.18 Yao and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ
coꢀworkers designed a molecular ion battery that shuttled hexafluorophosphate ions (PF6 )
between two redox polymer electrodes.19 A number of diꢀ and tetraꢀanionic organic salts
based on quinones balanced by Li ions and employed as both anode and cathode material
have been investigated, first by Poziot, Tarascon, Dolhem and coꢀworkers,20ꢀ21 and recently
by several other groups employing similar strategies and using Li or Na salts of terephthalic
acids,22ꢀ23 quinones,24 and imides.25ꢀ26 Generally, organic molecules are insulators, and all of
the above studies used large amounts of conductive additives, often different forms of carbon,
e.g., carbon black or carbon nanotubes, ranging from 20 to 80 % by weight. The use of use Li
or Na ions as chargeꢀcompensating electrolyte ions is prevalent in these studies, with research
into the more abundant Na ion gaining prominence. However, there is one ion that is even
more abundant, that is lighter and smaller than these, and that binds covalently to anionic
organic functional groups: the proton (H+). Armand and Tarascon predicted the future
development of entirely metalꢀfree allꢀorganic proton batteries,1 whilst Song and Zhou
suggested that a proton battery could consist of a conjugated carbonyl anode (denoted N) and
a conjugated carbonyl cathode (N’ꢀH+).5 We have now realized this concept; we have created
an allꢀorganic proton battery using the envisioned carbonyl compounds and developed a
protonꢀcontaining organic electrolyte, in addition to assembling a battery with a conductive
polymeric backbone. The allꢀorganic battery has a wellꢀdefined charge/discharge potential,
4
ACS Paragon Plus Environment

Page 5 of 23
Journal of the American Chemical Society
1
2
3
4
5
provides discharge capacities close to the theoretical limit, and has excellent rate capability
without any conductive additives.
6
7
8
9
Results and discussion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The design of the allꢀorganic proton battery in the discharged (1a ꢀ top half) and charged (1a ꢀ
bottom half) states is depicted in Figure 1 together with illustrations of the redox reactions and
how these processes are followed by different electrochemical techniques (1b, c and d). The
capacityꢀcarrying material is based on the two electron/two proton (2eꢀ/2H+)
quinone/hydroquinone (Q/QH2) redox couple and we used an anthraquinone
(discharged/oxidized form: AQ, charged/reduced form: AQH2), and a benzoquinone
(charged/oxidized form: BQ, discharged/reduced form: BQH2) as the negative and positive
electrode materials, respectively. The AQ and BQ moieties were attached as pendant groups
to a conducting polymer backbone, poly(3,4ꢀethylenedioxythiophene) (PEDOT), to form the
conducting redox polymers (CRPs)27 PEDOT-AQ and PEDOT-BQ (Figure 1a). The
conducting backbone ensures facile electron transport throughout the material without any
conducting additives while also anchoring the redox group to avoid its dissolution into the
electrolyte. Finally, the electrolyte consisted of a slurry of proton donors and acceptors in the
form of protonated pyridinium triflate derivates and the corresponding pyridines (Figure 1a –
centre). This slurry allows the AQ and BQ 2eꢀ/2H+ redox reaction to occur, with the quinones
switching between a carbonyl and an alcohol form, while the charge balance is maintained by
shuttling protons between the electrodes in a rockingꢀchair motion, precisely as in metalꢀion
batteries.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 23
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Depiction of the conducting redox polymers (electrode materials), the pyridineꢀ
based proton donors and acceptors (electrolyte), and the redox reactions which occur during
charge (cyan) and discharge (grey). a) The chemical structures of the negative electrode
material PEDOT-AQ (top left) and positive electrode PEDOT-BQH2 (top right) in the initial
discharged state. The redox reactions during charging are shown in cyan as the AQ form is
reduced to AQH2 (bottom left) while, simultaneously, BQH2 is oxidized to BQ (bottom right),
ending in the charged state. The pyridinium and pyridine electrolyte, seen in the middle,
enable the transfer of protons from the positive to the negative electrode while the electrons
are transported in the external circuit. During discharge the opposite reactions occur, as shown
6
ACS Paragon Plus Environment

Page 7 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
in grey, resulting in a net flow of protons towards the positive electrode. b) The threeꢀ
electrode voltammograms of PEDOT-AQ (left) and PEDOT-BQ (right) showing the redox
reactions depicted as charging (cyan) and discharging (grey) as they occur in a battery. The
potential difference between the peaks in the voltammograms is the expected cell potential of
the battery. c) The charge and discharge processes shown as the output from a constantꢀ
current, twoꢀelectrode, galvanostatic, chargeꢀdischarge experiment showing plateaus at cell
potentials where the redox reactions occur. d) A differential capacity plot using the
galvanostatic data from c) to illustrate these processes in a way that resembles a voltammetric
twoꢀelectrode experiment.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Behind this straightforward design lie several challenges. First: conducting polymers are only
conductive in certain potential regions, otherwise they are insulators.28 Thus, to take
advantage of the conductivity of the backbone and avoid the use of conductive additives, the
redox reaction of the pendant groups, in this case AQ/AQH2 and BQ/BQH2, needs to occur in
the conductive region of the backbone. Second: the system requires an electrolyte that allows
the pendant group redox reaction to happen reversibly, quickly, and without side reactions
within the conducting region of the polymer backbone. We have previously shown that the 2eꢀ
/2H+ redox reaction involving BQ attached to a polypyrrole backbone is sustained by using an
organic donor (acid) and acceptor (base) as the electrolyte.29 The resultant quinone redox
potential was controlled or tuned by the organic donor/acceptor strength. With a 1:1 ratio of
acid to base, the quinone potential could be shifted 0.5 V by varying the pyridinium pKa
value. PEDOT, which has been shown to provide an accessible and stable fundament in other
CRP investigations on organic energy storage, was chosen as the conductive backbone.30ꢀ31
First, we synthesized two functionalized monomers based on 3,4ꢀethylenedioxythiophene
(EDOT), one carrying an AQ pendant group (EDOT-AQ) and one a BQH2 (EDOT-BQH2)
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 23
1
2
3
4
5
6
7
8
9
(see the Supporting Information [SI] for synthetic procedures and full characterization).
These were subsequently either electropolymerized onto glassy carbonꢀ (for threeꢀelectrode
characterization) or graphite current collectors, or chemically polymerized and deposited onto
a graphite current collector (see the Methods section for the polymerization conditions and
details of the electrochemical investigations). Figures 2a and b show the threeꢀelectrode cyclic
voltammetry of PEDOT-AQ and PEDOT-BQ in MeCN solutions of different pyridinium
triflates and the corresponding free pyridines (0.1 M each) with pKa values of: ꢀ0.43 (2ꢀ
fluoropyridine), 2.76 (3ꢀchloropyridine), 5.23 (pyridine) and 7.46 (2,4,6ꢀtrimethylpyridine).
The expected Q/QH2 redox peak was evident for PEDOT-BQ in all electrolytes and, similar
to our previous study, the formal potential of the Q/QH2 reaction decreased by about 60 meV
per pKa (Figure 2c). A scan rate study revealed mixed diffusionꢀkinetic behavior for thicker
films of both polymers (SI section S5). Interestingly, for PEDOT-AQ, there was a significant
Q/QH2 redox peak only in the 2ꢀfluoropyridine electrolyte (Figure 2b). This is in contrast to
the monomers in solution where the AQ redox reaction is evident and follows a shift of ꢀ58
meV/pKa in different pyridine electrolytes.29 Transferring the polymerꢀcovered electrode from
any of the other pyridine electrolytes to cells containing the 2ꢀfluoropyridine electrolyte
revealed that the AQ functionality was still active; its redox chemistry was essentially silent in
the less acidic electrolytes (see SI section S6 for results from experiments on the monomers
and a discussion about chargeꢀtrapping phenomena on the oxidation sweep of these polymer
films). Monitoring the potential dependence of the conductance on the polymers in situ during
electrochemical redox conversion using interdigitated array (IDA) microelectrodes (Figures
2d and e)32 provided the answer to why the AQ redox reaction was silent in the less acidic
electrolytes. In the 2ꢀfluoropyridine electrolyte, the conductance onset potential, which
coincided with the onset of oxidation of the PEDOT backbone of PEDOT-AQ, was just
below the formal potential of the AQ/AQH2 redox reaction. The data obtained from the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment

Page 9 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
EDOTꢀAQ monomer in solution shows that in the less acidic electrolytes the AQ/AQH2 redox
reaction was well outside the conductive region of the polymer backbone (Figure 2c), In
contrast, even the least acidic 2,4,6ꢀtrimethylpyridine electrolyte was still not enough to shift
the BQ/BQH2 redox reaction outside the polymer conductivity window. In comparison, there
were no peaks associated with the quinone redox reaction in the voltammograms for Li ionꢀ
and Na ionꢀcontaining electrolytes. The reason for this is as discussed above; in a CRP, a
redox match between the pendant group redox reaction and the conductive potential region of
the conductive polymer backbone is an absolute requirement for successful polymerꢀpendant
group combinations. In Li ionꢀ and Na ionꢀbased electrolytes and with the less acidic
pyridinium ions, the BQ/BQLi2, BQ/BQNa2, AQ/AQLi2, AQ/AQNa2, and AQ/AQH2 redox
reactions, respectively, occur outside the conductive potential window and the requirement
above is not met (Figure 2c). The IDA measurement also revealed that the doping/dedoping
process of both polymers was relatively unaffected by the choice of pyridine electrolyte. The
attachment of the large AQ pendant groups to the PEDOT backbone caused the doping onset
to occur at a higher potential than for the smaller BQ, where the doping onset was close to
that of unsubstituted PEDOT (SI section S7). Presumably, this variance between the polymers
was due to the dissimilar steric bulks of the pendants, resulting in subtle differences in the
polymer ordering. Electrochemical quartz crystal microbalance (EQCM) experiments on both
polymers revealed a complex uptake and expulsion of pyridinium ions, triflate ions and ion
pairs during the doping/deꢀdoping processes. However, during the Q/QH2 redox reactions, the
mass change per charge was very close to one, indicating only uptake or expulsion of protons
(see SI section S8). Redox conversion of the pendant group was thus only associated with
charge compensation by protons and the mass transport relied only on the diffusion and
migration of the small, mobile proton. Unfortunately, like protons in water, pyridinium ions
undergo reduction and this process is dependent on the proton activity (Figure 2f). The
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 23
1
2
3
4
5
6
7
8
9
process somewhat overlaps with the AQ to AQH2 reduction and is detrimental to a battery cell
(SI section S9). Recent research into electrolytes for electrochemical energy storage has
revealed that it is possible to expand the potential window in which aqueous electrolytes are
stable and to suppress proton reduction by employing very concentrated electrolyte slurries.33ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34
We employed a similar technique and utilized an equimolar (1:1) slurry of 2ꢀ
fluoropyridinium triflate and 2ꢀfluoropyridine without any additional solvent, which
significantly shifted the proton reduction to lower potentials (Figure 2f).
Figure 2. The threeꢀelectrode electrochemical characterization of PEDOT-AQ and PEDOT-
BQ, including in situ conductance measurements and galvanostatic reduction of the
electrolyte. The different colors indicate the electrolyte used: 0.1 M 2ꢀfluoropyridinium
triflate ꢀ 2ꢀfluoropyridine in MeCN (black), 3ꢀchloropyridinium triflate ꢀ 3ꢀchloropyridine
(grey), pyridinium triflate ꢀ pyridine (orange), 2,4,6ꢀtrimethylpyridinium triflate ꢀ 2,4,6ꢀ
trimethylpyridine (yellow), and 1:1 slurry of M 2ꢀfluoropyridinium triflate / 2ꢀfluoropyridine
10
ACS Paragon Plus Environment

Page 11 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
without solvent (brown). a) and b) The voltammograms from PEDOT-BQ and PEDOT-AQ
in the different pyridine electrolytes. c) The formal potentials vs. pKa of the pyridium ion and
its linear dependence on PEDOT-BQ (squares) and PEDOT-AQ (circles). The halfꢀfilled
circles and dotted line indicate where the AQ/AQH2 reaction occurs in the EDOT-AQ
monomer. The pKaꢀindependent formal potentials of the first (filled) and second (empty)
BQ/BQLi2 (stars) and AQ/AQLi2 (triangle) redox reactions are also shown as references35ꢀ36
The colored regions indicate the conductive potential region for PEDOT-BQ (grey) and
PEDOT-AQ (dashed grey) obtained from IDA measurements. d) and e) IDA measurements
showing the conductance at different potentials from the same polymer film in different
electrolytes for PEDOT-BQ and PEDOT-AQ. f) The galvanostatic reduction of the 0.1 M
electrolyte solutions and, for comparison, the 1:1 electrolyte slurry (brown) using a graphite
working electrode and a similar current density to that used in the battery tests.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on the potential difference between the AQ/AQH2 and BQ/BQH2 reactions, a cell
potential of approximately 0.5 V can be expected from the redox reactions in the threeꢀ
electrode setup. We assembled the allꢀorganic batteries in the discharged state, i.e. with
PEDOT-AQ as the negative electrode (anode) and PEDOT-BQH2 as the positive electrode
(cathode) (Figure 1a ꢀ top). A swagelok cell was used for the battery tests and a glass
microfiber filter wetted with the 2ꢀfluoropyridine electrolyte slurry served as separator (see
the Experimental section and SI section S10 for full information and pictures). In order to
investigate the properties of each polymer type separately, we assembled cells with
significantly more of one of the polymers (n/p ratio 2:1 or 1:2 determined from the peak area
of the Q/QH2 redox reaction in threeꢀelectrode cyclic voltammetry of the electrode; see SI
section S11 for examples of pre and post voltammograms of the electrodes), which allowed
characterization of the limiting material. Constantꢀcurrent galvanostatic chargeꢀdischarge
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 23
1
2
3
4
5
6
7
8
9
cycles using different current densities for PEDOT-AQ as the limiting material gave a
discharge capacity of 103 mAh gꢀ1 (78% of the theoretical specific capacity) for the
AQ/AQH2 2eꢀ/2H+ redox reaction at a current density of 70 mA gꢀ1, approximately
corresponding to a Cꢀrate of C/2 (Figure 3a and b). Note, that the mass used is based on the
total deposited mass and energy dispersive Xꢀray spectroscopy experiments (see the
experimental section) indicate remaining electrolyte from polymerization. The discharge
capacity remained above 90 mAh gꢀ1 at current densities up to 400 mA gꢀ1 (3C) and then
gradually decreased to 25 mAh gꢀ1 at 20 000 mA gꢀ1 (160 C). A plateau centered at 0.5 V,
consistent with the Q/QH2 redox reaction, was apparent, contributing to most of the capacity
(see Figure 3b and SI section S12 for differential capacity plots). Figure 3a shows a
competing, irreversible process at low current densities during the charging; the majority drew
close to the cutꢀoff potential (0.8 V in this case, SI section S13), indicating that it occurred at
lower potentials than the reduction of AQ to AQH2. We attribute this to proton reduction
partly caused by the electrode imbalance, as a large capacity excess on the positive cathode
forced the potential of the anode into the proton reduction potential region (Figure 2f).
However, this process did not adversely influence the stability of the anode and, at higher
current densities; the coulombic efficiency was close to 100%. PEDOT-AQ was very stable
up to some 100 cycles at 400 mA gꢀ1, giving a discharge capacity of 88 mAh gꢀ1 (66 % of
theoretical) with a constant cell potential before it dropped off and reached 70 mhA gꢀ1 after
150 cycles (Figure 3c). Differential capacity plots of these results effectively demonstrate the
electrochemical process occurring and show the capacity associated with the Q/QH2 redox
reaction centered at 0.5 V and how it changes during cycling (Figure 3d). Postꢀbattery threeꢀ
electrode cyclic voltammetry of the electrode revealed that primarily PEDOT capacity
remained on the positive electrode while PEDOT-AQ was more or less unaffected. PEDOT-
BQ was thus responsible for the loss of capacity (SI Figure S17). In fact, the same PEDOT-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
ACS Paragon Plus Environment

Page 13 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
AQ electrode could be used repeatedly in several battery cells, exchanging the cathode, and
was stable for at least 1000 cycles (SI Figure S18). For PEDOT-BQ it was possible to
achieve a capacity of 120 mAh gꢀ1 (75%, at 140 mA gꢀ1, 1C), gradually decreasing to about 40
mAh gꢀ1 at 14000 mA gꢀ1 (100 C). In the stability test performed at 600 mA gꢀ1, the battery
delivered an initial capacity of 90 mAh gꢀ1 (56%), subsequently decreasing steadily to 40
mAh gꢀ1 over the next 125 cycles (Figure 3g and h). At lower current densities, the dropꢀoff
was faster and postꢀbattery threeꢀelectrode investigation of the electrode showed an
essentially unchanged PEDOT capacity with a minor broad Q/QH2 peak (SI Figure S19).
However, when PEDOT-BQ was used as the limiting material, the coulombic efficiency
improved at low current densities; this strengthened our belief that reduction of protons
mainly occurs at lower potentials reached after completing the AQ to AQH2 reduction. To
summarize, high coulombic efficiencies for the allꢀorganic proton battery with concentrated
acidꢀbase electrolytes can be achieved by avoiding depletion of the negative electrode. The
cause of capacity fading during cycling has been associated with processes occurring at the
positive electrode, and stabilizing the BQ group is expected to improve cyclability. Finally,
the battery device without conductive additives, as presented in this work, demonstrated good
rate performance that could be traced to fast proton cycling and the synergy between the
pendant group redox chemistry and the conducting polymer backbone, accomplished by redox
matching.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 23
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Battery cell characteristics measured in a twoꢀelectrode setup with PEDOT-AQ as
limiting material (top panels) and PEDOT-BQ as limiting material (bottom panels). a)
Specific capacity for the charge (cyan) and discharge (grey) processes at different current
densities for PEDOT-AQ. b) Selected discharge curves, 2nd cycle in the set of 4, from a). The
dashed cyan curve was the 38th cycle. c) Stability cycling at 400 mA gꢀ1 showing the specific
capacity (left axis) for the charge (cyan) and discharge (grey) processes as well as the
coulombic efficiency (black, right axis). d) The corresponding differential capacity plots from
c) for the 2nd, 50th, 100th and 150th cycles. e) Specific capacity at different current densities for
PEDOT-BQ. f) Selected discharge curves, 2nd cycle in the set of 4, from e). The dashed cyan
curve is the 28th cycle. g) Stability cycling at 600 mA gꢀ1 showing the specific capacity (left
axis) for the charge (cyan) and discharge (grey) processes as well as the coulombic efficiency
(black, right axis). h) The corresponding differential capacity plots from g) for the 2nd, 25th,
75th and 125th cycles.
14
ACS Paragon Plus Environment

Page 15 of 23
Journal of the American Chemical Society
1
2
Conclusions and further perspectives
3
4
5
6
7
8
In this proofꢀofꢀprinciple study, we have shown how two quinoneꢀbased CRPs can be used as
positive and negative electrode materials to assemble a proton battery by employing a slurry
of organic acids and bases as an ionic liquidꢀtype electrolyte. While an aqueous electrolyte
could theoretically also function with these electrode materials, the redox reaction of
PEDOT-AQ becomes very slow in an acidic aqueous electrolyte. We speculate that the large
hydrophobic AQ unit was causing this problem (SI section S14), and are working to solve this
issue. Importantly, in the presented batteries, no conductive additive was needed when the
redox reaction of the capacityꢀcarrying pendant groups overlapped with the conductive
potential region of the polymer backbone (Figure 4). This is in stark contrast to previous
metal–organic and allꢀorganic batteries in which it has not been unusual for the organic
electrode material to consist of at least 50 wt% conductive additives. Much work remains to
be done to realize industrializable devices; the cell potential is limited, the electrolyte can
likely be developed further, and the stability of the cathode material needs to be improved. On
the other hand, the stability and rate capability of PEDOT-AQ show the potential of this
concept. Importantly, the presented battery technology is highly versatile (Figure 4 ꢀ panels a,
b, and c). First, the developed electrolyte system makes it possible to shift the formal potential
of the pendant redox group without significantly affecting the conductive region of the
polymer by varying the donor/acceptor strength of the pyridine electrolyte through
substitution. Second, quinones can function as either anode or cathode Third, there is wide
structural diversity in, and many commercially available, quinone derivatives. Fourth, any
combination of a conductive polymer backbone and a capacityꢀcarrying group is, in theory,
possible. These factors allow for a high degree of tailoring and a large design space with
opportunities to increase the conductive region, and to find suitable, stable combinations of
capacityꢀcarrying units for the desired application and for maximal cell potential. Indeed,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 23
1
2
3
4
5
6
7
8
9
quinones substituted with electronꢀwithdrawing groups have significantly higher redox
potentials than their unsubstituted analogs.37ꢀ39 Utilizing such materials as the cathode and
keeping the AQ anode could theoretically create a proton battery with a cell potential well
above 1 V, bringing us closer to competitive and truly plastic electrochemical energy storage
(Figure 4 – panel c).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Depiction of the versatile nature of the CRP and electrolyte concept presented
herein. The formal potential of the quinone redox reaction is approximately marked by its
chemical structure while the conductive and insulating regions of PEDOT are depicted below
(also marked with a dashed line). The different background colors correspond to the
electrolyte slurry used and show the approximate point at which the potential pyridinium
reduction occurs (cyan – 2ꢀfluoropyridine, purple – pyridine). The plus and minus signs
indicate the use of Q as the positive or negative electrode. a) Shows the conditions we used to
provide a cell potential of ~0.5 V. b) Shows how BQ can be used as anode in a pyridineꢀbased
battery and how the undesired proton reduction shifts to lower potentials. Note that the AQ
moiety is outside of the conductive region, thus demonstrating that there is design space both
below and above the BQ redox reaction which can be accessed through substitution to
increase stability and/or cell potential. c) Shows how electronꢀwithdrawing substituents could
16
ACS Paragon Plus Environment

Page 17 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
be used to allow batteries with cell potential above 1 V. Note also that there are opportunities
to extend the cell potential even further if the conductive polymer region can be extended to
more negative potentials.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
General considerations
Both threeꢀ and twoꢀelectrode experiments were performed under ambient conditions. MeCN
was of anhydrous quality and stored over activated 4 Å molecular sieves. Electrolyte solutions
were purged with N2 before measurement. A Pt wire was used as counter electrode and in
organic solutions the reference electrode consisted of an Ag wire submerged in an AgNO3
solution of the electrolyte which was kept in a separate compartment. Threeꢀelectrode
potentials were measured and referenced against the ferrocene redox couple (vs. Fc0/+) which
is at 3.25 V vs. Li0/+ 40
.
Three-electrode investigations
Electropolymerization onto glassy carbon disk electrodes (Ø = 3 mm) was achieved by cyclic
voltammetry on 10 mM solutions of EDOT-AQ or EDOT-BQH2 (up to 1.0 V) at 50 mV/s or
100 mV/s in 0.1 M TBAPF6/MeCN. Polymerization was also achieved with the same result
using 0.1 M pyridinium triflate/MeCN. Note that, in the presence of an electrolyte containing
the free pyridine, no polymerization occurred. Polymerization onto the IDA microelectrode
was performed with a 1 mV bias. Conductivity was subsequently measured using a 10 mV
bias at 10 mV/s. Monomers where investigated using a 2 mM solution of EDOT-AQ or
EDOT-BQH2 using square wave voltammetry. Both forward and backward scans were
recorded and the mean of these are reported (SI section S6).
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 23
1
2
Two-electrode battery cells
3
4
5
6
7
8
Circular current collectors (Ø = 13 mm) were cut from a graphite sheet and the back side was
covered with a polyamide tape (Kapton) under which a rectangular piece of graphite was
attached to allow connection to the electrode holder during electropolymerization (see SI
section S10 for pictures). Electropolymerization was achieved using the same threeꢀelectrode
setup as above at 25 mV/s and 10 mM EDOT-AQ (up to 1.0 V) whilst EDOT-BQH2 needed
slightly higher potentials (up to 1.15 V). Subsequently, the electrodes were cycled in the 2ꢀ
fluoropyridine electrolyte and stopped in the discharged state, rinsed with MeCN and dried at
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50°C under vacuum overnight. Chemical polymerization of EDOT-AQ was achieved using a
10 mM monomer solution in MeCN with slow addition of 3 eq of FeCl3 (anh) dissolved in a
minimal amount of MeCN. After stirring for 1 h, EtOH was added, followed by centrifugation
and repeated washing with water and EtOH to yield a black powder. The powder was mixed
with carboxymethylcellulose (CMC) and styreneꢀbutadiene rubber (SBR) binders (90:5:5 by
weight) in water, ground, and applied to the current collectors before drying at 50 °C under
vacuum overnight. The electrodes were then characterized and put in the discharged state
before drying again. Under these conditions, EDOT-BQH2 produced only short polymers that
were poorly conducting; electropolymerized material was used for the positive electrode.
Electrochemically and chemically polymerized EDOT-AQ were completely interchangeable,
producing the same electrochemical response. The electrode loading (for the limiting
electrode) was between 0.5 – 1.5 mg with an n/p ratio of approximately 1:2 (negative
limiting) or 2:1 (positive limiting). Scanning electron microscopy (SEM) on electrode cross
sections of the electropolymerized material gives a thickness of between 5ꢀ10 ꢁm for
PEDOT-BQ while PEDOT-AQ is denser at 2ꢀ4 ꢁm. Energy dispersive Xꢀray spectroscopy
(EDX) indicate that TBAPF6 electrolyte remains in the electrode even after complete deꢀ
doping of the polymer film and extensive washing and constitutes between 10ꢀ20 wt% (SI
18
ACS Paragon Plus Environment

Page 19 of 23
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
section S15). The battery cells were assembled using a swagelok cell and the electrodes were
separated by a glass microfiber filter paper (grade GF/A) wetted with the equimolar 2ꢀ
fluoropyridinium triflate – 2ꢀfluoropyridine electrolyte slurry. The cutꢀoff potentials for the
galvanostatic cycles ranged from 0 V to 0.8 V.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Associated content
Supporting information
The Supporting Information is available free of charge on the ACS Publications website at
DOI:
Synthetic details and full characterization, further electrochemical results from the scan rate
study, charge trapping data, EQCM data, full description of the battery setup, preꢀ and postꢀ
battery cyclic voltammograms of the electrodes, and differential capacity plots of the organic
battery are supplied.
Author information
Corresponding author
* Martin.Sjodin@angstrom.uu.se
Notes
The authors declare no competing financial interests.
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 23
1
2
Acknowledgements
3
4
5
6
7
8
E. Wilhelms is gratefully acknowledged for performing the EQCM measurements. We also
gratefully acknowledge the Swedish Foundation for Strategic Research (SSF), the Swedish
Research Council (VR), the Carl Trygger Foundation, the Olle Engqvist Byggmästare
Foundation, the Swedish Energy Agency, and the European Union's Horizon 2020
Programme (H2020/2014ꢀ2020) under grant agreement n° 644631 (ROLLꢀOUT) for funding
this study.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
References
(1) Armand, M.; Tarascon, J. M. Nature 2008, 451, 652ꢀ657.
(2) Larcher, D.; Tarascon, J. M. Nat. Chem. 2015, 7, 19ꢀ29.
(3) Poizot, P.; Dolhem, F. Energy Environ. Sci. 2011, 4, 2003ꢀ2019.
(4) Gao, X.ꢀP.; Yang, H.ꢀX. Energy Environ. Sci. 2010, 3, 174ꢀ189.
(5) Song, Z.; Zhou, H. Energy Environ. Sci. 2013, 6, 2280ꢀ2301.
(6) Goodenough, J. B.; Park, K.ꢀS. J. Am. Chem. Soc. 2013, 135, 1167ꢀ1176.
(7) Nishide, H.; Oyaizu, K. Science 2008, 319, 737ꢀ738.
(8) Lewis, N. S. Science 2016, 351, Aad1920.
(9) Vlad, A.; Singh, N.; Galande, C.; Ajayan, P. M. Adv. Energy Mater. 2015, 5, 1402115.
(10) Schon, T. B.; McAllister, B. T.; Li, P.ꢀF.; Seferos, D. S. Chem. Soc. Rev. 2016, 45, 6345ꢀ
6404.
(11) Muench, S.; Wild, A.; Friebe, C.; Häupler, B.; Janoschka, T.; Schubert, U. S. Chem. Rev.
2016, 116, 9438ꢀ9484.
(12) Liang, Y.; Tao, Z.; Chen, J. Adv. Energy Mater. 2012, 2, 742ꢀ769.
(13) Häupler, B.; Wild, A.; Schubert, U. S. Adv. Energy Mater. 2015, 5, 1402034.
20
ACS Paragon Plus Environment

Page 21 of 23
Journal of the American Chemical Society
1
2
3
(14) Xu, K. Chem. Rev. 2014, 114, 11503ꢀ11618.
4
5
6
7
8
(15) Suga, T.; Ohshiro, H.; Sugita, S.; Oyaizu, K.; Nishide, H. Adv. Mater. 2009, 21, 1627ꢀ
1630.
9
(16) Suga, T.; Sugita, S.; Ohshiro, H.; Oyaizu, K.; Nishide, H. Adv. Mater. 2011, 23, 751ꢀ754.
(17) Zhu, L. M.; Lei, A. W.; Cao, Y. L.; Ai, X. P.; Yang, H. X. Chem. Commun. 2013, 49,
567ꢀ569.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(18) Deng, W.; Liang, X.; Wu, X.; Qian, J.; Cao, Y.; Ai, X.; Feng, J.; Yang, H. Sci. Rep.
2013, 3, 2671.
(19) Yao, M.; Sano, H.; Ando, H.; Kiyobayashi, T. Sci. Rep. 2015, 5, 10962.
(20) Chen, H.; Armand, M.; Courty, M.; Jiang, M.; Grey, C. P.; Dolhem, F.; Tarascon, J.ꢀM.;
Poizot, P. J. Am. Chem. Soc. 2009, 131, 8984ꢀ8988.
(21) Barrès, A.ꢀL.; Gengꢂ, J.; Bonnard, G.; Renaultꢂ, S.; Gottisꢂ, S.; Mentré, O.; Frayret, C.;
Dolhem, F.; Poizot, P. Chem. Eur. J. 2012, 18, 8800ꢀ8812.
(22) Wang, S.; Wang, L.; Zhang, K.; Zhu, Z.; Tao, Z.; Chen, J. Nano Lett. 2013, 13, 4404ꢀ
4409.
(23) Wang, S.; Wang, L.; Zhu, Z.; Hu, Z.; Zhao, Q.; Chen, J. Angew. Chem. Int. Ed. 2014, 53,
5892ꢀ5896.
(24) Hu, P.; Wang, H.; Yang, Y.; Yang, J.; Lin, J.; Guo, L. Adv. Mater. 2016, 28, 3486ꢀ3492.
(25) Iordache, A.; Delhorbe, V.; Bardet, M.; Dubois, L.; Gutel, T.; Picard, L. ACS Appl.
Mater. Interfaces 2016, 8, 22762ꢀ22767.
(26) Banda, H.; Damien, D.; Nagarajan, K.; Hariharan, M.; Shaijumon, M. M. J. Mater.
Chem. A 2015, 3, 10453ꢀ10458.
(27) Karlsson, C.; Huang, H.; Strømme, M.; Gogoll, A.; Sjödin, M. J. Electroanal. Chem.
2014, 735, 95ꢀ98.
(28) Heinze, J.; FrontanaꢀUribe, B. A.; Ludwigs, S. Chem. Rev. 2010, 110, 4724ꢀ4771.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 23
1
2
3
4
5
(29) Emanuelsson, R.; Huang, H.; Gogoll, A.; Strømme, M.; Sjödin, M. J. Phys. Chem. C
2016, 120, 21178ꢀ21183.
6
7
8
9
(30) Armand, M.; Grugeon, S.; Vezin, H.; Laruelle, S.; Ribiere, P.; Poizot, P.; Tarascon, J. M.
Nat. Mater. 2009, 8, 120ꢀ125.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(31) Lowe, M. A.; Kiya, Y.; Henderson, J. C.; Abruña, H. D. Electrochem. Commun. 2011,
13, 462ꢀ465.
(32) Karlsson, C.; Huang, H.; Strømme, M.; Gogoll, A.; Sjödin, M. Electrochim. Acta 2015,
179, 336ꢀ342.
(33) Gao, Q.; Demarconnay, L.; RaymundoꢀPinero, E.; Beguin, F. Energy Environ. Sci. 2012,
5, 9611ꢀ9617.
(34) Suo, L.; Borodin, O.; Gao, T.; Olguin, M.; Ho, J.; Fan, X.; Luo, C.; Wang, C.; Xu, K.
Science 2015, 350, 938.
(35) Yao, M.; Senoh, H.; Araki, M.; Sakai, T.; Yasuda, K. ECS Trans. 2010, 28, 3ꢀ10.
(36) Song, Z.; Zhan, H.; Zhou, Y. Chem. Commun. 2009, 448ꢀ450.
(37) Speelman, A. L.; Gillmore, J. G. J. Phys. Chem. A 2008, 112, 5684ꢀ5690.
(38) Karlsson, C.; Jämstorp, E.; Strømme, M.; Sjödin, M. J. Phys. Chem. C 2012, 116, 3793ꢀ
3801.
(39) Yokoji, T.; Kameyama, Y.; Maruyama, N.; Matsubara, H. J. Mater. Chem. A 2016, 4,
5457ꢀ5466.
(40) Yang, L.; Mihali, V.ꢀA.; Brandell, D.; Strømme, M.; Sjödin, M. J. Phys. Chem. C 2014,
118,25956ꢀ25963
.
22
ACS Paragon Plus Environment

Page 23 of 23
Journal of the American Chemical Society
1
2
TOC
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
ACS Paragon Plus Environment