doi.org/10.26434/chemrxiv.11366186.v1

Energetic Control of Redox-Active Polymers Towards Safe OrganicBioelectronic MaterialsAlexander Giovannitti, Reem B. Rashid, Quentin Thiburce, Bryan D. Paulsen, Camila Cendra, Karl J. Thorley,Davide Moia, J. Tyler Mefford, David Hanifi, Weiyuan Du, Maximilian Moser, Alberto Salleo, Jenny Nelson,Iain McCulloch, Jonathan Rivnay

Submitted date: 13/12/2019 • Posted date: 20/12/2019Licence: CC BY-NC-ND 4.0Citation information: Giovannitti, Alexander; Rashid, Reem B.; Thiburce, Quentin; Paulsen, Bryan D.; Cendra,Camila; Thorley, Karl J.; et al. (2019): Energetic Control of Redox-Active Polymers Towards Safe OrganicBioelectronic Materials. ChemRxiv. Preprint. https://doi.org/10.26434/chemrxiv.11366186.v1

Avoiding faradaic side reactions during the operation of electrochemical devices is important to enhance thedevice stability, to achieve low power consumption, and to prevent the formation of reactive side‑products.This is particularly important for bioelectronic devices which are designed to operate in biological systems.While redox‑active materials based on conducting and semiconducting polymers represent an exciting classof materials for bioelectronic devices, they are susceptible to electrochemical side‑reactions with molecularoxygen during device operation. We show that this electrochemical side reaction yields hydrogen peroxide(H2O2), a reactive side‑product, which may be harmful to the local biological environment and may alsoaccelerate device degradation. We report a design strategy for the development of redox-active organicsemiconductors based on donor-acceptor copolymers that prevent the formation of H2O2 during deviceoperation. This study elucidates the previously overlooked side-reactions between redox-active conjugatedpolymers and molecular oxygen in electrochemical devices for bioelectronics, which is critical for the operationof electrolyte‑gated devices in application-relevant environments.

File list (2)

download fileview on ChemRxivmanuscript_191208.pdf (716.69 KiB)

download fileview on ChemRxivSI_191208.pdf (3.22 MiB)

1

Energetic control of redox-active polymers towards safe organic bioelectronic materials

Alexander Giovannitti*, Reem B. Rashid, Quentin Thiburce, Bryan Paulsen, Camila Cendra, Karl Thorley, Davide Moia, J. Tyler Mefford, David Hanifi, Du Weiyuan, Max Moser, Alberto Salleo, Jenny Nelson, Iain McCulloch, and Jonathan Rivnay Dr. Alexander Giovannitti, Dr. Karl Thorley, Max Moser, Prof. Iain McCulloch Department of Chemistry, Imperial College London, London, SW7 2AZ, UK. Corresponding author: ag19@stanford.edu Dr. Alexander Giovannitti, Dr. Davide Moia, Prof. Jenny Nelson Department of Physics, Imperial College London, London, SW7 2AZ, UK Dr. Du Weiyuan, Prof. Iain McCulloch King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering Division, KAUST Solar Center (KSC), Thuwal 23955-6900, Saudi Arabia Reem R. Rashid, Dr Bryan Paulsen SJ, Prof. Jonathan Rivnay Simpson Querrey Institute, Northwestern University, Chicago, Illinois 60611, United States Department of Biomedical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States Dr. Alexander Giovannitti, Camila Cendra, Dr. Quentin Thiburce, David Hanifi, Dr. J. Tyler Mefford and Prof. Alberto Salleo Department of Materials Science and Engineering, Stanford University, CA 94305, USA

Keywords: donor-acceptor copolymer, organic-mixed-ionic-electronic-conductor, electrochemical transistor, oxygen reduction reaction, bioelectronics

2

Abstract

Avoiding faradaic side reactions during the operation of electrochemical devices is important to

enhance the device stability, to achieve low power consumption, and to prevent the formation of

reactive side-products. This is particularly important for bioelectronic devices which are designed to

operate in biological systems. While redox-active materials based on conducting and semiconducting

polymers represent an exciting class of materials for bioelectronic devices, they are susceptible to

electrochemical side-reactions with molecular oxygen during device operation. We show that this

electrochemical side reaction yields hydrogen peroxide (H2O2), a reactive side-product, which may be

harmful to the local biological environment and may also accelerate device degradation. We report a

design strategy for the development of redox-active organic semiconductors based on donor-acceptor

copolymers that prevent the formation of H2O2 during device operation. This study elucidates the

previously overlooked side-reactions between redox-active conjugated polymers and molecular

oxygen in electrochemical devices for bioelectronics, which is critical for the operation of

electrolyte-gated devices in application-relevant environments.

Introduction

Organic semiconductors with polar side-chains have been identified as a promising class of materials

for the field of bioelectronics. These materials, also called organic mixed ionic/electronic conductors

(OMIECs), can exchange ions with aqueous electrolytes when electronic charge carriers are injected,

transported, and stored in the bulk of the material.[1] Recent developments of OMIECs based on

redox-active conjugated polymers[2–8] and novel device concepts[9,10] have opened up new pathways

for bioelectronic devices including integrated circuits for electroencephalogram (EEG) monitoring[9] or

low-power voltage amplifiers based on organic electrochemical transistors (OECTs)[11]. Specifically, the

OECT has drawn significant attention in the field of organic bioelectronics. It operates by

electrochemically modulating the conductivity of a redox-active channel material with an electrolyte

that is often aqueous, through the application of a gate bias[12]. The electrochemical charging of

3

OMIECs can be described as a capacitive faradaic charging process, meaning that the OMIEC

undergoes a change in its oxidation state through an electron transfer with the contact (current

collector), while ions from the electrolyte penetrate inside the channel material to compensate the

charge carriers on the polymer backbone electrostatically with no change in the inserted ion’s

oxidation state.[13] OECTs can be operated in either depletion or enhancement mode, depending on

the choice of channel (and gate) materials. Depletion-mode devices are initially in their conductive,

charged state, under zero gate bias and their conductivity can be decreased by performing

electrochemical discharging reactions (de-doping), while enhancement-mode devices begin in an

intrinsically low conductivity state and become conductive during electrochemical charging reactions

(doping). The latter has the advantage of dissipating less static power when the device is not

operated[14], due to low OFF currents – which must be minimised as much as possible. One figure of

merit for OECTs is their transconductance gm = ∂ID/∂VG, where ID is the drain current and VG is the gate

voltage. The gm defines how efficiently the transistor can transduce signals and depends on the width

(W), length (L) and thickness (d) of the transistor’s channel. The normalised transconductance

𝑔𝑔m,norm = 𝑔𝑔m𝐿𝐿𝑊𝑊𝑊𝑊

is often reported to benchmark OECT materials, accounting for volumetric charging

of the channel.[15]

The conducting polymer poly(3,4-ethylenedioxythiophene)poly(styrene-sulfonate) PEDOT:PSS

exhibits high performance in depletion-mode OECTs[16] and is the most commonly used material in

organic bioelectronics. Recently, efficient OMIECs were developed for enhancement-mode OECT,

which, for the first time, exceeded the performance of PEDOT:PSS[3]. This was achieved by polymer

backbone and side-chain engineering to improve electronic and ionic charge transport[2,3]. So far,

chemical design strategies for enhancement-mode OECT materials focused mostly on the design of

polymer backbones with low ionisation potentials (IPs)[2], resulting in turn-on voltages around 0 V in

aqueous electrolytes, with few studies reporting on the electrochemical redox-stability of the

materials during operation.[4,5,17]

4

Electrochemical redox stability of OMIECs is accomplished when reversible capacitive faradaic

charging and discharging reactions occur in the absence of non-capacitive faradaic side-reactions

between the OMIEC and the electrolyte (e.g. solvent molecules or ions). The products of such

side-reactions may modify the chemical composition of the OMIEC and affect the performance of

electrochemical devices.[18] So far, little attention has been paid to non-capacitive faradaic reactions

in ambient, oxygen-containing aqueous electrolytes. This reaction is an electron-transfer reaction

from the OMIEC to molecular oxygen, also described as oxygen reduction reaction (ORR), which yields

either H2O2 (two-electron process)[19] or water (H2O) (four-electron process)[20], as well as charging

(oxidation) of the OMIEC. While the latter can sometimes be reversed by applying the appropriate

potential to reduce the oxidised OMIEC, if H2O2 has formed, it may remain in the electrolyte and

accumulate during device operation. Since oxygen has a low-lying lowest unoccupied molecular orbital

(LUMO), it can spontaneously reduce by an electron transfer from electron-rich OMIECs with an

appropriately aligned high-lying highest occupied molecular orbital (HOMO).[21] Electrodes made of

PEDOT:PSS, for example, show high faradaic efficiency (nearly 100%) of H2O2 production during

electrochemical discharging of the polymer in ambient conditions[19]. The formation of H2O2 during

device operation is a concern when operating in biological environments since corrosive damage to

the device materials, or lipid peroxidation can occur.[22–24]

This work focuses on the implementation of design rules for OMIECs for electrochemical devices such

as OECTs to prevent the ORR and hence the formation of H2O2 during device operation. We develop

OMIECs based on donor-acceptor copolymers that have large IPs to shift the operational voltages of

the OECT such that no ORR occurs in ambient conditions. Specifically, we show the importance of

functionalising the donor unit of the copolymer with electron-donating groups to enhance the

electrochemical redox stability of donor-acceptor copolymers in aqueous electrolytes. The proposed

design strategy is important for the future development of safe organic bioelectronic devices for in-

vivo implementation, and for other devices such as OECT-based sensors where unintended H2O2

formation could limit the accurate detection of analytes.

5

Optoelectronic and electrochemical properties of the donor-acceptor copolymers

Copolymers based on pyridine-flanked diketopyrrolopyrrole (PyDPP) with bithiophene (T2) or

3,3’-methoxybithiophene (MeOT2) (Figure 1a) were synthesised by a Stille polymerisation technique

where the choice of the comonomer affects the optoelectronic and electronic properties of the

copolymers. The bithiophene glycolated copolymer (p(gPyDPP-T2)) has an IP of 5.3 eV measured by

photoelectron spectroscopy in air (PESA) and 5.5 eV when measured by cyclic voltammetry (CV)

(Figure S10, Supporting Information). Employing the MeOT2 unit as the comonomer

(p(gPyDPP-MeOT2)) lowered the IP to 5.0 eV as measured by both PESA and CV. An increase of the

HOMO energy level was also observed by density functional theory (DFT) calculations (tuned

ωB97XD/6-31G*) for short oligomers of PyDPP-T2 and PyDPP-MeOT2 resulting in the calculated

HOMO energy levels of 5.36 and 4.92 eV, respectively, which compares well to the experimental IP

values. Both copolymers have good solubility in chloroform, enabling facile processing from solution,

and thin-films of the copolymers are insoluble in aqueous electrolytes. Mass spectrometry indicates

the formation of several repeat units for both copolymers (Figure S7 and S9, Supporting Information),

while gel permeation chromatography was inconclusive, indicating unrealistically high molecular

weights most likely due to aggregation of the copolymers in solution. Details about the synthesis and

characterisation of the copolymers are given in Sections 2-4, Supporting Information. The thin-film

microstructure of the copolymers in their dry, as-cast conditions was studied by grazing incidence

wide-angle X-ray scattering (GIWAXS) measurements. The copolymers show similar edge-on texture

and π-stacking distances (3.49 Å for p(gPyDPP-MeOT2) and 3.46 Å p(gPyDPP-T2)), while the lamellar

spacing increases from 16.87 Å p(gPyDPP-T2) to 19.19 Å p(gPyDPP-MeOT2), most likely due to

comonomer substitution as well as the attachment of a longer side-chain on the PyDPP unit to increase

the solubility of the copolymer in organic solvents (Figure S12 and Table S2, Supporting Information).

To study the electrochemical redox reactions of the copolymers in aqueous electrolytes, cyclic

voltammetry (CV) measurements and spectroelectrochemical measurements were carried out on thin

polymer films in aqueous solution. Figure 1b presents the CV measurements of p(gPyDPP-MeOT2) in

6

0.1 M aqueous NaCl solution. p(gPyDPP-MeOT2) has an oxidation onset potential of 0.3 V vs. Ag/AgCl

and shows high redox stability during 50 charging and discharging cycles between 0 V and 0.7 V vs.

Ag/AgCl in aqueous electrolytes. Due to the large IP of p(gPyDPP-T2), the oxidation onset is shifted to

higher potentials, measured to be 0.75 V vs. Ag/AgCl (Figure S13a, Supporting Information).

Compared to the MeOT2 copolymer, the T2 copolymer displays a low electrochemical redox stability

in water-based electrolytes (Figure S14, Supporting Information), making the polymer impractical for

applications in electrochemical devices. Since the donor-acceptor copolymers have large electron

affinities (EA) (Table S1), they can also be reduced and charged with electrons. The copolymers show

reversible electrochemical reduction reactions with a reduction onset of – 0.65 V vs Ag/AgCl for

p(gPyDPP-T2) and – 0.75 V vs Ag/AgCl for p(gPyDPP-MeOT2) (Figure S13e and Figure S13f, Supporting

Information).

To study the charging of polymer thin-films in aqueous electrolytes, we carried out

spectroelectrochemical measurements for both copolymers (Figure 1c and Figure S13d, Supporting

Information). This technique is most useful for studying the charging and discharging of OMIECs where

bias-dependent changes of the absorption spectrum can be related to the degree of charging,

providing information about polaron and bipolaron formation.[8,25] It is also applicable to identify

degradation processes during electrochemical charging/discharging of OMIECs, where irreversible

changes of the absorption spectrum can be related to chemical degradation.[18] As shown in Figure 1c,

the ground state absorption peak of p(gPyDPP-MeOT2) decreases upon oxidation of the copolymer

while a new absorption peak appears for the hole polaron with a similar absorption spectrum as was

previously reported for oxidised DPP copolymer analogues[26,27]. Electrochemically reversible polaron

formation up to 0.7 V vs Ag/AgCl is observed for p(gPyDPP-MeOT2). Applying higher potentials reveals

bipolaron formation, corresponding to the charging of the polymer repeat units with two electrical

charges, detected by a shift of the isosbestic point observed for the polaron formation as well as an

increase of the absorption peak at low energy (> 850 nm) (Figure S15d, Supporting Information).

Continuous charging of the copolymer to the bipolaron state results in chemical degradation

7

(Figure S15f, Supporting Information). Due to the low electrochemical redox stability of p(gPyDPP-T2),

reversible polaron formation is only observed at a low degree of charging (between 0.75 V and 0.85 V

vs Ag/AgCl, Figure 1d), while charging to higher positive potentials results in degradation instead of

bipolaron formation (Figure S14b, Supporting Information).

One explanation for the lower redox stability of p(gPyDPP-T2) compared to p(gPyDPP-MeOT2) is

revealed by comparing orbital and charge distribution along the chain (Figure S11, Supporting

Information). The MeOT2-unit provides more localisation of the wavefunction for the hole polaron

than the T2-unit, and is therefore expected to stabilise the hole polaron further and thus increase the

redox stability of the copolymer, as we have previously shown for other donor-acceptor

copolymers[18]. To give insights into the degradation mechanism during electrochemical charging of

the copolymers in aqueous electrolytes, we carried out additional spectroelectrochemical

measurements and monitored changes of the Fourier-transform infrared (FT-IR) spectra of the

copolymers after applying potentials at which irreversible electrochemical redox reactions occur

(Section 11, Supporting Information). For both copolymers, significant changes of the FT-IR spectrum

were observed, most pronounced for the C=O stretching vibration of the DPP-core at 1663 cm-1

(p(gPyDPP-T2)) and 1654 cm-1 (p(gPyDPP-MeOT2)). Additional surface characterisations by X-ray

photoelectron spectroscopy (XPS) of pristine and electrochemically-degraded polymer films revealed

changes of the N-C=O bond of the DPP unit (Section 12, Supporting Information), indicating

degradation of the DPP unit.

Performance of p(gPyDPP-MeOT2) in OECTs

Due to the superior electrochemical redox stability of p(gPyDPP-MeOT2) in aqueous electrolytes

compared to the p(gPyDPP-T2) copolymer, we analyse the performance of the former in p-type

enhancement-mode OECTs and compare it to state-of-the-art materials for aqueous electrolyte-gated

p-type OECTs, PEDOT:PSS and p(g2T-TT). Figures 2a-b present the output and transfer curves of the

p(gPyDPP-MeOT2) OECT where almost no hysteresis is observed. The OECT turns on at gate potentials

8

VG < – 0.35 V with an on/off ratio > 105. The device has a normalised peak transconductance (gm,norm)

of 19.5 S cm–1 +/– 2.5 S cm–1 (averaged over three devices) at VG = – 0.7 V. To compare the

performance of p(gPyDPP-MeOT2) to state-of-the-art OECT materials, we measured the hole mobility

(µh) of p(gPyDPP-MeOT2) in frequency-dependent OECT bandwidth measurements[28] as well as the

volumetric capacitance (C*) by electrochemical impedance spectroscopy (EIS). We recorded a hole

mobility of 0.030 +/– 0.007 cm2 V-1 s-1 (averaged over three devices) at gate potentials of VG = – 0.7 V,

which is comparable to mobility values reported for other PyDPP copolymers tested in OFETs[29]. The

volumetric capacitance of p(gPyDPP-MeOT2) is 60 F cm-3 at an offset potential of 0.7 V vs. Ag/AgCl

(Figure S21, Supporting Information), which is on par with values reported for other donor-acceptor

copolymers with glycol side-chains[18]. gm,norm of p(gPyDPP-MeOT2) OECTs is comparable to other p-

type OECTs based on benzodithiophene (BDT) copolymers[2] or propylenedioxythiophene (ProDOT)

copolymers[6] but is lower in performance compared to state-of-the-art polymers such as p(g2T-TT)[3]

or PEDOT:PSS[16], mostly due to lower values for both µh and C*[15]. Devices based on

p(gPyDPP-MeOT2) are found to be stable under long-term pulsed cycling, as presented in Figure 2c.

After 400 voltage pulses, no or little changes in the ON current is observed when applying VG = – 0.5

V, while a decrease of 8 % and 16 % is observed with VG = – 0.6 V and – 0.7 V, respectively. To highlight

the importance of backbone engineering for donor-acceptor polymers, we also tested the polymer

p(gPyDPP-T2) in an OECT and observed a low device stability during pulsed-cycling tests (Figure S20,

Supporting Information).

Faradaic side reactions of OMIECs in ambient conditions

The advantages of current state-of-the-art polymers for aqueous electrolyte-gated OECT materials

such as p(g2T-TT)[3] or PEDOT:PSS[16,30] are many-fold, including low operational voltages of the OECTs,

as well as a high hole mobility and volumetric capacitance, resulting in a high transconductance.

However, when operating these materials in electrochemical devices in ambient conditions, faradaic

non-capacitive side-reactions can occur with molecular oxygen that result in the formation of either

9

H2O2 or H2O. The formation of H2O2 would affect the performance of the device and could cause harm

to biological systems.

To examine the role of the ORR during electrochemical redox reactions of the polymers in ambient

conditions and to find out if H2O2 or H2O is predominantly formed, we carried out a set of experiments

for PEDOT:PSS and p(g2T-TT), as well as the herein reported p(gPyDPP-MeOT2). First, the ORR was

studied by probing the polymer’s reactivity towards oxygen. We monitored the oxidation of the

polymer by UV-Vis absorption spectroscopy with polymer thin-films immersed in aqueous electrolytes

in ambient conditions. For the already oxidised polymers PEDOT:PSS and p(g2T-TT), which accumulate

hole polarons due to the ORR when handling the polymers in air or due to potential chemical side-

reactions during the synthesis, a potential of – 0.6 V (PEDOT:PSS) or – 0.5 V (p(g2T-TT)) vs. Ag/AgCl

was applied to discharge the polymers before monitoring the ORR in open-circuit voltage (OCV)

conditions. As shown in Figure 3a, PEDOT:PSS oxidises rapidly in oxygen-containing aqueous

electrolytes and achieves a high degree of charging in less than 10 min, including bipolaron formation.

A spontaneous ORR was also observed for p(g2T-TT), albeit at a slower charging rate and to a lower

degree of charging as compared to PEDOT:PSS (Section 15, Supporting Information). In comparison,

p(gPyDPP-MeOT2) shows no spectral changes, as shown in Figure 3b, demonstrating that the

copolymer does not undergo ORR in ambient conditions. In order to probe how the oxygen

concentration affects the charging rates during ORR with the polymer, we performed additional

measurements in aqueous electrolytes with reduced oxygen concentration and observed slower

charging rates for both PEDOT:PSS and p(gT2-TT) (Figures S22c and S23c, Supporting Information).

That finding shows that de-doped thin films of PEDOT:PSS and p(g2T-TT) undergo spontaneous ORR .

Since the rate of the ORR highly depends on the pH, we lowered the pH value of the electrolyte and

observed faster charging as well as oxidation to a higher degree of charging for p(g2T-TT) (Figure S25a,

Supporting Information) as expected, while p(gPyDPP-MeOT2) does not become oxidised at low pH

(Figure S25d, Supporting Information). This result emphasises the low activity of p(gPyDPP-MeOT2)

towards ORR.

10

We hypothesise that the IPs of the polymers, and therefore their oxidation potentials, determine

whether or not they oxidise under ambient conditions. To show this relationship, we measured the

oxidation potentials of the polymers by CV in aqueous electrolytes and related them to the potential

recorded for ORR on a platinum (Pt) electrode, which is known to be an efficient catalyst for ORR. As

shown in Figure 3c, the ORR on Pt electrodes occurs at potentials < 0 V vs. Ag/AgCl. PEDOT:PSS has

the lowest oxidation potential of the polymer series (– 0.8 V vs Ag/AgCl), followed by p(g2T-TT)

(– 0.2 V vs. Ag/AgCl) and p(gPyDPP-MeOT2) (0.3 V vs. Ag/AgCl). The large overlap of the cyclic

voltammograms measured for PEDOT:PSS and p(g2T-TT) compared to the one measured for ORR on

the Pt electrode suggests that an electron-transfer to oxygen is more likely for these polymers than

for p(gPyDPP-MeOT2), which shows no or little overlap. A similar trend was also reported for chemical

doping of organic semiconductors, showing that the IP of conjugated polymers determines how many

charge carriers are transferred from a polymer to the dopant.[31] For the chemical doping of DPP-based

donor-acceptor copolymers with IP > 5.2 eV, specially designed dopants with large electron affinities

(EA) are required to achieve a high degree of charging.[32] To properly quantify the relative likelihood

of the oxidation reactions of the polymers, the thermodynamic driving forces could be worked out,

e.g., using DFT calculations[33] if the precise reaction mechanism were known. While the polyanion

PSS– in PEDOT:PSS might affect the rate of oxidation, since it is known to be a good proton

conductor[34], we believe that the rapid oxidation of PEDOT:PSS arises from its high-lying HOMO.

To verify whether the ORR predominantly yield H2O2 (two-electron process) or H2O (four-electron

process), we carried out rotating ring-disk electrode (RRDE) measurements[35] in oxygen-containing

aqueous electrolytes (Section 19, Supporting Information). The RRDE experiment enables to detect

redox-active species at the ring electrode, that are formed during the ORR at the polymer-coated disk

electrode. The measurements indicate that, at potentials < – 0.2 V vs. Ag/AgCl, both PEDOT:PSS and

p(g2T-TT) predominantly form H2O2 (and water) while p(gPyDPP-MeOT2) would form mostly H2O

(Figure S35, Supporting Information).To further illustrate that the ORR yields H2O2, we employed an

enzymatic reaction, that is highly selective for the detection of H2O2[36] and tested the electrolytes that

11

are used for charging and discharging of the polymers for H2O2. We observed H2O2 formation for

PEDOT:PSS and p(g2T-TT)), whereas no H2O2 formation is observed for p(gPyDPP-MeOT2)

(Figures S37-S38, Supporting Information).

To show the impact of the ORR and H2O2 formation during OECT operation, we recorded the transfer

curves of OECTs of PEDOT:PSS, p(g2T-TT) and p(gPyDPP-MeOT2) and compared the OFF currents of

the devices (Figure 3d). The OFF current of p(gPyDPP-MeOT2) is two orders of magnitude lower than

that of PEDOT:PSS or p(g2T-TT). The OFF currents of PEDOT:PSS and p(g2T-TT) can be decreased by

lowering the oxygen concentration in the electrolyte (decreasing the ORR as shown Figure S32,

Supporting Information) which also lowers the gate current of the OECTs (Figure S33, Supporting

Information). To quantify how much H2O2 is formed, we measure the gate current during pulsed

cycling experiments while varying VD, VG and ΔVG (Section 17, Supporting Information). For PEDOT:PSS

and p(g2T-TT), during continuous device operation under VG > 0.2 V, we observed an increase of the

injected charge per cycle (obtained by integrating the area-normalised gate current transient response

for each gate pulse) proportional to VD. We also monitored the drain current and observed largest

device degradation at potentials at which polymers are in their discharged, low conductivity state

(Figures S26-S28, Supporting Information). In comparison, the change in injected charge per cycle for

p(gPyDPP-MeOT2) is marginal, as expected, due to the absence of ORR during device operation

(Figure S31, Supporting Information). The findings highlight that H2O2 formation needs to be avoided

to avoid device degradation during long-term measurements.

Finally, we carried out OECT measurements in aqueous electrolytes with high and low oxygen

concentrations (Figure 4) and monitored changes of the OFF current in OECTs at OCV conditions after

de-doping the polymer by applying a gate potential. For PEDOT:PSS and p(g2T-TT), a spontaneous

turn-on of the device is observed with a faster rise of ID for PEDOT:PSS compared to p(g2T-TT) (Figures

4a-b), while p(PyDPP-MeOT2) remains in its low conductive state (Figure 4c). This further supports the

12

hypothesis that p(PyDPP-MeOT2) does not become oxidised in ambient conditions and hence does

not form H2O2 during device operation.

The findings highlight the importance of designing redox-active polymers that have a high

performance in OECTs while keeping non-capacitive faradaic side-reactions to a minimum to avoid the

formation of reactive side-products. The findings are also relevant for other applications including

energy storage devices, non-volatile memories or sensor devices where undesired oxidation of the

active materials and H2O2 formation can affect the retention of the charge[8], modify the charge of a

memory state[37], increase device degradation or possibly interfere with the sensing mechanism of

biosensors.

Conclusions

In this work, we elucidate electrochemical side-reactions of state-of-the-art p-type OECT materials

and show that the materials can undergo an ORR during device operation and form the detrimental

side-product H2O2. We explore polymer backbones that result in formation of organic semiconductors

with large IPs to avoid ORRs. We show that the engineering of the polymer backbone is important for

achieving high electrochemical redox stability in aqueous electrolytes and demonstrate that donor-

acceptor polymers are an interesting class of materials for the field of bioelectronics. Although the

copolymer displays a lower electronic charge carrier mobility in OECTs compared to state-of-the-art

polymers, we believe that the presented chemical design strategy is a viable route for developing next-

generation bioelectronic materials, especially for real-life applications where hazardous side-products

will need to be avoided and low OFF currents of devices are desired.

Acknowledgements

We thank Eric Daniel Głowacki for fruitful discussions about oxygen reduction reactions. A.G, J.N., and

I.M acknowledge funding from EPSRC project EP/G037515/1, EP/N509486/1; D.M and J.N. are grateful

for receiving funding from the Supersolar Hub (EP/P02484X/1). A.G. and A.S. acknowledge funding

from the TomKat Center for Sustainable Energy at Stanford University. Part of this work was

13

performed at the Stanford Nanofabrication Facilities (SNF) and Stanford Nano Shared Facilities (SNSF),

supported by the National Science Foundation as part of the National Nanotechnology Coordinated

Infrastructure under award ECCS-1542152. The project has also received funding from the European

Research Council (ERC) under the European Union's Horizon 2020 research and innovation program

(grant agreement No 742708). J.R. and B.P acknowledge funding from the National Science

Foundation (grant no. NSF DMR-1751308). This work utilised Northwestern University Micro/Nano

Fabrication Facility (NUFAB) and the Keck-II facility of NUANCE Center, which is partially supported by

Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF ECCS-1542205), the Materials

Research Science and Engineering Center (DMR-1720139) at the Materials Research Center, the State

of Illinois, and Northwestern University. NUANCE is further supported by the International Institute

for Nanotechnology (IIN); and the Keck Foundation. Measurements at Stanford Synchrotron Radiation

Lightsource, SLAC National Accelerator Laboratory, were supported by the U.S. Department of Energy,

Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515.

Contributions

A.G., I.M., and J.R. conceived the project. A.G. and M.M. synthesised the materials. R.B.R, B.P. and

Q.T. fabricated OECTs, R.B.R, B.P. and A.G. tested devices in inert and ambient conditions and

analysed the data. A.G. and D.M. performed electrochemical and spectroscopic measurements and

analysed data. C.C and D.H. recorded and analysed the GIWAXS data. D.W. recorded PESA

measurements. J.T.M. and A.G. carried out the RRDE measurements and analysed the data. K.T

carried out DFT calculations, A.G. and J.R. wrote the manuscript. All the authors contributed to the

discussion and manuscript preparation.

Competing Interests statement

The authors declare no competing interests.

14

References

[1] B. D. Paulsen, K. Tybrandt, E. Stavrinidou, J. Rivnay, Nat. Mater. 2019, DOI 10.1038/s41563-

019-0435-z.

[2] C. B. Nielsen, A. Giovannitti, D.-T. Sbircea, E. Bandiello, M. R. Niazi, D. A. Hanifi, M. Sessolo, A.

Amassian, G. G. Malliaras, J. Rivnay, I. McCulloch, J. Am. Chem. Soc. 2016, 138, 10252.

[3] A. Giovannitti, D.-T. Sbircea, S. Inal, C. B. Nielsen, E. Bandiello, D. A. Hanifi, M. Sessolo, G. G.

Malliaras, I. McCulloch, J. Rivnay, Proc. Natl. Acad. Sci. 2016, 113, 12017.

[4] A. Giovannitti, C. B. Nielsen, D.-T. Sbircea, S. Inal, M. Donahue, M. R. Niazi, D. A. Hanifi, A.

Amassian, G. G. Malliaras, J. Rivnay, I. McCulloch, Nat. Commun. 2016, 7, 13066.

[5] E. Zeglio, M. M. Schmidt, M. Thelakkat, R. Gabrielsson, N. Solin, O. Inganäs, Chem. Mater.

2017, 29, 4293.

[6] L. R. Savagian, A. M. Österholm, J. F. Ponder, K. J. Barth, J. Rivnay, J. R. Reynolds, Adv. Mater.

n.d., 0, 1804647.

[7] C. G. Bischak, L. Q. Flagg, K. Yan, C.-Z. Li, D. S. Ginger, ACS Appl. Mater. Interfaces 2019, 11,

28138.

[8] D. Moia, A. Giovannitti, A. A. Szumska, I. P. Maria, E. Rezasoltani, M. Sachs, M. Schnurr, P. R.

F. Barnes, I. McCulloch, J. Nelson, Energy Environ. Sci. 2019, 12, 1349.

[9] G. D. Spyropoulos, J. N. Gelinas, D. Khodagholy, Sci. Adv. 2019, 5, eaau7378.

[10] M. J. Donahue, A. Williamson, X. Strakosas, J. T. Friedlein, R. R. McLeod, H. Gleskova, G. G.

Malliaras, Adv. Mater. 2018, 30, 1705031.

[11] V. Venkatraman, J. T. Friedlein, A. Giovannitti, I. P. Maria, I. McCulloch, R. R. McLeod, J.

Rivnay, Adv. Sci. 2018, 5, 1800453.

15

[12] J. Rivnay, S. Inal, A. Salleo, R. M. Owens, M. Berggren, G. G. Malliaras, Nat. Rev. Mater. 2018,

3, 17086.

[13] L. Guan, L. Yu, G. Z. Chen, Electrochim. Acta 2016, 206, 464.

[14] H. Klauk, U. Zschieschang, J. Pflaum, M. Halik, Nature 2007, 445, 745.

[15] S. Inal, G. G. Malliaras, J. Rivnay, Nat. Commun. 2017, 8, 1767.

[16] D. Khodagholy, J. Rivnay, M. Sessolo, M. Gurfinkel, P. Leleux, L. H. Jimison, E. Stavrinidou, T.

Herve, S. Sanaur, R. M. Owens, G. G. Malliaras, Nat. Commun. 2013, 4, 2133.

[17] H. Sun, M. Vagin, S. Wang, X. Crispin, R. Forchheimer, M. Berggren, S. Fabiano, Adv. Mater.

2018, 30, 1704916.

[18] A. Giovannitti, K. J. Thorley, C. B. Nielsen, J. Li, M. J. Donahue, G. G. Malliaras, J. Rivnay, I.

McCulloch, Adv. Funct. Mater. 2018, 28, 1706325.

[19] E. Mitraka, M. Gryszel, M. Vagin, M. J. Jafari, A. Singh, M. Warczak, M. Mitrakas, M. Berggren,

T. Ederth, I. Zozoulenko, X. Crispin, E. D. Głowacki, Adv. Sustain. Syst. 2019, 3, 1800110.

[20] R. Kerr, C. Pozo-Gonzalo, M. Forsyth, B. Winther-Jensen, ECS Electrochem. Lett. 2013, 2, F29.

[21] D.-H. Chin, G. Chiericato, E. J. Nanni, D. T. Sawyer, B. H. J Photochem Photobiol, Oxygen and

Oxy-Radicals in Chem-Istry and Biology, Academic Press, 1982.

[22] K. Apel, H. Hirt, Annu. Rev. Plant Biol. 2004, 55, 373.

[23] R. L. Auten, J. M. Davis, Pediatr. Res. 2009, 66, 121.

[24] V. J. Watt BE, Proudfoot AT, Toxicol Rev 2004, 23, 51.

[25] C. Enengl, S. Enengl, S. Pluczyk, M. Havlicek, M. Lapkowski, H. Neugebauer, E. Ehrenfreund,

ChemPhysChem 2016, 17, 3836.

[26] M. Akbayrak, A. M. Önal, Polym. Chem. 2016, 7, 6110.

16

[27] C. Francis, ab D. Fazzi, S. B. Grimm, F. Paulus, S. Beck, de S. Hillebrandt, de A. Pucci de, J.

Zaumseil, J. Mater. Chem. C 2017, 5, 6176.

[28] J. Rivnay, M. Ramuz, P. Leleux, A. Hama, M. Huerta, R. M. Owens, Appl. Phys. Lett. 2015, 106,

43301.

[29] X. Zhang, C. Xiao, A. Zhang, F. Yang, H. Dong, Z. Wang, X. Zhan, W. Li, W. Hu, Polym. Chem.

2015, 6, 4775.

[30] D. Khodagholy, T. Doublet, P. Quilichini, M. Gurfinkel, P. Leleux, A. Ghestem, E. Ismailova, T.

Hervé, S. Sanaur, C. Bernard, G. G. Malliaras, Nat. Commun. 2013, 4, 1575.

[31] D. Kiefer, R. Kroon, A. I. Hofmann, H. Sun, X. Liu, A. Giovannitti, D. Stegerer, A. Cano, J.

Hynynen, L. Yu, Y. Zhang, D. Nai, T. F. Harrelson, M. Sommer, A. J. Moulé, M. Kemerink, S. R.

Marder, I. McCulloch, M. Fahlman, S. Fabiano, C. Müller, Nat. Mater. 2019, 18, 149.

[32] Y. Karpov, T. Erdmann, I. Raguzin, M. Al-Hussein, M. Binner, U. Lappan, M. Stamm, K. L.

Gerasimov, T. Beryozkina, V. Bakulev, D. V Anokhin, D. A. Ivanov, F. Günther, S. Gemming, G.

Seifert, B. Voit, R. Di Pietro, A. Kiriy, Adv. Mater. 2016, 28, 6003.

[33] S. Siahrostami, A. Verdaguer-Casadevall, M. Karamad, D. Deiana, P. Malacrida, B. Wickman,

M. Escudero-Escribano, E. A. Paoli, R. Frydendal, T. W. Hansen, I. Chorkendorff, I. E. L.

Stephens, J. Rossmeisl, Nat. Mater. 2013, 12, 1137.

[34] Y.-S. Ye, J. Rick, B.-J. Hwang, Y.-S. Ye, J. Rick, B.-J. Hwang, Polymers (Basel). 2012, 4, 913.

[35] N. A. Anastasijević, V. Vesović, R. R. Adžić, J. Electroanal. Chem. 1987, 229, 305.

[36] P. D. Josephy, T. Eling, R. P. Mason, J. Biol. Chem. 1982, 257, 3669.

[37] S. T. Keene, A. Melianas, Y. van de Burgt, A. Salleo, Adv. Electron. Mater. 2019, 5, 1800686.

17

Supporting Information

The supporting information contains the synthesis and characterisation of p(gPyDPP-MeOT2) and

p(gPyDPP-T2), electrochemical characterisation, spectroelectrochemical measurements, DFT

calculations, GIWAXS measurements, FT-IR and XPS measurements, OECT measurements, RRDE

measurements and enzymatic detection of hydrogen peroxide.

18

Figures

Figure 1. Characterisation of the copolymers. a) Chemical structure of the copolymers p(gPyDPP-T2)

and p(gPyDPP-MeOT2). b) CV measurements of p(gPyDPP-MeOT2) deposited on ITO-coated glass

substrates with a scan rate (ν) of 50 mV/s in 0.1 M NaCl vs. Ag/AgCl, showing 50 scans. c)

Spectroelectrochemical measurements of p(gPyDPP-MeOT2) during the charging of the film with

voltage steps of 0.05 V from 0.3 V to 0.7 V vs. Ag/AgCl. d) Electrochemical redox stability of

p(gPyDPP-T2) and p(gPyDPP-MeOT2) during charging of the polymer to 1.0 V vs. Ag/AgCl. Additional

information is provided in section 9 of the Supporting Information.

19

Figure 2. OECT performance of p(gPyDPP-MeOT2) in 0.1 M NaCl aqueous electrolyte in ambient

conditions with W = 100 μm, L = 10 μm, and d = 60 nm. a) Output curve (VG = 0 V to – 0.7 V,

ΔVG = 0.05 V, ν = 0.1 V s-1), b) transfer curve (VD = – 0.4 V, ν = 0.1 V s-1) and c) stability pulsing

experiment by applying alternating the gate potentials between VG = 0 V and (i) VG = -0.5 V, (ii)

VG = – 0.6 V or (iii) VG = – 0.7 V (with VD = – 0.4 V) with a pulse duration of 2 s. The device was operated

for 25 minutes, where ID is highlighted at the beginning and the end of the experiment.

20

Figure 3. Spectroelectrochemical and electrochemical measurements of the polymer series.

Monitoring changes of the absorption spectrum (oxidation of the polymer due to ORRs) for a)

PEDOT:PSS after discharging the polymer at – 0.6 V vs. Ag/AgCl for 100 s and switching to OCV

conditions and b) p(gPyDPP-MeOT2) after applying 0 V vs. Ag/AgCl for 100 s and switching to OCV

conditions (note, the polymer has no redox states at 0 V as shown in Figures 1b-c). c) Cyclic

voltammograms of PEDOT:PSS, p(g2T-TT), p(gPyDPP-MeOT2) using a rotating-disk electrode rotating

at 1600 RPM at low oxygen concentration (see Figure S34b (Supporting Information) for CV

measurement performed at saturated oxygen concentration) and monitoring the ORR at a rotating

Pt ring-electrode at saturated oxygen concentration in 0.1 M NaCl aqueous solution vs. Ag/AgCl

(ν = 5 mV s-1). d) Transfer curve of OECTs showing PEDOT:PSS (VD = – 0.1 V, ν = 1 V s-1), p(g2T-T)

(VD = – 0.3 V, ν = 1 V s-1), p(gPyDPPMeOT2) (VD = – 0.2 V, ν = 1 V s-1) with 0.1 M NaCl in ambient

conditions. W = 100 μm and L = 10 μm for all devices.

21

Figure 4. OECT performance in ambient and inert conditions. a) OECT performance of PEDOT:PSS,

applying gate potentials of VG = 0 V or VG = 0.7 V for 60 s (with VD = – 0.5 V) before switching to OCV

in inert conditions (black line) and ambient conditions (red line). b) OECT performance of pg2T-TT,

applying gate potentials of VG = – 0.4 V or VG = 0.4 V for 60 s (with VD = – 0.5 V) before switching to

OCV in inert conditions (black line) and ambient conditions (red line). c) OECT performance of

p(gPyDPP-MeOT2) in ambient conditions, applying gate potentials of VG = – 0.7 V for 60 s or VG = 0 V

(with VD = – 0.2 V) before switching to OCV in ambient conditions.

download fileview on ChemRxivmanuscript_191208.pdf (716.69 KiB)

1

Supporting information Energetic control of redox-active polymers towards safe organic bioelectronic materials

Alexander Giovannitti*, Reem B. Rashid, Quentin Thiburce, Bryan Paulsen, Camila Cendra, Karl Thorley, Davide Moia, J. Tyler Mefford, David Hanifi, Du Weiyuan, Max Moser, Alberto Salleo, Jenny Nelson, Iain McCulloch, and Jonathan Rivnay Dr. Alexander Giovannitti, Dr. Karl Thorley, Max Moser, Prof. Iain McCulloch Department of Chemistry, Imperial College London, London, SW7 2AZ, UK. Corresponding author: ag19@stanford.edu Dr. Alexander Giovannitti, Dr. Davide Moia, Prof. Jenny Nelson Department of Physics, Imperial College London, London, SW7 2AZ, UK Dr. Du Weiyuan, Prof. Iain McCulloch King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering Division, KAUST Solar Center (KSC), Thuwal 23955-6900, Saudi Arabia Reem R. Rashid, Dr Bryan Paulsen SJ, Prof. Jonathan Rivnay Simpson Querrey Institute, Northwestern University, Chicago, Illinois 60611, United States Department of Biomedical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States Dr. Alexander Giovannitti, Camila Cendra, Dr. Quentin Thiburce, David Hanifi, Dr. J. Tyler Mefford and Prof. Alberto Salleo Department of Materials Science and Engineering, Stanford University, CA 94305, USA

2

Table of Contents Materials and methods ........................................................................................................... 4

Materials ............................................................................................................................................. 4

MALDI .................................................................................................................................................. 4

Cyclic voltammetry ............................................................................................................................. 4

OECT fabrication and electrical measurement ................................................................................... 4

OECT measurements at high and low O2 concentration .................................................................... 5

Spectroelectrochemical measurements (UV Vis and FT-IR) ............................................................... 5

GIWAXS ............................................................................................................................................... 5

XPS measurements ............................................................................................................................. 6

Detection of hydrogen peroxide ......................................................................................................... 6

Synthesis overview .................................................................................................................. 7

. Monomer synthesis ............................................................................................................... 8

g6-PyDPP-Br2 ....................................................................................................................................... 8

g7-PyDPP-Br2 ..................................................................................................................................... 10

. Polymer synthesis................................................................................................................ 12

p(gPyDPP-T2) .................................................................................................................................... 12

p(gPyDPP-MeOT2) ............................................................................................................................ 13

. Properties of the polymers ................................................................................................. 14

. DFT calculations................................................................................................................... 15

. Grazing incidence wide angle scattering (GIWAXS) ............................................................ 16

. CV and spectroelectrochemical measurements in aqueous electrolytes ........................... 17

. Electrochemical stability of the polymers during oxidation in aqueous electrolytes ....... 18

p(gPyDPP-T2) .................................................................................................................................... 18

p(gPyDPP-MeOT2) ............................................................................................................................ 19

. FT-IR spectroscopy and analysis of the decomposition products during electrochemical oxidation ............................................................................................................................................... 20

p(gPyDPP-T2) .................................................................................................................................... 20

p(gPyDPP-MeOT2) ............................................................................................................................ 21

. XPS analysis of the polymer degradation .......................................................................... 22

p(gPyDPP-T2) .................................................................................................................................... 22

p(gPyDPP-MeOT2) ............................................................................................................................ 23

. OECT performance of polymer p(gPyDPP-T2) ................................................................... 24

. Capacitance measurements of p(gPyDPP-MeOT2) ........................................................... 25

3

Quantifying ORR of the polymers at low and high oxygen concentration ......................... 26

PEDOT:PSS ......................................................................................................................................... 26

p(g2T-TT) ........................................................................................................................................... 27

p(gPyDPP-MeOT2) ............................................................................................................................ 28

. Redox-reactions in aqueous electrolytes at different pH values ...................................... 29

OECT pulsed cycling measurements ................................................................................... 30

PEDOT:PSS (VD = – 0.1 V) ................................................................................................................... 31

PEDOT:PSS (VD = – 0.5 V) ................................................................................................................... 32

p(g2T-TT) ........................................................................................................................................... 33

p(g2T-TT) (VD = -0.3 V) ON to OFF ..................................................................................................... 33

p(g2T-TT) (VD = – 0.1 V) OFF to ON ................................................................................................... 34

p(g2T-TT) (VD = – 0.5 V) ..................................................................................................................... 35

p(gPyDPP-MeOT2) (VD = – 0.2 V) ...................................................................................................... 36

Performance of OECTs varying the oxygen concentration in the electrolyte ..................... 37

Monitoring the gate current with electrolytes containing low or high oxygen concentration ........ 37

RRDE measurements ........................................................................................................... 38

. Detection of hydrogen peroxide ....................................................................................... 40

. References ......................................................................................................................... 42

4

Materials and methods Materials Polymer p(g2T-TT) was synthesised according to the literature1. Indium tin oxide (ITO) coated glass substrates (70-100 Ω/sq), peroxidase horseradish and 3,3′,5,5′-tetramethylbenzidine were purchased from Merck Sigma-Aldrich. PEDOT:PSS was purchased from Heraeus Clevios PH1000. The polymers p(gPyDPP-MeOT2) and p(gPyDPP-T2) were synthesised as described in the supplementary section2-4.

MALDI Matrix-assisted laser desorption ionization time-of-flight (MALDI TOF) spectrometry was carried out on a Micromass MALDImxTOF with trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene]-malononitrile (DCTB) as the matrix as described previously2.

Cyclic voltammetry Electrochemical characterisation was carried out using an Ivium CompactStat potentiostat with a three-electrode setup with either 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) acetonitrile solution or 0.1 M NaCl aqueous electrolyte. The polymer thin films were employed as the working electrode with a Ag/AgCl 3 M NaCl reference electrode and a platinum-mesh counter electrode (active area 25 x 35 mm). The electrolyte was purged with argon for at least 15 minutes before carrying out the reduction reactions. IPs were calculated according to the literature: IP [eV] = E(Ox,onset vs Fc/Fc+)+5.13.

OECT fabrication and electrical measurement OECTs were fabricated using methods described previously but will be briefly mentioned below1,4. S1813 photoresist (Shipley) was spun onto a glass wafer at 3500 rpm for 30 s and baked at 110 °C for 1 min, followed by UV exposure using a SUSS MJB4 mask aligner and developed in AZ400 K (1:4 dilution). Metal contacts and interconnects of 5 nm chromium/100 nm gold were deposited using an AJA E-beam system. Metal lift off was performed in Microposit 1165 (MicroChem). Two ~1.5um layers of Paralyene C were then deposited (LabCoaterII) with a thin layer of a 2% soap solution (Micro90) cast in between. To simultaneously pattern the insulating Parylene C and active areas, AZ P4620 photoresist was then spun on at 3000 rpm for 1 min and baked at 110 °C for 2 min. It was then exposed to using the mask aligner and developed before etching with a RAMCO reactive ion etcher at 160 W with 50 sccm of O2 and 10 sccm of CHF3 for ~25 min. Lastly, the polymers p(gPyDPP-T2) and p(gPyDPP-MeOT2) were spun onto the OECT at 1000rpm for 1 min and peeled off after drying (no further annealing or cross-linking was carried out). OECTs with PEDOT:PSS and p(g2T-TT) were prepared according to the literature5,6. All electrical OECT Measurements were performed on a National Instruments PXIe-1082 system using custom made LabView Programs as previously described2. To obtain output and transfer curves 2 NI PXIe-4143 source measuring units were used. To obtain stability measurements a train of square pulses was applied to the gate using the NI PIXe-6363 DAQ while current was recorded using the NI PXIe-4143. Transconductance as a function of frequency was obtained in similar manner but with a small amplitude sine wave (10 mV) applied at the gate while recording current using the NI PXIe-4081 digital multimeters. All measurements were taken using an external Ag/AgCl pellet electrode as the gate in a 0.1 M NaCl solution. Transconductance was calculated using a custom Matlab script. Mobility was calculated by first extracting hole transit time by matching gate and drain current derived impedance measurements

5

using custom a Matlab script, a method described previously. All thickness measurements were taken with a Veeco Dektak-8 profilometer. Impedance, phase shift and effective capacitance data was generated using electrochemical impedance spectroscopy (EIS) (AutoLab). Ag/AgCl reference electrode (BASi), and Pt mesh were used as reference electrode and counter electrode, respectively, and a patterned Au electrode with p(gPyDPP-MeOT2) coating was used as the working electrode (fabricated with the same methods described above). EIS was performed in a 0.1 M NaCl solution with a 10 mV sine wave, with frequencies from 0.1 Hz to 100 kHz, with working electrode offsets between 0V to 0.7 V, as noted.

OECT measurements at high and low O2 concentration OECTs with channel dimensions of W = 100 µm and L = 10 µm were operated in an enclosed electrochemical cell (Redox.me) with a customized adjustable position Ag/AgCl gate electrode filled with ~10 mL of a 0.1 M NaCl aqeuoues solution. The cell was purged with N2 for ~16 hours to create a low O2 environment. After purging, the OECT was switched from an equilibrated ON state (VG = 0 V and VD = -0.5 V for PEDOT:PSS, VG = -0.4 V and VD = -0.4 V for pg(2T-TT), rendering both devices in the saturation regime) to a equilibrated OFF state (VG = 0.7 V for PEDOT:PSS, VG = 0.4 V for pg(2T-TT), drain voltages remain unchanged). After the OFF state equilibrated for 60 seconds, the Ag/AgCl gate was physically removed from the electrolyte and electrically disconnected. The channel current was then monitored for the 400s to indirectly monitor chemical oxidation of the film that will increase the conductance of the channel. The procedure was repeated on the same devices in an oxygen rich environment following ~8 hours of sparing with air.

Spectroelectrochemical measurements (UV Vis and FT-IR) The electrochemical measurements were carried out as described in the literatur2. The polymer thin films were immersed in a quartz cuvette filled with a 0.1 M NaCl solution and electrochemical measurements were performed while recording the UV Vis spectrum with an OceanOptics USB 2000+ spectrometer. The electrolyte was degassed with argon or nitrogen for at least 30 min for air-sensitive measurements. FT-IR measurements were recorded on a PerkinElmer Spectrum 100 FT-IR spectrometer with an ATR sampling unit. The neat polymers were dissolved in chloroform and drop cast on the ATR sampling unit and dried for 10 mins prior recording of the spectrum. For the analysis of the decomposition product, polymer p(gPyDPP-MeOT2) or p(gPyDPP-T2) were drop cast on ITO coated glass substrates. The polymer thin-films were electrochemically charged by applying a constant potential for 100 s [1.2 V for p(gPyDPP-MeOT2) and 1.0 V for p(gPyDPP-T2)], followed by transferring the polymer thin-films onto the ATR sampling unit.

GIWAXS Grazing incidence wide angle X-ray scattering (GIWAXS) was performed at the Stanford Synchrotron Radiation Lightsource (SSRL) on beam line 11-3 using an area detector (Rayonix MAR-225) and incident energy of 12.73 keV. The distance between sample and detector was calibrated using a LaB6 polycrystalline standard. The incidence angle (0.1°) was slightly larger than the critical angle, ensuring that we sampled the full depth of the film. X-ray measurements were performed in Helium environment to minimize air scattering and beam damage to samples. Raw data was reduced and analysed using a combination of Nika 1D SAXS7 and WAXStools8 software packages in Igor Pro.

6

XPS measurements XPS measurements were carried out on PHI Versaprobe 3 on Si-substrates coated with Cr (5 nm) and Au (35 nm). The samples were cleaned by sonication in acetone and IPA for 15 min before deposition of the polymers (5mg/mL) by spin coating (1000 RPM, 1 min). Cyclic voltammetry measurements (three electrode setup using a Ag/AgCl electrode as the reference electrode and a Pt mesh as the counter electrode, 0.1 M NaCl aqeuoues electrolyte) were carried out between 0 V - 1.0 V for pg(PyDPP-T2) and 0 V - 0.7 V for pg(PyDPP-MeOT2) vs. Ag/AgCl prior applying the voltage pulses that result in degradation of the polymers. The degradation of the polymer was initiated by applying a constant potential of 1.1 V for p(gPyDPP-T2) and 1.0 V for (p(gPyDPP-MeOT2) vs. Ag/AgCl for 100 s. Finally, CV measurements were carried out to monitor changes of the charging/discharging profile of the polymers. The polymer thin-films were washed with DI water and dried before conducting the XPS measurements.

Detection of hydrogen peroxide Hydrogen peroxide detection was carried out using the peroxidase/dye system (peroxidase horseradish/3,3′,5,5′-tetramethylbenzidine) following a protocol reported in the literature9. To proof that H2O2 was formed during the chronoamperometry measurements (as described in section 19), we used the electrolyte as the solution and monitored colour changes of the dye for 10 min.

7

Synthesis overview The pyridine-DPP core was synthesized according to the literature.10 A hexakis ethylene glycol side chain (-(CH2CH2O)6Me) was attached to the gPyDPP core for p(gPyDPP-T2) and a heptakis ethylene glycol side chain (-(CH2CH2O)7Me) was attached for polymer p(gPyDPP-MeOT2) to increase the polymers solubility in organic solvents. Preparation of the

N

N

O

O

N

N

OMe

OMe

SS n

6

6

N

N

O

O

N

N

OMe

OMe

SS

O

O

Me

Me

n

7

7

N

N

O

O

N NBr

BrROTs

H

H

K2CO3DMF

18-C-6

N

N

O

O

N NBr

Br

OMe

OMem

m

Pd2(dba)3P(o-tol)3

DMFSS

X

Xn

N

N

O

O

N N

OMe

OMem

m

NBr NaN

Me O

O

O

OMe

m = 6 (35 %)m = 7 (29 %)

X = H; m = 6 (79 %)X = OMe; m = 7 (81 %)

p(gPyDPP-T2) p(gPyDPP-MeOT2)

OH

54 % R =O

Memm = 6 m = 7

Figure S1. Synthesis of the polymers p(gPyDPP-MeOT2) and p(gPyDPP-T2).

8

. Monomer synthesis g6-PyDPP-Br2 (3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17-hexaoxanona-decan-19-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione)

N

N

O

O

N NBr

Br

OMe

OMe

6

6

A 100 mL two neck RBF was dried and purged with argon. Anhydrous potassium carbonate (351 mg, 2.32 mmol, 3.0 eq.) and 3,6-bis(5-bromopyridin-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (380 mg, 0.85 mmol, 1.0 eq.) were suspended in 20 mL of anhydrous DMF. 2,5,8,11,14,17-hexaoxanonadecan-19-yl 4-methylbenzenesulfonate (976 mg, 2.13 mmol, 2.5 eq.) was added and the reaction mixture was heated to 120 °C for 3 h. The reaction mixture was cooled to room temperature and 100 mL of water was added. The product was extracted with ethyl acetate (3 x 150 mL) and dried over MgSO4. The solvent was removed under vacuo. and the product was isolated by columns chromatography on silica gel with a mixture of 98 % ethyl acetate and 2 % methanol. The solvent was removed under vacuo. and the product was dried at 50°C under high vacuum for 16 h. 301 mg of the monomer was obtained with a yield of 35 % (0.3 mmol).

1H NMR (400 MHz, CDCl3) σ: 8.91 – 8.89 (d, J = 8.6 Hz, 2H), 8.76 – 8.75 (d, J = 2.3 Hz, 2H), 8.03 – 8.00 (dd, J = 8.6, 2.4 Hz, 2H), 4.58 – 4.55 (t, J = 6.1 Hz, 4H), 3.68 – 3.65 (m, 4H), 3.64 – 3.63 (m, 8H), 3.63 – 3.61 (m, 12H), 3.61 – 3.58 (m, 4H), 3.56 – 3.55 (m, 12H), 3.49 – 3.47 (m, 4H), 3.37 (s, 6H) ppm.

13C NMR (100 MHz, CDCl3): 162.3, 150.4, 145.0, 139.9, 133.1, 128.6, 122.8, 111.6, 72.1, 70.7 – 70.6 (multiple signals), 70.4, 69.7, 59.1, 41.7 ppm.

HRMS (ES-ToF): 1003.2577 [M-H+] (calc. 1003.2551).

9

Figure S2. 1H NMR spectrum of 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17-hexaoxanonadecan-19-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione in CDCl3.

Figure S3. 13C NMR spectrum of 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17-hexaoxanonadecan-19-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione in CDCl3.

10

g7-PyDPP-Br2

(3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17,20-heptaoxadocosan-22-yl)-2,5-dihydropyrrolo-[3,4-c]pyrrole-1,4-dione)

N

N

O

O

N NBr

Br

OMe

OMe

7

7

A 100 mL two neck RBF was dried and purged with argon. Anhydrous potassium carbonate (860 mg, 6.2 mmol, 3.0 eq.) and 3,6-bis(5-bromopyridin-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (927 mg, 2.1 mmol, 1.0 eq.) was suspended in 20 mL of anhydrous DMF. 2,5,8,11,14,17,20-heptaoxadocosan-22-yl 4-methylbenzenesulfonate (2.15 g, 4.4 mmol, 2.1 eq.) was added and the reaction mixture was heated to 120 °C for 3 h. The reaction mixture was cooled to room temperature and 100 mL of water was added. The product was extracted with ethyl acetate (4 x 150 mL) and dried over MgSO4. The solvent was removed under vacuo. and column chromatography on silica gel with ethyl acetate and 10 % methanol was carried out to isolate the product (Rf = 0.26). The solvent was removed under reduced pressure and the product was dried under high vacuum for 16 h at 50°C. 650 mg of a red semi-crystalline solid was obtained with a yield of 29 % (0.61 mmol).

1H NMR (400 MHz, CDCl3) σ: 8.91 – 8.88 (d, J = 8.6 Hz, 2H), 8.76 – 8.75 (d, J = 2.2 Hz, 2H), 8.01 (dd, J = 8.6, 2.4 Hz, 2H), 4.56 (t, J = 6.1 Hz, 4H), 3.66 (t, J = 6.1 Hz, 4H), 3.64 – 3.60 (m, 28H), 3.58 – 3.52 (m, 16H), 3.49 – 3.46 (m, 4H), 3.36 (s, 6H) ppm.

13C NMR (100 MHz, CDCl3): 162.3, 150.4, 146.0, 145.0, 139.9, 128.5, 122.8, 111.6, 72.1, 70.7 – 70.6 (multiple signals), 70.4, 69.7, 59.2, 41.6 ppm.

HRMS (ES-ToF): 1110.3080 [M-H2O+] (calc. 1110.3082).

11

Figure S4. 1H NMR spectrum of 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17,20-heptaoxadocosan-22-yl)-2,5-dihydropyrrolo-[3,4-c]pyrrole-1,4-dione in CDCl3.

Figure S5.13C NMR spectrum of 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17,20-heptaoxadocosan-22-yl)-2,5-dihydropyrrolo-[3,4-c]pyrrole-1,4-dione in CDCl3

12

. Polymer synthesis p(gPyDPP-T2) In a 5 mL vial, 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17-hexaoxanonadecan-19-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (62.88 mg, 62.58 µmol), 5,5'-bis(trimethylstannyl)-2,2'-bithiophene (30.78 mg, 62.58 µmol), Pd2(dba)3 (1.15 mg, 1.25 μmol) and P(o-tol)3 (1.52 mg, 5.00 μmol) were dissolved in 3.0 mL of anhydrous, degassed DMF. The reaction mixture was degassed for 15 min and then heated to 125 °C for 16 h. For the end-capping of the polymer, 0.1 mL of a solution made of 0.1 mL 2-(tributylstannyl)thiophene in 0.5 mL of chlorobenzene and 1.0 mg of Pd2(dba)3 was added and heated for 1 h to 135 °C, then 0.1 mL of a solution made of 0.1 mL 2-bromothiophene in 0.5 mL of chlorobenzene was added and heated for 1 h to 135 °C. The reaction mixture was cooled to room temperature and the blue solution was precipitated in ethyl acetate followed by addition of hexane. The suspension was filtered and transferred into a glass thimble. Soxhlet extraction was carried out with ethyl acetate, methanol, hexane, acetone and chloroform. The polymer dissolved in hot chloroform. Polymer p(gPyDPP-T2) was obtained as a dark blue solid with a yield of 79 % (50.1 mg, 49.6 µmol).

Figure S6. 1H NMR spectrum of p(gPy-DPP-T2) in CDCl3 measured at 55 °C.

Figure S7. MALDI spectrum of p(gPyDPP-T2).

13

p(gPyDPP-MeOT2) In a 5 mL vial, 3,6-bis(5-bromopyridin-2-yl)-2,5-di(2,5,8,11,14,17,20-heptaoxadocosan-22-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (98.19 mg, 89.9 µmol), (3,3'-dimethoxy-[2,2'-bithiophene]-5,5'-diyl)bis(trimethylstannane) (49.59 mg, 89.9 µmol), Pd2(dba)3 (1.65 mg, 1.80 μmol) and P(o-tol)3 (2.19 mg, 7.19 μmol) were dissolved in 3.0 mL of anhydrous, degassed DMF. The reaction mixture was degassed for 15 min and then heated to 125 °C for 16 h. For end-capping of the polymer, 0.1 mL of a solution made of 0.1 mL 2-(tributylstannyl)thiophene in 0.5 mL of chlorobenzene and 1.0 mg of Pd2(dba)3 was added and heated for 1 h to 135 °C, then 0.1 mL of a solution made of 0.1 mL 2-bromothiophene in 0.5 mL of chlorobenzene was added and heated for 1 h to 135 °C. The reaction mixture was cooled to room temperature and the blue solution was precipitated in ethyl acetate followed by addition of hexane. The suspension was filtered in a glass thimble and soxhlet extraction was carried out with ethyl acetate, methanol, hexane, acetone, acetonitrile and chloroform. The polymer dissolved in hot chloroform. Polymer p(gPyDPP-MeOT2) was obtained as a dark blue solid with a yield of 81 % (84 mg, 72.6 µmol).

Figure S8. 1H NMR spectrum of p(gPyDPP-MeOT2) in CDCl3 measured at 55 °C.

Figure S9. MALDI spectrum of p(gPyDPP-MeOT2).

14

. Properties of the polymers

Figure S10. (a) Thin film UV Vis spectrum of p(gPyDPP-T2) (red line) and p(gPyDPP-MeOT2) (black line) and (b) cyclic voltammograms of thin films of p(gPyDPP-T2) (red line) and p(gPyDPP-MeOT2) (black line) deposit on ITO coated glass substrates, measured with a scan rate of 100 mV/s with in 0.1 M NBu4PF6 acetonitrile solution.

Table S1. Properties of the polymers

Polymer IP [eV] [A]

IP [eV] [B]

EA [eV][C] Bandgap onset [nm]

Optical band gap [eV]

p(gPyDPP-T2) 5.3 5.5 3.8 745 1.7 p(gPyDPP-MeOT2) 5.0 5.0 3.6 908 1.4

[A] The IP was measured by photoelectron spectroscopy in Air (PESA). [B] CV measurements were carried out with a 0.1 M NBu4PF6 acetonitrile solution with a scan rate of 100 mV/s. IP = 5.1 eV + Eox(onset) vs Fc/Fc+. [C] Electron affinity was calculated by subtracting the optical band gap from IP (note: the calculation of the EA neglects the electron binding energy)

15

. DFT calculations Single point calculations on representative oligomers using an IP-tuned ωB97XD functional at the 6-31G* basis set level. These calculations were used to obtain the orbital population for the highest occupied molecular orbital (HOMO) divided into fragments along the polymer chain (red = T2-Py, black = DPP-Py) (structure shown below). Note that populations under 0.1 are not reported in the Gaussian output, and as such the remaining population is divided equally between the remaining fragments. A variety of approaches to define the fragments were tested, ultimately providing qualitatively similar conclusions. Approaches included using one pyridine with the T2 group and one with the DPP group to provide the overall best comparison between the T2 and DPP units, while avoiding creation of too many individual fragments and therefore a larger number of unreported HOMO occupation data points. Using the same fragment definitions, the charge per fragment was calculated using Hirshfeld charges in the case where the oligomer bears a positive charge.

Figure S11. DFT calculations of (a) HOMO occupation per fragment of PyDPP-MeOT2, (b) HOMO occupation per fragment of PyDPP-T2, (c) charge per fragment of PyDPP-MeOT2 (polaron), (d) charge per fragment of PyDPP-T2 (polaron).

16

. Grazing incidence wide angle scattering (GIWAXS)

Figure S12. 2D GIWAXS patterns of (a) p(gPyDPP-T2) and (b) p(gPyDPP-MeOT2). In-plane (Qxy) and out-of-plane (Qz) scattering lineouts for (c) p(gPyDPP-T2) and (d) p(gPyDPP-MeOT2). Grey dashed lines denote the dominant peak positions for the lamellar spacing distance family of planes (h00) and the π-stacking repeat distance (010).

Table S2. Predominant lattice spacings extracted from in-plane and out-of-plane GIWAXS lineouts.

Polymer d(100) / Å d(010) / Å

p(gPyDPP-MeOT2) 19.19 3.49

p(gPyDPP-T2) 16.87 3.46

17

. CV and spectroelectrochemical measurements in aqueous electrolytes Thin films of p(gPyDPP-MeOT2) and p(gPyDPP-T2) were prepared by spin coated on ITO coated glass substrates from a chloroform solution (5mg/mL). The electrolyte for the CV measurements was a 0.1 NaCl aqueous solution where the electrolyte was degassed with argon for 15 min for the electrochemical reduction measurements. The reference electrode for all applied potentials was Ag/AgCl.

Figure S13. (a) CV measurements of p(gPyDPP-MeOT2) and p(gPyDPP-T2) in 0.1 M NaCl with a scan rate of 50 mV/s. (b) Spectroelectrochemical measurements for CV measurements shown in (a), plotting changes of the absorption spectrum (λmax) for the indicated potentials vs Ag/AgCl. Changes of the absorption spectrum of p(gPyDPP-MeOT2) during (c) oxidation and (e) reduction as well as for p(gPyDPP-T2) (d) oxidation and (f) reduction.

18

. Electrochemical stability of the polymers during oxidation in aqueous electrolytes p(gPyDPP-T2) The polymer p(gPyDPP-T2) was first charged to 1.0 V vs. Ag/AgCl (3 scans), followed by carrying out chronoamperometry measurements where the indicated potentials were applied for 100 s, followed by reversing the potentials to 0 V vs. Ag/AgCl. The measurement are carried out to detect changes of the absoprtion spectrum during charging of the polymer and hence finding out the potential at which the polymer degrades.

Figure S14. (a) CV measurements of p(gPyDPP-T2) between 0 V and 1.0 V vs. Ag/AgCl with a scan rate of 50 mV/s and (b) changes of the absoprtion spectrum of p(gPyDPP-T2) when applying a constant potential for 100 s between 0.8 V to 1.2 V (voltage step is 0.1 V), followed by discharing the polymer by applying 0 V vs. Ag/AgCl

19

p(gPyDPP-MeOT2)

Figure S15. (a) CV measurements of p(gPyDPP-MeOT2), applying potentials between 0 V to 0.7 V vs. Ag/AgCl with a scan rate of 50 mV/s, (b) CV measurements for applying potentials between 0 V to 1.0 V vs Ag/AgCl with a scan rate of 50 mV/s, (c) changes of the UV Vis spectrum during charging of the polymer to 0.7 V, highlighting the isosbestic point (polaron) (red line), (d) changes of the UV Vis spectrum during charging of the polymer to 1.0 V, isosbetic point (polaron (red)) and shift of the isospectic point (blue) when the potential is increase >0.7 V vs Ag/AgCl. Comparison of the initial spectrum and the spectrum after 50 scans when applying (e) 0.7 V and (f) 1.0 V.

20

. FT-IR spectroscopy and analysis of the decomposition products during electrochemical oxidation For the analysis of the decomposition products, polymer p(gPyDPP-MeOT2) or p(gPyDPP-T2) were drop cast on ITO coated glass substrates. CV measurements were carried out prior to electrochemical decomposition of the polymers for which a constant potential was applied for 100 s to degrade the polymers [1.2 V for p(gPyDPP-MeOT2) and 1.0 V for p(gPyDPP-T2)]. A CV measurement was carried out afterwards to demonstrate that the charging of the material is irreversible.

FT-IR spectroscopy was used to analyses the decomposition products where the polymer was scratched off from the substrates and measured with an FT-IR ATR unit.

p(gPyDPP-T2)

Figure S16. (a) FT-IR spectra of the pyridine-DPP monomer (g6-PyDPP-Br2) with a hexakis ethylene glycol side chain (blue), neat p(gPyDPP-T2) (black) and p(gPyDPP-T2) after applying 1.0 V vs Ag/AgCl for 100s (red). (b) zoom-in of the spectra shown in (a) (1800 and 650 cm-1) highlighting the C=O stretching peak at 1664 cm-1 (C=O) of the DPP unit, (c) CV of drop casted film before and after applying 1.0 V vs Ag/AgCl with a scan rate of 50 mV/s.

Drop cast film of p(gPyDPP-MeOT2) before (left) and after (right) applying 1.0 V for 100s.

21

p(gPyDPP-MeOT2)

Figure S17. (a) FT-IR of the pyridine-DPP monomer (g6-PyDPP-Br2) with a heptakis ethylene glycol side chain (blue), neat p(gPyDPP-MeOT2) (black) and p(gPyDPP-MeOT2) after applying 1.2 V for 100s (red). (b) zoom-in of the spectra shown in (a) (1800 and 650 cm-1) highlighting the C=O stretching peak of the DPP unit at 1664 cm-1, (c) CV of the drop cast film before and after applying 1.2 V vs Ag/AgCl with a scan rate of 50 mV/s.

22

. XPS analysis of the polymer degradation Thin films of p(gPyDPP-T2) and p(gPyDPP-MeOT2) were prepared on gold coated silicon substrates. For the decomposition study of the polymers, a constant potential (1.1 V for p(gPyDPP-T2) or 1.0 V for p(gPyDPP-MeOT2)) was applied for 100 second, followed by discharging the polymer by applying 0 V for 100s. All XPS spectra were calibrated to the C-C peak (284.8 eV).

p(gPyDPP-T2) Changes of the XPS spectra before and after charging the polymer to 1.1 V vs. Ag/AgCl for p(gPyDPP-T2). This includes the disappearance of the N-C=O peak (C1s) as well as a decrease of the C-C peak intensity, suggesting that degradation of the backbones is associated with cleavage of the side chains attached to the PyDPP unit. Additionally, the N1s peaks are affected where two peaks are observed for the neat polymer likely to be the nitrogen atoms of the pyridine- and DPP units which merge to become one peak after applying 1.1 V vs Ag/AgCl. The C=O peak of the O1s peak shows a lower intensity after degradation in agreement with the disappearance of the peak for N-C=O. The S2p peaks are almost unaffected which indicates that degradation is mostly occurring on the DPP unit.

Figure S18. XPS spectra of p(gPyDPP-T2) of (a) C1s, (b) N1S, (c) Os1 and (d) S2p (neat polymer (top) and after applying 1.1 V vs Ag/AgCl (bottom)).

23

p(gPyDPP-MeOT2) For p(gPyDPP-MeOT2), the N-C=O peak of the C1s peak becomes lower in intensity after applying 1.0 V vs. Ag/AgCl, however they do not fully disappear completely as observed for p(gPyDPP-T2). A shift of the C-C peak (C1s), the N1s peaks and C=O (O1s) peaks can be observed, supporting the hypothesis that the degradation mostly effects of the DPP unit. Additional changes of the S2p peaks of p(gPyDPP-MeOT2) can be observed, which could hint to chemical side reactions of the MeOT2 unit during bipolaron formation.

Figure S19. XPS spectra of p(gPyDPP-MeOT2) of (a) C1s, (b) N1S, (c) Os1 and (d) S2p (neat polymer (top) and after applying 1.1 V vs Ag/AgCl (bottom)).

24

. OECT performance of polymer p(gPyDPP-T2)

Figure S20. OECT performance of p(gPyDPP-T2) a) output curve (VG =0 to – 1 V, ΔVG = – 0.05 V), b) transfer curve of the OECT (VD = – 0.4 V) after taking the output curve and c) stability measurements of the device during pulsed cycling measurements by applying voltage pulses with VG

= – 1 V (ON) and VG = – 0.7 V (OFF) for 2 s while monitoring ID in 0.1 M NaCl aqueous solution.

25

. Capacitance measurements of p(gPyDPP-MeOT2)

Figure S21. Electrochemical Impedance Spectroscopy for p(PyDPP-MeOT2) at working electrode offsets ranging from 0 to 0.7V (from red to blue). To quantify Ceff impedance was fit to a Randles Circuit (Rs(RP||C). The fit is shown in the dash green lined ( Rs = 336Ω , RP = 1.01 MΩ, C = 6.22 µF).

102

103

104

105

106

107

0

45

90

10-1 100 101 102 103 104 105

10-7

10-6

10-5

10-4

10-3

10-2

10-1

|Z|

(Ω)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 FIT

-pha

se (φ

) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Fit

C eff (

F/cm

2 )

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Fit

26

Quantifying ORR of the polymers at low and high oxygen concentration Measurements were conducted in 0.1 M NaCl aqueous electrolytes where the solution was purged for 30 mins with either nitrogen gas (low O2 conc.) or oxygen gas (high O2 conc.).

PEDOT:PSS

Figure S22. (a) Cyclic voltammograms of PEDOT:PSS at ν = 50 mV s-1 in 0.1 M NaCl aqeuoues solution with low and high oxygen concentration. (b) Evolution of the UV Vis absoprtion spectrum of PEDOT:PSS on an ITO coated glass substrate at low O2 concentration. Evolution of the UV Vis spectrum after discharging the polymer to – 1.0 V vs. Ag/AgCl for 10 seconds and monitoring changes of the absoprtion spectrum at OCV at c) low O2 concentration and d) high O2 concentration for 5 minutes.

27

p(g2T-TT)

Figure S23. (a) Cyclic voltammograms of p(g2T-TT) at ν = 50 mV s-1 in 0.1 M NaCl aqeuoues solution with low and high oxygen concentration. (b) Evolution of the UV Vis absoprtion spectrum of p(g2T TT) on an ITO coated glass substrate glass substrate at low O2 concentration. Evolution of the UV Vis spectrum after discharging the polymer to – 0.5 V vs. Ag/AgCl for 10 seconds, monitoring changes of the absoprtion spectrum at OCV at c) low O2 concentration and d) high O2 concentration for 5 minutes.

28

p(gPyDPP-MeOT2)

Figure S24. (a) Cyclic voltammograms of p(gPyDPP-MeOT2)at ν = 50 mV s-1 in 0.1 M NaCl aqeuoues solution with high oxygen concentration. (b) Evolution of the UV Vis absoprtion spectrum of p(gPyDPP-MeOT2) on an ITO coated glass substrate during the second CV scan at high O2 conc. shown in a). c) Evolution of the UV Vis spectrum and monitoring changes of the absoprtion spectrum at OCV at high O2 concentration for 60 minutes.

29

. Redox-reactions in aqueous electrolytes at different pH values We measured the charging of p(gPyDPP-MeOT2) in 0.1 M NaCl (pH = 6.6) and in 0.1 M NaSbF6 (pH = 2.5) and compare the redox-reaction of p(gPyDPP-MeOT2) to p(g2T-TT) which were previously reported in the literature11. A 0.1 M NaSbF6 aqueous solution was prepared which has a pH of 2.5.

Figure S25. Electrochemical and spectroelectrochemical measurements of polymer p(g2T-TT) and p(gPyDPP-MeOT2) in neutral and acidic aqeuoues electrolytes. (a) Evolution of the absorption spectrum of p(g2T-TT) when a film is immersed in a 0.1 M NaSbF6 aqueous solution in ambient conditions for 10 mins. (b) Spectroelectrochemical measurements when charging the polymer in 0.1 M NaSbF6 from -0.4 V to 0.4 V vs Ag/AgCl. (c) CV measurements of p(g2T-TT) showing 5 scans with a scan rate of 100 mV/s from -0.4 V to 0.4 V vs Ag/AgCl. (d) Evolution of the absorption spectrum of p(gPyDPP-MeOT2) when a film is immersed in a 0.1 M NaSbF6 aqueous solution in ambient conditions. (e) Spectroelectrochemical measurements of p(gPyDPP-MeOT2) when charging from 0 V to 0.7 V vs Ag/AgCl in 0.1 M NaSbF6. (f) CV measurements of p(gPyDPP-MeOT2) showing 5 scans with a scan rate of 100 mV/s from -0.4 V to 0.7 V vs Ag/AgCl.

30

OECT pulsed cycling measurements The electrochemical redox-stability of PEDOT:PSS, p(g2T-TT) and p(gPyDPP MeOT2) during OECT operation was monitored. In addition, the gate current was recorded during the pulsing experiment (integrated gate current per area), representing the accumulated charge during the pulsing experiment. [the area of the channel is AChannel =3.5*10^-5 cm2 with W = 104 μm and L = 34 μm].

The pulsed cycling measurements were carried out for 15 mins where gate potentials between 0 V (ON current) and the indicated gate potential [OFF current] were applied for 5 s. The indicated VD was applied throughout each experiment. The electrolyte was replaced after each pulsing experiment and the same device was used for each pulsing experiment series (same VD). The device stability was determined by monitoring changes of the ON current ΔID[ON][%] = ID[ON][pulse #1]/ ID[ON][pulse #90]

31

PEDOT:PSS (VD = – 0.1 V)

Figure S26. Pulsed cycling measurements of the OECT for PEDOT:PSS, applyig VD = – 0.1 V and gate voltage pulses for 5 s between VG = 0 V and a) VG = 0.2 V, b) VG = 0.4 V, c) VG = 0.6 V and d) VG = 0.8 V. Changes of the ON Current Δ ID[ON][%] during the pulsing experiment (90 pulses): a) – 0.8 % , b) +5.4 %, c) – 16.1 % and d) – 48.6 %. e) Integrated gate current recorded for the indicated VG during the pulsing experiment.

32

PEDOT:PSS (VD = – 0.5 V)

Figure S27. Pulsed cycling experiments of an OECT for PEDOT:PSS applyig VD = – 0.5 V and voltage pulses for 5 s between VG = 0 V a) VG = 0.2 V b) VG = 0.4 V, c) VG = 0.6 V and d) VG = 0.8 V. Changes of the ON Current Δ ID[ON][%] during the pulsing experiment (90 pulses): a) -5.0 % , b) – 6.5 % %, c) – 15.3 % and d) – 52.2 %. e) Integrated gate current recorded for the indicated VG during the pulsing experiment.

33

p(g2T-TT) p(g2T-TT) can act as a catalyst for the conversion of H2O2 to H2O (at VG < 0 V), we carried out two types of measurements: i) Pulsed cycing experiments beginning in the ON state of the device (VG = – 0.40 V) and applying gate potentials to turn the device OFF (VG > 0 V) (meaning that this pulsing sequence will convert H2O2 formed during ORR to H2O, resulting in a lower integratded gate current per area) and ii) beginning the experiment in an OFF state (VG > 0 V) and successive decreasing the gate potential to monitor the formation of H2O2 formation (without converting H2O2 to H2O). One can observe that the integrated gate current per area decreased when decreasing the gate potential.

It needs to be noted that the OFF currents of the device are increasing when the polymer VG < 0 V are avoided, indicating that formation of H2O2 affects the device operation, probably due to competing reactions between the discharging of the polymer (VG > 0 V) and oxidation of the polymer by H2O2.

p(g2T-TT) (VD = -0.3 V) ON to OFF

Figure S28. Pulsed cycling experiments for p(g2T-TT) applying VD = – 0.3 V and voltage pulses for 5 s between VG = – 0.4 V and a) VG = 0.2 V and b) VG = 0.4 V. Changes of the ON Current Δ ID[ON][%] during the pulsing experiment (90 pulses): a) – 3.9 % , b) – 21.1 % and c) Integrated gate current per area recorded for the indicated VG during the pulsing experiment.

34

p(g2T-TT) (VD = – 0.1 V) OFF to ON

Figure S29. Pulsed cycling experiments for p(g2T-TT) applying VD = – 0.1 V and voltage pulses for 5 s between VG = – 0.4 V and a) VG = 0 V, b) VG = – 0.2 V and b) VG = – 0.4 V. c) Integrated gate current recorded for the indicated VG during the pulsing experiment.

35

p(g2T-TT) (VD = – 0.5 V)

Figure S30. Pulsed cycling experiments for p(g2T-TT) applying VD = – 0.5 V and voltage pulses for 5 s between VG = 0.4 V and a) VG = 0 V, b) VG = – 0.2 V and b) VG = – 0.4 V. c) Integrated gate current recorded for the indicated VG during the pulsing experiment.

36

p(gPyDPP-MeOT2) (VD = – 0.2 V)

Figure S31. Pulsed cycling experiments of an OECT for p(gPyDPP-MeOT2), applying VD = – 0.2 V and voltage pulses for 5 s between VG = 0 V and a) VG = – 0.60 V, b) VG = – 0.65 V and c) VG = – 0.70 V. Stability ΔID[%] during a 15 min pulsing experiment (90 pulses): a) +6.0 % , b) – 4.3 % %, and c) – 12.9 %. c) Integrated gate current recorded for the indicated VG during the pulsing experiment. Note that the values reccorded for the integrated gate current per area are negative compared to p(g2T-TT) or PEDOT:PSS.

37

Performance of OECTs varying the oxygen concentration in the electrolyte

Figure S32. OECTs with PEDOT:PSS and p(g2T-TT) were tested in ambient and degassed aqueous electrolytes. Transfer curves were reccorded in a nitrogen box or in ambient conditions. PEDOT:PSS [VD(low O2 conc.) = – 0.4 V, VD (high O2 conc.) = – 0.5 V), scan rate of 1V/s]; p(g2T-TT) [VD (low O2 conc.) = – 0.3 V, VD(high O2 conc.) = – 0.3 V), scan rate of 1V/s].

Monitoring the gate current with electrolytes containing low or high oxygen concentration

Figure S33. OECT gate currents reccorded during measuring transfer curves in ambient (high O2 conc.) and low O2 conc. (solid line) for PEDOT:PSS [VD (low O2 conc. ) = – 0.4 V, VD (high O2 conc.) = – 0.5 V; p(g2T-TT) [VD (low O2 conc.) = – 0.3 V, VD (high O2 conc.) = – 0.3 V); and p(pPyDPP-MeOT2) VD (high O2 conc.) = – 0.3 V)] in 0.1 M NaCl aqeuoues solution.

38

RRDE measurements Sample preparation: Polymer solutions for p(g2T-TT) [2mg/mL] and p(gPyDPP-MeOT2) [10 mg/mL] were prepared in chloroform and 9.8 μL were spuncast at 700 rpm on the glassy carbon disk electrode of a rotating ring-disk electrode (E6R1 ChangeDisk RRDE, glassy carbon disk electrode, Pt ring electrode, Pine Research Instrumentation). The disk electrode has diameter of 5 mm (area = 0.19 cm2) and the ring electrode an outer diameter of 7.5 mm and inner diameter of 6.5 mm (area = 0.11 cm2). For PEDOT:PSS (PH-1000), 9.8 μL of a solution containing 6% ethylene glycol, 1% (3-glycidyloxypropyl)trimethoxysilane, dodecylbenzene sulfonic acid was spun cast at 700 rpm on the glassy carbon disk of the RRDE electrode, followed by cross-linking the polymer at 120 °C. The polymer loading on the disk electrode is 0.02 mg for p(g2T-TT) and 0.1 mg for p(gPyDPP-MeOT2) and 0.11 mg of PEDOT:PSS.

RRDE measurements: A 0.1 M NaCl solution was sparged with oxygen or argon for 30 mins. Cyclic voltammetry was performed using a Bio-Logic VSP-300 potentiostat at 5 mV/s on the disk electrode with a coiled Pt wire as the counter electrode and a Ag/AgCl reference electrode (ET069 Leakless Ag/AgCl Reference Electrode, eDAQ). The ring potential was held constant at 0.7 V vs Ag/AgCl to oxidize potentially formed H2O2 during the CV measurements. The RRDE rotation speed was set to 1600 RPM for all experiments (MSR Electrode Rotator, Pine Research Instrumentation). The collection efficiency of the RRDE was calibrated with a 20 mM K3[Fe(II)(CN)6] in 0.1 M NaCl and is 0.24.

Figure S34: RRDE measurements showing the CV measurement of PEDOT:PSS, p(g2T-TT) and p(gPyDPP-MeOT2) in 0.1 M NaCl solutions with a scan rate of 5 mV/s with a) a low oxygen concentration and b) a high oxygen concentration. The current values are the average over two scans. Note: Polymer p(gPyDPP-MeOT2) can also be reduced at potentials < – 0.5 V vs Ag/AgCl (also see Figure S13).

39

The electron transfer number (n) was calculated as shown in equation (i)12 by using the average disk and ring currents from the measurements at high oxygen concentrations, substrated by the currents measured at low oxygen concentration.

The results suggest that for PEDOT:PSS and p(g2T-TT), n reaches values close to 2, suggesting a preferential pathway towards H2O2 production during the oxygen reduction reaction (O2 + 2H2O + 2e- H2O2 + 2OH-). For polymer p(gPyDPP-MeOT2), n increases towards n = 4 for p(gPyDPP-MeOT2), which suggests that the polymer mostly undergoes a 4 electron process, where H2O2 intermediates are efficiently converted to H2O (H2O2 + 2e- 2OH-). It is interesting to note that the ORR of the reduced polymer at potentials <-0.6 V vs. Ag/AgCl mostly results in the formation of H2O.

Figure S35. The electron transfer number (n) for PEDOT:PSS, p(g2T-TT) and p(gPyDPP-MeOT2).

40

. Detection of hydrogen peroxide PEDOT:PSS, p(g2T-TT) and p(PyDPP-MeOT2) were deposited on ITO coated substrates and custom made 3D printed wells were attached with epoxy glue as shown in Figure S34c. 3 mL of 0.1 M NaCl aqueous electrolyte was added and chronoamperometry measurements were carried out by applying -0.4 V for 10 h. For each measurement, a fresh electrolyte solution was prepared and the electrodes where cleaned before the charging experiments were carried out. CV measurements were carried out before and after the charging experiments to ensure redox-stability of the materials during the chronoamperometry measurements.

Figure S36. Chronoamperometry measurements of PEDOT:PSS, p(g2T-TT) and p(gPyDPP-MeOT2) deposit on ITO as well as a bare ITO electrode in 0.1 M NaCl aqueous electrolytes in ambient conditions. (a) 10 h, (b) 1 h zoom-in and (c) picture of the electrochemical setup. CV measurements were carried out before (black) and after (red) perfoming the chronoamperometry measurements to ensure redox-stability of the polymers. Note that applying smaller potentials than -0.4 V vs Ag/AgCl for PEDOT:PSS result in decomposition of the polymer (results not shown).

41

Figure S37. Detection of hydrogen peroxide formed during chronoamperometry experiments in ambient conditions (Figure S34). a) Evolution of the absorption spectrum of the dye 3,3′,5,5′-tetramethylbenzidine after the addition of the horseradish peroxidase to the electrolyte solution used for the chronoamperometry experiments (Figure S34a) for PEDOT:PSS, p(g2T-TT) and p(gPyDPP-MeOT2). b) Control experiment for the detection of H2O2, showing the absorption spectrum of the dye and enzyme in deionised water as well as the spectrum for the oxidised dye after the addition of H2O2.

Figure S38. Detection of hydrogen peroxide of p(gPyDPP-MeOT2) after 500 charging/discharging cycles between 0 V and 0.7 V vs Ag/AgCl in 0.1 M NaCl oxygen-containing aqueous electrolyte.

42

. References

1. Giovannitti, A. et al. Controlling the mode of operation of organic transistors through side-chain engineering. Proc. Natl. Acad. Sci. 113, 12017–12022 (2016).

2. Moia, D. et al. Design and evaluation of conjugated polymers with polar side chains as electrode materials for electrochemical energy storage in aqueous electrolytes. Energy Environ. Sci. 12, 1349–1357 (2019).

3. Cardona, C. M., Li, W., Kaifer, A. E., Stockdale, D. & Bazan, G. C. Electrochemical Considerations for Determining Absolute Frontier Orbital Energy Levels of Conjugated Polymers for Solar Cell Applications. Adv. Mater. 23, 2367–2371 (2011).

4. Khodagholy, D. et al. High transconductance organic electrochemical transistors. Nat. Commun. 4, 2133 (2013).

5. Nielsen, C. B. et al. Molecular Design of Semiconducting Polymers for High-Performance Organic Electrochemical Transistors. J. Am. Chem. Soc. 138, 10252–10259 (2016).

6. Khodagholy, D. et al. High transconductance organic electrochemical transistors. Nat. Commun. 4, 2133 (2013).

7. Ilavsky, J. Nika: Software for two-dimensional data reduction. J. Appl. Crystallogr. 45, 324–328 (2012).

8. Oosterhout, S. D. et al. Mixing Behavior in Small Molecule:Fullerene Organic Photovoltaics. Chem. Mater. 29, 3062–3069 (2017).

9. Josephy, P. D., Eling, T. & Mason, R. P. The horseradish peroxidase-catalyzed oxidation of 3,5,3’,5’-tetramethylbenzidine. Free radical and charge-transfer complex intermediates. J. Biol. Chem. 257, 3669–3675 (1982).

10. Jung, J. W., Liu, F., Russell, T. P. & Jo, W. H. Synthesis of pyridine-capped diketopyrrolopyrrole and its use as a building block of low band-gap polymers for efficient polymer solar cells. Chem. Commun. 49, 8495–8497 (2013).

11. Cendra, C. et al. Role of the Anion on the Transport and Structure of Organic Mixed Conductors. Adv. Funct. Mater. 0, 1807034 (2018).

12. Zhou, R., Zheng, Y., Jaroniec, M. & Qiao, S.-Z. Determination of the Electron Transfer Number for the Oxygen Reduction Reaction: From Theory to Experiment. ACS Catal. 6, 4720–4728 (2016).

download fileview on ChemRxivSI_191208.pdf (3.22 MiB)