Redox Mechanisms in Li and Mg Batteries Containing Poly(phenanthrene quinone)/Graphene Cathodes using Operando ATR-IR Spectroscopy

Article abstract of DOI:10.1002/cssc.202000054

The redox reaction mechanism of a poly(phenanthrene quinone)/graphene composite (PFQ/rGO) was investigated using operando attenuated total reflection infrared (ATR-IR) spectroscopy during cycling of Li and Mg batteries. The reference phenanthrene quinone and the Li and Mg salts of the hydroquinone monomers were synthesized and their IR spectra were measured. Additionally, IR spectra were calculated using DFT. A comparison of all three spectra allowed us to accurately assign the C=O and C?O^? vibration bands and confirm the redox mechanism of the quinone/Li salt of hydroquinone, with radical anion formation as the intermediate product. PFQ/rGO also showed exceptional performance in an Mg battery: A potential of 1.8 V versus Mg/Mg²⁺, maximum capacity of 186 mAh g^?1 (335 Wh kg^?1 of cathode material), and high capacity retention with only 8 % drop/100 cycles. Operando ATR-IR spectroscopy was performed in a Mg/organic system, revealing an analogous redox mechanism to a Li/organic cell.

Full text of DOI:10.1002/cssc.202000054

Accepted Article
Title: Poly(phenanthrene quinone)/graphene cathode material and
investigation of its redox mechanism through operando ATR-IR
spectroscopy in Li- and Mg- batteries
Authors: Alen Vizintin, Jan Bitenc, Anja Kopač Lautar, Jože
Grdadolnik, Anna Randon-Vitanova, and Klemen Pirnat
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.202000054
Link to VoR: http://dx.doi.org/10.1002/cssc.202000054
A Journal of
www.chemsuschem.org

10.1002/cssc.202000054
ChemSusChem
Poly(phenanthrene quinone)/graphene cathode material and investigation of its redox
mechanism through operando ATR-IR spectroscopy in Li- and Mg- batteries
Alen Vizintin,1 Jan Bitenc,1 Anja Kopač Lautar,1 Jože Grdadolnik,1 Anna Randon Vitanova,2 Klemen
Pirnat1*
1 National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
2 Honda R&D Europe GmbH, Carl-Legien Strasse 30, 63703 Offenbach, Germany
A. Vizintin, J. Bitenc and A. Kopač Lautar contributed equally to this work.
*Corresponding author: K. Pirnat, email: klemen.pirnat@ki.si
Abstract
In this article, the redox reaction mechanism of poly(phenanthrene quinone)/graphene composite
(PFQ/rGO) was determined using operando attenuated total reflection infrared (ATR-IR) spectroscopy
during Li- and Mg- battery cycling. Reference phenanthrene quinone, Li- and Mg- salts of hydroquinone
monomers were synthesized and their IR spectra measured. Additonally, IR spectra were calculated using
density functional theory (DFT). A comparison of all three spectra: operando, chemically synthesized and
DFT, allowed us to properly assign C=O and C–O– vibration bands and confirm redox mechanism
quinone/Li salt of hydroquinone with radical anion formation as the intermediate product. PFQ/rGO was
also tested in Mg-battery where it showed exceptional performance: potential 1.8 V vs. Mg/Mg2+, maximum
capacity of 186 mAh/g (335 Wh/kg per cathode material), and high capacity retention, only 8% drop/100
cycles. Operando ATR-IR spectroscopy was performed in Mg-organic system as well, revealing analogous
redox mechanism to Li-organic cell.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Introduction
From wireless remotes, mobile phones and to wearable electronics, battery-powered personalized
devices have transformed the way we live. Development of low-price, safer, more powerful, and more
energy-dense battery technology will unlock innovation within the transport and facilitate a transition to
renewables by providing flexibility in producing and consummation of energy. Most state-of-the-art Li-ion
batteries employ inorganic cathodes. Used cathode materials vary depending on the requirements of the
application from lithium cobalt oxide (LCO), lithium nickel manganese cobalt oxide (NMC), lithium
manganese oxide (LMO), lithium iron phosphate (LFP) to nickel cobalt aluminum oxide (NCA).[1,2] Most
of these materials contain toxic or relatively rare elements (Ni, Co), their production uses plenty of energy,
and requires temperatures above 700 °C during ceramic synthesis.[3] Their theoretical capacities are limited
to around 200 mAh/g due to relatively high atomic weight of transition metals and a low number of
exchanged electrons. Currently, the best material in terms of energy density is NMC 811 with capacity and
voltage of 200 mAh/g and 3.8 V vs. Li/Li+ which results in a gravimetric energy density of around 760
Wh/kg per cathode material.[4]
Mg-batteries are becoming increasingly attractive alternative to Li-based batteries. There are several
reasons for that: lower price, sustainability, improved safety profile, and higher volumetric energy density.
Moreover, lithium deposits are located at some geopolitically sensitive locations and its extraction can have
big impact on the environment. On the other hand, Mg is 1000 times more abundant[5] and can be
economically produced even from seawater. As Mg does not form dendrites,[6] it can be safely used as metal
anode with very high volumetric capacity 3832 Ah/L (Li has 2062 Ah/L). However, there are also
disadvantages such as passivation of Mg anode, lack of suitable electrolytes,[7] difficult insertion of Mg2+
ion into inorganic materials and slow solid-state diffusion, which causes lack of suitable cathodes and
severely impedes development of practical Mg batteries.[8]
Organic materials, on the other hand, offer a better alternative to inorganic materials in terms of
versatility with different metal counter ions,[9,10] price, gravimetric energy density and sustainability.[11–15]
They can be produced from petrochemicals, biomaterials, organic waste[3,13,16–18] or even from CO2 as a
source of carbon[19] using low-temperature synthesis below 100 °C. Due to their low molecular mass and
high number of exchanged electrons, they can reach high practical capacities up to 600 mAh/g (Li-
rhodizonate), which could theoretically be increased up to 960 mAh/g (value for C=O carbonyl group). This
would result in theoretical energy density of almost 2000 Wh/kg per cathode for 2.0 V vs. Li/Li+ organic
materials. The most common drawback of organic materials is their solubility in organic electrolytes which
results in fast capacity drop during cycling and self-discharge.[12,17,20] Most promising organic materials are
actually redox-active polymers with very low solubility in electrolytes. Here are some promising examples,
such as poly(imides),[21] poly(anthraquinonyl sulfide) (PAQS),[22] poly(2,5-dihydroxy-1,4-benzoquinonyl
sulfide) (PDBS),[23,24] poly(vinylanthraquinone) (PVAQ),[25] polymer-bound pyrene-4,5,9,10-tetraone
(PPYT),[26] poly(anthraquinone) (PAQ),[27] poly(benzenequinonyl sulfide) (PBQS),[28,29] and
poly(diphenanthrene-quinone substituted norbornene) (PQN3).[30] Cathode or positive electrode materials
based on anthraquinone, such as poly(anthraquinonyl sulphide) (PAQS)[22] and poly(anthraquinone)
(PAQ)[27] have very stable electrochemistry, high Coulombic efficiency, good power performance (rate
capability) and were successfully exploited in both Li- and post Li-battery systems.[31–33] On the other hand,
their discharge potential is only around 2.2 V vs. Li/Li+ for discharge and correspondingly lower in Mg cell.
Thus, we recently developed an isomer of PAQ, which is named poly(phenanthrenequinone) (PFQ) with a
potential of 2.6 V vs Li/Li+.[34] Unexpectedly, this PFQ material is insoluble in organic solvents which
resulted in high cycling stability with only 9% drop of capacity after 500 cycles (50 mA/g or around C/5).
However, insolubility resulted in low material utilization. Therefore, a porous composite material containing
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
21 wt. % of reduced graphene oxide (PFQ/rGO) was synthesized with better capacity utilization. Recently,
it has been shown by our[29,31,35] and other research groups[32,36–38] that redox-active organic materials can be
successfully used as Mg cathode materials. The advantage of organic materials in Mg-systems can be
attributed to their relatively flexible structure and swelling of organic materials inside electrolyte, which
allow better accessibility of Mg ions. Additionally organic materials have lower redox-potentials that fit
well inside the Mg electrolyte stability window.
Herein we compare the redox reaction mechanism of PFQ/rGO in Li- and Mg- metal organic
battery. The redox mechanism in Li-battery was probed by collecting operando ATR-IR spectra during
charge/discharge of Li-PFQ/rGO battery. Then DFT calculations and IR spectra of chemically synthesized
monomer molecules were used to accurately assign bands from the operando measurements and assist their
interpretation. PFQ/rGO was also used as Mg battery cathode and its mechanism was analyzed with help of
ATR-IR and DFT.
Results and discussion
Li-PFQ/rGO battery:
Galvanostatic curves of Li-PFQ/rGO battery are displayed in Figure 1a. PFQ/rGO has two charge plateaus
at 2.47 V and 2.93 V and two discharge plateaus at 2.78 V and 2.26 V vs. Li/Li+ and can be more clearly
visualized by dQ/dE plot (Fig. S1). Two plateaus and reversible capacity well over 130 mAh/g (130 mAh/g
is theoretical capacity for a one-electron reaction) suggests that we have two-step redox reaction. In each
step, one electron should be exchanged. To confirm this hypothesis we have performed operando ATR-IR
measurements during cycling of Li-PFQ/rGO battery. For this measurement, we used a special spectro-
electrochemical pouch cell with Si wafer window, which is in contact with Ge ATR crystal. For this
operando ATR-IR setup penetration depth was estimated to 0.4–1.20 µm in the measured region 1800–600
cm–1.[39] Estimated thickness of our electrodes was around 150 μm, which means that we probed only a small
part of the electrode. In the case of insoluble PFQ polymer, where cathode active material does not dissolve,
this is not a problem. A strong IR signal from Si window and electrolyte was subtracted according to the
literature.[39–41] Subtraction visualizes only IR-active changes inside the cathode composite during cycling
(Figure 1b). For the background we used IR spectra of discharged cathode. Negative band in operando
spectra means a decrease and positive means increase of corresponding species compared to discharged
state. During charging intensity of the carbonyl band C=O at 1678 cm–1 increases, while the corresponding
band of C–O– group located at 1377 cm–1 decreases. During discharge, the process is reversed showing
reversible electrochemical redox reaction. From these measurements, we can conclude that the redox
mechanism is reversible oxidation of Li hydroquinone salt into quinone. These results are very analogous
to operando ATR-IR measurement of comparable compound PAQS (C=O stretching at 1670 cm–1, and
1650 cm–1 and C–O– stretching at 1370 cm–1).[40] Besides these two bands, we are able to observe many
other positive/negative bands, which are analyzed in detail in the following section.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Figure 1. (a) Galvanostatic cycling of Li-PFQ/rGO battery at C/5 current density in 1M LiTFSI in
DOL+DME electrolyte. (b) Operando subtraction ATR-IR spectra of 2nd charge and discharge.
The assignment of infrared bands in the operando ATR-IR spectra was achieved by comparison with DFT
calculated spectra and IR spectra of chemically synthesized monomer molecules. Firstly, the accuracy of
DFT calculations was verified by calculation of infrared spectra for chemically synthesized monomer
molecules 2,7-dibromo-9,10-phenanthrenequinone (FQ), dilithium salt of 2,7-dibromo-9,10-dihydroxy-
phenanthrene (LiFQ) and magnesium salt of 2,7-dibromo-9,10-dihydroxyphenanthrene (MgFQ). These
calculated spectra were compared with measured ones (Figure S2a–S2c). The spectra match surprisingly
well, especially for the most intensive bands such as carbonyl C=O stretching, C=C and C–Oˉ stretching
vibrations, ring vibrations and C–H out-of-plane deformations. However, some bands are shifted. The
biggest mismatch can be found for carbonyl band where calculation gives 65 cm–1 higher frequency than
measured value (Figure S2a). It can be also noted that calculation gives very inaccurate intensities for the
bands, especially at lower frequencies. Moreover, operando ATR measurement gives higher intensity at
lower frequencies. Due to these two reasons, we decided to zoom calculated DFT spectra 3–6× times in the
region below 1300 cm–1 to get more comparable results with operando ATR measurements. Noteworthy,
the calculation reveals the expected complex nature of the bands which belong to the ring vibrations. Those
bands are the combination of C–C, C=C, C–H and C–Oˉ or C=O modes that are strongly coupled. The
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
appearance of such coupled vibrations can be an additional source of observed differences between
calculated and measured spectra.
In the next step, DFT calculations were applied for LiFQ3 and FQ3 oligomers (Figure 2a), consisting of three
monomer units. Using larger oligomers as models would give us more accurate predictions but would extend
calculations tremendously. For more detailed explanation see Experimental part. Subtracted DFT spectra
were calculated using equation SDFT, Li = FQ3 – LiFQ3. For comparison, we also calculated subtracted spectra
for chemically synthesized reference monomers Schem, Li = FQ – LiFQ (Figure 2b), which were very similar
to the operando ATR-IR spectra (Figure 2c). For instance, carbonyl C=O stretching band[42,43] at 1678 cm–
1 from operando ATR-IR measurement is very close to the chemically synthesized monomer band at 1675
cm–1. C–Oˉ stretching vibration at 1377 cm–1 is close to the predicted value of 1348 cm–1 from DFT and the
measured value of 1383 cm–1 from chemically synthesized monomers. Other bands are explained in the
Supporting information and in Table S1.
Figure 2. Comparison of (a) DFT calculation of model trimer, 3-6× zoom was used below 1300 cm–1. (b)
measured IR spectra of chemically synthesized reference monomer. (c) Operando ATR-IR spectra of Li-
PFQ/rGO battery at the end of charge at 3.5 V vs. Li/Li+.
From the charge/discharge galvanostatic curves two plateaus can be easily distinguished, which can be
attributed to a two-electron reaction in two separate one-electron steps. Using the operando ATR-IR
technique (Figure 3a), we were able to detect intermediate radical anion LiPFQ* (Figure 3b). Two bands
1549 cm–1 and 1484 cm–1 start to appear towards the middle of the charge/discharge cycle and are in good
agreement with DFT calculated bands at 1562 cm–1 and 1487 cm–1 (Figure 3c and 3d). They correspond to
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
C O
C C
bonds. The position of those bands is in a good agreement
the stretching vibration of the
and
also with the data from the literature.[44] Other bands monotonically increase during charging and correspond
to C–Oˉ (1375 cm–1), ring vibrations (1254 cm–1 and 1048 cm–1), and C–H out-of-plane vibrations (808 cm–
1).
Figure 3. (a) Operando ATR-IR spectra of Li-PFQ/rGO battery during charging/discharging. (b) Proposed
redox reaction with intermediate radical anion LiPFQ*. (c) DFT spectra for radical anion trimer LiFQ3*,
discharged state LiFQ3, and their subtraction. 3-6× Zoom was used below 1300 cm–1. (d) Operando ATR-
IR spectra in the middle of the charge at 2.67 V vs. Li/Li+.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Mg-PFQ/rGO battery:
Electrochemical performance of PFQ/rGO composite material was evaluated in Mg-battery using 0.6 M
Mg(TFSI)2-2MgCl2 in dimethoxyethane (DME) electrolyte at 0.5C current density. Initial capacity is only
65 mAh/g, but it quickly increases to 130 mAh/g after only ten cycles (Figure 4a). Subsequent increases
increase to 186 mAh/g (72% of theoretical capacity Ctheo=260 mAh/g) is very gradual. Between 140-400
cycles capacity slowly decreases down to 146 mAh/g (8% drop/100 cycles). To the best of our knowledge,
this is one of the best results for Mg battery in terms of discharge capacity and cycle stability. Coulombic
efficiency in the first 14 cycles is above 100% due to large polarization in first cycles, which prematurely
finishes discharge. Consequently, some initial capacity (pristine material is in charged state) is retained and
discharge capacity is larger as charge capacity in previous charge. Efficiency with cycling drops to 94%
after 250 cycles and later increases back to 96% till 400th cycle. After 400 cycles, sporadic Coulombic
efficiency drops are observed and the capacity drop becomes more prominent. The reason for this is most
likely Mg anode degradation. If we look at galvanostatic curves, we can clearly see that the polarization
decreases until 100 – 200 cycles (Figure 4b). At 100th cycle two discharge plateaus at 2.19 V and 1.71 V vs.
Mg/Mg2+, and two charge plateaus at 2.25 V and 2.55 V vs. Mg/Mg2+ are observed and can be more clearly
visualized by dQ/dE plot (Fig. S3).. If we compare these results with a Li-system, the equilibrium potential
is only around 420 mV lower in Mg-system and the expected value is around 700 mV (E(Li/Li+) – E(Mg/Mg2+)
= 3.040 V – 2.372 V = 668 mV). Higher potential in Mg-battery system could be explained with an
energetically more favorable coordination of Mg2+ ion with two C–O– in the ortho position compared to two
Li+ ions.[35,45] We also tested C-rate performance of Mg-PFQ/rGO battery from 0.5C to 50C rate (Figure
4c). Due to a slow activation, the capacity is stabilized after 32 cycles (at 2C). However, we can still compare
the capacities relative to 0.5C at the end of the cycling (183 mAh/g, 100 cycles). At a current density of 5C,
the capacities decrease to around 1/3rd and at > 20C no capacity can be obtained. Coulombic efficiency
during rate capability tests decreases with higher C-rate from around 99% until 83%, which is rather unusual
– in most cases efficiency improves at high rates because there is less time for side reactions. In our case,
there may be again problem with premature cut-off at high rates, where polarization becomes significant.
More detailed information about C-rate experiments were obtained from galvanostatic curves (Figure 4d).
We can observe that the polarization increases with higher C-rate, which is expected. Capacity drop is
mainly attributed to the charge cut-off potential – we can clearly see that charging plateau at 5C and higher
rates is finished too early at 2.8 V vs. Mg/Mg2+. Thus, the capacity at high rates could be improved by setting
cut-off voltage to higher values, but we are limited with electrolyte stability window, which is around 2.8
V vs. Mg/Mg2+ for high surface area electrodes. In the same manner we can also explain low capacity in
early cycles: material could not be fully charged in first cycles until polarization gradually drops to allow
full charging of the material.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Figure 4. Cycling of Mg-PFQ/rGO battery under: (a) at 0.5C for 490 cycles (0.8-2.8 V vs. Mg/Mg2+), (b)
galvanostatic curves at different cycles numbers at 0.5C, (c) rate capability test from 0.5C to 50C (0.8-2.8
V vs. Mg/Mg2+), (d) galvanostatic curves at different current densities from 0.5C to 50C.
Operando ATR-IR spectra during cycling of Mg-PFQ/rGO battery were obtained in a similar manner as in
the case of Li battery system. During charging, carbonyl C=O band at 1680 cm–1 is increasing and C–O–
band at 1392 cm–1 is decreasing, although both are much smaller compared to Li-system, due to a lower
capacity utilization (Figure 5a). However, we can confirm a proposed redox reaction mechanism where Mg
hydroquinone salt MgPFQ is reversibly oxidized into quinone PFQ via radical anion MgPFQ* (Figure 5b).
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Figure 5. (a) Operando ATR-IR spectra of PFQ/rGO during cycling of Mg-PFQ/rGO battery at 0.5C current
density and (b) Proposed redox mechanism of PFQ during charge/discharge.
We followed the same strategy of band assignment as in the case of Li-system (Figure 6a and 6b).
Characteristic bands for C=O, C=C, C–O– and C–H out-of-ring vibrations have similar positions than in Li-
system (Table S1). On the other side, ring vibrations from 1350 cm–1 to 1000 cm–1 in Mg-system are quite
different compared to Li-system. When we compare this region with DFT calculations and chemically
synthesized monomers, the same pattern can be recognized: two bands (1350 cm–1 and 1330 cm–1), followed
by three bands (1183 cm–1, 1135 cm–1 and 1069 cm–1), and a negative band at 1050 cm–1.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Figure 6. Comparison of (a) DFT calculation of model trimer, 3-6× zoom was used below 1300 cm–1. (b)
Measured IR spectra of chemically synthesized reference monomer and (c) operando ATR-IR spectra of
Mg-PFQ/rGO battery at the end of charge at 2.8 V vs. Mg/Mg2+.
Two plateaus can be observed also in Mg-system during charging/discharging, suggesting that two-electron
redox reaction has two separate processes with intermediate relative stable radical anion (Figure 5b). To
confirm this hypothesis operando ATR-IR spectra were used (Figure 5a). Again, we can observe two
–1
–1
C O
C C
vibrations of anion
characteristic bands at 1544 cm and 1485 cm that correspond to
and
radical. However, if we compare these positions with those obtained by DFT (1648 cm–1 and 1587 cm–1),
we can observe quite a large difference between measured and calculated values (Figure 7). One of the
reasons could be attributed to the fact that in DFT calculations we were not able to use 1/2Mg2+ ion per
monomer unit. The MgCl+ ion was used instead. Use of MgCl+ was justified by the fact that in Mg battery
systems with Mg(TFSI)2 and MgCl2 based electrolyte, cathode material is exchanging both Mg2+ and
MgCl+.[39,46] All other bands are also present in charged spectra (Figure 6c, Table S1) and they are slightly
shifted.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Figure 7. (a) DFT calculation of model trimer: radical anion MgFQ3*, discharged state MgFQ3, and their
subtraction. 3-6× Zoom was used below 1300 cm–1. (b) Operando ATR-IR spectra in the middle of the
charge at 2.36 V vs, Mg/Mg2+.
Conclusion
PFQ/rGO active material is a next generation cathode material with extraordinary cycling stability
in both Li- and Mg-organic batteries. In Mg system material displays high capacity utilization of 186 mAh/g
and a small capacity fade 0.08% per cycle with an average discharge potential of 1.8 V vs. Mg/Mg2+. Rate
capability of Mg-PFQ/rGO is moderate, which can be mainly attributed to relatively high polarization of
Mg metal anode. Starting with Li electrolyte electrochemical mechanism of PFQ/rGO was investigated
through operando ATR-IR. Operando ATR-IR spectra were complemented with the synthesis of model
compounds and DFT calculations, which confirmed the quinone/hydroquinone salt electrochemical
mechanism. Formation of radical anion was identified as an intermediate step through appearance of new
bands at 1549 cm–1 and 1484 cm–1 which were the most intensive at the middle of the half cycle and assigned
,
with the help of DFT. Interestingly bands of radical anion were observed also in Mg system, which might
point to complex electrochemical mechanism with the possibility of MgCl+ or other complex monovalent
ions serving as counterions. Another option is that Mg2+ ions interact with two neighboring PFQ radical
ions. Nevertheless, PFQ/rGO cathode showed promising performance as a cathode in Mg battery, which
opens a path towards its application in other multivalent battery systems (Al, Zn, Ca…). Its high cycling
stability might in future lead to practical multivalent organic batteries.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
Experimental part
PFQ/rGO or Poly(phenanthrene quinone)/graphene composite: Synthesis was started by bromination of
phenanthraquinone with N-bromosuccinimide (NBS) in H2SO4 to obtain a brominated derivative of
monomer 2,7-dibromo-9,10-dihydroxyphenanthrene (FQ). Then FQ was reduced into 2,7-dibromo-9,10-
dihydroxyphenanthrene with Sn/HCl, which was further acetylated to monomer 9,10-diacetoxy-2,7-
dibromophenanthrene. The latter was polymerized using Ni(COD)2 in the presence of reduced graphene
oxide, dispersed in N-methyl pyrrolidone. As obtained polymer composite was deprotected with LiAlH4,
and then oxidized using 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ) to obtain the PFQ/rGO polymer
composite. More information is in the literature.[34]
FQ or 2,7-dibromo-9,10-phenanthrenequinone: 9,10-Phenanthrenequinone (7.00 g, 33 mmol) was
brominated using N-bromosuccinimide (NBS) (12.86 g, 72 mmol) in 190 mL of concentrated sulfuric acid
(97%) at room temperature for 8 h. Impure product (12.68 g, 103% yield) was further purified by
crystallization in boiling DMF (120 mL), filtrated, washed with cold DMF two times and dried overnight in
an oven at 50 °C to obtain a yellow crystalline product (8.13 g, 66% yield). More information is in the
literature.[47]
LiFQ or Dilithium salt of 2,7-dibromo-9,10-dihydroxyphenanthrene: 2,7-dibromo-9,10-phenanthrene-
quinone (FQ) (3.66 g, 10 mmol) was reduced using metallic tin (4 g, 34 mmol) in a mixture of 10 mL of
concentrated HCl and 50 mL of glacial acetic acid under reflux for 1 h yielding reduced 2,7-dibromo-9,10-
dihydroxyphenanthrene as a colorless crystalline solid (3.64 g, 99% yield) according to the literature.[34,48]
1H NMR (300Hz, DMSO-d6): δ 9.40 (2H, s), 8.67 (2H, d, J = 8.9 Hz), 8.30 (2H, d, J = 2.1 Hz), 7.66 (2H,
dd, J = 8.9, 2.1 Hz). ATR-IR: 3371, 1630, 1598, 1435, 1302, 1200, 1123, 1031, 928, 880, 792, 691 cm–1.
Reduced 2,7-dibromo-9,10-dihydroxyphenanthrene (30 mg, 0.082 mmol) was dissolved in dry THF (2 mL).
Then a solution of 2.5M n-butyl lithium in hexanes (Sigma Aldrich, 66 µL, 0.164 mmol, 2 equiv.) was
slowly added to the above solution. Color first changed from yellowish to dark green, then back to orange
at the end of the addition. The reaction mixture was stirred for 1 day at 30 °C, then the volatiles were
removed at 50 °C under reduced pressure to obtain orange powder (32 mg, 103% crude yield). All reaction
steps, isolation and characterization were performed inside a glovebox under inert Argon atmosphere as
1
these salts can be very easily oxidized in the air atmosphere. H NMR (300 MHz, DMSO-d6) δ 8.52 and
8.33 (4H, br m), 7.07 (2H, br s), ATR-IR: 2975, 2957, 2875, 1579, 1557, 1460, 1420, 1384, 1306, 1203,
1053, 991, 885, 814, 794, 699 cm–1.
MgFQ or Magnesium salt of 2,7-dibromo-9,10-dihydroxyphenanthrene: Was prepared in a similar manner
than Li salt above (LiFQ), with the only difference that 1M n-dibutil-Mg in heptanes (Sigma Aldrich, 82
1
µL, 0.082 mmol, 1 equiv.), was used instead of n-BuLi (33 mg, 104% crude yield). H NMR (300Hz,
DMSO-d6): δ 8.57 (2H, d, J = 2.3 Hz), 8.39 (2H, d, J = 8.8 Hz), 7.15 (2H, dd, J = 8.8, 2.3 Hz). ATR-IR:
2954, 2869, 1587, 1469, 1407, 1365, 1307, 1186, 1130, 1057, 1025, 994, 876, 816, 786, 708 cm–1.
ATR-IR spectra: Infrared spectra of reference monomers FQ, LiFQ, and MgFQ were measured inside the
glovebox under an inert Argon atmosphere on an IR spectrophotometer Bruker Alpha using Germanium
ATR crystal. Measuring range was 4000–600 cm–1. 32 consecutive scans were measured with a resolution
4 cm–1.
Operando ATR-IR spectroscopy: The ATR-IR measurements were done on a Bruker Vertex 80 equipped
with a Specac Silver Gate Ge crystal ATR and a liquid nitrogen cooled mercury cadmium telluride (MCT)
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
detector. The spectra were collected in absorbance mode with 64 scans at a resolution of 4 cm–1 in the range
from 4000 cm–1 to 600 cm–1. The IR spectra were collected in operando mode with a series of repetitive
scans every 2 min during galvanostatic cycling of the batteries with a current density of C/5 rate for a Li
battery system and 0.5C for a Mg battery system. The battery was assembled in a spectro-electrochemical
cell with a Si wafer window.[39,40] The atmospheric compensation was performed on the IR spectra in OPUS
version 7.8 software. The ATR difference spectra were obtained by subtracting the discharge spectrum from
the obtained IR spectrum at a specific point of the discharge/charge.
DFT calculations: DFT calculations were performed using Gaussian 09 software package.[49] M06-2X
hybrid functional[50] and 6-31+G(d,p) basis set were used to do geometry optimization and calculation of
vibrational frequencies and IR intensities. To model polymer PFQ and its reduced forms in Li and Mg
battery system trimer consisting of three monomer units that end with a methyl group was used (FQ3, LiFQ3,
LiFQ3*, MgFQ3, and MgFQ3*) in all cases (Figure S4). The models ensured suitable accuracy of the
calculation while keeping the computational cost reasonable. Vibrational modes of methyl groups that are
included in the model due to the finite size of the trimer, are not included in the analysis and plotted spectra.
To account for the effects of the surrounding environment, a dielectric constant of 7.4 was used – which is
a common value for ether-based electrolytes. Discrepancies between the theoretical and experimental
spectra arise due to approximations of the theoretical approach, such as the treatment of the electronic
Hamiltonian, finite basis set, and the harmonic approximation. This leads to overestimation of the calculated
frequencies by about 5%. Thus, we apply the scaling factor of 0.95 that allows easier visual comparison
between the theoretical and experimental values.
NMR spectra: NMR spectra were collected using 300 MHz Varian Unity Inova. Deuterated solvents CDCl3
or DMSO-d6 were used and TMS was used as a standard.
Assembling Li-batteries: Electrodes were prepared by mixing 60 mg of composite PFQ/rGO, 30 mg of
carbon black (Printex XE2) and 10 mg of polytetrafluoroethylene (PTFE) (60 wt.% water dispersion,
Aldrich) and 0.5 mL of isopropyl alcohol (IPA). All these ingredients were ball milled in 12 mL stainless
steel grinding jars (10 mm Ø balls) with a planetary ball mill (Retsch PM100) at 300 rpm for 30 min in an
air atmosphere. The obtained slurry was kneaded with a mortar and pestle to get a compact black gum. The
gum was rolled between two pieces of a non-adhesive paper with a roller to get an electrode film of
approximately 5×5 cm size. On top of this film, aluminum mesh (100 mesh size) was deposited and rolled
again to glue electrode composite and mesh together and dried in an air atmosphere. Afterward, electrode
discs with diameter 1.2 cm were cut and pressed with a load of 1 ton and further dried at 80 °C in vacuum
for 1 day. An average loading on electrode was 2.5 ± 0.4 mg of active material per cm2. Battery cells were
assembled in argon-filled glove box (water and oxygen levels < 1 ppm). Swagelok type battery cells were
assembled using the above-mentioned electrodes, a 13 mm glass fiber separator (Whatman GF/A) and
freshly rolled lithium (12 mm diameter, Aldrich). As an electrolyte 1M Bis(trifluoromethane)sulfonimide
lithium salt (Aldrich) in a mixture of dry 1,3-Dioxolane + Dimethoxyethane was used (1M
LiTFSI/DOL+DME).
Assembling Mg-batteries: Electrode composite gum for Mg tests was prepared in same way as that for Li.
Then gum was rolled in between two sheets of non-adhesive paper and afterwards electrode discs with 1.2
cm diameter were cut out to give self-standing electrodes. The obtained electrodes were subsequently dried
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
at 80 °C in vacuum for 1 day and transferred into an argon-filled glove box. An average loading on electrode
was 2.5 ± 0.4 mg of active material per cm2. Mg cells were assembled in Swagelok type battery cells using
graphite disc as cathode current collector, 13 mm glass fiber separator (Whatman GF/A) and freshly brushed
Mg foil (Gallium Sources, 99.95%, 0.05 mm). 0.6 M Mg(TFSI)2-2MgCl2 in DME was used as electrolyte.
Electrochemical measurements were performed on a potentiostat/galvanostat VMP3 (Bio-Logic, France)
at room temperature (25 °C). Batteries were cycled between 1.5–3.5 V vs. Li/Li+ at a current density of
around C/5. Mg-batteries were cycled from 0.8–2.8 V vs. Mg/Mg2+ at different C-rates from 0.5C until 50C.
Acknowledgement
We would like to thank the financial support from the Slovenian Research Agency ARRS (research projects
J2-8167, Z2-1863, Z2-1864 and research core funding P2-0393) and Honda R&D Europe (Germany).
Keywords: Ab initio calculations • operando ATR-IR spectroscopy • Mg-organic battery •
poly(phenanthrene quinone) • redox chemistry
References
[1]
[2]
[3]
[4]
C. Pillot, in Int. Batter. Semin. Exhib. Fort Lauderdale, Florida, USA, 2017.
K. Baes, M. Kolk, F. Carlot, A. Merhaba, Y. Ito, “Future of batteries, Winner takes all?,” 2018.
D. Larcher, J.-M. Tarascon, Nat. Chem. 2015, 7, 19–29.
D. Andre, S.-J. Kim, P. Lamp, S. F. Lux, F. Maglia, O. Paschos, B. Stiaszny, J. Mater. Chem. A
2015, 3, 6709–6732.
[5]
W. M. Haynes, D. R. Lide, T. J. Bruno, CRC Handbook of Chemistry and Physics. Abundance of
Elements in the Earth’s Crust and in the Sea, 2017.
[6]
[7]
D. Aurbach, Y. Cohen, M. Moshkovich, Electrochem. Solid-State Lett. 2001, 4, A113–A116.
J. Muldoon, C. B. Bucur, A. G. Oliver, T. Sugimoto, M. Matsui, H. S. Kim, G. D. Allred, J.
Zajicek, Y. Kotani, Energy Environ. Sci. 2012, 5, 5941–5950.
[8]
[9]
E. Levi, Y. Gofer, D. Aurbach, Chem. Mater. 2010, 22, 860–868.
N. Patil, A. Mavrandonakis, C. Jérôme, C. Detrembleur, J. Palma, R. Marcilla, ACS Appl. Energy
Mater. 2019, 2, 3035–3041.
[10] A. Ponrouch, J. Bitenc, R. Dominko, N. Lindahl, P. Johansson, M. R. Palacin, Energy Storage
Mater. 2019, 20, 253–262.
[11] J.-M. Tarascon, M. Armand, Nature 2001, 414, 359–367.
[12] Y. Liang, Z. Tao, J. Chen, Adv. Energy Mater. 2012, 2, 742–769.
[13] Z. Song, H. Zhou, Energy Environ. Sci. 2013, 6, 2280–2301.
[14] B. Häupler, A. Wild, U. S. Schubert, Adv. Energy Mater. 2015, 5, 1402034.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
[15] T. B. Schon, B. T. McAllister, P.-F. Li, D. S. Seferos, Chem. Soc. Rev. 2016, 45, 6345–6404.
[16] M. Armand, J.-M. Tarascon, Nature 2008, 451, 652–657.
[17] H. Chen, M. Armand, G. Demailly, F. Dolhem, P. Poizot, J.-M. Tarascon, ChemSusChem 2008, 1,
348–355.
[18] M. Miroshnikov, K. P. Divya, G. Babu, A. Meiyazhagan, L. M. Reddy Arava, P. M. Ajayan, G.
John, J. Mater. Chem. A 2016, 4, 12370–12386.
[19] S. A. Al-Saydeh, S. J. Zaidi, in IntechOpen, 2018, pp. 41–61.
[20] M. Yao, H. Senoh, S. Yamazaki, Z. Siroma, T. Sakai, K. Yasuda, J. Power Sources 2010, 195,
8336–8340.
[21] Z. Song, H. Zhan, Y. Zhou, Angew. Chemie - Int. Ed. 2010, 49, 8444–8448.
[22] Z. Song, H. Zhan, Y. Zhou, Chem. Commun. 2009, 448–450.
[23] K. Liu, J. Zheng, G. Zhong, Y. Yang, J. Mater. Chem. 2011, 21, 4125–4131.
[24] Z. Song, Y. Qian, X. Liu, T. Zhang, Y. Zhu, H. Yu, M. Otani, H. Zhou, Energy Environ. Sci. 2014,
7, 4077–4086.
[25] W. Choi, D. Harada, K. Oyaizu, H. Nishide, J. Am. Chem. Soc. 2011, 133, 19839–19843.
[26] T. Nokami, T. Matsuo, Y. Inatomi, N. Hojo, T. Tsukagoshi, H. Yoshizawa, A. Shimizu, H.
Kuramoto, K. Komae, H. Tsuyama, et al., J. Am. Chem. Soc. 2012, 134, 19694–19700.
[27] Z. Song, Y. Qian, M. L. Gordin, D. Tang, T. Xu, M. Otani, H. Zhan, H. Zhou, D. Wang, Angew.
Chemie Int. Ed. 2015, 54, 13947–13951.
[28] Z. Song, Y. Qian, T. Zhang, M. Otani, H. Zhou, Adv. Sci. 2015, 2, 1500124.
[29] J. Bitenc, K. Pirnat, G. Mali, B. Novosel, A. Randon Vitanova, R. Dominko, Electrochem.
commun. 2016, 69, 1–5.
[30] T. Kawai, S. Nakao, H. Nishide, K. Oyaizu, Bull. Chem. Soc. Jpn. 2018, 91, 721–727.
[31] J. Bitenc, K. Pirnat, T. Bančič, M. Gaberšček, B. Genorio, A. Randon-Vitanova, R. Dominko,
ChemSusChem 2015, 8, 4128–4132.
[32] B. Pan, J. Huang, Z. Feng, L. Zeng, M. He, L. Zhang, J. T. Vaughey, M. J. Bedzyk, P. Fenter, Z.
Zhang, et al., Adv. Energy Mater. 2016, 6, 1600140.
[33] J. Bitenc, N. Lindahl, A. Vizintin, M. E. Abdelhamid, R. Dominko, P. Johansson, Energy Storage
Mater. 2020, 24, 379–383.
[34] K. Pirnat, J. Bitenc, A. Vizintin, A. Krajnc, E. Tchernychova, Chem. Mater. 2018, 30, 5726–5732.
[35] J. Bitenc, K. Pirnat, E. Žagar, A. Randon-Vitanova, R. Dominko, J. Power Sources 2019, 430, 90–
94.
[36] L. Chen, J. L. Bao, X. Dong, D. G. Truhlar, Y. Wang, C. Wang, Y. Xia, ACS Energy Lett. 2017, 2,
1115–1121.
[37] F. Wang, X. Fan, T. Gao, W. Sun, Z. Ma, C. Yang, F. Han, K. Xu, C. Wang, ACS Cent. Sci. 2017,
3, 1121–1128.
[38] X. Fan, F. Wang, X. Ji, R. Wang, T. Gao, S. Hou, J. Chen, T. Deng, X. Li, L. Chen, et al., Angew.
Chemie Int. Ed. 2018, 57, 7146–7150.
[39] J. Bitenc, A. Vizintin, J. Grdadolnik, R. Dominko, Energy Storage Mater. 2019, 21, 347–353.
[40] A. Vizintin, J. Bitenc, A. Kopač Lautar, K. Pirnat, J. Grdadolnik, J. Stare, A. Randon-Vitanova, R.
Dominko, Nat. Commun. 2018, 9, 661.
[41] T. Bančič, J. Bitenc, K. Pirnat, A. Kopač Lautar, J. Grdadolnik, A. Randon Vitanova, R. Dominko,
J. Power Sources 2018, 395, 25–30.
[42] J. B. Lambert, Introduction to Organic Spectroscopy, Macmillan, 1987.
This article is protected by copyright. All rights reserved.

10.1002/cssc.202000054
ChemSusChem
[43] G. Socrates, Infrared and Raman Characteristic Group Frequencies, John Wiley & Sons, LTD,
West Sussex PO19 1UD, England, 2001.
[44] J.-R. Burie, A. Boussac, C. Boullais, G. Berger, T. Mattioli, C. Mioskowski, E. Nabedryk, J.
Breton, J. Phys. Chem. 1995, 99, 4059–4070.
[45] K. Hernández-Burgos, G. G. Rodríguez-Calero, W. Zhou, S. E. Burkhardt, H. D. Abruña, J. Am.
Chem. Soc. 2013, 135, 14532–14535.
[46] H. Dong, Y. Liang, O. Tutusaus, R. Mohtadi, Y. Zhang, F. Hao, Y. Yao, Joule 2019, 3, 782–793.
[47] T. H. Vo, M. Shekhirev, D. A. Kunkel, F. Orange, M. J.-F. Guinel, A. Enders, A. Sinitskii, Chem.
Commun. 2014, 50, 4172–4174.
[48] D. Xianying, W. Meilin, D. Bencai, L. Shujing, Z. Yuli, Z. Xiangjun, G. Libing, 2,7-Dibromo-9-
Hydroxyl Phenanthrene Derivatives and Preparation Method Thereof, 2014, CN20121287739
20120814.
[49] M. J., Frisch, G. W., Trucks, H. B., Schlegel, G. E., Scuseria, M. A., Robb, et al., 2009, Gaussian
09, Revision A.02.
[50] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
This article is protected by copyright. All rights reserved.