MRS Advances © 2017 Materials Research Society

DOI: 10.1557/adv.2017.267

Advanced anode materials for sodium ion batteries: carbodiimides

Aitor Eguia-Barrio1, Elizabeth Castillo-Martinez2,#,*, Xiaohui Liu3, Richard Dronskowski3, Luis

Lezama1, Michel Armand2 and Teofilo Rojo1,2,*
1
Dpto. de Química Inorgánica, Universidad del País Vasco UPV/EHU, P.O. Box 644, 48080
Bilbao, Spain.
2
CIC Energigune, Parque Tecnológico de Alava, C/Albert Einstein 48, 01510, Miñano, Spain
#
Current address: Department of Chemistry, University of Cambridge, Lensfield Road, CB2
1EW, Cambridge, UK.
3
Institute of Inorganic Chemistry, RWTH Aachen University, Landoltweg 1, D-52056, Aachen,
Germany
*
Corresponding author: [email protected], [email protected]

ABSTRACT

TMNCN (where TM = Mn2+, Fe2+, Co2+ or Ni2+) have been recently proposed as
electrochemically active materials for Na-ion insertion that operate via conversion reaction.
Their electrochemical performance for Na-ion batteries is presented here with an emphasis on
long-term cycling. With a very low voltage for Na insertion of ∼0.1V vs Na+/Na for MnNCN,
the overpotential observed in batteries of MnNCN plays a very important role in their
performance, evidencing big differences in the electrochemical performance between materials
produced with different nano- and micrometer particle sizes evidenced by SEM images. A more
suitable voltage for the conversion reaction accompanied by less overpotential is shown by
FeNCN, CoNCN and NiNCN. Despite the lower reversible capacity achieved by FeNCN (450
mAh/g) in comparison with CoNCN and NiNCN in the first cycle; the smallest first-cycle
irreversible capacity (220 mAh/g) and the lower voltage plateau (0.3 V vs Na+/Na) make FeNCN
a good candidate as an anode material for sodium ion batteries. The voltages of conversion
reaction are correlated with the calculated enthalpies of formation suggesting that
thermodynamics dominates the observed electrochemical conversion reaction.

INTRODUCTION

The current energy system is strongly based on fossil fuels. However, it is believed to
gradually move from the use of fossil fuels to deploying renewable or nuclear energy sources. A
way to compensate for the variability of energy production by renewable sources is by coupling
these sources with electrical energy storage devices, such as batteries. At least in principle,
batteries are able to store high energy densities and are flexible modular devices, able to use the
same chemistry in a wide range of sizes by adding small cells together, enabling its design from
portable to stationary applications. Up to now, lithium ion batteries have undergone a big
development1, dominating especially portable applications due to the possibility of operating at
room temperature. Despite sodium ion batteries delivering slightly smaller energy densities,
sodium ion technology is lower cost than lithium ion batteries because of the much higher

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 abundance of sodium vs lithium and the possibility of using an aluminum current collector in the
anode electrode (due to the lower density and price of aluminum with respect to copper).

Cathode materials in sodium ion batteries are mostly based on the same insertion
chemistry as lithium ion analogues, with fluorophosphates and layered oxides being the basis of
the first prototypes. However, on the anode side, the impossibility of intercalating bare sodium
ions is a drawback for this technology2,3 as well as an important driver for research. In this
context: high-capacity anodes such as phosphorus4 and alloys5; materials which operate at low
potentials like titanates6 and hard carbon7; and so-called environmentally friendly organic
compounds, have been investigated8. In the field of metal-organic compounds, the presence of
>C=N- and -C≡N- functionalities have shown to be beneficial for sodium ion batteries with poly-
Schiff bases acting as anodes9 and Prussian Blue related systems as cathodes10. We have
proposed and observed for the first time, that -(N=C=N)- containing compounds are
electrochemically active both for lithium and sodium ion batteries11. Transition-metal
carbodiimides are metal-organic materials which have many similarities with oxides: they have
the same charge (-2), similar electronegativity value, more covalency, a bit smaller electronic
bandgap12 and similar structures, but linked to transition-metal ions13 by (N=C=N)2- chains
instead of monoatomic oxide anions. A conversion reaction has been proposed to occur when
TMNCN inserts two Na ions to yield metal nanoparticles, TM, and presumably Na2NCN. This
fact was based on the disappearance of the reflections of the TMNCN, both by in-situ x-ray
diffraction in CuNCN, as well as ex-situ XRD of other phases11, and has been confirmed by in-
situ XRD and Mössbauer studies of FeNCN14.

In this report, the influence of particle size on the performance of MnNCN as anode
materials for sodium-ion batteries is first evaluated. Next, the voltage plateaux and the
irreversibility of first cycles of this and other transition metal carbodiimides (TM= Mn, Fe, Ni or
Co) are studied, evidencing that FeNCN shows the optimum performance among all the studied
phases, which is correlated with its thermodynamic properties.

EXPERIMENTAL

Manganese carbodiimide was prepared by two different routes. The solid state route
following the reported procedure by Liu et al.12 consists of heating a pellet of a mixture of
manganese chloride and zinc carbodiimide at 600 °C under vacuum. Then, the pellet was washed
with water to dissolve any residual chloride and the solution was filtered to obtain the MnNCN
powder. Alternatively, a metathesis reaction in organic media was performed mixing lithium
carbodiimide and manganese chloride in stoichiometric amounts, using THF as the solvent11 at
room temperature, inside the glovebox. After stirring overnight, the solution was centrifuged and
the MnNCN powder was separated from the LiCl which remains dissolved in THF. Lithium
carbodiimide was synthesized with butyl lithium and cyanamide as precursors with a ratio of
2.4:1. However, iron, cobalt and nickel carbodiimides were obtained by synthesizing first the
metal hydrogen dicyanamide in aqueous media (degassed water) under vacuum conditions
(Schlenk) using the metal chlorides and cyanamide as precursors. The second step is the
synthesis of the metal carbodiimide in a sealed glass ampoule. Thereby the hydrogen
dicyanamides are mixed with a eutectic mixture of LiCl and KCl and heated to 400 °C, yielding
the iron15, cobalt and nickel16 carbodiimides.

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 The powder X-ray diffractograms of the carbodiimides were collected on a Bruker D8
instrument with copper radiation (λ = 1.54056 Å) in the 2θ range from 5–80° with a step size of
0.021 for 2 hours using an air sensitive sample holder with Kapton film. For the electrochemical
characterization, the measurements were carried out in CR2032 coin-cell type cells using
metallic sodium as counter and reference electrodes. As the working electrode, 70% in weight of
the metal carbodiimides was hand-milled for 1 hour with 20% in weight of a mixture of C65
(Timcal) and Ketjen black carbon (50:50) in order to improve the electronic conductivity and
10% in weight of polyvinylidene in 1 mL of N-methyl-2-pyrrolidone fluoride as binder. As
carbodiimides are not very good electronic conductors, it is very important to mix the carbon
well with the materials. Then this mixture was casted by Dr. Blade technique in copper foil.
Before using these electrodes, they were dried at 80 °C under vacuum overnight. Between the
working and counter electrode, a Whatman glass fiber GFB/55 was used as separator. 1 M
NaFSI [Sodium bis(fluorosulfonyl)imide] (Solvionic) in methyl tetrahydrofuran (Aldrich) was
selected after testing different electrolyte mixtures. Galvanostatic discharge and charge of the
materials were measured in a MACCOR battery tester at C/10 rate between 0.005–3.0 V vs
Na+/Na (C/10 is the current needed to insert 2 Na ions per formula unit in 10 hours).

RESULTS and DISCUSSION

The differently synthesized MnNCN were characterized by x-ray diffraction. Different

diffractograms were obtained for both synthesized samples. On the one hand, following the solid
state synthesis reported by X. Liu et al.12 a crystalline single phase was obtained as deduced from
the narrowness of the reflections of the diffractogram and through the Rietveld refinement
analysis (Figure 1a). All reflections were in good agreement with the reported trigonal space
group R 3 mH (166) reported17. On the other hand, for manganese carbodiimide synthesized in
organic media, the diffractogram showed very broadened peaks due to the low crystallinity of the
sample (Figure 1b). Additionally, there is a noticeable loss of intensity of the 012 reflection,
which contains the Mn ions of adjacent Mn(NCN) layers. This could be indicative of Mn
vacancies but is also a result of the different sample preparation for XRD data collection, which
involved the use of kapton foil on top of the nanosized powders. The different crystallinity of the
samples is deduced from the different narrowness of the diffraction peaks in the XRD patterns
and confirmed by SEM. The particle size of the crystalline material is 4-5μm and the particles
have a regular bean-like shape as can be seen in the SEM images (Figure 2a) while for the
sample synthesized in organic media, the particle size was smaller (much less than 1μm) with
irregular shape (Figure 2b).

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 Figure 1. Rietveld refinement of MnNCN of crystalline sample (a) and the pattern matching of the sample
synthesized in organic media (b). Light grey points are the experimental pattern, the black curve is the calculated
pattern, dark grey line at the bottom is the difference between experimental and calculated pattern and grey lines are
the corresponding reflections for the space group R 3 mH.

Figure 2. SEM of the different MnNCN synthesized samples where in (a) the MnNCN by the reported method and
in (b) is the SEM image of the synthesized in organic media.

The X-ray powder diffractograms of iron, cobalt and nickel carbodiimides were indexed
using space group P63/mmc (194), which has a NiAs-type structure. Rietveld refinement
confirmed that these materials were synthesized as pure phases (as can be seen in Figure 3a for
FeNCN as representative for the isostructural compounds) and the refined structure is drawn in
Figure 3b. The agreement factors, cell parameters and structural characteristics obtained from the
Rietveld refinements of FeNCN, CoNCN and NiNCN are presented in Table 1.

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 Figure 3. Observed (light grey dots), calculated (solid black line), and difference (dark grey bottom curve) patterns
for the Rietveld analysis from the powder X-ray diffraction data of FeNCN and grey short vertical lines below the
profile indicate the peak positions of all the possible Bragg reflections (a) and the structure of FeNCN obtained from
the refinement (b).

Table 1. Space group, agreement factors, molar mass, cell parameters and structural data obtained from the
Rietveld refinement method of TMNCN (TM=Mn2+, Fe2+, Co2+ or Ni2+), ionic radius from ref. 18 and electronic
configuration of the TM2+.
MnNCN FeNCN CoNCN NiNCN
Space group R 3 mH P63/mmc P63/mmc P63/mmc
Rp 12.12 6.37 7.42 24.45
Rwp 9.98 7.25 8.02 30.89
RB 4.36 2.12 2.98 7.67
χ2 2.92 1.71 3.59 4.23
Molar mass (g/mol) 94.94 95.85 98.93 98.71
a=b (Å) 3.358 (3) 3.280 (2) 3.209 (3) 3.158 (2)
c (Å) 14.337 (4) 9.404 (1) 9.381 (2) 9.249 (4)
ϒ (⁰) 120 120 120 120
V (Å3) 140.037 (4) 87.60 (1) 83.64 (2) 79.91 (6)
V/Z (Å3) 46.679 43.8 41.82 39.96
TM–N (Å) 2.262 2.215 2.168 2.119
Ionic radius TM2+ (Å) 0.83 0.78 0.745 0.69
Electronic configuration 3d5 3d6 3d7 3d8

Electrochemical performance in sodium ion batteries

The reduction of these transition metals to the TM (I) oxidation state seems to be quite
unlikely and all TM2+ cations are expected to be reduced to TM0. At the same time, two sodium
ions could link with the carbodiimide group for a full displacement reaction where two electrons
would be involved. Thus, a C-rate of 1C was calculated as the current needed to insert 2 alkali
ions per transition metal in one hour. The electrochemical performance of MnNCN, produced by
different routes, in sodium ion batteries will be compared and next we extend the comparison of
the performance of the three isostructural members FeNCN, CoNCN and NiNCN.

i) The role of particle size in Manganese carbodiimide

Galvanostatic discharge and charge curves of MnNCN are shown in Figure 4. Crystalline
manganese carbodiimide had a very large 1st discharge capacity (380 mAh/g), at relatively low

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 potential (0.35 V vs Na+/Na average voltage). Despite this high capacity, just 80 mAh/g were
reversible in the first charge as can be seen in Figure 4a. It seemed that manganese carbodiimide
material was almost electrochemically inactive. The electrochemical inactivity of MnNCN
(which was previously reported by Milke et al.19) could be due to the proximity of the plateau to
sodium electroplating. In order to see if by reducing the polarization of the electrode a low
voltage plateau appeared, the common strategy of decreasing particle size was followed. This
way, MnNCN synthesized in organic media was tested. A much larger reversible capacity was
observed in the galvanostatic cycling of MnNCN synthesized in organic media, for sodium ion
batteries, with three different processes being observed in the first discharge at ∼ 0.7, 0.3 and 0.1
V vs Na+/Na (see Figure 4b). The smaller crystal and particle size could shorten the diffusion
path of sodium ions20. As a result, the polarization could be decreased making possible the
observation of electrochemical activity in MnNCN at voltages above Na plating.

Reaching 0.1 V vs Na+/Na, a plateau appeared which is likely due to the conversion reaction of
manganese carbodiimide with sodium. This plateau, as expected, was very close to 0 V vs
Na+/Na, which can be the reason why it did not appear in the crystalline sample as Milke et al
reported19. The redox processes that occur during charge are clearly shown in the zoomed δQδV
plots of the first and second cycle (Figure 4c and Figure 4d). On the first discharge all the redox
processes are seen below 1V vs Na+/Na and most of the capacity is achieved below 0.3V as
shown in the galvanostatic curve (not shown in the dQ/dV). In contrast, during the first and
second galvanostatic charges, the two main processes where the conversion reaction was taking
place are centered at 0.78V vs Na+/Na (which could also have a contribution from deintercalation
of Na ions from graphite layers) and at 2.5 V vs Na+/Na the second and most relevant process,
which implies a very high overpotential.

Figure 4. Galvanostatic first two cycles of crystalline MnNCN for sodium ion batteries in the voltage range of
0.005-3.0 V vs Na+/Na (a), synthesized in organic media (b) and first (c) and second (d) cycle differential capacity
plot of the MnNCN synthesized in organic media.

The cyclability of the crystalline sample was quite bad losing the few observed
electrochemical activity in the 5th cycle as can be seen in Figure 5a. For the less crystalline
material a reversible capacity of 175 mAh/g was achieved for more than 100 cycles with a

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 coulombic efficiency higher than 99% from the tenth cycle onwards (Figure 5b). Thus the main
drawback of this material to be a suitable anode for sodium ion batteries is the intrinsic
overpotential that this material exhibited.

Figure 5. Cyclability of crystalline MnNCN (a) and MnNCN synthesized in organic media (b) for sodium ion
batteries in the voltage range of 0.005-3.0 V vs Na+/Na.

Moreover, given the very low voltage of Na insertion (0.1 V vs Na+/Na), the material is
also prone to suffer from Na plating at the electrode surface if high current densities were used
and thus higher voltage materials would be preferred.

From the preliminary electrochemical measurements of Mn, Fe, Co and Ni

carbodiimides, higher voltage of reduction was expected on increasing the number of d electrons
on the TM2+ transition, i.e. from Mn to Ni11.

ii) Isostructural iron, cobalt and nickel carbodiimides as anode materials in sodium
ion batteries

The electrochemical performance of FeNCN exhibited a process at 1.5 V vs Na+/Na due

to sodium ion adsorption into the carbon layers, providing 30 mAh/g of capacity as can be seen
by the brown line in Figure 6a. At 0.8 V vs Na+/Na an additional process seemed to take place,
giving 150 mAh/g of irreversible capacity. This process was not observed in the published study
of the electrochemistry of FeNCN performed by Sougrati et al. 13. Since they used PC (Propylene
Carbonate) as solvent of the electrolyte, which does not come with a high donor number, thus
blocking the co-intercalation21 into the graphite interlayers, we can deduce that this process is
due to co-intercalation of sodium with solvent molecules into the graphite layers. Finally, at 0.3
V vs Na+/Na a large plateau of 400 mAh/g of capacity appeared which could be attributed to the
conversion reaction (reduction of Fe2+ compound to metallic iron).
The galvanostatic curve of CoNCN was similar to that for FeNCN. During the first
discharge, (see blue line of Figure 6a) a solid solution process appeared at 1.2 V vs Na+/Na
which could be attributed to sodium adsorption into the carbon layer. At 0.8 V vs Na+/Na the
sodium co-intercalation22 occurred in the same voltage as for FeNCN. At the end of the
discharge, there was a plateau corresponding to the conversion reaction centered at 0.5 V vs
Na+/Na. This plateau was assigned to the reduction of CoNCN forming metallic cobalt and the

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 sodium carbodiimide matrix. The 700 mAh/g of capacity delivered are a bit higher than
theoretical (542 mAh/g) which could be attributed to the contribution of the carbon to the
reversible capacity or to cobalt oxidation to higher than 2+ oxidation state. In the same manner,
NiNCN had almost identical galvanostatic cycling compared to CoNCN in terms of shape and
irreversible capacity of the first cycle. During the first discharge (Figure 6a, orange line), the
same processes are believed to occur. The solid solution process observed at 1.1 V vs Na+/Na is
likely due to the sodium adsorption into the carbon layers and the sodium co-intercalation21 with
electrolyte occurs at 0.85 V vs Na+/Na. The plateau centered at 0.65 V vs Na+/Na could be
assigned to the conversion reaction where the reduction of NiNCN into metallic nickel and a
sodium carbodiimide matrix occurred achieving a capacity of 600 mAh/g. In all three cases the
sharp feature observed in the dQ/dV during discharge (Figure 6b) becomes a much broader
feature upon charge which signals a different mechanism of reaction and likely different
products from the starting materials at the beginning of battery assembly.

Figure 6. Galvanostatic first two cycles of FeNCN (brown), CoNCN (blue) and NiNCN (orange) (a) and the
differential capacity plot of the first cycle (b).

Comparing the electrochemical performance of the described compounds, iron carbodiimide is

the most suitable anode material in sodium ion batteries because of its high capacity and the
smallest first cycle irreversible capacity and lowest voltage plateau among all TMNCN.

Besides the interesting electrochemical performance of FeNCN, the study of the electrochemical
reduction of this series of materials is interesting from the thermodynamic point of view. Indeed,
since the proposed reaction taking place in all these phases is the conversion reaction (TMNCN
+ 2 Na+ + 2e- ⇄ Na2NCN + TM), it is possible to rationalize the voltage of reduction of each
phase in terms of the thermodynamic properties of the reaction of formation of TMNCN from
the elemental metal and Na2NCN, although with opposite sign.

As is well known, the voltage of the redox reactions of the materials are related to the Gibbs free
energy values according to equation 1 in which R is the gas constant, T the temperature, K the
equilibrium constant, z the electron number, F the Faraday constant and E the electrochemical
force. Thus, from the experimental voltage of reduction, (E) an experimental Gibbs free energy
for the conversion reaction has been calculated and is listed in Table 2. Since the products of
oxidation are not confirmed to be TMNCN, the voltage of oxidation is not considered.
(Eq. 1)

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 Table 2. Experimentally observed voltage of reduction of the Transition-Metal Carbodiimides in sodium ion
batteries, the calculated Gibbs free energy (∆G) from the experimental reaction voltages and the reported values for
the enthalpy of formation (∆Hf) calculated by Launay and Dronskowski23.
Material Voltage ∆G ∆Hf
V vs Na+/Na kJ/mol kJ/mol
MnNCN 0.2 –38.6 +44.8
FeNCN 0.4 –77.2 +77.3
CoNCN 0.5 –96.5 +101.5
NiNCN 0.65 –125.45 +123.8

Given the recent synthesis of transition-metal carbodiimides (first reported in 2005)13, there are
no experimental reports on measured Gibbs free energies of formation, but a first-principles
theoretical study had predicted the enthalpy of formation for TMNCN (TM = Mn2+, Fe2+, Co2+ or
Ni2+) from their elements23 ahead of synthesis. These values, also listed in Table 2, predicted that
these metastable compounds could not be synthesized from their starting elements, so other
routes (described above) were followed. If we now consider that there is a constant term for the
formation of Na2NCN that we are neglecting for every material, and since the contribution of the
entropy term to the Gibbs free energy is usually much lower compared to the enthalpy term at
room temperature (TS ≪ H), the enthalpy values can be directly compared to the Gibbs
energy of each transition-metal carbodiimide as given in Eq. 2:

Greaction = H – TS ≈ H (Eq. 2)

Indeed, by comparing the two columns on the right in Table 2, a very good correlation is
observed between the here reported Gibbs free energies of the conversion reaction and the
calculated enthalpies of formation of the TMNCN. Therefore, despite that these calculations
have one experimental error which is based on the intrinsic overpotential that these materials
show in their electrochemical behavior and which in materials that operate through conversion
reactions can be very large (the experimentally observed voltage being lower than the
thermodynamic expected one) and despite a theoretical uncertainty in the theoretical formation
enthalpies due to the used exchange-correlation functional (GGA, Perdew-Wang 91) plus other
simplifications, the first reduction reaction in these systems seems to be dominated by the
thermodynamics of the reaction.

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 Figure 7. Scheme of the thermodynamic studies of MnNCN, FeNCN, CoNCN and NiNCN and the relationship with
Na2NCN compound.

A scheme summarizing these correlations (Figure 7) suggests a negative enthalpy of formation

of Na2NCN from its elements as well as the spontaneous reaction observed in the
electrochemical cells when TMNCN are left to react with sodium. Although there seems to be no
experimental value for the enthalpy of formation of Na2NCN, the related alkaline-earth phase
CaNCN is indeed exothermic with an enthalpy of formation of −350.6 kJ/mol24.

CONCLUSIONS

The electrochemical performance of conversion-based anode materials TMNCN (TM =

Mn, Fe, Co, Ni) have been investigated for sodium ion batteries. MnNCN, which has a very low
voltage for Na insertion of ∼0.1V vs Na+/Na shows a strong dependence of its electrochemical
performance on particle size according to SEM, with micron-sized MnNCN being almost
inactive and nano-sized MnNCN synthesized in organic media showing at least half of the
theoretical capacity for the conversion reaction. FeNCN, CoNCN and NiNCN show much higher
reversible capacity and lower overpotential. TMNCN improve the energy density and efficiency
in comparison with oxides due to the lower voltage where the conversion reaction is taking
place. CoNCN and NiNCN exhibited large capacities in the first cycle consisting of a 2e-
conversion process. Among all these TMNCN, FeNCN shows the smallest first cycle irreversible
capacity (220 mAh/g) and a lower suitable voltage plateau (0.3 V vs Na+/Na). This makes
FeNCN a good candidate as an anode material for sodium ion batteries. The voltages of the
conversion reaction in the first cycle, as deduced from the δQ/δV, increases with the number of d
electrons. The corresponding Gibbs free energies of the conversion reaction deduced from the
voltage are in good agreements with the calculated enthalpies of formation of the TMNCN,
suggesting that thermodynamics dominates the conversion reaction.

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 ACKNOWLEDGMENTS

The authors thank Prof. Adelhelm for fruitful discussions on the thermodynamic
calculations. AEB thanks UPV for a FPI-UPV PhD Fellowship. Financial support of FEDER
funds through the Spanish Ministry MINECO MAT-2013-41128-R project granted to the
University of Basque Country and ENE2013-44330-R granted to CIC Energigune is
acknowledged.

REFERENCES

1. J.-M. Tarascon and M. Armand “Issues and challenges facing rechargable lithium batteries”
Nature, 2001, 414, 359-367.
2. V. Palomares, P. Serras, I. Villaluenga, K. B. Hueso, J. Carretero-Gonzalez and T. Rojo “Na-
ion batteries, recent advances and present challenges to become low cost energy storage
systems.” Energy Environ. Sci. 2012, 5, 5884-5901.
3. V. Palomares, M. Casas-Cabanas, E. Castillo-Martínez, M. H. Han and T. Rojo, “Update on
Na-based battery materials. A growing research path”, Energy Environ. Sci., 2013, 6, 2312–
2337.
4. Y. Kim, Y. Park, A. Choi, N. S. Choi, J. Kim, J. Lee, J. H. Ryu, S. M. Oh and K. T. Lee, “An
Amorphous Red Phosphorus/ Carbon Composite as a Promising Anode Material for Sodium Ion
Batteries.” Adv. Mater. 2013, 25, 3045-3049.
5. S. Komaba, Y. Matsuura, T. Ishikawa, N. Yabuuchi, W. Murata and S. Kuze, “Redox reaction
of Sn-polyacrylate electrodes in aprotic Na cell.” Electrochem.Commun. 2012, 21, 65-68.
6. P. Senguttuvan, G. Rousse, V. Seznec, J.-M. Tarascon and R. Palacin, “Na2Ti3O7: Lowest
Voltage Ever Reported Oxide Insertion Electrode for Sodium Ion Batteries” Chemistry of
Materials, 2011, 23, 4109–4111.
7. Zhao J., Zhao L., Chihara K., Okada S., Yamaki J.i., Matsumoto S., Kuze S. and Nakane K.,
“Electrochemical and thermal properties of hard carbon-type anodes for Na-ion batteries” J.
Power Sources, 2013, 244, 752-757.
8. L. Zhao, J. M. Zhao, Y. -S. Hu, H. Li, Z. B. Zhou, M. Armand and L. Q. Chen, “Disodium
Terephthalate (Na2C8H4O4) as High Performance Anode Material for Low-Cost Room-
Temperature Sodium-Ion Battery” Adv. Energy Mater., 2012, 2, 962-965.
9. E. Castillo-Martinez, J. Carretero-Gonzalez and M. Armand; “Polymeric Schiff Bases as Low-
Voltage Redox Centers for Sodium-Ion Batteries” Angew. Chem. Int. Ed. 2014, 54, 5445-5449.
10. Y. Lu, L. Wang, J. Cheng and J.B. Goodenough; “Prussian blue: a new framework
of electrodematerials for sodium batteries” Chem. Commun., 2012, 48, 6544-6546.
11. A. Eguía-Barrio, E. Castillo-Martínez, X. Liu, R. Dronskowski, M. Armand and T. Rojo;
“Carbodiimides: new materials applied as anode electrodes for sodium and lithium ion batteries”
J. Mater. Chem. A 2016, 4, 1608-1611.
12. T. D. Boyko, R. J. Green, R. Dronskowski and A. Moewes; “Electronic Band Gap
Reduction in Manganese Carbodiimide: MnNCN”; J. Phys. Chem. C 2013, 117, 12754.

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267
 13. X. Liu, M. Krott, P. Müller, C. Hu, H. Lueken and R. Dronskowski; “Synthesis, Crystal
Structure, and Properties of MnNCN, the First Carbodiimide of a Magnetic Transition Metal”;
Inorg. Chem.; 2005, 44 (9), 3001.
14. M. T. Sougrati, A. Darwiche, X. Liu, A. Mahmoud, R. P. Hermann, S. Jouen, L. Monconduit,
R. Dronskowski and L. Stievano; “Transition-Metal Carbodiimides as Molecular Negative
Electrode Materials for Lithium- and Sodium-Ion Batteries with Excellent Cycling Properties”
Angew. Chem. Int. Ed. 2016, 55, 5090-5095.
15. X. Liu, L. Stork, M. Speldrich, H. Lueken and R. Dronskowski; “FeNCN and Fe(NCNH)2:
Synthesis, Structure, and Magnetic Properties of a Nitrogen-Based Pseudo-oxide and -hydroxide
of Divalent Iron”; Chem. Eur. J. 2009, 15, 1558.
16. M. Krott, X. Liu, B. P. T. Fokwa, M. Speldrich, H. Lueken and R. Dronskowski; “Synthesis,
Crystal-Structure Determination and Magnetic Properties of Two New Transition-Metal
Carbodiimides:  CoNCN and NiNCN”; Inorg. Chem. 2007, 46, 2204.
17. M. Krott, A. Houben, P. Müller, W. Schweika and R. Dronskowski; “Determination of the
magnetic structure of manganese carbodiimide with diffraction experiments using polarized
neutrons “; Phys. Rev. B; 2009, 80, 024117.
18. L. H. Ahrens; “The use of ionization potentials Part 1. Ionic radii of the elements”; Geochim.
Cosmochim. Acta, 1952, 2 (3), 155.
19. B. Milke, C. Wall, S. Metzke, G. Clavel, M. Fichtner and C. Giordano; “A simple synthesis
of MnN0.43@C nanocomposite: characterization and application as battery material”; J.
Nanopart. Res. 2014, 16, 2795.
20. J. Cui, C. Qing, Q. Zhang, C. Su, X. Wang, B. Yang and X. Huang; “Effect of the particle
size on the electrochemical performance of nano-Li2FeSiO4/C composites” Ionics 2014, 20, 23.
21 B. Jache and P. Adelhelm, “Use of Graphite as a Highly Reversible Electrode with Superior
Cycle Life for Sodium-Ion Batteries by Making Use of Co-Intercalation Phenomena”; Angew.
Chem. Int. Ed., 2014, 53 (38), 10169.
22. B. Jache, J. O. Binder, T. Abe and P. Adelhelm; “A comparative study on the impact of
different glymes and their derivatives as electrolyte solvents for graphite co-intercalation
electrodes in lithium-ion and sodium-ion batteries”; Phys. Chem. Chem. Phys., 2016, 18, 14299.
23. M. Launay and R. Dronskowski, “A Theoretical Study on the Existence and Structures of
Some Hypothetical First-Row Transition-Metal M(NCN) compounds” Z. Naturforsch. 2005,
60b, 437.
24 R. Blachnik: Elemente, anorganische Verbindungen und Materialen, Minerale; J. D’Ans, E.
Lax: Taschenbuch für Chemiker und Physiker; Springer-Verlag, Berlin, Heidelberg, New York
(1998).

Downloaded from https:/www.cambridge.org/core. University of Southern California, on 25 Mar 2017 at 10:15:19, subject to the Cambridge Core terms of use, available at
https:/www.cambridge.org/core/terms. https://doi.org/10.1557/adv.2017.267