Enhanced oxidative stability of non-Grignard magnesium electrolytes through ligand modification

Article abstract of DOI:10.1039/c3cc47277a

A series of non-Grignard Mg-electrolytes with various para-substituents was synthesized starting from commercially-available phenols. More electron-withdrawing substituents shift the anodic stability of the electrolyte by 400 mV. The p-CF₃ substituted phenol exhibits the highest stability of 2.9 V vs. Mg^2+/0, and cycles reversibly with the Chevrel-phase Mo₆S₈ Mg-ion cathode. the Partner Organisations 2014.

Full text of DOI:10.1039/c3cc47277a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enhanced oxidative stability of non-Grignard
magnesium electrolytes through ligand
modification†
Cite this: Chem. Commun., 2014,
50, 5193
Received 23rd September 2013,
Accepted 25th November 2013
Emily G. Nelson, Jeff W. Kampf and Bart M. Bartlett*
DOI: 10.1039/c3cc47277a
www.rsc.org/chemcomm
A
series of non-Grignard Mg-electrolytes with various para- deposition–dissolution characteristics.5 However, these electrolytes
substituents was synthesized starting from commercially-available are nucleophilic and quite sensitive to air and moisture due to
phenols. More electron-withdrawing substituents shift the anodic the Grignard components (RMgX and MgR2). These drawbacks
stability of the electrolyte by 400 mV. The p-CF3 substituted phenol decrease the likelihood of their adoption in battery production,
exhibits the highest stability of 2.9 V vs. Mg2+/0, and cycles reversibly and exclude their potential application in next-generation Mg/S8
with the Chevrel-phase Mo6S8 Mg-ion cathode.
and Mg/O2 batteries.6 The first non-nucleophilic electrolyte,
[Mg2Cl3ꢁ6THF][HMDSAlCl3] (HMDS = hexamethyldisilazide), is
Since Gaston Plante invented the lead-acid battery in 1859, research compatible with sulfur, however stability to 3.2 V vs. Mg2+/0 can
´
in rechargeable batteries has matured to fold several chemistries only be achieved through single-crystal formation.7
into our everyday lives. Notably, nickel-cadmium and nickel-metal
Prior efforts in non-Grignard reagents based on phenolate
hydride, as well as lithium-ion batteries power devices on all size Mg salts show good anodic stability to 2.6 V vs. Mg2+/0, reversible
scales—from portable electronics to hybrid electric vehicles (HEVs). Mg deposition–dissolution, and chemical compatibility with
At present, lithium-ion is the leading technology for HEVs, but they the Mo6S8 Chevrel-phase cathode.8 It also shows resistance to
are still not able to meet all the energy density requirements for air/moisture decomposition, with no loss in anodic stability after
all-electric vehicles (EVs) to become competitive in driving range stirring in air for 3 hours. The authors acknowledge an influence
and cost compared to traditional internal combustion engine of alkyl substituents on the electrochemical performance, but
vehicles. A promising candidate for high energy density batteries do not develop a hypothesis or predictive model to guide new
is Mg-ion batteries. Mg-ion batteries are still in the early stages of synthesis.8 Herein, we build a model rooted in physical organic
development, with the earliest research happening in the 1990s chemistry and a detailed examination of solution speciation to
and early 2000s.1,2 Magnesium is of interest due to its relatively elucidate the role of the ligand in enhancing electrochemical
negative potential (Mg2+/0 is ꢀ2.37 V vs. NHE) and high volumetric stability. We show an expanded electrochemical window of these
capacity (3832 mA h cmꢀ3 for the 2-electron couple). Magnesium air stable electrolytes by 400 mV, bringing them close to the air
also has the benefits of improved safety as there is no dendrite sensitive Grignard stability window, and cycle them as reversible
formation upon deposition,3 and it is low cost compared to lithium Mg-ion batteries.
due to high natural abundance (2.76% by weight of crustal rocks
compared to 18 ppm for lithium).4
We synthesized a series of six ROMgCl salts: [RO = 4-methoxy-
phenolate (MPMC), phenolate (PMC), 4-methyl-phenolate (MePMC),
Finding an electrolyte that displays both reversible Mg deposition 4-tert-butyl-phenolate (BPMC), 4-(trifluoromethyl)-phenolate (FMPMC),
and a wide electrochemical window is a major obstacle that con- and pentafluorophenolate (PFPMC)]. These phenolate salts were
strains research in rechargeable Mg batteries. Aurbach and co-workers then reacted with an AlCl3–THF solution to form the electrolytes.
discovered a breakthrough with Mg organohalo-aluminate salts Detailed synthesis is presented in the ESI.†
2
0
0
00
Mg(AlCl4ꢀnRn R n )2. The optimized composition (AlCl3-(PhMgCl)2)
Fig. 1 presents the cyclic voltammograms of (MPMC)2–AlCl3/THF
exhibits anodic stability to 3.3 V vs. Mg2+/0 and reversible Mg (electron-donating), (PMC)2–AlCl3/THF, and (FMPMC)2–AlCl3/THF
(electron-withdrawing); voltammograms for the other electrolytes
are presented in Fig. S1 of the ESI.† The (FMPMC)2–AlCl3/THF
Department of Chemistry, University of Michigan, 930 N. University Avenue Ann
electrolyte is referenced to the ferrocene Fc+/0 couple in Fig. S2
Arbor 48109-1055, USA. E-mail: bartmb@umich.edu; Fax: +1-734-647-4865
(ESI†). Reversible Mg deposition–dissolution is observed for
† Electronic supplementary information (ESI) available. CCDC 962673 and
all electrolytes with the exception of (MPMC)2–AlCl3/THF,
and the anodic stability of the solutions (shown in Table 1)
962674. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c3cc47277a
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5193--5195 | 5193

View Article Online
Communication
ChemComm
Fig. 2 Hammett plot of para-substituted phenols vs. oxidation onset
potential of non-Grignard electrolytes.
Fig. 1 Typical cyclic voltammograms of Pt electrodes in (a) 0.5
M
(FMPMC)2–AlCl3/THF, (b) (PMC)2–AlCl3/THF, and (c) (MPMC)2–AlCl3/THF
solutions. The scan rate is 25 mV sꢀ1 and the dashed lines are added as
guides to the eye.
Table 1 Conductivity, voltage, anodic current, and Hammett s+ values for
the substituted phenolate electrolyte complexes
Substituent
s+ value
Eona/V
Iab/mA
rc/mS cmꢀ1
Fig. 3 Cyclic voltammograms of (FMPMC)2–AlCl3/THF before (TT) and
after (---) exposure to air for 6 hours. Scan rate is 25 mV sꢀ1
.
OMe
Me
t-Bu
H
CF3
F5
ꢀ0.78
ꢀ0.31
ꢀ0.26
0
0.61
—
2.31
2.58
2.74
2.71
2.90
3.05
257.08
158.94
74.95
64.66
9.474
3.90
1052
1119
736
877
2240
2448
3 V vs. Mg2+/0 (Fig. S3, ESI†). The (BPMC)2–AlCl3/THF electrolyte
does not follow the main trend of s+ values, suggesting that the
steric bulk of the alkyl substituents contributes an additional
factor to the oxidative stability, though further studies are
needed. Fig. 3 shows that the oxidative stability of the electro-
lyte (FMPMC)2–AlCl3/THF remains unchanged after exposure to
air for 6 hours. Reversible Mg deposition–dissolution is still
observed. However, the decrease in efficiency and increased
a
Potential vs. Mg2+/0 at which the onset of anodic current is first
observed. Anodic current at 3 V vs. Mg2+/0
. Conductivity measured
b
c
at 25 1C at 0.5 M (based on Mg2+).
ranks as: (FMPMC)2–AlCl3/THF
> (BPMC)2–AlCl3/THF >
(PMC)2–AlCl3/THF > (MePMC)2–AlCl3/THF > (MPMC)2–AlCl3/THF. polarization is likely due to the slow oxidation of the Mg–Cl
The measured conductivity of the electrolyte solutions does not bonds, as has been previously reported.8
follow this trend; rather, all were B1 mS cmꢀ1 with the exception
Further support of imparting greater oxidative stability
of (FMPMC)2–AlCl3/THF, which exhibits a conductivity almost through electron-withdrawing substituents comes from preparing
twice the others at 2.24 mS cmꢀ1. The higher conductivity is the pentafluorophenolate complex. Although the s+ Hammett values
likely due to the stabilization of the overall negative charge on for all positions on the phenol ring are not known, the increased
the aluminate anion, creating a weaker ion pair.
number of electron-withdrawing groups results in an B100 mV
The electrochemistry illustrates the strong dependence of enhancement in the anodic stability, increasing the electrochemical
phenolate substitution on electrical performance. The Hammett plot window to 3 V for non-Grignard magnesium electrolytes (Fig. S5,
in Fig. 2 of onset potential vs. s+ values (s+ representing Hammett ESI†). This result represents the largest stability window for non-
parameters for phenol) show a linear negative slope, consistent with Grignard electrolytes, and is 400 mV more positive than what has
positive charge build up during oxidation.9 Withdrawing groups been previously reported,8 moving the non-Grignard systems close
destabilize positive charge increasing the difficultly of oxidation.10,11 to the stability window of the Grignard electrolytes.5 The conductivity
This same trend is observed with the anodic current recorded at of (PFPMC)2–AlCl3/THF is 2.44 mS cmꢀ1, similar to that of
5194 | Chem. Commun., 2014, 50, 5193--5195
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
In summary, through a controlled study of substituted phenols,
we have shown that the anodic stability of ROMgCl:AlCl3–THF non-
Grignard electrolytes can be pushed B400 mV past the phenolate
electrolytes previously published by placing very electron-
withdrawing substituents on the phenolate ring. This brings these
non-Grignard electrolyte systems close to the stability of previously
published Grignard based electrolytes. Using phenolate ligands
allows for improved stability in air and lower nucleophilicity, open-
ing the door to exploring high density cathodes such as Mg-air and
Mg-sulfur. The use of physical organic design principles and a
general synthesis method will allow us to further open the stability
window of these types of Mg electrolytes.
We gratefully acknowledge the University of Michigan
Fig. 4 Reversible galvanostatic cycling of Mo6S8 vs. Mg-foil in 0.5 M for generous start-up funding to support this work. We thank
Dr Eugenio Alvarado for assistance with 27Al NMR spectro-
(FMPMC)4–AlCl3/THF electrolyte at C/5.
scopy, Ms Laura Pfund for assistance with Raman spectroscopy,
(FMPMC)2–AlCl3/THF. The voltammogram of (FMPMC)2–AlCl3/ and Mr Xiaoguang Hao for helpful discussions.
THF is also recorded under near steady-state conditions (scan
rate of 1 mV sꢀ1) and presented as Fig. S6 (ESI†), and no new
features emerge. In order to verify the compatibility of the
Notes and references
1 T. D. Gregory, R. J. Hoffman and R. C. Winterton, J. Electrochem.
Soc., 1990, 137, 775–780.
2 D. Aurbach, Z. Lu, A. Schechter, Y. Gofer, H. Gizbar, R. Turgeman,
(FMPMC)2–AlCl3/THF electrolyte with a Mg-ion intercalation
cathode, Mo6S8 was synthesized by known solid-state methods,12
and galvanostatic cycling (C/5 rate = 50 mA cmꢀ2) versus a Mg-foil
auxiliary electrode shows reversible cycling in Fig. 4 (discharge
curves presented as Fig. S7, ESI†).
Y. Cohen, M. Moshkovich and E. Levi, Nature, 2000, 407, 724–727.
3 D. Aurbach, I. Weissman, Y. Gofer and E. Levi, Chem. Rec., 2003, 3,
61–73; Z. Feng, Y. NuLi, J. Wang and J. Yang, J. Electrochem. Soc., 2006,
153, C689–C693; L. Gaddum and H. French, J. Am. Chem. Soc., 1927,
49, 1295–1299.
4 N. Greenwood and A. Earnshaw, Chemistry of the Elements,
Butterworth-Heinemann, Oxford, 2nd edn, 1997.
5 O. Mizrahi, N. Amir, E. Pollak, O. Chusid, V. Marks, H. Gottleib, L. Larush,
E. Zinigrd and D. Aurbach, J. Electrochem. Soc., 2008, 155, A103–A109.
6 J. Muldoon, C. B. Bucur, A. G. Oliver, T. Sugimoto, M. Matsui,
H. S. Kim, G. D. Allred, J. Zajicek and Y. Kotani, Energy Environ. Sci.,
2012, 5, 5941–5950.
7 H. S. Kim, T. S. Arthur, G. D. Allred, J. Zajicek, J. G. Newman,
A. E. Rodnyansky, A. G. Oliver, W. C. Boggess and J. Muldoon,
Nat. Commun., 2011, 2(427), 1–6.
In solution, a mixture of complexes is present in a dynamic
equilibrium, previously reported for aluminium alkoxides.13,14 In an
effort to assign the 27Al NMR spectra of the as-synthesized electrolyte,
we prepared a series of solutions comprised of differing ratios between
the Lewis acid (AlCl3) and base (FMPMC). Details are presented in the
ESI,† and spectral assignments are based on prior work.15–17
The electron-withdrawing phenolates, FMP and PFP show the
greatest number of (RO)xAlCl4ꢀx species in solution, with x = 1 ꢀ 4
species all present. The remaining electrolytes do not show the
tetrakis-ligated aluminum species; the 27Al NMR spectrum of the
most electron-donating complex (MPMC)2–AlCl3/THF shows only
the bis- and mono-ligated complexes in smaller intensities relative
8 F. Wang, Y. Guo, J. Yang, Y. Nuli and S. Hirano, Chem. Commun.,
2012, 48, 10763–10765.
9 E. V. Anslyn and D. A. Dougherty, Modern Physical Organic Chemistry,
University Science Books, California, 2006, pp. 445–453.
10 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165–195.
to the Al2Cl6 species (Tables S1 and S2, ESI†). One possible explana- 11 Y. Guo, J. Yang, Y. Nuli and J. Wang, Electrochem. Commun., 2010,
12, 1671–1673.
tion for this observation is that ligand exchange between MPMC and
AlCl3 is slower or incomplete. The X-ray crystal structure of the com-
12 E. Lancry, E. Levi, Y. Gofer, M. Levi, G. Salitra and D. Aurbach, Chem.
Mater., 2004, 16, 2832–2838.
pound crystallized from (MPMC)2–AlCl3/THF (Fig. S8, ESI†) shows that 13 S. Hermanek, O. Kriz, J. Fusek, Z. Cerny and B. Casensky, J. Chem.
Soc., Perkin Trans. 2, 1989, 987–992.
14 A. Malmvik, U. Obenius and U. Henriksson, J. Chem. Soc., Perkin
ligand exchange is incomplete, resulting in a Mg cubane-like structure
with bridging methoxyphenolate ligands. In contrast, the X-ray crystal
Trans. 2, 1986, 1905–1910.
structure of (FMPMC)2–AlCl3/THF shows the expected ligand exchange 15 L. Ohman and U. Edlund, Aluminum-27 NMR of Solutions, eMagRes,
reaction forming [Mg2Cl3(THF)6]+ cations and [(FMP)AlCl3]ꢀ anions
2007; R. Benn and A. Rukinska, Angew. Chem., Int. Ed. Engl., 1986,
25, 861–881.
16 O. Mizrahim, N. Amir, E. Pollak, O. Chusid, V. Marks, H. Gottliedb,
(Fig. S9, ESI†). The 27Al NMR assignments are further supported by
the mass spectrometry and by Raman spectra (assignments based on
prior work18–20) presented in the ESI† (Fig. S10 and S11).
L. Larush, E. Zinigrad and D. Aurbach, J. Electrochem. Soc., 2008,
155, A103–A109.
17 H. Gizbar, Y. Vestfrid, O. Chusid, Y. Gofer, H. E. Gottlieb, V. Marks
and D. Aurbach, Organometallics, 2004, 23, 3826–3831.
In an effort to distinguish which solution species is responsible
for the wider potential window, bulk electrolysis of (FMPMC)2–AlCl3 18 Y. Vestfried, O. Chusid, Y. Goffer, P. Aped and D. Aurbach, Organo-
was carried out at 3.5 V vs. Mg2+/0 for 90 minutes. Comparing the
metallics, 2007, 26, 3130–3137.
19 A. Decken, H. Jenkins, G. Nikiforov and J. Passmore, Dalton Trans.,
27Al NMR spectra before and after electrolysis shows decomposition
2004, 2496–2504.
of the in situ formed aluminium complex (Fig. S12, ESI†) and 20 N. Tomar, E. Ghanti, A. Bhagi and R. Nagarajan, J. Non-Cryst. Solids,
2009, 355, 2657–2662.
21 Y. Guo, F. Zhang, J. Yang, F. Wang, Y. NuLi and S. Hirano, Energy
suggests that Lewis acids are responsible for the anodic stability of
these non-Grignard electrolytes, similar to what has been reported
Environ. Sci., 2012, 5, 9100–9106; Y. Guo, F. Zhang, J. Yang and
F. Wang, Electrochem. Commun., 2012, 18, 24–27.
for their Grignard congeners.17,21
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5193--5195 | 5195