Synthesis and characterization of a family of molecule-based magnets containing methyl-substituted phenyltricyanoethylene acceptors

Full text of DOI:10.1016/j.jmmm.2019.165953

JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials
journal homepage: www.elsevier.com/locate/jmmm
Research articles
Synthesis and characterization of a family of molecule-based magnets
containing methyl-substituted phenyltricyanoethylene acceptors
James A. King Jr., Christopher L. Houser, Ryan E. Corkill, Gordon T. Yee⁎
Virginia Tech, Department of Chemistry, Blacksburg, VA 24061, United States
1. Introduction
on the metal and organic radical containing unpaired electrons leads to
stronger antiferromagnetic direct exchange. Verdaguer has made a si-
milar argument for superexchange in Prussian Blue magnets [17]. We
noted in our original paper that steric effects due to substitution in the
2- and/or 6-positions (ortho- to the olefin) could also be playing a role
and were correlated with higher Tc as well. These effects were ascribed
to greater spin density on the nitrogen atoms of the nitrile functional
groups that coordinate to the V(II) [13]. For instance, we recently re-
ported the corresponding chloro-substituted PTCEs [18] which largely
follow the above trends but because chlorine atoms are larger than
fluorine atoms, steric effects appear to play a larger role in determining
Tc.
In 1991, Manriquez et al. reported the first room temperature mo-
lecule-based magnet, which was synthesized from bis(benzene)vana-
dium(0) and tetracyanoethylene (TCNE) in dichloromethane [1]. The
formula of the magnetic phase is given by V[TCNE]x·y(CH2Cl2) where x
is very close to 2 and y is typically small. The reaction is thought to
proceed by the production of two TCNE radical anions from one-elec-
tron reduction each by the vanadium(0). Each vanadium ultimately is
thus oxidized to vanadium(II). Coordinate covalent bonding by each
TCNE radical anion via two or more of its nitrogen atom lone pairs to
these cations gives rise to a network. However, direct evidence of this
structure is limited to EXAFS data [2] because the material is not
crystalline.
In this contribution, we explore the effects of analogous methyl
substitutions on the phenyl ring of PTCE. Because methyl groups are
electron donating, making the acceptors more difficult to reduce, the
obvious expectation was that Tc of the resulting magnets should always
be lower relative to that of the parent PTCE. This was not observed.
Reaction of these new acceptors with V(CO)6 results in new ferri-
magnets with ordering temperatures that range from 159 K to 244 K.
The observed trends in Tc confirm that the localization of spin density
due to loss of conjugation is very important and can be enough to
overcome the electron donating effects, resulting in unexpectedly
higher Tc for certain substitution patterns. These magnetic results are
evaluated in the context of trends in electrochemistry and the results of
density functional theory (DFT) calculations.
Since that time, there have been efforts to optimize Tc, the Curie
temperature, of this compound by changing to a different source of
metal, including hexacarbonylvanadium, V(CO)6 [3]. Researchers have
also explored replacing the vanadium with other metals [4–7] and re-
placing the TCNE with other organic acceptors [8–11] to create new
magnetic solids. Relevant to the latter strategy, we have previously
examined the replacement of one or two of the CN functional groups on
TCNE with phenyl rings to produce phenyltricyanoethylene (PTCE) and
dicyanostilbene [12] scaffolds, respectively. An advantage of this ap-
proach is that each of the positions on the phenyl ring is a potential site
for derivation, thus providing the opportunity to create not just new
acceptors, but new families of acceptors in which properties may be
systematically varied. The most obvious characteristic that can be
manipulated is the reduction potential of the acceptor.
2. Results and discussion
For example, we have reported families of fluoro-substituted PTCEs
where up to five fluorine atoms are placed on the phenyl ring [13,14].
Reactions of each of these building blocks with V(CO)6, yield a family of
magnetic coordination networks with Tc’s up to 315 K (42 °C) in the
case of 2,3,5,6-tetrafluorophenyltricyanoethylene [15]. Higher Tc’s in
this family were generally correlated with greater fluorine substitution,
hence easier reduction of the acceptor and a lower energy π* orbital in
which the unpaired electron resides. We modeled this behavior quali-
tatively based on an overlap model used by Carlegrim for V[TCNE]2
[16]. In this model, a better match between the energy of the orbitals
The three mono-methyl-substituted examples (b-d), as well as the
2,6-dimethyl- (e), 2,4,6-trimethyl- (f) and pentamethyl- (g) derivatives
were examined and compared to the parent PTCE(a). (Fig. 1) These
acceptors were synthesized by modification of a previously reported
three-step procedure [19] using commercially available starting mate-
rials, appropriately methyl-substituted benzaldehydes. Overall yields
from the benzaldehyde ranged from 14 to 30%. Each of these com-
pounds was purified by column chromatography and characterized by
1H NMR, IR, high resolution mass spectrometry and cyclic voltam-
metry.
⁎ Corresponding author.
E-mail address: gyee@vt.edu (G.T. Yee).
https://doi.org/10.1016/j.jmmm.2019.165953
Received 8 July 2019; Received in revised form 3 September 2019; Accepted 4 October 2019
Availableonline11October2019
0304-8853/©2019ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:JamesA.King,etal.,JournalofMagnetismandMagneticMaterials,https://doi.org/10.1016/j.jm
m
m.2019.165953

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
Table 2
Summary of magnetic data for six methyl-substituted acceptors paired with V
(CO)6.
Acceptor
Tc
(
3) K
Msat (emu-G/mol)
PTCE
215
244
200
182
241
209
159
5060
4080
6260
5420
3750
5130
4290
2-MePTCE
3-MePTCE
4-MePTCE
2,6-Me2PTCE
2,4,6-Me3PTCE
Me5PTCE
compounds and what would be expected for 4A ground state due to
vanadium(II) in a presumed pseudo-octahedral ligand field and organic
radical anions. A representative example of a plot of M vs. H is shown
below for V[4-MePTCE]2. (Fig. 3). The saturation magnetization is
roughly the expected value for one net unpaired electron per formula
unit (5585 emu-G/mol) and the materials are essentially saturated in a
field of 100 G. This is consistent with antiferromagnetic nearest
neighbor coupling between two S = 1/2 organic radical and the S = 3/
2 vanadium ion. In a three-dimensional coordination network, this
gives rise to bulk ferrimagnetism.
Fig. 1. Methyl-substituted PTCEs investigated herein.
Table 1
Summary of infrared data for six neutral methyl-substituted acceptors and
paired with V(CO)6.
Magnet
Neutral acceptor νC≡N
Magnetic phase νC≡N
(cm−1
)
(cm−1
)
V(2-MePTCE)2
V(3-MePTCE)2
V(4-MePTCE)2
V(2,6-Me2PTCE)2
V(2,4,6-Me3PTCE)2
V(Me5PTCE)2
2240, 2247
2234, 2196
2235, 2198
2240, 2242
2238, 2245
2242, 2264
2215, 2119
2207, 2116
2210, 2115
2202, 2116
2209, 2114
2219, 2113
The electronic effects of the methyl group on the properties of each
acceptor can be compared to those of the fluoro-group by examining the
acid properties of the analogous substituted phenols. The pKa’s of 2-
fluoro-, 3-fluoro- and 4-fluorophenol are 8.81, 9.28 and 9.81 [19], re-
spectively. These values are lower than the pKa of phenol (9.95), which
indicate the fluoro-groups are electron-withdrawing relative to a hy-
drogen atom. In contrast, the pKa’s of 2-methyl-, 3-methyl- and 4-me-
thylphenol are 10.28, 10.09 and 10.26, respectively, indicating their
weakly electron donating nature [20]. This trend is reflected in re-
duction potentials of the corresponding methyl-substituted PTCEs as
can be seen in Table 3 relative to Ag/AgCl reference electrode, where,
as expected, all three monomethyl reductions were more difficult than
the parent PTCE and the meta methyl group (3-Me) has the smallest
effect. As expected, addition of more methyl groups makes reduction
increasingly more difficult.
Subsequent reaction of each acceptor with hexacarbonylvanadium
(0), in dichloromethane in an inert atmosphere glovebox yields a black
insoluble solid that can be isolated by simple vacuum filtration and
washing with additional dichloromethane. The resulting materials
cannot be dissolved in non-coordinating solvents such as ether or
hexanes. The solids were characterized by IR (focusing on the CN
stretches) and elemental analysis. The observed CN stretching fre-
quencies for each acceptor are indicative of the oxidation state.
(Table 1) Specifically, the absorptions of the neutral acceptor are red-
shifted upon one-electron reduction to the radical anion, the form that
exists in the magnetic phase. We have previously reported this for PTCE
(neutral species; oxidation state = 0) for which the CN stretches are
observed 2235 and 2233 cm−1 to CN stretches at 2210 and 2129 cm−1
for the radical anion (oxidation state = −1) [13].
To address the question of whether there is anything unusual about
these data, we have used DFT calculations to determine gas phase
electron affinities. The electron affinities were calculated according to
protocols recommended by Rienstra-Kiracofe and co-workers [21],
where the electron affinity (EA) in our case is the energy of the opti-
mized, neutral acceptor with the energy of the optimized, anionic form
of the acceptor subtracted out. Because solvation effects should be si-
milar for all of the acceptors, we expect a linear correlation between
these two quantities, as is observed. (Fig. 4).
The elemental analysis for each magnetic phase suggests the stoi-
chiometric ratio of one donor to two acceptors, V[MexPTCE]2·yCH2Cl2
(vide infra). This is consistent with all previous reports of similar
compounds and results in an overall neutral network of vanadium 2+
cations bridged by radical anions. In each case, the elemental analysis
data include approximately 0.5 dichloromethane molecules which are
presumed to be solvents of crystallization. Although we have no
structural information on these amorphous materials, they are thought
to be three-dimensional networks of alternating S = 3/2 cations and
S = 1/2 organic radicals that are antiferromagnetically coupled
through direct exchange.
Plotting electron affinity vs. Curie temperature reveals not a simple
linear correlation, but two distinct linear series that are distinguished
by the presence or absence of substitution in the 2-position. (Fig. 5) In
the former series (blue diamonds), the radical anions are nearly planar
in the ground state (vide infra) and the expected electron donating
effects of the methyl groups are observed to lower the Curie tempera-
tures. This is based on an overlap model in which the strength of the
exchange is correlated with the energy match of the two orbitals that
contain unpaired electrons [13].
The observed Tc’s of the insoluble networks range from 159 to
244 K. (Table 2) Notably some Tc’s are higher than the parent un-
substituted compound, despite the fact that the methyl group is electron
donating, which was predicted to lead to lower Tc. Representative plots
of magnetization as a function of temperature (M vs. T) for all six
compounds are shown in Fig. 2. Tc values were determined by extra-
polating the steep part of the curve to zero and averaging at least two
independent measurements. The error in the measurements is 3 K.
The steepness of each transition suggests that the materials are rela-
tively uniform in composition.
For the latter series (blue circles), the presence of substitution only
ortho- to the olefin (2- or 2,6-), results in magnetic phases with sub-
stantially higher Tcs than that derived from PTCE. This surprising result
indicates something other than the simple electron donating effect is
operative and suggests that the change in geometry due to steric re-
pulsion can play a large role. From the data, it appears that further
addition of methyl groups lowers the Tc, which can be explained by the
electron donating effect asserting itself. However, the presence of more
methyl groups could also be inhibiting efficient network formation and
we cannot rule this out. To gain insight into the contributions to the
Every new magnet exhibits nearly zero coercivity, consistent with
the previously reported TCNE and fluoro- and chloro-substituted
2

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
Fig. 2. Magnetization vs. temperature for the six magnets described herein.
Fig. 3. Magnetization vs. applied field for V[4-MePTCE]2.
Fig. 4. Linear correlations between experimental redox potential and calcu-
lated electron affinity.
Table 3
Electrochemical data for the six acceptors.
Acceptor
Ered (V)
Acceptor
Ered (V)
PTCE
−0.397
−0.440
−0.402
−0.433
2,6-Me2PTCE
–0.456
–0.484
–0.515
2-MePTCE
3-MePTCE
4-MePTCE
2,4,6-Me3PTCE Me3PTCE
Me5PTCE
increased Tc, we used DFT calculations to evaluate the geometry and
the spin density distribution of the radical anion.
Within the PTCE molecule or radical anion, one can define a dihe-
dral angle between the plane of the phenyl ring and the plane of the
olefin moiety. When the dihedral angle is 0°, the π system of the phenyl
ring is fully conjugated with the pi orbitals of the olefin. If the angle is
90°, the two planes of the molecule are orthogonal, and there is no
conjugation. DFT calculations indicate that for PTCE, 3-MePTCE and 4-
MePTCE radical anions, the ground state is nearly planar at 12° and the
barrier to planarity is negligible. Substitution at only the 2-position
leads to a minimum in the dihedral angle of about 40° whereas if there
is substitution at both the 2- and 6-positions the minimum is at about
65°. (Fig. 6). The barriers to planarity are ~0.5 kJ/mol for no sub-
stitution in the 2- or 6-positions, ~24 kJ/mol for 2-MePTCE and
~53 kJ/mol for all acceptors with substitutions in both the 2- and 6-
positions.
Fig. 5. Electron affinity plotted against Curie temperature.
density on the nitrile nitrogen atoms. Fig. 7 shows that with no sub-
stitution in the 2- and/or 6-positions, the total spin density on the ni-
trogen atoms is largely constant. It increases from 0.42 electrons up to
0.48 electrons as the dihedral angle increases. The added spin density
comes at the cost of spin density on the phenyl ring, consistent with the
loss of conjugation.
The loss of conjugation results in localizing greater unpaired spin
3

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
identity of the acceptor and/or change the connectivity of the network,
providing another pathway for magnetic coupling. Since the methyl
group would be a poor leaving group for this reaction pathway, it
presumably does not occur, and so a “normal” depression of Tc is ob-
served due to the electron donating ability of the methyl.
To summarize, several molecule-based magnets derived from hex-
acarbonylvanadium(0) and various methyl-substituted phenyl-
tricyanoethylene one-electron acceptors were investigated. Two ex-
amples exhibited Tc’s that were surprisingly high based on the
expectation of electron-donation as the principal determining factor.
For compounds based on essentially planar acceptors that have similar
calculated spin densities on the nitrile nitrogen atoms (PTCE, 3-Me and
4-Me) there is a simple correlation between Tc and electron affinity.
Substitution at the 2- and/or 6-positions, however, profoundly changes
the geometry of the acceptor. DFT calculations suggest a twist to non-
coplanarity of the phenyl ring and olefin, placing greater unpaired spin
density on the nitrile nitrogen atoms. It is precisely these atoms that are
bonded to the vanadium(II) cations and are engaged in direct magnetic
exchange. This rationalizes why 2-MePTCE and 2,6-Me2PTCE support
higher Tc’s than the parent PTCE. These results suggest that spin density
plays a significant role in the determination of Tc, one at least as im-
portant as reduction potential. This, in turn, suggests that the dicya-
nostilbene family of acceptors, in which two nitrile function groups are
replaced with appropriately substituted phenyl rings, pushing even
more spin density onto the nitrile nitrogen atoms, could hold some
surprises.
Fig. 6. Dihedral angle plotted against Curie temperature.
3. Experimental
The cyclic voltammetry experiments were performed using a CH
Instruments model 600A potentiostat. The measurements were per-
formed on 5 mM solutions in acetonitrile/0.1 M [n-Bu4N][PF6] and
taken between the potential range of 0 and −700 mV at a scan rate of
100 mV/s using a polished carbon electrode with Ag/AgCl as the re-
ference. Magnetic experiments were performed on a 7 T Quantum
Design MPMS SQUID magnetometer. Samples for magnetic experiments
were loaded into specially designed tubes in the glovebox, and flame-
sealed under vacuum [24]. Experiments involving measurement of
magnetization as a function of temperature were performed by first
cooling the samples in 50 G and then measuring upon warming from
5 K to 300 K in a 5 G applied field. Measurements involving magneti-
zation as a function of applied field were performed at 5 K. All calcu-
lations were performed using the B3LYP density functional with the
6–31++G** basis set within Gaussian03 [25]. Beginning with the
optimized geometry of PTCE, the neutral and anionic forms of each
acceptor were calculated using geometry optimization. From each cal-
culation, the neutral and anionic energies, the Mulliken spin densities
and dihedral angles were determined.
Fig. 7. Total spin density on the nitrile nitrogen atoms plotted against Curie
temperature.
Greater spin density on the nitrogen atoms bonded to the vanadium
results in greater stabilization of the antiferromagnetically coupled
state for Heisenberg Hamiltonian, H = 2J S1 S2 [22] where J is the
coupling constant and S1 and S2 are the spins on the nitrogen atom and
vanadium ion, respectively. All other things being equal, larger values
of S1 would yield greater stabilization of the antiferromagnetically
coupled state, leading to higher Tc.
Where direct comparisons are possible, all but one of the Curie
temperatures of the methyl-substituted PTCE magnets are lower than its
fluorine- and chlorine-substituted analogs. (Table 4). The one exception
is when substitution is in the 4-(para) position. Although we previously
had no good explanation for the extremely low Tc for the magnets de-
rived from 4-FPTCE and 4-ClPTCE, (they are much lower than that for
PTCE (213 K), when it was expected to be much higher) we have since
identified a possible culprit: a fluorine or chlorine atom on a benzene
ring para- to a good electron withdrawing group is susceptible to nu-
cleophilic aromatic substitution [23]. The extremely electron-rich V(0)
could react to displace the halogen. Such a reaction would change the
The following procedure for the synthesis of the 2-methylphenyl-
tricyanoethylene is representative. It is based on a literature procedure
[18] and the syntheses for the 3-methyl-, 4-methyl-, 2,6-dimethyl-,
2,4,6-trimethyl-, and pentamethyl acceptors follow the same procedure
with minor alterations (noted). All reagents were used as received,
except for the following: the dichloromethane used in the magnet
synthesis was distilled from P2O5, and the V(CO)6 was prepared, using a
procedure from the literature, from [Et4N][V(CO)6] [26].
Table 4
3.1. 2-(2-methylphenyl)-1,1,2-tricyanoethylene, 2-MePTCE
Comparison of Tc for methyl-, fluoro-, and chloro-substituted [13,14,18] mag-
nets.
3.1.1. 2-(2-methylphenyl)-1,1-dicyanoethylene (1)
Substitution position
Methyl
Fluoro
Chloro
1 equivalent of 2-methylbenzaldehyde (3.573 g, 29.7 mmol) and an
excess of malononitrile (2.13 g, 32.2 mmol) were added to ~80 ml of
ethanol (100%) in a 100 ml beaker along with a stir bar. The reagents
and solvent were stirred until they became uniform, at which time 1
drop of piperidine was added. After stirring for 10 min, the stir bar was
removed, and the beaker was covered with parafilm and set aside for
2-
244
200
182
241
159
257
233
160
300
306
271
243
146
285
n/a
3-
4-
2,6-
Penta-
4

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
the night. White crystals began to form within two hours of the com-
pletion of the reaction. The white crystals were collected the following
day, washed with ethanol (95%), and weighed. 2.43 g (49%) of 1 were
2.44 (s, 3H) ppm versus TMS. HRMS-ESI (m/z, [M−H]-). Calcd for
C12H8N3: 194.2162. Found: 194.0724.
collected. IR (KBr): ν(CN), 2232 cm−1 and 2177 cm−1
.
1H NMR
3.3. 2-(4-methylphenyl)-1,1,2-tricyanoethylene, 4-MePTCE
(400 MHz, CDCl3) δ 8.06 (m, 1H), 7.46 (m, 2H), 7.31 (m, 2H), 2.41 (s,
3H) ppm versus TMS.
4-MePTCE was synthesized using a similar procedure, using 4-me-
thylbenzaldehyde as the starting material.
3.1.2. 2-(2-methylphenyl)-1,1,2-tricyanoethane (2)
One equivalent of 1 (1.20 g, 7.13 mmol) was dissolved in 100 ml of
ethanol (100%) in a 600 ml beaker with minimal heating. To this so-
3.3.1. 2-(4-methylphenyl)-1,1-dicyanoethylene
Yield: 89%. IR (KBr): ν(CN), 2229 cm−1 and 2202 cm−1
.
1H NMR
lution was added approximately
2 equivalents of KCN (1.04 g,
(400 MHz, CDCl3) δ 7.78 (d, 2H), 7.30 (d, 2H), 2.42 (s, 3H) ppm versus
16.0 mmol) dissolved in ice cold H2O, along with more ice-cold H2O in
order to fill the beaker to 400 ml. The reaction solution was im-
mediately placed in an ethanol/H2O ice bath at ~−7 °C. The reaction
was allowed to stir for approximately 1 hr, after which concentrated
HCl was added to acidify the solution. Upon adding the HCl, a white
solid began to precipitate from the solution. After attaining a pH around
6, the reaction was allowed to stir for 10 min, and the reaction beaker
was covered with parafilm containing holes to release gaseous HCN.
The beaker was then placed in the refrigerator overnight to allow the
product to continue forming. The crystals were collected and washed
with cold H2O the next day and were then dried under vacuum for
approximately 4 hr, weighed, and stored. 1.182 g (85%) of 2 were
TMS.
3.3.2. 2-(4-methylphenyl)-1,1,2-tricyanoethane
Yield: 88%. IR (KBr): ν(CN), 2249 cm−1 and 2259 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 7.35 (d, 2H), 7.29 (d, 2H), 4.38 (d, 1H), 4.17 (d,
1H), 2.38 (s, 3H) ppm versus TMS.
3.3.3. 2-(4-methylphenyl)-1,1,2-tricyanoethylene
Yield: 28%. Mp: 120.5–121.0 °C. IR (KBr): ν(CN), 2235 cm−1 and
2198 cm−1
.
1H NMR (400 MHz, CDCl3) δ 7.92 (d, 2H), 7.38 (d, 2H),
2.47 (s, 3H) ppm versus TMS. HRMS-ESI (m/z, [M−H]-). Calcd for
C12H8N3: 194.2162. Found: 194.0722.
isolated. IR (KBr): ν(CN), 2249 cm−1 and 2261 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 7.59 (m, 2H), 7.37 (m, 1H), 7.27 (m, 1H), 4.69 (d,
3.4. 2-(2,6-dimethylphenyl)-1,1,2-tricyanoethylene, 2,6-Me2PTCE
1H), 4.18 (d, 1H), 2.41 (s, 3H) ppm versus TMS.
2,6-Me2PTCE was synthesized using a similar procedure, using 2,6-
dimethylbenzaldehyde as the starting material.
3.1.3. 2-(2-methylphenyl)-1,1,2-tricyanoethylene
One equivalent of 2 (1.10 g, 5.64 mmol) was added to a 250 ml
round-bottom flask along with 40 ml of diethyl ether. An excess of n-
chlorosuccinimide (NCS) was then added with 80 ml of H2O to the
round-bottom, and the 2-layered solution was then stirred until all of
the NCS dissolved (~30 min). The organic layer turned light yellow in
color as more NCS dissolved. After stirring, the aqueous layer was re-
moved using a separatory funnel, and the organic layer was washed 3
times with H2O and dried with anhydrous Na2SO4. The ether was then
removed by rotoevaporation, leaving yellow oil in the bottom of the
flask that was dried under vacuum overnight. The resulting solid
(yellow, stained with purple) was dissolved in dichloromethane and
loaded onto a silica gel flash column with dichloromethane as the
mobile phase. The desired product was the first compound to elute off
of the column. The product was dried under vacuum, forming a pale-
yellow solid that was weighed and stored in the glovebox. Yield:
0.791 g (73%). Mp: 118.2–118.8 °C. IR (KBr): ν(CN), 2240 cm−1 and
3.4.1. 2-(2,6-dimethylphenyl)-1,1-dicyanoethylene
Yield: 43%. IR (KBr): ν(CN), 2237 cm−1 and 2201 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 8.13 (s, 1H), 7.28 (m, 2H), 7.16 (m, 1H), 2.55 (s,
6H) ppm versus TMS.
3.4.2. 2-(2,6-dimethylphenyl)-1,1,2-tricyanoethane
Yield: 87%. IR (KBr): ν(CN), 2246 cm−1 and 2269 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 7.27 (m, 2H), 7.15 (m, 1H), 4.92 (d, 1H), 4.38 (d,
1H), 2.54 (s, 6H) ppm versus TMS.
3.4.3. 2-(2,6-dimethylphenyl)-1,1,2-tricyanoethylene
Yield: 62%. Mp: 116.8–117.5 °C. IR (KBr): ν(CN), 2240 cm−1 and
2242 cm−1
.
1H NMR (400 MHz, CDCl3) δ 7.34 (m, 1H), 7.17 (d, 2H),
2.33 (s, 6H) ppm versus TMS. HRMS-ESI (m/z, [M−H]-). Calcd for
2247 cm−1
.
1H NMR (400 MHz, CDCl3) δ 7.52 (m, 2H), 7.37 (m, 2H),
C13H10N3: 208.243. Found: 208.088.
2.48 (s, 3H) ppm versus TMS. HRMS-ESI (m/z, [M−H]-). Calcd for
3.5. 2-(2,4,6-trimethylphenyl)-1,1,2-tricyanoethylene, 2,4,6-Me3PTCE
C12H8N3: 194.2162. Found: 194.0724.
2,4,6-Me3PTCE was synthesized using a similar procedure, using
2,4,6-trimethylbenzaldehyde as the starting material.
3.2. 2-(3-methylphenyl)-1,1,2-tricyanoethylene, 3-MePTCE
3-MePTCE was synthesized using a similar procedure, using 3-me-
thylbenzaldehyde as the starting material.
3.5.1. 2-(2,4,6-trimethylphenyl)-1,1-dicyanoethylene
Yield: 52%. IR (KBr): ν(CN), 2239 cm−1 and 2216 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 8.10 (s, 1H), 6.92 (m, 2H), 2.28 (s, 3H), 2.26 (s,
3.2.1. 2-(3-methylphenyl)-1,1-dicyanoethylene
Yield: 55%. IR (KBr): ν(CN), 2230 cm−1 and 2195 cm−1
.
1H NMR
6H) ppm versus TMS.
(400 MHz, CDCl3) δ 7.71 (s, 1H), 7.68 (m, 2H), 7.41 (m, 2H), 2.40 (s,
3.5.2. 2-(2,4,6-trimethylphenyl)-1,1,2-tricyanoethane
3H) ppm versus TMS.
Yield: 85%. IR (KBr): ν(CN), 2255 cm−1 and 2272 cm−1
.
1H NMR
(400 MHz, CDCl3) δ 6.94 (m, 2H), 4.86 (d, 1H), 4.34 (d, 1H), 2.46 (s,
3.2.2. 2-(3-methylphenyl)-1,1,2-tricyanoethane
Yield: 87%. IR (KBr): ν(CN), 2253 cm−1 and 2267 cm−1
.
1H NMR
6H), 2.26 (s, 3H) ppm versus TMS.
(400 MHz, CDCl3) δ 7.35 (m, 4H), 4.39 (d, 1H), 4.20 (d, 1H), 2.41 (s,
3H) ppm versus TMS.
3.5.3. 2-(2,4,6-trimethylphenyl)-1,1,2-tricyanoethylene
Yield: 33%. Mp: 108.2–108.9 °C. IR (KBr): ν(CN), 2238 cm−1 and
3.2.3. 2-(3-methylphenyl)-1,1,2-tricyanoethylene
2245 cm−1
.
1H NMR (400 MHz, CDCl3) δ 6.99 (s, 2H), 2.32 (s, 3H),
Yield: 62%. Mp: 109.1–109.8 °C. IR (KBr): ν(CN), 2234 cm−1 and
2.31 (s, 6H) ppm versus TMS. HRMS-ESI (m/z, [M−H]-). Calcd for
2196 cm−1
.
1H NMR (400 MHz, CDCl3) δ 7.79 (m, 2H), 7.45 (m, 2H),
C14H12N3: 222.270. Found: 222.104.
5

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
3.6. 2-(pentamethylphenyl)-1,1,2-tricyanoethylene, Me5PTCE
3.12. V[Me5PTCE]2
Me5PTCE was synthesized using a similar procedure, using penta-
methylbenzaldehyde as the starting material.
Yield: 69.6%. IR (KBr): ν(CN), 2219 cm−1 and 2113 cm−1. Anal.
Calc. for C32H30N6V1·0.35CH2Cl2: C, 67.07; H, 5.34; N, 14.51. Found: C,
67.08; H, 5.44; N, 14.49.
3.6.1. 2-(pentamethylphenyl)-1,1-dicyanoethylene
Yield: 63%. IR (KBr): ν(CN), 2235 cm−1 and 2203 cm−1
.
1H NMR
Funding
(400 MHz, CDCl3) δ 8.23 (s, 1H), 2.24 (s, 6H), 2.19 (s, 9H) ppm versus
This research was partially funded by the National Science
Foundation grant number CHM-0518220.
TMS.
3.6.2. 2-(pentamethylphenyl)-1,1,2-tricyanoethane
Yield: 78%. IR (KBr): ν(CN), 2242 cm−1 and 2264 cm−1
.
1H NMR
Acknowledgments
(400 MHz, CDCl3) δ 5.08 (d, 1H), 4.35 (d, 1H), 2.26 (s, 3H), 2.24 (s,
The Authors gratefully acknowledge Mr. Mark Harvey and Drs.
Guangbin Wang, Daniel Crawford and Diego Troya for helpful discus-
sions.
12H) ppm versus TMS.
3.6.3. 2-(pentamethylphenyl)-1,1,2-tricyanoethylene
Yield: 41%. Mp: compound charred at 180 °C. IR (KBr): ν(CN),
Declaration of Competing Interest
2239 cm−1 and 2249 cm−1
.
1H NMR (400 MHz, CDCl3) δ 2.27 (s, 3H),
2.23 (s, 12H) ppm versus TMS. HRMS-ESI (m/z, [M]). Calcd for
The authors declare no conflict of interest. The funders had no role
in the design of the study; in the collection, analyses, or interpretation
of data; in the writing of the manuscript, and in the decision to publish
the results.
C
16H15N3: 249.316. Found: 249.127.
All reactions involving the synthesis of the magnets were carried out
in a Vacuum Atmospheres Glovebox with N2 (ultra-high purity) as the
atmosphere.
References
3.7. V[2-MePTCE]2
[1] J.M. Manriquez, G.T. Yee, R.S. McLean, A.J. Epstein, J.S. Miller, A room tem-
perature molecular/organic-based magnet, Science 252 (1991) 1415–1417, https://
doi.org/10.1126/science.252.5011.1415.
Approximately 2.5 equivalents of the acceptor (2-MePTCE, 60.0 mg)
were dissolved in roughly 1.0 ml of dichloromethane in a 50 ml round-
bottom flask and left to stir (solution was yellow in color). 1 equivalent
of V(CO)6 (25.0 mg) was dissolved in 1.5 ml of dichloromethane in a
vial (solution was also yellow in color), and immediately added to the
acceptor in solution. The resulting solution turned almost immediately
black-green and eventually black. The reaction was allowed to stir for
30 min, after which the precipitate was collected on a medium frit by
vacuum filtration. The black solid was washed with dichloromethane
until the filtrate ran clear, dried under vacuum for an hour, and then
weighed and stored in a vial. Yield: 47.2 mg (89%). IR (KBr): ν(CN),
2215 cm−1 and 2119 cm−1. Anal. Calc. for C24H14N6V1·0.32CH2Cl2: C,
62.88; H, 3.18; N, 18.09. Found: C, 62.91; H, 3.63; N, 17.70.
[2] D. Haskel, Z. Islam, J. Lang, C. Kmety, G. Srajer, K.I. Pokhodnya, A.J. Epstein,
J.S. Miller, Local structural order in the disordered vanadium tetracyanoethylene
room-temperature molecule-based magnet, Phys. Rev. B 70 (2004) 054422, ,
https://doi.org/10.1103/PhysRevB.70.054422.
[3] J. Zhang, J.S. Miller, C. Vazquez, P. Zhou, W.B. Brinckerhoff, A.J. Epstein, Improved
synthesis of the V(tetracyanoethylene)x⋅y(solvent) room temperature magnet:
doubling of the magnetization at room temperature, ACS Symposium Series
American Chemical Society, Washington, DC, 1995, pp. 311–318.
[4] J. Zhang, J. Ensling, V. Ksenofontov, P. Gütlich, A.J. Epstein, J.S. Miller, [MII(tcne)
2]⋅ x CH2Cl2 (M=Mn, Fe Co, Ni) molecule-based magnets with Tc values above
100 K and coercive fields up to 6500 Oe, Angew. Chem. Int. Ed. 37 (1998) 657–660,
https://doi.org/10.1002/(SICI)1521-3773(19980316)37:5<657::AID-
ANIE657>3.0.CO;2-L.
[5] K.I. Pokhodnya, N. Petersen, J.S. Miller, Iron pentacarbonyl as a precursor for
molecule-based magnets: formation of Fe[TCNE]2 (Tc = 100 K) and Fe[TCNQ]2 (Tc
= 35 K), Inorg. Chem. 41 (2002) 1996–1997, https://doi.org/10.1021/ic011288b.
[6] K.I. Pokhodnya, E.B. Vickers, M. Bonner, A.J. Epstein, J.S. Miller, Solid solution
VxFe1-x[TCNE]2·zCH2Cl2 room-temperature magnets, Chem. Mater. 16 (2004)
3218–3223, https://doi.org/10.1021/cm049368b.
All other magnets were synthesized using a procedure similar to the
one used to synthesize V[2-MePTCE]2 and at similar scale.
3.8. V[3-MePTCE]2
[7] E.B. Vickers, A. Senesi, J.S. Miller, Ni[TCNE]2 · zCH2Cl2 (Tc = 13 K) and VxNi1 −
x[TCNE]y · zCH2Cl2 solid solution room temperature magnets, Inorg. Chim. Acta
357 (2004) 3889–3894, https://doi.org/10.1016/j.ica.2004.04.030.
[8] E.B. Vickers, T.D. Selby, J.S. Miller, Magnetically ordered (Tc = 200 K) bis(tetra-
cyanopyrazine)vanadium, V[TCNP]2·yCH2Cl2, J. Am. Chem. Soc. 126 (2004)
3716–3717, https://doi.org/10.1021/ja0397308.
Yield: 90.5%. IR (KBr): ν(CN), 2207 cm−1 and 2116 cm−1. Anal.
Calc. for C24H14N6V1·0.48CH2Cl2: C, 61.50; H, 3.15; N, 17.58. Found: C,
61.55; H, 3.51; N, 17.66.
[9] E.B. Vickers, I.D. Giles, J.S. Miller, M[TCNQ]y-based magnets (M = Mn, Fe Co, Ni;
TCNQ = 7,7,8,8-tetracyano-p-quinodimethane), Chem. Mater. 17 (2005)
1667–1672, https://doi.org/10.1021/cm047869r.
3.9. V[4-MePTCE]2
[10] M.L. Taliaferro, M.S. Thorum, J.S. Miller, Room-temperature organic-based magnet
(Tc≈50 °C) containing tetracyanobenzene and hexacarbonylvanadate(−I), Angew.
Chem. 118 (2006) 5452–5457, https://doi.org/10.1002/ange.200600988.
[11] J. Hao, R.A. Davidson, M. Kavand, K.J. van Schooten, C. Boehme, J.S. Miller,
Hexacyanobutadienide-based frustrated and weak ferrimagnets: M(HCBD)2·zCH2Cl2
(M = V, Fe), Inorg. Chem. 55 (2016) 9393–9399, https://doi.org/10.1021/acs.
inorgchem.6b01565.
Yield: 80.7%. IR (KBr): ν(CN), 2210 cm−1 and 2115 cm−1. Anal.
Calc. for C24H14N6V1·0.53CH2Cl2: C, 61.08; H, 3.15; N, 17.42. Found: C,
60.97; H, 2.95; N, 18.37.
3.10. V[2,6-Me2PTCE]2
[12] J.P. Fitzgerald, B.B. Kaul, G.T. Yee, Vanadium [dicyanoperfluorostilbene]2•yTHF: a
molecule-based magnet with Tc ≈ 205 K, Chem. Commun. 1 (2000) 49–50, https://
doi.org/10.1039/A907535F.
Yield: 85.8%. IR (KBr): ν(CN), 2202 cm−1 and 2116 cm−1. Anal.
Calc. for C26H18N6V1·0.26CH2Cl2: C, 64.70; H, 3.83; N, 17.24. Found: C,
64.68; H, 3.67; N, 17.71.
[13] M.D. Harvey, T.D. Crawford, G.T. Yee, Room temperature and near-room tem-
perature molecule-based magnets, Inorg. Chem. 47 (2008) 5649–5655, https://doi.
org/10.1021/ic702359g.
[14] M.D. Harvey, J.T. Pace, G.T. Yee, A room temperature ferrimagnet, vanadium
[pentafluorophenyltricyanoethylene]2, Polyhedron 26 (2007) 2037–2041, https://
doi.org/10.1016/j.poly.2006.09.097.
3.11. V[2,4,6-Me3PTCE]2
[15] M.K. Amshumali, M.D. Harvey, G.T. Yee, Room temperature and near-room tem-
perature coordination polymer magnets, Synth. Met. 188 (2014) 53–56, https://
doi.org/10.1016/j.synthmet.2013.11.023.
Yield: 79.7%. IR (KBr): ν(CN), 2209 cm−1 and 2214 cm−1. Anal.
Calc. for C28H22N6V1·0.57CH2Cl2: C, 63.33; H, 4.30; N, 15.51. Found: C,
63.32; H, 4.02; N, 16.19.
[16] E. Carlegrim, A. Kanciurzewska, M.P. de Jong, C. Tengstedt, M. Fahlman, The un-
occupied electronic structure of the semi-conducting room temperature molecular
6

J.A. King, et al.
JournalofMagnetismandMagneticMaterialsxxx(xxxx)xxxx
magnet V(TCNE)2, Chem. Phys. Lett. 452 (1–3) (2008) 173–177, https://doi.org/
https://doi.org/10.1021/acs.jchemed.6b00680.
10.1016/j.cplett.2007.12.049.
[24] S.P. Sellers, B.J. Korte, J.P. Fitzgerald, W.M. Reiff, G.T. Yee, Canted ferromagnetism
and other magnetic phenomena in square planar, neutral, manganese(II) and iron
(II) octaethyltetraazaporphyrins, J. Am. Chem. Soc. 120 (1998) 4662–4670,
https://doi.org/10.1021/ja973787a.
[17] S. Ferlay, T. Mallah, R. Ouahès, P. Veillet, M. Verdaguer, A room-temperature or-
ganometallic magnet based on Prussian blue, Nature 378 (1995) 701–703, https://
doi.org/10.1038/378701a.
[18] G. Yee, D. Tatum, J. Zadrozny, A new family of high Tc molecule-based magnetic
networks: V[x-ClnPTCE]2●yCH2Cl2 (PTCE = Phenyltricyanoethylene),
Magnetochemistry 5 (2019) 44–56.
[25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheesman,
J.A. Montgomery, T. Vreven, K.N.S. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson,
H. Naktsuji, M. Hada, M. Ehara, K. Totota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann,
O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala,
K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski,
S. Dapprich, A.D. Daniel, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck,
K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford,
J. Cioslowski, B.B. Stefanov, G. Lio, A. Liashenko, P. Piskorz, I. Komaromi,
R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara,
M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez,
J.A. Pople, Gaussian 03, Revision C.02, Gaussian, Inc., Wallingford CT, 2004.
[26] X. Liu, J.E. Ellis, Inorg. Synth. 34 (2004) 96–132.
[19] B.B. Corson, R.W. Stoughton, Reactions of alpha, beta-unsaturated dinitriles, J. Am.
Chem. Soc. 50 (10) (1928) 2825–2837, https://doi.org/10.1021/ja01397a037.
[20] P.J. Pearce, R.J.J. Simkins, Acid strengths of some substituted picric acids, Can. J.
Chem. 46 (1968) 241–248.
[21] J.C. Rienstra-Kiracofe, G.S. Tschumper, H.F. Schaefer, S. Nandi, G.B. Ellison,
Atomic and molecular electron affinities: photoelectron experiments and theoretical
computation, Chem. Rev. 102 (2002) 231–282, https://doi.org/10.1021/
cr990044u.
[22] R.L. Carlin, Magnetochemistry, Spring-Verlag, Berlin, Germany, 1986 ch. 6. ISBN 0-
387-15816-2.
[23] S.W. Goldstein, A. Bill, J. Dhuguru, O. Ghoneim, Nucleophilic aromatic substitution
– addition and identification of an Amine, J. Chem. Educ. 94 (2017) 1388–1390,
7