Substituent Effects on Exchange Coupling and Magnetic Relaxation in 2,2′-Bipyrimidine Radical-Bridged Dilanthanide Complexes

in 2,2 ′-Bipyrimidine Radical-Bridged Dilanthanide Complexes

Article

Substituent Eﬀects on Exchange Coupling and Magnetic Relaxation

Colin A. Gould, Edward Mu, Veacheslav Vieru, Lucy E. Darago, Khetpakorn Chakarawet,

Miguel I. Gonzalez, Selvan Demir, and Je ﬀ rey R. Long*

Cite This: https://dx.doi.org/10.1021/jacs.0c10612

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Systematic analysis of related compounds is crucial

to the design of single-molecule magnets with improved properties,

yet such studies on multinuclear lanthanide complexes with strong

magnetic coupling remain rare. Herein, we present the synthesis

and magnetic characterization of the series of radical-bridged

dilanthanide complex salts [(Cp* Ln) (μ-5,5′-R bpym)](BPh )

(

Ln = Gd, Dy; R = NMe (1), OEt (2), Me (3), F (4); bpym =

,2′-bipyrimidine). Modiﬁcation of the substituent on the bridging

,5′-R bpym radical anion allows the magnetic exchange coupling

constant, J

, for the gadolinium compounds in this series to be

Gd−rad

−

−1

tuned over a range from −2.7 cm (1) to −11.1 cm (4), with

electron-withdrawing or -donating substituents increasing or decreasing the strength of exchange coupling, respectively. Modulation

of the exchange coupling interaction has a signiﬁcant impact on the magnetic relaxation dynamics of the single-molecule magnets 1-

Dy through 4-Dy, where stronger J_G_d₋_r_a_dfor the corresponding Gd compounds is associated with larger thermal barriers to

magnetic relaxation (U ), open magnetic hysteresis at higher temperatures, and slower magnetic relaxation rates for through-barrier

eff

processes. Further, we derive an empirical linear correlation between the experimental U_e_f_fvalues for 1-Dy through 4-Dy and the

magnitude of J_G_d₋_r_a_dfor the corresponding gadolinium derivatives that provides insight into the electronic structure of these

complexes. This simple model applies to other organic radical-bridged dysprosium complexes in the literature, and it establishes clear

design criteria for increasing magnetic operating temperatures in radical-bridged molecules.

,10

INTRODUCTION

magnets have been made with lanthanide-based systems.

■

These breakthroughs have been enabled in part by the

development of clear design criteria for achieving large values

Single-molecule magnets exhibit slow magnetic relaxation that

is of molecular origin, arising from the presence of a bistable

magnetic ground state with a thermal barrier to inversion of

1−15

of U_e_f_fin mononuclear lanthanide complexes.

For

example, increasing the axiality of the ligand ﬁeld for oblate

the magnetic moment (U). When k T ≪ U, many such

Dy and Tb ions leads to larger thermal barriers.

molecular compounds have been shown to exhibit magnetic

hysteresis, a property that could potentially be utilized in

applications such as nanoscale information storage or spin-

Systematic studies conducted on related complexes have

been crucial in identifying these design principles and in

16−18

−6

verifying their predictions.

Importantly, these criteria link

based computing.

However, the vast majority of single-

magnetic properties to molecular structure, which can be easily

elucidated through experimental methods such as single-crystal

X-ray diﬀraction, thereby providing a facile method to identify

promising single-molecule magnet candidates. For instance,

the use of a sterically encumbered pentaisopropylcyclopenta-

molecule magnets retain their magnetic memory only at

prohibitively low temperatures. Signiﬁcant research has

therefore focused on increasing the maximum operating

temperature, which is typically deﬁned as the temperature at

which the magnetic relaxation time, τ, is equal to 100 s (T or

blocking temperature) or the highest temperature at which

iPr5

dienyl ligand (Cp ) and a pentamethylcyclopentadienyl

iPr5 +

ligand (Cp*) in [DyCp*Cp ] gives rise to shorter Ln−Cp

magnetic hysteresis is observed. Higher operating temper-

atures can be achieved by maximizing the eﬀective thermal

Received: October 6, 2020

barrier to magnetic relaxation (U ), which typically sets an

eff

upper limit to the value of T , and by mitigating through-

barrier relaxation pathways, such as quantum tunnelling of the

magnetization.

Over the past decade, the most signiﬁcant advances in

enhancing the operating temperatures of single-molecule

XXXX American Chemical Society

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

Scheme 1. Synthesis of the Radical-Bridged Dilanthanide Compounds [(Cp* Ln) (μ-5,5′-R bpym)](BPh ) (Ln = Gd, Dy; R =

NMe (1), OEt (2), Me (3), F (4))

Figure 1. Crystal structures of the complex cations in [(Cp* Dy) (μ-5,5′-R bpym)](BPh ) (R = NMe (1), OEt (2), Me (3), F (4)). Dark green,

gray, blue, red, and lime green spheres represent Dy, C, N, O, and F atoms, respectively; H atoms are omitted for clarity.

bond distances and a larger Cp−Ln−Cp angle than analogous

|J_G_d₋_r_a_d| of the corresponding gadolinium derivatives. This

19−21

[

DyCp ] complexes,

resulting in greater axial aniso-

empirical relation can be extended to other organic radical-

bridged dysprosium compounds in the literature and brings

into focus key design principles to access molecules in this

tropy and a 100-s magnetic blocking temperature of 65 K, the

highest value yet reported for any molecule.

Multinuclear lanthanide complexes featuring radical bridging

ligands also display some of the highest operating temperatures

class with larger U values and higher operating temperatures.

eff

yet reported for single-molecule magnets, as high as T = 20

RESULTS AND DISCUSSION

■

K. A small but growing number of these compounds has

emerged following the discovery of very strong magnetic

Synthesis and Structural Characterization. The mole-

cule 2,2′-bipyrimidine (bpym) serves as an ubiquitous bridging

ligand in coordination chemistry, however, symmetric 5,5′-

−

exchange in [({N(SiMe ) } (THF)Ln) (μ-N )] (Ln = Gd,

Tb, Dy).

4,25

For these molecules with strong coupling, both

R bpym variants have only yet been reported for alkyl, aryl,

35−38

single-ion anisotropy and magnetic exchange interactions

between the lanthanide and bridging radical dictate the

ether, and bromine substituents.

Bipyrimidine derivatives

are typically synthesized via Cu- or Ni-mediated coupling

magnitude of the thermal barrier to magnetic relaxation.

reactions, and therefore the synthesis of 5,5′-R bpym was ﬁrst

These properties are diﬃcult to experimentally quantify for

attempted via a Ni-catalyzed homocoupling of the correspond-

39−41

anisotropic lanthanide ions, and they can require intensive

ing 5-R-2-chloropyrimidine.

This approach furnished

27,28

calculations to theoretically predict.

As a result, identifying

,5′-R bpym with electron-donating substituents R = Me,

^w^h^e^t^h^e^r^aⁿⁱ^s^o^t^r^o^p^y^o^r^e^x^c^h^aⁿ^g^e^lⁱ^mⁱ^t^s^t^h^e^v^a^l^u^e^o^f^Ueff can be

quite challenging, thus forestalling the design of molecules with

improved magnetic properties. Toward this end, the systematic

analysis of related radical-bridged complexes with well-deﬁned

structural or electronic diﬀerences stands as an important goal

in the ﬁeld of molecular magnetism. Nevertheless, such studies

on radical-bridged complexes remain rare, in contrast to the

numerous studies on series of multinuclear lanthanide or

transition metal complexes with diamagnetic bridging

OEt, and NMe in isolated yields between 33 and 37% but did

not yield electron-deﬁcient derivatives with R = F or CF₃.

These results are consistent with reports of the synthesis of

,5′-R bpy (bpy = 2,2′-bipyridine), which found lower yields

for the Ni-catalyzed homocoupling of 5-R-2-chloropyridine

2−44

substrates bearing electron-withdrawing substituents.

Attempts to synthesize derivatives with R = F or CF via

Cu-mediated homocoupling reactions were also unsuccessful.

45,46

Instead, a Stille reaction

was used to couple 5-ﬂuoro-2-

9−34

ligands.

tributylstannylpyrimidine and 5-ﬂuoro-2-chloropyrimidine,

Herein, we report the synthesis and detailed magnetic

characterization of the series of organic radical-bridged

dilanthanide complex salts [(Cp* Ln) (μ-5,5′-R bpym)]-

aﬀording 5,5′-F bpym in 57% isolated yield. Although it was

not possible in our hands to isolate 5,5′-(CF ) bpym by an

analogous route, the Pd-catalyzed coupling of 5-triﬂuorometh-

yl-2-chloropyrimidine with 5-ﬂuoro-2-tributylstannylpyrimi-

(

BPh ) (Ln = Gd, Dy; R = NMe (1), OEt (2), Me (3), F

(

4)). Modiﬁcation of the substituent on 5,5′-R bpym radical

dine furnished 5,5′-F(CF )bpym in 46% isolated yield.

anion induces dramatic changes in the magnitude of the

intramolecular magnetic exchange, with electron-donating and

Analogous to the synthesis of the radical-bridged lanthanide

complex salts [(Cp* Ln) (μ-bpym)](BPh ), reaction of

withdrawing substituents decreasing and increasing the

[Cp* Ln](BPh ) with 5,5′-R bpym, followed by reduction

magnitude of the exchange coupling constant (J_L_n₋_r_a_d),

respectively, as quantiﬁed for the gadolinium congeners.

Signiﬁcantly, a linear correlation is demonstrated between

the experimental U_e_f_fvalues for 1-Dy through 4-Dy and

with potassium graphite, gave the compounds [(Cp* Ln) (μ-

2 2

5,5′-R bpym)](BPh ) (R = NMe (1), OEt (2), Me (3), and F

(4); Ln = Gd, Dy) in 27−69% isolated yield (Scheme 1). A

mixture of THF and toluene was used as the reaction solvent

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Figure 2. Dc magnetic susceptibility data for 1−4 under an applied magnetic ﬁeld of 10 kOe. The black lines represent ﬁts to the data for 1-Gd

through 4-Gd, as described in the main text, corresponding to a 4-fold increase in J across the series.

Gd−rad

in the synthesis of 1−3, but these conditions did not aﬀord the

desired compounds with 5,5′-F bpym or 5,5′-F(CF )bpym.

1/2 organic radical. This discrepancy is due to contributions

from temperature-independent paramagnetism, which is sup-

pressed under a 10 kOe ﬁeld (Figure 2), resulting in χ_MT

values of 16.1 to 16.2 emu K/mol for 1-Gd through 4-Gd at

Given the electron-deﬁcient nature of the ﬂuoro- and

triﬂuoromethyl-substituted ligands, it is likely that these

ligands coordinate more weakly to lanthanide ions, and thus

the presence of coordinating solvent may hinder complex

formation. Exclusion of THF from the reaction and

crystallization conditions enabled isolation of 4, although

300 K. As the temperature is lowered, χ T passes through a

shallow minimum between 75 and 140 K before rising to a

maximum value ranging from 18.6 to 23.3 emu K/mol (under

a 1 kOe ﬁeld), indicative of antiferromagnetic exchange

•−

attempts to isolate compounds with 5,5′-F(CF )bpym, the

coupling between the Gd centers and bpym . These

most electron-deﬁcient ligand in the series, were not successful.

The solid-state structures of compounds 1−4 were

determined from single-crystal X-ray diﬀraction data (Figure

maximum χ T values approach the theoretical value of 24.4

emu K/mol for an S = / system resulting from intra-

molecular antiferromagnetic Gd −ligand radical exchange

) with the exception of 1-Gd, for which suﬃcient quality

coupling.

single crystals could not be obtained. The C(2)−C(2′) bond

The χ T data for 1-Dy through 4-Dy exhibit qualitatively

similar behavior to those of the Gd compounds (see section

S5.1 of the Supporting Information and Figure 2). Under a 10

distance in a bpym ligand is indicative of its charge state, and

the corresponding distances in 1−4 range from 1.403(14) to

•

−

.432(18) Å, consistent with previously reported bpym

kOe ﬁeld, the χ T values at 300 K range from 26.3 to 28.6

7,49,50

complexes.

The average Dy−N and Gd−N bond

emu K/mol, slightly lower than the theoretical value of 28.7

distances range from 2.424(6) to 2.440(6) Å and 2.458(2)

to 2.469(5) Å, respectively, consistent with the average Ln−N

bond distances in [(Cp* Dy) (μ-bpym)](BPh ) (data for

emu K/mol for two noninteracting Dy centers and an S = 1/

2 organic radical. As the temperature is lowered, the χ_MT

curves rise to maxima between 32.3 and 37.1 emu K/mol,

indicative of the formation of a high-angular momentum

ground state arising from magnetic exchange. The peak in χ_MT

is followed by a precipitous drop at even lower temperatures,

indicative of magnetic blocking.

[

(Cp* Gd) (μ-bpym)](BPh ) were consistently low resolu-

2 2 4

tion and were not published as a result). Complexes 1−4

show average Dy−C and Gd−C distances of 2.627(19) and

.649(12) Å, respectively, and average Cp−Dy−Cp and Cp−

Gd−Cp angles of 140.1(3)° and 139.2(4)°, respectively,

The magnitude of the exchange coupling interaction was

consistent with [(Cp* Dy) (μ-bpym)](BPh ). The large

quantiﬁed for compounds of isotropic Gd (4f ) by ﬁtting the

uncertainties in these values arise from positional disorder of

the pentamethylcyclopentadienyl ligands. The Dy−C bond

distances and Cp−Dy−Cp angles in 1−4 are similar to those

reported previously for single-molecule magnets that contain

dc susceptibility data to the following spin-only Hamiltonian:

+ S_G_d₍₂₎), where J_G_d₋_r_a_dis the

Gd−rad rad Gd(1)

H = −2J

S (S

exchange constant corresponding to intramolecular Gd −

bpym coupling, S is the spin operator for bpym , and

•

−

•−

rad

Cp* Dy units, which typically display high single-ion

is the spin operator for each Gd ion (see Figure 2 and

Gd(n)

anisotropy due to the strong axial ligand ﬁeld exerted by the

S40, S43, S45, and S47). Terms that account for temperature-

independent paramagnetism and intermolecular exchange

interactions were also included in the ﬁts of all data (see

Tables S8−S11). Corresponding ﬁts of the full range of data

cyclopentadienyl ligands. The axis deﬁned by the two

lanthanide centers in 1−4 is nearly parallel with the plane of

the 5,5′-R bpym ligand (deﬁned by the 12 atoms of the

bipyrimidine fragment), deviating by only 0.6(1)° to 3.5(1)°.

Overall, the bond distances and angles in 1−4 are similar

across the series and follow no recognizable trend, suggesting

that any diﬀerences in the magnetic properties of these

compounds can be attributed to the electronic inﬂuence of the

for 3-Gd and 4-Gd under 10 kOe yield J

= −9.54(7) and

Gd−rad

−

−11.1(2) cm , respectively (Figure 2, black lines). The value

of J for 4-Gd is among the largest yet measured for a

Gd−rad

radical-bridged lanthanide complex and is slightly larger than

−

J_G_d₋_r_a_d= −10 cm determined for [(Cp* Gd) (μ-bpym)]-

substituent on the 5,5′-R bpym ligand.

(BPh₄). For 1-Gd and 2-Gd, a modiﬁed Hamiltonian

incorporating an additional term for intramolecular Gd −

Gd exchange coupling (J

Magnetic Exchange Coupling. Dc magnetic suscepti-

bility data were initially collected for 1−4 from 2 to 300 K

under an applied magnetic ﬁeld of 1 kOe (see section S5.1 of

) was necessary to ﬁt the rise in

Gd−Gd

χ T at low temperatures satisfactorily, yielding J

Gd−rad

−

the Supporting Information). At this ﬁeld, the values of χ T at

−2.66(12) and −4.16(25) cm and J_G_d₋_G_d= 0.125(8) and

00 K for the Gd³congeners range from 16.3 to 17.3 emu K/

0.145(5) cm , respectively, at 10 kOe. Weak ferromagnetic

−1

mol, slightly greater than the theoretical value of 16.1 emu K/

Ln ···Ln exchange coupling has been characterized

mol predicted for two noninteracting Gd³centers and an S =

previously in dilanthanide complexes bridged by 2,2′-

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

2,53

bipyrimidine.

Fits to the data for 3-Gd and 4-Gd using this

that the substituent eﬀect in 1−4 is achieved by installing

Hamiltonian yielded J_G_d₋_G_d< 1 × 10⁻cm and comparable

values of J_G_d₋_r_a_dto those obtained using the aforementioned

−1

substituents on ligand atoms that possess radical spin density

−

in the bpym SOMO. In contrast, a recent study on

Hamiltonian which considers only intramolecular Gd −

tetraoxolene radical-bridged diiron complexes showed that

substituent modiﬁcations at positions lacking radical spin

density in the ligand SOMO had no eﬀect on the magnitude of

bpym^•coupling.

−

The relative strength of the magnetic exchange coupling

interaction in 1-Gd through 4-Gd and [(Cp* Gd) (μ-

metal−radical exchange. These insights are key toward the

development of new and better-performing radical-bridged

lanthanide single-molecule magnets. Further, these insights

could enable the synthesis of improved single-chain magnets,

which show relaxation barriers that increase with J, and bulk

magnetic materials, which possess magnetic ordering temper-

atures that are directly proportional to J.

bpym)](BPh ) follows the trend R = NMe < OEt < Me <

H < F, which can be rationalized by considering the eﬀect of

•−

the substituent on the 5,5′-R bpym radical spin density. The

•

−

singly occupied molecular orbital (SOMO) for bpym has

coeﬃcients on C(2), C(2′), C(5), C(5′), and the nitrogen

4,55

atoms,

and the magnitude of J

is inﬂuenced primarily

Gd−rad

by the spin density on nitrogen. Substituents at the 5 and 5′

carbons can inﬂuence radical spin density through resonance

or inductive eﬀects at all positions that possess coeﬃcients in

Thermally Activated Magnetic Relaxation. The low-

temperature drop in the static magnetic susceptibility data for

1-Dy through 4-Dy are indicative of magnetic blocking, and

accordingly ac magnetic susceptibility data were collected to

investigate the presence of slow magnetic relaxation in each

compound. The data for 1-Dy show two peaks in the out-of-

phase magnetic susceptibility (χ″) in the frequency range of

0.1−1500 Hz between 4.5 and 11 K, indicative of two

independent magnetic relaxation processes occurring at these

temperatures (Figure S50 ). Similarly, for 2-Dy two relaxation

processes are present, although one process is clearly dominant

within the examined temperature range of 5−12.5 K (Figure

S54 ). In contrast, 3-Dy and 4-Dy each exhibit a single χ″ peak

in the same frequency range and for temperatures ranging from

9 to 17 K (Figures S56 and S58 ). Magnetic relaxation times, τ,

were extracted from simultaneous ﬁts of the ac susceptibility

data for 1-Dy through 4-Dy using a generalized Debye model

(see section S5.2 of the Supporting Information). Arrhenius

plots of inverse temperature versus the natural log of τ for 3-

Dy and 4-Dy, and for the dominant relaxation processes in 1-

Dy and 2-Dy, are linear, characteristic of magnetic relaxation

via a thermally activated Orbach mechanism (see section S5.4

•

−

the 5,5′-R bpym SOMO. For instance, introducing sub-

stituents that are electron-withdrawing by induction, such as

ﬂuorine, at the 5 and 5′ carbons decreases the radical

coeﬃcient at that position, thereby increasing the radical

spin density at C(2), C(2′), and the nitrogen atoms and thus

the strength of the magnetic exchange interaction. In contrast,

introducing electron-donating substituents at the 5 and 5′

carbons increases the radical coeﬃcient at that position,

thereby decreasing spin density at C(2), C(2′), and the

nitrogen atoms and the magnitude of J . Accordingly,

Gd−rad

introducing substituents that are weakly electron-donating by

induction, such as methyl, results in a slight decrease in the

magnitude of J_G_d₋_r_a_d, while introducing substituents that are

strongly electron-donating by resonance, such as dimethyla-

mino or ethoxy, results in a substantial decrease in the

magnitude of the coupling constant. Such dramatic substituent

eﬀects on radical spin density are supported by previous

electron paramagnetic resonance (EPR) characterization of

56,57

other radical anion species,

however, additional factors

may inﬂuence the magnitude of J_G_d₋_r_a_din 1-Gd through 4-Gd,

including substituent eﬀects on the energy of the 5,5′-

of the Supporting Information). Accordingly, all data were ﬁt

−

−1

to the equation τ = τ₀exp(U /k T), yielding thermal

eff

•

−

R bpym SOMO. EPR characterization of diamagnetic Y

barriers to magnetic relaxation (U ) of 31, 40, 82, and 93

eff

−1

or Lu derivatives of 1−4 and complementary computational

analysis could facilitate a more quantitative model of the

substituent eﬀect on magnetic coupling in 1-Gd through 4-Gd,

and such studies represent a clear step toward generalizing the

design principles identiﬁed here. While the magnitude of the

exchange coupling in 1-Dy through 4-Dy is challenging to

quantify using the experimental magnetic susceptibility data

due to the anisotropic nature of the Dy³ion, the relative

cm for 1-Dy through 4-Dy, respectively. For comparison, U

= 88 cm for [(Cp* Dy) (μ-bpym)](BPh ). Thus, the

magnitude of U_e_f_fincreases with the substituent on the radical

eff

−

ligand in the order NMe < OEt < Me < H < F, following the

same trend determined for |J_G_d_‑_r_a_d|. Plots of ln(τ) versus 1/T

for the minor relaxation process in 1-Dy and 2-Dy are also

linear with temperature, and the data could be ﬁt to an Orbach

−1

mechanism yielding U_e_f_f_,₂= 46 and 94 cm , respectively.

magnitude of the peak in the χ T versus T data and the

The value of U_e_f_ffor a radical-bridged lanthanide complex

depends on both single-ion anisotropy and exchange coupling

interactions. For dinuclear systems featuring lanthanides with

large single-ion anisotropy, the exchange interaction can be

temperature at which this maximum occurs follow a similar

trend as that characterized for the Gd derivatives, suggesting a

similar ordering of the exchange coupling strength.

Remarkably, these results demonstrate that the magnitude of

the exchange coupling constant in organic-radical bridged

dilanthanide systems can be tuned by more than a factor of 4,

described by the Ising model, and the splitting of electronic

states can be approximated with the following Hamiltonian: H

= −2J S (J

Ln−rad rad Ln(1)

+ J

) + ΣB O (i), where J

is the

Ln(2)

Ln−rad

−

from −2.66(12) to −11.1(2) cm , simply through ligand

modiﬁcation. While prior studies on radical-bridged complexes

have shown that large changes in magnetic coupling can be

induced by changing the donor atoms of the bridging ligand or

through installing ligand substituents that alter the molecular

structure of the complex, the substituent eﬀect in 1−4 arises

from the electronic inﬂuence of a substituent at a remote,

exchange constant corresponding to intramolecular Ln −

radical exchange, S is the spin operator for the organic

is the angular momentum

operator for each Ln ion, and B and O are parameters

rad

radical bridging ligand, J

Ln(n)

that describe crystal ﬁeld interactions for each lantha-

23,65−67

nide.

Excited state energies correspond to multiples of

J_L_n₋_r_a_d, and the energy diﬀerence between the ground and ﬁrst

58−61

nonmetal binding site.

Thus, magnetic coupling can be

excited state, typically deﬁned as U , is 15|J_D_y₋_r_a_d| for

eff

modulated substantially without signiﬁcant perturbation of the

metal coordination environment. It is important to emphasize

anisotropic radical-bridged dysprosium complexes (Figure 3,

left). Indeed, this relationship was recently demonstrated

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Figure 3. Qualitative energy diagrams for dinuclear radical-bridged

dysprosium complexes. The complex on the left displays high single-

ion anisotropy and can be described by an Ising model. The complex

on the right shows lower single-ion anisotropy and thus the exchange

interaction mixes the ground Kramers doublet with the excited

Figure 4. Plot of U_e_f_ffor 1-Dy through 4-Dy versus |J_G_d₋_r_a_d| of the

corresponding gadolinium complexes (dark red circles). A linear ﬁt to

these data provides an empirical model (black line) with which to

analyze the electronic structure of multinuclear radical-bridged

compounds (pink circles) and mono- or dinuclear dysprosium

complexes featuring a radical ligand which is coupled to one metal

center (blue circles). See Table S20 in the Supporting Information for

a full list of compounds and references.

^d^o^u^b^l^e^t^s^,^l^e^a^dⁱⁿ^g^t^o^a^l^o^w^e^r^v^a^l^u^e^o^f^Ueff^.

experimentally for the N₂³⁻radical-bridged dysprosium

Me4H

Me4H −

complex [K(crypt-222)][(Cp

Dy) (μ-N )] ([Cp

] =

tetramethylcyclopentadienyl anion). By contrast, a complex

with low single-ion anisotropy will possess low-lying crystal

ﬁeld states that are further mixed by the exchange interaction,

resulting in a comparatively lower value of U_e_f_f(Figure 3,

right). Substantial mixing of the crystal ﬁeld states can also

occur for complexes with large single-ion anisotropy and very

multinuclear radical-bridged dysprosium complexes in the

literature. For instance, values reported for [(Cp* Ln) (μ-

bpym)](BPh ) (Ln = Gd, Dy), (Cp* Ln) (μ-HAN) (HAN =

hexaazatrinaphthylene), and [CoCp ][((HBpz ) Ln) (μ-CA)]

3 2

large values of J_L_n₋_r_a_d

While the Ising model can oﬀer insight

(HBpz = hydrotris(pyrazol-1-yl)borate, CA = chloranilate) all

into the magnitude of U_e_f_fvalues in radical-bridged lanthanide

complexes, its utility is limited due to the diﬃculty of

experimentally determining J_L_n₋_r_a_dfor anisotropic lanthanide

ions and of quantifying single-ion anisotropy. Indeed, these

values are typically obtained through ab initio calcula-

fall near the best-ﬁt line describing the relation between U_e_f_f

and |J_G_d₋_r_a_d| for 1 through 4, which implies that these

47,72,73

complexes can also be described by an Ising model.

/−

By contrast, data for [(Cp* Ln) (μ-tppz)] (tppz = 2,3,5,6-

−

tetra(2-pyridyl)pyrazine) and [(Cp* Ln) (μ-ind)] (ind =

8−70

tions,

which can be time-intensive to execute and may

indigo) fall well below the line, which suggests that these

possess a wide margin of error.

complexes possess a value of U below the Ising limit (i.e., U

eff

74,75

We accordingly sought to develop an empirical model to

rationalize the values of U_e_f_ffor 1-Dy through 4-Dy that does

not require determination of J_D_y₋_r_a_d. In particular, the

< |15J_D_y₋_r_a_d|) due to low single-ion anisotropy.

This

analysis is consistent with the stronger equatorial interactions

observed for the Dy centers in these complexes. Indeed, ab

−

magnitude of magnetic exchange in radical-bridged Gd

initio calculations performed on [(Cp* Dy) (μ-ind)] predict

systems has been used as a proxy to estimate the relative

magnitude of the exchange coupling in isostructural systems of

that the ground Kramers doublet for each Dy ion is relatively

−

close in energy to the ﬁrst excited state (ΔE = 71 cm ),

further con ﬁ rming the low single-ion anisotropy in this

the anisotropic lanthanides. Given the similar trend in the

χ_MT data for the Dy and Gd congeners of 1 through 4, we

hypothesized that the J_G_d₋_r_a_dvalues obtained for 1-Gd through

−

complex (Table S23 ). For comparison, Δ E = ∼200 cm for

1-Dy through 4-Dy (see below and Table S23 ). Finally, the

empirical model presented here for 1 through 4 can also be

used to analyze complexes in which a single dysprosium ion

interacts with an organic radical (see section S5.5 of the

Supporting Information for additional details).

4-Gd may directly correlate with the exchange coupling in 1-

Dy through 4-Dy. Figure 4 features a plot of U_e_f_ffor the

dominant relaxation process in 1-Dy through 4-Dy versus

J _G _d ₋_r_a_d| for the corresponding Gd compounds (see also Table

S20 ). The resulting plot is linear, which implies that the Dy

The correspondence between the best-ﬁt line for 1 through

4 and available literature data for other dysprosium and

gadolinium pairs suggests that this empirical model may be a

useful predictive tool for identifying radical-bridged dyspro-

sium complexes exhibiting enhanced barriers to magnetic

relaxation. Indeed, the maximal value of the energetic splitting

between the ground and ﬁrst excited state, which typically

molecules can be described by an Ising modelthat is, U is

eff

directly proportional to |J

| for this series. This observation

Gd−rad

is consistent with the large single-ion anisotropy reported for

other dysprosium complexes with axial Cp* ligands. A linear

dependence of U on |J_D_y₋_r_a_d| was also previously reported for

eff

[

Cr Dy (OMe) (RN{(CH ) OH} ) (acac) (NO ) ] (R =

2 2 2 2 2 2 2 4 3 2

−

Me, Et, nBu) (acac = acetylacetonate) using computationally

represents U , can be estimated using only J

for the

eff

Gd−rad

determined values for the exchange coupling constants.

The empirical model proposed here for 1 through 4 can be

used more broadly to analyze the electronic structure of

corresponding Gd complex. This model can also aﬀord insight

into strategies for increasing the magnitude of the thermal

relaxation barrier. For instance, U_e_f_f= 35 cm⁻for

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Figure 5. Low-lying exchange spectra calculated for 1-Dy through 4-Dy, showing pathways for thermally activated magnetic relaxation (dotted blue

lines), quantum tunnelling of the magnetization (dotted red lines), and the experimental U_e_f_for U_e_f_f_,₂values (dashed gray lines). The numbers

associated with each path are the largest matrix element connecting each exchange doublet; the square of this value is proportional to the transition

rate between states.

−

[

(Cp* Dy) (μ-ind)] , but the trend presented here indicates

assumption that these complexes display large single-ion

anisotropy and can be described by an Ising model.

−

that U_e_f_fvalues as large as 93 cm could be achieved if the

ancillary cyclopentadienyl ligands were modiﬁed to maximize

the single-ion anisotropy of the Dy³ions. By contrast, larger

energetic splitting between the ground and ﬁrst excited state

will not be accessible simply by modifying the ancillary ligands

of the aforementioned HAN-bridged complex, and instead

modiﬁcations to the radical bridge are needed to increase the

magnitude of J_D_y₋_r_a_d. Notably, this empirical model is readily

implemented and relies on parameters that can be easily

extracted from experimental data, and therefore it could be

used to benchmark future computational studies or provide

rapid insight into the electronic structure of new radical-

bridged complexes.

The calculated exchange spectra for 1-Dy, 2-Dy, 3-Dy, and

4-Dy (Figure 5 and Table S25) yield Ueff values of 46, 59, 73,

−

and 83 cm , respectively, comparable to the experimental Ueff

−

values of 31, 40, 82, and 93 cm . Interestingly, the separation

between the ground state and second excited state in 1-Dy and

−

-Dy are 93 and 116 cm , respectively, which correspond to

−

^t^h^e^Ueff,2 values of 46 and 94 cm determined for the minor

relaxation process characterized for each complex. Orbach

relaxation via the second excited state has been characterized

previously in radical-bridged lanthanide complexes. This

mechanism is likely not observed for 3-Dy and 4-Dy due to the

higher energy of the second excited state in these complexes,

which renders an Orbach process thermally inaccessible in the

experimental temperature range. Altogether, these calculations

conﬁrm that 1-Dy through 4-Dy can be described by the Ising

model and that the magnitude of U_e_f_ffor these complexes is

dictated by the strength of the exchange coupling.

Two key assumptions are implicit to the foregoing analysis,

namely that J_G_d₋_r_a_dis proportional to J_D_y₋_r_a_dand that the

proportionality constant relating these two values is the same

for all represented complexes. These assumptions appear to be

_r_e_a_s_o_n_a_b_l_e_f_o_r_c_o_m_p_l_e_x_e_s_i_n_w_h_i_c_h_D_y3+ ions interact with

Through-Barrier Magnetic Relaxation. Dc relaxation

measurements were performed on 1-Dy through 4-Dy to

probe magnetic relaxation dynamics at lower temperatures. For

organic radicals. In contrast, although the complex

Me4H

−

[

(Cp

Dy) (μ-N )] does exhibit Ising exchange, it

2 2 2

deviates from the trend in Figure 4. This may be due to the

strong equatorial ligand ﬁeld of the N₂³bridge, which reduces

single-ion anisotropy, or due to a change in the proportionality

constant that relates J_G_d₋_r_a_dand J_D_y₋_r_a_d, which could result from

diﬀerences in the exchange mechanism for complexes bridged

versus organic radicals. The complex [({N-

SiMe ) } (THF)Dy) (μ-N )] likewise deviates from the

model, likely due to low single-ion anisotropy.

this model is unlikely to apply to complexes that exhibit low

values of |J_G_d₋_r_a_d| (<1 cm ), wherein single-ion eﬀects are

expected to dominate. Clearly, additional studies on series of

radical-bridged lanthanide complexes with diﬀerent organic

bridging ligands are necessary to elucidate the generality of the

empirical model for 1-Dy through 4-Dy.

To provide additional insight into the electronic structure of

this series and support for the above analysis, ab initio

calculations were performed on 1-Dy through 4-Dy. Complete

active-space self-consistent ﬁeld (CASSCF) calculations were

carried out to determine the energy of the crystal ﬁ eld states of

the individual Dy³ions in each complex (Table S23), and

broken-symmetry density functional theory calculations were

used to estimate the strength of the exchange interaction.

These calculations predict that the ground Kramers doublet of

-Dy and 4-Dy, magnetic relaxation times, τ, were extracted

−

from magnetization versus time plots by ﬁ tting the data to a

stretched exponential function (see section S5.3 of the

Supporting Information). Data for 1-Dy and 2-Dy could not

be ﬁt accurately by a single stretched exponential function,

likely due to the presence of the two independent relaxation

processes occurring for each complex. Further attempts to ﬁt

the dc relaxation data using a stretched exponential function

accounting for two relaxation processes led to overparamete-

rization, and unique ﬁts to the data for 1-Dy and 2-Dy could

therefore not be obtained.

−

by N₂

(

−

24,68

In addition,

−1

Plots of ln(τ) versus 1/T derived from the dc relaxation data

for 3-Dy and 4-Dy are nonlinear (Figures S81 and S83),

indicative of Raman relaxation and quantum tunnelling of the

magnetization. Accordingly, the data were ﬁt to the equation

−

−1

= τ_t_u_n_n_e_l+ CT , where τ

is the relaxation time for

tunnel

quantum tunnelling and C and n are free variables that describe

−

Raman relaxation, yielding τ_t_u_n_n_e_l= 660 s, C = 3.7 × 10 , and n

−

3.7 for 3-Dy, and τ_t_u_n_n_e_l= 350 s, C = 2.9 × 10 , and n = 3.0

for 4-Dy. As τ_t_u_n_n_e_lvalues can be inﬂuenced by intermolecular

dipolar interactions in the solid state, dc relaxation data were

also collected on dilute solutions of 3-Dy (7.8 mM) and 4-Dy

(8.0 mM) in 1,2-diﬂuorobenzene under zero ﬁeld and 500 Oe

(see section S5.3 of the Supporting Information and Figures

S82 and S84). Fits to the zero-ﬁeld data yielded τ_t_u_n_n_e_l= 94 and

each Dy ion in 1-Dy through 4-Dy is well-separated from the

ﬁrst excited doublet (ΔE = ∼200 cm ), consistent with the

−

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Figure 6. Variable-ﬁeld magnetization data collected for 1-Dy through 4-Dy, showing magnetic hysteresis. Sweep rates of 82(2) and 24(1) Oe/s

were used for |H| > 10 kOe and |H | < 10 kOe, respectively.

3 s for 3-Dy and 4-Dy, respectively, and ﬁts to the 500-Oe

factors on the rate of through-barrier magnetic relaxation in

these complexes, and additional investigations into this

behavior is ongoing.

data yielded τ_t_u_n_n_e_l= 34 000 and 35 000 s, respectively. Raman

relaxation parameters extracted from ﬁts to the solution phase

data under an applied ﬁeld are similar for 3-Dy and 4-Dy

Magnetic Relaxation in Terbium Derivatives. The

compounds [(Cp* Tb) (μ-5,5′-R bpym)](BPh ) (R = NMe

Tables S18 and S19). Overall, these results demonstrate that

through-barrier relaxation processes in 3-Dy and 4-Dy occur at

similar rates at low temperatures.

(1-Tb), F (4-Tb)) were also synthesized in order to examine

the impact of changing the lanthanide ion on single-molecule

magnet behavior (see section S1 of the Supporting

Information). Similar to the results for the Gd and Dy

Magnetic Hysteresis. Variable-ﬁeld magnetization data

were collected for 1-Dy through 4-Dy between ±35 kOe, using

a sweep rate of 82(2) Oe/s for |H| > 10 kOe and 24(1) Oe/s

for |H| < 10 kOe. All compounds exhibit magnetic hysteresis,

and the loops for 1-Dy through 4-Dy are open at zero ﬁeld to

temperatures as high as 4.0, 5.0, 6.5, and 7.0 K, respectively

compounds, χ T versus T data collected for 1-Tb and 4-Tb

under a 10 kOe ﬁeld (Figure S42 and S49) pass through

shallow minima at 80 and 170 K upon cooling from 300 K

before rising to maximum χ T values of 22.7 and 27.2 emu K/

(

Figure 6). For comparison, hysteresis loops are open at zero

mol, respectively. This behavior is again indicative of strong

antiferromagnetic exchange. Moreover, the relative temper-

ﬁeld up to 6.5 K in [(Cp* Dy) (μ-bpym)](BPh ) for data

collected with a comparable sweep rate. Interestingly, the

increase in the maximum hysteresis temperature upon moving

from 1-Dy to 4-Dy is consistent with the increase in |J_G_d₋_r_a_d|,

ature and magnitude of the χ T maximum for 1-Tb versus 4-

Tb is similar to what is observed for the corresponding Gd

derivatives, implying that |J_T_b₋_r_a_d| is larger in 4-Tb than in 1-

Tb.

where NMe < OEt < Me = H < F. This result can be

understood by considering that the magnetic relaxation at the

maximum hysteresis temperature for each complex is

dominated by an Orbach mechanism (Figures S78, S80, S81,

Dynamic magnetic susceptibility data were also obtained for

1-Tb and 4-Tb (Figures S52 and S60 ). For 4-Tb, relaxation

−

times extracted from these data were ﬁ t to the equation τ

eff

−

−1

and S83 ). The value of J_G_d₋_r_a_ddictates the magnitude of U in

τ₀exp(U /k T) + CT + τ

= 82 cm , C = 2.5 × 10 , n = 2.0, and τ

For comparison, U = 44 cm for [(Cp* Tb) (μ-bpym)]-

(Figure S85), yielding U

eff

tunnel

−

−2

−1

this series, which in turn determines the maximum temperature

at which the hysteresis loop remains open.

= 7.0 × 10 s.

tunnel

−

eff

The coercive ﬁeld (H ) can be employed to compare the

(BPh₄). The data for 1-Tb exhibit no temperature depend-

ence (Figure S79) and were used to extract τ_t_u_n_n_e_l= 2.9 × 10⁻

s. In contrast to the data for 1-Dy through 4-Dy, a plot of U_e_f_f

for 1-Tb, 4-Tb, and previously reported radical-containing

terbium complexes versus |J _G _d ₋_r _a _d | of the corresponding Gd

complex is not linear (Figure S86 and Table S21). This result

suggests that 1-Tb and 4-Tb possess low single-ion anisotropy

and deviate from an Ising exchange model, thus U_e_f_fis not

predicted to be directly proportional to J_T_b_‑_r_a_d. Indeed, both 1-

Tb and 4-Tb show fast quantum tunneling of the magnet-

ization (τ_t_u_n_n_e_l< 1 s) and magnetic hysteresis loops that are

closed at zero ﬁ eld at 2 K, consistent with low single-ion

anisotropy (Figures S88 and S96). The lower U_e_f_fvalue for

[Cp* Tb](BPh ) relative to [Cp* Dy](BPh ) further supports

rate of magnetic relaxation in a series of complexes if data are

collected at comparable sweep rates. As such, the value of H_c

provides an estimate of the rate of through-barrier relaxation

processes in 1-Dy through 4-Dy at low temperatures, where

Raman relaxation and quantum tunnelling of the magnet-

ization dominate. Values of H = 40, 430, 760, and 580 Oe

were determined for 1-Dy through 4-Dy, respectively, at 2 K.

For comparison, H = 600 Oe for [(Cp* Dy) (μ-bpym)]-

(

BPh ) at 2 K. Generally, these results demonstrate that

complexes with stronger exchange coupling display slower

through-barrier relaxation (e.g., H for 3-Dy and 4-Dy ≫ H

for 1-Dy). However, this correlation is not strictly linear. For

instance, |J_G_d₋_r_a_d| decreases by a factor of 2 from 2-Dy to 1-Dy,

while H decreases by over an order of magnitude. In addition,

this assertion. This analysis suggests that the lower U values

eff

J_G_d₋_r_a_d| follows the trend EtO (2-Dy) < Me (3-Dy) < H < F

observed for organic radical-bridged terbium(III) compounds

relative to their dysprosium(III) counterparts arises from lower

single-ion anisotropy, rather than weaker exchange coupling

interactions as has been previously proposed. Modiﬁcation of

the ancillary Cp* ligands to increase axiality or utilization of a

radical bridging ligand with weaker equatorial interactions thus

represents a clear step toward increasing the magnitude of U_e_f_f

and the temperature at which hysteresis is observed for related

terbium radical-bridged complexes. Future study of a series of

(

4-Dy), while H at 2 K follows the trend EtO < F < H < Me

for both solid-state and solution samples of 3-Dy and 4-Dy

see section S5.6 of the Supporting Information). This result is

(

intriguing, given that CASSCF calculations predict higher

single-ion anisotropy for 4-Dy than 3-Dy and that both

complexes are crystallographically centrosymmetric, which

results in a parallel arrangement of the anisotropy axes of the

Dy ions. This result may indicate the inﬂuence of other

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

high-anisotropy radical-bridged terbium(III) complexes could

furnish useful empirical correlations, such as the one derived

here for 1-Dy through 4-Dy.

(15 mL) and to this solution was added potassium carbonate (1.6 g,

2 mmol, 1.5 equiv). Subsequently, iodoethane (0.92 mL, 1.8 g, 12

mmol, 1.5 equiv) was added dropwise. The reaction was stirred at 25

C for 1 h, at which time water (50 mL) was added. The resulting

mixture was extracted with EtOAc (3 × 30 mL) and the organic layers

CONCLUSIONS

The foregoing analysis demonstrates that modifying the radical

■

were combined, dried over Na SO , and concentrated to yield a light

brown solid. This solid was dissolved in 25% EtOAc in hexanes and

puriﬁed via ﬁltration over silica gel. The product was obtained as a

colorless solid in 86% yield (1.05 g, 6.62 mmol).

ligand substituent in the compounds [(Cp* Ln) (μ-5,5′-

R bpym)](BPh ) (R = NMe (1), OEt (2), Me (3), and F

Synthesis of 5,5′-Di-(N,N-dimethylamino)-2,2′-bipyrimidine

(

4); Ln = Gd and Dy) induces drastic changes in both

(

(NMe ) bpym). Analogous to a literature synthesis of 2,2′-

intramolecular magnetic exchange interactions and single-

molecule magnet behavior. Electron-donating and -with-

drawing substituents decrease and increase the magnitude of

^t^h^e^e^x^c^h^aⁿ^g^e^c^o^u^p^lⁱⁿ^g^c^oⁿ^s^t^aⁿ^t^,^|^JGd−rad|, respectively.

Although the magnetic exchange coupling in 1-Dy through

bipyrimidine, nickel(II) chloride hexahydrate (360 mg, 1.5 mmol,

.25 equiv) and triphenylphosphine (1.6 g, 6.0 mmol, 1.0 equiv) were

placed in a Schlenk ﬂask and dried under reduced pressure (∼200

mTorr) for 20 min. The solids were subsequently dissolved in dry,

degassed DMF (30 mL) under argon with vigorous stirring, and to

this solution was added Zn powder (200 mg, 3.0 mmol, 0.5 equiv).

The resulting green solution became deep red in color and was stirred

for 1 h at 25 °C, at which point 5-(N,N-dimethyl)amino-2-

chloropyrimidine (950 mg, 6.0 mmol) was added. The solution

became black in color and was stirred for 1 h at 25 °C, followed by 48

h at 50 °C. The black solution was cooled to room temperature and

ﬁltered through diatomaceous earth, and the ﬁlter cake was washed

with DMF (30 mL). The ﬁltrate was concentrated under reduced

pressure (∼50 °C, 200 mTorr) to yield a brown oil, to which was

added a solution of ethylenediaminetetraacetic acid (2.6 g, 9.0 mmol,

-Dy was not quantiﬁed in this study, dc magnetic

susceptibility data indicate that the relative magnitudes can

be correlated with |J_G_d₋_r_a_d| for the Gd³congeners. Further, we

identify a linear relationship between the thermal barriers to

magnetic relaxation (U ) of the highly anisotropic 1-Dy

eff

through 4-Dy and the |J_G_d₋_r_a_d| values for the corresponding

Gd complexes. In the case of 1-Dy through 4-Dy, larger

|^JGd−rad^|^v^a^l^u^e^s^a^r^e^c^o^r^r^e^l^a^t^e^d^wⁱ^t^h^l^a^r^g^e^r^Ueff values, higher

maximum hysteresis temperatures, and smaller values of τ for

through-barrier relaxation processes, results that are supported

by ab initio calculations. Notably, the empirical correlation

1.5 equiv) in aqueous ammonia (7% NH in water; 20 mL). This

solution was stirred for 3 h, becoming blue in color and precipitating

colorless solids that were collected via vacuum ﬁltration. The solids

between U and |J_G_d₋_r_a_d| as shown here for 1-Dy through 4-Dy

eff

were triturated in Et O (3 × 20 mL) to remove triphenylphosphine

extends to other organic radical-bridged dysprosium com-

pounds in the literature and suggests clear design strategies to

and were then puriﬁed via recrystallization (1:10 CHCl /hexanes or

MeOH). The product was obtained as a pale yellow solid in 33% yield

increase U for molecules in this class. Together, these results

eff

(

480 mg, 2.0 mmol). H NMR (600 MHz, CDCl ) δ = 8.40 (s, 4H),

demonstrate the utility of systematic analysis of related

compounds in developing greater predictability in the design

of radical-bridged single-molecule magnets.

.10 (s, 12H) ppm. C NMR (150 MHz, CDCl ) δ = 150.0, 142.5,

40.4, 39.7 ppm. HRMS (ESI+) m/z calcd. for C H N [M+H ]:

45.1509; found: 245.1511.

Synthesis of 5,5′-Diethoxy-2,2′-bipyrimidine ((OEt) bpym).

EXPERIMENTAL SECTION

Analogous to the synthesis of (NMe ) bpym, nickel(II) chloride

■

2 2

hexahydrate (380 mg, 1.6 mmol, 0.25 equiv) and triphenylphosphine

1.7 g, 6.4 mmol, 1 equiv) were placed in a Schlenk ﬂask and dried

Materials and Methods. Unless otherwise mentioned, commer-

cial reagents were purchased from Sigma-Aldrich, Fischer, Acros,

Oakwood, Strem or Alfa Aesar and used without further puriﬁcation.

Tetrahydrofuran (THF), hexanes, 1,2-diﬂuorobenzene (DFB),

toluene, and N,N-dimethylformamide (DMF) were sparged with

argon and then dried by passing through alumina columns in a Glass

(

under reduced pressure (∼200 mTorr) for 20 min. The solids were

subsequently dissolved in dry, degassed DMF (40 mL) under argon

with vigorous stirring, and to this solution was added Zn powder (210

mg, 3.2 mmol, 0.5 equiv). The resulting green solution became deep

red in color and was stirred for 1h at 25 °C, at which point 5-ethoxy-

2-chloropyrimidine (1.0 g, 6.4 mmol) was added. The solution

became black in color and was stirred for 1 h at 25 °C, followed by 48

h at 50 °C. The black solution was cooled to room temperature,

ﬁltered through diatomaceous earth, and the ﬁlter cake was washed

with DMF (40 mL). The ﬁltrate was concentrated under reduced

pressure (∼50 °C, 200 mTorr) to yield a brown oil to which was

added a solution of ethylenediaminetetraacetic acid (2.7 g, 9.6 mmol,

Contour solvent puriﬁcation system from JC Meyer. iPr NH was

distilled and then dried over activated 4 Å molecular sieves. The

compounds 5,5′-dimethylbipyrimidine, [Cp* Gd](BPh ), and

[

Cp* Dy](BPh ) were prepared according to literature reports. 5-

2 4

Ethoxy-2-chloropyrimidine and 5-(N,N-dimethyl)amino-2-chloro-

pyrimidine were previously reported in the literature but were

prepared via alternative routes. All reactions were carried out in ﬂame-

dried glassware under an argon atmosphere using standard Schlenk

techniques or in a nitrogen atmosphere glovebox. Compositional C,

H, and N analyses were performed by the Microanalytical Facility at

the University of California, Berkeley using a PerkinElmer 2400 Series

II combustion analyzer.

1.5 equiv) in aqueous ammonia (7% NH

solution was stirred for 3 h, becoming green in color, and it was

in water; 30 mL). This

subsequently extracted with CHCl

were combined, dried over Na SO

solids were triturated in Et

triphenylphosphine and then puriﬁed via recrystallization (1:10

CHCl /hexanes or MeOH). The product was obtained as a colorless

solid in 37% yield (580 mg, 2.4 mmol). H NMR (500 MHz, CDCl

δ = 8.55 (s, 4H), 4.20 (q, J = 7.0 Hz, 4H), 1.46 (t, J = 7.0 Hz, 6H)

(3 × 30 mL). The organic layers

, and concentrated. The resulting

Synthesis of 5-(N,N-Dimethyl)amino-2-chloropyrimidine.

The molecule 5-amino-2-chloropyrimidine (4.5 g, 35 mmol) was

dissolved in 90% formic acid (7.3 mL, 30 equiv) and to this solution

was added an aqueous solution of formaldehyde (36% formaldehyde

in water, 7.8 mL, 18 equiv). The reaction was heated to reﬂux (∼105

O (3 × 30 mL) to remove excess

)

C) for 12 h at which time the solution was allowed to cool to 50 °C,

ppm. C NMR (125 MHz, CDCl ) δ = 154.8, 152.7, 144.3, 64.8,

and the solvent was removed under reduced pressure (∼200 mTorr).

The remaining waxy black solid was triturated with a saturated

aqueous solution of sodium bicarbonate (200 mL), and the resulting

brown solids were collected via vacuum ﬁltration and puriﬁed by

sublimation (∼80 °C, 200 mTorr). The product was obtained as a

colorless solid in 30% yield (1.65 g, 10.5 mmol).

14.7 ppm. HRMS (ESI+) m/z calcd. for C12H O N [M+H ]:

15 2 4

247.1190; found: 247.1191.

Synthesis of 5-Fluoro-2-tributylstannylpyrimidine. A sol-

ution of iPr NH (0.78 mL, 5.5 mmol, 1.3 equiv) in dry THF (10 mL)

was stirred at 0 °C in a Schlenk ﬂask under argon. To this solution

was added n-BuLi (2.5 M in hexanes, 2.0 mL, 5.0 mmol, 1.2 equiv)

dropwise, and the resulting pale yellow solution was stirred for 30 min

Synthesis of 5-Ethoxy-2-chloropyrimidine. The molecule 5-

hydroxy-2-chloropyrimidine (1.0 g, 7.7 mmol) was dissolved in DMF

at 0 °C. HSnBu (1.3 mL, 4.8 mmol, 1.1 equiv) was then added

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

dropwise, and the light green solution was stirred for 10 min at 0 °C,

cooled to −78 °C for 5 min, and then transferred via cannula to a

solution of 5-ﬂuoro-2-chloropyrimidine (580 mg, 0.40 mL, 4.4 mmol)

in dry THF (20 mL) that had been precooled to −78 °C. The

resulting deep orange-red solution was stirred at −78 °C for 2 h and

solution of ethylenediaminetetraacetic acid (0.47 g, 1.6 mmol, 1.5

equiv) in aqueous ammonia (7% NH in water; 20 mL) was added.

This solution was stirred for 3 h, becoming green in color, and then

was extracted with hexanes (3 × 20 mL) and subsequently with

CH Cl (3 × 20 mL). The CH Cl layers were combined, dried over

then 0 °C for 2 h, at which point it was quenched by addition of H O

Na SO , and concentrated to yield a brown solid that was puriﬁed via

2 4

(

30 mL). The reaction mixture was extracted with EtOAc (3 × 30

sublimation (∼120 °C, 200 mTorr). The product was obtained as a

mL), and the organic layers were dried over Na SO and then

colorless solid in 46% yield (102 mg, 0.5 mmol). H NMR (500 MHz,

concentrated under reduced pressure to yield a brown oil, which was

CDCl ) δ = 9.22 (s, 2H), 8.88 (s, 2H) ppm. C NMR (125 MHz,

puriﬁed via column chromatography on silica gel (R = 0.70, 10%

CDCl ) δ = 163.8, 158.2 (d, J = 270.0 Hz), 157.4, 155.6 (q, J = 3.3

EtOAc in hexanes). The product was obtained as a colorless oil in

Hz), 146.3 (d, J = 20.7 Hz), 125.0 (q, J = 34.0 Hz), 122.6 (q, J =

5% yield (590 mg, 1.5 mmol). H NMR (400 MHz, C D ) δ = 8.14

271.0 Hz) ppm. F-NMR (376 MHz, CDCl ) −61.5, −132.3 ppm.

(

s, 2H), 1.72 (m, 6H), 1.40 (m, 6H), 1.29 (t, J = 7.3 Hz, 6H), 0.92 (t,

HRMS (ESI+) m/z calcd. for C H N F [M+H ]: 245.0445; found:

4 4

J = 7.3 Hz, 9H) ppm. C NMR (150 MHz, C D ) δ = 184.2 (d, J =

245.0446.

Synthesis of [(GdCp* ) (μ-5,5′-(NMe ) bpym)](BPh ) (1-Gd).

0.7 Hz), 157.7 (d, J = 266.7 Hz), 142.6 (d, J = 15.3 Hz), 29.4, 27.7,

2 2

4.0, 10.9 ppm. ¹⁹F-NMR (376 MHz, C D ) −136.7 ppm. HRMS

The molecule [Cp* Gd](BPh ) (42.5 mg, 0.0568 mmol) was

(

ESI+) m/z calcd. for C H N FSn [M+H ]: 389.1410; found:

suspended in toluene (4 mL), and to this slurry was added

(NMe ) bpym (6.9 mg, 0.028 mmol, 0.5 equiv). The resulting

389.1410.

Synthesis of 5-Triﬂuoromethyl-2-tributylstannylpyrimidine.

solution was stirred for 30 min, during which time it changed color

Analogous to the synthesis of 5-ﬂuoro-2-tributylstannylpyrimidine, a

from pale yellow to bright orange. At this point, KC (3.8 mg, 0.028

solution of iPr NH (0.58 mL, 4.1 mmol, 1.3 equiv) in dry THF (10

mmol, 0.5 equiv) was added in THF (0.5 mL), and the reaction

became dark brown in color. After stirring 30 min, the solvent was

removed under reduced pressure and THF (2 mL) was added to form

a dark reddish-black solution. Black and white insoluble solids were

removed by ﬁltration through diatomaceous earth, and then the

solution was layered with toluene (2 mL) and cooled to −30 °C,

aﬀording red, block-shaped crystals of 1-Gd after 48 h (22.0 mg, 55%

mL) was stirred at 0 °C in a Schlenk ﬂask under argon. To this

solution was added n-BuLi (2.5 M in hexanes, 1.5 mL, 3.8 mmol, 1.2

equiv) dropwise, and the resulting solution was stirred for 30 min at 0

C. HSnBu (1.0 mL, 3.7 mmol, 1.1 equiv) was then added dropwise

and the solution was stirred for 10 min at 0 °C, cooled to −78 °C for

min, and then transferred via cannula to a solution of 5-

triﬂuoromethyl-2-chloropyrimidine (600 mg, 3.3 mmol) in dry THF

yield). Anal. Calcd. For C76

Found: C, 64.54; H, 6.78; N, 5.92.

Synthesis of [(GdCp* (μ-5,5′-(OEt)

Analogous to the synthesis of 1-Gd, [Cp*

0.0568 mmol) was stirred in toluene (4 mL) with (OEt)

mg, 0.028 mmol, 0.5 equiv) and then subsequently KC

H96BN Gd : C, 64.33; H, 6.82; N, 5.92.

6 2

(

20 mL) that had been precooled to −78 °C. The resulting solution

was stirred at −78 °C for 2 h and then 0 °C for 2 h, at which point it

was quenched by addition of H O (30 mL). The reaction mixture was

)

bpym)](BPh

Gd](BPh ) (42.5 mg,

bpym (7.0

(3.8 mg,

) (2-Gd).

extracted with EtOAc (3 × 30 mL), and the organic layers were dried

over Na SO and concentrated under reduced pressure to yield a

brown oil, which was puriﬁed via column chromatography on silica gel

R = 0.60, 10% EtOAc in hexanes). The product was obtained as a

colorless oil in 50% yield (720 mg, 1.6 mmol). H NMR (600 MHz,

0.028 mmol, 0.5 equiv) was added in THF (0.5 mL). Insoluble solids

were removed by centrifugation and then red, block-shaped crystals of

2-Gd were grown from a layered THF-toluene solution (2 mL THF,

2 mL toluene) cooled to −30 °C for 48 h (15.6 mg, 39% yield). Anal.

(

C D ) δ = 8.38 (s, 2H), 1.33 (m, 6H), 1.14 (m, 6H), 0.82 (m, 15H)

ppm. C NMR (150 MHz, C D ) δ = 162.4, 155.4 (t, J = 7.6 Hz),

Calcd. For C76

64.55; H, 6.43; N, 4.32.

Synthesis of [(GdCp*

Analogous to the synthesis of 1-Gd, [Cp*

0.0568 mmol) was stirred in toluene (4 mL) with Me

0.028 mmol, 0.5 equiv) and then subsequently KC

H94BN O Gd : C, 64.24; H, 6.67; N, 3.94. Found: C,

4 2 2

34.2 (q, J = 14.3 Hz), 133.2 (q, J = 289.7 Hz), 28.9, 27.5, 13.8, 10.4

ppm. F-NMR (376 MHz, C D ) −98.4 ppm. HRMS (ESI+) m/z

calcd. for C H N F Sn [M+H ]: 439.1378; found: 439.1384.

)

(μ-5,5′-Me

bpym)](BPh

Gd](BPh ) (42.5 mg,

bpym (5.3 mg,

(3.8 mg, 0.028

) (3-Gd).

2 3

Synthesis of 5,5′-Diﬂuoro-2,2′-bipyrimidine (F bpym). The

molecules 5-ﬂuoro-2-chloropyrimidine (130 mg, 1.0 mmol), 5-ﬂuoro-

-tributylstannylpyrimidine (370 mg, 1.0 mmol, 1.0 equiv), copper(I)

mmol, 0.5 equiv) was added in THF (0.5 mL). Insoluble solids were

removed by centrifugation and then red, block-shaped crystals of 3-

Gd were grown from a layered THF-toluene solution (2 mL THF, 2

mL toluene) cooled to −30 °C for 48 h (14.0 mg, 36% yield). Anal.

chloride (87 mg, 1.1 mmol, 1.1 equiv), and Pd(dppf)Cl (70 mg, 0.1

mmol, 10 mol %) were dissolved in dry THF (10 mL) in a Schlenk

ﬂask under argon. The resulting solution was heated to 70 °C for 18

h, ﬁrst turning yellow and then brown. The reaction was then allowed

to cool to ambient temperature, concentrated under reduced pressure,

and a solution of ethylenediaminetetraacetic acid (0.43 g, 1.5 mmol,

Calcd. for C74

65.21; H, 6.68; N, 3.86.

Synthesis of [(GdCp*

molecule[Cp* Gd](BPh ) (77.0 mg, 0.103 mmol) was suspended in

toluene (4 mL) and to this slurry was added F bpym (10.0 mg,

0.0515 mmol, 0.5 equiv). This reddish-orange solution was stirred for

30 min and then KC (6.9 mg, 0.052 mmol, 0.5 equiv) was added in

H90BN Gd : C, 65.31; H, 6.67; N, 4.12. Found: C,

4 2

)

(μ-5,5′-F bpym)](BPh ) (4-Gd). The

2 4

.5 equiv) in aqueous ammonia (7% NH in water; 20 mL) was

added. This solution was stirred for 3 h, becoming green in color, and

then was extracted with Et O (3 × 20 mL) and subsequently with

CH Cl (3 × 20 mL). The CH Cl layers were combined, dried over

Na SO , and concentrated under reduced pressure to yield a brown

toluene (0.5 mL). The resulting reddish-black solution was stirred for

24 h and then concentrated under reduced pressure, dissolved in 1,2-

diﬂuorobenzene (DFB, 2 mL), and ﬁltered through diatomaceous

earth. Red, block-shaped crystals of 4-Gd were grown from a layered

DFB-hexanes solution (2 mL DFB, 2 mL hexanes) stored at 25 °C for

solid that was puriﬁed via sublimation (∼120 °C, 200 mTorr). The

product was obtained as a colorless solid in 57% yield (106 mg, 0.6

mmol). H NMR (500 MHz, CDCl ) δ = 8.86 (s, 4H) ppm.

NMR (125 MHz, CDCl ) δ = ppm. F-NMR (376 MHz, CDCl )

−

134.1 ppm. HRMS (ESI+) m/z calcd. for C H N F [M+H ]:

48 h (30.6 mg, 43% yield). Anal. Calcd. For C76

H96BN Dy : C, 63.18;

6 2

4 2

95.0477; found: 195.0479.

Synthesis of 5-Fluoro-5′-triﬂuoromethyl-2,2′-bipyrimidine

F-CF -bpym). Analogous to the synthesis of F bpym, 5-triﬂuor-

H, 6.19; N, 4.00. Found: C, 63.08; H, 6.18; N, 3.79.

Mass Spectrometry. High-res mass spectrometric data were

obtained from the Mass Spectrometry Facility at the University of

California, Berkeley, on a Finnigan/Thermo LTQ-FT instrument

(ESI); data acquisition and processing were performed using the

Xcalibur software.

(

omethyl-2-chloropyrimidine (280 mg, 1.6 mmol, 1.5 equiv), 5-ﬂuoro-

-tributylstannylpyrimidine (400 mg, 1.1 mmol), copper(I) chloride

110 mg, 1.2 mmol, 1.1 equiv), and Pd(dppf)Cl (76 mg, 0.1 mmol,

(

0 mol %) were dissolved in dry THF (10 mL) in a Schlenk ﬂask

NMR Spectroscopy. NMR spectroscopic data were obtained for

solutions in deuterated solvents (CDCl or C D ) purchased from

under argon. The resulting solution was heated to 70 °C for 18 h, ﬁrst

turning yellow and then brown. The reaction was then allowed to cool

to ambient temperature, concentrated under reduced pressure, and a

Cambridge Isotope Laboratories, Inc. H NMR, C NMR, and

NMR data were recorded on Bruker AVQ-400, DRX-500, AV-500,

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

https://pubs.acs.org/doi/10.1021/jacs.0c10612.

Article

and AV-600 spectrometers. Chemical shifts (δ) are reported in ppm

were performed on the single crystal X-ray diﬀraction structures with

relative to the residual solvent peak (δ 7.26 for CDCl and δ 7.16 for

nearby Dy ions replaced with Lu . The methyl groups of the

C D for H NMR; δ 77.16 for CDCl and δ 128.06 for C D for

cyclopentadienyl ligand around each Lu ion were replaced by

NMR). Data for H NMR are reported in the following format:

chemical shift (ppm) (multiplicity (s = singlet, d = doublet, t = triplet,

q = quartet, m = multiplet), coupling constant (Hz), integration).

Data for ¹³C NMR and F-NMR are reported in terms of chemical

hydrogen atoms. The Cholesky decomposition threshold was set to 5

−

× 10 Hartree to save disk space. The ANO-RCC basis set was used

for all atoms (Table S22). Four point charges on the bridging carbon

and nitrogen atoms, each −0.25e, were included to consider the

•−

shift (ppm) with coupling constants for F− C coupling where

electrostatic potential from the unpaired electron of bpym

applicable.

CASSCF calculations comprised seven 4f type orbitals. Twenty-one

sextet, 128 quartet and 130 doublet states were admixed by spin−

orbit coupling within the RASSI program. Based on the obtained

spin−orbital states, local magnetic properties were calculated within

Infrared Spectroscopy. FT-IR spectra were recorded on a

PerkinElmer Avatar Spectrum 400 FTIR Spectrometer equipped with

attenuated total reﬂectance (ATR).

Single-Crystal X-ray Diﬀraction. Single-crystal X-ray diﬀraction

data were collected at small-molecule crystallography beamlines

the SINGLE_ANISO program. Finally, the exchange interaction

92,93

was included within the POLY_ANISO module.

To estimate the exchange coupling parameters between the Dy

ions and the bpym-radical ligand, broken-symmetry DFT calcu-

(

beamline 11.3.1 for 2-Gd, 3-Gd, and 3-Dy; and beamline 12.2.1 for

-Dy, 4-Gd, and 4-Dy) at the Advanced Light Source, Lawrence

Berkeley National Laboratory. Single crystals were coated with

Paratone-N oil, mounted on a MiTeGen loop, and frozen at 100 K

lations were employed by using ORCA 3.0.0 with SVP basis set,

B3LYP functional, Grid6 and TightSCF settings. Because Dy ions

under a stream of N from an Oxford Cryostems Cryostream 700 Plus

are multiconﬁgurational in their nature, they cannot be treated

on a Bruker AXS D8 diﬀractometer. Data were collected through a

combination of 4° and 1° ϕ and ω scans. Data reduction was

performed through SAINT and absorption correction through

adequately by DFT methods. Therefore, the Dy ions were replaced

with Gd , while preserving the position of all atoms intact. In this

way one can extract the exchange interaction between the Gd ions

80−82

SADABS (or TWINABS for 3-Dy).

and the radical and then rescale it to the spin of Dy to calculate

Single-crystal X-ray diﬀraction data for 1-Dy were collected at the

University of California, Berkeley using a Rigaku XtaLAB p200

equipped with a MicroMax-007 HF microfocus rotating anode and a

Dy−rad. This is achieved by multiplying the former value by 49/25.

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

■

Pilatus 200 K hybrid pixel array detector at 100 K under a N stream

of an Oxford Cryostems Cryostream with Mo Kα radiation (graphite

Pro

monochromator). The frames were integrated with CrysAlis

software, including a multiscan absorption correction that was applied

Pro 83

using the SCALE3 ABSPACK scaling algorithm within CrysAlis

Additional experimental details, NMR spectroscopy, IR

spectroscopy, X-ray crystallography reﬁnement details,

magnetic measurements, description of calculations, and

supplementary discussion (PDF)

Structure solutions were performed by SHELXT using direct

methods and were reﬁned by least-squares reﬁnement against F by

^S^H^E^L^X^L8 following standard procedures via OLEX2 crystallographic

software. For all structures, all non-hydrogen atoms were reﬁned

anisotropically. All hydrogen atoms were placed on geometrically

calculated positions using the riding model and reﬁned isotropically.

Magnetic Measurements. Each sample for magnetic measure-

ments was prepared by adding a polycrystalline powder (20.9 mg of 1-

Gd, 13.6 mg of 1-Dy, 7.0 mg of 1-Tb, 7.0 mg of 2-Gd, 17.4 mg of 2-

Dy, 6.8 mg of 3-Gd, 16.2 mg of 3-Dy, 4.6 mg of 4-Gd, 10.6 mg of 4-

Dy, and 12.4 mg of 4-Tb) to a 5 mm i.d./7 mm o.d. quartz tube with a

raised quartz platform. A layer of eicosane was added on top of the

sample (14.5 mg for 1-Gd, 16.7 mg for 1-Dy, 15.0 mg for 1-Tb, 15.7

mg for 2-Gd, 17.7 mg for 2-Dy, 18.1 mg for 3-Gd, 28.8 mg for 3-Dy,

Crystallographic structure data for 3-Dy (CIF )

Crystallographic structure data for 2-Gd (CIF)

Crystallographic structure data for 2-Dy (CIF)

Crystallographic structure data for 4-Dy (CIF )

Crystallographic structure data for 4-Gd (CIF)

Crystallographic structure data for 1-Dy (CIF )

Crystallographic structure data for 3-Gd (CIF)

6.6 mg for 4-Gd, 18.7 mg for 4-Dy, and 18.9 mg for 4-Tb) to provide

AUTHOR INFORMATION

Corresponding Author

good thermal contact between the sample and the bath and to prevent

crystallite torqueing. The tubes were ﬁtted with Teﬂon-sealable

adapters, evacuated using a glovebox vacuum pump, and then ﬂame-

sealed with an O /H ﬂame under vacuum. After ﬂame-sealing, the

eicosane was melted in a 45 °C water bath. Magnetic measurements

were also conducted on a 7.8 mM solution of 3-Dy and an 8.0 mM

solution of 4-Dy in 1,2-diﬂuorobenzene.

Magnetic susceptibility measurements were collected using a

Quantum Design MPMS2 SQUID magnetometer. All data were

corrected for diamagnetic contributions from the core diamagnetism

of the sample and for the diamagnetism of the eicosane used to

suspend the sample, estimated using Pascal’s constants to give

corrections of χ_d_i_a= −0.000814 for 1-Gd, −0.000812 for 1-Dy,

−

■

Jeﬀrey R. Long − Department of Chemistry and Department

of Chemical and Biomolecular Engineering, University of

California, Berkeley, Berkeley, California 94720, United

States; Materials Sciences Division, Lawrence Berkeley

National Laboratory, Berkeley, California 94720, United

States; orcid.org/0000-0002-5324-1321;

Email: jrlong@berkeley.edu

Authors

Colin A. Gould − Department of Chemistry, University of

California, Berkeley, Berkeley, California 94720, United

States; orcid.org/0000-0001-9539-1582

Edward Mu − Department of Chemistry, University of

California, Berkeley, Berkeley, California 94720, United

States

Veacheslav Vieru − Theory of Nanomaterials Group,

Katholieke Universiteit Leuven, 3001 Leuven, Belgium

Lucy E. Darago − Department of Chemistry, University of

California, Berkeley, Berkeley, California 94720, United

States

0.000812 for 1-Tb, −0.000806 for 2-Gd, −0.000869 for 2-Dy,

0.000773 for 3-Gd, −0.000771 for 3-Dy, −0.000756 for 4-Gd,

0.000754 for 4-Dy, and −0.000754 for 4-Tb.

Fits to the dc susceptibility data for 1-Gd through 4-Gd were

performed using PHI. Uncertainties for values of τ and magnetic

relaxation parameters were determined using the α and n values

extracted from ﬁts to ac susceptibility and dc relaxation measure-

88,89

ments, respectively.

Calculations. The local electronic and magnetic properties of the

Dy³⁺ions in 1-Dy through 4-Dy and [(Cp* Dy) (μ-ind)] were

calculated with the MOLCAS 8.2 program. Fragment calculations

−

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

(11) Rinehart, J. D.; Long, J. R. Exploiting Single-Ion Anisotropy in

Article

Khetpakorn Chakarawet − Department of Chemistry,

University of California, Berkeley, Berkeley, California 94720,

United States; orcid.org/0000-0001-5905-3578

Miguel I. Gonzalez − Department of Chemistry, University of

California, Berkeley, Berkeley, California 94720, United

States

Selvan Demir − Department of Chemistry, University of

California, Berkeley, Berkeley, California 94720, United

States; Department of Chemistry, Michigan State University,

East Lansing, Michigan 48824, United States

the Design of f-Element Single-Molecule Magnets. Chem. Sci. 2011, 2,

(

078−2085.

12) Chilton, N. F.; Collison, D.; McInnes, E. J. L.; Winpenny, R. E.

P.; Soncini, A. An electrostatic model for the determination of

magnetic anisotropy in dysprosium complexes. Nat. Commun. 2013,

, 2551.

13) Sievers, J. Z. Asphericity of 4 f -shells in their Hund ’s rule ground

states. Z. Phys. B: Condens. Matter Quanta 1982 , 45 , 289 −296.

14) Ungur, L.; Chibotaru, L. F. Magnetic anisotropy in the excited

states of low symmetry lanthanide complexes. Phys. Chem. Chem. Phys.

011, 13, 20086.

15) Chilton, N. F.; Goodwin, C. A. P.; Mills, D. P.; Winpenny, R. E.

https://pubs.acs.org/10.1021/jacs.0c10612

Complete contact information is available at:

P. The first near-linear bis(amide) f-block complex: a blueprint for a

high temperature single molecule magnet. Chem. Commun. 2015, 51,

Notes

01.

The authors declare no competing ﬁnancial interest.

(

16) Ding, Y.-S.; Han, T.; Zhai, Y.-Q.; Reta, D.; Chilton, N. F.;

Winpenny, R. E. P.; Zheng, Y.-Z. A Study of Magnetic Relaxation in

Dysprosium(III) Single-Molecule Magnets. Chem. - Eur. J. 2020, 26,

ACKNOWLEDGMENTS

■

17) Harriman, K. L.; Korobkov, I.; Murugesu, M. From a Piano

893.

This work was supported by NSF grant CHE-1800252. Single-

crystal X-ray diﬀraction experiments were performed at

beamlines 11.3.1 and 12.2.1 at the Advanced Light Source at

Lawrence Berkeley National Laboratory. The Advanced Light

Source is supported by the Director, Oﬃce of Science, Oﬃce

of Basic Energy Sciences, of the U.S. Department of Energy

under Contract No. DE-AC02-05CH11231. We thank the

National Science Foundation Graduate Research Fellowship

Program for support of C.A.G. and L.E.D. We thank Dr. Hasan

Celik and UC Berkeley’s NMR facility in the College of

Chemistry (CoC-NMR) for spectroscopic assistance. Instru-

ments in the CoC-NMR are supported in part by an NIH

Grant S10OD024998. V.V. acknowledges the postdoctoral

fellowship of Fonds Wetenschappelijk Onderzoek Vlaanderen

Stoll to a Sandwich: A Stepwise Route for Improving the Slow

Magnetic Relaxation Properties of Thulium. Organometallics 2017, 36,

4515−4518.

(18) Jiang, S.-D.; Liu, S.-S.; Zhou, L.-N.; Wang, B.-W.; Wang, Z.-M.;

Gao, S. Series of Lanthanide Organometallic Single-Ion Magnets.

Inorg. Chem. 2012, 51, 3079−3087.

(19) Goodwin, C. A. P.; Ortu, F.; Reta, D.; Chilton, N. F.; Mills, D.

P. Molecular magnetic hysteresis at 60 K in dysprosocenium. Nature

017, 548, 439−442.

20) Guo, F.-S.; Day, B. M.; Chen, Y.-C.; Tong, M.-L.;

Mansikkamaki, A.; Layfield, R. A. A Dysprosium Metallocene

(

Single-Molecule Magnet Functioning at the Axial Limit. Angew.

Chem., Int. Ed. 2017, 56, 11445−11449.

(21) McClain, K. R.; Gould, C. A.; Chakarawet, K.; Teat, S. J.;

Groshens, T. J.; Long, J. R.; Harvey, B. G. High-temperature magnetic

blocking and magneto-structural correlations in a series of

dysprosium(III) metallocenium single-molecule magnets. Chem. Sci.

(

FWO, Flemish Science Foundation) and the V435018N

FWO travel grant to UC Berkeley. We also thank Dr. Rebecca

L. Siegelman for assistance with crystallography, Dr. T. David

Harris for helpful discussions, and Dr. Katie R. Meihaus for

editorial assistance.

018, 9, 8492−8503.

22) Guo, F.-S.; Day, B. M.; Chen, Y.-C.; Tong, M.-L.;

Mansikkamaki, A.; Layfield, R. A. Magnetic hysteresis up to 80 K in

(

a dysprosium metallocene single-molecule magnet. Science 2018, 362,

1400−1403.

REFERENCES

■

23) Demir, S.; Gonzalez, M. I.; Darago, L. E.; Evans, W. J.; Long, J.

3) Bogani, L.; Wernsdorfer, W. Molecular spintronics using single-

1) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Magnetic

bistability in a metal-ion cluster. Nature 1993, 365, 141−143.

2) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets;

Oxford University Press: New York, 2006.

molecule magnets. Nat. Mater. 2008, 7, 179−186.

4) Vincent, R.; Klyatskaya, S.; Ruben, M.; Wernsdorfer, W.;

3 ‑

^R ^.^G ⁱ ^a ⁿ ^t ^c ^o ^e ^r ^c ⁱ ^v ⁱ ^t ^y ^a ⁿ ^d ^h ⁱ ^g ^h ^m ^a ^g ⁿ ^e ^t ⁱ ^c ^b ^l ^o ^c ^k ⁱ ⁿ ^g ^t ^e ^m ^p ^e ^r ^a ^t ^u ^r ^e ^s ^f ^o ^r ^N2

radical-bridged dilanthanide complexes upon ligand dissociation. Nat.

Commun. 2017, 8, 2144.

(

24) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. Strong

‑

^e ^x ^c ^h ^a ⁿ ^g ^e ^a ⁿ ^d ^m ^a ^g ⁿ ^e ^t ⁱ ^c ^b ^l ^o ^c ^k ⁱ ⁿ ^g ⁱ ⁿ ^N 2 -radical-bridged lanthanide

complexes. Nat. Chem. 2011, 3, 538−542.

(

3 ‑

25) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. A N

Balestro, F. Electronic read-out of a single nuclear spin using a

Radical-Bridged Terbium Complex Exhibiting Magnetic Hysteresis at

4 K. J. Am. Chem. Soc. 2011, 133, 14236−14239.

26) Demir, S.; Jeon, I. R.; Long, J. R.; Harris, T. D. Radical ligand-

containing single-molecule magnets. Coord. Chem. Rev. 2015, 289−

90, 149−176.

27) Giansiracusa, M. J.; Moreno-Pineda, E.; Hussain, R.; Marx, R.;

Martinez Prada, M.; Neugebauer, P.; Al-Badran, S.; Collison, D.;

Tuna, F.; van Slageren, J.; Carretta, S.; Guidi, T.; McInnes, E. J. L.;

Winpenny, R. E. P.; Chilton, N. F. Measurement of Magnetic

Exchange in Asymmetric Lanthanide Dimetallics: Toward a Trans-

ferable Theoretical Framework. J. Am. Chem. Soc. 2018, 140, 2504−

2513.

molecular spin transistor. Nature 2012, 488, 357−360.

Wernsdorfer, W. Electrically driven nuclear spin resonance in single-

5) Thiele, S.; Balestro, F.; Ballou, R.; Klyatskaya, S.; Ruben, M.;

molecule magnets. Science 2014, 344, 1135−1138.

(

(6) Heersche, H. B.; de Groot, Z.; Folk, J. A.; van der Zant, H. S. J.;

Romeike, C.; Wegewijs, M. R.; Zobbi, L.; Barreca, D.; Tondello, E.;

Cornia, A. Electron Transport through Single Mn12 Molecular

Magnets. Phys. Rev. Lett. 2006, 96, 206801.

8) Escalera-Moreno, L.; Baldovi, J. J.; Gaita-Arino, A.; Coronado, E.

7) Liddle, S. T.; van Slageren, J. Improving f-element single

molecule magnets. Chem. Soc. Rev. 2015, 44, 6655−6669.

Spin states, vibrations, and spin relaxation in molecular nanomagnets

and spin qubits: a critical perspective. Chem. Sci. 2018, 9, 3265 −3275.

(28) Meng, Y.-S.; Xiong, J.; Yang, M.-W.; Qiao, Y.-S.; Zhong, Z.-Q.;

Sun, H.-L.; Han, J.-B.; Liu, T.; Wang, B.-W.; Gao, S. Experimental

Determination of Magnetic Anisotropy in Exchange-Bias Dysprosium

Metallocene Single-Molecule Magnets. Angew. Chem., Int. Ed. 2020,

59, 13037−13043.

Single-Molecule Magnets. Chem. Rev. 2013 , 113 , 5110 −5148.

9) Woodruff, D. N.; Winpenny, R. E. P.; Layfield, R. A. Lanthanide

10) Lu, J.; Guo, M.; Tang, J. Recent Developments in Lanthanide

Single-Molecule Magnets. Chem. - Asian J. 2017, 12, 2772−2779.

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

29) Chaudhuri, P.; Wagner, R.; Weyhermu

ller, T. Ferromagnetic vs

(48) Berg, D. J.; Boncella, J. M.; Andersen, R. A. Preparation of

Antiferromagnetic Coupling in Bis(μ -phenoxo)dicopper(II) Com-

plexes. Tuning of the Nature of Exchange Coupling by Remote

Ligand Substituents. Inorg. Chem. 2007, 46, 5134−5136.

Coordination Compounds of Cp * Yb with Heterocyclic Nitrogen

Bases: Examples of Antiferromagnetic Exchange Coupling Across

Bridging Ligands. Organometallics 2002, 21, 4622−4631.

(49) Goudy, V.; Jaoul, A.; Cordier, M.; Clavaguera, C.; Nocton, G.

(

30) Ni, Z.-H.; Zhang, L.-F.; Tangoulis, V.; Wernsdorfer, W.; Cui,

A.-L.; Sato, O.; Kou, H.-Z. Substituent Effect on Formation of

Heterometallic Molecular Wheels: Synthesis, Crystal Structure, and

Magnetic Properties. Inorg. Chem. 2007, 46, 6029−6037.

Tuning the Stability of Pd(IV) Intermediates Using a Redox Non-

innocent Ligand Combined with an Organolanthanide Fragment. J.

Am. Chem. Soc. 2017, 139, 10633−10636.

(

31) Thompson, L. K.; Mandal, S. K.; Tandon, S. S.; Bridson, J. N.;

(50) Wang, D.; Moutet, J.; Tricoire, M.; Cordier, M.; Nocton, G.

Reactive Heterobimetallic Complex Combining Divalent Ytterbium

and Dimethyl Nickel Fragments. Inorganics 2019, 7 , 58 −70.

(51) Day, B. M.; Guo, F.-S.; Layfield, R. A. Cyclopentadienyl

Ligands in Lanthanide Single-Molecule Magnets: One Ring to Rule

Them All? Acc. Chem. Res. 2018, 51 , 1880 −1889.

Park, M. K. Magnetostructural Correlations in Bis(μ -phenoxide)-

Bridged Macrocyclic Dinuclear Copper(II) Complexes. Influence of

Electron-Withdrawing Substituents on Exchange Coupling. Inorg.

Chem. 1996, 35, 3117−3125.

(32) Liu, X.-W.; Wu, Z.; Chen, J.-T.; Li, L.; Chen, P.; Sun, W.-B.

Regulating the single-molecule magnetic properties of phenol oxygen-

bridged binuclear lanthanide complexes through the electronic and

spatial effect of the substituents. Inorg. Chem. Front. 2020, 7, 1229−

(52) Zucchi, G.; Le Goff, X. F. Magnetic properties of structurally

characterized binuclear lanthanide complexes bridged by 2,2 ′-

bipyrimidine. Polyhedron 2013, 52, 1262−1267.

(

238.

(53) Visinescu, D.; Fabelo, O.; Ruiz-Perez, C.; Lloret, F.; Julve, M.

−

33) Wang, W.-M.; Qiao, W.-Z.; Zhang, H.-X.; Wang, S.-Y.; Nie, Y.-

[Fe(phen)(CN) ] : a suitable metalloligand unit to build 3d-4f

Y.; Chen, H.-M.; Liu, Z.; Gao, H.-L.; Cui, J.-Z.; Zhao, B. Structures

and magnetic properties of several phenoxo-O bridged dinuclear

lanthanide complexes: Dy derivatives displaying substituent depend-

ent magnetic relaxation behavior. Dalton Trans. 2016, 45, 8182−8191.

heterobimetallic complexes with mixed bpym-cyano bridges (phen =

1,10 phenanthroline, bpym = 2,2 ′-bipyrimidine). CrystEngComm

2010, 12, 2454−2465.

(54) Geske, D. H.; Padmanabhan, G. R. An Electron Spin

Resonance Study of the Anion Radicals of 9,10-Diazaphenanthrene,

(

34) Habib, F.; Brunet, G.; Vieru, V.; Korobkov, I.; Chibotaru, L. F.;

Murugesu, M. Significant Enhancement of Energy Barriers in

Dinuclear Dysprosium Single-Molecule Magnets Through Electron-

Withdrawing Effects. J. Am. Chem. Soc. 2013, 135, 13242−13245.

2,2 ′-Bipyrimidine, and Δ ′-Biisobenzimidazolylidene. J. Am. Chem.

Soc. 1965, 87, 1651−1655.

(55) Kaim, W.; Kohlmann, S. Four Bridging Bis Chelate Ligands

with Very Low Lying π * Orbitals. MO Perturbation Calculations,

Electrochemistry, and Spectroscopy of Mononuclear and Binuclear

Group 6 Metal Tetracarbonyl Complexes. Inorg. Chem. 1987, 26, 68−

77.

(35) Inagaki, A.; Yatsuda, S.; Edure, S.; Suzuki, A.; Takahashi, T.;

Akita, M. Synthesis of Pd Complexes Combined with Photosensitiz-

ing of a Ruthenium(II) Polypyridyl Moiety through a Series of

Substituted Bipyrimidine Bridges. Substituent Effect of the Bridging

Ligand on the Photocatalytic Dimerization of α -Methylstyrene. Inorg.

Chem. 2007, 46, 2432−2445.

(56) Janzen, E. G. Substituent Effects on Electron Spin Resonance

Spectra and Stability of Free Radicals. Acc. Chem. Res. 1969, 2, 279−

288.

(36) Murata, K.; Araki, M.; Inagaki, A.; Akita, M. Syntheses,

photophysical properties, and reactivities of novel bichromophoric Pd

complexes composed of Ru(II)-polypyridyl and naphthyl moieties.

Dalton Trans. 2013, 42, 6989−7001.

(57) Dust, J. M.; Arnold, D. R. Substituent Effects on Benzyl Radical

ESR Hyperfine Coupling Constants. The σ • Scale Based upon Spin

Delocalization. J. Am. Chem. Soc. 1983, 105, 1221−1227.

37) Song, B.; Wu, G.; Wang, Z.; Zhang, X.; Smet, M.; Dehaen, W.

(58) Jeon, I. R.; Park, J. G.; Xiao, D. J.; Harris, T. D. An Azophenine

Metal-Ligand Coordination-Induced Self-Assembly of Bolaamphi-

Radical-Bridged Fe Single-Molecule Magnet with Record Magnetic

philes Bearing Bipyrimidine. Langmuir 2009, 25, 13306 −13310.

Exchange Coupling. J. Am. Chem. Soc. 2013, 135, 16845 −16848.

38) Schwab, P. F.; Fleischer, F.; Michl, J. Preparation of 5-

(59) Hua, C.; DeGayner, J. A.; Harris, T. D. Thiosemiquinoid

III

Brominated and 5,5 ′-Dibrominated 2,2 ′-Bipyridines and 2,2 ′-Bipyr-

Radical-Bridged Cr Complexes with Strong Magnetic Exchange

imidines. J. Org. Chem. 2002, 67, 443−449.

Coupling. Inorg. Chem. 2019, 58, 7044−7053.

(39) Nelson, T. D.; Crouch, R. D. Cu, Ni, and Pd Mediated

(60) Kanetomo, T.; Yoshitake, T.; Ishida, T. Strongest Ferromag-

netic Coupling in Designed Gadolinium(III)-Nitroxide Coordination

Compounds. Inorg. Chem. 2016, 55 , 8140 −8146.

Homocoupling Reactions in Biaryl Syntheses: The Ullmann Reaction.

Org. React. 2004, 63, 265−555.

(40) Nasielski, J.; Standaert, A.; Nasielski-Hinkens, R. Efficient

(61) Kanetomo, T.; Ishida, T. Strongest Exchange Coupling in

Gadolinium(III) and Nitroxide Coordination Compounds. Inorg.

Chem. 2014, 53, 10794−10796.

Coupling of 2-Halopyrimidines to 2,2 ′-Bipyrimidines. Synth. Commun.

991, 21, 901−906.

41) Vlad, G.; Horvat

Bipyrimidine. J. Org. Chem. 2002, 67, 6550−6552.

42) Li, H.; Oppenheimer, J.; Smith, M. R.; Maleczka, R. E.

Improved synthesis of electron deficient bipyridines. Tetrahedron Lett.

016, 57, 2231−2232.

43) Chan, K. S.; Tse, A. K.-S. Synthesis of Bis(trifluoromethyl)-

,2 ′-bipyridines by Nickel Catalyzed Homocoupling Reactions. Synth.

Commun. 1993, 23, 1929−1934.

44) Iyoda, M.; Otsuka, H.; Sato, K.; Nisato, N.; Oda, M.

(62) Thorarinsdottir, A. E.; Bjornsson, R.; Harris, T. D. Insensitivity

h, I. T. Improved Synthesis of 2,2 ′-

of Magnetic Coupling to Ligand Substitution in a Series of

Tetraoxolene Radical-Bridged Fe ₂ Complexes. Inorg. Chem. 2020,

59, 4634−4649.

(63) Finn, C. B. P.; Orbach, R.; Wolf, W. P. Spin-Lattice Relaxation

in Cerium Magnesium Nitrate at Liquid Helium Temperatures: A

New Process. Proc. Phys. Soc., London 1961 , 77 , 261 −268.

(64) Chibotaru, L. F.; Iwahara, N. Ising exchange interaction in

lanthanides and actinides. New J. Phys. 2015, 17, 103028.

Homocoupling of Aryl Halides Using Nickel(II) Complex and Zinc

(65) Prsa, K.; Nehrkorn, J.; Corbey, J. F.; Evans, W. J.; Demir, S.;

in the Presence of Et NI. An Efficient Method for the Synthesis of

Long, J. R.; Guidi, T.; Waldmann, O. Perspectives on Neutron

Scattering in Lanthanide-Based Single-Molecule Magnets and a Case

Study of the Tb (μ -N ) System. Magnetochemistry 2016, 2, 45−63.

Biaryls and Bipyridines. Bull. Chem. Soc. Jpn. 1990, 63, 80−87.

(45) Cordovilla, C.; Bartolome, C.; Martínez-Ilarduya, J. M.;

Espinet, P. The Stille Reaction, 38 Years Later. ACS Catal. 2015, 5,

040−3053.

46) Farina, V.; Krishnamurthy, V.; Scott, W. J. The Stille Reaction.

Org. React. 1997, 50, 1−652.

47) Demir, S.; Zadrozny, J. M.; Nippe, M.; Long, J. R. Exchange

Coupling and Magnetic Blocking in Bipyrimidyl Radical-Bridged

Dilanthanide Complexes. J. Am. Chem. Soc. 2012, 134, 18546−18549.

(66) Kofu, M.; Yamamuro, O.; Kajiwara, T.; Yoshimura, Y.; Nakano,

M.; Nakajima, K.; Ohira-Kawamura, S.; Kikuchi, T.; Inamura, Y.

Hyperfine structure of magnetic excitations in a Tb-based single-

molecule magnet studied by high-resolution neutron spectroscopy.

Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 064405−064407.

(67) Baker, M. L.; Tanaka, T.; Murakami, R.; Ohira-Kawamura, S.;

Nakajima, K.; Ishida, T.; Nojiri, H. Relationship Between Torsion and

(

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Anisotropic Exchange Coupling in a Tb -Radical-Based Single-

Article

III

(88) Reta, D.; Chilton, N. F. Uncertainty estimates for magnetic

relaxation times and magnetic relaxation parameters. Phys. Chem.

Chem. Phys. 2019, 21, 23567−23575.

Molecule Magnet. Inorg. Chem. 2015, 54, 5732−5738.

68) Vieru, V.; Iwahara, N.; Ungur, L.; Chibotaru, L. F. Giant

exchange interaction in mixed lanthanides. Sci. Rep. 2016 , 6 , 24046.

69) Zhang, Y.-Q.; Luo, C.-L.; Wang, B.-W.; Gao, S. Understanding

(89) Evans, P.; Reta, D.; Whitehead, G. F. S.; Chilton, N. F.; Mills,

D. P. Bis-Monophospholyl Dysprosium Cation Showing Magnetic

Hysteresis at 48 K. J. Am. Chem. Soc. 2019, 141, 19935−19940.

the Magnetic Anisotropy in a Family of N ₂ Radical-Bridged

‑

Lanthanide Complexes: Density Functional Theory and ab Initio

Calculations. J. Phys. Chem. A 2013, 117, 10873 −10880.

(

90) Aquilante, F.; Autschbach, J.; Carlson, R. K.; Chibotaru, L. F.;

Delcey, M. G.; De Vico, L.; Fdez Galvan

Gagliardi, L.; Garavelli, M.; Giussani, A.; Hoyer, C. E.; Li Manni, G.;

Lischka, H.; Ma, D.; Malmqvist, P. Å.; Muller, T.; Nenov, A.;

, I.; Ferre, N.; Frutos, L. M.;

́ ́

70) Zhang, Y.-Q.; Luo, C.-L.; Zhang, Q. Exchange coupling and

magnetic anisotropy in a family of bipyrimidyl radical-bridged

dilanthanide complexes: Density functional theory and ab initio

calculations. J. Comput. Chem. 2014, 35, 904−909.

71) Langley, S. K.; Wielechowski, D. P.; Vieru, V.; Chilton, N. F.;

Moubaraki, B.; Abrahams, B. F.; Chibotaru, L. F.; Murray, K. S. A

Cr(III) Dy(III) } single-molecule magnet: enhancing the blocking

temperature through 3d magnetic exchange. Angew. Chem., Int. Ed.

013, 52, 12014−12019.

72) Gould, C. A.; Darago, L. E.; Gonzalez, M. I.; Demir, S.; Long, J.

R. A Trinuclear Radical-Bridged Lanthanide Single-Molecule Magnet.

Angew. Chem., Int. Ed. 2017, 56, 10103−10107.

73) Zhang, P.; Perfetti, M.; Kern, M.; Hallmen, P. P.; Ungur, L.;

Lenz, S.; Ringenberg, M. R.; Frey, W.; Stoll, H.; Rauhut, G.; van

Slageren, J. Exchange coupling and single molecule magnetism in

redox-active tetraoxolene-bridged dilanthanide complexes. Chem. Sci.

Olivucci, M.; Pedersen, T. B.; Peng, D.; Plasser, F.; Pritchard, B.;

Reiher, M.; Rivalta, I.; Schapiro, I.; Segarra-Martí, J.; Stenrup, M.;

Truhlar, D. G.; Ungur, L.; Valentini, A.; Vancoillie, S.; Veryazov, V.;

Vysotskiy, V. P.; Weingart, O.; Zapata, F.; Lindh, R. MOLCAS 8:

New capabilities for multiconfigurational quantum chemical calcu-

lations across the periodic table. J. Comput. Chem. 2016 , 37 , 506 −541.

(

{

(91) Chibotaru, L. F.; Ungur, L. Ab initio calculation of anisotropic

magnetic properties of complexes. I. Unique definition of pseudospin

Hamiltonians and their derivation. J. Chem. Phys. 2012, 137, 064112.

(92) Chibotaru, L. F.; Ungur, L.; Aronica, C.; Elmoll, H.; Pilet, G.;

Luneau, D. Structure, Magnetism, and Theoretical Study of a Mixed-

(

III

Valence Co Co Heptanuclear Wheel: Lack of SMM Behavior

Despite Negative Magnetic Anisotropy. J. Am. Chem. Soc. 2008, 130,

12445−12455.

(93) Chibotaru, L. F.; Ungur, L.; Soncini, A. The origin of

nonmagnetic Kramers doublets in the ground state of dysprosium

triangles: evidence for a toroidal magnetic moment. Angew. Chem., Int.

Ed. 2008, 47, 4126−4129.

74) Demir, S.; Nippe, M.; Gonzalez, M. I.; Long, J. R. Exchange

018, 9, 1221−1230.

coupling and magnetic blocking in dilanthanide complexes bridged by

the multi-electron redox active ligand 2,3,5,6-tetra(2-pyridyl)pyrazine.

Chem. Sci. 2014, 5, 4701−4711.

(94) Shoji, M.; Koizumi, K.; Kitagawa, Y.; Kawakami, T.; Yamanaka,

S.; Okumura, M.; Yamaguchi, K. A general algorithm for calculation of

Heisenberg exchange integrals J in multispin systems. Chem. Phys.

Lett. 2006, 432, 343 −347.

75) Guo, F.-S.; Layfield, R. A. Strong direct exchange coupling and

single-molecule magnetism in indigo-bridged lanthanide dimers.

(95) Neese, F. The ORCA program system. Wiley Interdiscip. Rev.:

Chem. Commun. 2017, 53, 3130−3133.

76) Demir, S.; Zadrozny, J. M.; Long, J. R. Large Spin-Relaxation

Comput. Mol. Sci. 2012, 2, 73−78.

Barriers for the Low-Symmetry Organolanthanide Complexes

[

Cp * Ln(BPh )] (Cp *=pentamethylcyclopentadienyl; Ln = Tb,

Dy). Chem. - Eur. J. 2014, 20, 9524−9529.

(77) Evans, W. J.; Davis, B. L.; Champagne, T. M.; Ziller, J. W. C-H

bond activation through steric crowding of normally inert ligands in

the sterically crowded gadolinium and yttrium (C Me ) M com-

5 3

plexes. Proc. Natl. Acad. Sci. U. S. A. 2006 , 103 , 12678 −12683.

(78) Johnson, T. B.; Joyce, W. A. Researches on Pyrimidines.

LXXVI. New Methods of Synthesizing 2-Ketopyrimidines and Their

Sulfur Analogs. J. Am. Chem. Soc. 1915, 37, 2151−2164.

(79) Baram, S. G.; Shkurko, O. P.; Mamaev, V. P. Pyrimidines. IX.

Synthesis of 5-Substituted 2-Fluoropyrimidines. Izv. Sib. Otd. An.

Khim. 1977, 1, 106−109.

(80) wSAINT and APEX 2 Software for CCD Diﬀractometers; Bruker

AXS Inc.: Madison, WI, 2014.

81) Sheldrick, G. M. SADABS, Version 2.03, Bruker Analytical X-

Ray Systems, Inc.: Madison, WI, 2000.

82) Sheldrick, G. M. TWINABS.; University of Go

Gottingen, Germany, 2009.

83) Rigaku Oxford Diﬀraction CrysAlisPro Software system, version

.171.39.7a; Rigaku Corporation: Oxford, U.K., 2015

84) Sheldrick, G. M. SHELXT-Integrated Space-Group and

(

ttingen:

(

Crystal-Structure Determination. Acta Crystallogr., Sect. A: Found.

Adv. 2015, 71, 3−8.

(85) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr.,

Sect. A: Found. Crystallogr. 2008, 64, 112−122.

(86) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.

K.; Puschmann, H. OLEX2: A Complete Structure Solution,

Refinement and Analysis Program. J. Appl. Crystallogr. 2009, 42 (2),

(

39−341.

87) Chilton, N. F.; Anderson, R. P.; Turner, L. D.; Soncini, A.;

Murray, K. S. PHI: a powerful new program for the analysis of

anisotropic monomeric and exchange-coupled polynuclear d- and f-

block complexes. J. Comput. Chem. 2013, 34, 1164−1175.