inorganics
Article
Symmetric Assembly of a Sterically Encumbered
Allyl Complex: Mechanochemical and Solution
Synthesis of the Tris(allyl)beryllate, K[BeA0 3]
(A0 = 1,3-(SiMe3)2C3H3)
Nicholas C. Boyde 1 , Nicholas R. Rightmire 1 , Timothy P. Hanusa 1, * and William W. Brennessel 2
1
2
*
Department of Chemistry, Vanderbilt University, Nashville, TN 37235, USA;
[email protected] (N.C.B.); [email protected] (N.R.R.)
Department of Chemistry, University of Rochester, Rochester, NY 14627, USA;
[email protected]
Correspondence: [email protected]
Academic Editor: Matthias Westerhausen
Received: 23 April 2017; Accepted: 23 May 2017; Published: 27 May 2017
Abstract: The ball milling of beryllium chloride with two equivalents of the potassium salt of
bis(1,3-trimethylsilyl)allyl anion, K[A0 ] (A0 = [1,3-(SiMe3 )2 C3 H3 ]), produces the tris(allyl)beryllate
K[BeA’3 ] (1) rather than the expected neutral BeA’2 . The same product is obtained from reaction
in hexanes; in contrast, although a similar reaction conducted in Et2 O was previously shown to
produce the solvated species BeA’2 (OEt2 ), it can produce 1 if the reaction time is extended (16 h).
The tris(allyl)beryllate is fluxional in solution, and displays the strongly downfield 9 Be NMR shift
expected for a three-coordinate Be center (δ22.8 ppm). A single crystal X-ray structure reveals that the
three allyl ligands are bound to beryllium in an arrangement with approximate C3 symmetry (Be–C
(avg) = 1.805(10) Å), with the potassium cation engaging in cation–π interactions with the double
bonds of the allyl ligands. Similar structures have previously been found in complexes of zinc and
tin, i.e., M[M0 A0 3 L] (M0 = Zn, M = Li, Na, K; M0 = Sn, M = K; L = thf). Density functional theory (DFT)
calculations indicate that the observed C3 -symmetric framework of the isolated anion ([BeA0 3 ]− ) is
20 kJ·mol−1 higher in energy than a C1 arrangement; the K+ counterion evidently plays a critical role
in templating the final conformation.
Keywords: allyl ligands; beryllium; coordination modes; mechanochemistry; X-ray diffraction;
density functional theory calculations
1. Introduction
The physical and chemical properties of first-row elements often differ appreciably from their
second-row and heavier counterparts; for the group 2 metals, the outlier (”black sheep” [1]) designation
belongs to beryllium. To a considerably greater extent than its heavier congeners, even magnesium,
the small size of the Be2+ cation (0.27 Å for CN = 4; cf. 0.57 Å for Mg2+ ) [2] and its corresponding high
charge/size ratio ensures its bonds will be strongly polarized and possess substantial covalent character.
Not surprisingly, beryllium compounds with the same ligand sets commonly have different structures
from those of the other, more electropositive alkaline earth (Ae) metals. The bis(trimethylsilyl)amides
of Mg–Ba, for example, have a common dimeric bridged structure, [Ae(N(SiMe3 )(µ-N(SiMe3 )2 ]2 [3],
whereas that of beryllium is a two-coordinate monomer [4]. Similarly, the bis(cyclopentadienyl) complex
Cp2 Be has an η1 ,η5 -Cp structure [5] that is unlike that of the heavier metallocenes [6]. Investigation of
these differences, and indeed research with all beryllium compounds, has traditionally been limited
Inorganics 2017, 5, 36; doi:10.3390/inorganics5020036
www.mdpi.com/journal/inorganics
Inorganics 2017, 5, 36
Inorganics
Inorganics 2017,
2017, 5,
5, 36
36
2 of 11
22 of
of 11
11
because
concernsits
about
toxicity [7],
but serving
that
has as
not
prevented
its compounds
serving
as useful
has
prevented
compounds
from
benchmarks
of
steric
and
has not
not of
prevented
its
compounds
from
serving
as useful
useful
benchmarks
of the
thefrom
steric
and electronic
electronic
benchmarks
of
the
steric
and
electronic
consequences
of
crowded
metal
environments
[8–10].
consequences
consequences of
of crowded
crowded metal
metal environments
environments [8–10].
[8–10].
consequences
is theis
stabilitystability
of η1 - vs.of
η3 -bonded
ligands allyl
in compounds
of
One
of
these
consequences
the
ηη11-- vs.
ηη33-bonded
ligands
One of
ofthese
these
consequences
isrelative
the relative
relative
stability
of
vs. allyl
-bonded
allyl
ligands in
in
0 ]− (A0 = [1,3-(SiMe ) C
−− (A′
highly
electropositive
metals.
We
found
some
time
ago
that
the
bulky
allyl
[A
H
])
compounds
of
highly
electropositive
metals.
We
found
some
time
ago
that
the
bulky
allyl
[A′]
3 2 3(A′3=
compounds of highly electropositive metals. We found some time ago that the bulky allyl [A′]
=
1 -bonded A’ ligands1 in the solid
·
can
be
used
to
form
the
ether
adduct
BeA’
OEt
,
which
displays
η
[1,3-(SiMe
2 ether
2 adduct
[1,3-(SiMe33))22C
C33H
H33])
]) can
can be
be used
used to
to form
form the
the
ether
adduct BeA’
BeA’22·OEt
·OEt22,, which
which displays
displays ηη1-bonded
-bonded A’
A’
state
[11].
The
compound
is fluxional
in solution,
and exhibits
symmetric,
“π-type”symmetric,
bonding in“πits
ligands
in
state
The
is
in
and
ligands
in the
the solid
solid
state [11].
[11].
The compound
compound
is fluxional
fluxional
in solution,
solution,
and exhibits
exhibits
symmetric,
“πNMR
spectra
(e.g.,
only
one
peak
is
observed
for
the
SiMe
groups).
Density
functional
theory
(DFT)
type”
bonding
in
its
NMR
spectra
(e.g.,
only
one
peak
is
observed
for
the
SiMe
3
groups).
Density
type” bonding in its NMR spectra (e.g., only one peak is3 observed for the SiMe3 groups). Density
calculations
suggested
that
a base-free Be(C
SiH3 ) complex
slightly
more
functional
(DFT)
calculations
suggested
aa H,
base-free
Be(C
33E
H,
33)) complex
3 H3 E2 )that
2 (E =
functional theory
theory
(DFT)
calculations
suggested
that
base-free
Be(C33H
Hwould
E22))22 (E
(Ebe== more
H, SiH
SiH
complex
stable
with
delocalized,
π-type
allyls
than
with
monodentate,
sigma-bonded
ligands
(Scheme
1).
If so,
would
be
more
slightly
more
stable
with
delocalized,
π-type
allyls
than
with
monodentate,
sigmawould be more slightly more stable with delocalized, π-type allyls than with monodentate, sigmaberyllium
allyls
would
join
those
of
magnesium,
in
which
monodentate
allyl
ligands
are
uniformly
bonded
bonded ligands
ligands (Scheme
(Scheme 1).
1). If
If so,
so, beryllium
beryllium allyls
allyls would
would join
join those
those of
of magnesium,
magnesium, in
in which
which
found
in complexes
that areare
ether-solvated
[12], but
that in the that
absence
of ethers, cation–π
monodentate
allyl
uniformly
in
are
[12],
but
monodentate
allyl ligands
ligands
are
uniformly found
found
in complexes
complexes
that
are ether-solvated
ether-solvated
[12],interactions
but that
that in
in
with
the
metal
can
create
“slipped-π”
bonding
[13].
the
absence
of
ethers,
cation–π
interactions
with
the
metal
can
create
“slipped-π”
bonding
[13].
the absence of ethers, cation–π interactions with the metal can create “slipped-π” bonding [13].
Scheme
Scheme 1.
1. Optimized
Optimized geometries
geometries of
of Be(1,3-(SiH
Be(1,3-(SiH33))22C
C33H
H33))22.. At
At the
the B3PW91/aug-cc-pVDZ
B3PW91/aug-cc-pVDZ level,
level, the
the ππScheme 1. Optimized geometries−1of Be(1,3-(SiH3 )2 C3 H3 )2 . At the B3PW91/aug-cc-pVDZ level, the
−1
lower
in
energy
(∆G°)
than
the
σ-bound
geometry
(b)
[11].
bound
structure
(a)
is
4.0
kcal·mol
in energy (∆G°) than
the σ-bound geometry (b) [11].
bound structure (a) is 4.0 kcal·mol lower
−
1
◦
π-bound structure (a) is 4.0 kcal·mol lower in energy (∆G ) than the σ-bound geometry (b) [11].
The
The coordinated
coordinated ether
ether in
in BeA’
BeA’22·OEt
·OEt22 proved
proved impossible
impossible to
to remove
remove without
without destroying
destroying the
the
The
coordinated
ether
in
BeA’
OEt
proved
impossible
to
remove
without
destroying
the
·
2
2
complex
[11],
and
thus
we
investigated
mechanochemical
methods
of
synthesis
as
a
means
to
bypass
complex [11], and thus we investigated mechanochemical methods of synthesis as a means to bypass
complex
and thus
we investigated
mechanochemical
methods neutral
of synthesis
as a was
means
bypass
the
ethereal
solvents
[14].
below,
complex
the use
use of
of[11],
ethereal
solvents
[14]. As
As detailed
detailed
below, an
an unsolvated
unsolvated
neutral
complex
was not
nottoisolated
isolated
the
use
of
ethereal
solvents
[14].
As
detailed
below,
an
unsolvated
neutral
complex
was
not
isolated
via
via
via this
this route,
route, and
and the
the beryllate
beryllate anion
anion that
that was
was produced
produced instead
instead has
has structural
structural parallels
parallels with
with
this
route, and
the beryllate
anion thatof
instead
structural
with
previously
previously
described
-ate
Zn
[15]
[16].
In
of
species,
metal
previously
described
-ate complexes
complexes
ofwas
Znproduced
[15] and
and Sn
Sn
[16]. has
In all
all
of these
theseparallels
species, the
the alkali
alkali
metal
described
-ate
complexes
of
Zn
[15]
and
Sn
[16].
In
all
of
these
species,
the
alkali
metal
counterion,
usually
+
+
+
+
+
+
counterion,
counterion, usually
usually K
K but
but sometimes
sometimes Na
Na and
and Li
Li ,, appears
appears to
to play
play aa critical
critical role
role in
in the
the assembly
assembly of
of
+
K
sometimes
Na+ and Li+ , appears to play a critical role in the assembly of the symmetric complexes.
the
symmetric
complexes.
thebut
symmetric
complexes.
2. Results
Results and Discussion
Discussion
2.
2. Results and
and Discussion
2.1. Solid-State Synthesis
2.1.
2.1. Solid-State
Solid-State Synthesis
Synthesis
The reaction of BeCl2 and K[A’] was conducted mechanochemically with a planetary ball mill,
The
of
K[A’] was
mechanochemically
with
planetary
ball
The reaction
reaction
of BeCl
BeCl22 and
and
was conducted
conducted
mechanochemically
with aa ratios
planetary
ball mill,
mill,
followed
by an extraction
withK[A’]
hexanes.
Initial investigations
used 2:1 molar
of BeCl
2 and
followed
by
an
extraction
with
hexanes.
Initial
investigations
used
2:1
molar
ratios
of
BeCl
22 and
K[A’],
followed
by
an
extraction
with
hexanes.
Initial
investigations
used
2:1
molar
ratios
of
BeCl
and
K[A’],
K[A’], based on the assumption that the product formed would be BeA’2 (Equation (1)). Although the
based
the
assumption
that
the
product
formed
would
based on
onare
theoff-white
assumption
thatand
thewhite
product
formed
would be
be BeA’
BeA’22 (Equation
(Equation (1)).
(1)). Although
Although the
the
reagents
(K[A’])
(BeCl
2 ), the ground reaction mixture (15 min/600 rpm) is
reagents
are
off-white
(K[A’])
and
white
(BeCl
22),
the
ground
reaction
mixture
(15
min/600
rpm)
is
reagents
are
off-white
(K[A’])
and
white
(BeCl
),
the
ground
reaction
mixture
(15
min/600
rpm)
orange. Extraction with hexanes, followed by filtration, yielded an orange filtrate and ultimately is
a
orange.
Extraction
with
hexanes,
followed
by
filtration,
yielded
an
orange
filtrate
and
ultimately
aa
orange.
Extraction
with
hexanes,
followed
by
filtration,
yielded
an
orange
filtrate
and
ultimately
dark orange, highly air-sensitive solid (1) on drying.
dark
dark orange,
orange, highly
highly air-sensitive
air-sensitive solid
solid (1)
(1) on
on drying.
drying.
(1)
(1)
(1)
A
single
crystal
analysis
(described
below)
revealed
that
1
is
the
potassium
tris(allyl)beryllate,
A single crystal analysis (described below) revealed that 1 is the potassium tris(allyl)beryllate,
A 3single
1 is the potassium
tris(allyl)beryllate,
forms
in
of
that
2:1
of
used
optimum
K[BeA’
3].
]. This
Thiscrystal
forms analysis
in spite
spite (described
of the
the fact
fact below)
that the
therevealed
2:1 ratio
ratiothat
of reagents
reagents
used is
is not
not
optimum for
for its
its
K[BeA’
K[BeA’
].
This
forms
in
spite
of
the
fact
that
the
2:1
ratio
of
reagents
used
is
not
optimum
forand
its
3
production.
As
production.
As detailed
detailed below,
below, conducting
conducting the
the reaction
reaction with
with 1:1
1:1 and
and 3:1
3:1 molar
molar ratios
ratios of
of K[A’]
K[A’]
and
production.
As detailed
below,
the reaction
with 1:1It
3:1 molarthat
ratios
ofexcess
K[A’] and
BeClis
BeCl
11 as
sole
hexane-extractable
product.
is
the
halide
BeCl22 still
still yields
yields
as the
the
soleconducting
hexane-extractable
product.
Itand
is possible
possible
that
the
excess
halide
is2
still
yields
1
as
the
sole
hexane-extractable
product.
It
is
possible
that
the
excess
halide
is
captured
2−
2−
2−
2−
captured
captured in
in the
the form
form of
of polyhalide
polyhalide anions
anions such
such as
as [BeCl
[BeCl44]] or
or [Be
[Be22Cl
Cl66]] [17],
[17], although
although these
these have
have not
not
been
been definitively
definitively identified.
identified.
22 K[A’]
K[A’] ++ BeCl
BeCl22
BeA’
BeA’22 ++ 22 KCl
KCl (expected)
(expected)
Inorganics 2017, 5, 36
3 of 11
Inorganics 2017, 5, 36
3 of 11
inInorganics
the form
of 5,polyhalide
anions such as [BeCl4 ]2− or [Be2 Cl6 ]2− [17], although these have not been
2017,
36
3 of 11
definitively
identified.
2.2. Synthesis
Inorganics
2017, 5,in36Solution
3 of 11
2.2. Synthesis in Solution
The reaction
of K[A’] and BeCl2 was also examined in solution, using diethyl ether and hexanes.
2.2. Synthesis
in Solution
2.2.
Synthesis
in
Solution
TheseThe
results
are summarized
in Table
1.also
Previous
reactions
with diethyl
ether involved
stirring
for
reaction
of K[A’] and BeCl
2 was
examined
in solution,
using diethyl
ether and
hexanes.
reaction
of K[A’] and
BeCl
also
examined
inasolution,
using
diethyl
ether
and hexanes.
2 was 1.
2 h The
atThe
room
temperature,
which
formed
BeA’
2
·OEt
2
from
2:1
reaction
(#5);
Schlenk
equilibrium
was
These
results
are
summarized
in
Table
Previous
reactions
with
diethyl
ether
involved
stirring
for
reaction of K[A’] and BeCl2 was also examined in solution, using diethyl ether and hexanes.
These
results
are
summarized
in
Table
1.
Previous
reactions
with
diethyl
ether
involved
stirring
for
observed
in temperature,
a are
1:1 summarized
mixturewhich
that in
was
allowed
to2·OEt
react
for one
hour
(#4).
When
theinvolved
2:1 equilibrium
reaction
in Et
2O
2 h at room
formed
2 reactions
from
a 2:1
reaction
(#5);ether
Schlenk
was
These
results
Table
1.BeA’
Previous
with
diethyl
stirring
for
nics 2017, 5, 36
3 of Schlenk
11
2 is
h at room temperature,
which
BeA’
·OEt
a 2:1
reaction
(#5);
equilibrium
wasin
2 to
2 from
for
16
h,formed
however,
the
is
observed
(#6)Schlenk
exclusively.
Reaction
observed
intotemperature,
aproceed
1:1 mixture
that
was
allowed
react
for of
one
hour
(#4). When
the 2:1 equilibrium
reaction
in Et
2O
2 hallowed
at room
which
formed
BeA’
2formation
·OEt
2 from
a 12:1
reaction
(#5);
was
observed
in
a 1:1 mixture
that wasreactions,
allowed toinreact
for
one
hour
(#4). When
the 2:1 reaction
in Et2 O
hexanes
mimics
the
solid-state
that
1
is
the
exclusively
detected
organoberyllium
is
allowed
to
proceed
for
16
h,
however,
the
formation
of
1
is
observed
(#6)
exclusively.
Reaction
in a 1:1 mixture that was allowed to react for one hour (#4). When the 2:1 reaction in Et2in
O
Synthesis in Solutionisobserved
allowed
to
proceed
for 16 h,
however,
theLonger
formation
of 1 isand
observed
(#6)
exclusively.
Reaction
in
product
from
a 1:1the
reaction
1reactions,
h (#7).
reactions
a higher
ratio
of K[A’]
to
BeCl2 (e.g.,
hexanes
mimics
solid-state
that
1 isofthe
detected
organoberyllium
is
allowed
to proceed
for 16after
h, however,
theinformation
1 isexclusively
observed
(#6)
exclusively.
Reaction
in
hexanes
mimics
thealso
solid-state
reactions,
in that
1 is the
exclusively
detected
organoberyllium product
The reaction of K[A’]
2 was
examined
solution,
using
diethyl
ether
and
hexanes.
3:1)and
doBeCl
not
change
this
outcome.
product
from
a 1:1
reaction
afterin1reactions,
h (#7). Longer
reactions
and
a higher
ratio
of K[A’]
to BeCl2 (e.g.,
hexanes
mimics
the
solid-state
in that
1 is the
exclusively
detected
organoberyllium
from
a
1:1
reaction
after
1
h
(#7).
Longer
reactions
and
a
higher
ratio
of
K[A’]
to BeCl2 (e.g., 3:1) do not
e results are summarized
in
Table
1.
Previous
reactions
with
diethyl
ether
involved
stirring
for
3:1) do not
change
outcome.
product
from
a 1:1 this
reaction
after 1 h (#7). Longer reactions and a higher ratio of K[A’] to BeCl2 (e.g.,
change
this
outcome.
t room temperature,
which
formed
2·OEt2of
from
2:1 reaction
(#5); Schlenk
was as molar ratios.
Table
1.BeA’
Summary
K[A’]a and
BeCl2 reactions;
amountsequilibrium
of reagents given
3:1)
do not
change
this
outcome.
rved in a 1:1 mixture that was
allowed
to react
one
hour
When amounts
the 2:1 reaction
in given
Et2O as molar ratios.
Table
1. Summary
offor
K[A’]
and
BeCl(#4).
2 reactions;
of reagents
a
K[A’]:BeCl
Medium
Time (#6)amounts
Organoberyllium
Product(s)
Yield (%) b
Table
1. Summary
of
K[A’]of
and
BeCl
of reagents
given
ratios.
2 reactions;
owed to proceed for 16No.
h, however,
the2 formation
1 is
observed
exclusively.
Reaction
inasasmolar
Table 1. Summary
of K[A’]
and
BeCl
2 reactions;
amounts of reagents
given
molar ratios.
a
No.
K[A’]:BeCl
2
Medium
Time
Organoberyllium
Yield
1
1:1
15
min detected
K[BeA’3]Product(s)
97(%) b
nes mimics the solid-state
reactions,
in that
1 is the
exclusively
organoberyllium
a
b
No.
K[A’]:BeCl
Medium
Time
Organoberyllium
Product(s)
Yield
(%)
a
b
2 2 Medium
Time
Yield
12
1:1
15
min ratio ofOrganoberyllium
]Product(s)
97
2:1 Longer
21(%)
uct from a 1:1 reactionNo.
after
1K[A’]:BeCl
h (#7).
reactions and a higher
K[A’] toK[BeA’
BeCl23(e.g.,
15 min
K[BeA’3K[BeA’
]
97
1:1
15 min
33]
97
2:11:1
21
do not change this outcome.
25
312 1
3:1
o.
1
2
15 min
K[BeA’
] BeA’3]2 + BeCl
21
2:1
151min
21
3K[BeA’
25c,d
342 2
3:1
h
2A’BeCl
n/a
1:12:1
Et2O
Table 1. Summary of34K[A’]
and
BeCl
2
reactions;
amounts
of
reagents
given
as
molar
ratios.
15 min
K[BeA’
] BeA’
25
c
3K[BeA’
1512min
3]2 2+ BeCl
25c,d
3:1
h
2A’BeCl
n/a
1:1
Et2O
BeA’
2·OEt
77
5 3
2:13:1
c,d
4
1:1
Et
O
1
h
n/a
2A’BeCl
BeA’
+
BeCl
c,d
2
2 BeA’
12 hh Product(s) 2A’BeCl
n/a
1:1 Time
a
BeA’
2·OEt
77
16
K[BeA’
3b]2 2+ BeCl
98c
645
2:1
Et22OOrganoberyllium
K[A’]:BeCl2
Medium
Yield
(%)
5
2:1
Et2 O
2h
BeA’2 ·OEt2
77 c
21 hh 3]
2·OEt
77
K[BeA’
3] 2
98
675 6
2:12:115 min hexanes
Et2O
24c
1:1
1:1
K[BeA’
Et
16
h16
K[BeA’BeA’
98
2O
3 ] 97
16
3]
98
67= ball
2:11:115
Et2O
7 milling
1K[BeA’
hfor
K[BeA’3K[BeA’
] 21 been
a
24
1:1
hexanes
1 hhmechanical
at min
600 rpm.
The symbol
milling
has
proposed 24
in ref. [14];
2:1
3]
3
3:1
4
1:1
5
2:1
6
2:1
7
1:1
b
c Ref. [11]; d Products
9Be NMR,
a 7= ball
3] ref.
24 b
1:1limiting
hexanes
1 hmechanical
K[BeA’
Unrecrystallized;
reagent
taken
into
account;
observed
with
at 600
600rpm.
rpm.
symbol
for
milling
has were
been
proposed
in ref.
[14];
= ballmilling
milling
at
TheThe
symbol
for mechanical
milling has
been proposed
in
[14]; b Unrecrystallized;
a
15 min
K[BeA’3]
25
c Ref. [11]; d Products were observed with 9 Be NMR, and were not isolated.
limiting
reagent
taken into
account;taken
c Ref. [11]; d Products were observed with 9Be NMR,b
aUnrecrystallized;
and
were
not
isolated.
limiting
reagent
into
account;
=
ball
milling
at
600
rpm.
The
symbol
for
mechanical
milling
has
1h
2A’BeCl BeA’2 + BeCl
n/a c,dbeen proposed in ref. [14];
Et2O
c
d
and
were
not
isolated.
Unrecrystallized;
limiting
account;
Ref. [11]; Products
observed with 9Be NMR,
c
2 h reagent taken intoBeA’
2·OEt2
77 were
Et2O
2.3.
NMR
Spectroscopy
2.3.
NMR
Spectroscopy
and were not isolated.
16 h
K[BeA’3]
98
Et2O
2.3.The
NMR
Spectroscopy
1H
NMR
The1 H
NMRspectrum
spectrumofof11displays
displaysresonances
resonancestypical
typicalofofaa“π-bound”
“π-bound”A’A’ligand,
ligand,with
witha atriplet
triplet
24
hexanes
1h
K[BeA’3]
2.3.
NMR
Spectroscopy
representing
,
a
broad
resonance
(ν
=
39
Hz,
presumably
an
unresolved
doublet)
representing
(β)
,
a
broad
resonance
(ν
1/2
=
39
Hz,
presumably
an
unresolved
doublet)
representing
representing
H
The 1HH
NMR
spectrum
of
1
displays
resonances
typical
of
a
“π-bound”
A’
ligand,
with
a
triplet
1/2
(β)
= ball milling at 600 rpm. The1 symbol for mechanical milling has been proposed in ref. [14]; b
the
HH
H(γ)
,of
and
asinglet
singlet
the
two
equivalent
trimethylsilyl
groups
(Figure
1).
theequivalent
equivalent
(α)
and
H
, and
two
equivalent
trimethylsilyl
(Figure
The
(β)
, aspectrum
broad
resonance
(ν1/2 for
= for
39the
Hz,
presumably
unresolved
doublet)
representing
representing
H
The H NMR
1 adisplays
resonances
typical
of aan
“π-bound”
A’groups
ligand,
with
a1).
triplet
(α)
(γ)
crystallized; limiting reagent taken into account; c Ref. [11]; d Products were observed with 9Be NMR,
The
of
aHsymmetric
spectrum
even
when
σ-bound
ligands
expected
isrepresenting
consistent
appearance
ofHsuch
abroad
symmetric
when
σ-bound
ligands
are are
expected
is consistent
theappearance
equivalent
H
(γ)resonance
, and spectrum
a singlet
two
equivalent
trimethylsilyl
groups
(Figure
1).with
The
(β)(α)
,such
aand
(ν1/2 for
=even
39the
Hz,
presumably
an
unresolved
doublet)
representing
were not isolated.
with
a high
degree
of
fluxionality,
as
was
also
observed
in the
σ-bound
complex
BeA’
O)1).[11].
aappearance
high
degree
of(α)fluxionality,
asspectrum
alsofor
observed
in
the
σ-bound
complex
BeA’
2·(Et
O) 2[11].
The
of such
a symmetric
even
when
σ-bound
ligands
are
expected
is
consistent
with
the
equivalent
H
and
H(γ), and
awas
singlet
the
two equivalent
trimethylsilyl
groups
(Figure
2 ·2(Et
The
triplet
resonance
theallyl
allylligands,
ligands,
atδ6.97,
δ6.97,
shifted
downfield
from
ofof
BeA’
·2consistent
(Et
(δ6.53);
triplet
resonance
the
isis
shifted
downfield
from
that
BeA’
·(Et
2O)
(δ6.53);
high
degree
of of
fluxionality,
asspectrum
was at
also
observed
in
the
σ-bound
complex
BeA’
22·(Et
2O)
[11].
The
of such
aof
symmetric
even
when
σ-bound
ligands
arethat
expected
is
with
2 O)
NMR Spectroscopy aappearance
the
resonance
atat
δ3.19
isisslightly
upfield
(cf.
BeA’
·downfield
(Et
O)).
The
NMR
chemical
shifts
for
1
are
resonance
δ3.19
slightly
upfield
(cf.δ3.33
δ3.33
in
BeA’
2·(Et
2
O)).
The
NMR
chemical
shifts
for
1
are
triplet
resonance
of
the
allyl
ligands,
δ6.97,
isin
shifted
from
that
of
BeA’
2
·(Et
2
O)
(δ6.53);
athe
high
degree
of
fluxionality,
as
was at
also
observed
in 2the
σ-bound
complex
BeA’
2
·(Et
2
O)
[11].
The
2
The 1H NMR spectrum
of
1
displays
resonances
typical
of
a
“π-bound”
A’
ligand,
with
a
triplet
inin
line
with
those
observed
for
other
M[M’A’
]
complexes
(Table
2).
In
particular,
the
NMR
shifts
of
3]iscomplexes
(Table
InNMR
particular,
the2·(Et
NMR
shifts
of
line
with those
observed
for
other
M[M’A’
the
resonance
at δ3.19
slightly
upfield
(cf. 3δ3.33
in BeA’downfield
2·(Et
2O)).2).
The
shifts
1 are
triplet
resonance
of
theisallyl
ligands,
at δ6.97,
shifted
from
thatchemical
of BeA’
2O)for
(δ6.53);
resonance
(νat1/2
=observed
39sensitive
Hz,
presumably
an
unresolved
doublet)
representing
esenting H(β), a broad
the
ligands
are
sensitive
both
totothe
identity
of
central
divalent
metal
and
totothat
alkali
allyl
ligands
are
both
the
identity
the
divalent
metal
and
thatofof
alkali
3] complexes
(Table
2).
InNMR
particular,
the
NMR
shifts
of
in
line
with
those
for
other
M[M’A’
theallyl
resonance
δ3.19
is
slightly
upfield
(cf.
δ3.33
inofthe
BeA’
2central
·(Et
2O)).
The
chemical
shifts
for
1 are
quivalent H(α) andmetal
H
(γ), and
a
singlet
for
the
two
equivalent
trimethylsilyl
groups
(Figure
1).
The
counterion,
evidence
that
the
compounds
exist
as
contact
ion
pairs
in
solution.
Compound
1
metal
counterion,
evidence
that
the
compounds
exist
as
contact
ion
pairs
in
solution.
Compound
1
the
allyl
ligands
are
sensitive
both
to
the
identity
of
the
central
divalent
metal
and
to
that
of
alkali
in line with those observed for other M[M’A’3] complexes (Table 2). In particular, the NMR shifts of
+
arance of such a symmetric
spectrum
even
when
σ-bound
ligands
are
expected
is
consistent
with
+
and
the
greatest
similarities,
which
may
reflect
their
having
the
same
(K(K)1)
andK[ZnA’
K[ZnA’
] share
the
greatest
similarities,
which
may
reflect
their
having
the
same
counterion
metal
counterion,
evidence
that
thetocompounds
exist
as
contact
ion
pairsmetal
in
solution.
Compound
the
allyl
ligands
are
sensitive
both
the identity
of
the
central
divalent
andcounterion
to
that of alkali
3 ]3share
h degree of fluxionality,
as was
alsoofevidence
observed
in
the
complex
BeA’
2·(Et
2O) [18].
The
and
central
metals
similar
electronegativity
(χ
(1.57);
Zn
(1.65))
and
central
metals
of
similar
electronegativity
(χBeBe
(1.57);
Zn
(1.65))
[18].
K[ZnA’
3] share
the
greatest
similarities,
which
may
their
having
the
same counterion
(K+1)
metal
counterion,
that
theσ-bound
compounds
exist
asreflect
contact
ion[11].
pairs
in
solution.
Compound
9 Be
et resonance of the and
allylJohn
ligands,
at
δ6.97,
is
shifted
downfield
from
that
of
BeA’
2·(Et2O)
(δ6.53);
9Be
and
co-workers
have
demonstrated
that
NMR
chemical
shift
values
can
be
diagnostic
for
John
and
co-workers
have
demonstrated
that
NMR
chemical
shift
values
can
be
diagnostic
central metals
similar
electronegativity
(χ may
(1.57);
Zn (1.65))
[18]. the same counterion (K+)
K[ZnA’
3] shareof
the
greatest
similarities, which
reflect
their having
esonance at δ3.19 is
slightly
upfield
(cf.
δ3.33
in
BeA’
2
·(Et
2
O)).
The
NMR
chemical
shifts
for
1complexes
arevalues
9
coordination
numbers
solution
[19].
Typically,
organoberyllium
with
low formal
coordination
for coordination
numbers
in electronegativity
solution
[19]. Typically,
organoberyllium
with
formal
John and
co-workers
have
demonstrated
that
NMR
chemical
shift
can
below
diagnostic
and
central
metals
ofinsimilar
(χ
Be Be
(1.57);
Zncomplexes
(1.65)) [18].
3
]
complexes
(Table
2).
In
particular,
the
NMR
shifts
of
ne with those observed
for
other
M[M’A’
9
numbers,
such
BeMe
·
Et
O
(coordination
number
3,
δ20.8
ppm
in
Et
O),
are
observed
well
downfield
coordination
numbers,
such
as
BeMe
2
·Et
2
O
(coordination
number
3,
δ20.8
ppm
in
Et
2
O),
are
observed
for
coordination
numbers
in
solution
[19].
Typically,
organoberyllium
complexes
with
low
formal
John
andas
co-workers
have
demonstrated
that
Be
NMR
chemical
shift
values
can
be
diagnostic
2
2
2
9
llyl ligands are sensitive
both
to
the
identity
of
the
central
divalent
metal
and
to
that
of
alkali
9
ofwell
0
ppm.
BeA’
·
(Et
O)
has
a
Be
chemical
shift
of
δ18.2
ppm,
which
is
consistent
with
a
three-coordinate
downfield
of2numbers
0 ppm.
BeA’
2BeMe
·(Et2O)
has
Be chemical
shift of δ18.2
ppm,
which
consistent
with
coordination
such
2·Et
2Oa(coordination
number
3, δ20.8
ppm
in Etis
2with
O),
are
observed
for
coordination
inassolution
[19].
Typically,
organoberyllium
complexes
low
formal
2numbers,
l counterion, evidence
that
the
compounds
as
contact
ion
pairs
solution.
Compound
1which
9Be
geometry
in
solution
It exist
should
be
noted,
that
the
correlation
between
coordination
number
a three-coordinate
geometry
solution
It in
should
be
noted,
however,
that
theare
correlation
well
downfield
of 0[11].
ppm.
BeA’
2BeMe
·(Et
2O)
has
a[11].
chemical
shift
of
δ18.2
ppm,
consistent
with
coordination
numbers,
such
asin
2·Et
2Ohowever,
(coordination
number
3, δ20.8
ppm
in
Etis
2O),
observed
+)
9 Be chemical
K[ZnA’3] share theand
greatest
similarities,
which
may
reflect
their
having
the
same
counterion
(K
9Becan
9
shift
is
not
exact,
and
be
strongly
influenced
by
the
electronic
properties
of
theby
between
coordination
number
and
chemical
shift
is
not
exact,
and
can
be
strongly
influenced
a
three-coordinate
geometry
in
solution
[11].
It
should
be
noted,
however,
that
the
correlation
well downfield of 0 ppm. BeA’2·(Et2O) has a Be chemical shift of δ18.2 ppm, which is consistent with
central metals of similar
electronegativity
(χ
Be
(1.57);
Zn
(1.65))
[18].
9
ligands.
The
2-coordinate
complex
beryllium
bis(N,N´-bis(2,6-diisopropylphenyl)-1,3,2-diazaborolyl),
for
the
electronic
properties
of the
ligands.
The Itshift
2-coordinate
complex
between
coordination
number
and
Be chemical
is notbe
exact,
andhowever,
canberyllium
be strongly
influenced
by
a
three-coordinate
geometry
in
solution
[11].
should
noted,
thatbis(N,N´-bis(2,6the
correlation
9Be NMR chemical shift values can be diagnostic
John and co-workers
have
demonstrated
that of
9Be
example,
has
an extreme
downfield
shift
ofchemical
δ44
ppm
[20],
whereas
the
2-ccordinate
displays
diisopropylphenyl)-1,3,2-diazaborolyl),
for example,
has
an exact,
extreme
downfield
shift bis(N,N´-bis(2,6of influenced
δ44
[20],
the
electronic
properties
the
ligands.
The shift
2-coordinate
complex
between
coordination
number
and
is not
and
canberyllium
be Be(N(SiMe
strongly
by
3 )2 ppm
9 Bein
9 Be
oordination numbers
solution
[19].
Typically,
organoberyllium
complexes
with
low
formal
9Be
a the
NMR
shift
at
δ12.3
ppm
[4].
Nevertheless,
the
of
1
is
at
δ22.8
ppm,
which
to
our
knowledge
is
whereas
the
2-ccordinate
Be(N(SiMe
3
)
2
displays
a
NMR
shift
at
δ12.3
ppm
[4].
Nevertheless,
the
diisopropylphenyl)-1,3,2-diazaborolyl),
for example,
has an extreme
downfield
shift bis(N,N´-bis(2,6of δ44 ppm [20],
electronic properties of the ligands.
The 2-coordinate
complex
beryllium
9Be
dination numbers,the
such
as1BeMe
·Et2shift
O ppm,
(coordination
3, example,
δ20.8appm
inspecies.
Et2extreme
O),
are
9is
of
isthe
at 22-ccordinate
δ22.8
which
tonumber
our
most
positive
shift
yetwere
reported
for
a threemost
positive
yet reported
for
three-coordinate
DFT
methods
used
toppm
predict
whereas
Be(N(SiMe
3a
)2knowledge
displays
Bethe
NMR
shift
atobserved
δ12.3
ppm
[4].
Nevertheless,
the
diisopropylphenyl)-1,3,2-diazaborolyl),
for
has
an
downfield
shift
of δ44
[20],
9Be chemical shift of δ18.2 ppm, which9 is consistent with
9 Be chemical
downfield of 0 ppm.
BeA’
2species.
O)
has
appm,
9coordinate
DFT
methods
used
to predict
the
Beshift
chemical
shift
value
of
1 (B3LYP-D3/6the
shift
value
of 1 (B3LYP-D3/6-311+G(2d,p)//B3LYP-D3/6-31G(d)).
It
was
calculated
Be
of 12·(Et
isthe
at
δ22.8
which
towere
our
most
positive
shift
yet
reported
for a threewhereas
2-ccordinate
Be(N(SiMe
3)2knowledge
displays
a 9isBethe
NMR
at δ12.3
ppm
[4].
Nevertheless,
the
ree-coordinate geometry
in
solution
[11].
It
should
be
noted,
however,
that
the
correlation
9
9311+G(2d,p)//B3LYP-D3/6-31G(d)).
It
was
calculated
at
δ25.9
ppm,
in
reasonable
agreement
with
the
coordinate
species.
DFT methods
used to predict
the
Be chemical
shift yet
value
of 1 (B3LYP-D3/6Be of 1 is at
δ22.8 ppm,
which towere
our knowledge
is the
most
positive shift
reported
for a three9Be chemical shift is not exact,
een coordination number
and
and
can
be
strongly
influenced
by
2+
9Be
2)used
4] calculated
with
an isotropic
shielding
constant
ofagreement
108.98
ppm).
observed
(referenced
to [Be(OH
311+G(2d,p)//B3LYP-D3/6-31G(d)).
It was
atthe
δ25.9
ppm,
in reasonable
with the
coordinatevalue
species.
DFT methods
were
to predict
chemical
shift value
of 1 (B3LYP-D3/6electronic properties
of
the
ligands.
The
2-coordinate
complex
beryllium
bis(N,N´-bis(2,62+ with an
2
)
4
]
isotropic
shielding
constant
of
108.98
ppm).
observed
value
(referenced
to
[Be(OH
311+G(2d,p)//B3LYP-D3/6-31G(d)). It was calculated at δ25.9 ppm, in reasonable agreement with the
propylphenyl)-1,3,2-diazaborolyl),
for example,tohas
an extreme
downfield shift of δ44 ppm [20],
2)4]2+ with an isotropic shielding constant of 108.98 ppm).
observed value (referenced
[Be(OH
reas the 2-ccordinate Be(N(SiMe3)2 displays a 9Be NMR shift at δ12.3 ppm [4]. Nevertheless, the
f 1 is at δ22.8 ppm, which to our knowledge is the most positive shift yet reported for a three-
Complex
δ H(α)/H(γ)
δ H(β)
δ SiMe3
M–C (σ) Å
M’…C(olefin) Å
Ref.
Li[ZnA’3]
6.46
3.50
0.15
2.117(3)
2.745(4), 2.268(3)
[15]
Na[ZnA’3]
7.59
4.00
0.16
2.103(3)
2.857(3), 2.567(3)
6.97
3.19
0.20
1.805(10)
3.153(7), 2.940(7)
7.05
3.42
0.23
2.068(4)
3.205(3), 2.945(3)
6.43
4.42
0.42, 0.23
2.344(7)
3.201(7), 3.164(8), 3.065(8)
K[BeA’3]
Inorganics 2017, 5, 36
K[ZnA’3]
K(thf)[SnA’3]
a
b
b
[15]
this work
4 of 11
[15]
[16]
2+ with an
a Two
b Distance(s)
at δ25.9
ppm,
in reasonable
agreement
observed
(referenced
to [Be(OH
resonances
are observed
for the with
SiMe3the
groups,
as the value
A’ ligands
are not fluxional;
2 )4 ]
isotropic
shielding
constant of disorder.
108.98 ppm).
affected
by crystallographic
Figure 1. 11H NMR spectrum (500 Mhz) of isolated 1, recorded in C6D6. The starred peaks represent
Figure 1. H NMR spectrum (500 Mhz) of isolated 1, recorded in C6 D6 . The starred peaks represent
impurities: δ7.15 (residual protons of C6D6); δ0.9 and δ1.3, residual hexanes.
impurities: δ7.15 (residual protons of C6 D6 ); δ0.9 and δ1.3, residual hexanes.
2.4. Solid State Structure
Table 2. 1 H NMR shifts (ppm) and bond distances in M[M’A’3 ] complexes.
The structure of 1 was determined from single crystal X-ray diffraction. In the solid state, 1
Complex
δ H(α) /H
δ H(β) withδσ-bound
SiMe3
Ref. in
M–C
(σ) Å and aM’
···C(olefin)cation
Å
exhibits
approximate
C3(γ)
-symmetry,
A’
ligands
potassium
engaging
b
b
cation–π
with the three
bonds of the
allyls. It is isostructural
with the previously
Li[ZnA’interactions
6.46
3.50 double0.15
[15]
2.117(3)
2.745(4), 2.268(3)
3]
Na[ZnA’
0.16
2.103(3)
2.857(3),
3]
reported
M[ZnA’
3] 7.59
(M = Li, Na,4.00
K) and K(thf)[SnA’
3] complexes
[15,16].
The 2.567(3)
beryllium center[15]
is in a
K[BeA’
]
6.97
3.19
0.20
1.805(10)
3.153(7),
2.940(7)
this
work
3
nearly planar trigonal environment (sum of C–Be–C´ angles = 357.7°) (Figure 2).
K[ZnA’3 ]
7.05
3.42
0.23
2.068(4)
3.205(3), 2.945(3)
[15]
The average
Be–C
Åa has 2.344(7)
few direct3.201(7),
points 3.164(8),
of comparison
other
K(thf)[SnA’
6.43 distance
4.42of 1.805(10)
0.42, 0.23
3.065(8) with[16]
3]
− complex, the other being
molecules,
as
1
is
only
the
second
crystallographically
characterized
[BeR
3
]
a Two resonances are observed for the SiMe groups, as the A’ ligands are not fluxional; b Distance(s) affected by
3
lithium
tri-tert-butylberyllate
[21]. The latter’s Be center, like that in 1, is in a nearly perfectly planar
crystallographic
disorder.
trigonal environment (sum of C–Be–C angles = 359.9°). In the solid state, however, tri-tertbutylberyllate
is a dimer, [Li{Be(t-C4H9)3}]2, with some corresponding distortions in the Be–C bond
2.4. Solid State Structure
lengths; Be–C distances range from 1.812(4) Å to 1.864(4) Å, averaging to 1.843(6) Å. The Be–C length
The structure of 1 was determined from single crystal X-ray diffraction. In the solid state, 1 exhibits
in 1 is indistinguishable from the Be–Ccarbene length of 1.807(4) Å in the [Ph2Be(IPr)] (IPr = 1,3-bis(2,6approximate C3 -symmetry, with σ-bound A’ ligands and a potassium cation engaging in cation–π
di-isopropylphenyl)-imidazol-2-ylidene) complex, which also has a three-coordinate Be center [22].
interactions with the three double bonds of the allyls. It is isostructural with the previously reported
The anionic methyl groups in [Ph2Be(IPr)] are at a noticeably shorter distance, however (1.751(6) Å,
M[ZnA’ ] (M = Li, Na, K) and K(thf)[SnA’3 ] complexes [15,16]. The beryllium center is in a nearly
ave.). A 3similar relationship between the Be–C
carbene and Be–CH3 bond lengths exists in the related
planar trigonal environment (sum of C–Be–C´ angles = 357.7◦ ) (Figure 2).
[Me2Be(IPr)] [23] and [Me2Be(IMes)] (IMes = N,N’-bis(2,4,6-trimethylphenyl)imidazol-1-ylidene)
The average Be–C distance of 1.805(10) Å has few direct points of comparison with other molecules,
complexes [1]. A comparison of the Be–C length in 1 could also −
be made with the Be–C distance of
as 1 is only the second crystallographically characterized [BeR ] complex, the other being lithium
1.84 Å in lithium tetramethylberyllate, Li2[BeMe4], although the3bond distance would be expected to
tri-tert-butylberyllate [21]. The latter’s Be center, like that in 1, is in a nearly perfectly planar trigonal
be slightly longer in the latter owing to the higher
coordination number of beryllium and the greater
environment (sum of C–Be–C angles = 359.9◦ ). In the solid state, however, tri-tert-butylberyllate is a
negative charge [24].
dimer, [Li{Be(t-C4 H9 )3 }]2 , with some corresponding distortions in the Be–C bond lengths; Be–C
distances range from 1.812(4) Å to 1.864(4) Å, averaging to 1.843(6) Å. The Be–C length in 1 is
indistinguishable from the Be–Ccarbene length of 1.807(4) Å in the [Ph2 Be(IPr)] (IPr = 1,3-bis(2,6di-isopropylphenyl)-imidazol-2-ylidene) complex, which also has a three-coordinate Be center [22].
The anionic methyl groups in [Ph2 Be(IPr)] are at a noticeably shorter distance, however (1.751(6) Å,
ave.). A similar relationship between the Be–Ccarbene and Be–CH3 bond lengths exists in the related
[Me2 Be(IPr)] [23] and [Me2 Be(IMes)] (IMes = N,N’-bis(2,4,6-trimethylphenyl)imidazol-1-ylidene)
complexes [1]. A comparison of the Be–C length in 1 could also be made with the Be–C distance of
Inorganics 2017, 5, 36
5 of 11
1.84 Å in lithium tetramethylberyllate, Li2 [BeMe4 ], although the bond distance would be expected to
be slightly longer in the latter owing to the higher coordination number of beryllium and the greater
Inorganics
2017, 5,
36
5 of 11
negative
charge
[24].
Figure
2. Thermal
ellipsoid
plotofof1,1,illustrating
illustrating the
used
in the
text.text.
Ellipsoids
Figure
2. Thermal
ellipsoid
plot
the numbering
numberingscheme
scheme
used
in the
Ellipsoids
are drawn
at the
50%level,
level,and
andfor
for clarity,
clarity, hydrogen
removed
fromfrom
the the
are drawn
at the
50%
hydrogen atoms
atomshave
havebeen
been
removed
trimethylsilyl
groups.
Selected
bond
distances
(Å)
and
angles
(deg):
Be1–C1,
1.795(6);
Be1–C10,
1.810(6);
trimethylsilyl groups. Selected bond distances (Å) and angles (deg): Be1–C1, 1.795(6); Be1–C10,
Be1–C19, 1.811(6); C(2)–C(3), 1.351(5); C(11)–C(12), 1.350(5); C(20)–C(21), 1.358(5); K(1)–C(2), 3.138(4);
1.810(6); Be1–C19, 1.811(6); C(2)–C(3), 1.351(5); C(11)–C(12), 1.350(5); C(20)–C(21), 1.358(5); K(1)–C(2),
K(1)–C(3), 2.940(4); K(1)–C(11), 3.206(4); K(1)–C(12), 2.943(4); K(1)–C(20), 3.114(4); K(1)–C(21), 2.938(4);
3.138(4); K(1)–C(3), 2.940(4); K(1)–C(11), 3.206(4); K(1)–C(12), 2.943(4); K(1)–C(20), 3.114(4); K(1)–
C(1)–Be(1)–C(10), 119.1(3); C(1)–Be(1)–C(19), 119.0(3); C(10)–Be(1)–C(19), 119.4(3).
C(21), 2.938(4); C(1)–Be(1)–C(10), 119.1(3); C(1)–Be(1)–C(19), 119.0(3); C(10)–Be(1)–C(19), 119.4(3).
The C–C and C=C bonds in the alkyl groups in 1 are localized at 1.475(5) Å and 1.353(9) Å,
The C–C and C=C bonds in the alkyl groups in 1 are localized at 1.475(5) Å and 1.353(9) Å,
respectively. The K+ ···C(olefin) contacts average 3.153(7) Å and 2.940(7) Å to the carbon atoms β
+…
contacts
average
Å andatom,
2.940(7)
Å to the carbon
atoms β (C2,
respectively.
(C2, C11, The
and K
C20)C(olefin)
and γ (C3,
C12, and
C21) to3.153(7)
the beryllium
respectively.
These distances
+
C11, are
and
C20)
and
γ
(C3,
C12,
and
C21)
to
the
beryllium
atom,
respectively.
These
distances
comparable to, but slightly shorter than, the range of K ···C contacts found in the related zincate are
structureto,
(3.205(3)
Å and shorter
2.945(3) than,
Å, respectively),
which
the shorter
bonds inzincate
1.
comparable
but slightly
the range of
K+…reflects
C contacts
foundM–C
in the
(α) related
The distance
between
and K (3.59
Å) is long enough
to rule
out significant
metal-metal
interactions.
structure
(3.205(3)
Å andBe2.945(3)
Å, respectively),
which
reflects
the shorter
M–C(α) bonds
in 1. The
distance between Be and K (3.59 Å) is long enough to rule out significant metal-metal interactions.
2.5. Computational Investigations
It has previously
been suggested that the occurrence of C3 -symmetric M[M’A’3 L] (M’ = Zn, M = Li,
2.5. Computational
Investigations
Na, K; M’ = Sn, M = K; L = thf) complexes is the result of a templating effect of the associated alkali metal
It has previously
suggested
the occurrence
of C3-symmetric
3L] complexes
(M’ = Zn, M =
counterion
[25]. Thebeen
rationale
for thisthat
proposal
is that the neutral
MA’3 (M =M[M’A’
As, Sb, Bi)
Li, Na,
K;
M’
=
Sn,
M
=
K;
L
=
thf)
complexes
is
the
result
of
a
templating
effect
of
the
associated
alkali
always occur in two diastereomeric forms, with R,R,R (equivalently, S,S,S) and R,R,S (or S,S,R)
−
metalarrangements
counterion of
[25].
rationale
for this
proposal
is thatThe
theanionic
neutral
MA’
(M = As, Sb,
the The
allyl ligands
around
the central
element.
[MA
in Bi)
3 ´] 3 complexes,
contrast,
are
always
found
in
the
C
-symmetric
R,R,R
(or
S,S,S)
configuration,
and
it
is
not
unreasonable
complexes always occur in two diastereomeric
forms, with R,R,R (equivalently, S,S,S) and R,R,S (or
3
to arrangements
assume that theof
counterion
responsible
for the
S,S,R)
the allyl isligands
around
thedifference.
central element. The anionic [MA3´]− complexes,
A
DFT
investigation
was
undertaken
to
explore
possible
this effect. The and
geometry
in contrast, are always found in the C3-symmetricthe
R,R,R
(or origins
S,S,S) ofconfiguration,
it is not
−
of the free [BeA’3 ] anion was optimized with calculations employing the dispersion-corrected APF-D
unreasonable to assume that the counterion is responsible for the difference.
functional [26]. Three confirmations were examined: the C3 -symmetric form (S,S,S) found in the X-ray
A DFT investigation was undertaken to explore the possible origins of this effect. The geometry
crystal structure of 1, a related S,S,S form with one A’ ligand rotated antiparallel to the other two
of the free [BeA’3]− anion was optimized with calculations employing the dispersion-corrected APF(C1 symmetry), and a R,R,S form, also with one ligand antiparallel to the other two, derived from the
D functional
confirmations
examined:
structure [26].
of theThree
neutral
AlA’3 complexwere
(Figure
3) [27]. the C3-symmetric form (S,S,S) found in the X-
ray crystal structure of 1, a related S,S,S form with one A’ ligand rotated antiparallel to the other two
(C1 symmetry), and a R,R,S form, also with one ligand antiparallel to the other two, derived from the
structure of the neutral AlA’3 complex (Figure 3) [27].
Inorganics 2017, 5, 36
6 of 11
Inorganics 2017, 5, 36
Inorganics 2017, 5, 36
6 of 11
6 of 11
Figure 3. Geometry optimized structures of [BeA’3]− anions: (a) as found in the crystal structure of 1;
(b) related S,S,S form with one A’ ligand rotated (C1 symmetry); and (c) R,R,S form derived from the
− anions: (a) as found in the crystal structure of 1;
Figure
3.
Figure
3. Geometry
Geometry
optimized structures
structures of
of [BeA’
[BeA’33]]− anions:
(a) as found in the crystal structure of 1;
structure
of AlA’3. optimized
(b)
(b) related
related S,S,S
S,S,S form
form with
with one
one A’
A’ ligand
ligand rotated
rotated (C
(C11 symmetry);
symmetry); and
and (c)
(c) R,R,S
R,R,S form
form derived
derived from
from the
the
structure
of
Not surprisingly,
structure
of AlA’
AlA’33.. the calculated structures possess similar average Be–C bond lengths, ranging
from 1.782 Å (the C3-symmetric form (Figure 3a)) to 1.788 Å (for the rotated S,S,S form (Figure 3b)).
the
calculated
structures
possess
Not surprisingly,
calculated
structures
possess
similar
Be–C
bond−1 lengths,
Energetically,
the R,R,Sthe
form
is the most
stable; the
rotated
S,S,S average
form is 10.1
kJ·mol
higher inranging
energy
−1 in
form
3a))
to
Å
the
rotated
form
(Figure
3b)).
from
(the
C33-symmetric
-symmetric
form
(Figurestill
3a))(20.3
to 1.788
1.788
Å (for
(for
the
rotated
S,S,S
form
(Figure
3b)).
(∆G°),1.782
and Å
the
C3-symmetric
form
is (Figure
higher
kJ·mol
∆G°).
TheS,S,S
origin
of these
energy
−1
−
1
Energetically,
the
R,R,S
form
is
the
most
stable;
the
rotated
S,S,S
form
is
10.1
kJ·mol
higher
in
energy
Energetically,
form is the
most stable;
rotated
S,S,S to
form
10.1 kJ·amounts
mol higher
differences is the
notR,R,S
immediately
obvious,
but it the
may
be related
theisrelative
of interligand
−11 in ∆G°).
◦ ), and the C
−
◦ ).…
(∆G°),
C
3-symmetric form is higher
higher
still
(20.3
kJ·mol
The
origin
of
(∆G
still
(20.3
kJ
·
mol
in
∆G
The
origin
of
energy
3
congestion present. The low energy R,R,S form, for example, has no Me Me´ contactsthese
less than
4.0
but
relative
amounts
interligand
differences
is
not
immediately
obvious,
it
may
be
related
to
the
relative
amounts
of
interligand
Å, the sum of the van der Waals radii [18]. In contrast, the C3 symmetric form has multiple contacts
congestion
present.
The
energy
form,
for
Me
···…
Me´
contacts
less
than
4.04.0
Å,
congestion
present.
Thelow
low
energy
R,R,S
form,
forexample,
example,
hasnono
Me
Me´
less
between methyl
groups
of less
thanR,R,S
4.0 Å,
including
two ashas
short
as
3.76
Å.
Atcontacts
this level
ofthan
theory,
the
sum
of
the
van
der
Waals
radii
[18].
In
contrast,
the
C
symmetric
form
has
multiple
contacts
Å,
the
sum
of
the
van
der
Waals
radii
[18].
In
contrast,
the
C
3
symmetric
form
has
multiple
the energetics of the free anions do not provide a rationale3 for the exclusive formation of the S,S,S
between
asas
short
as as
3.76
Å. Å.
At At
thisthis
level
of theory,
the
methyl groups
groupsof
ofless
lessthan
than4.0
4.0Å,Å,including
includingtwo
two
short
3.76
level
of theory,
form. methyl
energetics
of the
do not
a +rationale
for
exclusive
formation
of the
S,S,S
the energetics
of free
the anions
free
anions
doprovide
not
provide
rationale
for
the alters
exclusive
formation
of theform.
S,S,S
Not surprisingly,
incorporation
of the
K
ionainto
the the
complex
the relative
stability
of
the
+ ion into the complex alters the relative stability of the
Not
surprisingly,
incorporation
of
the
K
form.
species. The optimized geometries of the C3-symmetric K[BeA’3] found in the X-ray crystal structure
species.
optimized
geometries
of calculated
the
C3 -symmetric
in anions,
the
crystal
structure
of
NotThe
surprisingly,
incorporation
of the
K+ ion
intoK[BeA’
thetocomplex
alters
theX-ray
relative
stability
of the
3 ] found
of 1 and
a related
S,S,R
form were
similarly
the
isolated
and
are
depicted
in
1species.
and
a
related
S,S,R
form
were
calculated
similarly
to
the
isolated
anions,
and
are
depicted
in
Figure
4.
The
optimized
geometries
of
the
C
3
-symmetric
K[BeA’
3
]
found
in
the
X-ray
crystal
structure
Figure 4.
of 1 and a related S,S,R form were calculated similarly to the isolated anions, and are depicted in
Figure 4.
Figure 4.
4. Geometry
structures of
of the
the K[BeA’
K[BeA’3]] complex:
complex: (a)
(a) as
Figure
Geometry optimized
optimized structures
as found
found in
in the
the crystal
crystal structure
structure
3
of
1;
and
(b)
related
S,S,R
form.
of 1; and (b) related S,S,R form.
Figure 4. Geometry optimized structures of the K[BeA’3] complex: (a) as found in the crystal structure
The
C3-symmetric
formform.
is 6.1 kJ·mol−−11more stable than the S,S,R form. This is not a consequence
of
1; and
(b) related S,S,R
The C3 -symmetric form is 6.1 kJ·mol more stable than the S,S,R form. This is not a consequence
of closer K++…(C=C) distances, which are nearly the same (avg. 3.91 Å in the C3 form; 2.86 Å in the
of closer
K
···(C=C) distances,
which
are−1nearly
the same
(avg.
3.91 form.
Å in the
C3isform;
2.86
Å in the
C3-symmetric
is 6.1
kJ·mol
more stable
than
the S,S,R
not a lead
consequence
S,S,RThe
arrangement).
Theform
asymmetric
arrangement
of the
ligands
in the
S,S,RThis
form
does
to closer
S,S,R arrangement).
The
asymmetric
arrangement
of
the
ligands
in
the
S,S,R
form
does
lead
to
closer
+…(C=C) distances, which are nearly the same (avg. 3.91 Å in the C3 form; 2.86 Å in the
of
closer
K
…
interligand
C
C
contacts
in
the
allyl
frameworks,
however,
as
small
as
3.37
Å,
whereas
there
are
no
interligand C···C contacts in the allyl frameworks, however, as small as 3.37 Å, whereas there are
S,S,R
arrangement).
The asymmetric
arrangement
of the ligands in the S,S,R form does lead to closer
similar
contacts
less
than
3.78
Å
in
the
C
3 form. The somewhat greater stability of the C3 form, possibly
no similar contacts less than 3.78 Å in the C3 form. The somewhat greater stability of the C3 form,
interligand
C…
C contacts
in crystal
the allyl
frameworks,
however, to
as the
small
as 3.37 appearance
Å, whereas there
no
coupled with
greater
of
packing,
contribute
exclusive
of thatare
form
possibly
coupled
withease
greater
ease of
crystal may
packing,
may contribute
to the exclusive appearance
of
similar contacts less than 3.78 Å in the C3 form. The somewhat greater stability of the C3 form, possibly
coupled with greater ease of crystal packing, may contribute to the exclusive appearance of that form
Inorganics 2017, 5, 36
Inorganics 2017, 5, 36
7 of 11
7 of 11
thatthe
form
in thestructure.
crystal structure.
It isthat
likely
that a similar
analysis
forisostructural
the isostructural
Zn and
in
crystal
It is likely
a similar
analysis
holdsholds
for the
Zn and
Sn
Sn complexes.
complexes.
The failure to produce an unsolvated BeA’22 in the absence of a coordinating solvent (i.e., either
mechanochemically or
or in hexanes) was also examined computationally
computationally with
with the
the aid of the Solid-G
mechanochemically
·
program
[28].
Both
BeA’
Et
O
and
1
are
found
to
have
coordination
sphere
coverage
program [28]. Both BeA’22·Et22
found to have coordination sphere coverage(G
(G
complex ) above
complex
90% (i.e., 97.0% (Figure 5a) and 92.6% (Figure 5b), respectively). Although the coverage of the metal
center in the hypothetical BeA’22 varies somewhat with the angle between the ligands, the minimum
energy position depicted in Figure 5c (C22 symmetry) has only 78.7% coverage. It is not unreasonable
bebe
too
coordinately
unsaturated
to be
isolable,
and and
will
to assume
assume that
thataamonomeric
monomericBeA’
BeA’
2 may
too
coordinately
unsaturated
to readily
be readily
isolable,
2 may
bindbind
an ethereal
solvent
molecule
during
synthesis,
or, if that
not available,
an additional
A’ ligand,
will
an ethereal
solvent
molecule
during
synthesis,
or, ifisthat
is not available,
an additional
A’
+
+
counterbalanced
with a Kwith
ion.
ligand,
counterbalanced
a K ion.
(a)
(b)
(c)
Figure
5. Visualization
Visualization of
of the
theextent
extentof
ofcoordination
coordinationsphere
spherecoverage
coverage(G(Gcomplex)) of:
of: (a)
(a) BeA’
BeA’2··Et
2O (the
Figure 5.
2 Et2 O (the
complex
coverage
from
the
two
allyls
are
assigned
blue
and
green;
that
from
the
ether
is
in
red);
(b)
1
(all
coverage from the two allyls are assigned blue and green; that from the ether is in red); (b) 1 (all three
three
allyls
allyls are
are in
in blue);
blue); and
and (c)
(c) BeA’
BeA’22,, using
using optimized
optimized coordinates
coordinates (APF-D/6-311G(2d)
(APF-D/6-311G(2d) (Be);
(Be); 6-31G(d)
6-31G(d) (other
(other
atoms))
and the
the program
program Solid-G
Solid-G[28].
[28]. The
TheG
Gcomplex value
takes into account the net coverage; regions
atoms)) and
complex value takes into account the net coverage; regions
of
the
coordination
sphere
where
the
projections
of
the
ligands
overlap are
are counted
counted only
only once.
once.
of the coordination sphere where the projections of the ligands overlap
3.
3. Materials
Materials and
and Methods
Methods
General Considerations:
Considerations: All
General
All syntheses
syntheses were
were conducted
conducted under
under rigorous
rigorous exclusion
exclusion of
of air
air and
and
moisture
using
Schlenk
line
and
glovebox
techniques.
(NOTE:
Beryllium
salts
are
toxic
and
should
moisture using Schlenk line and glovebox techniques. (NOTE: Beryllium salts are toxic and should be
be
handled
appropriate
protective
equipment.)
grinding
was completed,
jars
were
handled
withwith
appropriate
protective
equipment.)
After After
grinding
was completed,
the jarsthe
were
opened
opened
according
to
glovebox
procedures
to
protect
the
compounds
and
to
prevent
exposure
to
according to glovebox procedures to protect the compounds and to prevent exposure to dust dust
[10].
1H) and carbon (13C) spectra were obtained on Bruker DRX-500 or DRX-400
[10].
Proton
(
1
13
Proton ( H) and carbon ( C) spectra were obtained on Bruker DRX-500 or DRX-400 spectrometers
spectrometers
(Karlsruhe,
Germany),
and to
were
referenced
to residual
of9C6D6. Beryllium
(Karlsruhe,
Germany),
and were
referenced
residual
resonances
of C6 Dresonances
6 . Beryllium ( Be) spectra were
9Be) spectra were obtained on a Bruker DRX-500 at 70.2 MHz, and were referenced to BeSO4(aq).
(obtained
on a Bruker DRX-500 at 70.2 MHz, and were referenced to BeSO4 (aq). Combustion analysis
Combustion
analysis
was
performed by Tucson,
ALS Environmental,
Tucson, chloride
AZ, USA.
Beryllium
chloride
was
performed
by ALS
Environmental,
AZ, USA. Beryllium
was
purchased
from
was
purchased
from
Strem,
stored
under
an
N
2 atmosphere and used as received. The K[A’] (A’ =
Strem, stored under an N2 atmosphere and used as received. The K[A’] (A’ = 1,3-(SiMe3 )2 C3 H3 )
1,3-(SiMe
3)2C3H3) reagent was synthesized as previously described [29,30]. Toluene, hexanes, and
reagent was
synthesized as previously described [29,30]. Toluene, hexanes, and diethyl ether were
diethyl
ether
distilled
nitrogen
from potassium
ketyl
[31]. (C
Deuterated
distilled underwere
nitrogen
fromunder
potassium
benzophenone
ketyl benzophenone
[31]. Deuterated
benzene
6 D6 ) was
benzene
(C
6D6) was distilled from Na/K (22/78) alloy prior to use. Stainless steel (440 grade) ball
distilled from Na/K (22/78) alloy prior to use. Stainless steel (440 grade) ball bearings (6 mm,) were
bearings
(6 mm,)
were
thoroughly
cleaned
withprior
hexanes
andPlanetary
acetone prior
to use.
milling
thoroughly
cleaned
with
hexanes and
acetone
to use.
milling
wasPlanetary
performed
with
was
performed
with
a
Retsch
model
PM100
mill
(Haan,
Germany),
50
mL
stainless
steel
grinding
jar
a Retsch model PM100 mill (Haan, Germany), 50 mL stainless steel grinding jar type C, and safety
type
C,
and
safety
clamp
for
air-sensitive
grinding.
clamp for air-sensitive grinding.
3.1.
3.1. Mechanochemical
Mechanochemical Synthesis
Synthesis of
of K[BeA’
K[BeA’33]] (1)
(1)
Solid BeCl
BeCl22 (56.7
Solid
(56.7 mg,
mg, 0.71
0.71 mmol)
mmol) and
and K[A’]
K[A’] (319
(319 mg,
mg, 1.42
1.42 mmol)
mmol) were
were added
added to
to aa 50
50 mL
mL stainless
stainless
steel
grinding
jar
(type
C).
The
jar
was
charged
with
stainless
steel
ball
bearings
(6
mm
dia,
50
steel grinding jar (type C). The jar was charged with stainless steel ball bearings (6 mm dia, 50 count)
count)
and
closed tightly
tightly with
with the
the appropriate
appropriate safety
safety closer
closer device
device under
under an
an N
N22 atmosphere.
atmosphere. The
reagents
and closed
The reagents
were milled for 15 min at 600 rpm, resulting in a light orange solid. The product was extracted under
an inert atmosphere with minimal hexanes (< 100 mL) and filtered through a medium porosity
Inorganics 2017, 5, 36
8 of 11
were milled for 15 min at 600 rpm, resulting in a light orange solid. The product was extracted under
an inert atmosphere with minimal hexanes (<100 mL) and filtered through a medium porosity ground
glass frit, providing a dark orange filtrate. Drying under vacuum yielded a dark orange solid (61.5 mg,
21% yield of K[BeA’3 ]) which was recrystallized by the slow evaporation of toluene over one month
to provide dark orange-brown crystals of 1 suitable for single crystal X-ray diffraction. For a 3:1
K[A’]:BeCl2 reaction, 812 mg (3.62 mmol) K[A’] and 95.2 mg (1.19 mmol) BeCl2 were added to a
grinding jar. After extraction, 183 mg (25% yield) of orange solid was collected. Anal. Calcd. (%)
for C27 H63 BeKSi6 : C, 53.65; H, 10.51; Be, 1.49. Found: C, 52.09; H, 9.79; Be, 1.04. The values are
somewhat low, possibly from the high air-sensitivity of the compound, but the C:H molar ratio is
2.34:1.00, close to the expected 2.33:1.00. 1 H NMR (500 MHz, C6 D6 , 298K): δ 0.20 (s, 54H, SiMe3 ); 3.19
(br s (ν1/2 = 39 Hz), 6H, H(α,γ) ); 6.97 (t, 3H, J1 = 16 Hz, H(β) ). 13 C NMR (100 MHz, C6 D6 , 298K): δ1.02
(s, SiMe3 ); 70.71 (s, C(α,γ) ); 166.09 (s, C(β) ). 9 Be NMR (70.2 MHz, C6 D6 , 298K); δ22.8 (s) (ν1/2 = 360 Hz).
3.2. General Procedures for Synthesis of K[BeA’3 ] (1) with Solvents
Reactions were performed for either 1 or 16 h, and were run under inert atmosphere at room
temperature. The ratio of K[A’] and BeCl2 was varied such that the reactions of emphasis were 1:1,
2:1, and 3:1. A general reaction involved dissolving the beryllium chloride (ca. 0.1 g) in the solvent
of choice (Et2 O or hexanes); to this solution solid K[A’] was added slowly and solvent was used to
quantitatively transfer all material. Upon mixing, the solution was allowed to stir for the given time.
In the case of Et2 O, the solvent was removed in vacuo, and the resulting material was extracted with
hexanes, filtered through a medium porosity glass frit, and then dried in vacuo. In the case of reaction
in hexanes, the reaction mixture was filtered through a medium porosity fritted glass filter, and the
hexane was removed in vacuo. The resulting material in all cases was then analyzed with 1 H and
9 Be NMR.
3.3. Procedures for X-ray Crystallography
A crystal (0.20 × 0.20 × 0.08 mm3 ) was placed onto the tip of a thin glass optical fiber and mounted
on a Bruker SMART APEX II CCD platform diffractometer (Karlsruhe, Germany) for a data collection
at 100.0(5) K [32]. The structure was solved using SIR2011 [33] and refined using SHELXL-2014/7 [34].
The space group P1 was determined based on intensity statistics. A direct-methods solution was
calculated that provided most non-hydrogen atoms from the E-map. Full-matrix least squares/difference
Fourier cycles were performed which located the remaining non-hydrogen atoms. All non-hydrogen
atoms were refined with anisotropic displacement parameters. The allylic hydrogen atoms were found
from the difference Fourier map and refined freely. All other hydrogen atoms were placed in ideal
positions and refined as riding atoms with relative isotropic displacement parameters.
3.4. General Procedures for Calculations
All calculations were performed with the Gaussian 09W suite of programs [35]; an ultrafine
grid was used for all cases (Gaussian keyword: int = ultrafine). Each conformation of the [BeA’3 ]−
complexes was studied with the APF-D functional, a global hybrid with 23% exact exchange [26].
The 6-31+G(d) basis set was used for C,H,Si; the 6-311+G(2d) basis was used for Be. For the neutral
K[BeA’3 ] conformations, the APF-D functional was used with the 6-31G(d) basis set for C,H,Si;
6-311G(2d) was used for Be and K. The nature of the stationary points was determined with analytical
frequency calculations; all of these optimized geometries were found to be minima (Nimag = 0). For the
Solid-G calculations, the structures were preoptimized with the APF-D/6-311G(2d) (Be,K); 6-31G(d)
(C,H,Si) protocol.
Inorganics 2017, 5, 36
9 of 11
4. Conclusions
The generation of products from reagents that are not in the optimum stoichiometric ratio is
a known feature of some Group 2 reactions [36,37], a testament to the role that kinetic factors play
in s-block chemistry. It is perhaps not surprising that when mechanochemical activation is used
with alkaline earth reagents, a nonstoichiometric product such as the organoberyllate 1 is formed, as
grinding and milling environments are often far from equilibrium [38–41]. However, the fact that 1 is
also generated in hexanes indicates how non-ethereal synthesis can reveal features of reactions that are
obscured when they are conducted in coordinating solvents. It is now apparent that the production of
the previously described BeA’2 ·Et2 O, which was the expected complex from a 2:1 reaction of K[A’] and
BeCl2 in diethyl ether [11], actually depends critically on the presence of the solvent to prevent further
reaction of the beryllium center with an additional A’ ligand. (In a preliminary study, the reaction of
K[A0 ] and BeCl2 in a 2:1 molar ratio in THF (1 h) was found not to produce K[BeA0 3 ]. A species with a
9 Be NMR shift of δ16.6 ppm was present instead, tentatively identified as BeA0 (thf). If correct, this
2
indicates that THF, like Et2 O, can block the formation of the tris(allyl) anion with Be). Without such
ethereal solvent support, whether conducted mechanochemically or in hexanes, the reaction between
K[A’] and BeCl2 rapidly forms the kinetic product 1.
Parallels of the beryllium chemistry to the related tris(allyl) -ate complexes of Zn and Sn are
instructive, although they cannot be pushed too far. All the [MA3 ´]− species possess approximate C3
symmetry, and it is likely that the associated alkali metal cation is intimately involved in templating
their constructions. The formation of the zinc species K[ZnA’3 ] is also similar to that of 1 in that it
is formed from the reaction of 2 equiv. of K[A’] and ZnCl2 , i.e., in a non-stoichiometric reaction [15].
However, both it and K(thf)[SnA’3 ] are synthesized in THF, so it is clear that the driving force for -ate
formation compared to that for the neutral (Zn,Sn)A’2 species is greater than that for 1. This may reflect
the somewhat lesser covalency of Be–C versus Zn–C and Sn–C bonds, and the greater robustness of
M2+ ←:OR2 interactions with beryllium.
Supplementary Materials: The following are available online at http://www.mdpi.com/2304-6740/5/2/36/s1:
CIF and checkCIF file, and fractional coordinates of geometry-optimized structures.
Acknowledgments: Financial support by the National Science Foundation (CHE-1112181), The American
Chemical Society–Petroleum Research Fund (56027-ND3), and a Discovery Grant of Vanderbilt University is
gratefully acknowledged.
Author Contributions: The project concept was devised equally by all authors. Nicholas C. Boyde and
Nicholas R. Rightmire performed the experiments and analyzed the resulting data. William W. Brennessel
obtained and solved the single crystal X-ray data. The manuscript was written by Timothy P. Hanusa, with
contributions from all the co-authors.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
Arrowsmith, M.; Hill, M.S.; Kociok-Köhn, G. Activation of N-Heterocyclic Carbenes by {BeH2 } and
{Be(H)(Me)} Fragments. Organometallics 2015, 34, 653–662. [CrossRef]
Shannon, R.D. Revised effective ionic radii and systematic studies of interatomic distances in halides and
chalcogenides. Acta Crystallogr. Sect. A 1976, 32, 751–767. [CrossRef]
Westerhausen, M. Synthesis and Spectroscopic Properties of Bis(trimethylsilyl)amides of the Alkaline-earth
Metals Magnesium, Calcium, Strontium, and Barium. Inorg. Chem. 1991, 30, 96–101. [CrossRef]
Naglav, D.; Neumann, A.; Blaser, D.; Wolper, C.; Haack, R.; Jansen, G.; Schulz, S. Bonding situation in
Be[N(SiMe3 )2 ]2 —An experimental and computational study. Chem. Commun. 2015, 51, 3889–3891. [CrossRef]
[PubMed]
Nugent, K.W.; Beattie, J.K.; Hambley, T.W.; Snow, M.R. A precise Low-temperature crystal structure of
bis(cyclopentadienyl)beryllium. Aust. J. Chem. 1984, 37, 1601–1606. [CrossRef]
Burkey, D.J.; Hanusa, T.P. Structural Lessons from Main-Group Metallocenes. Comments Inorg. Chem. 1995,
17, 41–77. [CrossRef]
Inorganics 2017, 5, 36
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
10 of 11
Civic, T.M. Beryllium Metal Toxicology: A Current Perspective. In Encyclopedia of Inorganic and Bioinorganic
Chemistry; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2011.
Hanusa, T.P.; Bierschenk, E.J.; Engerer, L.K.; Martin, K.A.; Rightmire, N.R. 1.37—Alkaline Earth Chemistry:
Synthesis and Structures. In Comprehensive Inorganic Chemistry II, 2nd ed.; Reedijk, J., Poeppelmeier, K., Eds.;
Elsevier: Amsterdam, The Netherlands, 2013; pp. 1133–1187.
Iversen, K.J.; Couchman, S.A.; Wilson, D.J.D.; Dutton, J.L. Modern organometallic and coordination chemistry
of beryllium. Coord. Chem. Rev. 2015, 297–298, 40–48. [CrossRef]
Naglav, D.; Buchner, M.R.; Bendt, G.; Kraus, F.; Schulz, S. Off the Beaten Track—A Hitchhiker’s Guide to
Beryllium Chemistry. Angew. Chem. Int. Ed. 2016, 55, 10562–10576. [CrossRef] [PubMed]
Chmely, S.C.; Hanusa, T.P.; Brennessel, W.W. Bis(1,3-trimethylsilylallyl)beryllium. Angew. Chem. Int. Ed.
2010, 49, 5870–5874. [CrossRef] [PubMed]
Solomon, S.A.; Muryn, C.A.; Layfield, R.A. s-Block metal complexes of a bulky, donor-functionalized allyl
ligand. Chem. Commun. 2008, 3142–3144. [CrossRef] [PubMed]
Chmely, S.C.; Carlson, C.N.; Hanusa, T.P.; Rheingold, A.L. Classical versus Bridged Allyl Ligands in
Magnesium Complexes: The Role of Solvent. J. Am. Chem. Soc. 2009, 131, 6344–6345. [CrossRef] [PubMed]
Rightmire, N.R.; Hanusa, T.P. Advances in Organometallic Synthesis with Mechanochemical Methods.
Dalton Trans. 2016, 45, 2352–2362. [CrossRef] [PubMed]
Gren, C.K.; Hanusa, T.P.; Rheingold, A.L. Threefold Cation–π Bonding in Trimethylsilylated Allyl Complexes.
Organometallics 2007, 26, 1643–1649. [CrossRef]
Layfield, R.A.; Garcia, F.; Hannauer, J.; Humphrey, S.M. Ansa-tris(allyl) complexes of alkali metals: Tripodal
analogues of cyclopentadienyl and ansa-metallocene ligands. Chem. Commun. 2007, 5081–5083. [CrossRef]
[PubMed]
Neumüller, B.; Weller, F.; Dehnicke, K. Synthese, Schwingungsspektren und Kristallstrukturen der
Chloroberyllate (Ph4 P)2 [BeCl4 ] und (Ph4 P)2 [Be2 Cl6 ]. Z. Anorg. Allg. Chem. 2003, 629, 2195–2199. [CrossRef]
Allred, A.L. Electronegativity values from thermochemical data. J. Inorg. Nucl. Chem. 1961, 17, 215–221.
[CrossRef]
Plieger, P.G.; John, K.D.; Keizer, T.S.; McCleskey, T.M.; Burrell, A.K.; Martin, R.L. Predicting 9 Be Nuclear
Magnetic Resonance Chemical Shielding Tensors Utilizing Density Functional Theory. J. Am. Chem. Soc.
2004, 126, 14651–14658. [CrossRef] [PubMed]
Arnold, T.; Braunschweig, H.; Ewing, W.C.; Kramer, T.; Mies, J.; Schuster, J.K. Beryllium bis(diazaborolyl):
Old neighbors finally shake hands. Chem. Commun. 2015, 51, 737–740. [CrossRef] [PubMed]
Wermer, J.R.; Gaines, D.F.; Harris, H.A. Synthesis and molecular structure of lithium tri-tert-butylberyllate,
Li[Be(tert-C4 H9 )3 ]. Organometallics 1988, 7, 2421–2422. [CrossRef]
Gottfriedsen, J.; Blaurock, S. The First Carbene Complex of a Diorganoberyllium: Synthesis and Structural
Characterization of Ph2 Be(i-Pr-carbene) and Ph2 Be(n-Bu2 O). Organometallics 2006, 25, 3784–3786. [CrossRef]
Arrowsmith, M.; Hill, M.S.; Kociok-Köhn, G.; MacDougall, D.J.; Mahon, M.F. Beryllium-Induced C–N Bond
Activation and Ring Opening of an N-Heterocyclic Carbene. Angew. Chem. Inter. Ed. 2012, 51, 2098–2100.
[CrossRef] [PubMed]
Weiss, E.; Wolfrum, R. Über metall–alkyl-verbindungen VIII. Die kristallstruktur des lithium-tetramethylberyllats.
J. Organomet. Chem. 1968, 12, 257–262. [CrossRef]
Rightmire, N.R.; Bruns, D.L.; Hanusa, T.P.; Brennessel, W.W. Mechanochemical Influence on the
Stereoselectivity of Halide Metathesis: Synthesis of Group 15 Tris(allyl) Complexes. Organometallics 2016, 35,
1698–1706. [CrossRef]
Austin, A.; Petersson, G.A.; Frisch, M.J.; Dobek, F.J.; Scalmani, G.; Throssell, K. A Density Functional with
Spherical Atom Dispersion Terms. J. Chem. Theory Comp. 2012, 8, 4989–5007. [CrossRef] [PubMed]
Rightmire, N.R.; Hanusa, T.P.; Rheingold, A.L. Mechanochemical Synthesis of [1,3-(SiMe3 )2 C3 H3 ]3 (Al,Sc),
a Base-Free Tris(allyl)aluminum Complex and Its Scandium Analogue. Organometallics 2014, 33, 5952–5955.
[CrossRef]
Guzei, I.A.; Wendt, M. An improved method for the computation of ligand steric effects based on solid
angles. Dalton Trans. 2006, 3991–3999. [CrossRef] [PubMed]
Fraenkel, G.; Chow, A.; Winchester, W.R. Dynamics of solvated lithium(+) within exo,exo-[1,3-bis
(trimethylsilyl)allyl]lithium N,N,N0 ,N0 -tetramethylethylenediamine complex. J. Am. Chem. Soc. 1990, 112,
1382–1386. [CrossRef]
Inorganics 2017, 5, 36
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
11 of 11
Harvey, M.J.; Hanusa, T.P.; Young, V.G., Jr. Synthesis and Crystal Structure of a Bis(allyl) Complex of Calcium,
[Ca{C3 (SiMe3 )2 H3 }2 ·(thf)2 ]. Angew. Chem. Int. Ed. 1999, 38, 217–219. [CrossRef]
Perrin, D.D.; Armarego, W.L.F. Purification of Laboratory Chemicals, 3rd ed.; Pergamon: Oxford, UK, 1988.
APEX2; Version 2014.11–0; Bruker AXS: Madison, WI, USA, 2013.
Burla, M.C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano, G.L.; Giacovazzo, C.; Mallamo, M.;
Mazzone, A.; Polidori, G.; Spagna, R. SIR2011: A New Package for Crystal Structure Determination and
Refinement; Version 1.0; Istituto di Cristallografia: Bari, Italy, 2012.
Sheldrick, G.M. SHELXL-2014/7; University of Göttingen: Göttingen, Germany, 2014.
Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.;
Mennucci, B.; Petersson, G.A.; et al. Gaussian 09W; Revision D.01; Gaussian, Inc.: Wallingford CT, USA, 2009.
Glock, C.; Görls, H.; Westerhausen, M. N,N,N 0 ,N 0 -Tetramethylethylendiamine adducts of amido calcium
bases—Synthesis of monomeric [(tmeda)Ca{N(SiMe3 )2 }2 ], [(tmeda)Ca{NiPr2 }2 ], and dimeric Hauser
base-type [(tmeda)Ca(tmp)(µ-I)]2 (tmp = 2,2,6,6-tetramethylpiperidide). Inorg. Chim. Acta 2011, 374, 429–434.
[CrossRef]
Fitts, L.S.; Bierschenk, E.J.; Hanusa, T.P.; Rheingold, A.L.; Pink, M.; Young, V.G. Selective modification of the
metal coordination environment in heavy alkaline-earth iodide complexes. New J. Chem. 2016, 40, 8229–8238.
[CrossRef]
Hernández, J.G.; Friščić, T. Metal-catalyzed organic reactions using mechanochemistry. Tetrahedron Lett. 2015,
56, 4253–4265. [CrossRef]
Boldyreva, E. Mechanochemistry of inorganic and organic systems: What is similar, what is different?
Chem. Soc. Rev. 2013, 42, 7719–7738. [CrossRef] [PubMed]
Wang, G.-W. Mechanochemical organic synthesis. Chem. Soc. Rev. 2013, 42, 7668–7700. [CrossRef] [PubMed]
Waddell, D.C.; Thiel, I.; Clark, T.D.; Marcum, S.T.; Mack, J. Making kinetic and thermodynamic enolates via
solvent-free high speed ball milling. Green Chem. 2010, 12, 209–211. [CrossRef]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).