Synthesis, spectroscopic characterization, and reactivity of water-tolerant Eu3+-based precatalysts

Article abstract of DOI:10.1021/ic500790q

We report the synthesis, spectroscopic characterization, and reactivity of Eu³⁺ in the presence of a new set of chiral ligands designed for the aqueous, enantioselective Mukaiyama aldol reaction. Luminescence and NMR measurements were used to characterize the coordination environments of the Eu³⁺-based precatalysts, and this data is compared with yields and stereoselectivities. In addition to structure-function relationships, we found that, in the presence of excess hexadentate ligands, Eu³⁺ is coordinatively saturated, and subsequently, the reactivity of the precatalysts is reduced. These findings are helpful for the design of new ligands that bind Eu³⁺ without saturating the Eu³⁺ coordination sphere.

Full text of DOI:10.1021/ic500790q

Article
pubs.acs.org/IC
Synthesis, Spectroscopic Characterization, and Reactivity of Water-
Tolerant Eu3+-Based Precatalysts
Derek J. Averill and Matthew J. Allen*
Department of Chemistry, Wayne State University, 5101 Cass Avenue, Detroit, Michigan 48202, United States
S
* Supporting Information
ABSTRACT: We report the synthesis, spectroscopic charac-
terization, and reactivity of Eu3+ in the presence of a new set of
chiral ligands designed for the aqueous, enantioselective
Mukaiyama aldol reaction. Luminescence and NMR measure-
ments were used to characterize the coordination environ-
ments of the Eu3+-based precatalysts, and this data is compared
with yields and stereoselectivities. In addition to structure−
function relationships, we found that, in the presence of excess
hexadentate ligands, Eu3+ is coordinatively saturated, and
subsequently, the reactivity of the precatalysts is reduced.
These findings are helpful for the design of new ligands that
bind Eu3+ without saturating the Eu3+ coordination sphere.
INTRODUCTION
■
Many pharmaceuticals and natural products contain β-hydroxy
carbonyl moieties, making the synthesis of this functional group
an active area of research.1−7 The Mukaiyama aldol reaction is a
Lewis-acid-catalyzed, carbon−carbon bond-forming reaction
that produces β-hydroxy carbonyls (Scheme 1), and this
Scheme 1. Aqueous Mukaiyama Aldol Reaction
Figure 1. C2-symmetric ligands used to prepare Eu3+-based
precatalysts.
that excess ligand slows the reaction. Because of these
reaction requires chiral precatalysts to induce stereoselectivity.
observations, we hypothesized that equilibria exist between
Due to precatalyst instability toward hydrolysis, anhydrous
uncomplexed Eu3+ (racemic precatalyst), ligand-bound Eu3+
(stereoselective precatalyst), and Eu3+ encapsulated by more
than one ligand (deactivated precatalyst) and that an
understanding of the ligand features that influence this
equilibrium would enable rational ligand variations to improve
reactivity and selectivity.
To test our hypothesis, a variety of ligand features that we
suspected could affect reactivity and selectivity were studied.
Ligands 3−6 were synthesized and combined with Eu3+ to
investigate the influence of ligand donor type, chiral center
location, and chiral center bulk on reactivity. Eu3+ was chosen
over other trivalent lanthanides because it can act as an optical
probe to provide feedback regarding its coordination environ-
ment via luminescence-decay and steady-state luminescence
measurements. Additional information about the Eu3+ coordi-
solvents are used commonly,8−10 but recent efforts have
focused on aqueous versions of the enantioselective Mukaiyama
aldol reaction because of the financial and environmental
benefits of using aqueous media.11 Several examples of
enantioselective precatalysts that function in the presence of
water exist: Cu(OTf)2, Pb(OTf)2, and Ln(OTf)3 with crown
ethers, where OTf− is trifluoromethanesulfonate;12−16 Trost-
type semicrowns with Ga(OTf)3;17,18 and Zn(OTf)2 and FeCl2
with pybox-type ligands.19−22
Previously, we reported the synthesis and application of
water-tolerant Eu3+-based precatalysts for aqueous, enantiose-
lective Mukaiyama aldol reactions using ligands 1 and 2 shown
in Figure 1.23,24 The use of ligand 1 with Eu(OTf)3 at a ligand-
to-metal ratio of 2.4:1 resulted in high enantiomeric ratios
(95:5−99:1) for Mukaiyama aldol reaction products but
required long (168 h) reaction times. The excess of ligand
was required to achieve high stereoselectivity, but reaction
times were slower in the presence of excess ligand, suggesting
Received: April 3, 2014
Published: May 29, 2014
© 2014 American Chemical Society
6257
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
HRESIMS (m/z): [M + H]+ calcd for C14H31N2O4, 291.2284; found
291.2289. [α]2D3 +1.48 (c 1.35, CH3OH).
nation environment was obtained from NMR spectroscopy,
and the enantioselectivities and diastereoselectivities of reaction
products formed in the presence of ligands 3−6 with Eu3+ were
determined using chiral columns with high-performance liquid
chromatography (HPLC) for the aqueous, enantioselective,
Mukaiyama aldol reaction. In this Article, we describe the
relationships between ligand structure and precatalyst activity
with regard to efficiency and coordination.
(2S,2′S)-1,1′-(1,7-Dioxa-4,10-diazacyclododecane-4,10-diyl)bis-
(butan-2-ol) (4). To a stirring solution of 1,7-dioxa-4,10-diazacyclo-
dodecane (8) (0.040 g, 0.23 mmol) in methanol (2 mL) was added
(S)-(−)-1,2-epoxybutane (10) (0.51 g, 7.0 mmol) at ambient
temperature. After 12 h, the reaction mixture was filtered and
concentrated under reduced pressure to yield a clear, colorless oil.
Yield 64 mg, 88%. 1H NMR (400 MHz, CDCl3, δ): 5.30 (s, OH, 2H),
3.61−3.37 (m, CH and CH2, 10H), 2.83−2.71 (m, CH2, 4H), 2.59−
2.50 (m, CH2, 4H), 2.48−2.40 (m, CH2, 2H), 2.36−2.26 (m, CH2,
EXPERIMENTAL SECTION
■
2H), 1.51−1.28 (m, CH3CH2, 4H), 0.96 (t, J = 7.3 Hz, CH3, 6H). 13
C
Materials. Commercially available chemicals were used without
further purification. Water was purified using a PURELAB Ultra Mk2
(ELGA) water purification system. (2R,2′R)-Dimethyl-2,2′-(1,7-dioxa-
4,10-diazacyclododecane-4,10-diyl)dipropanoate (1),23 (2R,2′R)-2,2′-
(1,7-dioxa-4,10-diazacyclododecane-4,10-diyl)dipropanoic acid (2),24
and (Z)-trimethyl(1-phenylpropyl-1-enyloxy)-trimethylsilane (7) (Z/
E = 12:1)18 were synthesized according to previously published
procedures.
NMR (100 MHz, CDCl3, δ): 70.0 (CH), 69.5 (CH2), 62.6 (CH2),
55.5 (CH2), 27.5 (CH3CH2), 10.5 (CH3). HRESIMS (m/z): [M +
H]+ calcd for C16H35N2O4, 319.2597; found 319.2596. [α]2D3 +1.28 (c
1.34, CH3OH).
(2R,2′R)-2,2′-(1,7-Dioxa-4,10-diazacyclododecane-4,10-diyl)bis-
(propan-1-ol) (5). To a stirring solution of 1 (0.038 g, 0.11 mmol) in
tetrahydrofuran (1 mL) was added lithium aluminum hydride in
tetrahydrofuran (0.27 mL, 2.0 M, 0.54 mmol) at ambient temperature.
After 20 min of stirring, methanol (5 mL) was added followed by
water (5 mL). The mixture was washed with CH2Cl2 (3 × 15 mL).
Volatiles were removed under reduced pressure; water was added; the
mixture was filtered; and water was removed under reduced pressure
Characterization. Flash chromatography was performed using
silica gel 60, 230−400 mesh.25 Analytical thin-layer chromatography
(TLC) was carried out on TLC plates precoated with silica gel 60 F254
(250 μm layer thickness). TLC visualization was accomplished using a
UV lamp. NMR spectra were obtained in the Lumigen Instrumenta-
tion Center at Wayne State University. 1H NMR and correlation
spectroscopy (COSY) spectra were obtained using a Varian Mercury
400 (400 MHz) spectrometer, a Varian MR400 (400 MHz)
spectrometer, or a Varian 500-S (500 MHz) spectrometer. 13C
NMR, distortionless enhancement by polarization transfer (DEPT)
and heteronuclear multiple quantum coherence (HMQC) spectra
were obtained using a Varian Mercury 400 (100 MHz) spectrometer, a
Varian MR400 (100 MHz) spectrometer, or a Varian 500-S (125
MHz) spectrometer. DEPT, COSY, and HMQC spectra were used to
assign spectral peaks. Data for 1H NMR spectra are reported as
follows: chemical shift (ppm) relative to residual CHCl3 in CDCl3
(7.27 ppm) or CHD2OD in CD3OD (3.31 ppm); multiplicity (“s” =
singlet, “d” = doublet, “t” = triplet, “q” = quartet, “m” = multiplet, and
“brs” = broad singlet); coupling constant, J (Hz); assignment
(italicized elements are responsible for the shifts); and integration.
Data for 13C NMR spectroscopy are reported as ppm relative to
CDCl3 (77.23 ppm) or CD3OD (49.15 ppm) followed by assignment
(italicized elements are responsible for the shifts). High-resolution
electrospray ionization mass spectra (HRESIMS) were obtained on a
Waters Micromass LCT Premier XE mass spectrometer electrospray
time-of-flight high-resolution mass spectrometer. HPLC analyses were
carried out on a Shimadzu instrument equipped with a Chiralpak AS-
H column (Chiral Technologies Inc., 250 mm × 4.6 mm) using a
binary isocratic method (pump A, 2-propanol; pump B, hexanes; flow
rate 1.0 mL/min, isocratic, 90% A, 10% B, λ = 254 nm).
Luminescence-decay measurements26−28 for the determination of
water-coordination number, q, and steady-state luminescence measure-
ments were performed using a HORIBA Jobin Yvon Fluormax-4
spectrofluorometer. Titration mixtures were vortexed using a Fisher
Scientific vortex mixer before each measurement. Centrifugation was
performed using a Centrific Centrifuge at 7000 rotations per minute
(04-978-50, Fisher Scientific). Optical rotations were recorded using
an Autopol III automatic polarimeter.
1
to yield a clear, colorless oil. Yield 28 mg, 91%. H NMR (400 MHz,
CDCl3, δ): 5.01−4.90 (m, OH, 2H), 3.60−3.52 (m, CH2 ring, 4H),
3.45−3.24 (m, CH2 ring and HOCH2, 8H), 2.98−2.87 (m, CH, 2H),
2.83−2.74 (m, CH2, ring, 4H), 2.56−2.47 (m, CH2 ring, 4H), 0.82 (d,
J = 6.5 Hz, CH3, 6H). 13C NMR (100 MHz, CDCl3, δ): 69.9 (CH2
ring), 64.1 (HOCH2), 56.9 (CH), 49.5 (CH2 ring), 9.34 (CH3).
HRESIMS (m/z): [M + H]+ calcd for C14H31N2O4, 291.2284; found
291.2275. [α]2D3 +0.881 (c 0.267, CH3OH).
(2R,2′R)-2,2′-(1,7-Dioxa-4,10-diazacyclododecane-4,10-diyl)-
dipropanamide (6). To a mixture of 1,7-dioxa-4,10-diazacyclodode-
cane (8) (0.050 g, 0.29 mmol) and Cs2CO3 (1.0 g, 3.1 mmol) in
acetonitrile (4 mL) was added (S)-amino-1-oxypropan-2-yl-4-methyl-
benzenesulfonate (11) (0.21 g, 0.86 mmol), and the resulting mixture
was stirred for 72 h at ambient temperature and then centrifuged for
10 min. The supernatant was concentrated under reduced pressure
and washed with diethyl ether (6 × 15 mL) to yield a white solid. Yield
1
67 mg, 74%. H NMR (400 MHz, CDCl3, δ): 8.16 (brs, NH2, 2H),
5.44 (brs, NH2, 2H), 3.70 (t, CH2 J = 11.3 Hz, 4H), 3.44−3.31 (m,
CH2, CH, 6H), 2.94−2.80 (m, CH2, 4H), 2.60−2.48 (m, CH2, 4H),
1.25 (d, CH3, J = 7.5 Hz, 6H). 13C NMR (100 MHz, CDCl3, δ): 177.7
(CONH2), 68.0 (CH2), 59.1 (CH), 49.9 (CH2), 8.6 (CH3). HRESIMS
(m/z): [M + H]+ calcd for C14H29N4O4, 317.2189; found 317.2198.
[α]2D3 +0.349 (c 1.16, CH3OH).
(S)-Amino-1-oxypropan-2-yl-4-methylbenzenesulfonate (11). To
a stirring solution of (S)-2-hydroxypropanamide (1.00 g, 0.0112 mol)
in a mixture of triethylamine and dichloromethane (1:5 v/v, 10 mL)
was added 4-methylbenzene-1-sulfonyl chloride (2.35 g, 0.0123 mol)
at ambient temperature. After 96 h, the crude reaction mixture was
concentrated under reduced pressure and purified using flash
chromatography (1:1 ethyl acetate/hexanes) to afford a fluffy white
solid. Yield 704 mg, 26%. TLC Rf = 0.2 (1:1 ethyl acetate/hexanes).
1H NMR (400 MHz, CD3OD, δ): 7.83 (d, J = 8.3 Hz, CH, 2H), 7.45
(d, J = 8.3 Hz, CH, 2H), 4.82−4.73 (m, CH, 1H), 2.46 (s, CH3, 3H),
1.39 (d, J = 6.9 Hz, CH3, 3H). 13C NMR (100 MHz, CD3OD, δ):
174.2 (CONH2), 147.1, 134.7, 131.3 (CH), 129.3 (CH), 77.4 (CH),
21.8 (CH3), 19.4 (CH3). HRESIMS (m/z): [M + H]+ calcd for
C10H14NO4S, 244.0599; found 244.0644.
Titration Experiment General Procedure. Ligand-to-metal titra-
tions were performed in a cuvette by adding solutions of ligand (11.0
mM in 9:1 EtOH/H2O) to solutions of Eu(OTf)3 (0.834 mM in 9:1
EtOH/H2O). The resulting solutions were vigorously shaken using a
vortex mixer for 20 s and then allowed to stand for 5 min before
acquiring emission spectra. Emission spectra were obtained by exciting
at 395 nm (excitation and emission slit widths were set to 5 nm).
Mukaiyama Aldol Reaction General Procedure. To a mixture of
ligand and Eu(OTf)3 in 9:1 ethanol/water (0.4 mL) that was stirred
Synthesis and Characterization of Chiral Ligands 3−6.
(2S,2′S)-1,1′-(1,7-Dioxa-4,10-diazacyclododecane-4,10-diyl)bis-
(propan-2-ol) (3). To a stirring solution of 1,7-dioxa-4,10-
diazacyclododecane (8) (0.040 g, 0.23 mmol) in methanol (2 mL)
was added (S)-(−)-propylene oxide (9) (0.41 g, 7.0 mmol) at ambient
temperature. After 12 h, the reaction mixture was filtered and
concentrated under reduced pressure to yield a clear, colorless oil.
1
Yield 61 mg, 91%. H NMR (400 MHz, CDCl3, δ): 5.33 (brs, OH,
2H), 3.78−3.67 (m, CH, 2H), 3.58−3.47 (m, CH2, ring, 4H), 3.43−
3.33 (m, CH2 ring, 4H), 2.79−2.66 (m, CH2 ring, 4H), 2.60−2.50 (m,
CH2 ring, 4H), 2.44−2.35 (m, CH2 arm, 2H), 2.31−2.19 (m, CH2
arm, 2H), 1.05 (d, CH3, 6H). 13C NMR (100 MHz, CDCl3, δ): 69.5
(CH2 ring), 64.3 (CH2 arm) 64.2 (CH), 55.3 (CH2 ring), 19.8 (CH3).
6258
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
Luminescence Measurements for the Study of Eu3+
Coordination Environment. Luminescence measurements
were performed to study the interaction of Eu3+ with ligands
1−6. Eu3+ is a useful optical probe for studying coordination
environments because of its coordination-environment sensi-
tive luminescence-lifetime26−30 and the hypersensitivity of the
for 5 min at ambient temperature and then cooled to −25 °C was
added benzaldehyde (12) (3.2 μL, 32 μmol, 1.0 equiv) followed by
(Z)-trimethyl-(1-phenylpropyl-1-enyloxy)trimethylsilane (7) (11 μL,
48 μmol, 1.5 equiv). The mixture was stirred for 168 h at −25 °C. Silyl
enol ether 7 hydrolyzes throughout the course of the reaction as
observed by TLC, but it is not completely consumed until 168 h. The
mixture was then purified using silica gel chromatography (1:9 ethyl
acetate/hexanes, Rf = 0.2), and volatiles were removed under reduced
pressure. The enantiomeric and diastereomeric ratios of the products
were determined by HPLC.8
7
5D0 → F2 transition (emission at ∼616 nm) to perturbations
of symmetry.31−33 Consequently, luminescence-lifetime meas-
urements were used to determine water-coordination numbers
(q), and steady-state luminescence measurements were used to
monitor changes in the crystal field of Eu3+during ligand-to-
metal titrations.
RESULTS AND DISCUSSION
■
Ligands 1−6. With ligands 1−6 and Eu3+, we were able to
study the influence of ligand structure on reactivity, selectivity,
and complex formation. The functional groups that bind to the
metal were changed as part of this strategy to incorporate esters
(1), carboxylic acids (likely carboxylates under reaction
conditions) (2), alcohols (3, 4, and 5), and amides (6).
These functional groups were chosen because esters, carboxylic
acids, alcohols, and amides have different Lewis basicities that
affect the ability of ligands to bind Eu3+. In addition to
functional groups, the location and size of the chiral center on
the ligand side arms were investigated with ligands 3, 4, and 5
to study the influence of structural features on reactivity and
selectivity.
To calculate q for Eu3+ in each mixture, we used eq 1, where
1.2, 0.23, 0.44, and 0.075 are empirically derived constants. The
value of 1.2 is independent of the solvent system, and the
correction factor of 0.23 was used because our solvent systems
were composed of EtOH/H2O (9:1) and not water.34 The
terms nOH and nNH are the number of non-water-based inner-
sphere alcohol (O−H) or amide (N−H) oscillators,
respectively.29,34 The terms τ
and τ
are the measured
−1
−1
H2O
D2O
luminescence-decay rates of Eu3+ in EtOH/H2O and EtOD/
D2O, respectively (decay rates are in the Supporting
Information). Water-coordination numbers of Eu3+ were
measured in mixtures of EtOH/H2O 9:1 (v/v) with ligand-
to-Eu3+ ratios of 1:1, 2:1, and 6:1 (Table 1). This solvent
The syntheses of ligands 3−6 are shown in Scheme 2. Briefly,
3 and 4 were synthesized by the ring-opening of epoxides with
Table 1. Water-Coordination Numbers for Eu3+ with
Ligands 1−6
Scheme 2. Syntheses of Ligands 3−6
a,
d
,
g
a,
e,
g
b,
d
,
g
b,
e,
g
c
,
d
,
g
c,e,g
ligand
q
q
q
q
q
q
f
1
3.5
2.2
2.2
1.9
2.1
3.8
3.5
2.2
1.1
0.8
1.0
3.4
2.1
0.8
2.0
1.8
2.1
3.0
2.1
0.8
1.0
0.8
1.0
2.6
1.4
0.0
1.1
nd
1.4
0.0
0.0
nd
f
2
3
4
5
6
0.5
1.7
0.0
1.3
a
b
c
Ligand-to-metal ratio of 1:1. Ligand-to-metal ratio of 2:1. Ligand-
d
to-metal ratio of 6:1. Calculated for complexes with Eu3+
e
coordination by one ligand. Calculated for complexes with Eu3+
f
coordination by two ligands. Ligands 1 and 2 do not have chelator-
based inner-sphere O−H or N−H oscillators; therefore, qd = qe. nd =
g
not determined. The error associated with water-coordination
number determination is 0.1 water molecules.29
composition was chosen because it is commonly used for
lanthanide-catalyzed Mukaiyama aldol reactions.13,23,24 Also,
ligand-to-Eu3+ ratios of 1:1, 2:1, and 6:1 were chosen to
determine if more than 1 equiv of ligand can bind Eu3+. Due to
the possibility of more than 1 equiv of ligand being coordinated
to Eu3+, we calculated q values for both 1 and 2 equiv of ligand
bound to Eu3+ (we did not calculate q values for greater than 2
equiv of ligand bound to Eu3+ because this type of binding is
unlikely to occur based on steric interactions and entropy)
assuming that all N−H or O−H oscillators on each ligand
would be in the inner sphere of the metal ion. This assumption
allowed us to obtain maximum and minimum q values with
ratios of 1:1 ligand-to-Eu3+ (maximum) and 2:1 ligand-to-Eu3+
(minimum). For example, maximum and minimum q values for
Eu3+ with ligand 6 were calculated for 1 equiv of ligand (2
amides or 4 N−H oscillators) for a maximum q value and for 2
equiv of ligand (4 amides or 8 N−H oscillators) for a minimum
q value.
macrocycle 8 in MeOH. Ligand 5 was prepared by reducing 1
with lithium aluminum hydride. The reduction of 1 to produce
5 was chosen instead of preparing a chiral-alcohol side arm to
react with 8 because the stereoisomers of starting material 1 can
be readily purified by HPLC. Ligand 6 was synthesized by
functionalizing 8 using (S)-amino-1-oxypropan-2-yl-4-methyl-
benzenesulfonate (11). Ligands 3−6 were all synthesized in
good yields (74−91%) in one step from commercially available
or reported starting materials.
6259
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
ratio allows the determination of end points for ligand-to-Eu3+
−1
q = 1.2[(τ
− τD−1O) − 0.23 − 0.44nOH − 0.075nNH]
H2O
2
binding stoichiometries for ligands 1−6 with Eu3+. To monitor
(1)
5
changes in luminescence intensities, we compared the D0 →
Eu3+ usually has a coordination number between 8 and 9;
therefore, because q determinations represent the average of all
species in solution, water-coordination numbers greater than 3
in the presence of hexadentate ligands suggest the presence of
uncomplexed Eu3+ (Figure 2) or incomplete coordination by
5
7
7F1 and D0 → F2 transitions of Eu3+. Figure 3 illustrates the
Figure 2. Representation of unchelated Eu3+ and a Eu3+-containing
complex of 3 with q values of 9 and 3, respectively (counteranions
have been omitted for clarity). One complex is shown on the right of
the equilibrium to demonstrate the point of the figure, but multiple
species are likely in the actual solution including sandwich-type
structures that involve incomplete chelation from more than one
multidentate ligand.
Figure 3. Normalized emission spectra (λex = 395 nm) of Eu(OTf)3 in
EtOH/H2O 1:9 with (dotted line) and without (solid line) 0.6 equiv
of ligand 3. The bands at 577, 591, and 616 nm arise from the 5D0 →
7F0, 5D0 → 7F1, and 5D0 → 7F2 transitions, respectively. Similar spectra
were observed for the other ligands in this Article. The sensitivity of
the ligands. By measuring the water-coordination numbers of
Eu3+ at different ligand-to-metal ratios, we were able to monitor
the complexation of Eu3+ with ligands 1−6. Mixtures of Eu3+
with ligands 1 and 6 at ligand-to-metal ratios of 1:1 have Eu3+
water-coordination numbers greater than 3, indicating incom-
plete complexation of Eu3+. This observation suggests that
ligands 1 and 6 form labile complexes with Eu3+ under these
conditions. Ligands 2−5 with Eu3+ at ligand-to-metal ratios of
1:1 have q values between 0.8 and 2.2. These values of q suggest
that ligands 2−5 (ligands with side arms that have functional
groups with the ability to become deprotonated) chelate Eu3+
more strongly than 1 and 6 (ligands with side arms that have
functional groups that lack the ability to become deproto-
nated).
Average values of q are lower at ligand-to-metal ratios of 2:1
compared to 1:1 mixtures prepared from the same ligands. This
observation is a demonstration of Le Chatelier’s principle
where the equilibrium system shown in Figure 2 can be driven
to the right (increased chelation of Eu3+) in the presence of
excess ligand. Water-coordination numbers were lower than 3
at ligand-to-Eu3+ ratios of 6:1 in the presence of ligands 1−3, 5,
and 6 indicating that Eu3+ can be bound by more than one
ligand at a time. Mixtures of 4 and Eu3+ at a ligand-to-Eu3+ ratio
of 6:1 became turbid in EtOD/D2O; therefore, we did not
calculate q for Eu3+ with 4 at this ratio. Interestingly, Eu3+ in the
presence of 6 equiv of ligand 2 had a q value of 0.0; the most
reasonable explanation for such a low q is complete
encapsulation of Eu3+ by 2 or more equiv of ligand 2. These
observations of q at different ligand-to-Eu3+ ratios support our
hypothesis that an equilibrium system with multiple species
occurs with ligands 1−6 and Eu3+. We suspect that there are
multiple equilibria in solution but that the predominant species
are unbound Eu3+, a single multidentate ligand coordinated to
Eu3+, and two multidentate ligands coordinated to Eu3+ (Eu3+ is
potentially moving between the macrocycles of each ligand
while being coordinated by the appendages of both ligands).
To gain a better understanding of the Eu3+ equilibrium
system, we titrated solutions of ligand into solutions containing
Eu3+ and monitored the changes in coordination by steady-state
luminescence measurements. Monitoring changes in steady-
state luminescence intensity as a function of ligand-to-Eu3+
5
7
5
7
the D0 → F2 transition combined with the less sensitive D0 → F1
transition makes Eu3+ a useful ratiometric sensor for concentration-
independent sensing of changes in coordination [(5D0 → 7F2)/(5D0 →
7F1) emission intensity ratios are independent of concentration]. The
intensity ratio increase is expected upon ligand coordination to Eu3+
because the 5D0
→
7F2 emission intensity is highly sensitive to
perturbations of the Eu3+ crystal field by coordinated ligands.32,33,35,36
sensitivity of the 5D0
→
7F2 transition to changes in
coordination environment (with and without 0.6 equiv of
multidentate ligand 3). The emission spectrum of Eu3+ in the
presence of ligand 3 was normalized to the magnetic-dipole
7
(5D0 → F1) band at 591 nm (relatively insensitive to changes
in coordination environment) for visual comparison of the 5D0
7
→ F2 emission intensities.
Compared to Eu(OTf)3 in the absence of hexadentate
ligands, the intensity of the 5D0 → 7F2 transition relative to the
5D0
→
7F1 transition increased in the presence of every
hexadentate ligand in this study. Changes in Eu3+ coordination
environments were monitored by plotting the emission
intensity ratios versus ligand-to-Eu3+ ratios (Figure 4).
Ligand-to-Eu3+ titrations with 1−6 caused an increase in the
5
7
5
7
intensity ratio of the D0 → F2 to D0 → F1 transitions of
Eu3+ between 1 and 3 equiv of ligand (indicating change in Eu3+
coordination). Ester functionalized ligand 1 does not have
titration end points in the range studied, indicating weak
interactions between ligand and Eu3+. Lastly, the titration of
ligand 6 and Eu3+ has an apparent end point near 6 equiv of
ligand, which indicated weak interactions between ligand and
Eu3+. On the basis of the titration data in Figure 4, ligand donor
type affects coordination more than chiral center location and
size. The relative strength of interactions with Eu3+ for these
ligands is carboxylate > alcohol > amide > ester. Alcohol-based
ligands might appear out of order, but if they are at least
partially deprotonated when coordinated to Eu3+, then they
would be expected to bind more tightly than amides or esters.
This stability trend is in agreement with other studies of
macrocyclic ligands with Eu3+.37−40
1H NMR Characterization of Ligands with Eu3+. In
addition to luminescence measurements, NMR spectroscopy
6260
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
Figure 4. Emission intensity ratios of Eu3+ (616 nm/591 nm) versus
equiv of ligand for Eu3+ with 1 (Δ), 2 ( ), 3 ( ), 4 ( ), 5 ( ), and 6
⧫
◊
▲
○
□
( ). Increases in magnitude of the emission intensity quotient (616
nm/591 nm) arise from increases in crystal field splitting of Eu3+.33
was used to investigate the interactions between ligands and
Eu3+. By monitoring changes in the 1H NMR spectra of ligands
in the presence of Eu3+ at different ligand-to-Eu3+ ratios, we
found that at least three different species can exist in solution.
1
Figure 5 contains H NMR spectra of ligand 3 that were
acquired at ligand-to-Eu3+ ratios of 1:0, 4:1, and 0.5:1. These
ligand-to-Eu3+ ratios were chosen because they allowed us to
investigate 3 without Eu3+ (1:0), with an excess of ligand (4:1),
and with an excess of Eu3+ (0.5:1). The NMR spectra were
different at each ratio, indicating the presence of different ligand
environments. With a ligand-to-Eu3+ ratio of 4:1 and an
acquisition temperature of −40 °C, we observed new signals
downfield from the signals arising from only ligand. We
attribute the downfield signals to a Eu3+−Ln (n > 1) species
because an excess of ligand increases the probability for
multiple ligands to bind Eu3+. The NMR spectral evidence for
the presence of a Eu3+−Ln (n > 1) species is also consistent
with q measurements for Eu3+ in the presence of excess 3.
Although these experiments were performed at different
temperatures that likely influence the rates associated with
ligand binding (−40 °C for NMR spectroscopy to avoid
coalescence and ambient temperature for luminescence
measurements because of instrument capabilities), the con-
clusions with respect to precatalyst structure are consistent
because the luminescence values represent an average of all
species in solution, and this value changes as a function of the
stoichiometry. At a ligand-to-metal ratio of 0.5:1, we observed
signals upfield relative to signals arising from the ligand. We
attribute the upfield signals to a 1:1 Eu3+-to-ligand complex
because an excess of metal will shift the equilibrium toward 1:1
ligand-to-metal complexes and unchelated Eu3+, or possibly a
Eu3+n−L (n ≥ 1) species. From the 1H NMR spectra of ligands
1−6 with Eu3+ at different ligand-to-metal ratios, we found that
at least three different species of Eu3+ exist depending on the
ligand-to-Eu3+ ratio, and these findings corroborate our
luminescence observations.
1
1
Figure 5. (A) H NMR spectrum of 3 in 9:1 EtOD/D2O. (B) H
NMR spectrum of 3 in 9:1 EtOD/D2O at −40 °C with 0.25 equiv of
Eu(OTf)3. Arrows point to signals observed in the presence of excess
ligand (the temperature of −40 °C was required to resolve the signals
between 4 and 2 ppm). We attribute these new signals to a Eu3+−Ln (n
1
> 1) species. (C) H NMR spectrum of 3 in 9:1 EtOD/D2O with 2
equiv of Eu(OTf)3. Arrows point to signals observed in the presence of
excess Eu3+. We attribute the new upfield signals to a Eu3+n−L (n ≥ 1)
species.
Testing of Precatalyst Reactivity and Selectivity in the
Aqueous Enantioselective Mukaiyama Aldol Reaction.
We tested the reactivity and selectivity of mixtures of Eu3+ with
ligands 1−6 in the aqueous Mukaiyama aldol reaction shown in
Scheme 3. We chose the reaction in Scheme 3 because it has
been previously used for testing of precatalysts for Mukaiyama
aldol reactions;41−44 therefore, our results can be compared to
results obtained using other precatalysts. Because reactivity and
6261
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
Scheme 3. Aqueous Mukaiyama Aldol Reaction Performed
with Ligands 1−6
selectivity of Mukaiyama aldol reactions depend on ligand-to-
metal ratios,14,23,24 the reactions in this study were performed at
ligand-to-Eu3+ ratios of 1.2:1 and 2.4:1 (Table 2) to enable
comparison to published values.18,23
Table 2. Aqueous Mukaiyama Aldol Yields and Selectivities
Catalyzed by Eu(OTf)3 with Ligands 1−6
Figure 6. (left) Proposed selective nucleophilic attack at the si face of
benzaldehyde (high selectivity and reactivity is observed). (middle)
Precatalysts are nonselective when R is OH or NH2. (right) Ligands
3−5 with Eu3+ are unable to block nucleophilic attack at either face of
benzaldehyde likely due to the stereochemistry at the position of R2
causing less blocking than with ligand 1. Coordinated water has been
omitted for clarity; additionally, ligand 2 has been drawn in the
carboxylic acid form although it likely exists in the deprotonated state
while coordinated to Eu3+ in aqueous media.
1
2
3
4
5
6
a
,
b
c
yield (%)
96
8
0
17
12
5
0
20
6
b
,d
e
yield (%)
92
0
0
a,
f
c
dr (syn:anti)
2.1:1
0.72:1
nd
1.3:1
nd
1.1:1
nd
1.1:1
nd
0.93:1
0.55:1
0.78:1
0.66:1
d
,
f
e
dr (syn:anti)
32:1
a,
f
c
er (syn)
2:1
1.03:1
nd
1.2:1
1:1
nd
1:1
d
,f
e
er (syn)
96:1
nd
nd
a
b
c
24 mol % ligand, 20 mol % Eu(OTf)3. Isolated yield. Reference 18.
d
e
f
48 mol % ligand, 20 mol % Eu(OTf)3. Reference 23. Determined by
chiral HPLC analysis; nd = not determined.
Ligand 1 with Eu3+ was previously reported to give 96% and
92% yields at ligand-to-Eu3+ ratios of 1.2:1 and 2.4:1 with
respective diastereomeric ratios of 2.1:1 and 32:1 and
enantiomeric ratios of 2:1 and 96:1.23,24 In this study, ligand
2 with Eu3+ gave an 8% yield at a ligand-to-Eu3+ ratio of 1.2:1
with a diastereomeric ratio of 0.72:1 and an enantiomeric ratio
of 1.03:1 with no products being formed at the 2.4:1 ligand-to-
Eu3+ ratio. We hypothesize that, in the presence of excess 2, the
inner-coordination sphere of Eu3+ nears saturation and cannot
act as a precatalyst for Mukaiyama aldol reactions, which is
consistent with observed q numbers for Eu3+ in the presence of
excess 2. Chiral alcohol ligands 3−5 with Eu3+ gave low yields
(5−17%) and low diasteriomeric (1.1:1−1.3:1) and enantio-
meric (1:1−1.2:1) ratios at ligand-to-Eu3+ ratios of 1.2:1. The
low selectivity is likely due to the lack of steric bulk on the
ligand side arms. A lack of steric bulk near the coordinated
atom might not allow the ligand to block nucleophilic attack at
either the re or si face of the aldehyde from the incoming silyl
enol ether (Figure 6, left). This blocking is the suspected
mechanism that imparts selectivity.24 No products were
observed for mixtures of 2.4:1 ligand-to-Eu3+ with ligands 3−
5, and this lack of reactivity is likely due to an equilibrium
system that involves displacement of inner-sphere water by a
second equivalent of ligands 3−5 (Figure 7). Saturation of Eu3+
by more than 1 equiv of ligand could inhibit the reaction by
preventing aldehyde coordination to Eu3+ to become activated
for nucleophilic attack. The q data for Eu3+ with ligands 3−5
support our hypothesis that excess ligand prevents the aldehyde
from becoming activated for nucleophilic attack.
Figure 7. Proposed equilibria involving multiple Eu3+ species with
reactivity summarized under the species. These equilibria are based on
luminescence-decay, steady-state luminescence measurements, and 1H
NMR data with reactivity and selectivity from Mukaiyama aldol
reaction results.
reactivity of Eu3+ in the presence of 6 is likely explained by the
resonance structure of amides that place electron density on the
carbonyl oxygen, lowering the Lewis acidity of Eu3+. Because
low selectivity was observed with Eu(OTf)3 in the presence of
6, we hypothesize that the steric bulk of the primary amide
ligand is not large enough to selectively block the nucleophilic
attack by the incoming silyl enol ether (Figure 6).
From Mukaiyama aldol reactions, we found that substitution
at the carbonyl carbon and the carbon adjacent to the
macrocycle nitrogen is required for high stereoselectivity in
the products (Figure 7). Finally, we found that carboxylic acid,
alcohol, and amide functionalized ligands (2, 3−5, and 6,
respectively) cause a reduction in reactivity compared to 1. We
hypothesize that ligand 1 likely donates the lowest amount of
electron density to Eu3+ out of all of the ligands studied,
resulting in the strongest Lewis acid precatalyst (most reactive)
of the series when combined with Eu3+. This result suggests
that improvements to reactivity would need to include donors
that are weaker bases than esters.
Amide functionalized ligand 6 with Eu(OTf)3 at ligand-to-
Eu3+ ratios of 1.2:1 and 2.4:1 gave low yields (20 and 6%,
respectively) and low diasteriomeric and enantiomeric
selectivity. It is likely that excess 6 slowed the Mukaiyama
aldol reaction by decreasing the Lewis acidity at the metal
center. These findings suggest that amide side arms deactivate
the metal center compared to ester side arms. The reduced
CONCLUSION
■
We synthesized a series of new, chiral, Eu3+-binding ligands. By
spectroscopically characterizing the ligands and Eu3+ inter-
actions using luminescence-decay, steady-state luminescence,
and NMR spectroscopies, we found evidence suggesting that
saturation of the Eu3+ coordination sphere by multiple
6262
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263

Inorganic Chemistry
Article
(16) Hamada, T.; Manabe, K.; Ishikawa, S.; Nagayama, S.; Shiro, M.;
Kobayashi, S. J. Am. Chem. Soc. 2003, 125, 2989−2996.
(17) Li, H.-J.; Tian, H.-Y.; Chen, Y.-J.; Wang, D.; Li, C.-J. Chem.
Commun. 2002, 2994−2995.
(18) Li, H.-J.; Tian, H.-Y.; Wu, Y.-C.; Chen, Y.-J.; Liu, L.; Wang, D.;
Li, C.-J. Adv. Synth. Catal. 2005, 347, 1247−1256.
(19) Mlynarski, J.; Jankowska, J. Adv. Synth. Catal. 2005, 347, 521−
525.
(20) Jankowska, J.; Mlynarski, J. J. Org. Chem. 2006, 71, 1317−1321.
(21) Jankowska, J.; Paradowska, J.; Rakiel, B.; Mlynarski, J. J. Org.
Chem. 2007, 72, 2228−2231.
(22) Ollevier, T.; Plancq, B. Chem. Commun. 2012, 48, 2289−2291.
(23) Mei, Y.; Averill, D. J.; Allen, M. J. J. Org. Chem. 2012, 77, 5624−
5632.
hexadentate ligands occurs in the presence of excess ligand. A
comparison of ligand functional groups and chiral center
location and bulk provided insight into the factors that affect
reactivity and selectivity of water-tolerant chiral precatalysts.
We found that the reactivity of Eu3+-based precatalysts depends
on the type of coordinating functional groups on the ligands,
and that the selectivity of Eu3+-based precatalysts depends on
chiral center location and functional group size. The selectivity,
reactivity, and spectroscopic characterizations described here
contribute to the understanding of lanthanide-based precata-
lysts, and we expect these findings to be useful in the design of
new precatalysts with high activity and selectivity for
asymmetric carbon−carbon bond-forming reactions in aqueous
media.
(24) Mei, Y.; Dissanayake, P.; Allen, M. J. J. Am. Chem. Soc. 2010,
132, 12871−12873.
(25) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923−
ASSOCIATED CONTENT
2925.
■
S
(26) Averill, D. J.; Dissanayake, P.; Allen, M. J. Molecules 2012, 17,
2073−2081.
(27) Dissanayake, P.; Allen, M. J. J. Am. Chem. Soc. 2009, 131, 6342−
6343.
* Supporting Information
1H and 13C NMR spectra for ligands 3−6, H and 13C NMR
1
spectra for 11, chiral HPLC chromatograms for Mukaiyama
aldol products, tabulated emission intensity ratios from
(28) Horrocks, W. D., Jr.; Sudnick, D. R. J. Am. Chem. Soc. 1979, 101,
334−340.
1
titrations, tabulated luminescence-decay rates, and H NMR
spectra of ligands 1−6 with and without Eu3+. This material is
(29) Supkowski, R. M.; Horrocks, W. D., Jr. Inorg. Chim. Acta 2002,
340, 44−48.
available free of charge via the Internet at http://pubs.acs.org.
(30) Lin, Z.; Allen, M. J. Dyes Pigm. 2014, DOI: 10.1016/
j.dyepig.2014.03.020.
AUTHOR INFORMATION
■
(31) Brittain, H. G. Inorg. Chem. 1980, 19, 640−643.
(32) George, M. R.; Golden, C. A.; Grossel, M. C.; Curry, R. J. Inorg.
Chem. 2006, 45, 1739−1744.
(33) Bemquerer, M. P.; Bloch, C., Jr.; Brito, H. F.; Teotino, E. E. S.;
Miranda, M. T. M. J. Inorg. Biochem. 2002, 91, 363−370.
(34) Dissanayake, P.; Mei, Y.; Allen, M. J. ACS Catal. 2011, 1, 1203−
1212.
(35) Binnemans, K.; Herck, K. V.; Gorller-Warland, C. Chem. Phys.
Lett. 1997, 266, 297−302.
(36) Billard, I.; Georg, S. Helv. Chim. Acta 2009, 92, 2227−2237.
Corresponding Author
*E-mail: mallen@chem.wayne.edu. Phone: (313) 577-2070.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by a CAREER Award from the
National Science Foundation (CHE-0955000). D.J.A. was
supported by a Schaap Fellowship.
(37) Baranyai, Z.; Ban
́
yai, I.; Brucher, E.; Kiral
́
y, R.; Terreno, E. Eur.
̈
J. Inorg. Chem. 2007, 3639−3645.
́
(38) Tot
4070−4076.
́
h, E.; Brucher, E.; Laz
́
ar
́
, I.; Tot
́
h, I. Inorg. Chem. 1994, 33,
̈
REFERENCES
■
(1) Xuan, R.; Oh, H.-S.; Lee, Y.; Kang, H.-Y. J. Org. Chem. 2008, 73,
1456−1461.
(39) Wu, S. L.; Horrocks, W. D., Jr. Inorg. Chem. 1995, 34, 3724−
3732.
(2) Park, J. W.; Oh, H.-S.; Jung, W. S.; Park, S. R.; Han, A. R.; Ban,
Y.-H.; Kim, E. J.; Kang, H.-Y.; Yoon, Y. J. Chem. Commun. 2008,
5782−5784.
(40) Chin, A. K. O.; Morrow, J. R.; Lake, C. H.; Churchill, M. R.
Inorg. Chem. 1994, 33, 656−664.
(41) Kobayashi, S.; Nagayama, S.; Busujima, T. J. Am. Chem. Soc.
1998, 120, 8287−8288.
(3) Tosin, M.; Smith, L.; Leadlay, P. F. Angew. Chem., Int. Ed. 2011,
50, 11930−11933.
(42) Kobayashi, S.; Wakabayashi, T.; Nagayama, S.; Oyamada, H.
Tetrahedron Lett. 1997, 38, 4559−4562.
(4) Pearson, A. J.; Panda, S. Org. Lett. 2011, 13, 5548−5551.
(5) Lee, Y.; Choi, J. Y.; Fu, H.; Harvey, C.; Ravindran, S.; Roush, W.
R.; Boothroyd, J. C.; Khosla, C. J. Med. Chem. 2011, 54, 2792−2804.
(6) Oh, H.-S.; Kang, H.-Y. J. Org. Chem. 2012, 77, 1125−1130.
(7) Morar, M.; Pengelly, K.; Koteva, K.; Wright, G. D. Biochemistry
2012, 51, 1740−1751.
(8) Denmark, S. E.; Wong, K.-T.; Stavenger, R. A. J. Am. Chem. Soc.
1997, 119, 2333−2334.
(9) Denmark, S. E.; Heemstra, J. R., Jr. Org. Lett. 2003, 5, 2303−
2306.
(43) Kobayashi, S.; Hachiya, I. Tetrahedron Lett. 1992, 33, 1625−
1628.
(44) Nomiya, K.; Ohta, K.; Sakai, Y.; Hosoya, T.-a.; Ohtake, A.;
Takakura, A.; Matsunaga, S. Bull. Chem. Soc. Jpn. 2013, 86, 800−812.
(10) Desimoni, G.; Faita, G.; Piccinini, F.; Toscanini, M. Eur. J. Org.
Chem. 2006, 5228−5230.
(11) Kobayashi, S.; Hachiya, I. J. Org. Chem. 1994, 59, 3590−3596.
(12) Kobayashi, S.; Nagayama, S.; Busujima, T. Chem. Lett. 1999,
71−72.
(13) Nagayama, S.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122,
11531−11532.
(14) Kobayashi, S.; Hamada, T.; Nagayama, S.; Manabe, K. Org. Lett.
2001, 3, 165−167.
(15) Kobayashi, S.; Nagayama, S.; Busujima, T. Tetrahedron 1999, 55,
8739−8746.
6263
dx.doi.org/10.1021/ic500790q | Inorg. Chem. 2014, 53, 6257−6263