Inorganic Chemistry
Article
(m, 16H), 5.35 (m, 4H), 4.28−3.12 (m), 2.18 (bs), 2.09 (s), 1.86−1.39
(m), 1.34 (m). 13C NMR (125 MHz, DMSO-d6): δC 159.5, 159.3,
156.1, 153.3, 153.2, 153.1, 152.9, 143.2, 137.6, 131.4, 129.1, 128.1,
127.9, 126.2, 124.6, 116.1, 116.0, 115.8, 115.7, 115.6, 115.2, 113.1,
113.0, 70.8, 70.6, 70.5, 70.4, 70.2, 7.1, 69.68, 69.65, 69.58, 69.4, 69.3,
69.2, 69.1, 68.7, 68.5, 68.1, 68.0, 67.8, 67.7, 67.6, 67.64, 67.57, 67.4,
30.9, 30.2, 29.0, 28.63, 28.58, 28.52, 25.7, 25.6, 25.5, 25.3. 19F NMR
(471 MHz, DMSO-d6): δF −69.4, −70.9. LRMS-ESI calcd for
C234H238Cu2FeN14O40: m/z = 1016.4 [M]4+, 1052.9 [M + H +
PF6]4+; found 1017.2 [M]4+, 1052.7 [M + H + PF6]4+. MALDI-TOF
calcd for C234H238Cu2FeN14O40: m/z = 3950.7 [M − Fe − Cu]+, 2006.8
[M − Fe]2+; found 3950.2 [M − Fe − Cu]+, 2006.3 [M − Fe]2+.
Synthesis of [4]C. A solution of [4]C-M (0.0051 g, 0.00113 mmol,
1 equiv) was prepared in 10 mL of MeCN. To this was added 10 mL of
saturated sodium ethylenediaminetetraacetic acid (EDTA) dibasic in
H2O, 2 mL of saturated NH4OH, and 2 mL of 30% H2O2. The reaction
mixture was heated to 80 °C for 15 min, at which point the color of the
solution changed from red to colorless. The mixture was diluted with 50
mL of H2O and extracted with 3 × 25 mL of CHCl3. The organic layer
was dried over Na2SO4 and was filtered. The solvent was removed via a
rotary evaporator to afford an orange film, which was washed with 50
mL of MeOH and 50 mL of MeCN to remove any remaining metalated
species, leaving the product behind as a faint orange-pink film (0.0013
g, 30%). 1H NMR (500 MHz, DMSO-d6): δH 8.55−8.34 (m), 8.31 (s),
8.27−8.12 (m), 8.04−7.77 (m), 7.57−7.28 (m), 7.15 (d, J = 8.5 Hz),
7.04 (bs), 7.01 (d, J = 7.9 Hz), 6.93−6.57 (m), 6.08−4.40 (m), 4.31−
3.33 (m), 1.35 (s), 1.24 (s). 13C NMR (125 MHz, DMSO-d6): δC
167.0, 160.1, 159.8, 159.7, 159.5, 154.9, 154.8, 154.5, 152.6, 152.5,
152.4, 147.8, 147.7, 145.3, 137.1, 137.0, 136.9, 131.7, 131.6, 131.4,
131.3, 131.2, 131.1, 128.7, 128.5, 127.3, 125.7, 121.3, 119.1, 118.9,
115.6, 115.3, 115.2, 115.1, 114.8, 114.7, 114.6, 114.5, 114.3, 69.9, 69.0,
68.8, 67.4, 38.1, 31.3, 29.8, 29.0, 28.7, 28.5, 28.4, 28.1, 25.1, 25.0, 23.2,
22.4, 14.9, 10.8. LRMS-ESI calcd for C234H238N14O40: m/z = 649.1 [M
+ 6H]6+, 660.5 [M + Na(HCOO) + 6H]6+, 778.8 [M + 5H]5+, 792.4 [M
+ Na(HCOO) + 5H]5+, 972.9 [M + 4H]4+, 989.9 [M + Na(HCOO) +
4H]4+, 1006.9 [M + 2Na(HCOO) + 4H]4+; found m/z = 648.4 [M +
6H]6+, 659.3 [M + Na(HCOO) + 6H]6+, 778.0 [M + 5H]5+, 790.5 [M +
Na(HCOO) + 5H]5+, 972.4 [M + 4H]4+, 987.4 [M + Na(HCOO) +
4H]4+, 1003.7 [M + 2Na(HCOO) + 4H]4+. MALDI-TOF calcd for
C234H238N14O40: m/z = 3885.7 [M + H]+, 1942.9 [M + 2H]2+, 1964.8
[M + 2Na]2+; found 3885.7 [M + H]+, 1944.2 [M + 2H]2+, 1966.2 [M +
2Na]2+.
washes. After the organic layer was dried with Na2SO4, Fe-
(OTPM)2 was obtained in 89% yield (see Scheme S20 for
further details). To prove that two simultaneous RCM steps
were possible on the ternary complex to afford a [2]catenate
([2]C-M), Fe-(OTPM)2 was dissolved in CHCl3 with 0.2 equiv
of second-generation Grubbs’ catalyst, and the mixture was
heated at 45 °C for 16 h while stirring under N2 (Figure 2a,
Scheme S21). Proton nuclear magnetic resonance (1H NMR)
analysis of an aliquot revealed some starting material remained,
so an additional 0.2 equiv of the catalyst was added, and the
reaction mixture stirred for another 4 h. To obtain pure [2]C-M,
column chromatography was performed on basic alumina using
a gradient mobile phase (CHCl3 to 5% MeOH/CHCl3).
Comparison of the 1H NMR spectra of OTPM, Fe-(OTPM)2,
and [2]C-M (Figure 2b) revealed changes in key diagnostic
proton resonances associated with the terpy ligand and the
terminal olefin of OTPM. Specifically, metalation of OTPM
with Fe2+ to form Fe-(OTPM)2 resulted in significant shifting of
the terpy proton resonances (a−e), while the chemical shifts of
the phen proton resonances (1−5) remained relatively
unchanged (Figures 2b and S4), supporting the claim of
orthogonal metal ion selectivity of OTPM. However, peak
broadening and new splitting patterns were observed for the
phen protons, which could be attributed to a slight change in
chemical environment afforded by the rigidification of the
ligands in the dimeric complex Fe-(OTPM)2, relative to free
OTPM. It is also possible that, in solution, the unbound phen
ligands in Fe-(OTPM)2 are in close proximity to one another
further differentiating their chemical environment, resulting in
peak broadening and new splitting patterns. Similarly, the olefin
proton resonances are no longer in the same chemical
environment, likely due to the difference in rigidity of the
metalated terpy half of the macrocycle compared to the
unmetalated phen half. After RCM to form [2]C-M, some
broadening was observed, and new splitting patterns emerged
that can be attributed to the shielding and deshielding
commonly observed in MIMs. Additionally, the terminal olefin
proton resonances α and λ were consumed, suggesting the
successful conversion to internal olefins of the catenate product.
The product identity was further corroborated by high-
resolution mass spectrometry via electrospray ionization
(HRMS-ESI) (Figure 2c). Data for the parent molecular ion
[M]2+ showed mass peaks centered around m/z = 1300, which
closely matched the simulated spectrum. Similarly, the [M +
H]3+ ion was also determined and matched the calculated
spectrum.
RESULTS AND DISCUSSION
■
To test the feasibility of the one-pot strategy, the preparation of
OTPM was carried out first via the convergent synthesis of
asymmetric terpy- and phen-containing halves that were
coupled together using a standard nucleophilic substitution
reaction under high pressure. The terpy ligand was chosen due
to its selectivity toward the formation of hexacoordinate
complexes with bivalent (M2+) metals such as Fe2+ and Ru2+.
Conversely, the phen ligand favors a tetracoordinate geometry
when complexed with monovalent (M+) metals such as Cu+. In
their seminal work, Sauvage and co-workers utilized a similar
orthogonal templation strategy but for the purpose of synthesiz-
ing [2]catenates. In the present case, the synthesis of the larger
macrocycle precursor OTPM has been demonstrated for the
purpose of synthesizing a [4]catenate (see Supporting
Analytical high-pressure liquid chromatography (HPLC) was
performed (Figure 2d) on OTPM and [2]C-M to assess product
purity. Although 1H NMR (Figure 2b) indicates that the OTPM
precursor is very pure, the HPLC trace displayed a small
subpopulation of peaks at shorter retention times. These peaks
using LC-MS-ESI, which is likely the result of hydrogen bonding
with the nonmetalated pyridine rings. The purity of OTPM was
further demonstrated using analytical GPC in DMF (Figure
purity of [2]C-M was also evaluated using analytical HPLC, and
the resultant trace revealed a narrow, unimodal peak supporting
the claim that it was isolated cleanly. Additionally, retention of
Fe2+ in [2]C-M during purification was confirmed by UV−vis
absorption spectroscopy (Figure 2e). The resultant stability of
the metal complex in [2]C-M is important in order for a one-pot
synthetic strategy to be successful.
To form the first metal complex (Fe-(OTPM)2), OTPM was
dissolved in tetrahydrofuran (THF) and heated to 60 °C,
followed by the addition of 1.3 equiv of Fe(BF4)2·6H2O in
deionized (DI) H2O, resulting in a red-colored solution. The
reaction mixture was heated for 1 h before being cooled to room
temperature and extracted in CHCl3 against three DI H2O
D
Inorg. Chem. XXXX, XXX, XXX−XXX