Fluorous effects in amide-based receptors for anions

Article abstract of DOI:10.1021/jo9000788

(Figure Presented) Hybrid receptors designed to recognize both the sulfonate headgroup and the fluorous tail of perfluorooctanesulfonate (CF ₃(CF₂)₇SO₃^-, "PFOS") were prepared by coupling fluorinated

Full text of DOI:10.1021/jo9000788

Fluorous Effects in Amide-Based Receptors for Anions
Jesse V. Gavette, Jacqueline M. McGrath, Anne M. Spuches,* Andrew L. Sargent,* and
William E. Allen*
Department of Chemistry, Science and Technology Building, East Carolina UniVersity,
GreenVille, North Carolina 27858-4353
allenwi@ecu.edu
ReceiVed January 13, 2009
Hybrid receptors designed to recognize both the sulfonate headgroup and the fluorous tail of
-
perfluorooctanesulfonate (CF3(CF2)7SO3 , “PFOS”) were prepared by coupling fluorinated carboxylic
-
acids onto poly(aminomethyl)benzene scaffolds. Binding to PFOS, CF3SO3 , p-TsO-, and Cl- was
1
monitored by H NMR and isothermal titration calorimetry (ITC). In chloroform solvent, hydrogen-
bonding to anions is accompanied by downfield shifts in the amide NH protons of the fluorinated receptors
and by evolution of heat. Association constants for 1:1 complexation (Kassoc) are >1000 M-1. An analogous
hydrocarbon receptor binds weakly to anionic guests (Kassoc < 50 M-1). Ab initio calculations indicate
that the differences in 1:1 binding strengths between fluorous and nonfluorous hosts cannot be ascribed
to differences in NH donor acidities.
Introduction
that has gained notoriety as a globally dispersed persistent
organic pollutant.5,6 Synthetic receptors for anions typically
present an array of hydrogen-bond donor atoms to their intended
guests.7 The sulfonate headgroup of PFOS is expected to be a
poor acceptor of H-bonds, however, because resonance stabi-
Highly fluorinated organic compounds tend to partition into
solvents and stationary phases that are likewise “fluorous”,
making it possible to separate the components of a mixture on
the basis of their degree of fluorine content.1 In the field of
supramolecular chemistry, fluorous interactions have been used
to control large-scale assembly. Targeted replacement of leucine
residues with hexafluoroleucines imparts stability to peptide
oligomers,2 fluorinated dendrimers can assume columnar shapes
that are not observed in their hydrocarbon analogues,3 and a
“teflon-footed” resorcinarene forms hexameric capsules in wet
fluorous media.4 To examine whether these effects can enhance
anion binding in organic solvents, we have chosen to target
-
lization within the SO3 moiety and inductive electron with-
drawal (by F atoms) limit the negative charge density at each
oxygen. The hybrid receptors described here feature perfluo-
rocarbon arms in proximity to amidic NH donor units. They
are designed to weakly anchor the oxoanionic portion of PFOS
by H-bonding, allowing any subtle contributions from host-guest
fluorous contacts to be observed. At present, none of the
(5) (a) Hori, H.; Hayakawa, E.; Yamashita, N.; Taniyasu, S.; Nakata, F.;
Kobayashi, Y. Chemosphere 2004, 57, 273–282. (b) Hebert, G. N.; Odom, M. A.;
Bowman, S. C.; Strauss, S. H. Anal. Chem. 2004, 76, 781–787. (c) Tseng, C.-
L.; Liu, L.-L.; Chen, C.-M.; Ding, W.-H. J. Chromatogr. A. 2006, 1105, 119–
126.
-
perfluorooctanesulfonate (CF3(CF2)7SO3 , “PFOS”), a surfactant
(1) Zhang, W. Chem. ReV. 2004, 104, 2531–2556.
(2) (a) Bilgic¸er, B.; Kumar, K. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
15324–15329. (b) Lee, K.-H.; Lee, H.-Y.; Slutsky, M. M.; Anderson, J. T.; Marsh,
E. N. G. Biochemistry 2004, 43, 16277–16284. (c) Lee, H.-Y.; Lee, K.-H.; Al-
Hashimi, H. M.; Marsh, E. N. G. J. Am. Chem. Soc. 2006, 128, 337–343.
(3) Percec, V.; Glodde, M.; Johansson, G.; Balagurusamy, V. S. K.; Heiney,
P. A. Angew. Chem., Int. Ed. 2003, 42, 4338–4342.
(6) (a) Moriwaki, H.; Takagi, Y.; Tanaka, M.; Tsuruho, K.; Okitsu, K.;
Maeda, Y. EnViron. Sci. Technol. 2005, 39, 3388–3392. (b) Hori, H.; Nagaoka,
Y.; Yamamoto, A.; Sano, T.; Yamashita, N.; Taniyasu, S.; Kutsuna, S. EnViron.
Sci. Technol. 2006, 40, 1049–1054.
(7) (a) Gale, P. A. Acc. Chem. Res. 2006, 39, 465–475. (b) Steed, J. W.
Chem. Commun. 2006, 2637–2649. (c) Kang, S. O.; Hossain, M. A.; Bowman-
James, K. Coord. Chem. ReV. 2006, 250, 3038–3052. (d) Lee, C.-H.; Miyaji,
H.; Yoon, D.-W.; Sessler, J. L. Chem. Commun. 2008, 24–34.
(4) Shimizu, S.; Kiuchi, T.; Pan, N. Angew. Chem. Ed. Int. 2007, 46, 6442–
6445.
3706 J. Org. Chem. 2009, 74, 3706–3710
10.1021/jo9000788 CCC: $40.75 2009 American Chemical Society
Published on Web 04/09/2009

Fluorous Effects in Amide-Based Receptors
a
1
TABLE 1. Association Constants from H NMR Titrations in CDCl3 at 295 K
3
4
5
6
7
b
b
b
b
b
b
b
PFOS
CF3SO3
p-TsO-
Cl-
4.0 × 106 M-2
2.4 × 106 M-2
2.0 × 106 M-2
1500 M-1
3.1 × 106 M-2
4.0 × 106 M-2
8.2 × 106 M-2
8.9 × 105 M-2
39 M-1
41 M-1
36 M-1
12 M-1
1400 M-1
nd
130 M-1
nd
c
c
-
5500 M-1
1000 M-1
190 M-1
60 M-1
a Values represent averages of at least two replicate titrations. Anions were added as TEA salts. b Binding stoichiometry of 2 hosts per 1 guest. c From
ref 11. nd ) not determined.
SCHEME 1
SCHEME 2
of 0.0003-0.0005 M are freely mobile and yield sharp 1H NMR
spectra. Six signals are observed in the 19F NMR spectrum of
3 (CF3 at -81.7 ppm, CF2 units from -115.5 to -127.0 ppm).13
19F-based measurements could not be used in quantitative
determinations of anion binding, however, as the presence of
putative guests induced negligible shifts in these peaks. Two
equivalents of PFOS (as tetraethylammonium (TEA) salt) caused
the signal assigned to CH2CF2CF2 to move upfield by 0.08 ppm,
while 2 equiv of p-toluenesulfonate (p-TsO-) induced a down-
field shift of 0.03 ppm.
methods to detect5 or decompose6 PFOS rely upon its selective
binding by organic hosts.
Results
Association constants with PFOS, p-TsO-, CF3SO3 , and Cl-
-
(Table 1) were derived from graphs of receptor δNH vs
Synthesis. Steric gearing predisposes the functional arms of
1,3,5-trisubstituted 2,4,6-triethylbenzenes to lie on the same face
of the central arene,8 and several anion receptors that take
advantage of such preorganization have been described.9 Here,
two tripodal hosts with >40% fluorine content by weight were
prepared by coupling fluorinated carboxylic acids onto tris(ami-
nomethyl) scaffold 110 (Scheme 1) using standard amidation
conditions.11 Upon cooling of the reaction mixtures and/or
adding water, 3 and 4 precipitated as analytically pure powders
in good yield. Hydrocarbon control compound 511 was synthe-
sized from 1 and nonanoic acid. As expected, the coupling
strategy was unsuccessful when applied to trifluoroacetic acid.
Short-armed host 6 was instead obtained by reaction of 1 with
the corresponding anhydride. Diamine 212 served as starting
material for bidentate molecule 7 (Scheme 2), which was
isolated via column chromatography using standard silica gel
and a nonfluorous eluent (CH2Cl2/EtOAc).
[anion]total.
14,15 A single set of peaks appeared throughout each
experiment, indicative of fast exchange on the NMR time scale.
Although the presence of perfluorocarbon arms in a host
enhances anion affinities (e.g., fluorous, bidentate 7 forms more
robust complexes than nonfluorous, tridentate 5), it does not
impart selectivity for PFOS in chloroform. For 3 and 4, fitting
of sulfonate binding isotherms to a 1:1 stoichiometry model
led to large errors. A continuous variations plot for 3 with
TEAPFOS (Figure 1) confirmed that higher-order species are
formed.16 Reliable fits for δNH vs [sulfonate] were ultimately
obtained by assuming that all complexes in solution have 2:1
host-guest composition. The simple 1:1 binding behavior of
7, which is 51% fluorine by weight, suggests that high F content
alone is not sufficient to promote higher-order complexation.
Chloride binding stoichiometry is sensitive to the structure
of the host and the tetraalkylammonium countercation.17 The
ratio of 3/Cl-, verified by continuous variations analysis to be
1:1 when TEACl is the chloride source (see the Supporting
Information), is 2:1 for titrations performed using the tetrabu-
tylammonium (TBA) salt. The ꢀ value for 2:1 complexes of 3
Anion Binding in Solution: NMR. In preparation for NMR
titrations, the behavior of fluorous hosts 3, 4, and 7 was
examined in deuterated chloroform. These compounds can be
dissolved in CDCl3 to concentrations approaching 0.005 M,
though such samples tend to flow poorly and to foam. Solutions
(13) Buchanan, G. W.; Smits, R.; Munteanu, E. J. Fluorine Chem. 2003,
123, 255–259.
(14) Hynes, M. J. J. Chem. Soc., Dalton Trans. 1993, 311–312.
(15) When 1H NMR binding experiments were conducted at higher receptor
concentrations ([3] or [4] ) 0.004 M) the resultant isotherms were not smoothly
hyperbolic and the ꢀ values obtained from them were not reproducible. See the
Supporting Information.
(8) Hennrich, G.; Anslyn, E. V. Chem.sEur. J. 2002, 8, 2219–2224.
(9) (a) Zyryanov, G. V.; Palacios, M. A.; Anzenbacher, P. Angew. Chem.,
Int. Ed. 2007, 46, 7849–7852. (b) Turner, D. R.; Paterson, M. J.; Steed, J. W.
Chem. Commun. 2008, 1395–1397. (c) Berryman, O. B.; Sather, A. C.; Hay,
B. P.; Meisner, J. S.; Johnson, D. W. J. Am. Chem. Soc. 2008, 130, 10895–
10897.
(10) Wallace, K. J.; Hanes, R.; Anslyn, E.; Morey, J.; Kilway, K. V.; Siegel,
J. Synthesis 2005, 2080–2083.
(11) Gavette, J. V.; Sargent, A. L.; Allen, W. E. J. Org. Chem. 2008, 73,
3582–3584.
(12) Diamine 2 appears in the following paper, but details of its synthesis
are not given: Lee, K. H.; Hong, J.-I. Tetrahedron Lett. 2000, 41, 6083–6087.
We obtained 2 in 28% yield by reduction of the corresponding diazide using
PPh3 in THF/H2O
(16) Both 1:1 and 2:1 complexes containing 3 were observed by electrospray
ionization mass spectrometry. In negative ion mode, the base peak for chloroform
solutions of 3 and TEAPFOS in molar ratios of 10:1, 2:1, and 1:8 appears at
m/z ) 1870, corresponding to 3 · PFOS. The 10:1 samples also have a low-
abundance (< 1%) signal at m/z ) 3242 assigned to 3 · 3 · PFOS.
(17) Sessler, J. L.; Gross, D. E.; Cho, W.-S.; Lynch, V. M.; Schmidtchen,
F. P.; Bates, G. W.; Light, M. E.; Gale, P. A. J. Am. Chem. Soc. 2006, 128,
12281–12288.
J. Org. Chem. Vol. 74, No. 10, 2009 3707

Gavette et al.
FIGURE 1. Continuous variations plot for 3 with TEAPFOS in CDCl3.
TABLE 2. Thermodynamic Parameters from ITC in CDCl3 at
298 Ka
K
∆H
(kcal/mol)
Τ∆S
(kcal/mol)
N(anion/host
stoichiometry)
(M-1
)
3 + PFOS
3 + CF3SO3
3 + Cl-
3500
1600
1200
1200
-11.0
-10.1
-3.4
-6.1
-5.7
0.8
0.54
0.54
0.99
0.76
FIGURE 2. ITC response during addition of TEAPFOS (12.5 mM in
chloroform) to 3 (0.401 mM in chloroform). The fit line corresponds
to K ) 3400 M-1, ∆H ) -11.6 kcal/mol, T∆S ) -6.8 kcal/mol, N )
0.50.
-
6 + PFOS
-6.9
-2.6
a Values represent averages of at least two replicate titrations. Anions
were added as TEA salts.
is consistent with aggregates of 3 undergoing disassociation.
Exothermic behavior was restored by using more dilute titrant
solutions ([3] < 0.001 M), and thermodynamic parameters
derived from these runs were statistically identical to those
obtained by “forward” titrations.
and TBACl (8.2 × 105 M-2) is of the same order of magnitude
as that measured for 4 + TEACl.
Anion Binding in Solution: ITC. Selected combinations
from Table 1 were analyzed by isothermal titration calorimetry
(ITC). To facilitate comparison with NMR results, host mol-
ecules in the sample cell were prepared to concentrations of
0.0003-0.0005 M in CDCl3. Interactions between 3 and PFOS
or CF3SO3- are driven entirely by enthalpy (Table 2), and their
similar thermodynamic profiles are indicative of a common
mode of complexation. Chloride binding is characterized by a
positive, albeit small, T∆S term. Positive entropies have been
observed previously for calix[4]pyrrole-chloride complexes in
DMSO17 and for binding of sulfate ion by ditopic guanidinium
receptors in methanol18 and are associated with release of
ordered solvent into the bulk solution. Host 3 was unresponsive
to the TEA salt of a weakly coordinating anion (PF6-). Thus,
the data in Table 2 do not arise from binding of the counter-
cation. Stoichiometries, N, for complexes containing 3 follow
those determined by NMR. In the case of 3 + TBACl (data not
shown), N tended toward zero during the curve fitting process.
Fixing N at either 1.00 or 0.500 allowed for convergence, but
the latter fit had significantly lower error. A representative plot
for a 2:1 interaction is shown in Figure 2.
Discussion
Inductive Effects. For synthetic hosts that operate via H-bond
donation, anion affinities are enhanced by the presence of
electron-withdrawing groups near the donor atoms. Halogenation
of the ꢀ-positions of pyrrole-based receptors, for example, makes
the NH groups more acidic.19 A similar effect is observed when
the aromatic ring of phenyl-substituted amides is substituted.20
In the case of 3, 4, and 7, two CH2 units separate the donor
amides from (CF2)nCF3. Previous studies showed that phos-
phine21 and 2,2′-bipyridine22 ligands require eight or more
methylene spacers to be fully insulated from perfluoroalkyl
groups, but for the present systems, 1H NMR data indicate that
-CH2CH2- is almost as effective. In chloroform, the NH
resonances of 3 and 4 appear at 5.28 and 5.29 ppm, only slightly
downfield of that observed for nonfluorous 5 (5.23 ppm).11
Natural population analyses of charges on the amide NH atoms,
at the B3LYP/6-31+G* level of theory, further support that the
(19) (a) Anzenbacher, P.; Try, A. C.; Miyaji, H.; Jurs´ıkova´, K.; Lynch, V. M.;
Marquez, M.; Sessler, J. L. J. Am. Chem. Soc. 2000, 122, 10268–10272. (b)
Maeda, H.; Ito, Y. Inorg. Chem. 2006, 45, 8205–8210.
(20) Zhang, Y.; Qin, J.; Lin, Q.; Wei, T. J. Fluorine Chem. 2006, 127, 1222–
1227.
(21) Jiao, H.; Le Stang, S.; Soo´s, T.; Meier, R.; Kowski, K.; Rademacher,
P.; Jafarpour, L.; Hamard, J.-B.; Nolan, S. P.; Gladysz, J. A. J. Am. Chem. Soc.
2001, 124, 1516–1523.
“Reversed” ITC experiments were conducted by injecting
chloroform solutions of 3 into the instrument cell containing
TEAPFOS. Endothermic peaks were observed when the con-
centration of 3 in the syringe exceeded 0.002 M. This finding
(22) Bennett, B. L.; Robins, K. A.; Tennant, R.; Elwell, K.; Ferri, F.; Bashta,
I.; Aguinaldo, G. J. Fluorine Chem. 2006, 127, 140–145.
(18) Schmidtchen, F. P. Coord. Chem. ReV. 2006, 250, 2918–2928.
3708 J. Org. Chem. Vol. 74, No. 10, 2009

Fluorous Effects in Amide-Based Receptors
TABLE 3. Calculated Binding Energies and Hydrogen-Bond
Lengths for 1:1 Complexesa
b
∆Egas phase ∆Echloroform NH · · · acceptor OdCCH · · · acceptor
(kcal/mol) (kcal/mol)
(Å, av)
(Å, av)
3·PFOS
3·CF3SO3
3·Cl-
4·CF3SO3
5·PFOS
5·CF3SO3
5·Cl-
-31.4
-35.0
-45.3
-34.8
-24.9
-27.4
-34.7
-13.6
-15.4
-16.2
nd
-11.1
nd
2.07
2.04
2.59
2.05
2.18
2.13
2.71
2.74
2.73
3.58
2.70
2.80
2.79
3.71
-
-
-
nd
a Computations employed the Becke-Tsuneda-Hirao gradient-corrected
exchange-correlation functional24 and a double numerical plus polarization
basis set. b COSMO solvent model25 applied. nd ) not determined.
binding atoms of 3, 4, and 5 are electronically similar. These
calculations predict average values of +0.428, +0.429, and
+0.426 per hydrogen atom, respectively, for the three hosts.
Although the perfluoroalkyl groups in 3 have negligible local
impact on amide acidity, they have a pronounced influence on
the global distribution of charge density, relative to hydrocarbon
5. The calculated dipole moment of 3 is 4.43 D, with its positive
end located near the aromatic scaffold of the host and its
negative end directed toward the perfluorocarbon arms. There-
fore, from a “dipole-engineering”23 perspective, 3 is properly
matched to its intended PFOS guest. For 5 the dipole is of
similar magnitude (5.28 D) but in the opposite direction. DFT-
calculated models of PFOS, CF3SO3-, and Cl- were docked to
the lowest-energy conformations of receptors 3, 4, and 5 and
the resultant 1:1 systems were optimized.24 Binding energies
follow the trend 3 > 5 for all three guests, and average
NH· · ·anion distances are about 0.1 Å shorter for complexes
of 3 (Table 3). Application of a solvent field25 corresponding
to chloroform renders anion binding less favorable.
Countercation Effects. DFT was further applied to two
complexes with included countercations, 3 · TEACl and
3·TEACF3SO3. Figure 3 shows the lowest-energy gas-phase
structures for each. In Figure 3a, the spherical Cl- ion forms
three hydrogen-bonds to 3 while remaining tightly paired to
TEA. Desolvation of both salt components would be required
for binding in this manner in solution and may give rise to the
favorable entropic term observed by ITC. Such close approach
between charges is impossible for a directional species like
FIGURE 3. (a) DFT-optimized view of 3·TEACl, showing the
countercation inside of the binding pocket. The NTEA · · ·Cl- separation
is 4.21 Å, and the average NH· · ·Cl- hydrogen-bond length is 2.99 Å.
(b) DFT-optimized view of 3·TEACF3SO3, showing the countercation
outside of the binding pocket. The NTEA · · ·Striflate separation is 6.42 Å,
the closest NTEA · · ·Otriflate approach is 5.52 Å, and the average
NH· · ·Otriflate hydrogen-bond length is 2.16 Å.
be enhanced in solvents of higher polarity. In acetone-d6, the
NMR-derived ꢀ values for 2:1 binding27 of 3 + PFOS and 3 +
-
CF3SO3 once it has bound to a tripodal receptor. Thus, the
-
CF3SO3 are 1.3 × 104 and 8.6 × 102 M-2, respectively,
cation resides outside of the cleft (Figure 3b). We propose that
an “outer-sphere” cation is probable for the larger sulfonates
PFOS and p-TsO-, which occupy more space than CF3SO3- in
the binding pocket, but also for chloride complexes when the
cation cannot comfortably fit within the receptor cleft (e.g., 3
+ TBACl). If so, the interaction stoichoimetries observed for 3
by NMR and ITC are correlated with cation location.
indicating that 3 can distinguish between sulfonates that have a
long fluorinated chain or a short one. Furthermore, selectivity
for PFOS over CF3SO3- is predicted to disappear in a solvent-
free environment (Table 3). Comparing absolute binding
strengths in acetone and chloroform, any stabilization from
fluorous interactions in the former solvent is apparently over-
whelmed by acetone’s ability to compete for H-bonding sites
on the host. In the absence of guests, hydrogen-bonding between
acetone-d6 and the NH protons of 3 causes them to resonate at
7.18 ppm (vs 5.28 ppm in CDCl3, see above). The limited
solubility of 3 in DMSO-d6, CD3CN, and perfluorohexane
prevented measurement of ꢀ values in these solvents.
Solvent Effects. Perfluorocarbons are hydrophobic.26 As such,
the affinity between the fluorous regions of host and guest should
(23) Kimura, S. Org. Biomol. Chem. 2008, 6, 1143–1148.
(24) (a) Becke, A. D. J. Chem. Phys. 1988, 88, 1053–1062. (b) Delley, B.
J. Chem. Phys. 1990, 92, 508–517. (c) Tsuneda, T.; Hirao, K. Chem. Phys. Lett.
1997, 268, 510–520. (d) Delley, B. J. Chem. Phys. 2000, 113, 7756–7764, The
large size of 3 and the absence of obvious modes of 2:1 assembly prevented
evaluation of higher-order complexes using DFT.
Conclusions
(25) (a) Klamt, A.; Schueuermann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799–805. (b) Klamt, A. J. Chem. Phys. 1995, 99, 2224–35. (c) Klamt, A.; Jonas,
V. J. Chem. Phys. 1996, 105, 9972–9981.
(26) (a) Dunitz, J. D. ChemBioChem 2004, 5, 614–621. (b) Biffinger, J. C.;
Kim, H. W.; DiMagno, S. G. ChemBioChem 2004, 5, 622–627. (c) Ja¨ckel, C.;
Koksch, B. Eur. J. Org. Chem. 2005, 4483–4503. (d) Jing, P.; Rodgers, P. J.;
Amemiya, S. J. Am. Chem. Soc. 2009, 131, 2290–2296.
The presence of multiple fluorine atoms in tripodal anion
receptors enhances binding affinities without significantly alter-
(27) The maximum in the continuous variations plot for 3 + TEAPFOS in
acetone-d6 appears at g0.6 mol fraction 3. See the Supporting Information.
J. Org. Chem. Vol. 74, No. 10, 2009 3709

Gavette et al.
0.103 g (28%) of 6 as a white solid. mp ) 216-218 °C; 1H NMR
(CDCl3) δ 1.25 (t, 9H), 2.72 (q, 6H), 4.63 (d, 6H), 6.14 (s, 3H);
13C (CD3CN) δ 16.3, 24.0, 39.1, 111.5, 115.3, 119.1, 122.9, 131.5,
ing the acidity of their H-bond donor groups. Such hosts form
2:1 complexes with sulfonates, complicating direct comparison
with nonfluorous analogues. Efforts are underway to obtain
binding data in fluorous solvents,28 using systems with greater
fluorine content than those described here.
146.1, 156.9, 157.4, 157.9, 158.3; IR (film) νmax 3284, 1695 cm-1
;
Anal. Calcd for C21H24N3O3F9: C, 46.93; H, 4.50; N, 7.82. Found:
C, 47.26; H, 4.59; N, 7.69.
N,N′-(2,4,6-Triethyl-1,3-phenylene)bis(methylene)bis-
(4,4,5,5,6,6,7,7,8,8,9,9,9-tridecafluorononanamide) (7). Diamine
212 (0.125 g, 0.567 mmol), 4,4,5,5,6,6,7,7,8,8,9,9,9-tridecafluo-
rononanoic acid (0.455 g, 1.16 mmol), HBTU (0.530 g, 1.40 mmol),
and DIPEA (0.220 g, 1.70 mmol) were combined in 6 mL of DMF.
Under N2, the flask was heated to 60 °C with stirring. After 3 h,
the reaction mixture was poured into 100 mL of water and extracted
with CH2Cl2. The organic phase was washed with water, dried over
Na2SO4, filtered, and evaporated to give a brown oil, which
solidified while standing under vacuum. The crude product was
purified by flash column chromatography over silica gel using
CH2Cl2/EtOAc (2:1, v:v) as the eluent, affording 90 mg (16%) of
7 as an off-white solid. mp ) 198-200 °C; TLC (CH2Cl2/EtOAc,
2:1) Rf ) 0.65; 1H NMR (CDCl3) δ 1.18 (t, 3H), 1.23 (t, 6H), 2.43
(t, 4H), 2.46-2.60 (m, 4H), 2.65 (q, 4H), 2.70 (q, 2H), 4.48 (d,
4H), 5.30 (s, 2H), 7.02 (s, 1H); 13C NMR (CDCl3) δ 15.9, 16.6,
22.7, 26.2, 26.5, 26.8, 27.0, 37.9, 128.0, 130.7, 143.4, 144.3, 169.3;
Anal. Calcd for C32H30F26N2O2: C, 39.68; H, 3.12; N, 2.89. Found:
C, 39.68; H, 3.15; N, 2.85.
ITC. Samples of 3 or 6 were dissolved in freshly opened CDCl3
to concentrations of 0.0004-0.0005 M then transferred into the
calorimetry cell. A solution of the appropriate anion (as tetraalky-
lammonium salt), prepared to be ∼30× more concentrated than
that of the host, was injected in 4 µL aliquots until [anion] was at
least 3[host]. To determine the enthalpies of solution for TEAPFOS,
TEACF3SO3, TEACl, and TBACl, the anion salt in question was
titrated into an equilibrated 1:6 mixture of host and anion salt. The
heat changes associated with the last four injections of these
titrations were averaged. The resulting background values were
subtracted from the original binding data, then iterative fitting to a
“single set of identical binding sites” model was performed. Values
of K, ∆H, and N (interaction stoichiometry) were allowed to vary
freely.
Experimental Section
N,N′,N′′-(2,4,6-Triethylbenzene-1,3,5-triyl)tris(methylene)tris-
(4,4,5,5,6,6,7,7,8,8,9,9,9-tridecafluorononanamide) (3). Triamine
110 (0.125 g, 0.501 mmol), 4,4,5,5,6,6,7,7,8,8,9,9,9-tridecafluo-
rononanoic acid (0.609 g, 1.55 mmol), O-(benzotriazol-1-yl)-
N,N,N′,N′-tetramethyluronium hexafluorophosphate (HBTU, 0.664
g, 1.75 mmol), and N,N-diisopropylethylamine (DIPEA, 0.259 g,
2.00 mmol) were combined in 8 mL of DMF. Under N2, the mixture
was heated to 60 °C for 2 h with stirring, during which time it
became red-brown. Upon cooling, the mixture deposited a precipi-
tate. The solid was collected by filtration, washed with water, and
dried under vacuum to afford 0.575 g (84%) of 3 as a faintly pink
powder. mp ) 208-210 °C; 1H NMR (CDCl3) δ 1.20 (t, 9H), 2.44
(t, 6H), 2.58 (m, 6H), 2.69 (q, 6H), 4.50 (d, 6H), 5.28 (t, 3H);13C
NMR (acetone-d6) δ 16.5, 23.4, 26.8, 27.2, 27.4, 38.4, 108.4, 112.0,
116.1, 119.6, 123.1, 133.2, 144.6, 169.8;19F NMR (CDCl3, neat
CF3COOH (-78.5 ppm) as external reference) δ -81.7, -115.5,
-122.8, -123.8, -124.4, -127.0; IR (film) νmax 3306, 1644, 1548
cm-1; Anal. Calcd for C42H36F39N3O3: C, 36.78; H, 2.65; N, 3.06.
Found: C, 36.91; H, 2.67; N, 3.08.
N,N′,N′′-(2,4,6-Triethylbenzene-1,3,5-triyl)tris(methylene)tris-
(4,4,5,5,6,6,6-heptafluorohexanamide) (4). The procedure de-
scribed for 3 was followed, using 4,4,5,5,6,6,6-heptafluorohexanoic
acid (0.389 g, 1.61 mmol). After 24 h, the mixture was allowed to
cool with stirring. Dropwise addition of cold water precipitated the
crude product, which was collected by filtration and washed with
water. Drying under vacuum yielded 0.412 g (89%) of 4 as a light
orange solid. mp ) 212-213 °C; 1H NMR (CDCl3) δ 1.20 (t, 9H),
2.46 (t, 6H), 2.52 (m, 6H), 2.69 (q, 6H), 4.50 (d, 6H), 5.29 (t, 3H);
13C NMR (acetone-d6) δ 16.5, 23.4, 26.8, 26.9, 38.4, 133.2, 144.6,
169.8; IR (film) νmax 3320, 1643 cm-1; Anal. Calcd for
C33H36F21N3O3: C, 43.01; H, 3.94; N, 4.56. Found: C, 43.26; H,
3.96; N, 4.58.
N,N′,N′′-(2,4,6-Triethylbenzene-1,3,5-triyl)tris(methylene)tris-
(2,2,2-trifluoroacetamide) (6). Triamine 110 (0.173 g, 0.694 mmol)
and DIPEA (1.49 g, 11.5 mmol) were combined in 10 mL of DMF.
While the mixture was stirred, a solution of trifluoroacetic anhydride
(0.468 g, 2.22 mmol) in 10 mL of DMF was added dropwise over
several minutes. A condenser was affixed to the flask, and the
contents were heated to 60 °C under N2 for 24 h. Upon cooling,
the reaction was quenched by dropwise addition of 10 mL of water.
The solvents were removed under reduced pressure, and the crude
product was dissolved in CH2Cl2. The organic phase was washed
with water, dried over Na2SO4, filtered, and evaporated to afford
Acknowledgement. W.E.A. thanks the ECU Division of
Research and Graduate Studies for a Research Development
Award. A.L.S. thanks the ECU CACS and the National Science
Foundation (CNS-0619285). J.V.G. was supported by a Bur-
roughs-Wellcome Fellowship for 2007-2008.
1
Supporting Information Available: H/13C NMR spectra
for all new compounds, representative binding curves from 1H
NMR and ITC, Job plots, views of DFT-calculated structures
and associated atomic coordinates. This material is available
free of charge via the Internet at http://pubs.acs.org.
(28) (a) O’Neal, K. L.; Geib, S.; Weber, S. G. Anal. Chem. 2007, 79, 3117–
3125. (b) O’Neal, K. L.; Weber, S. G. J. Phys. Chem. 2009, 113, 149–158.
JO9000788
3710 J. Org. Chem. Vol. 74, No. 10, 2009