Metal Complexation of Novel Thia-Crown Ether Macrocycles by ESI-MS

1
Metal Complexation of Novel Thia-Crown Ether
Macrocycles by ESI-MS Sheldon M. Williams, Wendi
M. David and Jennifer S. Brodbelt Department of
Chemistry and Biochemistry University of Texas at
Austin, Austin, TX 78712
2

• Overview
• Purpose
• Evaluate relative heavy metal ion binding
  affinities of crown-ether
• macrocycles with S, O, and N heteroatoms.
• Determine avidities for extracting heavy metals
  from aqueous
• solution.
• Method
• ThermoFinnigan LCQ Duo
• 50/50 methanol/ chloroform solutions
• Extractions from water to chloroform

3

• Results
• Most thia-crown ether macrocycles tested were
  found to bind
• exclusively to mercury(II) ion in
  competitive assays with
• cadmium(II), lead(II), mercury(II), and
  zinc(II) chlorides
• Extraction studies with chloroform and water
  revealed that several of
• the thia-crown ether macrocycles extracted
  mercury(II) ion efficiently
• and exclusively in the presence of cadmium,
  lead, mercury, and zinc
• ions

4
Introduction Novel macrocycles are
currently being developed and evaluated for use
as selective, recyclable ligands for extraction
of heavy metals from contaminated water. Fast,
efficient feedback about metal selectivities and
avidities will aid the design and development
process. Electrospray ionization mass
spectrometry (ESI-MS) shows promise for rapid
screening of binding selectivities in host-guest
chemistry 1-5, offering versatility in a
variety of solvent systems and requiring minimal
sample consumption. In the present study, ESI-MS
is used to evaluate the metal binding
selectivities of an array of novel caged
macrocycles (Figure 1) for mercury(II), lead(II),
cadmium(II), and zinc(II) ions. It is found that
the type of heteroatom (S, O, N), cavity size,
and presence of other substituents influence the
metal selectivities. The desired
structure of a heavy metal extraction agent
should be one that minimizes its solubility in an
aqueous environment and yet is able to
efficiently extract the desired metal ion from
the wastewater. For this reason, several
water-insoluble macrocycles that exhibited
superior
5
affinity for particular heavy metal ions in our
initial binding assays have been tested for their
ability to extract mercury(II), lead(II),
cadmium(II), and zinc(II) ions from an aqueous
environment into an organic environment.
6
Methods Solutions containing a single
host with multiple metals are analyzed for each
thia-crown ether macrocycle in 50/50 methanol/
chloroform. The concentration of the host and
each metal are 2.5 x 10-5 M. Initial extractions
of aqueous heavy metal salts to host in
chloroform are conducted with 0.033 M of each
metal salt in aqueous solution and 2 x 10-4 M
host in chloroform. Experiments conducted for the
purpose of detecting the extraction of low metal
ion concentrations used a 11 hostmetal chloride
molar ratio with metal concentrations varying
from 1 x 10-4 M to 1 x 10-5 M. For extraction, 1
ml of organic solution with host and 1 ml aqueous
solution with metal are vortexed for five minutes
in a closed 4 ml vial. The organic phase from the
extraction experiments is then analyzed by
ESI-MS. All mass spectrometry experiments are
performed on a ThermoQuest LCQ Duo ESI-MS with a
needle voltage of 5 kV and a heated-capillary
temperature of 150oC. A flow rate of 10 ?l/min
was used for all ESI-MS experiments except the
low metal ion concentration experiments where a
flow rate of 60 ?l/min was used.
7
Figure 1 Thia-crown Ether Macrocycle Structures
8
Figure 1, cont.
9
Results Macrocycle/ Heavy Metal Binding Survey
in 50/50 Methanol/ Chloroform As
presented in Figure 2, the ESI-mass spectra for
the solutions containing a macrocycle and the
metal perchlorates in 50/50 methanol/ chloroform
typically consist of signals for complexes of a
host and a doubly-charged metal ion as well as
singly charged tertiary complexes including a
single counter-ion. The signal intensities of the
metal complexes in the ESI-mass spectra were used
to estimate the relative binding selectivities
and avidities of the hosts. A comparison of the
selectivities of every macrocycle tested is
compiled in Table 1.
10
Figure 2A ESI-MS of 7 with Cd(ClO4)2, Pb(ClO4)2,
Zn(ClO4)2 (1111) in 50/50 Methanol/Chloroform
(7CdClO4)
100
(7PbClO4)
(7ZnClO4)
(7Cu)
(7Cd)2
(7Pb)2
(7Zn)2
0
200
400
600
800
1000
1200
m/z
11
Figure 2B ESI-MS of 7 with Cd(ClO4)2, Pb(ClO4)2,
Hg(ClO4)2, Zn(ClO4)2, (11111) in 50/50
Methanol/Chloroform
12
It was observed that Zn2 complexed favorably
with hosts with four ethylene-heteroatom units in
the crown ring, but less than three non-sulfur
heteroatoms (4, 6, 12, 14), whereas Cd2
complexed favorably to the hosts with four or
more ethylene-heteroatom units with two or more
non-sulfur heteroatoms (5, 7, 9, 10, 11, 13, 14,
15). Macrocycles 2 and 3, which are hosts with
four propyl-sulfur heteroatom units, also showed
large Zn2 and Cd2 affinities. Pb2 complexed
favorably to 1 and 7. Mercury(II) complex was
usually the dominant species or only species
observed for the solutions containing this ion,
thus suggesting that many of the thio-crown
ethers have great Hg2 selectivities. One host,
8, which contains only three ethylene-heteroatom
units in its ring, appears to prefer binding to
residual sodium ion impurity over the much
greater quantity of heavy metal ions in solution,
indicating the cavity to be too small to bind any
of the heavy metals efficiently. These initial
findings were used to select candidates for
extraction of metal ions from an aqueous phase
into a chloroform organic phase in order to
further evaluate the ability of these hosts as
agents for extracting heavy metals from an
aqueous environment. Using these selected
compounds, the extraction process served as a
model
13
for wastewater extraction and to determine the
selectivities of hosts in this process towards
heavy metal cations.
Selective Mercury Ion Extraction from Aqueous
Solution Due to the exclusive selectivity
most of the thia-crown ether macrocycles showed
towards mercury(II) ion, the eight macrocycles
providing the best combination of clean spectra
and high signal-to-noise ratio for the (Host Hg
ClO4) complex were used for studying
mercury(II) ion extraction. Extractions using
mercury(II) perchlorate were generally
unsuccessful, and the observed spectra were
similar to that shown for 7 in Figure 3A.
However, 2 appears to have extracted mercury (II)
ion with a perchlorate counter-ion, though rather
poorly as shown in Figure 3B. Extraction of
mercury(II) chloride was much more successful as
shown in the spectra of 2 and 7 in Figures 4A and
4B, respectively, and for all the macrocycles
tested in Table 2.
14
Figure 3 ESI-MS Macrocycle-Containing Chloroform
Phase After Extraction of Aqueous Phase
Containing Cd(ClO4)2 Pb(ClO4)2 Hg(ClO4)2
Zn(ClO4)2 (11111)
100
Host 7
(7HgCl)
(7Cu)
0
200
400
600
800
1000
m/z
100
(2Cu)
Host 2
(2HgCl)
(2HgClO4)
0
200
400
600
800
1000
m/z
15
Table 2 Relative Efficiencies of Mercury(II)
Extraction by Sulfur Containing Macrocycles
16
Since in almost all cases, the metal ion must
transfer from the aqueous solution into the
chloroform to complex with a water-insoluble
macrocycle, the metal ion must be bound to two
counter-ions to form a neutral molecule before
transfer can occur. In order for the metal ion
bound to two anions to complex with the
macrocycle, one of the anions must pass through
the central cavity of the macrocycle. It is
believed that the effective diameter of the
metal-bound perchlorate ion, which is likely in a
tetrahedral geometry, is too large too
efficiently pass through the macrocycles cavity.
In addition, the perchlorate anion is more
hydrophilic than the chloride ion due to its four
oxygen atoms, which increases the water
desolvation energy needed to transfer to the
chloroform from the aqueous phase. Of
the macrocycles tested, 7 gave the best
extraction results. Its superior performance is
likely due to a combination of several factors.
Primarily, the four sulfurs appear close to the
geometry needed for a square-planar geometry,
with the two chloride counter-ions binding at the
axial positions of a near-ideal octahedral
structure. The two additional oxygens can
provide ion-dipole interactions to stabilize
the
17
mercury ion in the cavity. For 15, the two
additional oxygens may add more sites for
interaction with the mercury ion, but the added
space between the sulfurs probably interferes
with the sulfurs attaining a geometry as
favorable as is achieved for 7. The three sulfurs
and one oxygen of 6 likely allow for a similar
square-planar geometry, though the presence of
the oxygen, and perhaps the smaller size, reduce
its performance compared to 7 and 15. Macrocycles
6 and 7 may be superior in mercury extraction
performance to 2 and 3 because the propyl units
between every sulfur in 2 and 3 may result in a
deviation from the ideal square-planar geometry,
similar to the effect with 15, except that there
are only four heteroatom interaction sites in the
macrocycle cavity versus six and eight for 7 and
15, respectively. Although all the macrocycles
tested in the extraction experiments had at least
four heteroatoms, those with less than three
sulfurs performed the least well. As a
final experiment, 7 was used to determine the
lower limit for detecting extraction of mercury
from water into chloroform. Figure 5 presents a
plot of the signal-to-noise ratio for the
(macrocycle Hg Cl) peak with decreasing
equimolar concentrations of HgCl2 and 7 in the
aqueous and organic phase, respectively.
18
Figure 5 ESI-MS Signal-to-Noise Ratio Versus
HgCl2 Concentration
HOST 7
19
A lower-limit-of-detection of 1 x 10-5 M (2 ppm
Hg) was determined, though using a higher flow
rate (gt100 ?l/min), using a greater, constant
macrocycle concentration in the organic phase (gt1
x 10-4 M), and improving tuning of the ESI-MS
interface lenses on the LCQ Duo would likely
improve the detection limit by greater than an
order of magnitude. These probable methods of
improving the limit-of-detection were not
examined in the current study due to limited
sample quantities, but are planned when
additional 7 becomes available.
20

• Conclusions
• Crown ether macrocycles with several sulfur
  heteroatoms and a ring
• composed of at least four ethylene
  heteroatom units are necessary for
• efficient, selective mercury extraction.
• Macrocycle cavities with several sulfurs and/or
  additional nitrogens
• and oxygens arranged to bind to mercury with
  a square-planar
• geometry appear the most ideal.
• Macrocycles with a pair of sulfurs separated by
  an ethylene unit on
• opposite sides of the cavity with a flexible
  tether between, and with
• additional nucleophilic heteroatoms on the
  tether appears to create the
• most ideal mercury extraction agent of those
  studied.
• The presence of small, low hydrophilicity anions
  in the aqueous
• medium greatly enhances mercury ion
  extraction for the macrocycles
• tested in this study.
• Sulfur containing crown ether macrocycles have
  been shown to have
• potential as agents for selectively
  extracting and detecting aqueous
• mercury ion over a large concentration range.

21

• Future Work
• Optimize lower-limit-of-detection methodology
• Examine selectivity in the presence of other
  common metal ions (alkali
• and alkaline earths)
• Molecular modeling and ab initio calculations
• Acknowledgements
• The laboratory of Alan P. Marchand, Department of
  Chemistry,
• University of North Texas, is gratefully
  acknowledged for synthesizing
• macrocycles 5 through 15.
• The National Science Foundation, the Welch
  Foundation, and the Texas
• Advanced Technology Program are gratefully
  acknowledged.

22
References 1. Kempen, E.C., Brodbelt, J.S.,
Bartsch, R.A., Blanda, M.T., Farmer, D.B.,
Anal. Chem., 2001, 73, 384. 2. Blair, S.M,
Brodbelt, J.S., Marchand, A.P., Chong, H.-S.,
Alihodzic, S., J. Am. Soc. Mass Spectrom.,
2000, 11, 884. 3. Kempen, E.C., Brodbelt, J.S.,
Anal. Chem., 2000, 72, 5411. 4. Blair, S.M.,
Brodbelt, J.S., Marchand, A.P., Kumar,
Kalpenchery, A., Chong, H.-S., Anal. Chem.,
2000, 72, 2433. 5. Kempen, E.C., Brodbelt, J.S.,
Bartsch, R.A., Jang, Y., Kim, J.S., Anal.
Chem., 1999, 71, 5493.