Raman spectroscopic investigation of tetraethylammonium polybromides

Article abstract of DOI:10.1021/ic100869r

A large number of polyhalides, especially polyiodides, have been discovered and studied, but definitive studies on polybromides remain scarce. Br ₃^- is the only monovalent polybromide with a known structure. Higher-order monovalent polybromide anions have been proposed but not structurally confirmed as discrete species. In this study tetraalkylammonium polybromides with molecular formulas R₄NBr_2x+1 (R = ethyl; x = 1-4) were prepared by reacting tetraalkylammonium monobromide or tribromide salts with gas-phase bromine. Distinct and characteristic Raman spectra were obtained from the solid polybromides in the spectral range between 100 and 400 cm^-1. Experimental Raman spectra were compared to ab initio calculations to propose the structure of these polybromide anions. A general agreement between the experimental and theoretical results was observed. This study demonstrates that Raman spectroscopy is a sensitive technique for probing the structure of discrete monovalent polybromides.

Full text of DOI:10.1021/ic100869r

8
684 Inorg. Chem. 2010, 49, 8684–8689
DOI: 10.1021/ic100869r
Raman Spectroscopic Investigation of Tetraethylammonium Polybromides
Xiaoyun Chen,* Mark A. Rickard, John W. Hull Jr., Chao Zheng, Anne Leugers, and Petra Simoncic
Analytical Sciences, 1897 Building, The Dow Chemical Company, Midland, Michigan 48667
Received March 1, 2010
A large number of polyhalides, especially polyiodides, have been discovered and studied, but definitive studies on
polybromides remain scarce. Br3 is the only monovalent polybromide with a known structure. Higher-order mono-
-
valent polybromide anions have been proposed but not structurally confirmed as discrete species. In this study tetra-
alkylammonium polybromides with molecular formulas R NBr
(R = ethyl; x = 1-4) were prepared by reacting
2xþ1
tetraalkylammonium monobromide or tribromide salts with gas-phase bromine. Distinct and characteristic Raman
4
-
1
spectra were obtained from the solid polybromides in the spectral range between 100 and 400 cm . Experimental
Raman spectra were compared to ab initio calculations to propose the structure of these polybromide anions.
A general agreement between the experimental and theoretical results was observed. This study demonstrates that
Raman spectroscopy is a sensitive technique for probing the structure of discrete monovalent polybromides.
9
Introduction
infinite polybromide network. Tribromide is the only mono-
valent polybromide that has been systematically studied by
Anionic polyhalides are an important class of inorganic
ions that have found wide application in the field of super-
molecular architecture design. On the basis of conceptually
simple Lewis acid-base interactions, polyhalides may adopt
complex structures and serve as versatile building blocks. While
extensive research has been carried out for polyiodides, poly-
bromide anions have received less attention, despite many3
practical applications of polybromides for water treatment,
battery applications, selective bromination of alkenes,
and as a potential dopant for carbon nanotube property
10
crystallographic and vibrational spectroscopic methods. It
has been demonstrated that the Raman spectrum of a tri-
bromide salt can be correlated with its structure. The relative
intensities of the symmetric and antisymmetric Br-Br stretch-
ing modes and their frequencies were shown to be highly
sensitive to the bond angle and length. The existence of
higher-order monovalent polybromides (with five or more
bromine atoms per negative charge) was suggested based on
traditional analytical techniques such as titration, melting
1
2
4
5
6
11
modification.
point, and electrochemical methods. However, they have
The crystal structures of several polyvalent polybromides
not been confirmed as discrete species based on spectroscopic
evidence nor was any structural information available for
these monovalent polybromides. Raman bands around 230-
2- 7
-
2- 8
have beenreported, including Br10 , (Br ) (Br ), and an
2
4
-
1
2
50 cm were only tentatively assigned to pentabromides or
*
To whom correspondence should be addressed. E-mail: xchen4@dow.com.
1) Metrangolo, P.; Resnati, G.; Arman, H. D. Halogen bonding: funda-
mentals and applications; Springer: Berlin, 2008; pp 105-136.
2) (a) Svensson, P. H.; Kloo, L. Chem. Rev. 2003, 103, 1649–1684.
b) Deplano, P.; Ferraro, J. R.; Mercuri, M. L.; Trogu, E. F. Coord. Chem. Rev.
6a,10,12
(
polybromides in a few publications.
To the authors’
knowledge, this paper is the first systematic investigation of
polybromide anions. Tetraethylammonium was used as the
cation in this study, and its tribromide, pentabromide, hepta-
bromide, and nonabromide salts were prepared and char-
acterized by Raman spectroscopy. In addition, ab initio
calculations were carried out and compared to the experi-
mental results. The most likely structure of each polybromide
is proposed.
(
(
1
999, 188, 71–95.
3) Goodenough, R. D.; Mills, J. F.; Place, J. Environ. Sci. Technol. 2002,
(
3
, 854–856.
(
(
4) Singh, P.; Jonshagen, B. J. Power Sources 1991, 35, 405–410.
5) (a) Ruasse, M.-F.; Lo Moro, G.; Galland, B.; Bianchini, R.; Chiappe, C.;
Bellucci, G. J. Am. Chem. Soc. 1997, 119, 12492–12502. (b) Bellucci, G.;
Bianchini, R.; Chiappe, C.; Ambrosetti, R. J. Am. Chem. Soc. 2002, 111, 199–202.
ꢀ
(
6) (a) Liu, B.; Cui, Q.; Yu, M.; Zou, G.; Carlsten, J.; Wagberg, T.;
Sundqvist, B. J. Phys.: Condens. Matter 2002, 14, 11255–11259. (b) Sung, D.;
Park, N.; Park, W.; Hong, S. Appl. Phys. Lett. 2007, 90, 093502-3.
(
7) Cunningham, C. W.; Burns, G. R.; McKee, V. Inorg. Chim. Acta 1990,
67, 135–137.
8) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Hursthouse, M. B.;
(10) Schnittke, A.; Stegemann, H.; Fuellbier, H.; Gabrusenoks, J. J. Raman
Spectrosc. 1991, 22, 627–631.
1
(
(11) (a) Chattaway, F. D.; Hoyle, G. J. Chem. Soc., Trans. 1923, 123, 654–
662. (b) Bloch, R.; Farkas, L.; Schnerb, J.; Winogron, F. J. Phys. Colloid Chem.
2002, 53, 1117–1125. (c) Mercier, P. L.; Kraus, C. A. Proc. Natl. Acad. Sci.
U.S.A. 1956, 42, 65–67. (d) Rallo, F.; Silvestroni, P. J. Electrochem. Soc. 1972,
119, 1471–1474.
Huth, S. L.; Isaia, F.; Lippolis, V.; Mancini, A.; Ogilvie, H. Inorg. Chem.
Commun. 2005, 8, 79–82.
(
9) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Isaia, F.; Lippolis, V.;
Mancini, A.; Pala, L.; Slawin, A. M. Z.; Woollins, J. D. Chem. Commun.
003, 2226–2227.
2
(12) Evans, J. C.; Lo, G. Y. S. Inorg. Chem. 1967, 6, 1483–1486.
pubs.acs.org/IC
Published on Web 09/02/2010
r 2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 19, 2010 8685
Figure 1. Series of tetraethylammonium polybromide samples prepared via gas-phase reactions. The numbers below each sample refer to the molar ratios
of Br to Et NBr.
2
4
1
perturbation theory (MP2) was utilized with Pople’s split-
4
Experimental Section
15
valence double-ζ basis set withd polarization functions [6-31G(d)]
for geometry optimizations and frequency calculations. Calcu-
Materials and Polybromide Synthesis. Bromine and tetraethyl-
ammonium bromide were obtained from Aldrich and used without
further purification. Bromine and tetraethylammonium bromide
were gravimetrically measured to prepare the samples with varied
molar ratios of bromine to tetraethylammonium bromide. Liquid
bromine was placed in a small glass dish at the bottom of a glass
desiccator. Tetraethylammonium bromide was ground to a fine
powder with a mortar and pestle and was placed in a glass dish on a
porcelain tray in the desiccator above the bromine. The desiccator
lid was attached (silicone grease was used as the sealant) to allow
bromine vapors to contact the solid tetraethylammonium bromide.
After standing for 1-4 days at ambient temperature, all liquid
bromine in the bottom dish had evaporated. The resulting orange
to orange-red solid was removed from the dish and ground to a
finer powder consistency with a mortar and pestle before being
placed in glass vial for Raman spectral collection.
-
-
lations for Br
3
and Br
5
anions were carried out with the 6-31þ
G(d) basis set, which includes diffuse functions. Adding diffuse
functions increased the computational requirements but did not
improve the predicted harmonic frequencies (see the Supporting
Information for details). Although computational requirements
anion are low, MP2 Raman intensity calculations
-
for the Br
3
-
for the Br9 anion are prohibitively expensive with moderately
large basis sets like 6-31þG(d). Therefore, MP2/6-31G(d) cal-
culations were used for all polybromide anions.
Results
Polybromides have been demonstrated to form under
various conditions including in the solid state, in suspension,
11a
and in solution. Free bromine was observed to be released
Raman Spectroscopy Characterization. A Kaiser RXN1 Raman
system was used for all Raman spectral characterization. Because
of the potential hazards of bromine vapor, spectra were collected
from polybromides in sealed glass vials in a backscattering geo-
metry. Laser excitation at 785 nm with approximately 30 mW
power was delivered using a PhAT probe (Kaiser Optical Systems,
Inc.), which spread the laser power uniformly over a 6 mm spot
area to minimize local sample heating. All studied polybromides
had strong Raman signals, and care was taken to ensure that there
was no detector saturation.
from polybromides until the tribromide form was reached.
The volatility and hazards of bromine make it difficult to
study these species using traditional analytical techniques.
The formation of polybromides with three, five, seven, and
nine bromine atoms per anion has been systematically studied
by our group. A series of polybromides with apparent mole-
cular formulas (CH CH ) NBr
via gas-phase reactions between Br and Et NBr based on
procedures similar to those in a literature method. Poly-
bromides with intermediate molar ratios of Br to Et NBr
(n=1-5) were prepared
3
2 4
2nþ1
2
4
11a
Neutron Activation Analysis (NAA). The total bromine was
measured in tetraethylammonium polybromide samples by NAA.
The samples were prepared by weighing aliquots of approxi-
mately 20 and 30 mg of the bromide salts into polyethylene vials,
adding 5 mL of N,N-dimethylformamide (DMF), and heat-
sealing the vials. Triplicate bromine standards were prepared by
transferring aliquots of 4-bromobenzoic acid to 2 dram vials,
adding 5 mL of DMF, and heat-sealing the vials. All vials were
mixed on a vortex mixer until no visible solids were observed.
The prepared samples and standards were irradiated for 10 min
at a power level of 5 kW in the rotary specimen rack of The Dow
TRIGA Research Nuclear Reactor. Spectra of the emitted γ radi-
ation were acquired using two high-purity germanium detectors
coupled to a ND/Canberra analyzer system (MCA). The 554,
2
4
were also prepared. The resulting polybromides changed
color from orange to dark red as the Br content increased
2
(
Figure 1). The sample with a 5:1 Br /Et NBr molar ratio
2 4
was the only one with substantial bromine vapor in the vial
headspace. The bromine content was independently verified
with NAA (Table 1).
Raman Spectroscopy Characterization. Raman spectra
of the polybromides listed in Table 1 are shown in Figure 2.
The Raman spectrum from the sample with a 1:1 Br2/
Et NBr ratio is consistent with the literature results. It has
4
-1
a strong symmetric Br-Br stretching band at 163 cm
and a weak antisymmetric band at 198 cm . Upon
-
1
698, and 777 keV γ-rays from 82-Br were used to quantify the
bromine content using 4-bromobenzoic acid (Alfa Aesar lot
reaction with more Br , two new features are observed
2
-
1
at 210 and 253 cm . For the samples with 1.75 and 2:1
1
0003270) as the external standard.
Ab Initio Calculations. Calculations were performed with the
-
1
ratios, another band at 270 cm appears. This band is the
1
3
Gaussian03 software package. Second-order Møller-Plesset
(15) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,
(
13) Frisch, M. J.; et al. Gaussian03, revision C.02; Gaussian, Inc.: Wallingford,
CT, 2004.
14) Møller, C.; Plesset, M. S. Phys. Rev. 1934, 46, 618–622.
724. (b) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257.
(c) Hariharan, C.; Pople, J. A. Mol. Phys. 1974, 27, 209. (d) Hariharan, P. C.;
Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(

8
686 Inorganic Chemistry, Vol. 49, No. 19, 2010
Chen et al.
explained based on the sequential formation of tribromide,
pentabromide, heptabromide, and nonabromide. The spec-
tral evolution observed in situ from a heptabromide to a
nonabromide sample is shown in Figure 3 as an example.
Ab initio calculations provide more insight into the struc-
ture of these polybromide anions.
Ab Initio Calculations. Several structures for each
polybromide were optimized and are shown in Table 2.
A summary of the corresponding calculation results is
shown in Table 3. These structures were selected using the
VSEPR model (see the Supporting Information for
2
a
details) and known polyiodide structures. Closed-ring
structures were not stable for any polybromide. Higher-
level ab initio calculations at the MP2/6-31G(d) level were
carried out for the lowest-energy structure of each poly-
bromide, and Raman bands were assigned based on a
comparison of the theoretically and experimentally obser-
ved frequencies (Table 4). Low-frequency bands below
-
1
1
00 cm , which are generally deformations, were weak
Figure 2. Raman spectraofthe tetraethylammonium polybromidesam-
ples shown in Figure 1.
and are not listed in Table 4. Vibrational modes des-
cribed as terminal stretches primarily involve the outer
bromine atoms, and apex stretches involve the central
bromine atom.
Table 1. Summary of the Sample Preparation and NAA Results
x in (CH CH
3
2 4 x
) NBr
Br
2
/(CH
3
CH
2
)
molar ratio
4
NBr
Br wt %
(by NAA)
Discussion
expected
NAA results
Although it was difficult to prepare pure polybromides
based on the gas-phase reaction approach used here, the
Raman spectrum of pure polybromides could be obtained by
spectral subtraction. Shown in Figure 4 are “pure” Raman
spectra of all four polybromides after subtraction of neigh-
boring polybromide contributions, based on the assumption
that they share no common Raman bands. This assumption
is supported by the results observed when the bromination
process is monitored with in situ Raman spectroscopy (an
example involving heptabromide and nonabromide is shown
in Figure 3). It was found that, during a slow gas-phase
bromination process, the monobromide was sequentially
converted to tribromide, pentabromide, heptabromide, non-
abromide, and then to some even higher-order bromides,
which have not been successfully prepared via the batch-type
reaction described in the Experimental Section. The “pure”
spectra shown in Figure 4 are eigenspectra whose linear
combination can be used to reconstruct most experimentally
observed spectra. Normally only two neighboring poly-
bromides coexist at any stage of the bromination process
when bromine vapor is slowly exposed to the sample. The
only exception was observed during the conversion from
tribromide to pentabromide when the sample occasionally
became liquid at room temperature before resolidifying.
The liquid phase is likely a eutectic formed between tribro-
mide and some intermediate species between tribromide
and pentabromide (or pentabromide itself). Much broader
bands were observed from the liquid polybromide material,
and they could not be well characterized by Raman spectros-
copy. Further study will be needed to understand this
phenomenon.
1
1
1
1
2
2
2
3
4
.00
.25
.50
.75
.00
.25
.50
.00
.00
65.6 ( 1.5
67.3 ( 1.5
70.7 ( 1.5
72.6 ( 1.5
72.8 ( 1.5
74.4 ( 1.5
78.5 ( 1.5
81.0 ( 1.5
83.0 ( 2.0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
7.0
9.0
3.10 (2.91-3.32)
3.35 (3.13-3.59)
3.93 (3.66-4.23)
4.31 (4.01-4.66)
4.36 (4.04-4.71)
4.73 (4.38-5.13)
5.95 (5.45-6.52)
6.94 (6.32-7.68)
7.95 (6.94-9.23)
dominant feature for the sample with a 3:1 Br /Et NBr
2
4
-
1
molar ratio. Two other weaker bands at 165 and 209 cm
are also discernible from that sample. Upon further in-
corporation of Br , the sample with a 4:1 Br /Et NBr
molar ratio has two bands at 257 and 276 cm , and they
remain unchanged when the ratio is raised to 5:1.
The spectral changes in Figure 2 provide strong evi-
dence that discrete polybromide species are sequentially
formed as the Br /Et NBr molar ratio is changed from
to 4, with distinct and characteristic Raman features obser-
ved for samples with 1:1 (tribromide), 3:1 (heptabromide),
and 4:1 (nonabromide) ratios. The samples with 1.75:1
and 2:1 ratios, however, have a band at 270 cm that can
be attributed to heptabromide. The fact that this band
was absent from samples with 1.25:1 and 1.5:1 ratios fur-
ther confirms that the 270 cm band should be attributed
to heptabromide instead of pentabromide. The presence
of heptabromide in a sample with a 2:1 ratio could be a
2
2
4
-
1
2
4
1
-
1
-
1
result of equilibrium between Br , tribromide, pentabro-
2
mide, and heptabromide. It could also be a consequence
of a heterogeneous reaction between tribromide and Br2,
leading to a mixture of pentabromide and heptabromide
in the final sample.
To further investigate the sequential formation of these
polybromides, in situ Raman spectra were also collected
during the bromination and debromination processes.
All of the polybromide anions in Figure 4 have relatively
simple Raman spectra, which indicates that these anions have
high symmetry. A comparison between the theoretical and
experimental results is summarized in Figure 4 and Table 4.
Good agreement was found between the theoretical and
experimental frequencies with the exception of heptabromide.
Although the Br loading information from such in situ
2
experiments was unavailable, the spectral changes can be

Article
Inorganic Chemistry, Vol. 49, No. 19, 2010 8687
Figure 3. Representative Raman spectra collected during in situ bromination from heptabromide to nonabromide (left panel) and the corresponding
debromination (right panel). Spectra 1-5 were collected sequentially.
Table 2. Potential Structures for the Polybromide Anions
-
Table 3. Theoretical Calculation Summary for the Polybromide Anions
Listed in Table 2
symmetry for Br7 provides acceptable agreement with the
experimental results, but a better correlation may require the
inclusion of the cation in the calculation. A tetrahedral
relative
energy
-
method/
basis set
point
group (kJ/mol)
imaginary
frequency
structure with T symmetry is proposed for the Br9 anion.
d
molecule label
These structures are similar to but not identical with their
2
-
polyiodide counterparts.
Br
Br
3
1
1
2
3
1
2
3
4
5
1
2
3
4
5
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
MP2/6-31G(d)
HF/6-31G(d)
HF/6-31G(d)
HF/6-31G(d)
HF/6-31G(d)
HF/6-31G(d)
D
¥h
2v
¥h
0
0
2
0
0
0
1
0
1
0
1
0
3
2
-
-
5
C
D
0.0
17.2
21.9
0.0
14.7
19.6
31.3
125.8
0.0
35.6
45.1
49.8
240.1
Many polyiodide structures can be described as iodide (I )
-
2a
or triiodide (I3 ) coordinated with iodine (I ). Likewise, the
2
C
C
C
C
C
2v
3v
2h
2v
2v
-
-
-
-
lowest-energy structures for Br5 , Br7 , and Br9 have a
central Br coordinated with two or more Br molecules.
Calculated bond lengths for Br and Br3 to Br9 are shown
Br
7
9
-
2
-
-
2
-
in Table 5. Bond lengths for the outer bromine atoms in Br
,
5
D
2h
-
-
-
Br7 , and Br9 are much shorter than the inner bond lengths
Br
T
C
C
d
and are similar to those of Br . As the series progresses from
s
2
-
-
2v
Br5 to Br9 , the outer bonds grow shorter and the inner
bonds grow longer.
D
D
3h
4h
While earlier papers postulate the existence of pentabro-
mides, heptabromides, and nonabromides mainly based on
the reactant composition, there was no experimental evidence
regarding whether these polybromides were indeed discrete
species or simply composed of monobromides/tribromides
with loosely bound bromine. The Raman spectroscopic
results presented above provide strong evidence that discrete
monovalent polybromide anions exist and the transitions
1
6
The tribromide results agree with literature observations.
The pentabromide structure is proposed to be V-shaped with
C2v symmetry. The trigonal-pyramidal structure with C3v
(
16) Schnittke, A.; Stegemann, H.; Fullbier, H. J. Raman Spectrosc. 1991,
2
2, 627–631.

8
688 Inorganic Chemistry, Vol. 49, No. 19, 2010
Chen et al.
a
Table 4. Comparison between the Experimental and Theoretical Results for the Lowest-Energy Polybromide Anions
frequency
calcd intensity
4
Raman (A /amu)
˚
molecule
symmetry
band assignment
exptl
calcd
IR (km/mol)
-
gþ
Br
3
5
Σ
Σ
A
sym. stretch
asym. stretch
163
198
253
171
218
252
147
217
136
282
135
250
109
289
265
121
78.1
þ
u
219.7
76.8
23.0
211.5
172.3
47.7
22.7
123.2
70.7
-
-
-
Br
1
1
1
terminal sym. stretch
apex sym. stretch
terminal asym. stretch
apex asym. stretch
terminal sym. stretch
apex sym. stretch
terminal asym. stretch
apex asym. stretch
terminal sym. stretch
terminal asym. stretch
apex asym. stretch
139.9
19.8
136.8
1.2
171.0
6.0
245.5
25.8
B
A
E
A
2
210
270
Br
7
Br
9
276
257
438.8
425.3
82.2
T
2
140.6
69.5
a
Only strong bands from the experimental results are listed.
-
-
Table 5. Calculated Bond Lengths for Br
2
and Br
3
to Br
9
a
bond lengths (pm)
molecule
inner atoms
outer atoms
Br
Br
Br
Br
Br
2
3
5
7
9
231
258
-
-
-
-
271
282
285
246
240
239
a
MP2/6-31G(d), from calculations in Table 4 except for Br
2
.
samples contained in sealed glass vials. Far-IR experi-
ments were also attempted by pressing polybromide sam-
ples in a polyethylene powder. It was not possible to
prevent bromine vapor evolution except for tribromides.
Some examples of far-IR spectra are shown in the Sup-
porting Information.
X-ray diffraction (both powder and single-crystal) experi-
ments were attempted on these polybromide samples to
obtain definitive structural information. However, the
results were inconclusive because of the difficulty of the
sample preparation and handling. Powder X-ray diffrac-
tion results indicate that all polybromides studied here
were highly crystalline. It was difficult, however, to ascer-
tain the true bromine loading in these samples because of
the slow release of bromine. Differential scanning calo-
rimetry (DSC) characterization was also attempted with
polybromide samples in small flame-sealed ampoules.
DSC results further confirmed the highly crystalline nature
of these polybromides. Multiple sharp phase transitions
were observed because most samples were a mixture
of polybromides. This observation, combined with the
possibility of eutectic formation between various poly-
bromides, made it difficult to use DSC to gain further
insights into these polybromides. Future X-ray diffrac-
tion and DSC studies of these polybromides may yield
valuable structural information if these challenges are
overcome.
Figure 4. Comparison of the experimental Raman spectra (after sub-
traction of neighboring polybromides’ contributions) and theoretical
Raman spectra.
Conclusion
In conclusion, we demonstrate for the first time that
discrete monovalent pentabromides, heptabromides, and
nonabromides exist based on their characteristic Raman
spectra. These spectra were compared to theoretically calcu-
lated spectra to further acquire structural insights into these
between them are relatively sharp. Because of the high
volatility of bromine, Raman spectroscopy with the PhAT
probe’s diffuse illumination was found to be the best char-
acterization tool because it allows spectral collection from

Article
Inorganic Chemistry, Vol. 49, No. 19, 2010 8689
monovalent polybromide anions. Such results provide a frame-
work for the future characterization of various polybromides
in various environments and applications.
Jamie Stanley, Bryan Haskins, Dee Dickerson, and Dow’s
High Performance Computing Cluster.
Supporting Information Available: Far-IR spectra of tribro-
mide, pentabromide, and heptabromide and ab initio calcula-
tion details. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. The authors are grateful for inspir-
ing discussions with Joseph Bonadies, Jr., John Kramer,
Xue Han, and Mark Beach and for technical support from