Poly(diiododiacetylene): Preparation, isolation, and full characterization of a very simple poly(diacetylene)

Article abstract of DOI:10.1021/ja8011403

Poly(diiodiacetylene), or PIDA, is a conjugated polymer containing the poly(diacetylene) (PDA) backbone but with only iodine atom substituents. The monomer diiodobutadiyne (1) can be aligned in the solid state with bis(nitrile) oxalamide hosts by hydrogen bonds between oxalamide groups and weak Lewis acid-base interactions (halogen bonds) between nitriles and iodoalkynes. The resulting cocrystals start out pale blue but turn shiny and copper-colored as the polymerization progresses. The development of a crystallization methodology that greatly improves the yield of PIDA to about 50% now allows the full characterization of the polymer by X-ray diffraction, solid-state ¹³C MAS NMR, Raman, and electron absorption spectroscopy. Comparison of a series of hosts reveals an odd-even effect in the topochemical polymerization, based on the alkyl chain length of the host. In the cocrystals formed with bis(pentanenitrile) oxalamide (4) and bis(heptanenitrile) oxalamide (6), the host/guest ratio is 1:2 and the monomer polymerizes spontaneously at room temperature, while in the case of bis(butanenitrile) oxalamide (3) and bis(hexanenitrile) oxalamide (5), where the host and guest form cocrystals in a 1:1 ratio, the polymerization is disfavored and does not go to completion. The topochemical polymerization can also be observed in water suspensions of micrometer-sized 6.1 cocrystals; the size distribution of these microcrystals, and the resulting polymer chains, can be controlled by sonication. Completely polymerized PIDA cocrystals show a highly resolved vibronic progression in their UV/vis absorption spectra. Extensive rinsing of the crystals in organic solvents such as methanol, THF, and chloroform separates the polymer from the soluble host. Once isolated, PIDA forms blue suspensions in a variety of solvents. The UV/vis absorption spectra of these suspensions match the cocrystal spectrum, without the vibronic resolution. However, they also include a new longer-wavelength absorption peak, associated with aggregation of the polymer chains.

Full text of DOI:10.1021/ja8011403

Published on Web 05/20/2008
Poly(diiododiacetylene): Preparation, Isolation, and Full
Characterization of a Very Simple Poly(diacetylene)
Liang Luo, Christopher Wilhelm, Aiwu Sun, Clare P. Grey, Joseph W. Lauher, and
Nancy S. Goroff*
Department of Chemistry, State UniVersity of New York at Stony Brook,
Stony Brook, New York 11794-3400
Received February 14, 2008; E-mail: nancy.goroff@stonybrook.edu
Abstract: Poly(diiodiacetylene), or PIDA, is a conjugated polymer containing the poly(diacetylene) (PDA)
backbone but with only iodine atom substituents. The monomer diiodobutadiyne (1) can be aligned in the
solid state with bis(nitrile) oxalamide hosts by hydrogen bonds between oxalamide groups and weak Lewis
acid-base interactions (halogen bonds) between nitriles and iodoalkynes. The resulting cocrystals start
out pale blue but turn shiny and copper-colored as the polymerization progresses. The development of a
crystallization methodology that greatly improves the yield of PIDA to about 50% now allows the full
characterization of the polymer by X-ray diffraction, solid-state 13C MAS NMR, Raman, and electron
absorption spectroscopy. Comparison of a series of hosts reveals an odd-even effect in the topochemical
polymerization, based on the alkyl chain length of the host. In the cocrystals formed with bis(pentanenitrile)
oxalamide (4) and bis(heptanenitrile) oxalamide (6), the host/guest ratio is 1:2 and the monomer polymerizes
spontaneously at room temperature, while in the case of bis(butanenitrile) oxalamide (3) and bis(hexaneni-
trile) oxalamide (5), where the host and guest form cocrystals in a 1:1 ratio, the polymerization is disfavored
and does not go to completion. The topochemical polymerization can also be observed in water suspensions
of micrometer-sized 6·1 cocrystals; the size distribution of these microcrystals, and the resulting polymer
chains, can be controlled by sonication. Completely polymerized PIDA cocrystals show a highly resolved
vibronic progression in their UV/vis absorption spectra. Extensive rinsing of the crystals in organic solvents
such as methanol, THF, and chloroform separates the polymer from the soluble host. Once isolated, PIDA
forms blue suspensions in a variety of solvents. The UV/vis absorption spectra of these suspensions match
the cocrystal spectrum, without the vibronic resolution. However, they also include a new longer-wavelength
absorption peak, associated with aggregation of the polymer chains.
Introduction
sensing materials; many PDAs undergo a visible color change
in response to environmental perturbations like temperature, pH,
Conjugated organic polymers have attracted great interest for
their electronic and optical properties and for potential applica-
tions including organic light-emitting diodes (OLEDs),1–3 field-
effect transistors,4 and chemical and biological sensors.5,6
Poly(diacetylenes) (PDAs) are a widely investigated class of
conjugated polymers composed of an alternating poly(ene-yne)
structure. The nonlinear optical properties resulting from this
quasi-one-dimensional π system make PDAs excellent mul-
tiphoton absorbers.7–9 In addition, they are good candidates for
and ligand-receptor interactions.10–13
Recently, we reported the first preparation of poly(diiododi-
acetylene), or PIDA, a poly(diacetylene) with single iodine atom
substitutents.14 Compared with other PDAs that have been
reported, PIDA has a nearly unadorned backbone and therefore
offers the potential for new insights into the inherent optical
and electronic properties of PDAs without side-chain interfer-
ence. In addition, PIDA can be considered as an iodinated
carbyne, and thermally or electrically induced carbon-iodine
bond dissociation may provide a route from PIDA to structurally
homogeneous and fully characterized linear carbyne. PIDA also
represents a potent precursor for general PDA preparation via
post-polymerization modification at the iodinated sites.
Because of the simplicity of its structure, PIDA can provide
insights into the color transitions observed in other PDAs, which
(1) Samuel, I. D. W.; Turnbull, G. A. Chem. ReV. 2007, 107, 1272–1295.
(2) Kulkarni, A. P.; Tonzola, C. J.; Babel, A.; Jenekhe, S. A. Chem. Mater.
2004, 16, 4556–4573.
(3) Gross, M.; Muller, D. C.; Nothofer, H. G.; Scherf, U.; Neher, D.;
Brauchle, C.; Meerholz, K. Nature 2000, 405, 661–665.
(4) Zaumseil, J.; Sirringhaus, H. Chem. ReV. 2007, 107, 1296–1323.
(5) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. ReV. 2000, 100,
2537–2574.
(6) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem. ReV. 2007, 107,
1339–1386.
(10) Jonas, U.; Shah, K.; Norvez, S.; Charych, D. H. J. Am. Chem. Soc.
1999, 121, 4580–4588.
(7) Carter, G. M.; Thakur, M. K.; Chen, Y. J.; Hryniewicz, J. V. Appl.
Phys. Lett. 1985, 47, 457–459.
(11) Mino, N.; Tamura, H.; Ogawa, K. Langmuir 1991, 7, 2336–2341.
(12) Charych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D. Science
1993, 261, 585–588.
(8) Abe, S.; Schreiber, M.; Su, W. P.; Yu, J. Phys. ReV. B 1992, 45, 9432–
9435.
(13) Reppy, M. A.; Pindzola, B. A. Chem. Commun. 2007, 4317–4338.
(14) Sun, A. W.; Lauher, J. W.; Goroff, N. S. Science 2006, 312, 1030–
1034.
(9) Guo, D.; Mazumdar, S.; Dixit, S. N.; Kajzar, F.; Jarka, F.; Kawabe,
Y.; Peyghambarian, N. Phys. ReV. B 1993, 48, 1433–1459.
9
7702 J. AM. CHEM. SOC. 2008, 130, 7702–7709
10.1021/ja8011403 CCC: $40.75
2008 American Chemical Society

Preparation and Characterization of Poly(diiododiacetylene)
A R T I C L E S
Scheme 1. Synthesis of Hosts 4-6
on the stronger pyridine-iodine halogen bonds, but in these
cocrystals the monomers are even harder to polymerize.
Although PIDA can be obtained by subjecting cocrystals 7·1
and 8·1 to high pressure of 3-6 GPa,21 the increased mosaicity
of the crystals and the complexity of the pressure method limit
the potential for further studies and applications of PIDA.
We report here a new methodology for preparing cocrystals
of 1 with host 4, as well as hosts 5 and 6, so that PIDA crystals
with well-defined structures can be produced reliably and in
large quantities. This development makes it realistic to carry
out studies of the unique optical, electronic, and chemical
properties of the material and to explore using PIDA as a starting
material for the synthesis of novel PDAs, especially those
without any appropriate “handle” to control their solid-state
assembly for topochemical polymerization. This paper presents
new conditions for the formation of PIDA via single-crystal-
to-single-crystal topochemical polymerization and a more
comprehensive characterization of PIDA than was previously
possible.
Figure 1. Diiodobutadiyne (1) may be aligned by Lewis-basic hosts 3-8
for topochemical polymerization to form PIDA 2.
have been attributed to side-chain-induced conformational
changes in the PDA backbone.15 Chromatic changes in PDAs
in response to environmental stimuli have served as the basis
for the majority of PDA-based sensing materials. However, to
elucidate the mechanisms of the color and emission changes in
PDA materials has proven to be a great challenge. Studies have
shown that the factors affecting the chromic state and the
susceptibility to chromic change of PDA materials include
rotational changes in the polymer backbone,15 changes in the
structure of the side-chain assembly,16 and hydrogen-bonding
interactions at the side-chain head-groups.10,17 The aggregation
states and the transition of the polymer backbone from extended
rod-like segments to discrete coils18 further confuse discussion
of the interplay between the conjugated polymer backbone and
the ensemble material structure in determining the observed
optical properties. PIDA has no side chains, presenting an
important test case in the study of PDA optical properties.
The preparation of PDAs requires assembly of the monomer
diynes in an appropriate arrangement, either on their own or
by interaction with cocrystallizing agents, to favor the head-
to-tail 1,4-polymerization.19,20 To align monomer diiodob-
utadiyne (1) with the required spacing for this topochemical
polymerization, we use host-guest supramolecular scaffolds
based on halogen bonds, the Lewis acid-base interactions
between 1 and functionalized Lewis bases (see Figure 1). In
our report of the first preparation of PIDA,14 bis(pentanenitrile)
oxalamide (4) was employed to produce cocrystals with
monomer 1, in which 1 spontaneously polymerizes at room
temperature. However, preparing large quantities of PIDA
proved challenging due to the insolubility of the host and the
relatively weak halogen bonds between the host and guest.
Bispyridyl oxalamides 7 and 8 also form cocrystals with 1 based
Results and Discussion
Synthesis of Hosts and Guest. Hosts 4-6 were prepared from
amines 11a-c (see Scheme 1) as described previously for host
4.14 The route to the amines was altered slightly from the
previously reported method, replacing H2O/triphenylphospine
reduction of azide 1022 with a cobalt(II) chloride-catalyzed
sodium borohydride reduction because of the mild reaction
conditions, high chemoselectivity, and efficiency of this
method.23 The monomer diiodobutadiyne (1) was prepared in
one step from commercially available 1,4-bis(trimethylsilyl)-
butadiyne by silver nitrate-catalyzed iodination with N-iodo-
succinimide.24
Preparation of Host-Guest Cocrystals. In previous attempts,
crystallization of monomer 1 with host 4 gave crystals of varying
quality and polymer yield. Investigation of a variety of crystal-
lization techniques revealed the importance of removing all
particulate impurities from the solution at the start of the
recrystallization. In the optimized method, the host is first
dissolved in methanol, and then diyne 1 is added to make a 1:1
or 1:2 host/guest ratio. The solution is subjected to centrifugation
and then decanted and transferred to a shallow crystallization
dish. The dish is covered with needle-punctured aluminum foil,
(15) Schott, M. J. Phys. Chem. B 2006, 110, 15864–15868.
(16) Dautel, O. J.; Robitzer, M.; Lere-Porte, J. P.; Serein-Spirau, F.; Moreau,
J. J. E. J. Am. Chem. Soc. 2006, 128, 16213–16223.
(17) Kew, S. J.; Hall, E. A. H. Anal. Chem. 2006, 78, 2231–2238.
(18) Chu, B. Xu, R. L. Acc. Chem. Res. 1991, 24, 384–389. Wenz, G.;
Müller, M. A.; Schmidt, M.; Wegner, G. Macromolecules 1984, 17,
837–850.
(21) Wilhelm, C.; Boyd, S. A.; Chawda, S.; Fowler, F. W.; Goroff, N. S.;
Halada, G. P.; Grey, C. P.; Lauher, J. W.; Luo, L.; Martin, C. D.;
Parise, J. B.; Tarabre, C.; Webb, J. A. J. Am. Chem. Soc. 2008, 130,
4415–4425.
(22) Da Ros, T.; Prato, M.; Lucchini, V. J. Org. Chem. 2000, 65, 4289–
4297.
(19) Baughman, R. H. J. Polym. Sci. Part B-Polym. Phys. 1974, 12, 1511–
1535.
(23) Fringuelli, F.; Pizzo, F.; Vaccaro, L. Synthesis-Stuttgart 2000, 646–
650.
(20) Wegner, G. Naturforsch. B 1969, B24, 824ff.
(24) Gao, K.; Goroff, N. S. J. Am. Chem. Soc. 2000, 122, 9320–9321.
9
J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008 7703

A R T I C L E S
Luo et al.
Figure 3. (A) Blue single crystal grown from a solution of diyne 1 and
host 4 at 4 °C. (B) The same crystal, which turned shiny and copper-colored
after staying at room temperature for 1 week. The corresponding crystal
structures are shown below.
the long dimension corresponding to the (100) axis, the direction
of polymer growth. The slow evaporation of solvent from a
solution of diyne 1 and host 5 or 6 produces similar blue crystals
but in greater yields, which we attribute to the longer side chains
and the resulting better solubility of the hosts. No highly
reflective material is isolated initially from either system. The
pale blue cocrystals made with host 5 (Figure 2B) turned dark
gray after staying at room temperature for 1 week, while those
made with host 6 transformed to a shiny gold color under the
same conditions, very similar to the cocrystals made with host
4. Scanning electron microscopy (SEM) images of the gold
crystals made with host 6 reveal that they have very smooth
surfaces, with one dimension (010) much smaller than the other
two dimensions (inset, Figure 2C).
To identify the structure of the blue material produced with
host 4, crystals were grown at 4 °C, thereby reducing the solvent
evaporation rate. The blue crystals acquired under these condi-
tions are similar in appearance to those grown at room
temperature, but they are larger, allowing for structural deter-
mination by X-ray diffraction. Diffraction studies demonstrate
that these initial cocrystals contain primarily the monomer
(Figure 3A), analogous to the polymer cocrystals. In the 4·1
cocrystals, half the iodine atoms of 1 are halogen-bonded to
the nitrile nitrogen atoms, and half are in close contact with
the oxalamide oxygens. Compared to those in the polymer
cocrystals, the halogen bonds of the monomer have a more
favorable geometry, with an N-I-C angle of 177° and an
O-I-C angle of 164° (compared to 165° and 131°, respectively,
in the polymer cocrystal, Figure 4). These observations match
our expectation that the arrangement of host and guest in the
cocrystal is determined by interactions of the monomer, before
polymerization begins.
Figure 2. Microscopic images of host-guest cocrystals from hosts 4 (A),
5 (B), and 6 (C). Insets in panels A and C are SEM images of corresponding
cocrystals.
and the methanol is allowed to evaporate slowly. Blue crystals
appear when all the solvent has evaporated. For cocrystals with
hosts 4 and 6, a 1:2 ratio works best, and the initially formed
blue crystals become shiny and copper- or gold-colored after
about 1 week at room temperature. In the case of host 4, some
of the cocrystals are highly reflective and copper-colored even
at the start. With host 5, a 1:1 ratio is better, leading to blue
crystals that turn dark over time but never become shiny (Figure
2).
Single-Crystal-to-Single-Crystal Topochemical Polymeri-
zation. The improvements in the crystallization method increase
the reliability and yield for the production of cocrystals. The
first crystals isolated upon evaporation of solvent from a solution
of diyne 1 and host 4 include some copper-colored, shiny
material. These highly reflective crystals tend to be the largest
and highest quality among those isolated, and their structure
matches our previous report for the 4·2 cocrystals. The blue
crystals that make up the bulk of the material are not of X-ray
quality. However, during a week at room temperature, the blue
crystals steadily darken in color, finally becoming shiny and
very similar in appearance to previously isolated copper-colored
crystals (Figure 2A). The cocrystals look like filaments, with
Note that the blue color of the initial cocrystal indicates the
presence of some polymer. Weiser has demonstrated that PDAs
forming in a monomer crystal can be observed by UV/vis
absorption at very low mole fractions, <10-5 25
Initially, the
.
polymerization concentration is too low to be detected by X-ray
diffraction, but it is still observable by the eye. During a week
at room temperature, these crystals undergo a transition in
(25) Weiser, G.; Berrehar, J. Phys. ReV. Lett. 2007, 99.
9
7704 J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008

Preparation and Characterization of Poly(diiododiacetylene)
A R T I C L E S
13C MAS NMR. The insolubility of the PIDA cocrystals
makes it difficult to characterize them by solution-phase NMR.
However, magic-angle spinning NMR (MAS NMR) techniques
allow for NMR characterization of the solid samples, particularly
the cocrystals formed with hosts 5 and 6 that cannot be identified
completely by X-ray diffraction. The 13C MAS NMR spectrum
of cocrystal 4·1, characterized previously by X-ray diffraction,
contains peaks at 161.2, 122.6, 39.2, 29.2, 24.6, and 16.7 ppm
(corresponding to the host) and at 80.0 ppm (monomer
ꢀ-carbon), while the signal for the R-carbon of the monomer, a
broad peak in the range of 10-20 ppm,21 is hidden by the host
methylene carbon peaks, as shown in Figure 5A. The spectrum
of polymer cocrystal 4·2 exhibits two new peaks at 110.6 and
80.1 ppm, attributed to the ꢀ (sp) and R (sp2) carbons of PIDA,
respectively (Figure 5B). The polymer R-carbon has a chemical
shift very similar to that of the monomer ꢀ-carbon, but it is
much broader because of the interaction of the R-carbon with
the quadrupolar iodine atom.
Using these 13C NMR spectra as reference, we can determine
the composition of the cocrystals formed with hosts 5 and 6.
The spectrum of the dark blue, week-old cocrystals with host
5, as shown in Figure 5C, does not include clear resonances
corresponding to either monomer or polymer carbons. Instead,
there is a very broad peak or collection of peaks in the range of
70-100 ppm. After heating of the dark cocrystals in vacuum
at 30 °C for 1 month, the NMR spectrum shows a new resonance
at 110 ppm and a broader signal centered at 81 ppm (Figure
5D). The cocrystals with host 5 evidently polymerize very
slowly at room temperature, turning dark gradually. Heating
the samples increases the rate of polymerization, but the crystals
do not regain crystallinity. Figure 5E shows the NMR spectrum
of the gold cocrystals with host 6, with one peak at 109.6 ppm
and another double-peak signal centered at 85.0 ppm. The reason
is not yet clear for this splitting in the peak attributed to the
R-carbon, but this double peak integrates at a ratio of ∼1:1 with
the peak at 110 ppm, corresponding to the ꢀ-carbon, indicating
the formation of fully polymerized PIDA within the cocrystals.
Figure 4. The lighter background shows the structure of the monomer
diiodobutadiyne cocrystal; the bold foreground drawing shows the structure
of the resulting PIDA. The symbols in the figure represent the halogen bond
angle changes between guest and host (blue, monomer; red, polymer; dashed
line, pointing out of the plain). The table below summarizes the changes in
the crystal structure upon polymerization.
appearance from opaque blue to copper-colored and highly
reflective, with a concurrent transformation in the crystal
structure to the previously observed PIDA cocrystal. This single-
crystal-to-single-crystal topochemical polymerization is shown
in Figure 3, while Figure 4 indicates the intermolecular changes
that occur as a result of the polymerization. As the mole fraction
of polymer nears 100%, the crystals take on a characteristic
metal-like appearance.
Within the 4·1 cocrystals, the repeat distance of aligned
monomers is 4.98 Å, the C4-C1′ contact distance is 3.88 Å,
and the molecular tilt angle is 52°. After polymerization, the
repeat distance shortens to 4.94 Å. The topochemical polym-
erization occurs spontaneously at room temperature, requiring
relatively little atomic motion, other than movement of the
carbon atoms that are directly involved in the bonding changes,
as demonstrated in Figure 4.
Efforts to determine the structures of the cocrystals obtained
with host 5 and 6 by X-ray diffraction have been inconclusive.
The mosaicity of the cocrystals prevents complete analysis, but
the diffraction data do provide well-defined unit cells of both
cocrystals, allowing us to determine the ratio of host to guest
in each case. In the case of cocrystal 5·1, assuming a host/
guest ratio of 1:1, the measured non-hydrogen atomic volume
is 19.1 Å3, and the calculated density is 1.94 Mg/m3, both values
that match well with those for previous fully determined
cocrystals. In contrast, the unit cell parameters of cocrystal 6·2
suggest a 1:2 ratio of host and guest within the cocrystal; the
non-hydrogen volume (22.0 Å3) and calculated density (2.02
Mg/m3) for a 1:2 stoichiometry are both consistent with those
observed in cocrystal 4·2 (Table S1, Supporting Information).
Within cocrystal 6·2, the iodine atoms are resolvable, and they
are consistent with the proposed all-planar polymer structure.
Raman Spectroscopy. Raman spectroscopy is a sensitive tool
for identifying the alternating ene-yne structure of poly(diacety-
lenes). Figure 6 shows the progressive change in the Raman
spectrum of cocrystals with host 4 over the course of the
topochemical polymerization of 1. The Raman spectrum of the
initial blue 4·1 cocrystals has a relatively low scattering
intensity, shown as the blue spectrum in Figure 6. The band at
1413 cm-1, corresponding to a carbon-carbon double bond
stretching mode, indicates the presence of some polymer,
consistent with the crystals’ blue color. After staying at room
temperature for 3 days, when the crystals have turned deeper
blue and have an approximately 50% degree of polymerization,
as determined by X-ray diffraction, the Raman scattering
intensity has increased greatly, and three major bands are visible
at962,1415,and2067cm-1,correspondingtothecarbon-carbon
single bond, double bond, and triple bond stretches, respectively
(ν(C-C), ν(CdC), and ν(Ct C)). When the crystals have
polymerized fully, gaining a coppery appearance, the Raman
intensity (red spectrum, Figure 6) increases to about 100 times
that of the initial spectrum. Here, ν(C-C) is found at 971 cm-1
,
ν(CdC) is at 1415 cm-1, and ν(C≡C) is at 2071 cm-1
.
These observations are consistent with the increased concen-
tration of polymer 2 within these cocrystals. The polymer itself
exhibits a very strong Raman spectrum because of the large
polarizability of the PDA backbone. Initially, the cocrystals
contain only small amounts of polymer. As the reaction
9
J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008 7705

A R T I C L E S
Luo et al.
Figure 6. Raman spectra of monomer (blue), partially polymerized (green),
and polymer (red) cocrystals made by host 4.
Table 1. Odd-Even Effect of Host Side Chain Lengths on
Topochemical Polymerization of Guests 1
host-guest
stoichiometry
repeat
distance (Å)
spontaneous
polymerization
hosta
colorb
3
4
5
6
brown to dark
blue to coppery
blue to dark
1:1
1:2
1:1
1:2
5.25
4.94
5.07
4.89
no
yes
no
blue to gold
yes
a Host compound numbers, corresponding to the number of CH2
groups in the host side chains. b Color transitions of cocrystals staying at
room temperature for 1 week.
from the less polarizable monomer or short intermediate
oligomers.
Odd-Even Effect on Topochemical Polymerization. The
results of the X-ray diffraction, 13C MAS NMR, and Raman
spectroscopy studies demonstrate an odd-even effect based on
the length of the alkyl chains in the host molecules (see Table
1). Host 3 forms cocrystals with monomer 1 in a ratio of 1:1,
and the repeat distance of packed monomers in these cocrystals
is 5.25 Å.14 These brown cocrystals turn dark and decompose
at room temperature, with the Raman spectrum showing no
evidence of polymerization (Figure S5, Supporting Information).
The cocrystals of host 5 with diyne 1 have the same host-guest
stoichiometry of 1:1. Their initial blue appearance indicates that
polymer chains have been formed at low concentration in a
monomer matrix. These 5·1 cocrystals turn dark at room
temperature but do not gain the metallic appearance character-
istic of a high degree of polymerization. Only after the cocrystals
have been heated for an extended period can the polymer be
detected by 13C MAS NMR. As demonstrated by the increase
in mosaicity of the crystals, the host does not provide an ideal
scaffold for single-crystal-to-single-crystal polymerization of
monomer 1.
In contrast, the cocrystals with hosts 4 and 6, which each
contain an even number of methylene groups in their alkyl side
chains, have a stoichiometry of two monomers for every host
molecule. These cocrystals undergo spontaneous topochemical
polymerization as single crystals. Odd-even effects have been
reported previously in the polymerization and chromism of PDA
materials.26–28 Our results provide a new example in which the
alkyl chain’s parity determines the packing of host and guest
Figure 5. Solid-state 13C NMR spectra of cocrystals: (A) cocrystal 4·1;
(B) cocrystal 4·2; (C) black cocrystals prepared with host 5; (D) cocrystals
from panel C after being heated under vacuum at 30 °C for 1 month; and
(E) gold cocrystals with host 6. Blue asterisk marks the peak for monomer
1; red asterisks mark signals for PIDA 2.
proceeds, the spectrum for the polymer increases in intensity.
However, the energies of the Raman scattering peaks do not
change significantly, indicating that the observed signal, even
at the start, comes from the low concentration of polymer in
the sample. The scattering from the polymer dwarfs any signal
(26) Fujita, N.; Sakamoto, Y.; Shirakawa, M.; Ojima, M.; Fujii, A.; Ozaki,
M.; Shinkai, S. J. Am. Chem. Soc. 2007, 129, 4134–5.
(27) Tachibana, H.; Kumai, R.; Hosaka, N.; Tokura, Y. Chem. Mater. 2001,
13, 155–158.
9
7706 J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008

Preparation and Characterization of Poly(diiododiacetylene)
A R T I C L E S
Figure 7. UV/vis absorption spectrum of gold 6·2 cocrystals, plotted as a
function of wavenumber.
in the cocrystals and the success of the desired polymerization.
When the host has an odd number of methylene groups in each
alkyl chain, each iodine atom in the guest forms a halogen bond
with a terminal nitrogen of the guest, as designed. When the
number of methylene groups in each host side chain is even,
the stoichiometry shifts to 2:1, and one iodine atom in each
guest is in close contact with a host carbonyl oxygen. The
interaction between iodine and oxalamide oxygen is not likely
to provide significant stabilization, given that the carbonyls are
also simultaneously acting as hydrogen-bond acceptors. There-
fore, the difference in observed stoichiometry must come from
different spatial requirements for the odd vs even hosts.
Electron Absorption Spectroscopy. An important method to
characterize the formation of the PDA backbone is UV/vis
absorption spectroscopy. The layered morphology of PIDA
cocrystal 6·2 (Figure 2C) provides a very good substrate to
measure the electronic absorption spectrum of the polymer
where the individual PIDA strands are completed planar and
isolated from one another. Figure 7 shows the absorption
spectrum of gold 6·2 crystals plotted as a function of wave-
number (cm-1). On this frequency scale, the major absorption
peak at 684 nm corresponds to an energy of 14 617 cm-1 (1.8
eV), consistent with the π-π* transition of the polymer
backbone. This transition is followed by a detailed vibrational
progression, resulting from exciton coupling to the double-bond
(D ≈ 1400 cm-1, as determined by Raman) and triple-bond (T
≈ 2100 cm-1) stretching modes, and an extended high-energy
tail, attributed to free motion of excitons along the chain.25
Similar vibronic progressions have been observed in the
reflectance spectra of substituted PDAs and the derived absorp-
tion spectra from these measurements.29 Weiser, Schott, and
co-workers have also observed some vibrational overtones in
the absorption spectra of PDA chains diluted in a monomer
matrix.30 However, to our knowledge, this is the first directly
acquired absorption spectrum of a fully polymerized PDA to
show this detailed vibronic resolution. The low energy of the
Figure 8. Absorption spectroscopic change of a water suspension of
cocrystals made with host 6. (A) Blue 6·1 cocrystals dispersed in water.
(B) Water suspension from panel A turns dark blue after 3 days at room
temperature. (C) Suspension from panel A appears dark gray after 6 days
at room temperature.
major absorption in the UV/vis spectrum of the cocrystals and
the highly resolved vibronic satellites are consistent with PIDA’s
highly planar structure. The clean, detailed spectrum indicates
that each of the polymer chains has the same geometry and
environment, without significant interchain interactions. Previ-
ously, such a well-resolved PDA spectrum could only be
approached by theoretical models.31 According to theoretical
studies by Weiser and co-workers,25,29,32 vibronic progressions
in optical spectra of PDA crystals derive from the high
polarizability of the singlet excitons, which are also responsible
for the large nonlinear susceptibility ꢁ(3) observed in PDAs.
Topochemical Polymerization in Microcrystal Suspensions.
To investigate the polymerization process, we examined the
spectroscopic changes in water suspensions of the 6·1 cocrystals
(Figure 8). Cocrystals of monomer 1 and host 6 can be dispersed
into water to form quasi-homogeneous microcrystalline suspen-
sions. The blue curve in Figure 8 shows the absorption spectrum
of the suspension formed when blue crystals made with host 6,
determined by Raman spectroscopy to contain primarily mono-
mer (Figure S7, Supporting Information), were dispersed in
water (Figure 8A). The major absorption peak at 650 nm and
the vibrational satellite at 594 nm suggest the presence of
oligomers or polymer chains, dilute in a monomer crystal matrix
at a mole fraction that cannot be detected by X-ray diffraction
or NMR. The blue suspension turns darker in color after 3 days
(28) Odd-even effects have also been observed by F. W. Fowler, J. W.
Lauher, and co-workers in cocrystal systems for the polymerization
of triynes. See: Liu, Y. The supramolecular synthesis of polydiacety-
lene and polytriacetylene. Master’s Thesis, The State University of
New York: Stony Brook, NY, 2001.
(31) Spagnoli, S.; Berrehar, J.; Lapersonne-Meyer, C.; Schott, M. J. Chem.
Phys. 1994, 100, 6195–6202.
(29) Weiser, G. Phys. ReV. B 1992, 45, 14076–14085.
(32) Barisien, T.; Legrand, L.; Weiser, G.; Deschamps, J.; Balog, M.; Boury,
B.; Dutremez, S. G.; Schott, M. Chem. Phys. Lett. 2007, 444, 309–
313.
(30) Horvath, A.; Weiser, G.; Lapersonne-Meyer, C.; Schott, M.; Spagnoli,
S. Phys. ReV. B 1996, 53, 13507–13514.
9
J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008 7707

A R T I C L E S
Luo et al.
Figure 10. Suspensions and aggregates of PIDA. (A) Gold 6·2 cocrystals
dispersed in H2O. (B) Suspension of isolated PIDA (2) in THF. (C)
Suspension from panel B after 2 h at room temperature, with visible
aggregation. (D) SEM image of PIDA aggregates. (E) TEM image of a
PIDA fiber with a diameter of about 20 nm.
Figure 9. Size distributions of microcrystalline particles obtained by
dispersing 6·1 cocrystals in water via sonication over varying times: red,
30 min; green, 60 min; blue, 120 min. Insets are typical SEM images of
the particles after the corresponding treatment (scale bars are 2 µm).
microcrystal.36,37 Because PIDA is formed here in a single-
crystal-to-single-crystal transformation, the particle size should
be directly related to the resulting polymer length, providing a
route to control the molecular weight distribution of the polymer.
at room temperature (Figure 8B), and the two major absorption
peaks shift to 679 and 621 nm. These changes indicate an
increase in the effective conjugation length of the polymer as
the reaction progresses within the microcrystalline particles.
After 3 more days, the absorption spectrum of the suspension
shows a maximum absorption peak at 689 nm and a clearly
resolved vibronic progression (Figure 8C). The excellent match
between the absorption spectra of the water suspension and solid
crystals indicates complete topochemical polymerization within
the suspended particles.
Formation of these water suspensions, before polymerization,
can also be used to control the size distribution of the resulting
polymer cocrystals.33–35 Transmission electron microscopy
(TEM) images of the micrometer-sized monomer cocrystals
demonstrate that the particle size decreases as the sonication
time increases. Figure 9 indicates the size distributions of the
particles, based on a statistical comparison of over 100 particles
for each. When blue 6·1 cocrystals are initially dispersed in
water by sonication, individual particles exceed the TEM
dimensions. After 30 min of sonication, the cocrystals are broken
into particles with average diameters of 5 µm. Further sonication
results in smaller size and a finer dispersion of the particles.
For example, after 120 min of sonication, most particles have
diameters of approximately 2 µm, and they are dispersed more
evenly throughout the suspension. Topochemical polymeriza-
tions of PDAs in optimal crystalline lattices have been shown
to proceed quantitatively, spanning the length of each
Isolation and Aggregation of PIDA. Although an oxalamide
host is necessary to form PIDA, the polymer can be separated
from the host after the polymerization. The gold 6·2 cocrystals
can be dispersed in water (Figure 10A) or organic solvents such
as methanol, THF, or chloroform. The host oxalamide 6 is
insoluble in water but much more soluble in organic solvents.
After extensive sonication and rinsing, these solvents give a
clear blue suspension, which is fully depleted of host 6, as
judged by FTIR spectroscopy (Figure 10B; see Figure S9,
Supporting Information, for details). The blue suspensions start
out transparent but begin to show visible aggregates after
minutes or hours without sonication (Figure 10C). SEM and
TEM images (Figure 10D,E) of drop-cast fibers from a THF
suspension demonstrate that these fibers have diameters of
approximately 10-50 nm, suggesting aggregates of hundreds
of individual polymer strands. The aggregation of PIDA is not
surprising, given expected favorable interactions between the
electron-rich π backbone of the polymer and the Lewis-acidic
iodine atom side groups.
The removal of host, and subsequent aggregation of polymer,
can be followed by UV/vis absorption spectroscopy. Figure 11A
shows the absorption spectra of several different suspensions,
each prepared using fully polymerized gold cocrystals of PIDA
and host 6. The red spectrum was measured using an aqueous
suspension, such as that pictured in Figure 10A. This spectrum
is indistinguishable from the previously measured spectrum of
gold 6·2 cocrystals in the solid state (Figure 7). If the gold 6·2
(33) Katagi, H.; Kasai, H.; Okada, S.; Oikawa, H.; Komatsu, K.; Matsuda,
H.; Liu, Z. F.; Nakanishi, H. Jpn. J. Appl. Phys. Part 2, Lett. 1996,
35, L1364–L1366.
(34) Wang, X.; Yang, K.; Ye, H.; Wang, Y. P.; Lee, J. S.; Sandman, D. J.
J. Macromol. Sci., Pure Appl. Chem. 2006, 43, 1937–1943.
(35) For a recent example of sonication as a method to control particle
size in a cocrystal system, see: Bucar, D. K.; MacGillivray, L. R. J. Am.
Chem. Soc. 2007, 129, 32–33.
(36) Iida, R.; Kamatani, H.; Kasai, H.; Okada, S.; Oikawa, H.; Matsuda,
H.; Kakuta, A.; Nakanishi, H. Mol. Cryst. Liq. Cryst. 1995, 267, 95–
100.
(37) Takahashi, S.; Miura, H.; Kasai, H.; Okada, S.; Oikawa, H.; Nakanishi,
H. J. Am. Chem. Soc. 2002, 124, 10944–10945.
9
7708 J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008

Preparation and Characterization of Poly(diiododiacetylene)
A R T I C L E S
increase of conjugation length or interchain aggregation. We
do not have complete X-ray diffraction data on PIDA in the
6·2 cocrystals to determine unambiguously that the polymer
backbone is fully planar, but our other spectroscopic studies
(Raman and MAS NMR) indicate that the polymer in these
crystals is indistinguishable from the material in the 4·2
cocrystals. Thus, it seems most likely that the UV/vis spectrum
of the gold cocrystals, with its λmax at approximately 690 nm,
represents the polymer in a completely planar conformation.
We believe interchain interactions, demonstrated in the SEM
and TEM images, are the most reasonable explanation of the
observed low-energy absorption.
To test the stability of these aggregates, we examined the
UV/vis absorption behavior upon dilution of the suspensions.
Figure 11B shows the UV/vis spectrum of the initial host-
depleted suspension, as well as the spectra after successive
dilution. Plots of the absorption at multiple wavelengths as a
function of relative concentration (inset, Figure 11B) obey
Beer’s law, indicating that the aggregates are not in dynamic
equilibrium under these conditions. However, high-speed cen-
trifugation (10 000 rpm, 1 h) leads to removal of the absorbing
particles, confirming that the blue sample is a suspension, rather
than a true solution.
Conclusion
The isolation of PIDA from the host molecules allows us to
explore further applications and properties of this minimally
substituted poly(diacetylene). For example, we will explore the
transformation of PIDA into other functionalized PDAs via
transition-metal-catalyzed couplings. We will also examine the
possibility of reductive dehalogenation of PIDA to give linear
carbon structures, i.e., the elusive carbyne. In addition, the ability
to prepare PIDA in aqueous suspensions gives us an effective
method to control the molecular weight distribution of the
polymer. PIDA has potential as a useful material in its own
right, and we will continue to study its physical, chemical,
optical, and electronic properties.
Figure 11. (A) UV/vis absorption spectra of suspensions of gold-colored
6·2 cocrystals: red, aqueous suspension of 6·2 cocrystals, as pictured in
Figure 10A; brown, 6·2 cocrystals immediately after dispersing in THF;
blue, host-depleted PIDA suspension in THF, as pictured in Figure 10B.
(B) Absorption spectra of the host-depleted PIDA in THF, measured at a
series of relative concentrations (c, 0.67c, 0.50c, 0.33c) prepared by
successive dilution. Inset: Plot of absorbance at several different wavelengths
vs relative concentration.
Acknowledgment. We thank the NSF (CHE-9984937, CHE-
0446749, and CHE-0453334) for support of this research. We thank
the research group of G. Halada for help in measuring the Raman
spectra of the materials described here. We also thank the research
groups of N. S. Sampson and D. P. Raleigh for help in measuring
the electronic absorption spectra of PIDA.
cocrystals are dispersed in an organic solvent such as methanol,
THF, or chloroform, the spectrum quickly takes on a different
appearance. As shown in the brown curve in Figure 11A, the
initial spectrum upon forming the suspension in THF (or the
other organic solvents) includes a new broad shoulder at about
750 nm and a simultaneous broadening of the shorter-
wavelength vibronic satellites. After removal of the host by
sonication and rinsing, as described above, the UV/vis spectrum
loses all vibrational definition, and the long-wavelength absorp-
tion becomes more prominent.
Supporting Information Available: Complete experimental
and characterization information for hosts 4-6 and the resulting
cocrystals, including crystallographic data in CIF format for
cocrystals 4·1 and 4·2. This material is available free of charge
via the Internet at http://pubs.acs.org.
The appearance of the absorbance at longer wavelength can
be attributed to a new lower-energy exciton generated by either
JA8011403
9
J. AM. CHEM. SOC. VOL. 130, NO. 24, 2008 7709