Synthesis and evaluation of fully (5-amidoisophthalic acid)-functionalised polyacrylamides as selective inhibitors of the beta crystal polymorph of L-glutamic acid

Article abstract of DOI:10.1016/j.tet.2010.05.023

Poly-N-5-acrylamidoisophthalic acid (4), poly-N-(5-(N-(3,5-dicarboxyphenyl) carbamoyl)pentyl)acryl-amide (10a) and poly-N-(11-(N-(3,5-dicarboxyphenyl) carbamoyl)undecyl)acrylamide (10b) were prepared and assessed as polymorph-selective crystallization inhibitors of the stable b form of L-glutamic acid.Polymerization was carried out as the final step in the preparation of 10a and 10b to ensure the preparation of fully functionalized polymers.Polymers 4, 10a and 10b were effective as complete inhibitors of the stable b form of L-glutamic acid in quantities of 0.5% w/w or greater, whereas the corresponding 'monomeric' additives 2 and 11 required quantities of 3% or greater to completely inhibit the b form, demonstrating a cooperative binding effect by the polymeric additives.Within the series of polymers 4, 10a and 10b, polymer 10a, which features a short tethering chain, was the most effective.

Full text of DOI:10.1016/j.tet.2010.05.023

Tetrahedron 66 (2010) 5459e5466
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis and evaluation of fully (5-amidoisophthalic acid)-functionalised
polyacrylamides as selective inhibitors of the beta crystal polymorph
of -glutamic acid
L
*
Dawn M. Kelly, Humphrey A. Moynihan
Department of Chemistry/Analytical and Biological Chemistry Research Facility, University College Cork, Cork, Ireland
a r t i c l e i n f o
a b s t r a c t
Poly-N-5-acrylamidoisophthalic acid (4), poly-N-(5-(N0-(3,5-dicarboxyphenyl)carbamoyl)pentyl)acryl-
amide (10a) and poly-N-(11-(N0-(3,5-dicarboxyphenyl)carbamoyl)undecyl)acrylamide (10b) were pre-
Article history:
Received 1 December 2009
Received in revised form 19 April 2010
Accepted 10 May 2010
pared and assessed as polymorph-selective crystallization inhibitors of the stable
b form of L-glutamic
acid. Polymerization was carried out as the final step in the preparation of 10a and 10b to ensure the
preparation of fully functionalized polymers. Polymers 4, 10a and 10b were effective as complete in-
Available online 13 May 2010
hibitors of the stable
responding ‘monomeric’ additives 2 and 11 required quantities of 3% or greater to completely inhibit the
form, demonstrating a cooperative binding effect by the polymeric additives. Within the series of
polymers 4, 10a and 10b, polymer 10a, which features a short tethering chain, was the most effective.
Ó 2010 Elsevier Ltd. All rights reserved.
b form of L-glutamic acid in quantities of 0.5% w/w or greater, whereas the cor-
Keywords:
L-Glutamic acid
b
Crystal polymorphism
‘Tailor-made’ Additives
1. Introduction
crystallization additives.15 Rationally designed additives, known as
‘tailor-made’ additives, have been used to control crystal mor-
Crystal polymorphism of organic compounds has become an
area of considerable current interest,1 in particular because of the
significance of polymorphism for pharmaceuticals.2 This has driven
much new research in crystallisation methods and polymorph
discovery. Examples of new crystallisation technologies include co-
crystallization,3 supercritical fluid crystallization,4 sonocrystalliza-
tion,5 laser-induced crystal nucleation6 and capillary-space
crystallization.7 New approaches to polymorph discovery have
been developed utilising high-throughput screening technologies.8
Recent research has also attempted to examine the molecular level
events occurring prior to crystal nucleation, using techniques such
as small-angle neutron scattering9 and NMR spectroscopy.10
It is widely appreciated that molecular species other than the
crystallizing solute or the crystallization solvent can have a signifi-
cant impact on the outcome of crystallizations. Process impurities
can affect crystal morphology11 and form.12 Impurities can be
deliberately added to influence process outcomes. Polymeric ad-
ditives have been used as crystal heteronuclei in polymorph
screening.13 Structurally similar materials have been added as
pseudoseeds.14 Some recent research on co-crystallization has
resulted in novel or unexpected crystal forms being obtained as
a consequence of putative co-crystallizing compounds acting as
phology and polymorphism.16 Usually, such additives will be
structurally analogous to molecules of the crystallizing substance,
as analogues are likely to have sufficient structural similarity to the
crystallizing compound to allow incorporation into pre-critical
crystal nuclei, after which any structural dissimilarity can impede
addition of further molecules of crystallising substance.17 That
portion of the analogue structure, which permits selective in-
teraction with pre-critical nuclei has been referred to as the
‘binder’, while that portion which impedes further addition has
been referred to as the ‘perturber’.18
Davey et al. have shown that certain conformationally restricted
diacids can act as selective binder groups for the stable
polymorph of -glutamic acid, inhibiting crystallisation of that
form.19 This effect is based upon mimicry of the
-glutamic acid
conformation found in the form. The conformations of -glutamic
form and the metastable form are il-
form can be seen to have a more open or
b crystal
L
L
b
L
acid found in the stable
lustrated in Figure 1. The
b
b
a
extended conformation, with the carboxyl groups further apart;
while the form possesses a more closed or folded conformation
with the carboxyl groups closer together. Addition of conforma-
tionally restricted mimics of the form conformation, such as tri-
mesic acid, to -glutamic acid crystallisations was found to result in
crystallisation of the metastable form under conditions, which
would normally produce the more stable
form.19 This effect is
attributed to selective addition of molecules of trimesic acid to
crystal nuclei of the form and inhibition of their growth into
a
b
L
a
b
* Corresponding author. E-mail address: h.moynihan@ucc.ie (H.A. Moynihan).
b
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.05.023

5460
D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
nuclei of this form. This is illustrated in Figure 2. The amino group
of 5-aminoisophthalic acid is hence free to be used as a point of
attachment by an amide formation.
Figure 1. Conformations of L-glutamic acid found in the
a and b crystal polymorphs.
mature crystals, whereas crystal nuclei of the metastable
not similarly affected.
a form are
We have previously shown that the efficiency of these inhibitors
can be improved through the use of a cooperative binding effect.20
This effect involves the use of polymeric additives, rather than their
monomeric counterparts. Whereas each molecule of a monomer-
like additive, such as trimesic acid, binds independently to crystal
nuclei, once one residue of a polymeric additive binds to a crystal
nucleus, binding of further residues is assisted. Hence, while 10% w/
w of trimesic acid is required to fully inhibit crystallization of the
Figure 2. Mimicry of the conformation found in the
acid by benzene-1,3-dicarboxylic acid units.
b crystal polymorph of L-glutamic
The simplest completely functionalised polymer, which could be
prepared would hence be just the polyacrylamide formed from 5-
acrylamidoisophthalic acid. This was prepared from 5-amino-
isophthalic acid (2) by a reaction with acryloyl chloride, followed by
free radical polymerisation to give compound 4 (Scheme 1).
b
polymorph of L-glutamic acid, a corresponding polyacrylamide-
Scheme 1. Preparation of compound 4.
borne analogue was found to fully inhibit
b
form crystallization at
The details of the molecular level events, which precede the
appearance of crystal nuclei are still very uncertain and are a sub-
ject of current interest.9 For this reason, it is often beneficial to
allow compounds intended to interact with crystal nuclei to have
a degree of conformational flexibility, so as to allow the binder
groups to achieve a ‘best fit’ with the crystal nucleus. The 12-carbon
chain in the polyacrylamide precursor 1 is intended to provide such
flexibility. While the polyacrylamide 4 is fully functionalized with
5-amidoisophthalic acid groups, it lacks the conformational
flexibility provided by such a tethering carbon chain. Hence, the
preparation of analogues of polyacrylamide 4 with inserted six and
12-carbon chains (compounds 10a and 10b) was also carried out.
Fully functionalised tethered polyacrylamides 10a and 10b were
prepared as outlined in Scheme 2. The corresponding six and 12-
carbon terminal amino acids (5a and 5b) were first N-protected
with Boc groups, after which the 4-amidoisophathalic acid group
was introduced by DCC/HOBt coupling to give compounds 7a and
7b. The Boc N-protection was removed by treatment with tri-
fluoroacetic acid, after which it was found to be preferable to isolate
the deprotected amines as the hydrochloride salts (8a and 8b).
These were converted to the corresponding acrylamides (9a and
9b), which were polymerised in the final step to give the required
polyacrylamides.
1% w/w quantities.21 The preparation of this polymeric analogue
involved treatment of an acid chloride functionalized poly-
acrylamide (compound 120) with 5-aminoisophthalic acid. This
direct polymer functionalization process resulted in a material in
which only approximately 50% (as determined by 1H NMR) of the
polymer residues were functionalized with 5-aminoisophthalic
acid residues. Any attempts to improve on the efficiency of these
tailor-made crystallization additives would need to address the
additive preparation so as to, preferably, provide polymeric mate-
rial, which is completely functionalized. In this paper we described
the preparation of fully functionalized polymeric additives and
their efficacy as inhibitors of the
b crystal form of L-glutamic acid.
2. Results and discussion
Our previously reported additive was based on the use of 5-
aminoisophthalic acid (2), rather than trimesic acid, as the ‘binder’
group.21 5-Aminoisophthalic acid and its derivatives are usable in
this way as only two of the three carboxylic acid groups of trimesic
acid are necessary for its mimicry of the conformation found in the
Compounds 4, 10a and 10b were assessed for their effects on the
crystallisation of
samples of -glutamic acid can be initially assigned on the basis of
crystal morphology. Typically, crystals of the form of -glutamic
acid have the prismatic morphology shown in Figure 3(left),
whereas those of the form have the needle-like morphology
L-glutamic acid from water. The crystal form of
L
a
L
b
crystal form of
L-glutamic acid, and its selective binding to crystal
b

D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
5461
Scheme 2. Reagents and conditions: (a) Boc2O, dioxane, aq NaHCO3; (b) 5-aminoisophthalic acid, DCC, HOBt, DIPEA, DMF; (c) TFA, DCM; (d) HCl, EtOH; (e) acryloyl chloride, aq
NaOH; (f) AIBN, DMF, 80 ꢀC.
shown in Figure 3(right). However, more rigorous assignment can
be achieved using powder X-ray diffraction (PXRD). Theoretical
PXRD patterns, shown in Figure 4, were generated from crystal
structural data obtained from the Cambridge Structural Database.
The PXRD patterns of the two forms are clearly different, allowing
unambiguous assignment of crystal form. For example, Figure 5
Addition of sufficient quantities of 5-aminoisophthalic acid de-
rivatives generally inhibits formation of the more stable form, and
promotes crystallization of the metastable form. However, the
quantity of additive necessary to completely suppress form crys-
tallization varies considerably with structure. Addition of 10% w/w
of 5-aminoisophthalic acid (2) is required to fully inhibit formation
b
a
b
shows typical PXRD patterns of actual samples of the
crystal forms obtained in this study. It can be seen that these
corresponded very well with their theoretical counterparts.
a
and
of the b form. Addition of 5% w/w of 2 results in crystallization of
b
a mixture of polymorphs, while compound 2 has no inhibitory effect
when added in 2 or 1% w/w quantity (Table 1). The acetamido de-
rivative 11 shows somewhat greater inhibitory activity, with 3% w/w
completely suppressing the appearance of the b form. Addition of 1%
w/w of 11 has no effect on the polymorphic outcome.
Addition of poly(5-acylamidoisophthalic acid) 4 in quantities of
0.5% w/w or greater resulted in a complete inhibition of the
b form.
This effect was lost upon reduction of the added quantity to 0.1% w/
w (Table 1). The fact that polymer 4 is an order of magnitude more
active than, for example, 5-aminoisophthalic acid 2, is a demon-
stration of the cooperative binding effect mentioned above.
Polymers 10a and 10b additionally have tethering chains, con-
sisting of five or eleven carbons plus an amide group, inserted
between the 5-aminoisophthalic acid groups and the poly-
acrylamide backbone found in polymer 4. These tethering chains
should provide the isophthalic acid groups with the flexibility to
achieve a ‘best fit’ when binding to crystal nuclei. In fact, polymer
10a, which contains the shorter tethering chain, gives complete
inhibition of the
a cooperative binding effect, which is superior to that shown by the
simple polyacrylamide 4. However, the form reappears upon re-
duction of the amount of 10a added to 0.1% w/w. Polymer 10b,
which contains the longer tethering group, is only equally active to
b form at 0.2% w/w or greater. This clearly shows
b
Figure 3. Typical prismatic morphology of the
a crystal form of L-glutamic acid (left);
typical needle-like morphology of the crystal form of L-glutamic acid (right).
b
the simple polyacrylamide 4, with complete inhibition of the
observed at 0.5% w/w quantities, but not at lesser quantities.
In general, the polymeric additives 4, 10a and 10b were found to
fully inhibit the form in quantities of approximately an order of
b form
Table 1 summarises the outcomes of crystallizations of L-gluta-
mic acid is the absence and presence of quantities of compounds 4,
10a and 10b. For comparison, data on the use of the ‘monomeric’
analogues 5-aminoisophthalic acid 2 and its acetamido derivative
11 are also presented.
b
magnitude less than those required by the ‘monomeric’ additives 2
and 11, demonstrating the effectiveness of cooperative binding.
While inclusion of the tethering group in polymer 10a gives an
improvement over the simple polyacrylamide 4, possibility through
increased ability to achieve a ‘best fit’ with crystal nuclei, that im-
provement was effectively lost on moving to the longer tethering
group in polymer 10b. As these crystallizations are from water, the
hydrophility/hydrophobicity of the polymeric additive becomes
a possible issue. The polyacrylamide backbone is itself reasonably
hydrophilic, allowing polymers 4 and 10a to act effectively in water.
However, the 11-carbon tethering group in polymer 10b may be too
hydrophobic for efficient solvation in water, reducing the efficacy of
polymer 10b compared to that of polymers 4 and 10a.
Crystallization of
L-glutamic acid from water reliably gives the
b
form when carried out at 38 ꢀC using 35 g of solute per litre of
solvent, and gives the a
form when carried out at 18 ꢀC using 20 g of
solute per litre of solvent, as described in the literature24; likewise,
crystallization from a solution containing 45 g per litre of solvent at
45 ꢀC gives a mixture of
a and b forms (Table 1).

5462
D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
STOE Powder Diffraction System
19-Dec-2008
4000
3500
3000
2500
2000
1500
1000
500
0
10.0
20.0
30.0
40.0
50.0
2Theta
19-Dec-2008
STOE Powder Diffraction System
4000
3500
3000
2500
2000
1500
1000
500
0
10.0
20.0
30.0
40.0
50.0
2Theta
Figure 4. Theoretical PXRD pattern of
a b
form of L-glutamic acid (CSD Ref. LCLUAC0222) (top); theoretical PXRD pattern of -form of L-glutamic acid (CSD Ref. LGLUAC23) (bottom).
Our previously reported additives were prepared by the direct
functionalization of polymer 1.21 These suffered from the drawback
that the extent of functionalization was variable, typically around
50%. This presented an uncertainty when comparing the perfor-
mance of these polymers with their ‘monomeric’ counterparts.
However, the polymeric additives described in this paper do not
suffer from this drawback, as polymerization is kept until the final
step, hence ensuring 100% functionalization. The direct comparison
of these polymers with their ‘monomeric’ analogues as crystalli-
zation additives is therefore placed on a sounder footing.
3. Conclusions
The 5-amidoisophthalic acid functionalized polymers 4, 10a and
10b demonstrated a cooperative binding effect compared to their
‘monomeric’ analogues 2 and 11, in that approximately one order of

D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
5463
STOE Powder Diffraction System
14-Mar-2008
3600
3200
2800
2400
2000
1600
1200
800
400
0
10.0
20.0
30.0
40.0
50.0
2Theta
18-Mar-2008
STOE Powder Diffraction System
4000
3500
3000
2500
2000
1500
1000
500
10.0
20.0
30.0
40.0
50.0
2Theta
Figure 5. PXRD pattern of a sample of the
a form of L-glutamic acid (top); PXRD pattern of a sample of the b-form of L-glutamic acid (bottom).
magnitude less than the polymeric compounds was sufficient to
fully inhibit the appearance of the more stable crystal form of L-
effectiveness of the polymers lies in a balance between the efficient
solvation, and the ability to achieve a ‘best fit’ with pre-critical
crystal nuclei.
Crystallization additives, which reliably control polymeric out-
come in quantities in the order of 0.1% w/w or less are approaching
a level of efficacy, which may be acceptable for many industrial
applications, for example, in the manufacturing of metastable
b
glutamic acid. Polymerization was kept until the final step in the
preparations of 4,10a and 10b, ensuring that the resulting polymers
were 100% functionalized with 5-aminoisophthalic acid group,
hence removing any doubt that incomplete functionalization might
be a factor affecting efficacy. Our findings suggest that the

5464
D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
Table 1
a range from aprroximately 160 to 6,000,000 Da. Data is presented
as weight average molecular weight (Mw) and polydispersity index
(Mw/Mn) where Mn is the number average molecular weight.
Outcome of crystallisations of
tives 2, 4, 10a, 10b and 11
L-glutamic acid from water in the presence of addi-
Entry no.
Concna (g Lꢁ1
)
Additive no.
% w/wb
Form
4.1.1. 5-Acryloylaminoisophthalic acid (3)25. 5-aminoisophthalic
acid 2 (1 g, 5.52 mmol) and sodium hydroxide (0.66 g, 16.56 mmol)
wereadded to water (11 ml) and the mixturewas cooled to 0 ꢀC in an
ice bath. The excess sodium hydroxide served to maintain the pH
above 12. Acryloyl chloride (0.62 ml, 6.10 mmol) was added in 0.1 ml
portions every 2 min over a 12 min period to the rapidly stirred
aqueous suspension. After an additional 2 min the reaction mixture
was acidified to pH 2.6, using a pH metre, by the drop-wise addition
of concd HCl at 0 ꢀC with stirring. The precipitate was collected by
filtration, washed with a little ice cold waterand dried in a desiccator
to afford 3 (0.76 g, 60%) as a light pink solid: mp 267e272 ꢀC; IR (KBr)
1
2
3
4
5
6
7
8
35
20
45
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
None
None
None
2
2
2
d
d
d
10
5
2
1
3
1
b
ac
a
a
a
b
b
a
b
a
a
a
a
b
a
a
a
a
b
and bd
and b
2
11
11
4
4
4
9
10
11
12
13
14
15
16
17
18
19
2.2
1
0.8
0.5
0.1
0.3
0.2
0.1
0.5
0.4
4
4
n
3345, 2965, 1716, 1605, 1559, 1201; 1H NMR (DMSO-d6)
d 5.81 (1H,
10a
10a
10a
10b
10b
dd J¼1.5, 10.5 Hz), 6.30 (1H, dd J¼1.5, 16.5 Hz), 6.49 (1H, dd J¼10.5,
16.5 Hz), 8.17 (1H, s), 8.54 (2H, s), 10.69 (1H, s). MS m/z (%) 236 (Mþ,
14), 146 (15), 116 (4), 115 (56), 105 (100), 74 (38), 64 (16). Calcd for
C11H10NO5: M, 236.0559. Found: m/z 236.0548.
and
b
a
Crystallizations were carried out at 35 g Lꢁ1 at 38 ꢀC as described in Ref. 24.
% w/w of additive.
b
c
4.1.2. Poly-N-5-acrylamidoisophthalic
acid
(4)20. 5-Acryl-
Crystallizations of entry no. 2 were carried out at 20 g Lꢁ1 at 18 ꢀC as described
oylaminoisophthalic acid 3 (0.53 g, 2.72 mmol) was refluxed in
5 ml THF at 70 ꢀC for 72 h in the presence of AIBN (30 mg,
0.18 mmol). The solvent was removed using a rotary evaporator to
afford 3 (0.53 g, 100% recovery) as a yellow solid, degree of grafting
(determined from the relative integration of the arylH and
eCH2CH2(CO)e resonances) 100%: mp 250e251 ꢀC (decomp.); IR
by Ref. 24.
d
Crystallizations of entry no. 3 were carried out at 45 g Lꢁ1 at 45 ꢀC as described
by Ref. 24.
crystal forms of high-value fine chemicals such as pigments, high-
energy compounds, organic magnets or organic conducting
compounds. Further improvement in the design of the additives
described here, which may make them effective in quantities of
0.1% w/w or less includes improved solvation of the tethering
sidechains, or alternative structured scaffolds such as dendrimers.
The additives also demonstratethat relativelyminorquantities of
impurities can have a significant impact of the outcome of crystal-
lization. This is an especially relevant issue for the manufacture of
highly regulated compounds such as pharmaceuticals.
(KBr)
1.20e1.80 (br m), 8.17 (1H, s), 8.54 (2H, s), 10.71 (1H, s); Mw
n
3000e3500, 1711, 1603, 1559, 1203; 1H NMR (DMSO-d6)
d
21,600 Da, Mw/Mn 9.0.
4.2. General procedure for the N-protection of amino acids26
A suspension of amino acid (20 mmol) and di-tert-butyldicar-
bonate (33.6 mmol) in dioxane (30 ml) and saturated aq sodium
bicarbonate solution (40 ml, freshly prepared) was stirred at room
temperature for 5 days. The bulk of the dioxane was removed under
reduced pressure. Ethyl acetate (30 ml) was added and the mixture
was acidified to pH 3e4 by addition of concd HCl. The mixture was
extracted with ethyl acetate (3ꢂ20 ml). The organic layers were
combined, washed with water (2ꢂ25 ml), dried over MgSO4 and
the solvent evaporated to afford a white waxy solid.
4. Experimental
4.1. General
Materials used were purchased from SigmaeAldrich. IR spectra
were recorded on a PerkineElmer 1000 spectrometer in the range
of 4000e500 cmꢁ1. All melting points were recorded on a Vickers
Microscope model M14/2 and are uncorrected. 1H and 13C NMR
spectra were recorded on a Bruker Avance 300 spectrometer at 300
and 75 MHz, respectively. The NMR spectra were recorded in either
deuteriochloroform (CDCl3) or deuterated dimethylsulphoxide
(DMSO-d6) and tetramethylsilane (TMS) was used as an internal
4.2.1. N-tert-Butoxycarbonyl-6-aminohexanoic acid (6a)27. Obtained
by the N-protection of 6-aminohexanoic acid to afford 4.54 g (98%):
mp 34e36 ꢀC (lit.27 38e39 ꢀC); IR (thin film) 3336, 2977, 1710, 1169;
1H NMR (CDCl3)
d 1.26e1.45 (2H, m), 1.44 (9H, s), 1.62e1.69 (4H, m),
2.35 (2H, t, J¼7.5 Hz), 3.10e3.14 (2H, q, J¼9 Hz), 4.60 (1H, br s), 5.79
standard. Chemical shifts (
d
) are reported in parts per million (ppm)
(1H, br s).
downfield of TMS. Coupling constants (J) are quoted in hertz (Hz)
and splitting patterns are indicated as s (singlet), br s (broad sin-
glet), d (doublet), t (triplet), q (quartet), dd (doublet of doublets)
and m (multiplet). For 13C NMR spectra, assignments are made from
13C DEPT spectra run in DEPT-90 and DEPT-135 modes. Liquid
chromatography mass spectra (MS) were obtained using an LCT
Premier Time of Flight instrument and a Quattro Micro triple
quadrupole instrument. Elemental analysis for carbon, hydrogen
and nitrogen was recorded on a CE440 elemental analyser. Thermal
analysis was recorded on a DSC Q1000 instrument. Polymer mo-
lecular weights were determined by Intertek MSG, Redcar, UK, by
Gel Permeation Chromatography using a Viscotek GPC Max TDA302
instrument and a 2ꢂ30 cm Plgel Omnisec GPC column with
dimethylacetamide as eluent, flow rate of 0.8 ml minꢁ1, tempera-
ture of 50 ꢀC and detection by ELSD@95C. The system was cali-
brated against a series of linear polystyrene standards covering
4.2.2. N-tert-Butoxycarbonyl-12-aminododecanoic acid (6b)27. Ob-
tained by the N-protection of 12-aminododecanoic acid to afford
7.65 g (96%): mp 83e85 ꢀC (lit.27 83.5e84.5 ꢀC); IR (thin film)
n
3369, 2985, 1724, 1687, 1523, 1172; 1H NMR (CDCl3)
d 1.27 (14H, m),
1.53 (9H, s), 1.50e1.53 (2H, m) 1.58e1.65 (2H, m), 2.34 (2H, t,
J¼7.5 Hz), 3.07e3.13 (2H, q, J¼12 Hz), 4.51 (1H, br s), 5.62 (1H, br s).
4.3. General procedure for the DCC coupling of N-protected
amino acids 6a and 6b5
To a solution of N-protected amino acid (20 mmol) in dry DMF
(200 ml) was carefully added DCC (4.12 g, 22 mmol) at 0 ꢀC followed
by HOBt (2.97 g, 22 mmol) and DIPEA (10.45 ml, 60 mmol). After
continuous stirring for 10 min at 0 ꢀC, a solution of 5-amino-
isophthalic acid 2 (3.99 g, 22 mmol) in dry DMF (10 ml) was added

D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
5465
dropwise to the reaction mixture at 0 ꢀC. Stirring was continued for
4.5. General procedure for acrylating 8a and 8b25
1 h at 0 ꢀC and then overnight at room temperature. The reaction
mixture was then cooled to below 4 ꢀC for 30 min to promote the
precipitation of the white crystalline side product, DCU. This was
then filtered from the reaction mixture and the filtrate was collected
and evaporated using a rotary evaporator to remove the bulk of the
DMF. The remaining solution was poured onto ice with stirring and
left overnight to remove residual DMF and HOBt. The product was
collected by suction filtration to afford a white solid.
The functionalised hydrochloride salt (7.35 mmol) and sodium
hydroxide (1.03 g, 25.73 mol) were added to water (14.5 ml) and
the mixture was cooled to 0 ꢀC. The excess sodium hydroxide
served to keep the pH above 12. Acryloyl chloride (0.67 ml,
8.3 mmol) was added in 0.1 ml portions every 2 min over a 12 min
period to the rapidly stirred aqueous suspension. After an addi-
tional 2 min stirring, the reaction mixture was acidified to pH 2.6 by
the drop-wise addition of concd HCl while maintaining the tem-
perature at 0 ꢀC. The product precipitated out of solution and was
collected by filtration and washed with ice cold water to afford
a light pink solid.
4.3.1. 5-(6-(tert-Butoxycarbonylamino)hexanamido)isophthalic acid
(7a). Obtained using DCC coupling of N-protected amino acid 6a to
afford 7.19 g (91%): mp 179e182 ꢀC; IR (KBr)
n
3361, 3315, 1683,
1605, 1538, 1282, 1167; 1H NMR (DMSO-d6)
d 1.23e1.40 (2H, m),
1.35 (9H, s), 1.56e1.61 (4H, m), 2.32 (2H, t, J¼7.5 Hz), 2.90 (2H, q,
4.5.1. 5-(6-Acylamidohexanamido)isophthalic acid (9a). Obtained
by reaction of acryloyl chloride with 4c to afford 0.97 g (38%): mp
J¼9 Hz), 6.79 (1H, t, J¼6 Hz), 8.14 (1H, s), 8.43 (2H, s), 10.24 (1H, br
s), 12.72 (2H, br s); 13C NMR (DMSO-d6)
d
24.70, 25.89, 28.21, 29.27,
138e142 ꢀC; IR (KBr)
(DMSO-d6) d 1.09e1.32 (2H, m), 1.33e1.69 (4H, m), 2.27 (2H, t,
n
3372, 3215, 1702, 1656, 1564, 1260; 1H NMR
33.30, 36.33, 77.26, 123.35, 124.31, 131.60, 139.81, 155.55, 166.60,
171.68. Anal. Calcd for C19H26N2O7: C, 57.86; H, 6.64; N, 7.10. Found:
C, 57.86; H, 7.16; N, 7.62.
J¼7.5 Hz), 3.11e3.15 (2H, m), 5.48 (1H, dd, J¼3, 9 Hz), 5.98 (1H, dd,
J¼3, 18 Hz), 6.15 (1H, dd J¼10.5, 16.5 Hz), 7.80 (1H, s), 7.89 (2H, br s),
8.11 (1H, s), 8.45 (2H, s), 10.21 (1H, br s); 13C NMR (DMSO-d6)
d 25.4,
4.3.2. 5-(6-(tert-Butoxycarbonylamino)dodecanamido)isophthalic
26.1, 26.7, 28.8, 33.3, 40.3, 123.0, 124.6, 131.9, 133.3, 139.4, 164.4,
167.3, 171.6. MS m/z (%) 349 (Mþ, 30), 278 (16), 266 (25), 264 (18),
225 (60), 190 (10), 136 (10), 130 (19), 115 (100), 74 (60), 64 (18).
Calcd for C17H21N2O6: M, 349.1400. Found: m/z 349.1389.
acid (7b). Obtained using DCC coupling of N-protected amino acid
6b to afford 8.20 g (86%): mp 182e184 ꢀC; IR (KBr)
n
3352, 1687
(CO), 1607, 1541, 1280, 1169; 1H NMR (DMSO-d6)
d 1.27 (14H, br s),
1.44 (9H, s), 1.42e1.44 (2H, m) 1.61e1.65 (2H, m), 2.34 (2H, t,
J¼7.5 Hz), 2.91e2.96 (2H, q, J¼12 Hz), 6.71 (1H, br s), 8.14 (1H, s),
8.28 (2H, s), 10.21 (1H, br s), 12.72 (2H, br s); 13C NMR (DMSO-d6)
4.5.2. 5-(12-Acylamidododecanamido)isophthalic acid (9b). Obtained
by reaction of acryloyl chloride with 5c to afford 2.77 g (87%): mp
d
26.2, 28.2, 28.6, 28.8, 28.9, 28.9, 28.9, 29.4, 36.4, 40.3, 77.2, 123.3,
140e143 ꢀC; IR (KBr)
NMR (DMSO-d6) d 1.39 (14H, br s), 1.42e1.45 (2H, m), 1.47e1.50 (2H,
n
3283, 3115, 1702, 1657, 1622, 1552, 1229; 1H
124.3, 131.8, 139.8, 155.5, 166.5, 171.8. Anal. Calcd for C25H38N2O7: C,
62.74; H, 8.00; N, 5.85. Found: C, 63.22; H, 8.28; N, 6.40.
m), 2.24 (2H, t, J¼7.5 Hz), 2.99e3.06 (2H, q, J¼12 Hz), 5.48 (1H, dd,
J¼3, 9 Hz), 5.98 (1H, dd J¼3, 15 Hz), 6.14 (1H, dd J¼9.6, 17.2 Hz), 8.02
(1H, s), 8.30 (2H, s), 10.11 (1H, br s), 12.72 (2H, br s); 13C NMR (DMSO-
4.4. General procedure for the trifluoroacetic acid cleavage of
the functionalised tert-butyl cpds 7a and 7b and the
subsequent formation of their hydrochloride salts6
d6) d 25.8, 26.4, 27.1, 28.3, 28.4, 28.5, 28.7, 28.9, 36.4, 38.5, 122.9,
124.6, 131.9, 133.2, 139.4, 164.4, 167.1, 167.4, 171.6. MS m/z (%) 433
(Mþ, 50), 381 (2), 380 (18), 379 (100), 339 (6), 270 (12), 244 (6), 192
(4), 116 (4), 115 (74), 75 (5), 74 (78). Calcd for C23H33N2O6: M,
433.2339. Found: m/z 433.2331.
To a solution of the functionalised tert-butyl compound (2 mmol)
in DCM (10 ml) was added trifluoroacetic acid (10 ml) in one portion
at room temperature. The reaction mixture was stirred for 1 h and
then evaporated under vacuum to remove the solvent and pumped
dry using a high vacuum pump. This afforded the crude tri-
fluoroacetate salt as a light pink solid. This was then dissolved in
ethanol (2 ml) containing 1 ml of concd HCl. The mixture was evap-
oratedundervacuumtoaffordthehydrochloridesaltasayellowsolid.
4.6. General procedure for the preparation of the
functionalised polymers (10a) and (10b) with 100%
functionality3
The acryloyl compound (1.2 g, 3.45 mmol) was polymerized in
DMF (10 ml) at 80 ꢀC for 72 h with AIBN (30 mg, 0.18 mmol). The bulk
of the DMF was removed using rotary evaporation and the resulting
polymer was precipitated fromwaterand filtered to give ayellowsolid.
4.4.1. 5-(3,5-Dicarboxyphenylcarbamoyl)pentylammonium chloride
(8a). Obtained by trifluoroacetic acid cleavage of 7b to afford 0.64 g
(97%): mp 300e301 ꢀC (decomp.); IR (KBr)
n
2931, 1699, 1654, 1609,
1.20e1.41 (2H, m), 1.43e1.73 (4H,
m), 2.32 (2H, t, J¼7.5 Hz), 2.70e2.81 (2H, m), 7.89 (2H, br s), 8.11
(1H, s), 8.45 (2H, s), 10.41 (1H, br s); 13C NMR (DMSO-d6)
24.4,
4.6.1. Poly-N-(5-(N0-(3,5-dicarboxyphenyl)carbamoyl)pentyl)acryl-
1565, 1219; 1H NMR (DMSO-d6)
d
amide (10a). Compound 10a (1.05 g, 88% recovery): mp above
250 ꢀC (decomp.); IR (KBr)
n H
3338, 3115, 1708, 1651, 1556, 1219; 1
d
NMR (DMSO-d6) d 1.09e1.32 (2H, br m), 1.33e1.69 (4H, m),
25.4, 26.7, 33.3, 36.0, 123.4, 124.3, 131.6, 139.9, 166.5, 171.6. MS m/z
(%) 295 (Mþ, 100), 225 (2), 191 (6), 177 (10), 136 (16), 130 (42), 105
(4). Calcd for C14H19N2O5: M, 295.1294. Found: m/z 295.1284.
2.24e2.27 (2H, m), 3.10e3.15 (2H, m), 7.89 (2H, br s), 8.11 (1H, s), 8.45
(2H, s), 8.82 (1H, s), 10.21 (1H, br s); Mw 84,000 Da; Mw/Mn 17.5.
4.6.2. Poly-N-(11-(N0-(3, 5-dicarboxyphenyl)carbamoyl)undecyl)ac-
4.4.2. 11-(3,5-Dicarboxyphenylcarbamoyl)undecylammonium chlo-
rylamide (10b). Compound 10b (1.49 g, 100% recovery): mp above
ride (8b). Obtained by the trifluoroacetic acid cleavage of 7b to
250 ꢀC (decomp.); IR (KBr)
n
(KBr) 3337, 3114, 1705, 1652, 1555,
afford 0.78 g (97%): mp 300 ꢀC (decomp.); IR (KBr)
n
3490, 2922,
1218; 1H NMR (DMSO-d6)
d 01.01e1.42 (14H, br s), 1.42e1.45 (2H,
1710, 1606, 1562, 1204; 1H NMR (DMSO-d6)
d
1.19 (14H, m),
m), 1.43e1.56 (2H, m), 2.34 (2H, t, J¼7.5 Hz), 2.70e3.74 (2H, q,
J¼12 Hz), 8.10 (1H, s), 8.43 (2H, s), 10.23 (1H, br s), 12.72 (2H, br s);
Mw 68,000 Da, Mw/Mn 26.2.
1.38e1.42 (2H, m) 1.47e1.52 (2H, m), 2.26 (2H, t, J¼7.5 Hz),
2.66e2.82 (2H, q, J¼12 Hz), 7.67 (3H, br s), 8.07 (1H, s), 8.38 (2H, s),
10.25 (1H, br s), 12.72 (2H, br s); 13C NMR (DMSO-d6)
d 25.0, 25.8,
26.9, 28.5, 28.7, 28.8, 40.0, 123.4, 131.8, 139.8, 166.5, 171.8. MS m/z
(%) 379 (Mþ, 100), 285 (4), 244 (5), 216 (16), 130 (4), 115 (34), 74
(25). Calcd for C20H31N2O5: M, 379.2233. Found: m/z 379.2223.
4.6.3. 5-Acetamidoisophthalic acid (11)21. To a stirred solution of 5-
aminoisophthalic acid 2 (3.62 g, 20 mmol) in DMF (30 ml), acetic
anhydride (2.83 ml, 30 mmol) and acetic acid (0.6 ml, 10 mmol)

5466
D.M. Kelly, H.A. Moynihan / Tetrahedron 66 (2010) 5459e5466
were added. The reaction mixture was stirred at room temperature
for 4 h. Water (20 ml) was then added and the reaction mixture was
allowed to stir for a further 30 min at 0 ꢀC. The precipitate, which
formed was isolated by filtration and recrystallised from ethanol to
afford 2 (4.21 g, 94%) as a white solid: mp 313e315 ꢀC (decomp.); IR
References and notes
1. Bernstein, J. Polymorphism in Molecular Crystals; Clarendon: Oxford, 2002.
2. Hilfiker, R. Polymorphism: In the Pharmaceutical Industry; John Wiley & Sons:
Weinheim, 2006.
3. (a) Almarsson, Ö; Zaworotko, M. J. Chem. Commun. 2004, 1889e1896; (b) Re-
menar, J. F.; Morissette, S. L.; Peterson, M. L.; Moulton, B.; MacPhee, J. M.;
Guzman, H. R.; Almarsson, Ö J. Am. Chem. Soc. 2003, 125, 8456e8457; (c) Os-
wald, I. D. H.; Motherwell, W. D. S.; Parsons, S.; Pidcock, E.; Pulham, C. R.
Crystallogr. Rev.. 2004, 10, 57e66.
(KBr)
2.16 (3H, s), 8.22 (1H, s), 8.49 (2H, s), 10.39 (1H, br s).
n
3546, 3450,1722, 1671,1616, 1573,1205; 1H NMR (DMSO-d6)
d
4. Edwards, A. D.; Shekunov, B. Y.; Kordikowski, A.; Forbes, R. T.; York, P. J. Pharm.
Sci. 2001, 90, 1115e1124.
5. McCausland, L. J.; Cains, P. W.; Martin, P. D. Chem. Eng. Prog. 2001, 56e61.
6. Zaccaro, J.; Matic, J.; Myerson, A. S.; Garetz, B. A. Cryst. Growth Des. 2001, 1, 5e8.
7. Chyall, L.; Tower, J. M.; Coates, D. A.; Houston, T. L.; Childs, S. L. Cryst. Growth
Des. 2002, 2, 505e510.
8. Monissette, S. L.; Almarsson, Ö; Peterson, M. L.; Remenar, J. F.; Read, M. J.; Lemmo,
A. V.; Ellis, S.; Cima, M. J.; Gardner, C. R. Adv. Drug Deliv. Rev. 2004, 56, 275e300.
9. Hughes, C. E.; Hamad, S.; Harris, K. D. M.; Catlow, C. R. A.; Griffiths, P. C. Faraday
Discuss 2007, 136, 71e89.
10. Spitaleri, A.; Hunter, C. A.; McCabe, J. F.; Packer, M. J.; Cockcroft, S. L. Crys-
tEngComm. 2004, 6, 489e493; Chiarella, R. A.; Gillon, A. L.; Burton, R. C.; Davey,
R. J.; Sadiq, G.; Auffret, A.; Cioffi, M.; Hunter, C. A. Faraday Discuss. 2007, 136,
179e193.
4.7. Crystallizations
All recrystallisations were carried out in 100 ml round bottom
flasks. Solutions of L-glutamic acid were prepared in deionised
water at the following concentrations: 20 g Lꢁ1, 35 g Lꢁ1, 45 g Lꢁ1
,
maintained at 18 ꢀC, 38 ꢀC and 45 ꢀC, respectively. The solutions
were heated to dissolve the acid and additives. In some cases fil-
tration was necessary prior to recrystallisation. Samples of the
form of -glutamic acid were prepared by the method of Garti and
Zour7 as follows: Monosodium-
-glutamate, (5 g, 29.5 mmol), was
a-
L
L
11. Berkovitch-Yellin, Z.; Addadi, L.; Idelson, M.; Lahav, M.; Leiserowitz, L. Angew.
dissolved in water (50 ml). Concd HCl (2 ml) was added dropwise
and the solution was stirred for 10 min at room temperature. A
further 1 ml concd HCl was then added dropwise and the resulting
precipitate was collected by filtration to yield rhombic crystals
(1.99 g, 13.5 mmol, 45.7%), mp 199e204 ꢀC.
Chem. Suppl. 1982, 1336e1345.
12. Blagden, N.; Davey, R. J.; Rowe, R.; Roberts, R. Int. J. Pharm. 1998, 172, 169e177;
Mukuta, T.; Lee, A. Y.; Kawakami, T.; Myerson, A. S. Cryst. Growth Des. 2005, 5,
1429e1436.
13. Lang, M.; Grzesiak, A. L.; Matzger, A. J. J. Am. Chem. Soc. 2002, 124,
14834e14835; Price, C. P.; Grzesiak, A. L.; Matzger, A. J. J. Am. Chem. Soc. 2005,
127, 5512e5517.
14. Miura, H.; Ushio, T.; Nagai, K.; Fujimoto, D.; Lepp, Z.; Takahashi, H.; Tamura, R.
Cryst. Growth Des. 2003, 3, 959e965.
15. Vishweshwar, P.; McMahon, J. A.; Oliveira, M.; Peterson, M. L.; Zaworotko, M. J.
4.7.1. Crystal habit. Crystal habits were observed using a Nikon
ECLIPSE 50i POL Polarizing Microscope. Crystal pictures were
obtained using a Nikon COOLPIX 8400 digital camera with 8.0 ef-
fective megapixels.
J. Am. Chem. Soc. 2005, 127, 16802e16803.
16. Weissbuch, I.; Berkovic, G.; Leiserowitz, L.; Lahav, M. J. Am. Chem. Soc. 1990, 112,
5874e5875; Weissbuch, I.; Frolow, F.; Addadi, L.; Leiserowitz, L.; Lahav, M. J. Am.
Chem. Soc. 1990, 110, 561e567; Bonafede, S. J.; Ward, M. D. J. Am. Chem. Soc.
1995, 117, 7854e7861; Weissbuch, I.; Leisorowitz, L.; Lahav, M. Adv. Mater. 1994,
6, 952e956.
17. Berkovitch-Yellin, Z.; van Mil, J.; Addadi, L.; Idelson, M.; Lahav, M.; Leiserowitz,
L. J. Am. Chem. Soc. 1985, 107, 3111e3122; Addadi, L.; Berkovitch-Yellin, Z.;
Weissbuch, I.; van Mil, J.; Shimon, L. J. W.; Lahav, M.; Leiserowitz, L. Angew.
Chem., Int. Ed. Engl. 1985, 24, 466e485; Domopoulou, A.; Michaelides, A.;
Skoulika, S.; Kovala-Demertzi, D. J. Cryst. Growth 1998, 191, 166e168.
18. Torbeev, V. Y.; Shavit, E.; Weissbuch, I.; Leiserowitz, L.; Lahav, M. Cryst. Growth
Des. 2005, 5, 2190e2196.
4.7.2. Powder X-ray diffraction (PXRD). Powder X-ray diffraction
was performed at ambient temperature using a Stoe Stadi MP PXRD
operating in transmission mode with a linear PSD detector with an
anode current of 40 mA, an accelerating voltage of 40 kV and
Cu Ka1 X-radiation (l¼1.5406 Å) over a scan range of 3.5ꢀ to 60ꢀ 2
q,
scanning in steps of 2ꢀ for 90 s per step. Samples were held be-
tween acetate foils and were not ground. Calculated patterns were
generated from crystallographic information files downloaded
from the Cambridge Structural Database, using the THEO function
on the Stoe WinXPOW software with a pseudo-Voigt profile-shape
function and a Gauss component of 0.8.
19. Davey, R. J.; Blagden, N.; Potts, G. D.; Docherty, R. J. Am. Chem. Soc. 1997, 119,
1767e1772.
20. Staab, E.; Addadi, L.; Leiserowitz, L.; Lahav, M. Adv. Mater. 1990, 2, 40e43.
21. Kelleher, J. M.; Lawrence, S. E.; McAuliffe, M. T.; Moynihan, H. A. CrystEngComm.
2007, 9, 72e77.
22. Hirayama, N.; Shirahata, K.; Ohashi, Y.; Sasada, Y. Bull. Chem. Soc. Jpn. 1980, 53,
30e35.
23. Hirokawa, S. Acta Crystallogr. 1955, 8, 637e641.
Acknowledgements
24. Garti, N.; Zour, H. J. Cryst. Growth 1997, 172, 486e498.
25. Pless, D. D.; Lee, Y. C.; Roseman, S.; Schnaar, R. L. J. Biol. Chem. 1983, 258,
2340e2349.
26. Shendage, D. M.; Froehlich, R.; Haufe, G. Org. Lett. 2004, 6, 3675e3678.
27. Jakobsen, C. M.; Denmead, S. R.; Isaacs, J. T.; Gady, A.; Olsen, C. E.; Christensen,
S. B. J. Med. Chem. 2001, 44, 4696e4703.
The authors are grateful for the award of on Irish Research
Council of Science, Engineering and Technology Embark Post-
graduate Scholarship (to DMK).