Discovery and recovery of delta p-aminobenzoic acid

Article abstract of DOI:10.1039/C8CE01882K

We explore the polymorphism of p-aminobenzoic acid (pABA) under high-pressure conditions. We have been able to isolate a new high-pressure form (δ-pABA) at pressures exceeding 0.3 GPa from three different pressure-transmitting media, water, water?:?ethanol and pure ethanol. We explore the compression behaviour of α-pABA in each of these media using neutron powder diffraction and observe that the transition is kinetically hindered using the aqueous ethanol and ethanol solutions compared with the pure aqueous medium. δ-pABA is sufficiently stable to be recovered to ambient pressure to enable its characterisation via X-ray powder diffraction and differential scanning calorimetry. At ambient pressure we have observed that δ-pABA converts into α-pABA on heating beyond 70 °C.

Full text of DOI:10.1039/C8CE01882K

View Article Online
View Journal
CrystEngComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. R. Ward, S.
Younis, A. J. Cruz-Cabeza, C. L. Bull, N. P. Funnell and I. D. H. Oswald, CrystEngComm, 2019, DOI:
1
0.1039/C8CE01882K.
Volume 18 Number 1 7 January 2016 Pages 1–184
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
CrystEngComm
Accepted Manuscripts are published online shortly after
www.rsc.org/crystengcomm
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
HIGHLIGHT
Tiddo J. Mooibroek, Antonio Frontera et al.
Towards design strategies for anion–π interactions in crystal engineering
rsc.li/crystengcomm

Page 1 of 18
CrystEngComm
View Article Online
Discovery and Recovery of delta p-aminobenzoic acid
DOI: 10.1039/C8CE01882K
1
1
2
3
Martin R. Ward, Shatha Younis, Aurora J. Cruz-Cabeza, Craig L. Bull, Nicholas P.
3
1
Funnell, and Iain D.H. Oswald
1
Strathclyde Institute of Pharmacy & Biomedical Sciences (SIPBS), University of
Strathclyde, 161 Cathedral Street, G4 0RE, Glasgow, U.K.
2
School of Chemical Engineering and Analytical Science, University of Manchester, M13
9
PL Manchester, United Kingdom
3
ISIS Neutron and Muon Source, Science and Technology Facilities Council, Rutherford
Appleton Laboratory, Harwell, Didcot, OX11 0QX, UK
Abstract
We explore the polymorphism of p-aminobenzoic acid (pABA) under high-pressure conditions. We
have been able to isolate a new high-pressure form (δ-pABA) at pressures exceeding 0.3 GPa from
three different pressure-transmitting media, water, water:ethanol and pure ethanol. We explore the
compression behaviour of α-pABA in each of these media using neutron powder diffraction and observe
that the transition is kinetically hindered using the aqueous ethanol and ethanol solutions compared with
the pure aqueous medium. δ-pABA is sufficiently stable to be recovered to ambient pressure to enable
its characterisation via X-ray powder diffraction and differential scanning calorimetry. At ambient
pressure we have observed that δ-pABA converts into α-pABA on heating beyond 70 °C.
Keywords: p-aminobenzoic acid, neutron powder diffraction, high pressure, recovery
Introduction
The discovery of and relationships between different solid-state forms of molecular materials
is of paramount importance to industry due to the changes in physicochemical properties that
1
may exist between phases. The nucleation of specific polymorphs of a pharmaceutical, for
example, can be initiated through seeding experiments to ensure that the correct polymorph is
isolated from a crystallisation. From that point on, the behaviour of the solid is assumed to be
well controlled and hence the ability to process into a formulated product can be achieved.
However, at every stage of the operation there is the possibility of transformation of the
desired polymorph into a new, as yet, unidentified polymorph that can cause issues in the
formulated product.2
To explore the solid-state landscape fully we need to be able to probe both the
thermodynamic variables of temperature and pressure. The investigation of solid-state

CrystEngComm
Page 2 of 18
View Article Online
DOI: 10.1039/C8CE01882K
polymorphs using temperature has become routine with the advent of variable temperature
devices for analytical equipment such as calorimetry and diffraction.3–5 Whilst significant
improvements and developments of both hardware and software have been made, which
permit laboratory based characterisation of the effects of pressure, there are still relatively
few such studies reported to date.6–17 Phase transformations of organic materials at high
pressure can potentially be observed by purely compressing materials into new forms via a
single crystal-to-single crystal mechanism,18–23 or through reconstructive mechanisms that
result in the degradation of the crystals.24,25 The characterisation of the latter can prove
challenging, however, depending on the pressure at which the transition occurs, one can use a
solution of the material and perform a recrystallisation at high pressure. This method has
been used to facilitate the growth of single crystals of these degraded samples for single
crystal X-ray diffraction measurement.11,26
Para-aminobenzoic acid (pABA) has three polymorphs and has been the subject of numerous
crystallisation studies at various temperatures.27–35 For a more comprehensive review of
pABA polymorphism readers are directed to the article by Cruz-Cabeza et al. following this
one.{REF TO BE PROVIDED ONCE DOI ASSIGNED} Its simple molecular structure
provides a rich phase behaviour that signifies the potential for the discovery of new phases at
high pressure. Yan et al investigated the pressure dependence of α- and β-pABA (both
P2 /n) observing that both phases are stable to 13 GPa using an inert pressure-transmitting
1
medium, silicone oil.36 Through ab initio approaches they calculated the compression of the
crystal structure assuming there was no change in phase and attributed the stability of the
polymorphs to the hydrogen bonded dimers and tetramers in α-pABA and β-pABA
respectively. Following this work, we wanted to explore how pABA would react to
compression using different pressure transmitting media in which the solubility of pABA
varied from low (water) to high (ethanol). In the following paper we describe the
crystallisation behaviour using neutron diffraction and spectroscopic techniques.
Methods
Sample preparation
p-Aminobenzoic acid (pABA; purity 99%, Sigma-Aldrich A9878; LOT MKBZ3723V) was used to
prepare individual samples prior to loading Diamond Anvil cells. An excess of pABA was weighed in
1
ml of the selected solvent (water; 50:50 water:ethanol; ethanol) and stirred for approximately 2-4 hours

Page 3 of 18
CrystEngComm
View Article Online
at room temperature (293 K). The solution was filtered into a new vial using a 0.2 µm filter, prior to
DOI: 10.1039/C8CE01882K
loading the cells.
High pressure
3
7
A Merrill- Bassett Diamond Anvil Cell (DAC) was equipped with two Boehler-Almax gem-diamonds
with culets of 600 µm seated in tungsten carbide backing discs. A 250 µm hole was drilled in a pre-
indented tungsten gasket with a thickness of 90 µm using a Boehler-Almax microdriller. Small pieces
3
8
of ruby were loaded into the sample chamber to act as a pressure calibrant. For the compression studies
a powdered sample of pABA was added to the sample chamber along with a drop of the pre-prepared
solution described above.
For the recrystallisation experiments using ethanol or the water:ethanol mixture a saturated solution
was added to the DAC together with a piece of ruby (to act as a pressure marker by the standard
3
9
fluorescence method). No further addition of solid material was required as there was enough solute
in the solution to allow precipitation on application of pressure. In the case of water, the solubility of
pABA was low enough that a small crystallite of α-pABA was required along with the saturated solution
so that on precipitation at pressure there was enough sample for analysis by X-ray diffraction. At each
pressure point (Ethanol: 0.22 and 0.49 GPa; mixture: 0.8 and 1.2 GPa; water: 0.27, 0.33 and 0.71 GPa),
the cell was heated to dissolve the small crystallite and subsequently cooled to allow the precipitation.
In some instances, pressure-cycling was required to initiate precipitation. Once a solid material
appeared the temperature was cycled from room temperature to 323 K so as to anneal the small
crystallites into one singular crystal or only a few smaller crystals that were in different orientations
providing the potential for extra data given the limited access via the pressure cell.
Raman spectroscopy
A Horiba Xplora Raman Microscope from Horiba Scientific John Yvon coupled with a 532nm
excitation laser was used to measure the Raman spectra of the samples at high pressure. Slit, hole,
filters and accumulation times were varied to maximise the sample signal and reduce any fluorescence
that may have come from the sample environment.
Infrared Spectroscopy
Fourier Transform Infrared spectroscopy (FT-IR) data was collected using a Bruker Tensor-II
spectrometer equipped with a DigiTect 24 Bit ADC detectors. For each spectrum 16 scans were
-1
-1
performed with a resolution of 4 cm over the range 400-4000 cm .
Differential scanning calorimetry
Differential Scanning Calorimetry (DSC) data was collected using a Netzsch DSC 214 Polyma
calibrated for both temperature and sensitivity over the temperature range -93°C to 605°C, at a heating

CrystEngComm
Page 4 of 18
-1
View Article Online
DOI: 10.1039/C8CE01882K
rate of 20°C min using thermal standards (indium, tin, bismuth and zinc) supplied by Netzsch.
Approximately 6 mg of sample material was weighed into pre-weighed aluminium pans before sealing
with a pierced lid. Samples were subject to 2 heating and cooling cycles from 20 to 200 °C at a rate of
-1
2
0 °C min . A 5-minute isothermal hold was programmed after each heat and cool step. The furnace
-1
was purged with helium (60 mL min ) during the experiments.
Single Crystal X-ray Diffraction.
Single crystal data were collected on a Bruker D8 Venture diffractometer equipped with an Incoatec
IµS microfocussed Cu X-ray source (Kα = 1.5406 Å). Suitable crystals were identified and mounted
1
on a low-background Kapton microloop (200 μm). Crystals were indexed using a fast scan experimental
method from which the collection strategy was determined using Bruker Apex3 software. Scaling40
and data reduction were also performed using Apex3 software. For low-temperature analysis, data were
collected at 100 K using an Oxford Cryostream cooling system.41 Structures were solved by intrinsic
4
2
43
phasing using ShelXT through Olex2 (v1.2) software. Full matrix least-squares refinement of data
was also performed using Olex2 with ShelXL. All non-hydrogen atoms were treated anisotropically.
Hydrogen atoms were placed geometrically (followed by a positional refinement for the acidic/amine
hydrogens) and were subsequently constrained to ride on their parent atoms.
High-pressure diffraction data were collected on a Bruker APEX2 diffractometer equipped with an
Incoatec IµS microfocussed Mo X-ray source (Kα = 0.71073 Å). Data collection procedures followed
1
4
4
those of Dawson et al. with the addition of four runs at χ = 90°. The data were reduced in APEX3
using the dynamic masking procedures implemented in the software. Absorption corrections were
4
0
applied using SADABS as implemented in the Scale function in APEX3. Ambient pressure structures
4
3
were used as the starting models for the refinements using Olex2. RIGU restraints were used to aid
the modelling of the displacement parameters. Hydrogen atoms were placed geometrically and allowed
to refine before riding constraints were applied.
Neutron powder diffraction
4
High-pressure neutron powder diffraction data were collected for pABA-d (CDN isotopes,
4
5
#
D4209, lot #X-496) using the PEARL diffractometer at the ISIS spallation neutron source located
at the STFC Rutherford Appleton Laboratory. Three experiments were conducted with different
pressure-transmitting media. The sample was ground and placed into a standard Ti-Zr alloy
4
6
encapsulated gasket. The media was added dropwise to the sample to provide quasi-hydrostatic
conditions during the compression. A lead pellet (~80 mg) was added to the gasket to act as a
suitable pressure marker. The resulting capsule assembly was then compressed within a type V3b
Paris-Edinburgh (P-E) press27 equipped with standard profile anvils with cores fabricated from
zirconia toughened alumina (ZTA). The P-E cell piston pressure was monitored and controlled by
means of an automated hydraulic system. For the pure D O system we observed poor wetting of
2

Page 5 of 18
CrystEngComm
View Article Online
the solid during initial attempts at loading the sample. A successful loading was performed where
DOI: 10.1039/C8CE01882K
4
the pABA-d sample was loaded with a pellet of frozen D O. The D O melted within the timeframe
2
2
of loading the press onto the diffractometer which was confirmed through the lack of ice peaks in
the diffraction patterns.
Time-of-flight (TOF) neutron powder diffraction data suitable for Pawley refinement were
collected and the diffraction pattern intensities were corrected for the wavelength and scattering-
angle dependence of the neutron attenuation by the P-E cell anvil and gasket (Ti-Zr) materials.45,47
Sample pressures were calculated from the refined lead lattice parameters and the known equation
of state.
Each sample was compressed and data were collected at regular intervals to a maximum
pressure of 1.104(9) GPa (D O) and ~2 GPa (EtOD:D O; EtOD). In general, data were collected at
2
2
each pressure point for 30-60 minutes with longer collections at the maximum pressure (2 hrs). For
all samples, data were collected at various pressures on decompression (EtOD: 0.38 GPa, ambient;
EtOD:D O: 1.57, 1.46, 0.94, 0.89, 0.64, 0.02 GPa, ambient; D O: only ambient). The final ambient
2
2
pressure collection was to observe the recovery of the high-pressure phase.
Large volume press
7
48
A large volume press (LVP) was used to prepare δ-pABA-h at University of Strathclyde.
The starting material was identified as α-pABA by X-ray powder diffraction (XRPD). The material
was lightly ground before loading ca. 300 mg in to a polytetrafluoroethylene (PTFE) sample chamber
that was comprised of a cylindrical tube (ID = 8mm, OD = 10mm) the ends of which were sealed using
PTFE caps and PTFE sealing tape. The remaining volume of the sample chamber was filled with a
saturated aqueous pABA solution (298 K). The PTFE sample chamber was transferred to the large
volume press cell assembly and a maximum load of 7 tonnes was applied to the sample (equivalent
pressure = 0.8 GPa). The sample was held at elevated pressure conditions for a period of ca. 16 hours.
Over the course of the 16 hours, the load on the sample had decreased to 6 tonnes (P = 0.69 GPa)
indicating a loss of pressure over the time period. Sample solids were rapidly isolated from the aqueous
solution by filtration. During filtration it was noted that the recovered sample was comprised of well-
defined crystals with a pink hue. Recovered samples were stored in a sealed container prior to analysis.
The samples were characterized by means of X-ray diffraction (SC-XRD and XRPD), Raman, FT-IR
and DSC.
Results and Discussion
Single crystal diffraction
We investigated pABA using the established recrystallisation at high-pressure methodology where
a saturated solution of solute in a solvent system is loaded into a pressure vessel and pressure

CrystEngComm
Page 6 of 18
applied to induce crystallisation.49,50 The morphology of α-pABA is a distinctive needle shape
View Article Online
DOI: 10.1039/C8CE01882K
3
4
compared with the prism and block morphology of β- and γ-forms, respectively. Hence, we would
have indications of new solid forms depending on the morphology of the crystals grown at pressure.
The γ-form has also been identified as needles which makes the separation from the α-form
difficult.32 Through sequential application of pressure and heating, we were able to observe the
growth of the crystals and their morphology with subsequent identification using single-crystal X-
ray diffraction. From all three media block-like crystals were isolated at pressures beyond 0.3 GPa
(
Table ES1). These crystals were indexed as a new monoclinic phase (Pn, Z’ = 1) which we will
designate as δ-pABA. δ-pABA has a similar packing as β-pABA with a perpendicular hydrogen
bonding arrangement between the acid group and the amine to form chains along the [1 0 1]
direction (Figure 1b), however it is the interchain interaction that differs. In δ-pABA, the
neighbouring π-stacked molecule is orientated in the same direction whilst the molecules are anti-
parallel in the β-polymorph. The parallel chains are linked to each other through the amine
hydrogen bond to the carbonyl moiety of the acid group. The second hydrogen on the amine remains
unfulfilled indicating the packing of molecules is of paramount importance rather than the
fulfilment of the hydrogen bonding groups.
Figure 1: a) the numbering scheme for pABA; b) the amine-carboxylic acid hydrogen bond forms
‘
head-to-tail’ chains (central three molecules) whilst the carboxylic acid hydrogen bonds to the
amine in a perpendicular interaction between neighbouring molecules (vertical interactions); and
c) the packing of the δ-phase viewed down the b-axis.

Page 7 of 18
CrystEngComm
View Article Online
Compression of saturated solutions to obtain the δ-polymorph led to the inevitable issue of
DOI: 10.1039/C8CE01882K
dissolution of the sample on decompression. To overcome this problem we began to explore the
effect of the solute/solution properties on the behaviour of pABA under high pressure conditions
with a view that the increase in solubility in the pressure-transmitting medium may overcome the
kinetic barriers to interconversion. Previous work on the α- and β-forms at high pressure using
3
6
Raman spectroscopy indicated that each of the phases were stable on compression to 13 GPa. The
stability of the α- and β-forms was attributed to the favourable dimer and tetramer hydrogen
bonding configurations observed in the two phases. Both of these studies were performed in an
inert silicone oil medium which may have helped to supress the conversion into the δ-form from
either of these initial phases due to the significant molecular rearrangement required. The change
to media in which there is some solubility of pABA may facilitate the transition.
High pressure Neutron diffraction
To follow the single crystal measurements and provide structural information with respect to the
compression behaviour we used neutron powder diffraction for phase identification. The advantage
of not requiring heat to anneal the solid into a single crystal allowed us to monitor the transition
purely with pressure and bypass any effects of temperature. The unit cell parameters of the
previously-determined α-polymorph of pABA at 100 K are a = 18.562(1) Å, b = 3.732(<1) Å, c =
2
8
1
8.568(1) Å, β = 93.99(<1)°. The large unit cell made the fitting of the data extremely challenging
due to peak overlap in the d-spacing range available on the Pearl instrument however, we were able
to obtain a Pawley fit to the data at all pressures (Figure S1). Figure 2 shows the powder diffraction
patterns of pABA in the three different solvent compositions on increasing pressure. From these
experiments we observe that the behaviour of pABA is different depending on the solvent that is
used as the pressure-transmitting medium. The pABA sample in D O converts cleanly from α-
2
pABA to δ-pABA (Figure 2a). The conversion to δ-pABA appears to begin at approximately
0
.79(8) GPa and is complete by 0.956(10) GPa. The Pawley fits to the ambient pressure data and
those collected at 1.104(9) GPa are shown in Figure 3. These refinements show that α-pABA is
present from the beginning of the experiment rather the thermodynamically stable β-pABA. The
sample at 1.104(9) GPa shows a misfit at 2.5 Å which we attribute to ice VI. The samples in ethanol
and water:ethanol show no conversion to δ-pABA on compression (Figure 2b&c). Our original
hypothesis that enhanced solubility in the pressure medium facilitates the α- to δ-pABA
transition (in part informed by the absence of transformation when using low-solubility silicone
3
6
oil ) cannot be reconciled with our observations here. We believe that the viscosity of the
solution may be causing difficulties in molecular mobility hence the nucleation of the high-pressure
phase. An increase in viscocity with pressure is observed in pure methanol and 4:1
methanol:ethanol liquid systems and so one can assume that becomes even more significant with

CrystEngComm
Page 8 of 18
5
1
View Article Online
DOI: 10.1039/C8CE01882K
solute present. During loading of the samples, either in the DAC or PE press, there is little control
over the quantity of sample and solvent loaded hence no way in which to measure the solubility or
supersaturation which can change with pressure.52 We later followed the same thermodynamic
route using optical microscopy study within a DAC offline. For the sample in water α-pABA
converts to δ-pABA within 45 minutes at 0.85 GPa which was easily identifed using Raman
spectroscopy (Figure 4). This was a similar timespan to that observed in the neutron diffraction
experiment. For the 50:50 mixture we observed that over a timespan of 8 hours at 1.35 GPa the
conversion of α-pABA into another new phase occurs, which remains unidentified. The Raman
spectrum for this sample was collected and shown to be different to the known phases for which
we have Raman data (Figure S2). Annealing the powder of the new phase resulted in the formation
of δ-pABA coupled with a reduction of pressure to 0.80(5) GPa . At 1.85(5) GPa the α-form persists
for 75 hours. We reduced the pressure to 0.99(5) GPa for 24 hours and over this time period the
sample did not change and remained as the α-form. We further reduced the pressure to 0.75(5) GPa
and we observed that the transition between α-form and δ-form occurs within 7 hours after which
the pressure was determined to be 0.64(5) GPa.
4
2
Figure 2: Neutron diffraction data for the compression of pABA-d in (a) water-d ; b) 50:50 water-
2
6
6
d :ethanol-d mixture; and c) pure ethanol-d . The conversion to the δ-form is complete in the
2
water-d by 0.956(10) GPa whereas the conversion to δ-pABA does not occur in the other two
media. Asterisks indicate the peaks of δ-pABA emerging from α-pABA.

Page 9 of 18
CrystEngComm
View Article Online
DOI: 10.1039/C8CE01882K
Figure 3 a) The Pawley fit of the α-form to the ambient pressure neutron diffraction dataset of
4
2
pABA-d in water-d . b) The Pawley fit of the δ-form to the ambient pressure neutron diffraction
4
2
dataset of pABA-d in water-d collected at 1.104(8) GPa. There is a misfit at ~2.5 Å that can be
attributed to the (2 1 1) reflection of ice VI.
Figure 4: Raman spectra for the α-form and δ-form showing the large difference in the 50-1000
-
1
cm range.

CrystEngComm
Page 10 of 18
View Article Online
On decompression to ambient pressure we observe in both the 50:50 water:ethanol and pure ethanol
DOI: 10.1039/C8CE01882K
samples that there is conversion to δ-pABA. This begins at 0.638(16) GPa in the mixture whilst it
is already complete by 0.376(6) GPa in ethanol (Figure 5). One of the key features is that, in all
three cases, δ-pABA is sufficiently stable at ambient pressure for recovery and charaterisation. This
is an exciting observation as it is a further example of a material whose high-pressure polymorph
is sufficiently stable to be recovered to ambient pressure. A number of other materials have been
1
6
53
shown to exhibit this phenomenon, for example γ-amino butyric acid (GABA).H O, paracetamol,
2
maleic acid,53 3-hydroxy-4,5-dimethyl-1-phenylpyridazin-6-one54 and glycolide.24,48 Due to the
constraints of allocated measurement time on the Pearl instrument we were only able to monitor
the recovered forms for 2 hours however over this time period there was no sign of conversion.
Figure 5: The diffraction patterns of pABA on decompression from the highest pressure achieved
2
6
6
in each medium a) 50:50 water-d :ethanol-d ; and b) ethanol-d . The α-form transforms cleanly to
the δ-form on decompression and persists at ambient pressure. The growth of the δ-form from the
α-form is shown by asterisks in a).
Recovery of δ-pABA – Large Volume Press
The ability to analyse material properties in-situ at high pressure is challenging but with pABA-h7
we had the opportunity to recover the high-pressure phase and investigate its properties under
ambient pressure. Using the LVP we were able to recreate the conditions of the Pearl experiment
but in a much larger quantity (300 mg scale). The maximum pressure generated by the press is 0.8
7
GPa which is beyond the transition point hence we held α-pABA-h in a saturated aqueous solution
at this pressure for 16 hours. On downloading, the recovered sample showed a plate morphology
in contrast with the needle-like morphology of α-pABA (Figure 6a); this is consistent with the
block morphology observed in the cell where the environment can geometrically constrain the
morphology. The unit cell parameters of the δ-form can be fitted to the X-ray diffraction pattern of
a sample recovered from the press that has not been ground (Figure 6b). We attempted to collect

Page 11 of 18
CrystEngComm
View Article Online
capillary data on the sample but it appears that on grinding the powder δ-pABA converts to a
DOI: 10.1039/C8CE01882K
mixture of β-pABA and α-pABA (Figure 6c).
Figure 6: a) Crystals of recovered δ-pABA at ambient pressure. The plate morphology is distinct
from the needle and block morphologies of α-pABA and β-pABA. b) The Pawley fit of the
recovered δ-pABA without prior grinding. Black tick marks represent the positions of the δ-pABA
reflections. c) The Pawley fit of data from collection in a capillary indicating that after 30 minutes
and grinding the sample the conversion to a mixture of β-pABA (black ticks) and α-pABA (red
ticks) has occurred.
The distinct morphology allowed us to have confidence throughout our analysis that δ-pABA was
stable for a matter of weeks as long as we did not grind the powder. Differential scanning
calorimetry was performed on the recovered phase and its trace is shown in Figure 7. On heating
δ-pABA exhibits a small shallow endotherm (+2 kJ/mol) caused by the reconstructive transition to
α-pABA which melts at 187.7ºC. On cooling, the solid nucleates. Subsequent analysis of the solid
has identified the solid as the γ-polymorph through use of the IR absorbance bands at 891 and 911
-
1
34
-1
cm (Figure S3); these peaks are far more separated in α-pABA (890 and 922 cm ). The IR
analysis and DSC trace leads us to believe that the γ-form is stable until the melt although there

CrystEngComm
Page 12 of 18
View Article Online
may be the possibility that both γ-pABA and α-pABA are present due to the double peak on melting
DOI: 10.1039/C8CE01882K
in the second cycle.
Figure 7: The differential scanning calorimetry trace for δ-pABA. Data were collected at 20°Cmin-
1
from 20-200°C. The black trace is the original material from the LVP identified as δ-pABA that
undergoes the transition to α-pABA on heating (76°C). The red trace is the cooling of the melt
with recrystallisation at ~160°C. The blue trace is the reheating of the recrystallised material
which we have identified through IR spectroscopy to be γ-pABA. The pink trace is the cooling of
the melt with recrystallisation taking place at 156°C. Note the reduced peak due to some
decomposition on melting as well as a change in endotherm shape due to the change in contact
between the sample and the base of the pan.
The discovery and recovery of δ-pABA has demonstrated that high pressure can be a powerful
tool for the exploration of the solid state. The result presented herein joins a growing list of
compounds whose high-pressure polymorphs are recoverable to ambient pressure as a metastable
form. Of those polymorphs recovered to ambient pressure there are a number for which high
pressure routes are the only method for their production which highlights the potential for
pressure as a screening tool for polymorphism. Aminobutyric acid, glycolide, 3-hydroxy-4,5-

Page 13 of 18
CrystEngComm
View Article Online
DOI: 10.1039/C8CE01882K
dimethyl-1-phenylpyridazin-6-one and paracetamol form novel, recoverable phases that have not
been observed at ambient pressure despite many crystallisation experiments. Fabbiani and co-
workers were able to demonstrate the recovery of γ-aminobutyric acid monohydrate to ambient
5
5
pressure. In this case, pressure was applied to 4 M solutions of aminobutyric acid and through
pressure precipitation the hydrate was observed at 0.44 GPa. The authors were able to
decompress the hydrate and use the crystal to seed ambient pressure solutions. The key
observation in this work was that the high-pressure precipitation method was the only way in
which to observe the hydrated form despite an extensive screen of crystallisation conditions.
Seeding was also possible in the case of glycolide. Our group has recovered a new high-pressure
polymorph of glycolide in large quantities (300mg). Using the LVP we were able to isolate and
recover the high-pressure form to ambient pressure and use these crystals as seeds for
crystallisation experiments to obtain single crystal for identification; the high pressure form was
stable for up to 12 days.48
Pressure-preciptiation was the sole route for the isolation of the γ-polymorph of 3-hydroxy-4,5-
5
4
dimethyl-1-phenylpyridazin-6-one. Roszak et al discovered that the high-pressure phase (γ-form)
could only be isolated at high pressure despite the polymorph showing extensive metastability (over a
year) in open vials on the bench at ambient pressure. The authors did not state whether seeding
experiments were performed on these materials. One key observation of the structure was the high
energy molecular conformation adopted by the molecules in the γ-form (forced by efficient packing)
may have prevented isolation through ambient pressure routes.
Finally, Oswald et al. have demonstrated that the metastable orthorhombic polymorph of
paracetamol can be succesfully isolated to ambient pressure using a variety of concentrations.53
The stability of the orthorhombic form was altered depending on the concentration of solutions
used for the precipitation experiment. At >140mg/ml of paracetmol in water the authors
demonstrated that the solid converts to the monoclinic form over the course of several hours but
that the conversion can be retarded by cooling to 275-280 K. A sample at 100mg/ml was stable
for a period of days and provided crystals satisfactory for single crystal X-ray analysis. With
concentrations <100 mg/ml the monohydrate of paracetamol was obtained (sole route to isolation)
and recovered. In both cases, the crystals recovered from high pressure could be used as seed
crystals.
In each of these examples in the literature and the present one in this study, the discovery and
isolation of these forms has been serendipitous hence there is a strong need to be able to develop
strategies to identify materials that are likely to i) undergo a transition to a new form; and ii) be
recoverable to ambient pressure. Only once these strategies have been developed can high-
pressure precipitation be a robust tool for industry as a polymorph screening methodology.

CrystEngComm
Page 14 of 18
View Article Online
DOI: 10.1039/C8CE01882K
In conclusion, we have shown that pABA undergoes a phase transformation at high pressure from
α-pABA to a new polymorph that we have designated δ-pABA. Through precipitation from
solution we have identified that the pressure of the phase transition is relatively low at ~0.3 GPa.
We have observed that whilst having low solubility in water the phase transformation takes place
easily over a short timescale (45 mins). In solvents in which pABA is more soluble (e.g. ethanol)
the transition is hindered (8 hours) but it is accelerated by decompression of the solution to
ambient pressure. Molecular mobility in the solvent media is cited as a possible explanation for
this. The high-pressure δ-pABA is recoverable to ambient pressure as a metastable form and
converts to α- and β-pABA on grinding at room temperature (20-25°C); heating converts δ-pABA
to α-pABA .
Acknowledgements
The authors would like to thank Roger Davey of University of Manchester for useful discussions
and Lloyd Farquahar for help during beamtime. We would like to thank Science and Technology
5
6
Facilites Council for the award of beamtime (RB1810710). ACC would like to acknowlege the
Royal Society for an Industry Fellowship in Astra Zeneca and IDHO & MRW thank the EPSRC for
funding (EP/N015401-1). All data underpinning this publication are openly available from the
University of Strathclyde KnowledgeBase at XXX (to be validated once article is accepted). CCDC
deposition numbers are 1876686-1876690.
References
1
2
J. Bernstein, Cryst. Growth Des., 2011, 11, 632–650.
J. Bauer, S. Spanton, R. Henry, J. Quick, W. Dziki, W. Porter and J. Morris, Pharm. Res.,
2
001, 18, 859–66.
3
4
5
6
A. Fouskova, O. Sedmik and V. Svoboda, Ceskoslov. Cas. Pro Fys. Sekce a, 1971, 21, 275-.
F. Smutný, Phys. Status Solidi, 1972, 9, K109–K110.
J. Cosier, A. M. Glazer and IUCr, J. Appl. Crystallogr., 1986, 19, 105–107.
Y. V Seryotkin, T. N. Drebushchak and E. V Boldyreva, Acta Crystallogr. Sect. B, 2013, 69,
7
7–85.
7
8
B. A. Zakharov, Y. V Seryotkin, N. A. Tumanov, D. Paliwoda, M. Hanfland, A. V Kurnosov
and E. V Boldyreva, RSC Adv., 2016, 6, 92629–92637.
E. Patyk, J. Skumiel, M. Podsiadlo and A. Katrusiak, Angew. Chemie-International Ed., 2012,
5
1, 2146–2150.

Page 15 of 18
CrystEngComm
View Article Online
9
1
R. Lee, D. S. Yufit, M. R. Probert and J. W. Steed, Cryst. Growth Des., 2017, 17, 1647–1653.
DOI: 10.1039/C8CE01882K
0
D. I. A. Millar, H. E. Maynard-Casely, A. K. Kleppe, W. G. Marshall, C. R. Pulham and A. S.
Cumming, CrystEngComm, 2010, 12, 2524–2527.
1
1
2
N. Abbas, I. Oswald and C. Pulham, Pharmaceutics, 2017, 9, 16.
1
S. A. Moggach, W. G. Marshall and S. Parsons, Acta Crystallogr. Sect. B Struct. Sci., 2006,
6
2, 815–825.
1
3
C. L. Bull, G. Flowitt-Hill, S. de Gironcoli, E. Küçükbenli, S. Parsons, C. Huy Pham, H. Y.
Playford and M. G. Tucker, 2017, 4, 569–574.
1
1
1
4
5
6
I. D. H. Oswald and C. R. Pulham, CrystEngComm, 2008, 10, 1114–1116.
F. P. A. Fabbiani, G. Buth, B. Dittrich and H. Sowa, CrystEngComm, 2010, 12, 2541.
F. P. A. Fabbiani, G. Buth, D. C. Levendis and A. J. Cruz-Cabeza, Chem. Commun. (Camb).,
2
014, 50, 1817–9.
1
1
1
7
8
9
L. E. Connor, C. A. Morrison, I. D. H. Oswald, C. R. Pulham and M. R. Warren, Chem. Sci., ,
DOI:10.1039/c7sc01379e.
I. D. H. Oswald, A. R. Lennie, C. R. Pulham and K. Shankland, CrystEngComm, 2010, 12,
2
533.
A. Delori, I. B. Hutchison, C. L. Bull, N. P. Funnell, A. J. Urquhart and I. D. H. Oswald,
Cryst. Growth Des., 2018, 18, 1425–1431.
2
2
0
1
B. A. Zakharov and E. V Boldyreva, J. Mol. Struct., 2014, 1078, 151–157.
S. Bergantin, M. Moret, G. Buth and F. P. A. Fabbiani, J. Phys. Chem. C, 2014, 118, 13476–
1
3483.
2
2
2
2
3
4
B. A. Zakharov, E. A. Losev and E. V Boldyreva, CrystEngComm, 2013, 15, 1693.
N. P. Funnell, W. G. Marshall and S. Parsons, CrystEngComm, 2011, 13, 5841–5848.
I. B. Hutchison, C. L. Bull, W. G. Marshall, S. Parsons, A. J. Urquhart and I. D. H. Oswald,
Acta Crystallogr. Sect. B Struct. Sci. Cryst. Eng. Mater., 2017, 73, 1151–1157.
2
2
5
6
S. Li, Q. Li, J. Zhou, R. Wang, Z. Jiang, K. Wang, D. Xu, J. Liu, B. Liu, G. Zou and B. Zou, J.
Phys. Chem. B, 2012, 116, 3092–3098.
A. Y. Fedorov, D. A. Rychkov, E. A. Losev, B. A. Zakharov, J. Stare and E. V. Boldyreva,
CrystEngComm, 2017, 19, 2243–2252.

CrystEngComm
Page 16 of 18
View Article Online
2
7
R. A. Sullivan, R. J. Davey, G. Sadiq, G. Dent, K. R. Back, J. H. ter Horst, D. Toroz and R. B.
DOI: 10.1039/C8CE01882K
Hammond, Cryst. Growth Des., 2014, 14, 2689–2696.
2
2
3
8
9
0
R. A. Sullivan and R. J. Davey, CrystEngComm, 2015, 17, 1015–1023.
S. Gracin and Å. C. Rasmuson, Cryst. Growth Des., 2004, 4, 1013–1023.
M. Svard, F. L. Nordstrom, E. M. Hoffmann, B. Aziz and A. C. Rasmuson, Crystengcomm,
2
013, 15, 5020–5031.
3
3
3
1
2
3
R. C. G. Killean, P. Tollin, D. G. Watson, D. W. Young and IUCr, Acta Crystallogr., 1965, 19,
82–483.
4
R. Benali-Cherif, R. Takouachet, E.-E. Bendeif, N. Benali-Cherif and IUCr, Acta Crystallogr.
Sect. C Struct. Chem., 2014, 70, 323–325.
N. Kamali, C. O’Malley, M. F. Mahon, A. Erxleben and P. McArdle, Cryst. Growth Des.,
2
018, 18, 3510–3516.
3
3
4
5
N. Kamali, A. Erxleben and P. McArdle, Cryst. Growth Des., 2016, 16, 2492–2495.
S. Athimoolam, S. Natarajan and IUCr, Acta Crystallogr. Sect. C Cryst. Struct. Commun.,
2
007, 63, o514–o517.
3
3
6
7
T. Yan, K. Wang, D. Duan, X. Tan, B. Liu and B. Zou, RSC Adv., 2014, 4, 15534–15541.
S. A. Moggach, D. R. Allan, S. Parsons and J. E. Warren, J. Appl. Crystallogr., 2008, 41, 249–
2
51.
3
3
4
8
9
0
G. J. Piermarini, S. Block, J. D. Barnett and R. A. Forman, J. Appl. Phys., 1975, 46, 2774–
780.
2
G. J. Piermarini, S. Block, J. D. Barnett and R. a. Forman, J. Appl. Phys., 1975, 46, 2774–
780.
2
G. M. Sheldrick, SADABS, Programs Scaling Absorpt. Correct. Area Detect. Data, 2008,
University of Göttingen: Göttingen, Germany.
4
4
4
1
2
3
J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105–107.
G. M. Sheldrick and IUCr, Acta Crystallogr. Sect. A Found. Adv., 2015, 71, 3–8.
O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann and IUCr, J.
Appl. Crystallogr., 2009, 42, 339–341.
4
4
A. Dawson, D. R. Allan, S. Parsons and M. Ruf, J. Appl. Crystallogr., 2004, 37, 410–416.

Page 17 of 18
CrystEngComm
View Article Online
4
4
4
5
6
7
C. L. Bull, N. P. Funnell, M. G. Tucker, S. Hull, D. J. Francis and W. G. Marshall, High Press.
DOI: 10.1039/C8CE01882K
Res., 2016, 36, 493–511.
P. Taylor, M. Street, L. Wt, D. I. A. Millar, W. G. Marshall, I. D. H. Oswald, C. R. Pulham, I.
A. David, G. William, D. H. Iain and C. R. High-, Crystallogr. Rev., 2010, 16, 115–132.
O. Arnold, J. C. Bilheux, J. M. Borreguero, A. Buts, S. I. Campbell, L. Chapon, M. Doucet, N.
Draper, R. Ferraz Leal, M. A. Gigg, V. E. Lynch, A. Markvardsen, D. J. Mikkelson, R. L.
Mikkelson, R. Miller, K. Palmen, P. Parker, G. Passos, T. G. Perring, P. F. Peterson, S. Ren,
M. A. Reuter, A. T. Savici, J. W. Taylor, R. J. Taylor, R. Tolchenov, W. Zhou and J.
Zikovsky, Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc.
Equip., 2014, 764, 156–166.
4
8
I. B. Hutchison, A. Delori, X. Wang, K. V. Kamenev, A. J. Urquhart and I. D. H. Oswald,
CrystEngComm, 2015, 17, 1778–1782.
4
5
9
0
F. P. A. Fabbiani and C. R. Pulham, Chem. Soc. Rev., 2006, 35, 932.
F. P. A. Fabbiani, D. R. Allan, A. Dawson, W. I. F. David, P. A. McGregor, I. D. H. Oswald,
S. Parsons and C. R. Pulham, Chem. Commun. (Camb)., 2003, 3004–3005.
5
5
1
2
B. Grocholski and R. Jeanloz, J. Chem. Phys., 2005, 123, 204503.
N. T. Morgan, T. C. Frank, R. J. Holmes and E. L. Cussler, Cryst. Growth Des., 2016, 16,
1
404–1408.
5
3
I. D. H. Oswald, I. Chataigner, S. Elphick, F. P. A. Fabbiani, A. R. Lennie, J. Maddaluno, W.
G. Marshall, T. J. Prior, C. R. Pulham and R. I. Smith, CrystEngComm, 2009, 11, 359–366.
5
5
4
5
K. Roszak, A. Katrusiek and A. Katrusiak, Cryst. Growth Des., 2016, 16, 3947–3953.
F. P. A. Fabbiani, G. Buth, D. C. Levendis and A. J. Cruz-Cabeza, Chem. Commun., 2014, 50,
1
817–1819.
5
6
I. D. H. Oswald, A. J. Cruz-Cabeza, C. L. Bull, M. R. Ward and L. Farquhar, Effect of
pressure-transmitting medium on the phase transformation of para-aminobenzoic acid.,
https://data.isis.stfc.ac.uk/doi/investigation/90588565.

CrystEngComm
Page 18 of 18
View Article Online
DOI: 10.1039/C8CE01882K
A new high-pressure recoverable form has been observed in the model system, p-aminobenzoic
acid.