Neutron diffraction of deuterated tripalmitin and the influence of shear on its crystallisation

Article abstract of DOI:10.1016/j.chemphyslip.2019.02.011

This neutron diffraction study of deuterated tripalmitin has provided further insight into a forensic observation of the crystallisation of lipids under high-shear conditions. To achieve this, an experimental set up was designed to enable simultaneous rheological data from a Couette cell to be recorded with neutron powder diffraction, enabling the influence of shear on the polymorph transformation on cooling to be monitored in real time. Tripalmitin was observed to directly transform from a liquid phase to a β polymorph under the influence of shear. Although the liquid to β transition was not observed to be influenced by shear rate, the degree of crystallinity, qualitatively denoted by an increase in the sharpness of the diffraction peaks, was observed at higher shear rates. Evidence is also presented that the rate of cooling influences the ordering in the β-polymorph produced in zero shear conditions.

Full text of DOI:10.1016/j.chemphyslip.2019.02.011

Accepted Manuscript
Title: Neutron diffraction of deuterated tripalmitin and the
influence of shear on its crystallisation
Authors: B.H. Stuart, H.E. Maynard-Casely, N. Booth, A.E.
Leung, P.S. Thomas
PII:
S0009-3084(18)30192-0
DOI:
Reference:
https://doi.org/10.1016/j.chemphyslip.2019.02.011
CPL 4746
To appear in:
Chemistry and Physics of Lipids
Received date:
Revised date:
Accepted date:
15 October 2018
8 January 2019
25 February 2019
Please cite this article as: Stuart BH, Maynard-Casely HE, Booth N, Leung
AE, Thomas PS, Neutron diffraction of deuterated tripalmitin and the influence
of shear on its crystallisation, Chemistry and Physics of Lipids (2019),
https://doi.org/10.1016/j.chemphyslip.2019.02.011
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1
Neutron diffraction of deuterated tripalmitin and the influence of shear on its crystallisation
1*
2
2
3+
4
B.H. Stuart , H.E. Maynard-Casely , N. Booth , A.E. Leung and P.S. Thomas
1Centre for Forensic Science, University of Technology Sydney, 15 Broadway, NSW 2007, Australia
2Australian Centre for Neutron Scattering, Australian Nuclear Science and Technology Organisation,
Locked Bag 2001, Kirrawee DC, NSW 2232, Australia
3National Deuteration Facility, Australian Nuclear Science and Technology Organisation, Locked Bag
2
001, Kirrawee DC, NSW 2232, Australia
4School of Mathematical and Physical Sciences, University of Technology Sydney, 15 Broadway, NSW
007, Australia
2
*Corresponding author
+Current address: Scientific Activities Division, European Spallation Source ERIC, Lund 224 84 Sweden.
Highlights




Polymorphic structures adopted by deuterated tripalmitin on cooling have been characterised
using neutron diffraction.
For shear studies, simultaneous rheological data from a Couette cell was collected with
neutron diffraction data.
Tripalmitin was observed to directly transform from a liquid phase to a  polymorph under the
influence of shear.
Temperature at which  transition occurs is independent of shear rate, but shear forces
influence crystalline order.
Abstract
This neutron diffraction study of deuterated tripalmitin has provided further insight into a forensic
observation of the crystallisation of lipids under high-shear conditions. To achieve this, an
experimental set up was designed to enable simultaneous rheological data from a Couette cell to be
recorded with neutron powder diffraction, enabling the influence of shear on the polymorph
transformation on cooling to be monitored in real time. Tripalmitin was observed to directly

2
transform from a liquid phase to a  polymorph under the influence of shear. Although the liquid to β
transition was not observed to be influenced by shear rate, the degree of crystallinity, qualitatively
denoted by an increase in the sharpness of the diffraction peaks, was observed at higher shear rates.
Evidence is also presented that the rate of cooling influences the ordering in the -polymorph
produced in zero shear conditions.
Keywords:
tripalmitin; neutron diffraction; shear; crystallisation
1. Introduction
The appearance of unusual crystalline structures observed during a forensic post-mortem
examination of a motor vehicle accident victim was the topic of an earlier study [1]. The crystal
structures were found to be comprised of triacylglycerols (TAGs), the dominant lipid structure found
in human adipose tissue, and the morphology of the crystalline material was proposed to be a result
of shear stresses. However, the complexity of the TAG composition of adipose tissue made it difficult
to elucidate the precise mechanism by which the crystalline structures were formed. In order to gain
a better understanding of the influence of shear on adipose tissue, the shear behaviour of a model
TAG, tripalmitin, has been investigated.
Tripalmitin is a monoacid TAG derived from palmitic acid. Tripalmitin is able to crystallise to three
main polymorphs: α, β’ and β. Each polymorph demonstrates specific hydrocarbon chain packing and
thermal stability [2-4]. The  phase adopts hexagonal chain packing, while the β’ phase exhibits
orthorhombic packing. The most stable  polymorph forms triclinic chain packing. The formation of
the various polymorphs depends upon the crystallisation conditions.
A number of studies have employed x-ray diffraction (XRD) methods to investigate the formation of
polymorphism of tripalmitin [5-8]. XRD is a popular tool for the study of TAG polymorphism as
characteristic diffraction patterns can be identified for the different packing of each polymorph. Also,

3
the influence of shear on the crystallisation of fats has been widely studied due to its industrial
importance, particularly in the food industry. The application of shear is known to influence the
nucleation and crystallisation processes of lipids [10,11]. In order to understand shear induced effects
in lipids, experiments involving the combination of XRD with a rheometer have been developed to
provide a more direct examination of the nature of the crystallisation process under shear [10, 12-14].
These studies have demonstrated that a successful means of controlling shear has been with the use
of a Couette cell, which consists of two concentric cylinders. The sample is placed between the
cylinders and the inner cylinder is then rotated to generate the shear required.
The use of neutron diffraction for the study of TAG polymorphism has been limited due to the
requirement of access to specialised instrumentation [9]. There is, however, a number of potential
benefits to undertaking neutron diffraction measurements of tripalmitin. Neutrons scatter from the
nucleus of each atom, an effective point source, which means that neutron diffraction is not impeded
by the form factor inherent in XRD and a wider angle of data to be more easily accessed. Additionally,
compared to X-rays, neutrons are highly penetrating, which allows for in situ data to be collected
from complex sample environments, such as shear cell environments. However, hydrogen (a large
component in tripalmitin) is an incoherent scatterer of neutrons, which for experimental practicality
leads to low signal-to-noise ratios of the measured diffraction. A common path to overcoming this
limitation is to substitute deuterium into the structure, and as such, the deuterated tripalmitin
structure has been investigated for this study.
In this study, neutron diffraction has been used to monitor the crystallisation processes of deuterated
tripalmitin during cooling and in shear stress experiments controlled with a Couette cell. The
tripalmitin was examined using a high intensity powder diffractometer in conjunction with a
rheometer to shear the lipid sample while collecting diffraction data.
2. Materials and methods
2.1 Materials
Reagents were used as received from Sigma-Aldrich (USA). Solvents were used as received from
Sigma-Aldrich or were purified by literature methods. The solvents for nuclear magnetic resonance

4
(
NMR) spectroscopy were used as received from Cambridge Isotope Laboratories Inc. (USA).
Deuterium oxide (D O) (99.8%) was purchased from Atomic Energy of Canada Limited (Canada).
2
2.2 NMR spectroscopy
NMR spectroscopy was used to confirm the composition of the synthetic deuterated tripalmitin. 1H
2
13
NMR (400 MHz), H NMR (61.4 MHz) and C NMR (100 MHz) spectra were recorded using a Bruker
00 MHz spectrometer at 298 K. Chemical shifts were referenced to the residual signal of the solvent.
2H NMR spectroscopy was performed using the probe’s lock channel for direct observation.
4
2.3 Mass spectrometry
Mass spectrometry was used to determine the extent of deuteration in the tripalmitin synthesised.
Electrospray ionisation mass spectra were recorded using a 4000 QTrap AB SCIEX mass spectrometer.
The overall percentage deuteration of the molecules was calculated using enhanced resolution mass
spectrometry (ER-MS). The isotope distribution of the different isotopologues was determined by
analysing the area under each MS peak, which corresponds to a defined number of deuterium atoms.
The contribution of the carbon-13 (natural abundance) to the value of the area under each [X+1] MS
signal was subtracted based on the relative amount found in the unlabelled version [15].
2.4 Synthesis of deuterated tripalmitin
Palmitic acid-d31 (Figure 1) was first produced via metal-catalysed hydrothermal H/D exchange
2
according to a method reported in the literature for azelaic acid-d14 and nonanoic acid-d17 [16]. H
NMR (61.4 MHz, CDCl
3
) δ 0.83 (s), 1.20 (s), 1.59 (s), 2.31 (s). MS (ESI) m/z calculated for C16
31 2
D O

[
MH] as 286.4; found: 286.5 (100%). Deuteration: 97.9% by MS: isotope distribution: d27 1.3%, d28
2
.9%, d29 10.0%, d30 30.4%, d31 55.5%.
Palmitic acid-d31 (Figure 1) (6.08 g, 21.1 mmol, 97.5% D by MS) and glycerol (588 mg, 6.38 mmol) were
dissolved in dry dichloromethane (DCM) (75 mL). 4-Dimethylaminopyridine (DMAP) (826 mg, 6.76
mmol) was added. When all of the reagents had dissolved, the vessel was covered with foil and a 1.0
M solution of N,N'-dicyclohexylcarbodiimide (DCC) in DCM was added in aliquots: 10 mL (10 mmol)
initially; after 2 h another 5 mL (5 mmol); after an additional 18 h another 3.5 mL (3.5 mmol); after an

5
additional 2 h another 3 mL (3 mmol) (total: 21.5 mL, 21.5 mmol). After 2 h, additional palmitic acid-
31 (562 mg, 1.85 mmol) and DCC (1.0 M in DCM, 5 mL, 5 mmol) were dissolved in dry DCM (10 mL)
d
and added to the reaction mixture, which was allowed to stir at room temperature overnight. The
mixture was filtered and concentrated under reduced pressure, then dissolved in 5% ethyl acetate
(
EtOAc) in petroleum ether and filtered again. The crude product was purified firstly by manual flash
column chromatography (5% EtOAc in petroleum ether), then the fractions containing the product
were combined, concentrated and the resulting solid recrystallised from EtOH to afford a white solid
1
(
1.88 g, 33%). H NMR (400 MHz, CDCl
3
) δ 0.82 (s, residual), 1.19-1.25 (complex, residual), 1.54 (s,
2
residual), 2.27 (s, residual), 4.22 (AB quartet, J = 11.6, 6.0, 4.3 Hz, 4 H), 5.26 (m, 1 H). H NMR (61.4
13
MHz, CDCl
3
) δ 0.81 (s), 1.12 (s), 1.56 (s), 2.27 (s). C NMR (100 MHz, CDCl ) δ 13.1 (m), 21.5 (m), 23.9
3
(m), 28.5 (m), 30.7 (m), 33.6 (m), 62.2 (s), 69.0 (s), 173.1 (s), 173.5 (s). MS (ESI+) m/z calculated for
+
C
d
51
H
5
D
93NaO [M+Na] as 923.3; found: 922.5 (100%). Deuteration: 92.8% by MS: isotope distribution:
6
87 0.7%, d88 2.2%, d89 5.7%, d90 13.2%, d91 24.7%, d92 29.0%, d93 24.5%.
2.5. Neutron diffraction
Neutron diffraction data was collected using the WOMBAT high intensity powder diffractometer at
the Australian Centre for Neutron Scattering, Australian Nuclear Science and Technology Organisation
(ANSTO) [17]. The instrument was set up with a focusing germanium monochromator optimised to
the [113] reflection so the wavelength of incoming neutrons was λ = 2.41 Å. Before the shear
experiments were conducted, cooling experiments were undertaken within an orange cryostat
mounted onto the WOMBAT instrument [18] with the sample placed within a 70 mm length, 6mm
diameter vanadium can with thermocouples and heaters placed above and below the sample for
temperature control. Three runs were conducted; two with the cooling controlled at 0.1 and 0.5 K
-
1
min and one ‘rapid’ cooling run where the heating was switched off, the rate of this cooling was
-
1
determined to be 1.8 K min on average. Collection times for each pattern during these cooling runs
were 5 min, 1 min and 1 min, respectively. After these experiments were completed, the cryofurnace
was removed and replaced with the rheometer. The deuterated tripalmitin specimens were then
subjected to shear stress via the use of an Anton Paar MCR500 rheometer with an ANSTO designed
silica Couette cell suitable for neutron scattering experiments. The layout of the cell within the
instrument is illustrated in Figure 2 [19], positioning of the apparatus was optimised with copper foil
placed within the cell. The inner cylinder had a diameter of 48 mm and the gap between the inner and

6
outer cylinders was 0.5 mm. The temperature was controlled by an external water jacket shaped so as
not to obstruct the incoming neutron beam or the neutron detector. Two experimental runs were
-
1
completed with this apparatus, both with the cooling rate controlled to 0.5 K min , one with shear
applied at 10 Hz and one with shear applied at 1000 Hz. For both of these experimental runs patterns
were each 1 min.
2
.6. Data analysis
Each diffraction pattern was collected on WOMBAT’s continuous detector bank, located between 20-
60° degrees around the sample position. The detector is 200 mm high and each collected pattern
1
was integrated vertically to produce a one-dimensional plot. Each temperature’s pattern was then
stacked to produce thermodiffractograms presented in subsequent figures. Fitting of the diffraction
data was undertaken using the TOPAS suite of programs [20]. Counting times of the diffraction data
were varied during the course of the experiment and where data is compared in this publication,
counts have been normalised to the incoming beam monitor before plotting.
3. Results and discussion
3
.1 Neutron diffraction of deuterated tripalmitin
The formation of polymorphs by deuterated tripalmitin without shear was initially investigated by
examining the structures formed under differing cooling rates within the cryofurnace set up. A cooling
-
1
rate of 1.8 K min from 350 to 280 K was employed to produce an α phase in the tripalmitin. A slower
-
1
rate of 0.1 K min from 360 to 290 K was used to produce a β phase. Figure 3 shows the neutron
diffraction data as a function of temperature collected for each of these rates, as well as the data for
-
1
0
.5 K min , the cooling rate chosen for the shear experiments. The corresponding diffraction patterns
at a temperature of 300 K for each cooling rate are also shown in Figure 3.
Table 1 lists the 2 angles and the calculated d-spacing values for the peaks shown in the 300 K
-
1
diffraction patterns produced at cooling rates of 0.1 and 1.8 K min and may be attributed either the
-
1

or  phases. For the faster cooling rate of 1.8 K min , a strong peak at 33.6° (4.17 Å) is indicative
-
1
that the sample had formed into the  phase [6,7]. The slower cooling rate of 0.1 K min resulted in a

7
number of diffraction peaks, notably at d-spacings of 4.54, 3.81 and 3.65 Å. Based on Kellen et al.
reporting of three strong peaks in the XRD pattern at d-spacings of 4.60, 3.85 and 3.70 Å [7], it can be
deduced that the sample had formed the  phase. The shift of ~0.05 Å in the peak positions can be
attributed to the differences in the sample positioning relative to the detectors between the XRD and
neutron experiments. Table 1 also lists multiple peaks above 60° corresponding to d-spacings in the
range 1.5-2.5 Å, that had not been noted by Kellen et al. These are higher angle peaks that are
difficult to observe in XRD, such as the peak at 68.4° (2.14 Å), are resolved in the neutron diffraction
data because of the lack of form factor.
The quality of the data is such that a refinement of the data was able to be undertaken against a
model of the β tripalmitin crystal structure proposed by van Langevelde et al. [21]. This is presented
̅
in Figure 4, and qualitatively shows a good fit to the triclinic 푃1 unit cell with lattice parameters of a =
.42 Å, b = 11.98 Å, c = 40.31 Å, α = 84.28°, β = 87.35° and γ = 79.65°, resulting in a volume of 2561 Å3.
5
Aside from the hydrogen atoms being replaced by deuterium atoms in the model, the refined
parameters during the refinement in addition to the lattice parameters were scale factor, peak shape
and an 8-term spherical harmonic preferred orientation model. This serves as excellent confirmation
of the model determined by van Langevelde et al. from single crystal x-ray diffraction data.
-
1
The intermediate cooling rate used (0.5 K min ) also shows a  phase in the diffraction data collected
at 300 K (Figure 3). At 335 K, the pattern immediately changes to that of a  phase, indicating that this
is the temperature at which the liquid   transition occurs in the absence of shear. The 300 K
-
1
diffraction pattern for the 0.5 K min cooling rate shows slightly broader peaks than those observed
-
1
in the pattern collected for the 0.1 K min cooling rate, indicating differences in crystal size or
perfection due to different cooling rates [7].
3
.2 The influence of shear on the crystallisation of deuterated tripalmitin
Figure 5 illustrates the background contribution of the shear cell, which is dominated by the silica
from the cell and with some effects of air scattering at higher angles. The air scattering effects are
removed when the cell is filled. This shows it is challenging to identify the presence of an α phase

8
from within the cell other than in a possible broadening of the peak at ~35°, which is revealed in more
detail with the subtraction shown in Figure 5.
Figure 6 illustrates the diffraction data collected during cooling of the deuterated tripalmitin at a rate
-
1
-1
of 0.5 K min and shearing at 10 s . As the tripalmitin is cooled from 350 to 323 K, the viscosity values
early in the experiment are low and correspond to a liquid phase. The patterns collected in this range
show broad peaks near 35° and 68°. Although these peaks correspond to those predicted for the -
phase of tripalmitin, an inspection of the background contribution to the pattern by the shear cell
reveals that these two peaks are from the cell itself.
As was observed in the diffraction data collected at zero shear, there is a clear transition to the 
-
1
phase occurring as the tripalmitin cools under shear at 10 s (Figure 5). Although the background
contribution of the shear cell is also overlapping with the -phase peaks, the sharper contributions
made by this polymorph is more readily discernible in the diffraction data shown in Figure 5. The
temperature at which the transition to the  phase appears to occur is lowered to 323 K compared to
the 335 K liquid   transition temperature observed in the cryofurnace measurements where no
shear was applied. Viscosity measured during the experiment showed a gradual increase as the
sample was cooled to 323 K, where the transition was observed in the diffraction data. The instigation
of the transition also coincides with a very rapid increase in viscosity as cooling proceeds [22,23].
-
1
-1
Similar findings to those observed at 10 s were obtained for the data collected at a shear of 1000 s .
-
1
The diffraction data collected during cooling of the deuterated tripalmitin at a rate of 0.5 K min and
-
1
shearing at 1000 s is shown in Figure 7. A comparison of the peaks associated with the  phase
-
1
observed in the diffraction data collected below 325 K for the 10 and 1000 s shear rates indicates
-
1
that sharper peaks are produced under 1000 s (Figures 5 and 7). The narrower peaks at the higher
shear rate may be associated with a more ordered structure produced when higher shear forces are
applied and are likely to be due to an increased degree of orientation of the TAG molecules at higher
shear rates. Note that the instrument configuration used in this study is not set up to identify
orientation effects.
-
1
The transition from a liquid phase to the  phase is observed to occur at 325 K at 1000 s , again
below that observed in the cryofurnace measurements with no shear, and solidification is also

9
-
1
accompanied by a sharp increase in the viscosity. The increase in shear from 10 to 1000 s appears to
have no major influence on the crystallisation temperature of tripalmitin indicating that the reduction
in the transition temperature is most likely to be due to differences in sample geometry rather than
extraneous effects such as shear induced heating.
4. Conclusions
Neutron diffraction has been successfully applied to characterise the different polymorphic structures
adopted by deuterated tripalmitin as it is cooled from the liquid state, and has confirmed previous
single crystal determination of the triclinic β tripalmitin crystal structure. Through the use of a shear
cell in combination with the high intensity powder diffractometer it was possible to monitor the effect
of shear stress on tripalmitin. A clear transition to a -phase for this TAG was observed. The
temperature at which the transition occurs is independent of shear rate used, but shear forces play a
role in the degree of crystalline order when the -polymorph is formed; a higher degree of order at
higher shear rates indicates shear rate influences orientation. A limitation to the interpretation of any
-phase of tripalmitin was identified in this study due to the nature of the shear cell. However, there
is scope to modify the materials employed and develop methods for the removal of the background
contribution in further studies. This will enable the influence of shear to be further investigated for
TAG systems of interest in a range of applications.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
This work was supported by ANSTO’s Australian Centre for Neutron Scattering Neutron Beam
Instrument Access funding [grant numbers 3549 and 3640] and National Deuteration Facility funding
[
grant number 3640].
References

1
0
1. Stuart, B.H., Notter, S.J., Langlois, N., Maynard, P., Ray, A., Berkahn, M. Characterisation of the
triacylglycerol crystal formation in adipose tissue during a vehicle collision, 2007. J. Forensic Sci. 52,
9
2
3
38-942.
. Sato, K., 2001. Crystallisation behaviour of fats and lipids – a review. Chem. Eng. Sci. 56, 2255-2265.
. Sato, K., 2001. Molecular aspects in fat polymorphism. In: Widlak, N., Hartel, R.W., Narine, S.S.
(Eds.), Crystallisation and Solidification Properties of Lipids. AOCS Press, Champaign, pp. 1-15.
4. Lawler, P.J., Dimick, P.S., 2002. Crystallisation and polymorphism of fats. In: Akoh, C.C. (Ed.), Food
Lipids: Chemistry, Nutrition and Biotechnology. Marcel Dekker, New York, pp 275-300.
5. Sato, K., Kuroda, T., 1987. Kinetics of melt crystallisation and transformation of tripalmitin
polymorphs. J. Amer. Oil Chem. Soc. 64, 124–127.
6. Kellens, M., Meeussen, W., Reynaers, H., 1990. Crystallisation and phase transition studies of
tripalmitin. Chem. Phys. Lipids. 55, 163–178.
7. Kellens, M., Meeussen, W., Riekel, C., Reynaers, H., 1990. Time resolved X-ray diffraction studies of
the polymorphic behaviour of tripalmitin using synchrotron radiation. Chem. Phys. Lipids. 52, 79-98.
8. Hongisto, V., Lehto, V.P., Laine, E., 1996. X-ray diffraction and microcalorimetry study of the α→β
transformation of tripalmitin. Thermochim. Acta 276, 229-242.
9. Cebula, D.J., McClements, D.J., Povey, M.J.W., Smith, P. R., 1992. Neutron diffraction studies of
liquid and crystalline trilaurin. J. Am. Oil. Chem. Soc. 69, 130-136.
10. Mazzanti, G., Marangoni, A.G., Guthrie, S.E., Idziak S.H.J., Sirota, E.B., 2005. Crystallisation of bulk
fats under shear. In: Dutcher, J.R., Marangoni A.G. (Eds.), Soft Materials – Structure and Dynamics,
Marcel Dekker, New York, pp. 279–299.

1
1
11. Tran, T., Rousseau, D., 2016. Influence of shear on fat crystallisation. Food Res. Int. 81, 157-162.
12. Mazzanti, G., Guthrie, S.E., Sirota, E.B., Marangoni, A.G., Idziak, S.H.J., 2003. Orientation and phase
transitions of fat crystals under shear. Cryst. Growth Des. 3, 721-725.
13. Mazzanti, G., Marangoni, A.G., Idziak, S.H.J., 2009. Synchrotron study on crystallisation kinetics of
milk fat under shear flow. Food Res. Int. 42, 682-694.
14. Mazzanti, G., Marangoni, A.G., Idziak S.H.J., 2005. Modeling phase transitions during the
crystallisation of a multicomponent fat under shear. Phys. Rev. E 71, 041607.
15. Duffin, K.L., Henion, J.D., Shieh, J.J., 1991. Electrospray and tandem mass spectrometric
characterisation of acylglycerol mixtures that are dissolved in nonpolar solvents. Anal. Chem. 63,
1781-1788.
16. Darwish, T.A., Luks, E., Moraes, G., Yepuri, N.R., Holden P.J., James, M., 2013. Synthesis of
deuterated [D32] oleic acid and its phospholipid derivative [D64] dioleoyl-sn-glycero-3-
phosphocholine. J. Label. Compd. Radiopharm. 56, 520-529.
17. Studer, A.J., Hagen, M.E., Noakes, T.J., 2006. Wombat: The high-intensity powder diffractometer
at the OPAL reactor. Physica B 385-386, 1013-1015.
18. Brochier, D., 1977. Institut Laue-Langevin Technical Report 77/74.
19. Booth, N., Davidson, G., Imperia, P., Lee, S., Stuart, B.H., Thomas, P.S., Komatsu, K., Yamane, R.,
Prescott, S., Maynard-Casely, H.E., Nelson, A., Rule, K., 2017. Three impossible things before lunch -
the task of a sample environment specialist. J. Neutron Res. 19, 49-56.
20. Coelho, A., 2008. TOPAS 4.1. Bruker AXS, Wisconsin, USA.

1
2
21. van Langevelde, A. J., van Malssen, K., Hollander, F., Peschar, R., Schenk, H., 1999. Structure of
mono-acid even-numbered β-triacylglycerols. Acta Cryst. B55, 114-122.
22. Sonwai, S., Mackley, M.R., 2006. The effect of shear on the crystallisation of cocoa butter. J. Amer.
Oil Chem. Soc. 83, 583-596.
2
3. de Graef, V., Goderis, B., Van Puyvelde, P., Foubert, I., Dewettinck K., 2008. Development of a
rheological method to characterise palm oil crystallising under shear. Eur. J. Lipid Sci. Technol. 110,
21-529.
5

1
3
Figure 1. Structures of palmitic acid-d31 and tripalmitin-d93
.
palmitic acid-d31
tripalmitin-d93

1
4
Figure 2. Geometry of neutron instrument and shear cell. Solid arrows indicate the path of the
incident neutron beam, and dashed arrows indicators of paths of the diffracted neutron beams. The
dashed circle indicates the 50 mm diameter from the central sample position that is accepted by the
radial collimator mounted before the detector.
silica Couette
cell with direction
of shear
beam slits

1
5
Figure 3. Neutron diffraction data collected at a wavelength of 2.41 Å for different cooling rates within
the cryofurnace (not from within the rheometer) and corresponding diffraction patterns collected at
-
1
-1
3
00K. Collection times for each pattern in the 0.1 K min data was 5 min, for 0.5 K min was 1 min
-
1
and for the 1.8 K min was also 1 min. Colours indicate the intensity of the diffraction pattern as per
the scale bar.

1
6
Figure 4. Refinement of the neutron diffraction data collected at a wavelength of 2.41 Å during the
-
1
0
.1 K min cooling run at 300 K (I(obs)) versus the model of β phase of tripalmitin determined by
Langevelde et al. (I(calc)) [21]. The variables of this model described in the text produced a fit RWP of
.2 % and GOF of 6.1 %.
6

1
7
Figure 5. Comparison of the diffraction patterns of the empty shear cell (black line) with that of the
cell including α-tripalmitin (red line) collected at a wavelength of 2.41 Å and at 300 K. Data have been
normalised to incident beam monitor counts. The blue line is the subtraction of the empty shear cell
pattern from that including the α-tripalmitin, which at low angles shows the characteristic features of
the α-tripalmitin diffraction pattern.

1
8
-
1
-
Figure 6. Left hand image - diffraction data collected during cooling at 0.5 K min and shearing at 10 s
1.
Colours indicate the intensity of the diffraction pattern (left axis) as per the scale and black squares
chart measured viscosity (right axis). Right hand image - comparison between diffraction data
collected at 350 and 300 K. Each pattern was collected for 1 min.

1
9
-
1
Figure 7. Left hand image - diffraction data collected during cooling at 0.5 K min and shearing at 1000
-
1
s . Colours indicate the intensity of the diffraction pattern (left axis) as per the scale and black
squares chart measured viscosity (right axis). Right hand image - comparison between diffraction
data collected at 350 and 300 K. Each pattern was collected for 1 min.

2
0
Table 1. Diffraction parameters for deuterated tripalmitin polymorphs obtained using neutron
diffraction.
2 (°) d-spacing (Å)
polymorph
3
3
3
4
6
6
6
7
8
9
9
0.8
3.6
6.9
1.5
4.1
8.4
9.5
6.5
0.4
1.6
9.3
4.54
4.17
3.81
3.40
2.27
2.14
2.11
1.95
1.87
1.68
1.58










