frequency with pressure, with a dv/dp of 4.5 cm-1 GPa-1, over
the 0-30 GPa interval, indicating a progressive strengthening of
the hydrogen bonds between neighbouring molecules. Ionisation
capillary should any degradation products have been formed
during the sealing process. Datasets were collected at 210 K and
also at 150 K (standard data collection temperature); no phase
transitions were observed on cooling to this temperature.
of the molecules above 30 GPa, with the formation of the NO3 H+
-
salt, was suggested due to the apparent absence of all other
bands apart from the vs(NO3 ) of the nitrate ion. The sharp
X-ray diffraction intensities were collected with Mo-Ka
-
˚
radiation (l
=
0.71073 A) on
a
Bruker SMART Apex
discontinuity in dv/dp of the vs(NO2) symmetric elongation mode
at 10 GPa could not be fully rationalised from the experimental
data and, more recently, Me´reau and Mathieu15 have conducted
a theoretical investigation of the high-pressure systematics of
nitric acid to establish a structural mechanism for this behaviour.
They proposed that there is a two-step ionisation of the HNO3
molecules, with 75% of this process completed by 10 GPa and total
ionisation occurring at 30 GPa via a structural phase transition
induced by proton-tunnelling. In the absence of a high-pressure
structure determination of nitric acid, their calculations assumed
that the crystal structure obtained on initial compression into the
solid phase from the liquid at 1.4 GPa and room temperature is the
same as that obtained on freezing below 232 K at ambient pressure,
which was termed the a phase. The calculations also assumed that
the transition to the b phase, at 30 GPa, involves only a migration
of the hydrogen atoms, which link the HNO3 molecules in a dimeric
arrangement in the a phase and shift to form the ribbon-like chains
of the b phase, with minimal effect on the heavy atom positions
and with no change in the crystal symmetry.
Here we report the crystal structure of the high-pressure phase
of nitric acid (here termed phase-II to distinguish it from the
theoretically proposed b phase of Me´reau and Mathieu15), which
is the first crystalline phase obtained by compression of the liquid
to pressures just in excess of 1.0 GPa, along with a redetermination
of the low-temperature, ambient-pressure, phase-I structure (here
we relabel the a phase as phase-I for consistency). Both structures
are found to adopt the monoclinic P21/c symmetry and are
characterized by a herring-bone packing of catemeric hydrogen
bondedmolecular chains which run parallelto the crystallographic
b-axis. The high-pressure phase-II crystal structure has two
molecules in the asymmetric unit making it somewhat simpler than
that of phase-I and although the two structures are superficially
alike there are some significant differences between them.
CCD diffractometer17 equipped with an Oxford Cryosystems
Cryostream-Plus variable-temperature device.18 Absorption cor-
rections were carried out using the multiscan procedure SAD-
ABS.19,20 The structure of nitric acid at 210 K was solved by direct
methods21 and refined by full-matrix least-squares against F2 using
all data.22 Hydrogen atoms were located in difference maps and all
˚
OH distances were refined with a distance restraint of 0.82(5) A.
All non-H atoms were modeled with anisotropic displacement
parameters.
The diffraction intensities collected showed unusual systematic
absences that did not fit those expected for an orthorhombic or
monoclinic lattice. The data showed absences signifying 21-screw
axes along all three directions as well as absences for a c-glide
perpendicular to the b-direction; no accompanying mirrors were
observed for the other two directions. Solving the data by direct
methods in monoclinic P21/c gave a reasonable solution, although
the R-factor remained rather high (see below). The anomaly in
the systematic absences observed for this dataset can be explain by
the position of the molecules within the unit cell. The fractional
coordinates of the central nitrogen atom for each independent
1
4
1
2
molecule have an x, y or z value close to 0,
,
,
34 , which creates
artificial systematic absences that are not related to the presence
of a true symmetry operation. These observations were also made
by Luzzati13 in his earlier study.
To further compound the problem of anomalous systematic
absences, the low temperature form of nitric acid crystallises with
a b-angle close to 90◦ and is therefore susceptible to twinning.
The refinement of the original model that was output by SIR-92
gave a final R-factor of 23.15%. By adding a twin component for
a two-fold rotation around the a-direction this value dropped to
3.54% with a refined twin scale factor of 52%.
Determination of the high-pressure phase-II crystal structure of
nitric acid
Experimental
The sample was prepared by loading nitric acid into a Merrill-
Bassett diamond-anvil cell23 equipped with 600 mm culet diamonds
and a tungsten gasket. The sample was pressurised at room tem-
perature until several crystallites were observed. The temperature
was then increased, using a hand-held hot air blower, so that the
polycrystalline sample was partially remelted, and subsequently
cycled close to this elevated melting temperature in order to reduce
the number of crystallites. A single-crystal was eventually obtained
at a pressure of 1.6 GPa, as determined using the ruby fluorescence
technique.24
Preparation of nitric acid and d1 nitric acid
Nitric acid was prepared by adding sulfuric acid (98 wt%, 10.0 g,
5.4 cm3, 0.10 mol) to potassium nitrate (10.5 g, 0.10 mol) and
distilling off the product at 83-84 ◦C. The resulting nitric acid was
briefly degassed under vacuum to remove excess dissolved nitrogen
oxides. The d1-nitric acid was prepared by the same method using
d2-sulfuric acid.
X-ray diffraction data (Mo-Ka) were collected on a Bruker
kappa CCD diffractometer using a strategy that is a slight
modification of that described by Dawson et al.25 where four
new additional runs were included with c-settings of +90◦ and
-90◦. The data were integrated using the program SAINT26 using
‘dynamic masks’ to avoid integration of regions of the detector
shaded by the body of the pressure cell.25 Absorption corrections
were carried out with the programs SADABS19 and SHADE.27
Redetermination of the ambient-pressure phase-I crystal structure
of nitric acid
A sample was drawn into a capillary (o.d. 0.48 mm) and flame-
sealed. A polycrystalline mass was produced by freezing the liquid
at 210 K. A single crystal was grown using the laser zone refinement
technique of Boese and Nussbaumer.16 Before the crystal was
grown the sample was zone-refined to purify the sample in the
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3736–3743 | 3737
©
Allan
