Date Published: June 01, 2020
Publisher: International Union of Crystallography
Author(s): Damian Mroz, Ruimin Wang, Ulli Englert, Richard Dronskowski.
Anisotropic displacement parameters for the isomorphous compounds 1-(halomethyl)-3-nitrobenzene (halo = chloro and bromo) were calculated from first principles and determined by X-ray diffraction experiments. Unexpectedly, the experiment for the bromo compound proved more challenging than theory.
Careful diffraction experiments on crystals of reasonable quality provide reliable intensity data from which atomic positions and anisotropic displacement parameters (ADPs) can be derived almost routinely. The alternative route towards ADPs, namely, their calculation from first principles, has made good progress (George et al., 2015a ▸,b ▸, 2016 ▸, 2017 ▸; Deringer et al., 2014 ▸, 2016 ▸, 2017 ▸; Baima et al., 2016 ▸; Lane et al., 2012 ▸; Madsen et al., 2013 ▸; Pozzi et al., 2013 ▸; Dittrich et al., 2012 ▸).
Compounds 1 and 2 were obtained from Sigma–Aldrich and recrystallized from methanol by slow evaporation at room temperature. The elevated vapour pressure of these compounds does not permit their storage for periods longer than a few weeks. An Oxford Cryostream device was used to maintain a constant data-collection temperature of 100 K.
Our initial diffraction experiments were conducted with in-house equipment at 100 K. Mo Kα radiation from a microfocus tube was used, and data collections extended to a resolution of 0.62 Å (λmax = 35°). We will refer to these data sets as 1a and 2a. A first comparison between the experimental and energy-minimized crystal structures in terms of lattice parameters and overall residuals of mean Cartesian displacements (RMS) is provided in Table 2 ▸ and documents a good match.
We set out to benchmark ADPs based on dispersion-corrected DFT calculations on the harmonic approximation, and it turned out that our in-house experiment, despite elevated redundancy and resolution, was not really able to do so. An alternative experiment at a synchrotron beamline at the same temperature but on a smaller crystal and with a short wavelength gave results in better agreement with theory. We do not dwell on compiling all possible sources of error but rather draw three optimistic conclusions: (i) the quality of theoretically calculated ADPs may challenge that of standard experiments, (ii) the directionality of the ADPs based on the intensity data of our in-house diffractometer match that obtained at the synchrotron beamline even if the amplitudes do not agree and (iii) for the (many!) crystal structures with minor absorption effects only, ADPs from good in-house data match those obtained at the synchrotron beamline; compound 1, with its unexceptional absorption properties, provides a good example for that statement. In our future work, we will attempt to gain insight into the various sources of experimental error. The calculation of absorption-affected data by analytical methods, followed by their treatment with a multi-scan correction program might be a suitable approach.