Second order superstructure reflexions in tilted perovskites
Richard · 2025
Read Second order superstructure reflexions in tilted perovskites by Richard (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-16024.txt.
Introduction
Richard Beanland and Robin Sjokvist extend their diffraction rules for tilted perovskites in this 2025 paper. Perovskites with tilted oxygen octahedra show extra faint spots, called superstructure reflexions. A first paper derived simple yes-no rules for those spots.
That work assumed tilts were vanishingly small and kept only the first term of an expansion. Real crystals tilt enough that the second term matters. It creates a fresh set of spots with a half-even-even-odd index pattern. The authors derive yes-no rules for when those second-order spots appear. The rules help readers decode real diffraction patterns from tilted perovskite materials without heavy computation.
Catalog record
| CREATOR | Richard |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-16024
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Second order superstructure reflexions in tilted perovskites Richard Abstract In a previous article [ B+S ] , we derived Boolean conditions for the appearance of superstructure reflexions in diffraction patterns from perovskites with tilted oxygen octahedra, using the structure factor equation. Assuming that the deviation from the untilted prototype perovskite structure was infinitesimally small, we expanded the structure factor as a Taylor series, truncated at the first term. This gave an elegant and simple method giving conditions on the presence or absence of superstructure reflexions. However, in real perovskite materials these distortions are sufficiently large for higher order terms to become significant, giving an additional set of superstructure reflexions. Here, we consider the second term in the expansion and show that it gives rise to ‘second order’ reflexions with indices of the form 1 2 \frac{1}{2} even-even-odd. Boolean conditions for their presence are given. Octahedral tilting rules Beanland \cauthor [a]RobinSjö[email protected] \aff [a]Department of Physics, University of Warwick, Coventry CV4 7AL, UK 1 Introduction In [ B+S ] , we derived Boolean conditions for the appearance of superstructure reflexions in tilted perovskites, taking the displacements of the oxygen atoms from their positions in the prototype unit cell to be \boldsymbol{\delta}^{(j)} . In this framework, the structure factor F_{g} for a reflexion g with Miller indices g_{1}\penalty\ g_{2}\penalty\ g_{3} (or hkl in the convention used in the International Tables for Crystallography, [ IntTables ] ) can be written F_{g}=\sum_{j=1}^{n}f_{g}^{(j)}\exp{\left(2\pi i\textbf{g}\cdot\textbf{r}^{(j)}\right)}\exp{\left(2\pi i\textbf{g}\cdot\boldsymbol{\delta}^{(j)}\right)} (1) where the sum is taken over all j atoms in the unit cell, each having atomic scattering factor f_{g}^{(j)} and fractional coordinates \textbf{r}^{(j)} in the prototype structure. A simple equation giving the type of reflexions produced by a given tilt system was derived by expanding the last term as a series \exp{\left(2\pi i\textbf{g}\cdot\boldsymbol{\delta}^{(j)}\right)}=1+2\pi i\textbf{g}\cdot\boldsymbol{\delta}^{(j)}-4\pi^{2}(\textbf{g}\cdot\boldsymbol{\delta}^{(j)})^{2}+... (2) which can be truncated after the first-order term, if the displacements \boldsymbol{\delta}^{(j)} are taken to be infinitesimally small, which was the premise of our previous article. However, experimental electron diffraction patterns show additional sets of weak superstructure spots that are not given by the previous analysis, indicating that oxygen displacements in tilted perovskites are sufficiently large for second-order terms to be important. These additional spots were termed ‘concert’ reflexions by [ Woodward ] . Here, we derive equations for these sets of weak second-order superstructure reflexions. We find that their type is complementary to first-order superstructure reflexions, and that they are only present in perovskites with oxygen octahedral tilts about more than one axis, at least one of which must be an in-phase tilt system. 2 Calculation From the last term in Eq. 2 and following the approach used in [ B+S ] the structure factor of a second-order superstructure reflexion it produces g is F_{g}=-\sum_{j=1}^{n}f_{g}^{(j)}4\pi^{2}(\textbf{g}\cdot\boldsymbol{\delta}^{(j)})^{2}A^{s^{(j)}_{1}}B^{s^{(j)}_{2}}C^{s^{(j)}_{3}}, (3) where A , B and C take values of \pm 1 or \pm i for even or odd g_{i} respectively, and the position of oxygen atoms is written as \textbf{s}^{(j)}=4\textbf{r}^{(j)} . Importantly, when Eq. 3 is evaluated for a specific tilt system, terms appear that are only non-zero for even, or odd, reflexion indices; for example, (1+A^{2}) is only non-zero when g_{1} is even, while (1-B^{2}) is only non-zero when g_{2} is odd. A unit cell doubled in all three dimensions is used, large enough to be a unit cell for any Glazer octahedral tilting pattern, which contains 24 oxygen atoms and hence gives 24 terms in Eq. 3 . The doubling of the unit cell in the calculation means that, in the reference frame of the prototype perovskite cell, superstructure reflexions have half-order indices and, for example, an equation of the form (1-A^{2})(1-B^{2})(1+C^{2}) indicates that superstructure spots must have the form hkl=\frac{1}{2}odd-odd-even , which is shortened to \frac{1}{2}ooe . It is helpful to split Eq. 3 into two by expanding the \textbf{g}\cdot\boldsymbol{\delta}^{(j)} term, i.e. (\textbf{g}\cdot\boldsymbol{\delta}^{(j)})^{2}=(g^{2}_{1}\delta^{2}_{1}+g^{2}_{2}\delta^{2}_{2}+g^{2}_{3}\delta^{2}_{3})+2(g_{1}g_{2}\delta_{1}\delta_{2}+g_{2}g_{3}\delta_{2}\delta_{3}+g_{3}g_{1}\delta_{3}\delta_{1}), (4) and we will consider the two terms in brackets in this equation separately. First, we have F_{g}=-\sum_{j=1}^{n}f_{g}^{(j)}4\pi^{2}\penalty\ (g^{2}_{1}\delta^{2}_{1}+g^{2}_{2}\delta^{2}_{2}+g^{2}_{3}\delta^{2}_{3})\penalty\ A^{s^{(j)}_{1}}B^{s^{(j)}_{2}}C^{s^{(j)}_{3}}. (5) which is, substituting the 24 oxygen coordinates \textbf{s}^{(j)} \begin{split}\frac{F_{g}}{2\pi^{2}f_{g}}=\penalty\ &-(g^{2}_{1}\delta^{2}_{1}+g^{2}_{2}\delta^{2}_{2})\\ &\penalty\ \penalty\ \penalty\ (AB+AB^{3}+A^{3}B+A^{3}B^{3}+ABC^{2}+AB^{3}C^{2}+A^{3}BC^{2}+A^{3}B^{3}C^{2})\\ -\penalty\ &(g^{2}_{2}\delta^{2}_{2}+g^{2}_{3}\delta^{2}_{3})\\ &\penalty\ \penalty\ \penalty\ (BC+BC^{3}+B^{3}C+B^{3}C^{3}+A^{2}BC+A^{2}BC^{3}+A^{2}B^{3}C+A^{2}B^{3}C^{3})\\ -\penalty\ &(g^{2}_{1}\delta^{2}_{1}+g^{2}_{3}\delta^{2}_{3})\\ &\penalty\ \penalty\ \penalty\ (AC+AC^{3}+A^{3}C+A^{3}C^{3}+AB^{2}C+AB^{2}C^{3}+A^{3}B^{2}C+A^{3}B^{2}C^{3})\end{split} (6) and this reduces to \begin{split}\frac{F_{g}}{2\pi^{2}f_{g}}=&-\left((g^{2}_{1}\delta^{2}_{1}+g^{2}_{2}\delta^{2}_{2})AB+(g^{2}_{2}\delta^{2}_{2}+g^{2}_{3}\delta^{2}_{3})BC+(g^{2}_{1}\delta^{2}_{1}+g^{2}_{3}\delta^{2}_{3})AC\right)\\ &\penalty\ \penalty\ \penalty\ (1+A^{2})(1+B^{2})(1+C^{2})\end{split} (7) which is only non-zero for hkl=\frac{1}{2}eee , i.e. hkl are all integers and are thus coincident with the matrix reflexions. In other words, no new set of superstructure reflexions is produced by this term. The second part of Eq. 4 is F_{g}=-\sum_{j=1}^{n}f_{g}^{(j)}4\pi^{2}\penalty\ (g_{1}g_{2}\delta_{1}\delta_{2}+g_{2}g_{3}\delta_{2}\delta_{3}+g_{3}g_{1}\delta_{3}\delta_{1})\penalty\ A^{s^{(j)}_{1}}B^{s^{(j)}_{2}}C^{s^{(j)}_{3}}. (8) This equation indicates that weak second-order superstructure reflexions are only produced by perovskites with at least two tilt systems, since all terms contain products of two different tilt systems \delta_{i}\delta_{j} , i\neq j . Furthermore, it is found that any system that is comprised only of anti-phase tilts just produces terms containing (1+A^{2})(1+B^{2})(1+C^{2}) which, like the example considered above, are coincident with matrix reflexions. To summarise this point: no second-order superstructure reflexions are produced in the seven Glazer tilt systems a^{0}a^{0}c^{+} , a^{0}a^{0}c^{-} , a^{0}b^{-}b^{-} , a^{0}b^{-}c^{-} , a^{-}a^{-}a^{-} , a^{-}b^{-}b^{-} and a^{-}b^{-}c^{-} . They are found in the seven tilt systems a^{+}a^{+}a^{+} , a^{0}b^{+}b^{+} , a^{+}b^{+}c^{+} , a^{0}b^{+}c^{-} , a^{+}a^{+}c^{-} , a^{+}b^{-}b^{-} and a^{+}b^{-}c^{-} . (Note that, following [ H+S ] , we do not consider tilt systems with a space group that coincides with one of the fourteen listed above, since they produce no additional reflexions.) The results are summarised below in Table 1 , including the first-order superstructure reflexions derived in [ B+S ] . It is also instructive to examine particular examples. One of the tilt systems that satisfies the minimum requirements for the presence of second-order superstructure reflexions is a^{0}b^{+}c^{-} , which has \frac{1}{2}ooo first-order superstructure reflexions that result from the antiphase c^{-} tilts and \frac{1}{2}oeo reflexions from the in-phase b^{+} tilts [ B+S ] . The second-order superstructure reflexions are of the type \frac{1}{2}eoe , i.e. complementary to the first-order \frac{1}{2}oeo reflexions. Similarly, the tilt system a^{+}a^{+}c^{-} has first-order superstructure reflexions of the type \frac{1}{2}eoo , \frac{1}{2}oeo and \frac{1}{2}ooo and second-order superstructure reflexions of the type \frac{1}{2}oee and \frac{1}{2}eoe . That is, in general, any odd-odd-even first-order reflexion produced by in-phase tilts produces a complementary even-even-odd reflexion in the presence of an antiphase tilt . There are three structures with multiple in-phase tilt systems and no anti-phase tilts, i.e. a^{0}b^{+}b^{+} , a^{+}a^{+}a^{+} and a^{+}b^{+}c^{+} . In these structures, no even-even-odd second-order reflexions are produced and for the latter two cases the second order reflexions simply coincide with first-order \frac{1}{2}ooe , \frac{1}{2}oeo and \frac{1}{2}eoo reflections, but with fewer systematic absences. For the a^{0}b^{+}b^{+} tilt system, second-order superstructure reflections take the form \frac{1}{2}eoo while first-order superstructure reflections have indices \frac{1}{2}oeo and \frac{1}{2}ooe . In this case we would therefore expect all three types of odd-odd-even reflexion to be present, but with the second-order reflexions having a significantly weaker intensity. Table 1: Pseudo-cubic second-order superstructure reflections in perovskites with octahedral tilting. Miller indices are given in the form \textbf{g}=hkl . No. Tilt system Space group First order Second order 1 a^{+}a^{+}a^{+} Im\bar{3} \frac{1}{2}ooe,\penalty\ |h|\neq|k|; \frac{1}{2}oeo,\penalty\ |h|\neq|l|; \frac{1}{2}eoo,\penalty\ |k|\neq|l| \frac{1}{2}ooe,\penalty\ h,k\neq 0; \frac{1}{2}oeo,\penalty\ h,l\neq 0; \frac{1}{2}eoo,\penalty\ k,l\neq 0 2 a^{0}b^{+}b^{+} I\frac{4}{m}mm \frac{1}{2}ooe,\penalty\ |h|\neq|k|; \frac{1}{2}oeo,\penalty\ |h|\neq|l| \frac{1}{2}eoo,\penalty\ k,l\neq 0 3 a^{0}a^{0}c^{+} P\frac{4}{m}bm \frac{1}{2}ooe,\penalty\ |h|\neq|k| None 4 a^{0}a^{0}c^{-} I\frac{4}{m}cm \frac{1}{2}ooo,\penalty\ |h|\neq|k| None 5 a^{0}b^{-}b^{-} Imma \frac{1}{2}ooo,\penalty\ !(|h|=|k|=|l|),\penalty\ k\neq l,