Competition between Anion-Deficient Oxide and Oxyhydride Phases during the Topochemical Reduction of LaSrCoRuO6

Binary metal hydrides can act as low-temperature reducing agents for complex oxides in the solid state, facilitating the synthesis of anion-deficient oxide or oxyhydride phases. The reaction of LaSrCoRuO6, with CaH2 in a sealed tube yields the face-centered cubic phase LaSrCoRuO3.2H1.9. The reaction with LiH under similar conditions converts LaSrCoRuO6 to a mixture of tetragonal LaSrCoRuO4.8H1.2 and cubic LaSrCoRuO3.3H2.13. The formation of the LaSrCoRuOxHy oxyhydride phases proceeds directly from the parent oxide, with no evidence for anion-deficient LaSrCoRuO6–x intermediates, in contrast with many other topochemically synthesized transition-metal oxyhydrides. However, the reaction between LaSrCoRuO6 and LiH under flowing argon yields a mixture of LaSrCoRuO5 and the infinite layer phase LaSrCoRuO4. The change to all-oxide products when reactions are performed under flowing argon is attributed to the lower hydrogen partial pressure under these conditions. The implications for the reaction mechanism of these topochemical transformations is discussed along with the role of the hydrogen partial pressure in oxyhydride synthesis. Magnetization measurements indicate the LaSrCoRuOxHy phases exhibit local moments on Co and Ru centers, which are coupled antiferromagnetically. In contrast, LaSrCoRuO4 exhibits ferromagnetic behavior with a Curie temperature above 350 K, which can be rationalized on the basis of superexchange coupling between the Co1+ and Ru2+ centers.


Supporting Information
. Observed, calculated and difference plots from the structural refinement of a P21/n symmetry model against SXRD data collected from LaSrCoRuO6 at room temperature.Table S1.Parameters extracted from the structural refinement of LaSrCoRuO6 against SXRD data.

Thermogravimetric analysis of Sample A.
Figure S2.Thermogravimetric data collected while heating Sample A under flowing oxygen.

Iodometric Titration of Sample A
4. Thermogravimetric analysis of Sample B. Figure S3.Thermogravimetric data collected while heating Sample B under flowing oxygen.

Structural characterization of Sample B.
Figure S6.Observed, calculated and difference plots from the structural refinement of Sample B against SXRD data using model described in Table 2 in the main text.S2.Parameters from the structural and magnetic refinement of LaSrCoRuO4 against NPD data collected at 5 K.   S1.Parameters extracted from the structural refinement of LaSrCoRuO6 against SXRD data.

Thermogravimetric analysis of Sample A.
Figure S2.Thermogravimetric data collected while heating Sample A under flowing oxygen.

Iodometric Titration of Sample A
The hydride content of Sample A was determined via Iodometric titration.As the oxidation states of cobalt and ruthenium are equal or less than 2, the major redox reactions of the oxidative iodometric titration are as follows: 10 ml of a standardised 0.00166 M KIO3 solution was pipetted into a 3-necked flask under flowing argon and an excess of KI (~70 mg) was added, to liberate 4.98 × 10 -5 moles of I2.To this solution 10 ml of aqueous, 2 M HCl was added along with a carefully weighed portion of Sample A (~30 mg).The solution was stirred under argon and Sample A dissolved in the solution, consuming I2 according to the reaction scheme above.The quantity of unreacted I2 was then determined by titration with a standardised solution of Na2S2O4, using starch as an indicator.A constant argon flow was maintained to avoid oxidation of the samples by air throughout the whole titration process.The titration was repeated 5 times to establish the quantity of I2 consumed on dissolution of Sample A, and thus the average oxidation states of the transition metals.Combining these values and the oxygen content obtained from the TGA experiment, the hydride content of Sample A can be determined.

Synthesis of Sample C.
Sample C was prepared by grinding LaSrCoRuO6 with 8 mole equivalents of LiH in an argon filled glove box.The resulting mixture was then poured into an open-ended Pyrex tube that was placed within a silica flow-tube which could be sealed at each end with valves, so that the flow-tube assembly could be inserted into a clam-shell furnace while maintaining an argon atmosphere over the sample mixture.The flow-tube was then purged with argon for 20 minutes before being heated as described below, under a constant flow of argon, as shown schematically in Figure S8.

Magnetic measurements in the presence of elemental Co impurities via the 'ferrosubtraction' method.
Procedure used to measure the magnetization of samples containing elemental cobalt: The magnetization of elemental Co is observed to saturate in applied magnetic fields of more than 2 T. Thus the paramagnetic susceptibility of a bulk sample can be measured in the presence of elemental Co impurities by measuring the gradient of magnetization-field isotherms in applied fields larger than 2 T. As shown in Figure S7.
To this end the magnetization of samples was measured in a series of 5 fields between 3 T and 5 T. The magnetization vs. field data were fitted to a linear function, the gradient of which is the paramagnetic susceptibility of the bulk sample and the intercept is the saturated ferromagnetic moment of the sample.Data points with large errors were excluded from fits.All fits had at least 4 data points.This procedure was repeated at 5 K intervals between 5 K and 300 K to measure the temperature dependent susceptibility of samples.

Figure S4 .
Thermogravimetric data (top) and m/z = 18 mass-spectrum signal (bottom) collected as a function of temperature during the reoxidation of sample B back to LaSrCoRuO6 under oxygen.

Figure S5 .
Thermogravimetric data (top) and m/z = 18, m/z = 2 mass-spectrum signals (bottom) collected as a function of temperature during the reoxidation of sample B back to LaSrCoRuO6 under N2.

Figure S7 .
Electron diffraction data collected from Sample B. 6. Synthesis of Sample C. Figure S8.Schematic diagram of the experimental setup for the synthesis of Sample C. 7. Thermogravimetric analysis of Sample C. Figure S9.Thermogravimetric data collected while heating Sample C under flowing oxygen.8. Magnetic measurements in the presence of elemental Co impurities via the 'ferrosubtraction' method.Figure S10.Magnetisation of Sample A measured as a function of applied field at 300 K. 9. Magnetic Characterization of Sample A. Figure S11.Plot of inverse magnetic susceptibility against temperature for Sample A. Linear fit for T > 150 K, consistent with Curie-Weiss law.10.Magnetic Characterization of Sample B. Figure S12.Plot of inverse magnetic susceptibility against temperature for Sample B. Linear fit for T > 150 K, consistent with Curie-Weiss law.11.Magnetic Characterization of Sample C. Figure S13.Magnetization data collected from Sample C using the ferrosubtraction procedure.

Figure S14 .
Observed calculated and difference plots from the structural and magnetic refinement of LaSrCoRuO4 against NPD data collected at 5 K. Table

Figure S1 .
Figure S1.Observed, calculated and difference plots from the structural refinement of a P21/n symmetry model against SXRD data collected from LaSrCoRuO6 at room temperature.

Figure S5 .
Figure S5.Thermogravimetric data (top) and m/z = 18, m/z = 2 mass-spectrum signals (bottom) collected as a function of temperature during the reoxidation of sample B back to LaSrCoRuO6 under N2 gas.

Figure S6 .
Figure S6.Observed, calculated and difference plots from the structural refinement of Sample B against SXRD data using model described in Table 2 in the main text.Red tick marks indicate peak positions of LaSrCoRuO4.8H1.2, black ticks LaSrCoRuO3.3H2.13.

Figure S8 .
Figure S8.Schematic diagram of the experimental setup for the synthesis of Sample C.

Figure S9 .
Figure S9.Thermogravimetric data collected while heating Sample C under flowing oxygen.

Figure S10 .
Figure S10.Magnetisation of Sample A measured as a function of applied field at 300 K.A linear fit to high-field region (H > 25000 Oe) yields a gradient which is the paramagnetic susceptibility of the sample, and an intercept which is the saturated ferromagnetic moment of the sample.

Figure S11 .
Figure S11.Plot of inverse magnetic susceptibility against temperature for Sample A. Linear fit for T > 150 K, consistent with Curie-Weiss law.

Figure S12 .
Figure S12.Plot of inverse magnetic susceptibility against temperature for Sample B. Linear fit for T > 150 K, consistent with Curie-Weiss law.

Figure S13 .
Figure S13.Magnetization data collected from Sample C using the ferrosubtraction procedure.

Figure S14 .
Figure S14.Observed calculated and difference plots from the structural and magnetic refinement of LaSrCoRuO4 against NPD data collected at 5 K. Black tick marks indicate peak positions for LaSrCoRuO4, red ticks LaSrCoRuO5.