# Fileset
[FAO_2022_8acceptedversion_forMDR.pdf](https://mdr.nims.go.jp/filesets/b6eb90fe-4c49-47dd-89a7-ccea7d075d32/download)
## Creator
[Takayuki Nakane](https://orcid.org/0000-0003-0282-169X), Satoshi Ishii, [Tetsuo Uchikoshi](https://orcid.org/0000-0003-3847-4781), [Takashi Naka](https://orcid.org/0000-0002-0645-6952)
## Rights
This is the peer reviewed version of the following article: Nakane T, Ishii S, Uchikoshi T, Naka T. Influence of fabrication conditions on the structural characteristics and the magnetic properties of FeAl2O4. J Am Ceram Soc. 2023; 106: 2317–2325, which has been published in final form at https://doi.org/10.1111/jace.18915. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article may not be enhanced, enriched or otherwise transformed into a derivative work, without express permission from Wiley or by statutory rights under applicable legislation. Copyright notices must not be removed, obscured or modified. The article must be linked to Wiley’s version of record on Wiley Online Library and any embedding, framing or otherwise making available the article or pages thereof by third parties from platforms, services and websites other than Wiley Online Library must be prohibited.[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)
## Other metadata
[Influence of fabrication conditions on the structural characteristics and the magnetic properties of FeAl2O4](https://mdr.nims.go.jp/datasets/6fc9f4ee-3c8d-420a-b8a7-26e8029a0466)
## Fulltext
Influence of Fabrication Conditions on the Structural Characteristics and the Magnetic Properties of FeAl2O4 Takayuki Nakane1, Satoshi Ishii2, Tetsuo Uchikoshi1 and Takashi Naka1 1. National Institute for Materials Science, Research Center for Functional Materials, 1-2-1, Sengen, Tsukuba, Ibaraki 305-0047, Japan 2. Tokyo Denki University, Department of Physics, Hatoyama, Hiki-gun, 350-0394 Saitama, Japan This is the peer reviewed version of the following article: Journal of the American Ceramics Society, which has been published in final form at https://doi.org/10.1111/jace.18915. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article may not be enhanced, enriched or otherwise transformed into a derivative work, without express permission from Wiley or by statutory rights under applicable legislation. Copyright notices must not be removed, obscured or modified. The article must be linked to Wiley’s version of record on Wiley Online Library and any embedding, framing or otherwise making available the article or pages thereof by third parties from platforms, services and websites other than Wiley Online Library must be prohibited. https://doi.org/10.1111/jace.18915Abstract We report a systematic evaluation of the magnetic properties of FeAl2O4 focusing on the relationship between the fabrication conditions and its structural characteristics, in order to improve ceramics processing in applications of this material. For this purpose, the most important factor to control is the inversion parameter, expressed as y in the (Fe1−yAly)(Al1−y/2Fey/2)2O4 composition, which is relatively high for the spinel aluminate of a transition metal. The magnetic properties of these samples all shows the spin glass phenomenon at low temperatures, and the cusp temperature depends systematically on this y value. This means that the evaluation of these magnetic properties will be an effective way to predict some characteristics of product FeAl2O4. Additionally, this study found an anomaly in the structural and magnetic characteristics of FeAl2O4 fabricated at a low temperature. This is thought to originate in a tiny and a small amount of impurity. It will be key for discussing the quality of chemically synthesized FeAl2O4, which is typically produced at low temperatures. Introduction There is no doubt that the crystal structure plays an important role in the investigation of functional ceramics. Following this concept, the spinel structure, with its isotropic cubic symmetry and flexible tuning of its chemical composition, is particularly attractive. The compositional formula of a spinel oxide is expressed as AB2O4, where A and B indicate di- and tri-valent metal cations, respectively. The cations that exist at each type of site (i.e. A2+ and B3+) respectively form characteristic tetrahedral and octahedral structures with the coordinating anion (i.e. O2-). The various attractive-properties found in spinel oxides mainly originate in the electronic states of these two metal cations. Consequently, site-exchange phenomenon between these two metal cations often becomes the substantive issue for discussing the properties of spinel oxides.1,2 This phenomenon is so prevalent that the degree of site-exchange is often defined through inversion parameter, y, meaning the degree of distribution of A2+ in the B site, i.e. y of a (A2+1−yB3+y)(B3+2−yA2+y)O4 composition. Differences in y values sometimes result in the markedly different properties of the spinel material, even though the chemical composition is often expressed simply as AB2O4 in many case3. This point is often the hidden reason for contradictions between experimental results reported by different groups. This y value should not be thought of as a parameter that can be neglected in the study of all spinel materials. Application of spinel materials have been widely investigated in a wide range of scientific fields including magnetism4,5, catalytic chemistry6-9, optical science10-12, pigments13 and so on. In particular, catalytic applications are meaningful and promising challenges. However, some issues can be imagined at the nascent stage for the roadmap to fully launch a catalytic application studies of the spinel materials., For instance, catalytic materials commonly requires effective chemical synthesis procedures for industrial application, since they should be produced at low cost. In addition, the processes should be applicable for the mass or large-scale production. For example, mass production is required for nanoparticle forms and large-scale production is required for the porous bulk forms used as a heterogeneous catalyst. The chemical synthesis of ceramic materials is usually performed at relatively low temperatures under 200 oC, and this tends to increase the inversion parameter in case of spinel materials.14 Increasing the inversion parameter is generally thought of as a negative factor for the physical properties of the spinel materials. However, it is not a simple in case of the catalytic applications. Appropriate amount of the site exchange phenomenon is considered as one of the possibility of enhancing the catalytic performance in case of transition metal aluminates expressed as MAl2O4 (M: transition metal), since it increases the amount of Al3+ in tetrahedral coordination.15,16 That is, chemical synthesis of spinel material should control the inversion parameter precisely. In this case, an effective evaluation method is also required in order to accurately measure this quantity. Recently, Braga et al reported the attractive potential of hercynite (FeAl2O4) for application as a chemical synthesis catalyst to produce styrene, one of the important chemical component in the modern worldwide.17,18 An important consideration is that FeAl2O4 consist of the ubiquitous elements, rather than expensive metals such as rhodium or platinum. Also, the chemical industry has recently faced a drastic paradigm shift in fabrication processes brought about by the change of natural resources for these components from naphtha to methane hydrate. These considerations encourage us to further investigate the potential of FeAl2O4 as a synthesis catalyst of chemical components like styrene from new feedstocks. There is a long history of the study of transition metal aluminates with the composition of MAl2O4 but we have found few fabrications that discuss FeAl2O4.17-19 There are probably two main reasons for this lack. One is that attractive applications for this oxide had not been discovered before the report by Braga et al.17 The second is that the fabrication of FeAl2O4 is relatively difficult compared to other metal spinel aluminates, since it requires preventing the oxidation of Fe2+.19 Unfortunately, the synthesis conditions used for spinel aluminates easily oxidize Fe2+ to Fe3+, so this point will probably be a future issue in investigating the catalytic performance of FeAl2O4. There are few reports on the relationship between the precise structure and the physical properties of FeAl2O4. In particular, the influence of the fabrication conditions on the inversion parameter and the resulting physical properties have not yet been clarified for FeAl2O4. Given this background, we planned to obtain systematic data about the relationship between the fabrication conditions and the precise structure of FeAl2O4 as the first step for starting its study of its application. Here, we focus on the inversion parameter of the products as the most important structural parameter. Additionally, the magnetic properties of all the samples were also evaluated, since they often give us a meaningful principle to understand the usefulness of the product in the case of the processing studies of the materials including magnetic elements. Experimental The FeAl2O4 samples were prepared as polycrystalline powders. The starting materials of FeO (Kojundo chemical Laboratory: 99.5 %) and -Al2O3 (STREM chemicals: min 97 %) were mixed at the nominal ratio of 1:2 and pelletized. The pelletized mixture was placed in an alumina crucible (99.9%) and encapsulated in a quartz tube under vacuum atmosphere and set in a furnace. This tube was heated up to the sintering temperature for 3 hours, kept at that temperature for 24 hours and cooled down to 200 oC slowly. The sintering temperature was 900 oC or 1100 oC or 1200 oC. On the other hand, the cooling time was controlled in the range between quenching directly from the furnace ( 0 hour) and 200 hours. This manuscript labels all samples with the fabrication conditions in the style of Temp_time, i.e. Temp and time are the sintering temperature and the cooling time from Temp to 200 oC, respectively. For instance, the sample labeled 1200_40 is the powder sintered at 1200 oC for 24 h and cooled from this temperature to 200 oC for 40 hours (25 oC/h). The product FeAl2O4 samples were evaluated by X-ray diffraction (XRD, Mini-FLEX: Rigaku) measurements using Cu K (l = 1.542 Å) radiation. In addition, the structural information of these samples are analyzed from the Synchrotron XRD (SXRD) patterns obtained at the SPring-8 (BL15XU) beamline at Harima in Japan.20,21 The powder diffractometer using X-ray beam with λ = 0.65298 Å (Nb K edge Bearden) was used by scanning in steps of 0.003 degree over the 2θ range of 5 ~ 80 degree. Samples packed in the capillary tube with a diameter of 0.2 mm was used for the measurements and the data were collected at room temperature. The measured SXRD data were all refined by the Rietveld method22 using the RIETAN-FP software.23 The structural model was cubic with the space group of Fd-3m (No. 227-2). The total occupancies of Fe2+ and Al3+ at the A site and Al3+ and Fe2+ at the B site were both fixed at 1.0, and the total amount of Fe2+ and Al3+ was also restricted in order to express the composition to be (Fe1−yAly)(Al1−y/2Fey/2)2O4. The coordinate values of A and B sites are (0.125, 0.125, 0.125) and (0.5, 0.5, 0.5) for our Rietveld refinement, respectively, thus the refined characteristic structural parameters, except for some functional parameters, are basically the lattice constant, a, and u of the oxygen site (u, u, u). It means that our Rietveld refinement is a quite simple analysis. Fourier transform infrared (FT-IR, FT/IR-6200: JASCO) spectra performed on samples in KBr pellets provided useful confirmatory evidence of structural changes. The magnetic properties of the products were evaluated by using a superconducting quantum interference device magnetometer (SQUID, MPMS-XL: Quantum Design). Firstly, the temperature dependence of the magnetic susceptibilities, χDC, was measured at 10 kOe DC magnetic fields. The Weiss temperature, Θ, and the Curie constant for 1 mol amount, Cmol, were calculated from the experimental data of each temperature, T, between 120 and 300 K using the Curie−Weiss (CW) law, χmol = Cmol/(TΘ). The effective moment, peff, was calculated as the equation, peff μB = √3kBCmol/NA, where kB, NA, and μB are the Boltzmann constant, Avogadro number and Bohr magneton, respectively. On the other hand, the theoretical effective moment, peff-cal can be calculated with the equation, peff-cal = g√s(s+1) , where g and s are the g-factor and spin quantum number of the magnetic element, respectively. Thus, if we regard peff = peff-cal and fix s = 4/2 as the value for Fe2+, the g-factor of the samples are calculated from the obtained Cmol.8,24 Furthermore, this study tried to check the temperature dependence of the AC susceptibilities, χAC, with the magnitude of 5 Oe for some samples in order to see the frequency dependence in the range between 15 and 1500 Hz. Structural Characteristics Figure 1 (a) shows the XRD patterns of all product samples. They are judged to be single phase FeAl2O4 spinel, although high resolution SXRD measurement detected very small amount of Al2O3 peaks as the prime impurity for the samples sintered at 900 oC (see Figure 1 (b)). The intensity of the Al2O3 peaks are quite small, hence we ignored them in all structural analysis in this study. On the other hand, this information about Al2O3 peaks indicates the existence of a very small amount of iron-rich regions or phase in the sample that may has possibility affect the magnetic properties. Table 1 summarizes the main results of the Rietveld refinement for the SXRD results of all samples. Here, Rwp and S indicates the reliability value of weighted pattern and goodness of the fitting, respectively. These data were easily refined, and thus the values of the reliability indicators were good. We found that the inversion parameters were relatively all higher than that of other spinel aluminates like CoAl2O4.8 It is one of the unique characteristics of the FeAl2O4 system. Figure 2 plots the lattice constant against inversion parameter of our samples. The data looks scattered, but it does show as general trend.25-27 That is, both increasing the heating temperature and slowing the cooling rate contribute to reduce the inversion parameter of the spinel oxide. This trend is also roughly seen in the result of FT-IR measurements (see Figure 3). Here, an increase in the inversion parameter relatively enhances the intensity of the peak shoulder around 750 20 30 40 50 60 70 809 12 15 18 21 24 27 30 332θ / degreesIntensity(A.U.)1200_2001200_401200_01100_01100_36900_36900_28(a)(b) 900_28Al2O3220400311422 511440620533620533422 511 440400220311Figure 1Figure 1 (a) XRD patterns of product FeAl2O4 samples. (b) SXRD patterns pf 900_28. cm-1 and decreases the peak intensity around 520 cm-1 (see the arrowed shoulder and peak in the figure). These changes are considered to be strongly influenced by the site exchange phenomenon.28 The inversion parameter is commonly believed to be the most closely connected factor that determines the lattice parameter of a spinel oxide. However, Figure 2 shows that the lattice constant depends on the heating temperature rather than the inversion parameter. As the reason, we considered as follows. At first, slowing the cooling rate apparently reduces the inversion parameter. On the other hand, high temperature sintering is generally thought to enhance the homogenization of the cation distribution.29 Hence, the heating temperature is not considered to be the critical parameter to reduce the inversion parameter. However, higher heating temperature results in the elongation of the cooling time. It is thought to promote the 0.10 0.12 0.14 0.16 0.188.1588.1608.1628.1648.166( a ) ( b )( c )1200 oC1100 oC900 oCInversion parameter, yLatticeconstant/ ÅFigure 2Figure 2 Relationship between the inversion parameter and the lattice constant of product FeAl2O4 samples. The plotted color of each sample corresponds to that in Figure 1. The square symbol (■) indicates the quenched sample. The other symbols indicate the rate of cooling as a circle (● : 25 oC/h), an inverted triangle (▼ : 19.4 oC/h) and a triangle (▲ : 5 oC/h) , respectively. appropriate ordering of cations (i.e. reducing the inversion parameter). In fact, the inversion parameter appears to be similar between 1200_40 and 1100_36 (in both cases the cooling rate is 25 oC/h) and between 1200_0 and 1100_0. Probably, the homogeneity of the overall cation distribution was more important for reducing the lattice constant than reducing the inversion parameter. Table 1 Experimental results of product FeAl2O4 samples with respect to the refined structure and magnetic characteristics 1200_200 1200_40 1200_0 1100_36 1100_0 900_36 900_28 a (Å) 8.15958(1) 8.16070(2) 8.15999(1) 8.16285(2) 8.16151(1) 8.16253(3) 8.16376(6) u 0.26399(4) 0.26377(6) 0.26387(4) 0.26381(5) 0.26354(4) 0.26382(5) 0.26375(7) y 0.1220(14) 0.1348(16) 0.1606(12) 0.1358(14) 0.1726(10) 0.1578(14) 0.1710(20) Rwp 4.160 4.016 3.118 2.816 2.500 2.907 3.508 S 2.893 2.636 2.011 1.768 1.585 1.839 2.21 Cmol (emu K/Oe) 3.80 3.93 3.85 3.90 3.92 4.47 7.07 peff 5.51 5.60 5.54 5.55 5.59 5.96 7.53 g 2.25 2.29 2.26 2.28 2.28 2.44 3.07 Θ (K) -124.5 -128.9 -132.1 -134.0 -138.9 -164.8 -265.1 Tg (K) 11.4 12.5 12.2 13.2 13.4 15.2 39.5 800 750 700 650 600 550 500( a ) ( b )( c )Wave number/ cm-1NormalizedIntensities（A.U.)1100_361100_0900_361200_401200_01200_200Figure 3Figure 3 FT-IR spectra normalized by the highest peak in the band around 675 cm-1 of product FeAl2O4 samples. The plotted color of each sample corresponds to that in Figure 1. Magnetic Properties Figure 4 plots the reciprocal χDC of 1200_200 as a function of the temperature. This figure shows the typical curvature of 1/χDC-T plot for all samples. The data of high temperature region over 120 K can be easily fitted by the CW law (see the red line drawn in this figure). The Cmol (emu K/Oe), peff, g and Θ (K) values analyzed by this fitting procedures are also summarized in Table 1. Figure 5 plots the g values of samples against the inversion parameter. If we ignore the data of 900_36 and 900_28 from the discussion, thr site-exchange phenomenon looks to slightly increase the g value of Fe2+ (as in the dashed line drawn in Figure 5) within the margin of the analysis error. The systematic change of g value influenced by the y value 0 5 10 15 20 25 30 35 401.61.82.02.22.42.62.83.03.20 50 100 150 200 250 300020406080100120( a ) ( b )( c )Temperature / K1/χDC/ (emu/g Oe)-1Temperature / KχAC/ 10-4emu/g Oe15 Hz1500 HzFigure 4Figure 4 Temperature dependence of the reciprocal χDC measured for 1200_200 as a typical 1/χDC-T curve of product FeAl2O4 samples. The insert shows the frequency dependence of the χAC-T curve of the same sample. indicated by this line is quite similar to the data for some other spinel aluminate series (MAl2O4),30-33 and is basically caused by the difference in the coordination of Fe2+ between tetrahedral (A site) and octahedral (B site). Because, the B site of the spinel structure tend to require the trivalent cation,34 its appearance is also supposed to be the reason for the changes of g value. In fact, the peff of Fe2+ is in the range of 5.2 ~ 5.5 in general, but our all calculated data show a slightly higher value. On the other hand, it is difficult to explain the exceptionally large value of peff and g values of 900_36 and 900_28. This abnormally deviated data presumably is caused by the existence of another type of ( a ) ( b )( c )gvalueInversion parameter, y1200 oC1100 oC900 oC0.10 0.12 0.14 0.16 0.182.252.302.352.402.452.502.20Figure 5Figure 5 Calculated g values of product FeAl2O4 samples plotted against the inversion parameter. The plotted color of each sample corresponds to that in Figure 1. Plotted symbols are distinguished by the cooling speed of the sample as a square (■: quenched), a circle (●: 25 oC/h), an inverted triangle (▼: 19.4 oC/h) and a triangle (▲: 5 oC/h), respectively. anomaly in these samples. The site-exchange phenomenon itself is not considered to form some kind of long-range spin ordering in the spinel structure. Also, the y values of 900_36 and 900_28 do not seem to be enough large to form such ordering. Therefore, we speculate that these exceptionally large g values imply the localization of Fe2+ and/or Fe3+ in the spinel oxide. The discussions in the next section also support this speculation. As the next step, we should discuss the 1/χDC-T plot in Figure 4 in the low temperature region under 120 K. The experimental data departs from the CW law, and we can see the cusp at around 12 K. The inset figure shows the frequency dependence of this cusp observed for the χAC-T plot of the same sample (1200_200). This cusp is considered to originate in the spin glass phenomenon, since it is often reported for some spinel aluminates including a 3d-transition-metal element as the A site cation.30-33 The inset shows that enhancing the frequency of the applied AC field increases the cusp temperature and decreases the magnitude of the susceptibility peak. These trends are reasonable to understand it as the cusp symbolizing the spin glass phenomenon. Note that, these cusp temperatures, Tg, are slightly higher than that of other spinel aluminates like CoAl2O4.10 This is considered to be due to the relatively higher inversion parameter of FeAl2O4 samples. The magnetic behaviour of FeAl2O4 is considered to be antiferromagnetism in the low temperature region in case of a sample not subject to the without site-exchange phenomenon (i.e. y = 0).35 However, preparing antiferromagnetic FeAl2O4 without site-exchange is quite difficult. In practice, the magnetic ordering among the A site Fe2+ ions is randomly disrupted by non-magnetic Al3+ by the site-exchange phenomenon. The magnetic ordering of the A site cation in this case has been actively discussed for some MAl2O4 series recently from the physics viewpoint.36-40 Here, we should not restrict the magnetic ordering via the exchange interaction only between nearest neighbors of M2+ at the A site, but we should additionally take into account the influence of the next nearest neighbor cations. Thus, the magnetic ordering of Fe2+ becomes quite complicate when some disorder like site-exchange phenomenon is introduced. Furthermore, the magnetic moment of Fe2+ at the B site is also considered to be a non-isolated state. The distance between a Fe2+ at the A site and one at the B site is shorter than that to the A site one of the next nearest neighbors. The y values of our samples were all larger than 0.125 (= 2/16), and the number of B site cations is 16 in the unit cell of spinel structure. In that case, there is the possibility to form rectangular bonding between two Fe2+ and O2- in the unit cell. This rectangular bonding pattern has the potential to form a different magnetic ordering from the antiferromagnetism of A site according to the Kanamori-Goodenough rule.41-43 These complexities in the magnetic ordering of Fe2+ are considered to give rise to the strong magnetic frustration for the long range spin-ordering of FeAl2O4. This scenario is the considered as the reason for the spin glass phenomenon to appear as the overall magnetism of FeAl2O4 in the temperature region.33 In that case, the Tg should correspond to the inversion parameter systematically. Table 1 includes the Tg values for all our samples, and Figure 6 plots these Tg against the y values. The trend that increasing the y value results in enhancing the Tg of FeAl2O4 can been seen from Figure 6. If we ignore the data of 900_36 and 900_28 from the discussion (see the dashed line in the figure), this trend is quite consistent with the discussion about the g value. Hence, we concluded that the site-exchange phenomenon essentially increases Tg of the FeAl2O4 system as seen in the dashed line in Figure 6. This figure also plots the |Θ| using the values in Table 1 with the 10 times larger scale of Tg. These plots are good agreement for all samples except for 900_28. The difference between Tg and Θ is often regarded as an indicator of the magnetic frustration in discussions of the spin glass materials.30,44 From this viewpoint, the 10 times scale is enough to judge the existence of the magnetic frustration. This result also supports our explanation about the spin glass phenomenon observed for our FeAl2O4. ( b )Tg/KInversion parameter, yTg|Θ||Θ|/K0.10 0.12 0.14 0.16 0.1810152025303540100150200250300350400Figure 6Figure 6 Relationship between the inversion parameter and the cusp temperature, Tg, (filled symbols plotted on the left y-axis scale) and the absolute values of the Weiss temperature, |Θ|, (cross mark, , plotted on the right y-axis scale) of product FeAl2O4 samples. Plotted symbols of Tg are distinguished by the cooling speed of the sample as a square (■: quenched), a circle (●: 25 oC/h), a inverted triangle (▼: 19.4 oC/h) and a triangle (▲: 5 oC/h), respectively. The plotted color of each sample corresponds to that in Figure 1. Discussion for the anomaly in FeAl2O4 sintered at 900 oC This study succeeded in showing systematic trends with respect to the relationship between the structural characteristics and the magnetic properties in FeAl2O4 with the exceptions of the experimental data of 900_36 and 900_28 which were drastically different from our expectations. Figure 7 compares the χDC-T curves of 1200_200 and 900_28. The cusp of 900_28 is apparently broader than that of 1200_200, and its magnitude is also higher. In addition, the χDC-T curve of 900_28 drastically changed with the value of the applied magnetic field (see the figure insert). This field dependence of 900_28 is apparently larger than that of the other our samples, and at first sight, looks as though it might be originating in superparamagnetism rather than the spin glass phenomenon. Proceeding from this viewpoint, we found a shoulder peak at around 20 K for the χDC-T curve of 900_28 at 0.1 kOe (see the insert in Figure 7). This shoulder peak was prominently observed only for 900_28. The Tg in Table 1 was determined from the main peak with the same criteria. However, the shoulder observed at around 20 K of 900_28 is probably the effective Tg originating in the site exchange phenomenon. This speculation is moreover consistent with the value of |Θ| discussed in Figure 6. If we regard the Tg of 900_28 as 20 K plotted as the dashed circle with a dashed arrow in Figure 6, the general trend between Tg , |Θ| and y seems to be more reasonable for our samples. In other words, the origin of the magnetic anomaly of 900_28 is considered to relate to the main peak of the χDC-T curve measured at 0.1 kOe. Then, the drastic shift produced by increasing the applying magnetic-field and the broader shape of this main peak are considered to be the blocking temperature of the superparamagnetism signal. Therefore, the origin of the magnetic anomaly of 900_28 is considered to relate to the influence of superparamagnetism signal. The origin of this ferromagnetic signal is quite difficult to ascertain in principle, although several workers have reported on the ferromagnetism of FeAl2O4.45-47 If the y of 900_28 is very high value, the ferromagnetic signal in this sample can be explained as the influence of Fe2+ at the B site. However, the y is not considered to be high enough for producing ferromagnetism. In addition, we found a tiny amount of 0 20 40 60 80 100 120 1401.02.03.04.05.06.07.00 50 100 150 2000.01.02.03.04.05.0( a )( c )Temperature / KχDC/ emug-1Oe-11200_200900_280.1 kOe1 kOe10 kOe50 kOeTemperature / KχDC/ emug-1Oe-1Figure 7Figure 7 Dependence of the DC susceptibility at 10 kOe on the temperature for the zero field cooled measurement. The data are plotted for 1200_200 (brown circle) and 900_28 (green circle). The insert shows the field dependence of the χDC-T curve for 900_28. impurity Al2O3 from the SXRD peak. Therefore, we considered the possibility that a small amount of Fe3O4 is the origin of the ferromagnetism in our sample. In fact, we can identify a small kink around 120 K in the χDC-T curves of 900_28 at 0.1 kOe and 1 kOe (see the arrow of the insert in Figure 7), that is speculated to be the Verway transition48,49. The amount of Fe3O4 must be small, since the kink is not visible in the data measured at high magnetic fields. This consideration is supported by the SXRD patterns of 900_28 (see Figure 8), which are prominentlt asymmetric. These asymmetric peaks are speculated to imply the existence of a different phase with a similar lattice. Therefore, we concluded that the origin of the magnetic anomalies of 900_28 and 900_36 are the influence of a superparamagnetism signal caused by a small amount of Fe3O4 impurity. 25.8 26.0 26.2 26.415.1 15.2 15.3 15.4( a )( c )2 θ / degreesIntensity(A.U.)1200_2001100_01100_36900_36900_28311 440Figure 8Figure 8 Deatil of SXRD peaks of some FeAl2O4 samples in this study. The plotted color of each sample corresponds to that in Figure 1. Small differences in the fabrication conditions have the potential to change the homogeneity of the sample. Iron oxide, especially, has various stable phases corresponding to small changes in the fabrication conditions. Their importance is already recognized by most scientists, but this study indicates the necessity of further strict discrimination. For instance, the preparation conditions of 900_28 and 900_36 are quite similar, the difference being only the cooling rates of 25 oC/hour and 19 oC/hour respectively. This difference tends to be regarded as the “(almost the) same”. However the magnetism of these two samples were quite different. According to the Figure 1 (XRD data of the solid state reaction), we tended to think that 900 oC is sufficiently high for to obtain a single phase of FeAl2O4. However, low temperature synthesis causes high inversion and involves the small amounts of impurities like Fe3O4, which can strongly influence the magnetism data. Probably, this point will be key for discussing the quality of chemically synthesized FeAl2O4, since it is basically fabricated at low temperatures. Summary The spinel FeAl2O4 has potential to be a synthesis catalyst for some valuable chemical component. However, there are few studies reporting on the relationship between the precise structure and the physical properties of this material. In particular, the influence of the fabrication conditions on the inversion parameter and its link to physical properties is yet unclear, a situation which promoted a systematic evaluation of this relationship. Fabricated FeAl2O4 samples were essentially single phase, so that the structural analysis was straightforward. These experimental results reveals that the inversion parameter, y, of (Fe1−yAly)(Al1−y/2Fey/2)2O4 is lowered by increasing the sintering temperature and lengthening the cooling time. However, the absolute inversion parameter values were relatively higher than that of other spinel aluminates, which leads to magnetic frustration of the overall magnetism of (Fe1−yAly)(Al1−y/2Fey/2)2O4. Therefore, our all FeAl2O4 samples showed the spin glass phenomenon in the low temperature region of the χDC-T curve, and characteristics magnetic values obtained from this curve, i.e. Cmol, peff, g, Θ and Tg, were all dependent on the y of (Fe1−yAly)(Al1−y/2Fey/2)2O4. In particular, Tg is consistently linked to the y value, and it corresponds to 1/10 of the |Θ| value. This appears to be universal for all samples sintered above 1100 oC. Consequently, the evaluation of these magnetic characteristics values will be an effective guidelines to judge the quality of synthesized FeAl2O4 as a chemical catalyst. This result is believed to be meaningful for the future investigation of FeAl2O4 for applying it as the catalyst. On the other hand, this study found structural and magnetic abnormalities in the data for FeAl2O4 fabricated at 900 oC. They appear to originate in the existence of a tiny and a small amount of Fe3O4 or related impurity. The effects of small amounts of impurities will be an important consideration when discussing the quality of chemically synthesized FeAl2O4, which is usually fabricated at a low temperature. Acknowledgements This work was partially supported by a Grant-in-Aid for Challenging Exploratory Research, KAKENHI(16K13999). References [1] O’neill H S C. Temperature dependence of the cation distribution in CoAl2O4 spinel. Eur. J. Mineral. 1994;6:603-609. [2] Nakatsuka A, Ikeda Y, Yamasaki Y, Nakayama N, Mizota T. Cation distribution and bond lengths in CoAl2O4 spinel. Solid State Commun. 2003;128:85-90. [3] Tielens F, Calatayud M, Franco R, Recio J M, Pérez-Ramírez J, Minot C. J. Phys. Chem. B. 2006;110: 988-995. [4] Hagiwara M, Kimura S, Nishihagi N, Suzuki T, Nohara M, Takagi H, Kindo K. High Field Magnetism of the Spinel Oxide CoAl2O4. J. Low Temp. Phys. 2010;159: 11-14. [5] Maljuk A, Tsurkan V, Zestrea V, Zaharko O, Cervellino A, Loidl A, Argyriou D N. Floating-zone growth of large high-quality CoAl2O4 single crystals. J. Cryst. Growth, 2009; 311:3997–4000. [6] Ji L, Tang S, Zeng H C, Lin J, Tan K L. CO2 reforming of methane to synthesis gas over sol–gel-made Co/g-Al2O3 catalysts from organometallic precursors. Appl. Catalysis A. 2001;207:247-255. [7] Walsh A, Yan Y, Al-Jassim M M, Wei S-H. Electronic, Energetic, and Chemical Effects of Intrinsic Defects and Fe-Doping of CoAl2O4: A DFT+U Study. J. Phys. Chem. C, 2008;112:12044-12050. [8] Ahn K-S, Yan Y, Kang M-S, Kim J-Y, Shet S, Wang H, Turner J, Al-Jassim M M. CoAl2O4-Fe2O3 p-n nanocomposite electrodes for photoelectrochemical cells. Appl. Phys. Lett., 2009, 95, 022116/1-3. [9] Ishii S, Nakane T, Furusawa T, Naka T. Influence of pH tuning at the precursor preparation process on the structural characteristics and catalytic performance of hydrothermally synthesized ZnAl2O4 nanoparticles. J. Asian Ceram. Soc. 2018;6:7-12. [10] Nakane T, Naka T, Sato K, Taguchi M, Nakayama M, Mitsui T, Matsushita A, Chikyow T. Dalton Trans. Spectroscopic and crystallographic anomalies of (Co1-xZnx)Al2O4 spinel oxide. 44, 997-1008 (2015). [11] Chen X Y, Ma C, Zhang Z J, Wang B N. Ultrafine gahnite (ZnAl2O4) nanocrystals: Hydrothermal synthesis and photoluminescent properties. Mater. Sci. Eng. B. 2008;151: 224-230. [12] Miron I, Enache C, Vasile M, Grozescu I. Optical properties of ZnAl2O4 nanomaterials obtained by the hydrothermal method. Physica Scripta. 2012;T149:014064. [13] Kock L D, De Waal D. Raman studies of the underglaze blue pigment on ceramic artefacts of the Ming dynasty and of unknown origins. J. Raman Spect. 2007;38:1480−1487. [14] Ishii S, Nakane T, Furusawa T, Naka T. Synthesis of single-phase ZnAl2O4 nanoparticles via a wet chemical approach and evaluation of crystal structure characteristics. Cryst. Res. Technol. 2016;51:324–332. [15] Knözinger H, Ratnasamy, P. Catalytic Aluminas: Surface Models and Characterization of Surface Sites. Catal. Rev. 1978;17;31-70. [16] Zecchina A, Coluccia S, Morterra C. Infrared Sectra of Molecules Adsorbed on Oxide durfaces. Appl. Spectrosc. Rev. 1985;21;259-310. [17] Braga T P, Longhinotti E, Pinheiro A N, Valentini A. Synthesis of hybrid spheres for the dehydrogenation of ethylbenzene in the presence of CO2. Appl. Catal. A: Gen. 2009;362:139–146. [18] Batista A H de M, Ramos F S O, Braga T P, Lima C L, Sousa F F de, Barros E B D, Filho J M, Oliveira A S de, Sousa J R de, Valentini A, Oliveira A C. Mesoporous MAl2O4 (M= Cu, Ni, Fe or Mg) spinels: Characterisation and application in the catalytic dehydrogenation of ethylbenzene in the presence of CO2. Appl. Catal. A: Gen. 2010; 382: 148-157. [19] Chen J, Yu L, Sun J, Li Y, Xue W. Synthesis of hercynite by reaction sintering. J. Euro. Ceram. Soc. 2011;31:259-263. [20] Tanaka M, Katsuya Y, Yamamoto A. A new large radius imaging plate camera for high-resolution and high-throughput synchrotron x-ray powder diffraction by multiexposure method. Rev. Sci. Instrum. 2008;79:075106. [21] Tanaka M, Katsuya Y, Matsushita Y, Sakata O. Development of a synchrotron powder diffractometer with a one-dimensional X-ray detector for analysis of advanced material. J. Ceram. Soc. Jpn. 2013;121:287-290. [22] Rietveld H M. A profile refinement method for nuclear and magnetic structures. J. Appl. Cryst. https://jlc.jst.go.jp/DN/JALC/00360496147?type=list&lang=ja&from=J-STAGE&dispptn=11969;2:65-71. [23] Izumi F, Momma K. Three-dimensional visualization in powder diffraction. Sol. Stat. Phenom. 2007;130:15-20. [24] Naka T, Sato K, Matsushita Y, Terada N, Ishii S, Nakane T, Taguchi M, Nakayama M, Hashishin T, Ohara S, Takami S, Matsushita A. Multistage ordering and critical singularities in Co1−xZnxAl2O4 (0 ≤ x ≤ 1): Dilution and pressure effects in a magnetically frustrated system. Phys. Rev. B. 2015;91:224412. [25] Larsson L, O’Neill H St C, Annersten H. Crystal chemistry of synthetic hercynite (FeAl2O4) from XRD structural refinements and Moessbauer spectroscopy. Eur. J. Mineral. 1994;6:39-51. [26] Peterson R C, Lager G A, Hitterman R L. A time-of-flight neutron powder diffraction study of MgAl2O4 at temperatures up to 1273 K. Am. Mineral. 1991;76:1455–1458. [27] Shishin D, Prostakova V, Jak E, Decterov S A. Critical assessment and thermodynamic modeling of the Al–Fe–O system. Metall. Mater. Trans. B. 2016;47:397-424. [28] Tatarchuk T, Shyichuk A, Lamkiewicz J,Kowalik J. Inversion degree, morphology and colorimetric parameters of cobalt aluminate nanopigments depending on reductant type in solution combustion synthesis. Ceram. Int. 2020;46:14674-14685. [29] Harrison R J, Redfern S A T, O’Neill H St C. The temperature dependence of the cation distribution in synthetic hercynite (FeAl2O4) from in-situ neutron structure refinements. Am. Mineral. https://www.scientific.net/ssp.130.15https://jlc.jst.go.jp/DN/JALC/00028162261?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/00028162261?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/00116722032?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/00116722032?type=list&lang=ja&from=J-STAGE&dispptn=11998;83:1092–1099. [30] Tristan N, Hemberger J, Krimmel A, Krug von Nidda H A, Tsurkan V, Loidl A. Geometric frustration in the cubic spinel MAl2O4 (M = Co, Fe, and Mn). Phys. Rev. B. 2005;72:174404. [31] Dutta D P, Sharma G. Synthesis and magnetic behavior of spinel FeAl2O4 nanoparticles. Mater. Sci. Eng. B. 2011;176:177-180. [32] Hanashima K, Kodama Y, Akahoshi D, Kanadani C, Saito T. Spin Glass Order by Antisite Disorder in the Highly Frustrated Spinel Oxide CoAl2O4. J. Phys. Soc. Jpn. 2013;82:024702. [33] Nair H S, Kumar K R, Strydom A M. Spin freezing in the spin-liquid compound FeAl2O4. Phys. Rev. B 2015;91: 054423. [34] Jastrzębska I, Szczerba J, Stoch P, Błachowski A, Ruebenbauer K, Prorok R, Śnieżek E. Crystal structure and Mössbauer study of FeAl2O4. NUKLEONIKA. 2015;60:47-49. [35] Roth W L. Magnetic properties of normal spinels with only a-a interactions. J. Phys. France 1964;25:507-515. [36] Krimmel A, Mutka H, Koza M M, Tsurkan V, Loidl A. Spin excitations in frustrated A-site spinels investigated with inelastic neutron scattering. Phys. Rev. B. 2009;79:134406. [37] Fichtl R, Tsurkan V, Lunkenheimer P, Hemberger J, Fritsch V, Krug von Nidda H-A, Scheidt E-W, Loidl A. Orbital freezing and orbital glass state in FeCr2S4. Phys. Rev. Lett. 2005;94:027601. [38] Fritsch V, Hemberger J, Büttgen N, Scheidt E W, Krug von Nidda H A, Loidl A, Tsurkan V. Spin and https://jlc.jst.go.jp/DN/JALC/70016436339?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/70016436339?type=list&lang=ja&from=J-STAGE&dispptn=1Orbital Frustration in MnSc2S4 and FeSc2S4. Phys. Rev. Lett. 2004;92:116401. [39] Bergman D, Alicea J, Gull E, Trebst S, Balents L. Order-by-disorder and spiral spin-liquid in frustrated diamond-lattice antiferromagnets. Nature Phys. 2007;3:487-491. [40] Chen G, Balents L, Schnyder A P. Spin-Orbital Singlet and Quantum Critical Point on the Diamond Lattice: FeSc2S4. Phys. Rev. Lett. 2009;102:096406. [41] Goodenough J B, Loeb A L. Theory of Ionic Ordering, Crystal Distortion, and Magnetic Exchange Due to Covalent Forces in Spinels. Phys. Rev. 1955;98:391-408. [42] J. Kanamori. Superexchange interaction and symmetry properties of electron orbitals. J. Phys. Chem. Sol. 1959;10:87-98. [43] Wilkinson M K, Cable J W, Wollan E O, Koehler W C. Neutron Diffraction Investigations of the Magnetic Ordering in FeBr2, CoBr2, FeCl2, and CoCl2. Phys. Rev. 1959;113:497-507. [44] Gingras M J, Stager C V, Raju N P, Gaulin B D, Greedan J E. Static Critical Behavior of the Spin-Freezing Transition in the Geometrically Frustrated Pyrochlore Antiferromagnet Y2Mo2O7. Phys. Rev. Lett. 1997;78:947-950. [45] Fukushima J, Hayashi Y, Takizawa H. Structure and magnetic properties of FeAl2O4 synthesized by microwave magnetic field irradiation. J. Asian Ceram. Soc. 2013;1:41-45. [46] Mukhtar F, Kayani Z N, Riaz S, Naseem S. Effect of in-situ oxidation on structure and ferromagnetic properties of Fe3Al and FeAl2O4 thin films prepared by electrodeposition. Ceram. Int. 2018;44:9550-9560. [47] Yaseen M, Mahmood Q, Ramay S M, Ali I, Naz M Y, Mahmood A. The First-Principle Study of the Electronic Structure, Ferromagnetic and Thermoelectric Properties of Spinel Alloy FeAl2O4 Using mBJ Functional Approach. J. Supercond. Nov. Mag. 2018;31:1435-1441. [48] Verwey E J W. Electronic Conduction of Magnetite (Fe3O4) and its Transition Point at Low Temperatures. Nature 1939;144:327-328. [49] Verwey E J W, Haaymann P W. Electronic conductivity and transition point of magnetite (“Fe3O4”). Physica. 1941;8:979-987. Table 1 Experimental results of product FeAl2O4 samples with respect to the refined structure and magnetic characteristics Figure captions Figure 1 (a) XRD patterns of product FeAl2O4 samples. (b) SXRD patterns pf 900_28. Figure 2 Relationship between the inversion parameter and the lattice constant of product FeAl2O4 samples. The plotted color of each sample corresponds to that in Figure 1. The square symbol (■) indicates the quenched sample. The other symbols indicate the rate of cooling as a circle (● : 25 oC/h), an inverted triangle (▼ : 19.4 oC/h) and a triangle (▲ : 5 oC/h) , respectively. Figure 3 FT-IR spectra normalized by the highest peak in the band around 675 cm-1 of product FeAl2O4 samples. The plotted color of each sample corresponds to that in Figure 1. Figure 4 Temperature dependence of the reciprocal χDC measured for 1200_200 as a typical 1/χDC-T curve of product FeAl2O4 samples. The insert shows the frequency dependence of the χAC-T curve of the same sample. Figure 5 Calculated g values of product FeAl2O4 samples plotted against the inversion parameter. The plotted color of each sample corresponds to that in Figure 1. Plotted symbols are distinguished by the cooling speed of the sample as a square (■: quenched), a circle (●: 25 oC/h), an inverted triangle (▼: 19.4 oC/h) and a triangle (▲: 5 oC/h), respectively. Figure 6 Relationship between the inversion parameter and the cusp temperature, Tg, (filled symbols plotted on the left y-axis scale) and the absolute values of the Weiss temperature, |Θ|, (cross mark, , plotted on the right y-axis scale) of product FeAl2O4 1200_200 1200_40 1200_0 1100_36 1100_0 900_36 900_28 a (Å) 8.15958(1) 8.16070(2) 8.15999(1) 8.16285(2) 8.16151(1) 8.16253(3) 8.16376(6) u 0.26399(4) 0.26377(6) 0.26387(4) 0.26381(5) 0.26354(4) 0.26382(5) 0.26375(7) y 0.1220(14) 0.1348(16) 0.1606(12) 0.1358(14) 0.1726(10) 0.1578(14) 0.1710(20) Rwp 4.160 4.016 3.118 2.816 2.500 2.907 3.508 S 2.893 2.636 2.011 1.768 1.585 1.839 2.21 Cmol (emu K/Oe) 3.80 3.93 3.85 3.90 3.92 4.47 7.07 peff 5.51 5.60 5.54 5.55 5.59 5.96 7.53 g 2.25 2.29 2.26 2.28 2.28 2.44 3.07 Θ (K) -124.5 -128.9 -132.1 -134.0 -138.9 -164.8 -265.1 Tg (K) 11.4 12.5 12.2 13.2 13.4 15.2 39.5 samples. Plotted symbols of Tg are distinguished by the cooling speed of the sample as a square (■: quenched), a circle (●: 25 oC/h), a inverted triangle (▼: 19.4 oC/h) and a triangle (▲: 5 oC/h), respectively. The plotted color of each sample corresponds to that in Figure 1. Figure 7 Dependence of the DC susceptibility at 10 kOe on the temperature for the zero field cooled measurement. The data are plotted for 1200_200 (brown circle) and 900_28 (green circle). The insert shows the field dependence of the χDC-T curve for 900_28. Figure 8 Deatil of SXRD peaks of some FeAl2O4 samples in this study. The plotted color of each sample corresponds to that in Figure 1.