**1. Introduction**

Aluminum titanate (AT (Al2TiO5)) is a compound with low thermal expansion, excellent thermal shock resistance, and low thermal conductivity. In view of these properties,

**Citation:** Fan, K.; Jiang, W.; Ruiz-Hervias, J.; Baudín, C.; Feng, W.; Zhou, H.; Bueno, S.; Yao, P. Effect of Al2TiO<sup>5</sup> Content and Sintering Temperature on the Microstructure and Residual Stress of Al2O3–Al2TiO<sup>5</sup> Ceramic Composites. *Materials* **2021**, *14*, 7624. https://doi.org/10.3390/ ma14247624

Academic Editors: Donatella Giuranno and Christos G. Aneziris

Received: 7 October 2021 Accepted: 8 December 2021 Published: 11 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

it is a suitable second-phase candidate for improving the thermal and mechanical properties of alumina (Al2O3)-based ceramic systems [1,2]. Alumina–AT (A–AT) ceramics have attracted considerable attention as flaw-tolerant ceramics in thermal and structural applications, such as thermal insulation liners, diesel particulate filters, vehicle emissions control, and high-temperature flue gas filtration supports, because of their improved flaw tolerance, toughness, and superior thermal properties [3–6]. In addition to the typical investigation of mechanical properties, such as flaw tolerance, crack resistance, and thermal shock resistance, many studies have focused on the reinforcement mechanisms of A–AT ceramics [7–9]. The key factors presumed to be responsible for the toughness and improved flaw tolerance of ceramic composites are residual stresses [7].

High tensile and/or compressive residual stresses are expected to develop in aluminabased composites owing to the thermal expansion mismatch among different phases and thermal expansion anisotropy in each phase during cooling after sintering [10]. In the micro-mechanical interaction between matrix and reinforcing particles, residual stresses appear to perform an important function in the mechanical properties of A–AT composites. Padture et al. [11] and Asmi et al. [12] showed that the addition of AT to alumina could improve flaw tolerance and crack-growth resistance. They found that this resulted from the existence of residual stresses arising from the thermal mismatch between alumina and AT. Skala et al. [7] reported that the residual stresses in A–AT composites are responsible for a wide variety of possible toughening mechanisms (e.g., crack deflection, crack bridging, and microcracking). Residual stresses could influence the capability of materials to absorb energy from external loading and distribute damage, consequently leading to enhanced quasi-plasticity and flaw tolerance. The occurrence of spontaneous microcracking in AT was also reported to be caused by residual stresses induced by the strong thermal expansion anisotropy of AT (*αa*,*AT*25−<sup>1000</sup> ◦<sup>C</sup> <sup>=</sup> <sup>−</sup>2.4 <sup>×</sup> <sup>10</sup>−<sup>6</sup> ◦<sup>C</sup> −1 , *<sup>α</sup>b*,*AT*25−<sup>1000</sup> ◦<sup>C</sup> = 11.9 <sup>×</sup> <sup>10</sup>−<sup>6</sup> ◦<sup>C</sup> −1 and *<sup>α</sup>c*,*AT*25−<sup>1000</sup> ◦C= 20.8 <sup>×</sup> <sup>10</sup>−<sup>6</sup> ◦<sup>C</sup> −1 ) [13], resulting in a low elastic modulus, approximately 25–40 GPa [14]. In contrast, in Botero's study [9], a significantly improved elastic modulus was attained in fine-grained AT using nanoindentation in dense A–AT composites without microcracks; this was attributed to the residual stress interaction between matrix and reinforcing particles. These statements show that the mechanical properties of A–AT composites are considerably related to residual stresses and microstructure. To improve A–AT ceramic properties, the strict control of the microstructure and clear comprehension of the nature as well as the magnitude of residual stresses generated during fabrication are necessary.

The interest in developing laminated ceramics to achieve superior flaw tolerance is growing [15]. In addition to the effect of laminate stacking design (e.g., layer thickness and stacking structure), detailed residual stress information (e.g., residual stresses between phases and grains) regarding each layer is crucial. This is extremely important to understand the reinforcing mechanism and optimize the resulting residual stress field, which is useful to produce optimum laminated structures with operative reinforcing layers. It was supposed that the residual stresses in ceramics could be modified by selecting the appropriate material design and fabrication process [16,17]. For A–AT ceramics, the addition of AT and the implementation of sintering treatments can considerably influence their microstructure and mechanical properties [18,19]. Because the co-sintering of layers with different compositions in laminated materials is required, the effects of introducing AT and sintering treatments on residual stress development in addition to those related to laminate stacking design must be considered.

Several studies on the residual stress analysis of A–AT composites have been conducted. Singh [20] and Skala et al. [7] used synchrotron radiation diffraction and X-ray diffraction (XRD) techniques in their investigation. They employed single-peak shift analysis for residual strain measurement in functionally graded A–AT composites. However, they only considered the shifts of individual peaks of Al2O<sup>3</sup> and AT; residual strain was presented without residual stresses. Fluorescence piezo-spectroscopy was implemented to detect stress in each layer of the multilayered Al2O3–Al2TiO<sup>5</sup> composites. The results

indicated weak tension in AT layers (5–20 MPa) and compressive hydrostatic stresses in alumina layers (20 MPa) [10]. The residual stresses in each A–AT laminate layer were also evaluated using simplified model calculations [21]. The reported tensile stresses in A-10AT (composite containing 10 vol.% AT) and compressive stresses in A-30AT (containing 30 vol.% AT) layers were expected to be approximately 15 and 90 MPa, respectively. According to previous work, the residual stress results were presented in different values, depending on fabrication routes, composition design, and residual stress determination techniques. Most studies focus on uniform residual stresses in between layers, laminates, or phases. However, they do not account for thermal and elastic crystal anisotropies at the grain-scale level, which may perform a significant function in crack initiation processes, such as fatigue or failure. Furthermore, owing to crystal anisotropy and microstructure complexity, the residual strains from single-peak diffraction measurements or model calculations may not represent bulk material behavior [22]. To date, direct and reliable experimental data on multi-scale residual stress in A–AT composites remain lacking. To accurately quantify residual stresses under different fabrication conditions or material compositions, the most suitable residual stress measurement technique for each case may be implemented.

The time-of-flight (TOF) neutron diffraction technique allows for the non-destructive measurement of residual strains in bulk materials because it is capable of high neutron penetration. In addition, the entire diffraction pattern, which is important for determining the residual stress in complex materials, such as ceramic composites, can be determined [23]. Information on multi-scale residual stresses, such as average stress among phases, intergranular stress state within a single phase, and even nonuniform microstrains at the atomic scale, can be derived from the analysis of the entire diffraction pattern [22].

In this study, TOF neutron diffraction and Rietveld analysis were applied to precisely determine the residual stresses in a series of A–AT composites with different AT contents and sintering treatments and, therefore, with significantly different microstructures. Multiphase qualitative and quantitative analyses as well as crystal structure determination were performed using Rietveld refinement. The through-thickness residual stress profiles of the mean phase stress and intergranular stress state of each phase were obtained for all the examined A–AT composites. The effects of second-phase AT addition and sintering temperature were discussed based on the observed microstructures and detected residual stresses.
