Multi-probe microstructure tracking during heat treatment without an in-situ setup: Case studies on martensitic steel, dual phase steel and β-Ti alloy

Highlights

• A multi-probe method to track microstructures during heat treatment is developed.

• It enables the analysis of various complex phenomena, even those at atomistic scale.

• It circumvents some of the free surface effects of classical in-situ experiments.

Abstract

In-situ scanning electron microscopy observations of the microstructure evolution during heat treatments are increasingly demanded due to the growing number of alloys with complex microstructures. Post-mortem characterization of the as-processed microstructures rarely provides sufficient insight on the exact route of the microstructure formation. On the other hand, in-situ SEM approaches are often limited due to the arising challenges upon using an in-situ heating setup, e.g. in (i) employing different detectors, (ii) preventing specimen surface degradation, or (iii) controlling and measuring the temperature precisely. Here, we explore and expand the capabilities of the “mid-way” solution by step-wise microstructure tracking, ex-situ, at selected steps of heat treatment. This approach circumvents the limitations above, as it involves an atmosphere and temperature well-controlled dilatometer, and high resolution microstructure characterization (using electron channeling contrast imaging, electron backscatter diffraction, atom probe tomography, etc.). We demonstrate the capabilities of this approach by focusing on three cases: (i) nano-scale carbide precipitation during low-temperature tempering of martensitic steels, (ii) formation of transformation-induced geometrically necessary dislocations in a dual-phase steel during intercritical annealing, and (iii) the partial recrystallization of a metastable β-Ti alloy.

Introduction

Current alloy design strategies for expanding the achievable property spectrum often lead to microstructures with increased complexity, e.g. by incorporating additional constituents, grain size distributions, and compositional heterogeneities at different scales. However, regardless of whether an austenitic–martensitic transformation-induced-plasticity (TRIP) steel, an age hardenable Al alloy, an α + β-Ti alloy, or any other alloy that is considered, the microstructure evolution cannot be easily back-tracked from the final microstructure. This is due to the various sequentially or simultaneously occurring micro-processes involved. Thus, there is a growing motivation for the employment of in-situ techniques for the analysis of the microstructure development. In-situ studies in transmission electron microscopy (TEM) [1], [2], [3], lab-scale X-ray systems [4], [5], and high-energy beamlines [6], [7], [8], [9] have enabled unprecedented leaps in the understanding of various micro-phenomena [2], [4], [7], [9]. However, when considering microstructure development during heat treatment, each of these techniques have their specific limitations. Thin TEM foils are affected strongly by surface effects [10], and provide limited possibilities of follow-up analyses by other techniques (e.g. by atom probe tomography (APT)). In-situ X-ray diffraction (XRD), synchrotron X-ray diffraction (SXRD) or neutron diffraction studies access large volumes of integrated microstructural information (e.g. phase fraction, texture), however not allowing for the direct visualization of the microstructure itself. 3D X-ray microscopy [11], [12] provides an improvement in this regard, yet still with limited spatial resolution not allowing for studying complex nanostructured alloys.

On the other hand, in-situ heating experiments in the scanning electron microscope (SEM) [13], [14], [15], [16], [17], [18], [19] provide an optimal combination of investigated material volume, surface effects, and varieties of imaging modes, resolution and practicality. Yet, the presence of a miniaturized heating stage in the SEM chamber can create certain limitations as well. Employing different detectors (backscattered electrons (BSE) and electron backscatter diffraction (EBSD) [20]) or imaging modes (electron channeling contrast imaging (ECCI) [21]) during heat treatment is limited due to degradation of the resolution and signal/noise ratio (caused by the thermal expansion, radiation and the spurious magnetic or electrical fields [13], [22]), or even not possible due to other practical limitations (e.g. space, measurement duration). In fact, even with in-situ setups in chamber, EBSD measurements were often carried out after cooling down to room temperature [17], [18] because of the limited applicable temperature and time resolution at high temperatures [14], [16], [22]. Additionally, precise control and measurement of temperature on the sample is difficult. Often the thermocouple is fixed to the heating stage rather than on the sample [14], [22]. Achievable heating/cooling rates are usually lower than 100 K·s^− 1, typically from 0.2 to 10 K·s^− 1 [1], [13], [18], [22], [23]). Furthermore, even with the advanced vacuum systems of current SEMs, avoiding oxidation at high temperatures can be an important challenge.

In this work, we explore the capabilities of an approach which can be applied without an in-situ setup, that consists of (i) interrupted heat treatments [24] in a dilatometer, and (ii) microstructure characterization in-between by EBSD, ECCI, APT, etc. In what follows we refer to this methodology as “quasi-in-situ” due to the similarity of the data generated in classical in-situ experiments, although the probing process itself is not in-situ. The case studies which focus on three different alloys and heat treatment regimes demonstrate that through careful design of this experimental methodology, this practical approach can deliver significant amount of insight to complex micro- and nano-processes.