**2. Materials and Experiments**

### *2.1. Materials*

The as-forged GH4698 alloy ingot used in this study had a size of Φ300 mm × 1000 mm. The chemical composition of the GH4698 alloy was quantified via x-ray fluorescence (XRF1800, Shimadzu Inc., Kyoto, Japan) and an infrared carbon-sulfur analyzer (CS2800, Eltra Inc., Haan, Germany), as shown in Table 1 [23,24]. It should be noted that the carbon content of the material used in this research was at the upper limit allowed for GH4698. The specimens for elemental analysis were prepared by wire electrode cutting, turning, and mechanical polishing to Φ33 mm × 12 mm cylinders. The hot compression specimens were prepared by wire electrode cutting and turning to Φ8 mm × 12 mm cylinders.


**Table 1.** Composition of the GH4698 alloy.

### *2.2. Experiments*

Hot compressions were conducted on a compression machine (Gleeble3500, Dynamic Systems Inc., Austin, TX, USA). The compression temperatures were determined to be 950 ◦C, 1000 ◦C, 1050 ◦C, 1100 ◦C, and 1150 ◦C, respectively. These temperatures covered the usual hot working temperature range of GH4698 [1]. To cover the strain rate range of the billet forging process, preforging process, and final-forging process of large forgings on hydraulics, the strain rates were selected to be 0.001 s<sup>−</sup>1, 0.01 s−1, 0.1 s−1,1s−1, and 3 s−1, respectively. The experimental procedure is shown in Figure 1. The specimens were heated to 1150 ◦C at 3.3 ◦C/s, held for 180 s, cooled to deformation temperatures, held for 180 s, compressed to the strain of 0.95, and quenched. The true stress and logarithmic strain were calculated accordingly. The compressive strains, which had negative values, were written as positive for simplicity in this research. Smoothing was applied on the stress-strain curves.

**Figure 1.** Illustration of the experiment procedure.

#### **3. Results and Discussion**

#### *3.1. Flow Behaviors*

The stress-strain curves of GH4698 at various strain rates are shown in Figure 2. Basically, single-peak flow stress curves were obtained. In the starting stage of deformation, the flow stress increased because of the contradicting effect of work hardening by dislocation pile ups and cross slips, as well as dynamic softening, by creep and recovery. As the deformation proceeded, the dislocation densities at the grain boundaries reached the critical value for recrystallization, and the dynamic recrystallized grains nucleated, resulting in the softening and a gradual drop in the flow stress [25]. When the dynamic recrystallization was completed, the flow stress remained nearly constant, as the hardening effect and the softening effect were balanced. It can also be seen from Figure 2 that the flow stress decreased with increasing temperature. This could be explained by the observation that the thermal movement of atoms was more intense at higher temperatures. The creep and dynamic recrystallization were more likely to occur owing to fewer obstacles for the dislocation motion to overcome, resulting in the overall softening of the GH4698 alloy. With the increasing strain rate, the flow stress increased. This is because at a faster strain rate, it was more difficult for the dynamic recovery to fully occur, leading to a higher average dislocation density and a greater deformation resistance. It is worth noting that the peak flow stress shown in Figure 2 agreed well with the results of Zhang et al. [22], but the curve shapes were different, which may be attributed to differences in the composition of as-received material. As mentioned in Section 2.1, the carbon content of the material used in this research, 0.08%, was obviously higher than that in the literature [22], 0.048%. For nickel-based alloys, Chen et al. [4] and Lin et al. [7] proved that dynamic recrystallization occurred during hot compressions. Therefore, a refined and uniform microstructure could be obtained by

selecting optimal hot working parameters. In this way, the hot working ability could be improved, and cracking defects of large forgings in the billet opening process could be avoided.

**Figure 2.** Stress strain curves of GH4698 compressed at the strain rates of (**a**) 0.001 s<sup>−</sup>1, (**b**) 0.01 s<sup>−</sup>1, (**c**) 0.1 s<sup>−</sup>1, (**d**)1s<sup>−</sup>1, and (**e**)3s−1.
