Acta Metall Sin  2018, Vol. 54 Issue (5): 773-788    DOI: 10.11900/0412.1961.2017.00525
 Special Issue for the Solidification of Metallic Materials Current Issue | Archive | Adv Search |
Solutions in Improving Homogeneities of Heavy Ingots
Jun LI1,2, Mingxu XIA1, Qiaodan HU1, Jianguo LI1,2()
1 School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
2 Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration, Shanghai Jiao Tong University, Shanghai 200240, China
Abstract

The inhomogeneity in large ingots not only decides the final properties of the product, but also restricts downstream hot working processing severely. It is very important to improve the homogeneity of ingots for saving energy, improving material utilization ratio, increasing performance of component, and the construction of key equipment. In this paper, the general inhomogeneity problem in large ingots, such as macrosegregation, inclusion, shrinkage porosity, and large crystal have been introduced. The evolutions of this inhomogeneity in the subsequent hot working processing have also been discussed, based on which the concept of homogeneity window for large ingots has been proposed. The research progress of numerical simulation of macrosegregation in large ingots and some new methods for improving the homogeneity of large ingot have also been introduced and analyzed. Three fundamental reasons for the inhomogeneity of ingots were concluded, i.e. the uneven cooling rate, the uncontrollable multiphase flow, and the solute redistribution during solidification. Aiming at these three fundamental reasons, a novel casting method called layer casting (LC), which has been proposed by our team recently, was introduced to modify the serious inhomogeneity problem in large ingots. In this method, molten alloy was poured into the mold separately and layer upon layer. As soon as the poured molten alloy solidified to a critical volume fraction range, the next layer amount of molten alloy was poured into the mold. For each layer, the mass, composition, and pouring temperature of poured molten alloy could be artificially designed and controlled based on the target homogeneity window. Both experiment and numerical simulated results shown that, in comparison with conventional ingot fabrication method, the LC method can significantly decrease the uncontrollable multiphase flow, uniform the cooling rate, and improve the solute redistribution, subsequently, improve the homogeneity of ingots. For large ingots fabrication, the LC method has the potential to substantially decrease the energy consumption, materials consumption, and the investment of large equipment. Its wide application prospect for high quality large ingots is also expected.

 ZTFLH: TG244
Fund: Supported by National Key Research and Development Program of China (No.2017YFB0305300), Joint Funds of the National Natural Science Foundation of China (No.U1660203), National Natural Science Foundation of China (No.51404152) and Shanghai Pujiang Program (No.14PJ1404800)
 Fig.1  Schematic of the influence of composition and ingot size on inhomogeneity of ingots Fig.2  Simulated distributions of C (a) and Mn (b) contents varied with annealing time for SA508-3 alloy Fig.3  Variation of required weight of ingot with final weight of forging (h—thickness, SR—surface radius) Fig.4  Schematic of the homogeneity control window of nuclear island closure head of AP1000 (Each symbol indicates one parameter. All solid symbols indicate the failure of materials. σb—tensile strength, δ—elongation, Akv—impact energy) Fig.5  Schematic of the relationship of pouring amount and solidification fraction with time elapsed for mould casting Fig.6  Schematic of the smelting and pouring processing of 600 t steel ingot[22] Fig.7  Pouring processing of vacuum multi-ladle casting Fig.8  Carbon distribution along ingot centerlines of large ingots for different pouring methods: conventional vs multiple pouring (MP)[29] Fig.9  Typical characteristics of macrosegregation (a) and sulphur print (b) in large scale ingot[30] and schematic macrosegregation in ingot (c)[31] Fig.10  Final macrosegregation color maps of 3.25 t steel ingot predicted by simplified dendritic model (a), globular model (b), and the comparison between simulated and experimental macrosegregation distributions along the central line for different models (c)[84] Fig.11  Macrosegregation of 55 t steel ingot[83] (a) 2D case with 4-phase shrinkage model (b) experimental result(c) shrinkage (d) segregation index along centerline Fig.12  Characteristic of 3D A-segregation[83](a) 3D channel segregation pattern(b) longitudinal section of channel segregation (c) experimental A-segregation Fig.13  Schematic of layer casting (LC) (c1, c2,$?$, cn—melt composition of each ladle; T1, T2,$?$,Tn—pouring temperature of each ladle; t1, t2,$?$, tn—pouring time interval for each ladle)[88] Fig.14  Schematic of layer casting procedure on 600 t steel ingot with four 50 t furnaces (The whole ingot was casted by 12 layer pouring) Fig.15  As-cast grain structures of conventional processing (CP) ingot (a1) and LC ingot (a2), microstructural of CP ingot (b-a1-1, b-a1-2 and b-a1-3) and LC ingot (b-a2-1, b-a2-2 and b-a2-3) and distribution of grain size[89] Fig.16  Distribution of copper composition in term of segregation index (cmix-c0/c0) of CP (a1) and LC (a2) ingots and distribution of segregation index along the centerline (b)[89] Fig.17  Macrosegregation distributions curves, on which the macrosegregation indexs and corresponding coordinates along the center line of some points are given, along the center line of 100 t steel ingots fabricated by conventional casting method, LC method of 5 ladles and 10 ladles, predicted by numerical simulations[88]