Mixed Grain Structure Analysis of P91 Thick-Walled Steel Pipes

Literature Overview

This 2018 paper by Ning Fangkun, Jia Weitao, Zhu Fuxian, and Wang Ping from Northeastern University investigates the mixed grain (混晶) phenomenon observed in P91 thick-walled steel pipes during metallographic examination. Funded by the National Natural Science Foundation of China (51690161), the study analyzes the relationship between the mixed grain structure, the hot rolling process, and heat treatment practices, and proposes a corrective approach involving the conversion of non-equilibrium microstructure to equilibrium microstructure before final heat treatment. Published in the Journal of Northeastern University (Vol. 39, No. 1, pp. 50–54), the paper addresses a critical quality issue in the production of thick-walled P91 steel pipes used in power plant applications.

Core Technical Content

Background: P91 Steel and Its Applications

P91 (ASTM A335 P91 / EN 10216-2 10CrMo9-10) is a 9Cr-1Mo martensitic stainless steel widely used in ultra-supercritical (USC) power plant applications, including thick-walled piping for steam lines, superheaters, and reheaters operating at temperatures up to 650 °C. The material's excellent creep resistance and high-temperature strength are derived from a fine martensitic microstructure with precipitated M₂₃C₆ carbides. However, the manufacturing of thick-walled P91 pipes is challenging because the material is prone to mixed grain structures, which severely degrade high-temperature performance.

Mixed Grain Phenomenon

Mixed grain (also called mixed grain structure or dual grain structure) refers to the coexistence of fine and coarse grains within the same cross-section of a steel pipe. In P91 thick-walled pipes, this phenomenon manifests as:

• Fine grain regions: Typically in the outer layers of the pipe wall, where the cooling rate during hot rolling is higher.
• Coarse grain regions: Typically in the inner layers or core of the pipe wall, where the cooling rate is lower and grain growth is more pronounced.

The coarse grain regions are particularly problematic because they have significantly lower creep strength and fracture toughness than the fine grain regions, creating a heterogeneous microstructure that can lead to premature failure under high-temperature service conditions.

Root Cause Analysis

The authors identify three key factors contributing to the mixed grain structure:

The Piercing Stage as the Critical Phase

A key finding of the study is that the mixed grain structure originates during the piercing (皮尔格轧制) stage of the hot rolling process. The authors observed that under identical hot rolling parameters, the seamless pipe blanks (毛管) showed no mixed grain, while the elongated pipes (荒管) exhibited abnormal grain growth. This indicates that the piercing stage introduces the initial microstructural heterogeneity that cannot be fully corrected by subsequent rolling or heat treatment.

The mechanism is as follows:

1. During piercing, the billet is deformed at high temperature with a mandrel, creating a hollow shell (seamless pipe blank).
2. The deformation is highly non-uniform: the outer surface experiences greater strain and higher cooling rate, while the inner surface (near the mandrel) experiences less strain and slower cooling.
3. This non-uniform deformation creates a gradient in the recrystallized grain size across the wall thickness.
4. Subsequent elongation rolling does not fully homogenize this gradient because the strain levels are insufficient to cause complete recrystallization of the coarse grain regions.
5. The resulting non-equilibrium microstructure carries through to the final heat treatment, where the coarse grain regions continue to exhibit larger grain sizes even after normalizing.

Corrective Approach: Non-Equilibrium to Equilibrium Microstructure Conversion

The authors propose a two-stage heat treatment approach:

1. Stage 1 – Non-equilibrium to equilibrium conversion: A high-temperature austenitizing treatment (e.g., 1100–1150 °C for extended time) is applied to fully homogenize the austenite grain structure. At this temperature, the pre-existing grain size differences are eliminated through complete austenite grain growth, resulting in a uniform (equilibrium) grain structure.
2. Stage 2 – Final normalizing + tempering: The standard heat treatment is then applied to the homogenized material. Because the starting microstructure is now uniform, the final martensitic structure is also uniform, eliminating the mixed grain phenomenon.

Comparison of Heat Treatment Approaches

Engineering Practice Integration

Impact on Power Plant Safety

The mixed grain phenomenon in P91 thick-walled pipes is a serious safety concern in ultra-supercritical power plants. The coarse grain regions have:

• Reduced creep strength: Creep rupture life can be reduced by 30–50% in coarse grain regions compared to fine grain regions at the same temperature and stress level.
• Reduced fracture toughness: Coarse grains have lower Charpy V-notch impact energy, increasing the risk of brittle fracture during transients or seismic events.
• Accelerated oxidation: Coarse grain boundaries are more susceptible to oxidation and intergranular attack at high temperatures.

In service, these heterogeneous regions can act as preferential sites for crack initiation and propagation, leading to unexpected pipe failures. Several power plant incidents have been attributed to mixed grain structures in P91 piping.

Quality Control Recommendations

Based on the findings of this paper and general industry practice, the following quality control measures are recommended for P91 thick-walled pipe production:

FMEA Analysis of Mixed Grain Failure

Study Insights

This paper makes a significant contribution to the understanding of P91 steel pipe manufacturing, particularly in identifying the piercing stage as the critical phase for mixed grain formation. The proposed corrective heat treatment approach—converting the non-equilibrium microstructure to an equilibrium microstructure before final normalizing and tempering—is a practical and effective solution that can be implemented in existing production facilities without major equipment modifications.

From a metallurgical perspective, the study reinforces the principle that microstructural homogeneity is not achieved by the final heat treatment alone but must be established during the hot working process. The piercing stage, which is often treated as a purely geometric operation (converting a solid billet to a hollow shell), is in fact a critical metallurgical operation that determines the initial grain structure. This insight should influence the design of piercing parameters (temperature, reduction ratio, rolling speed) to minimize microstructural heterogeneity at the source.

For engineers involved in the procurement and qualification of P91 thick-walled pipes for power plant applications, this paper provides a clear basis for specifying metallographic examination requirements and for evaluating the effectiveness of heat treatment procedures. The recommendation to measure grain size at multiple wall locations and to require a uniformity ratio of less than 2:1 is a practical and achievable quality criterion that should be incorporated into purchase specifications.

The broader implication is that the manufacturing of thick-walled high-performance alloy pipes requires a holistic approach that integrates metallurgical understanding with process engineering. The traditional approach of relying on final heat treatment to correct all microstructural defects is insufficient for materials as demanding as P91; instead, each manufacturing stage must be optimized to produce a uniform microstructure that can be maintained through subsequent processing. This is a lesson that extends beyond P91 to other high-performance alloy systems used in demanding applications such as nuclear power, aerospace, and petrochemical processing.