Quantitative Investigation on Strengthening and Toughening Mechanism of 1000 MPa Grade Hydropower Steel

Abstract

The microstructure and properties in the 1000 MPa grade hydropower steel during “quenching + tempering” process was investigated. Key microstructural parameters such as effective grain size, carbon content of martensitic matrix, and dislocation density were analyzed by scanning electron microscope, x-ray diffraction and electron backscatter diffraction. Tensile properties and Charpy V-notch (CVN) impact toughness of experimental steels were measured, enabling quantitative analysis of the relationship between microstructure and yield strength, ductile-to-brittle transition temperature (DBTT). Results show that, after quenching at 890 °C and tempering at 630 °C for 120 min, the effective grain size was limitedly refined to 1.70 μm. The carbon content of martensitic matrix is reduced to 0.0154 pct, and the dislocation density decreased to 3.63 × 10¹⁴ m⁻². Accordingly, the grain refinement strengthening, solid solution strengthening and dislocation strengthening amounted to 422, 138 and 93 MPa, respectively. The DBTT was reduced to −81 °C, and the contribution from grain refinement, dislocation and precipitation strengthening and microstructure heterogeneity was −308, 140 and 124 °C, respectively. The improvement in DBTT compared to that tempered at 630 °C for 15 min was mainly related to the decrease of dislocation density and ripening of precipitated particles, i.e., the increased plastic deformability of martensitic lattice.

Appendix

The calculation procedure of dislocation density and parameters employed are presented as follows (Ref 20). The $\Delta K$, K and $\overline{C}$ in Eq 5 can be expressed as

$${ }\Delta K = \frac{{2cos\theta \left( {\Delta \theta } \right)}}{\lambda }$$

(11)

$${ }K = \frac{{2{\text{sin}}\theta }}{\lambda }$$

(12)

$$\left\{ {\begin{array}{*{20}c} {\overline{C} = \overline{{C_{h00} }} \left( {1 - qH^{2} } \right)} \\ {} \\ {H^{2} = \frac{{h^{2} k^{2} + h^{2} l^{2} + k^{2} l^{2} }}{{\left( {h^{2} + k^{2} + l^{2} } \right)^{2} }}} \\ \end{array} } \right.$$

(13)

Then, Eq 5 can be reformulated as

$$\frac{{\left[ {\frac{{2cos\theta \left( {\Delta \theta } \right)}}{\lambda } - \frac{0.9}{D}} \right]^{2} }}{{\left( {\frac{2sin\theta }{\lambda }} \right)^{2} }} \cong 2\pi b^{2} \rho \overline{{C_{h00} }} \left[ {1 - q\frac{{h^{2} k^{2} + h^{2} l^{2} + k^{2} l^{2} }}{{\left( {h^{2} + k^{2} + l^{2} } \right)^{2} }}} \right]$$

(14)

where $\Delta \theta$ is the FWHM (radian) of the diffraction peak (Ref 9), and $q$ is a constant. H is a constant affected by the indices of crystallographic plane. $\overline{{C }_{\mathrm{h}00}}$ is the average dislocation contrast factor for the {h00} reflections and is determined by the dislocation contrast factor for the {h00} reflections (C_h00) of pure edge and screw dislocations (${C}_{h00\left(edge\right)},{C}_{h00\left(screw\right)}$) as well as their respective fractions $(f^{edge} ,f^{screw} )$.

$$\left\{ \begin{gathered} \overline{{C_{h00} }} = C_{{h00\left( {edge} \right)}} f^{edge} + C_{{h00\left( {{\text{screw}}} \right)}} f^{screw} \hfill \\ f^{edge} = \frac{{q_{screw}^{th} - q}}{{q_{screw}^{th} - q_{edge}^{th} }} = 1 - f^{screw} \hfill \\ \end{gathered} \right.$$

(15)

where C_h00 is related to the elastic parameters, i.e., C₁₁, C₁₂ and C₄₄ of metals. The ${q}_{i}^{th}$, where i stands for edge or screw) represents the theoretical parameter value of pure edge or screw dislocations and was calculated as follows (Ref 41)

$$\left\{ {\begin{array}{*{20}c} {q_{i}^{th} = a_{i}^{q} \left[ {1 - \exp \left( {\frac{ - A}{{b_{i}^{q} }}} \right)} \right] + c_{i}^{q} A + d_{i}^{q} } \\ {C_{h00i} = a_{i}^{{C_{h00} }} \left[ {1 - \exp \left( {\frac{ - A}{{b_{i}^{{C_{h00} }} }}} \right)} \right] + c_{i}^{{C_{h00} }} A + d_{i}^{{C_{h00} }} } \\ {A = \frac{{2C_{44} }}{{C_{11} - C_{12} }}} \\ \end{array} } \right.$$

(16)

where A is the elastic anisotropy parameter. The $a_{i}^{{C_{h00} }} ,{ }b_{i}^{{C_{h00} }} ,c_{i}^{{C_{h00} }} ,d_{i}^{{C_{h00} }}$ and $a_{i}^{q} ,b_{i}^{q} ,c_{i}^{q} , d_{i}^{q}$ are all constants. The C₁₁, C₁₂ and C₄₄ of martensite was determined to be 230, 135 and 117 MPa, respectively (Ref 42). Important model parameters in Eq 16 and 15 are listed in Table 4 and 5, respectively.

Table 4 The α_i, b_i, c_i and d_i in Eq. 16 from (Ref 41)

Table 5 The calculated parameters in Eq. 15 for steel 0 T

In this work, the calculated $\overline{{C }_{h00}}$ for steel 0 T is used when calculating the dislocation density of the tempered sample. Then, $\overline{C }$ was calculated by Eq 13 and presented in Table 6 along with K for each tempered sample.

Table 6 The K and $\overline{C}$ for each tempered sample

It should be mentioned that the two necessary parameters D and q for the calculation of $\overline{{C }_{h00}}$ were determined from the linear correlation between $\Delta K$ and K in Eq 5, $\left[ {\frac{{cos\theta \left( {2\Delta \theta } \right)}}{\lambda } - \frac{0.9}{D}} \right]^{2} /\left( {\frac{2sin\theta }{\lambda }} \right)^{2}$ and $\frac{{h^{2} k^{2} + h^{2} l^{2} + k^{2} l^{2} }}{{\left( {h^{2} + k^{2} + l^{2} } \right)^{2} }}$ in Eq 14, respectively (see Fig. 13). Finally, the dislocation density for steel 0 T was calculated to be 2.862 × 10¹⁵/m².

Fig. 13

Linear fitting of “$\Delta K$” vs. “K” to determine D in (a) and “$\left( {\Delta K - 0.9/D} \right)^{2} /K^{2}$” vs. “$H^{2}$” to determine q in (b)