길이 변화 곡선을 이용한 소둔 조건이 AHSS의 상변태 거동에 미치는 영향 분석

Abstract

냉간 압연 공정 (cold-rolled process)에 의하여 제조되는 AHSS (Advanced High Strength Steels)는 소둔 종료 시점의 오스테나이트 상태에 의하여 상변태 거동이 크게 변화하여 최종 미세 조직과 기계적 성질에 큰 영향을 미치는 것으로 알려져 있다. 그러므로 연속 냉각 전 시점에서 소둔 조건에 따른 오스테나이트 상태를 정확하게 예측하여야 한다. 열처리 전 초기 미세 조직은 color tint 에칭된 광학현미경 (OM, Optical Microscopy) 분석과 전계방사 주사전자현미경 (FE-SEM, Field-Emission Scanning Electron Microscopy) 분석 결과 70 %의 페라이트와 15 %의 베이나이트 및 마르텐사이트로 형성되어 있었다. 또한 석출 상태는 조대한 TiN과 미세한 NbC가 관찰되었으며, 900 ℃에서 120 sec 동안 소둔한 이후 석출 상태에 변화가 없었기 때문에 열처리 중 석출물의 고용 및 추가 석출은 진행되지 않았다. 치환형 원소의 이동이 고려된 ortho-평형 조건하에서 계산된 페라이트 변태 온도 (OAe3)보다 낮은 온도인 780 ℃에서 소둔 시 1200 sec 유지 동안 오스테나이트로의 역변태는 완료되지 않았다. 또한 20.0 sec 이상 소둔 시 역변태 속도가 감소하는 것이 관찰되었기 때문에 20.0 sec 미만의 소둔 시간은 탄소 확산 제어 역변태가 진행되며 그 이상의 소둔 시간은 페라이트 내의 Mn의 확산이 역변태를 제어하는 구간이라고 할 수 있었다. 또한 승온 중 역변태가 완료되지 않은 840 ℃ 소둔 시 18.9 sec 동안 오스테나이트로의 역변태는 완료되었다. 연속 냉각 중 상변태 거동은 소둔 온도에 따라 서로 다르게 관찰되었다. 우선 오스테나이트로의 역변태가 승온 중 완료되는 900 ℃ 소둔 시 냉각 속도 증가에 따라 상변태 시작 온도와 상변태가 가장 빠르게 진행되는 온도 (TP, peak temperature)가 감소하였다. 오스테나이트로의 역변태가 종료되지 않은 780 ℃에서 120 sec 동안 소둔한 후 연속 냉각 중 전체적인 상변태 거동은 냉각 속도 증가에 따라 오히려 촉진되었으며, 상변태 종료 온도는 일반적인 거동과 같이 냉각 속도 증가에 따라 감소하는 현상이 관찰되었다. 소둔 중 오스테나이트로의 역변태가 종료된 840 ℃, 120 sec 소둔 조건의 경우 느린 냉각 속도에서 이단 변태 거동이 관찰되었으며, 10 ℃/s의 냉각 속도를 가진 경우 900 ℃ 소둔한 경우와 유사한 거동을 나타냈다. 이와 같이 냉각 속도에 따라 비이상적인 상변태 거동이 나타나는 원인은 오스테나이트 내의 탄소 불균일에 의한 것으로 예상하였다. 이러한 탄소 불균일에 의하여 840 ℃ 소둔 시 냉각 중 상변태가 진행되는 온도 영역이 900 ℃ 소둔한 경우와 비교하여 넓게 관찰되었으며 이는 탄소 함량이 낮은 오스테나이트가 먼저 변태가 진행되어 변태 시작 온도를 증가시키고, 탄소 함량이 높은 오스테나이트는 상변태가 충분히 지연된 이후 저온 영역에서 변태가 진행되기 때문에 변태 종료 온도를 감소시키는 것으로 판단하였다. 900 ℃에서 120 sec 동안 소둔한 경우 오스테나이트 내 탄소는 균질하게 분포되어 있다고 가정하였는데, 840 ℃에서 1200 sec 동안 소둔한 실험은 냉각 중 상변태 거동이 900 ℃ 소둔 실험과 매우 유사하게 관찰되었다. 그러므로 840 ℃에서 120 sec 소둔 시 탄소는 불균일하게 분포되어 있으며, 1200 sec 소둔 시 오스테나이트 내 탄소의 불균일이 해소되었다고 할 수 있다. 또한 연속 냉각 중 상변태가 진행되는 온도 영역이 900 ℃ 소둔 실험과 유사해지는 소둔 시간은 780 ℃ 역시 1200 sec인 것으로 확인되었기 때문에 오스테나이트 내 탄소의 재분배가 완료되어 탄소 불균일 분포가 완전히 해소되는 시간은 780, 840 ℃ 모두 1200 sec라고 할 수 있었다. 승온 중 오스테나이트로의 역변태가 완료되지 않은 780, 840 ℃ 소둔 온도에서 소둔 시간에 따른 냉각 중 상변태 거동을 분석한 결과 상변태 시작 시점의 변태 거동이 감소되는 소둔 시간이 각각 300, 180 sec인 것으로 측정되었다. 그러므로 탄소의 불균일 분포가 냉각 중 상변태에 큰 변화를 나타내지 않을 정도로 충분히 해소된 시간은 780 ℃의 경우 300 sec, 840 ℃의 경우 180 sec로 판단하였다. 780 ℃에서 120 sec 소둔한 후 10 ℃/s의 냉각 속도로 냉각한 경우에 형성된 최종 미세 조직은 단일 결정립이 내부의 베이나이트와 결정립 계면의 마르텐사이트로 이루어진 상이 관찰되었다. 이러한 상의 형성은 Mn이 오스테나이트 결정립 계면으로 이동하고 쌓여 있는 결과로서 C, Mn의 함량이 주위의 다른 상에 비하여 높게 측정되었다. 그러므로 오스테나이트 역변태 과정은 ortho-평형 조건을 따르며, 20.0 sec 이후의 역변태는 페라이트 내 Mn이 오스테나이트 계면으로 이동하여 성장하는 DMn 제어 변태라고 할 수 있다. 그러므로 본 연구에서는 오스테나이트로의 역변태 및 연속 냉각 중 상변태 거동을 길이 변화 곡선 및 이의 1차 미분 곡선을 통해 분석하였으며, 최종 미세 조직을 OM 및 FE-SEM을 이용하여 관찰하고 EPMA (Electron Probe Micro-Analyzer)를 이용하여 상간 원소 분포를 측정하였다. |It was well-known that final microstructures and mechanical properties were strongly influenced by the state of austenite at the end of soaking in cold-rolled AHSS (Advanced High Strength Steels). Therefore, it was necessary that the state of austenite was accurately expected before continuous cooling. Initial microstructure of as-rolled specimen was composed of 70 % ferrite and 15 % bainite and martensite as results of the analysis of optical micrograph by color tint etching and FE-SEM (Field-Emission Scanning Electron Microscopy). Coarse TiN and fine NbC precipitation were observed on TEM (Transmission Electron Microscopy) images of as-rolled specimen. Because similar fraction and morphology of precipitation were measured in the experiment of cooling rate of 10 ℃/s after 120 sec soaking at 900 ℃, it was decided that precipitations were not dissolved and supplementarily formed during heat treatments. Re-austenization was not completed during 1200 sec soaking at 780 ℃, because this temperature was lower than the ortho-equilibrium temperature of the transformation of ferrite (OAe3). The rate of re-austenization was decreased over 20.0 sec soaking at 780 ℃. Therefore, re-austenization was controlled by carbon diffusion under 20.0 sec and manganese diffusion in ferrite over 20.0 sec at 780 ℃. In the case of soaking temperature of 840 ℃, re-austenization was completed for 18.9 sec. Transformation behaviors with cooling rates were very different as soaking temperatures. First, the start, finish, and peak temperatures (TP), which had the fastest transformation rate, of the transformation were decreased with the increase of cooling rate at 900 ℃, 120 sec soaking condition, which was complete re-austenization during continuous heating (3 ℃/s). The finish temperatures of the transformation were decrease with the increase of cooling rate, but the start temperatures of the transformation were increased at 780 ℃, 120 sec soaking condition. Re-austenization was completed during 120 sec soaking at 840 ℃. In this case, two different transformations took place during continuous cooling of 0.2 ℃/s, but transformation behavior of 10 ℃/s was quite similar to the case of 10 ℃/s at 900 ℃, 120 sec soaking condition. It was expected that the reason of abnormal transformation behaviors as soaking temperatures was the inhomogeneous distribution of carbon in austenite. When the specimen was soaked at 840 ℃, wide transformed regions were observed than the case of 900 ℃ soaking condition due to the inhomogeneity of carbon in austenite. Because carbon depleted austenite was transformed at relatively high temperature, the start temperature of the transformation was increased. The finish temperature of the transformation was decreased due to the increase of the hardenability of carbon enriched austenite. It was assumed that carbon was homogeneously distributed at 900 ℃, 120 sec soaking condition, and transformation behavior during continuous cooling after 1200 sec soaking at 840 ℃ was quite similar to the case of 900 ℃, 120 sec soaking condition irrespective of cooling rate. Therefore, it was determined that the distribution of carbon in austenite was homogeneous for 1200 sec soaking at 840 ℃. Because transformed region of the transformation during continuous cooling after 1200 sec at 780 ℃ soaking was similar to the case of 900 ℃, 120 sec soaking condition, it was decided that carbon in austenite was homogeneously distributed for 1200 sec soaking at 780, 840 ℃. It was observed that soaking times of the retardation of transformation behaviors at the early stage of the transformation during cooling were 300 and 180 sec respectively, at 780 and 840 ℃ soaking. Therefore, it was expected that soaking times, which were the start times of the sufficiently homogeneous distribution of carbon in austenite, was 300 sec at 780 ℃ and 180 sec at 840 ℃. It was observed that the single grain was composed of bainite inside and martensite at rim in the experiment soaked at 780 ℃ except 10 sec soaking. The reason of the formation of this phase was the pile-up of manganese at austenite grain boundary. Moreover, higher carbon and manganese concentration were measured at this martensite rim than bainite inside this single grain. Because of re-austenization controlled by DMn, manganese in ferrite diffused into austenite grain boundary, it was determined that re-austenization at 780 ℃ obeyed ortho-equilibrium condition. In this work, re-austenization and transformation behaviors during continuous cooling were analyzed by dilatations and the 1st derivatives of dilatations. Final microstructures were observed by OM (Optical Microscopy) and FE-SEM, and the distribution of carbon, manganese was measured by EPMA (Electron Probe Micro-Analyzer).; It was well-known that final microstructures and mechanical properties were strongly influenced by the state of austenite at the end of soaking in cold-rolled AHSS (Advanced High Strength Steels). Therefore, it was necessary that the state of austenite was accurately expected before continuous cooling. Initial microstructure of as-rolled specimen was composed of 70 % ferrite and 15 % bainite and martensite as results of the analysis of optical micrograph by color tint etching and FE-SEM (Field-Emission Scanning Electron Microscopy). Coarse TiN and fine NbC precipitation were observed on TEM (Transmission Electron Microscopy) images of as-rolled specimen. Because similar fraction and morphology of precipitation were measured in the experiment of cooling rate of 10 ℃/s after 120 sec soaking at 900 ℃, it was decided that precipitations were not dissolved and supplementarily formed during heat treatments. Re-austenization was not completed during 1200 sec soaking at 780 ℃, because this temperature was lower than the ortho-equilibrium temperature of the transformation of ferrite (OAe3). The rate of re-austenization was decreased over 20.0 sec soaking at 780 ℃. Therefore, re-austenization was controlled by carbon diffusion under 20.0 sec and manganese diffusion in ferrite over 20.0 sec at 780 ℃. In the case of soaking temperature of 840 ℃, re-austenization was completed for 18.9 sec. Transformation behaviors with cooling rates were very different as soaking temperatures. First, the start, finish, and peak temperatures (TP), which had the fastest transformation rate, of the transformation were decreased with the increase of cooling rate at 900 ℃, 120 sec soaking condition, which was complete re-austenization during continuous heating (3 ℃/s). The finish temperatures of the transformation were decrease with the increase of cooling rate, but the start temperatures of the transformation were increased at 780 ℃, 120 sec soaking condition. Re-austenization was completed during 120 sec soaking at 840 ℃. In this case, two different transformations took place during continuous cooling of 0.2 ℃/s, but transformation behavior of 10 ℃/s was quite similar to the case of 10 ℃/s at 900 ℃, 120 sec soaking condition. It was expected that the reason of abnormal transformation behaviors as soaking temperatures was the inhomogeneous distribution of carbon in austenite. When the specimen was soaked at 840 ℃, wide transformed regions were observed than the case of 900 ℃ soaking condition due to the inhomogeneity of carbon in austenite. Because carbon depleted austenite was transformed at relatively high temperature, the start temperature of the transformation was increased. The finish temperature of the transformation was decreased due to the increase of the hardenability of carbon enriched austenite. It was assumed that carbon was homogeneously distributed at 900 ℃, 120 sec soaking condition, and transformation behavior during continuous cooling after 1200 sec soaking at 840 ℃ was quite similar to the case of 900 ℃, 120 sec soaking condition irrespective of cooling rate. Therefore, it was determined that the distribution of carbon in austenite was homogeneous for 1200 sec soaking at 840 ℃. Because transformed region of the transformation during continuous cooling after 1200 sec at 780 ℃ soaking was similar to the case of 900 ℃, 120 sec soaking condition, it was decided that carbon in austenite was homogeneously distributed for 1200 sec soaking at 780, 840 ℃. It was observed that soaking times of the retardation of transformation behaviors at the early stage of the transformation during cooling were 300 and 180 sec respectively, at 780 and 840 ℃ soaking. Therefore, it was expected that soaking times, which were the start times of the sufficiently homogeneous distribution of carbon in austenite, was 300 sec at 780 ℃ and 180 sec at 840 ℃. It was observed that the single grain was composed of bainite inside and martensite at rim in the experiment soaked at 780 ℃ except 10 sec soaking. The reason of the formation of this phase was the pile-up of manganese at austenite grain boundary. Moreover, higher carbon and manganese concentration were measured at this martensite rim than bainite inside this single grain. Because of re-austenization controlled by DMn, manganese in ferrite diffused into austenite grain boundary, it was determined that re-austenization at 780 ℃ obeyed ortho-equilibrium condition. In this work, re-austenization and transformation behaviors during continuous cooling were analyzed by dilatations and the 1st derivatives of dilatations. Final microstructures were observed by OM (Optical Microscopy) and FE-SEM, and the distribution of carbon, manganese was measured by EPMA (Electron Probe Micro-Analyzer).