Proceedings of the 7th Annual ISC Graduate Research Symposium
ISC-GRS 2013
April 24, 2013, Rolla, Missouri
1
CHARACTERIZATION OF NON-METALLIC INCLUSIONS AND MICRO-POROSITY IN STEEL: FORGED INGOTS, CENTRIFUGAL AND NET SHAPE CASTINGS
Jianxun He
Department of Materials Science and Engineering,
Missouri University of Science and Technology
218 McNutt, 1400N. Bishop, Rolla, MO, USA 65401
Email:
1
1
Abstract
Micro-porosity and non-metallic inclusions are in many cases similar in size (micron/submicron size ranges) and appearance, and both negatively affect the mechanical properties of cast steel. However these defects typically have different origins, and improving steel properties by eliminating or suppressing their negative effects poses different technological challenges. Therefore, differentiation of micro-porosity and non-metallic inclusions is practically important.
In this article, practical aspects of sample preparation and automated SEM/EDX analysis are discussed. The developed experimental procedures are applied for analysis of different industrial cast steel products including centrifugal steel casting, investment casting and ingot casting products. The observed results were compared for different steels. The possibilities of healing micro-porosity by hot plastic deformation and hot isostatic pressure treatment are also evaluated.
1. INTRODUCTION
Non-metallic inclusions and porosity are common defects in steel. Their negative effects are evident in many areas of steel production and product application. Non-metallic inclusions affect steel’s castability and may cause problems such as nozzle clogging during continuous casting [1, 2], reduce its ductility, induce inferior surface defects, cracks and poor finish, jeopardize steel’s mechanical properties and change steel’s corrosion resistance when it is put into service. Porosity seems to have some of the same effects as non-metallic inclusions, acting as areas of stress concentration, surface defects, reducing fatigue resistance, etc. In other aspects, porosity has different effects from non-metallic inclusions, for instance, porosity can lead to leaks in fluid-containing castings [3]. Moreover, non-metallic inclusions may interact and combine with porosity, which has been observed in many cases.
The origins of non-metallic inclusions and porosity are different. Non-metallic inclusions can be divided into exogenous inclusions, inclusions primarily derived from external sources such as re-oxidation, slag entrainment and lining erosion, and indigenous inclusions, inclusions formed during solidification by the thermodynamic affinity of reactive components with the remaining impurities in the melt. Porosity sources in steel castings include air entrainment, blow holes during filling, gas evolution from solute during solidification, solidification shrinkage, mold or core blows, and volatile binder [4].
Improving steel properties by eliminating or suppressing negative effects of non-metallic inclusions and porosity is of great importance and poses different technological challenges. Today, techniques such as filtration of the melt, inert gas stirring [5], flux and slag absorption [6], inclusion modification [7], flow control [8, 9], dissolved oxygen control, protective atmosphere, etc. are available to control non-metallic inclusions [10, 11]. Porosity can usually be suppressed by degassing [12], vacuum treatment [13], flow control [14], solidification control [15], hot isostatic pressing [16-19], or hot plastic deformation [20], to name a few. The ultimate goal is to reduce both non-metallic inclusions and porosity as much as possible.
The current study focuses on the characterization of non-metallic inclusions and micro-porosity in centrifugal and investment castings and comparison to forged ingots on the basis of previous work completed at Missouri S&T [21-23].
2. STEELS INCLUDED IN STUDY AND ANALYSIS METHODOLOGY
Industrial steel samples were collected at different plants, including castings and ingots. Steels used are shown in Table 1.
Table 1: steel types studied and processing.
Casting Process / Type of SteelCentrifugal casting / Medium carbon low alloyed cast steel
Forged ingot / 316 L Stainless steel
Investment casting / High strength Eglin steel
Centrifugal casting: Medium carbon low alloy cast steel was melted in an electric arc furnace followed by an argon oxygen decarburization (AOD) treatment. Then the melt was transported through a ladle to the centrifugal caster. In total 3077 kg of cast steel was melted and poured into the centrifugal mold at a temperature of 1570 ºC. The ladle was argon stirred for approximately 15 minutes before casting.
The chemical composition measured in the AOD is shown in Table 2. During the centrifugal casting the maximum rotation speed was 860 rpm. The centrifugal force at the beginning of the casting process was between 95-160 G. It was then reduced at a constant rate to 30 G until solidification was completed. The centrifugally cast tube is 6.7 m (22 feet) long, it’s inside diameter (ID) is 0.336 m (14 inch) and its outside diameter (OD) is 0.508 m (20 inch). The wall thickness of the tube is 0.076 m (3 inch). Centrifugal cast samples were taken from the end of the tube which is opposite to the pouring side. Samples from five different thickness locations were taken, which is shown in Figure 1.
1
Table 2: Chemical composition in AOD, wt. %.
1
C / S / Si / Mn / Cr / Mo / Ni / Cu / V / Ti / Al / P / W0.20 / 0.001 / 0.16 / 0.628 / 2.70 / 0.76 / 0.10 / 0.13 / 0.05 / 0.004 / 0.062 / 0.013 / 0.015
1
Table 3: Composition of 316L Stainless steel in wt. %.
1
C / Si / Mn / S / P / Cr / Ni / Mo / Al / N0.02 / 0.43 / 1.50 / 0.004 / 0.022 / 16.31 / 10.28 / 2.10 / 0.002 / 0.064
1
1
Figure 1. Schematic representation of sampling of the centrifugally cast tube.
Forged ingot: 7168 kg of 316 stainless steel was bottom poured into the ingot mold. The chemical composition of 316 L stainless steel is shown in Table 3. The teeming temperature was 1534ºC. Finally the as-cast square ingot was hot forged into round product with a reduction ratio of 2.04:1. Samples at different heights and radii from the forged ingot were collected.
Investment casting: Eglin steel (ES) was developed by the U.S. Air Force as cost effective, ultra high-strength steel for military applications [24]. ES was investment cast into test bars. The chemical composition of the ES is shown in Table 4. The wax pattern for the investment casting is shown in Figure 2. Each test bar is 0.0254 m (1 inch) in diameter and 0.140 m (5.5 inch) in length. The shell material is fused silica. Before pouring, the shell was fired to 982 ºC for 1 h. The heat size was 68 kg. Argon gas was introduced through a porous plug in the bottom and under an insulation blanket on the top of the furnace. The pouring temperature was 1566 ºC. The melt was poured into 4 hot test bar trees for each heat.
Table 4: Composition of ES (wt %).
Fe / C / Cr / Mn / Mo / Nb / Ni / Si / Ti / V / W92.984 / 0.265 / 2.6 / 0.65 / 0.42 / 0.01 / 1.0 / 1.0 / 0.006 / 0.065 / 1
1
Figure 2. Wax pattern of the investment cast ES test bars (left) and cast test bars (right)
1
Hot Isostatic Pressing (HIP): HIP was employed to treat the investment cast ES bar at a participating plant in a high temperature pressurized furnace under an inert argon gas atmosphere. The HIP treatment was conducted at 1163 ºC for 4 h at 103.4 MPa. The test bars were cut at four different locations along the longitudinal direction.
Automated SEM/EDX analysis: An ASPEX PICA 1020 was used to simultaneously analyze the inclusions and porosity in the steel samples. Information such as size, morphology and composition can be obtained quickly using an automated SEM. Rule files and vector files were applied for the analysis. The user-defined rules are divided into two categories: classification and zero elements. Starting with the first rule in the list, ASPEX categorizes each detected spot if the rule is true. If it is not, the following rule in the list will be applied. Images with EDS analysis can be obtained simultaneously during scanning [25]. The zero element rules are used to distinguish inclusions and porosity from stains, which may cause error in the results. Zero rules for the base elements such as Fe, Ni and Cr can reduce the computational work of EDX spectra and make the scanning process more efficient. For different steels, the rule definition of inclusion and porosity is also different. An example of applied rules is given in Table 5.
1
Table 5: Example of classification rules of porosity and inclusions for centrifugal casting.
Process / Classification / RulesCentrifugal Casting / Vector File / C, Na, Mg, Al, Si, S, K, Ca, Ti, Cr, Mn, Fe, Ni and O
Carbon / If C>5
FeO Stains
Porosity
MnS
CaS
Ca-Mn-S
Other Sulfides
Al2O3
CaO
SiO2
MnO / Fe>=30
Fe >=90 and O<5
Mn>30 and S>20
Ca>30 and S>20
Ca>10 and Mn>10 and S>20
S>20
Al>20 and Mn<20 and Ca<10 and Si<20 and S<20
Ca>5 and S<20 and Al<10 and Ca<10
Si>30 and Mn<20 and S<20 and Al<10 and Ca<10
Mn>20 and S<20 and Al<20 and Si<20
Zero Elements
Cr=0 / If Cr>=0.1
Fe=0 / If Al>=2.5 or Mn>=2.5 or Ca>=2.5 or C>=2.5 or Si>=2.5
C=0 / If Al>=2.5 or Mn>=2.5 or Ca>=2.5 or Si>=2.5
1
3. RESULTS AND DISCUSSION
Centrifugal casting: Micro-porosity and non-metallic inclusions with diameters less than 10 microns were observed in the samples from centrifugal casting. Two types of pore shapes were observed. The first type included spherical pores with attached inclusions on internal surface. Figure 3 shows an example of a spherical micro-porosity with noticeable non-metallic inclusions at the boundaries of the micro-pore. The size of the micro-pore is around 5 μm. The second type was irregular shape pores which are shown in Figure 4. The size of the micro-pore is approximately 3 μm and different types of non-metallic inclusions were associated to the micro-pore.
1
Figure 3. Centrifugally cast steel sample (X=0.5) showing inclusions around the boundaries of micro-porosity in a steel matrix [22].
1
Average sizes of micro-porosity and non-metallic inclusions of the centrifugally cast tube at different thickness locations are given in Figure 5a. It appears that from inside diameter to outside diameter, the average sizes of both micro-porosity and non-metallic inclusions decrease, with the exception at the half thickness.
In this study we intentionally preserved the ID and OD regions to verify the distribution of micro-porosity and non-metallic inclusions. However, practically speaking, the ID and OD regions from centrifugal casting in industry are frequently removed during final machining to remove the large regions of inclusions and micro-porosity (see Figure 5b).
1
Figure 4. Centrifugally cast steel sample (X=0.5) exhibiting a ring of inclusions in the former boundary of a irregular shape micro-pore [22].
1
1
a) b)
Figure 5. Average sizes (a) and areas (b) covered by micro-porosity and non-metallic inclusions in the centrifugally cast tube at different thickness locations.
1
Investment cast ES: As-cast and HIPed samples were investigated. Metallographic images of as-cast and HIP treated ES samples are shown in Figure 6. Two groups of porosity with differing dimensions were observed in as-cast conditions: (i) macro-porosity with a pore diameter above 10 μm (up to 100 μm) on the top of the as-cast test bars and (ii) interdendritic micro-porosity with pore size below 10 μm in all casting regions. After HIP treatment, the large macro-porosity disappeared from the samples, while the micro-porosity still remained.
The micro-porosity size distribution of the as-cast and HIP treated ES samples is shown in Figure 7a. As-cast samples have wide size ranges of porosity, while the HIP treated samples exhibit a narrow size range between 0 to 5 μm. The largest porosity nearly all disappeared in the HIP treated samples, which is in agreement with the observation in Figure 6. Non-metallic inclusion size distribution of the as-cast and HIP treated ES samples is shown in Figure 7b. Non-metallic inclusions were not as affected by the HIP treatment as the porosity with large inclusions still evident. However, it does appear that the inclusions are slightly smaller after HIPing and the quantity of large size non-metallic inclusions decreased slightly.
1
a) b)
b) Figure 6. Macro- and micro porosity observed in the as-cast ES (a) and small size of micro-porosity after HIP treatment (b). Metallographic images, samples were etched with Nital.
1
1
a) b)
Figure 7. Histograms of the porosity and inclusion size distribution detected in ES for the as-cast and HIP treated samples: a) by number and b) by covered area.
1
Ingot and forged stainless steel: For the ingot casting, the concentration density and size distribution of non-metallic inclusions is shown in Figure 8. It reveals a concentration of inclusions within the range of 0 to 5 μm in the ingot. Inclusions with the diameter between 1 to 2 μm have the largest concentration density in the samples. Another notable characteristic of the forged ingot in this study is that no porosity defects were detected through the ASPEX analysis. This can be attributed to the hot forging process of the ingot. This result suggests that even a relatively small reduction ratio (2:1) in hot forging is an effective method of reducing the major porosity in cast products.
1
Figure 8. Histogram of non-metallic inclusion size distribution detected in the 316 L ingot. No micro-porosity was found.