|
PQ-CGI Inmold(英文）
|
Introduction
to Thermal Analysis of Metals When liquid metal cools off, the temperature is reduced, which causes energy to be released. Temperature represents the total amount of thermal energy in a melt and is related to the kinetic energy of molecules. When a melt cools down the thermal agitation of its molecules is reduced. This energy is expressed as specific heat e.g. as kJ/ Kg and degree C. When the liquid reaches a temperature called liquidus the bond between the atoms becomes more rigid on a macro scale level. More energy is then released until the liquid is transformed into solid state. The energy released at this stage is called latent heat of fusion, fusion enthalpy or just latent heat for short. The temperature stays constant until the transformation is completed. Latent heat is measured as kJ/Kg. Solidus is the temperature where the metal or a precipitated phase is completely solid. Liquidus and solidus coincide for a pure metal. An alloy usually solidifies with a solidification interval. Thermal analysis is based on recording temperatures at certain time intervals during the solidification process. Cooling curves can thereby be constructed and used to analyze and classify an alloy. A cooling curve is thus a plot of the temperature as the function of time for a sample of an alloy poured into a standardized mold with a thermocouple, usually positioned in the center. Arrest temperatures such as liquidus and solidus in a cooling curve as well as cooling rates during various phases of the solidification can be used as metallurgical attributes to classify a melt and to correlate it to the behavior when poured in a mould.
Fig 1 shows a phase diagram for two metals soluble in each other,, A and B. Fig 2 shows the corresponding cooling curve for an alloy with 50% A and 50% B. The first solid phase is precipitated when the liquidus temperature is reached . It will contain about 90% A and 10% B. Latent heat is released from this moment on., which explains why the cooling curve is less steep until solidus is reached. Fig 3 shows the cooling curve for a melt consisting of 100% B. In this case the temperature will drop until the liquidus temperature is reached and stay at that temperature until all is solid. Commercial alloys consist of a mixture of two or more metals and in some cases metalloids. Metals are often soluble in each other or form compounds or phases during the solidification process. Different compounds can also be formed between metalloids and gases present in the liquid. In such cases the alloy usually solidifies over a temperature range and different phases are precipitated from the liquid. The released latent heat is different for the different phases, which will influence the slope of the cooling curve. In some cases the temperature stays at a certain constant value until all the liquid has been transformed into solid state. This type of solidification where two or more phases are precipitated at a constant temperature is called eutectic freezing. It is found in many commercial alloys such as cast iron and aluminum silicon alloys. The progressing solidification from liquid state is studied in thermal analysis (TA) for metals. However, in some cases it is valuable to study a sample when it is heated from solid state to liquid state. Special standardized sensors or test cups with thermocouples are commercially available (e.g.Quik-Cups). Cooling curves are often constructed with temperature on the Y-axis and time on the X-axis. The cooling curve can be derivated, which makes it easier to trace changes in the solidification behavior. A derivated curve shows the rate of change in temperature over time - thus the Y-axis shows temperature / time unit. The 2:nd derivative can be useful in some cases. Another method of tracing changes in the solidification behavior is to compare the cooling curve with a standardized sample. This technique is called DTA, differential thermal analysis. In practice a base line curve is constructed using the cooling rates before liquidus and after solidus. The differences between the real curve and the constructed base line curve can be correlated to the latent heat released by the different phases in the melt. Why is chemical analysis not sufficient for cast iron alloys? Cast iron alloys are very complex and several of the mechanisms behind crystallization and growth of austenite and graphite are only partially understood. While alloying elements may alter the physical properties of cast iron, the final properties and especially the behavior of the alloy when poured in a mold can not be fully predicted and controlled by chemical analysis alone. The practical foundryman experiences this daily in the form of unexpected casting defects, low yields and variations in physical properties. The amount and precipitation patterns of austenite and graphite influence the behavior of gray, ductile and compacted graphite iron during mold filling and solidification. The main variables involved are:
Among these variables only the composition can be controlled by chemical analysis. In practice only a limited number of elements is being tested, tramp elements are often not analyzed, which further complicates the situation. A spectrometer analysis does not tell us anything about compounds such as SiO2, FeO, silicates, other oxides, particle sizes, dissolved gases and others that have a profound effect on the solidification. Nor does it tell us anything about what actually happens during solidification, such as amount and type of precipitated graphite. Alloys with identical chemistry can behave completely differently depending on variations in the other variables. The liquidus temperature and the amount of primary austenite might for example vary as much as 10° C for the same analysis, as a result of different melting conditions. This will of course have an impact on the risk for defects, e.g. macro shrinkages. The amount of eutectic graphite especially at the end of freezing might vary depending on the nucleation properties and may result in problems with micro shrinkage and porosity. Bad nucleation properties might result in chill, etc. It can be concluded that chemical analysis alone does not qualify as an efficient process control method for melting and treatment of cast iron alloys. The development of the NovaCast process control method originally started as a project between NovaCast and the Swedish Foundry Association. Its purpose was to develop an efficient control method for cast iron based on thermal analysis. At that time most commercial thermal analysis systems were based on using tellurium-coated test cups to make the iron solidify "white". By doing this, liquidus and especially the "white" eutectic temperature were easy to trace. Regression analysis proved that the carbon equivalent, carbon and silicon could be estimated using these two temperatures. The method works fine but it does not tell us anything about the most important feature of cast iron, namely how the carbon precipitates into graphite. Our thermal analysis methods are therefore based on the usage of test cups without any tellurium, allowing the sample to solidify "gray", i.e. according to the stable phase diagram. The method is called ATAS, an acronym for Adaptive Thermal Analysis System. It is used for gray and ductile iron. The reason for "adaptive" is that during the investigation we found that an interpretation system for cooling curves must be adapted to various conditions in each foundry. The ATAS system uses artificial intelligence methods to interpret the cooling curves individually for each alloy. A more recent development is our system for production of compacted graphite iron. This system is called PQ-CGI (Prime Quality CGI) and makes use of a more specialized type of thermal analysis to control the precipitation of graphite. Why use thermal analysis for process control? The main purpose of metallurgical process control is to solve three different types of problems that make the difference between success and failure. The first problem is to avoid casting defects. In a normal foundry about 30-40% of all defects have a metallurgical cause or are influenced by the metallurgical status of the iron. The main types of defects are:
The second problem is achieving a high yield. The yield depends on the safety margins that the methods engineer has to use when designing the gating and risering system. If the metallurgical process control is insufficient, which it usually is if the foundry relies on chemistry alone, the engineer finds by trial and error that he has to use large margins in order to cope with the worst conditions. A large variance in the thermal parameters makes it necessary to use these large margins. The result is oversized risering systems and low yield. The third problem is to reduce costs. Unnecessary scrap is expensive – the scrap costs are usually about 70% of the manufacturing costs. Scrap also reduces production capacity. Low yield means higher consumption of energy and refractory as well as higher consumption of inoculants and Mg-alloys. Insufficient knowledge about the oxygen and nucleation status of a melt will result in too high additions of inoculants and Mg-alloys. Thermal analysis using "gray" samples is a suitable method for revealing how an alloy behaves during solidification. Temperature and time recordings displayed as cooling curves show the thermodynamic behavior of the alloy. It is a function of several complex events, mainly of nucleation properties, precipitation of solid phases and changes in heat conductivity during solidification. The purpose of our thermal analysis systems is to open a new metallurgical "window" into the process. By using thermal parameters as control features, ATAS allows foundrymen to optimize and control the complex cast iron processes. |
|
|---|---|---|
| Copyright © NovaCast AB 2002 |
