【正文】 C, but with varying stress levels in three phases. In the ?rstphase, the stress was increased stepwise at an average rate of MPa/min, in the second phase, the stress was decreased at thesame rate, in the third phase。 C Huang et al. 2006。 C, at which point a phasechange takes place as discussed previously and the thermalstrain bees nearly constant up to 800176。 C. This variation can be attributed to many factors, primarily variable heating and strain/load rates during thetest. For example, when the heating rate of the stressed specimenis small, the specimen will be subjected to stress at elevated temtemperaturestressstrain curves of the tested specimen.The yield strength and elastic modulus constitutive relationships from the ASCE manual, Eurocode, and those proposed byPoh 2001 are also shown in Figs. 3 and 4, respectively. Thehightemperature reduction factors for the yield strength and elastic modulus of steel are also presented in the “Mechanical Properties” section of the Appendix for ASCE, EC3, and Poh.As is the case with test data, there is also a considerable variation in the constitutive models for yield strength and modulus ofelasticity. These variations in constitutive models are due to thelarge variation in the test data used to pile the respectiveconstitutive models. A review of the models shows that the Eurocode model predicts less reduction in yield strength of steel withtemperature as pared to the ASCE or Poh models. However,the Eurocode model provides a higher reduction in elastic modulus of steel with temperature as pared to the ASCE and Pohmodels as shown in Fig. 4. Also, the Eurocode model assumes noreduction in steel yield strength up to 400176。 C seen in Fig. 2. The variation between the testdata and the models shown in Fig. 2 is partly due to the fact thatthe majority of the existing data on speci?c heat originates fromstudies carried out on iron and nonstructural steel alloys. Additionally, the maximum temperature reached in these studies ofiron and nonstructural steel was below 750176。 ASCE 1992 , as well as published test data,were used to pile Figs. 1 and 2 Rempe and Knudson 2008。 Structural steel.IntroductionStructural ?re safety is one of the primary considerations in thedesign of highrise buildings where steel is often the material ofchoice for structural members. At present, structural ?re safety?re resistance of steel members is generally achieved throughprescriptive approaches which are based on either standard ?reresistance tests or empirical calculation methods. These prescriptive based approaches have major drawbacks and do not providea rational or realistic ?re resistance assessment. The recent movetoward performancebased ?re design has increased the focus onthe use of puter simulations for evaluating ?re resistance ofstructural members. Knowledge of hightemperature properties ofsteel is critical for evaluating ?re resistance using numerical models.Steel has excellent strength properties at ambient temperature,however, like other materials。 Mahmud Dwaikat2。 Fire resistance。ASCE,ISSN 08991561/2010/5423–434/$.Research Signi?canceResearch on hightemperature material properties for use in putational models has not kept pace with the development ofcalculation techniques for predicting the ?re response of steelstructures. In this paper, the available hightemperature constitu. Candidate, Civil and Environmental Engineering, MichiganFig. 1. Thermal conductivity of steel as predicted by different models and as measured in different test programsFig. 2. Speci?c heat of steel as predicted by different models and asmeasured in different teststive relationships for steel are pared with each other, andremendations are made on the applicability of these relationsfor structural ?re safety design. This will facilitate the use ofrational approaches to ?re engineering of steel structures, andpromote performancebased ?re safety design.HighTemperature Properties: State of the ArtMaterial Properties for Fire Safety DesignThe response of steel structures exposure to ?re is governed by:a thermal。 C, as is seen in Fig. 2. In general, thespeci?c heat of steel increases with an increase in temperaturewith a large spike occurring around 750176。 Anderbergconstant stress and temperature , can in?uence the resulting1988 . Despite the fact that strain rate has a signi?cant effect onthe test results, a large amount of test data on conventional steel ispublished without the information on strain rates. Therefore, teststandards are still concerned with de?ning limits for strain rates intests Outinen 2007。 Cooke 1988。 C see Fig. 7 .Fig. 6. De?nition of yield point and proportionality limit in ASCE,EC3, and Poh 2001Fig. 7. Thermal strain of steel as predicted by different models andas measured in different tests