Numerical Modelling of Lean Duplex Stainless Steel Hollow and Concrete-filled Columns of Square, L-, T- and +-shape Sections under Pure Axial Compression

No Thumbnail Available
Journal Title
Journal ISSN
Volume Title
Since the last two decades, there has been an accelerating interest in the use of stainless steel in the construction industry (Mann, 1993, Gardner 2005) owing to the benefits of stainless steels in terms of its good structural performance, aesthetic features and ease of construction of structural members. Generally, the austenitic stainless steel grades which contain ~ 8% - 11% nickel feature most prominently within the constructional industry. However, with increasing nickel prices there is an escalation in the demand for Lean Duplex Stainless Steel (LDSS) with low nickel content of ~ 1.5%, such as grade EN 1.4162 (Gardner, 2005b; Baddoo, 2008; EN 10088-4, 2009). LDSS, such as, grade EN 1.4162, has twice the mechanical strength of conventional austenitic and ferritic stainless steel and has a potential for use as main structural members (Gardner, 2005b). Even though there are several favourable characteristics involved in the use of LDSS material, often it may not be economical due to its relatively high material costs. Hence, a promising and innovative way to reduce the cost is through composite construction, where concrete is filled inside the stainless steel hollow sections (also known as Concrete-Filled Stainless Steel Tubular (CFSST) structures), which combines the advantages of both stainless steel and concrete, and thus providing not only an increase in the load-carrying capacity but also rapid construction (i.e. formwork is not required) (e.g. Uy and Patil, 2006; Zhao et al., 2010; Young and Ellobody, 2006; Lam and Gardner, 2008; Uy et al., 2011). In addition to the benefits gained in terms of economy and strength with the use of LDSS material and a composite construction, further benefits may be gained by combining it with the use of special-shaped sections, such as, L-, T-, and +-shape sections (Non-Rectangular Sections or NRSs). In engineering structures, NRSs are mainly used in reinforced concrete columns because of their convenient construction at beam-column joints, larger moment of inertia of the cross-sections that leads to much higher capacity for resisting lateral load along with the advantage of providing a flushed wall face, resulting in an enlarged usable and regular indoor floor space area. To the best of author’s knowledge, till date, there are no literatures available regarding studies on the structural behavior of LDSS hollow and concrete-filled stub and slender tubular columns with NRSs. Hence, the primary objective of the study undertaken in this research was to explore the structural behavior of LDSS hollow and concrete-filled tubular stub and slender columns with NRSs by comparing with a representative square section having same LDSS material cross-sectional area (i.e. equal LDSS material consumption adopted) under pure axial compressive loading through finite element (FE) analyses using general-purpose FE software package Abaqus (2009). In the first component of the present study, the effect of cross-sectional shapes and thickness of the LDSS tube on the strength and deformation capacities as well as the failure modes of the LDSS hollow tubular stub and slender columns are reported towards understanding the behavior of NRSs over the representative square section. For LDSS hollow tubular stub columns, it was found that, for LDSS tube thickness < 30 mm, NRSs showed higher strength capacities by about 36 % to 3 %, 72 % to 3 %, and 134 % to 9 % for L-, T- and +-shape section, respectively, over the representative square section, with the percentage difference in strength decreasing with increasing thicknesses of the LDSS tube. But for LDSS tube thickness ≥ 30 mm, the ratio of the strengths for the NRSs and the representative square section tends to 1.0. The axial deformation at ultimate load of the NRSs hollow tubular stub columns are higher compared to the epresentative square section, with +-shape section estimating a much higher deformation capacity, suggesting that there is a provision for achieving a better ductility in the use of NRSs. And for LDSS hollow tubular slender columns, a range of column lengths viz., 3.0 m – 10.0 m, thereby providing a range of non-dimensional member slenderness ( ) from 0.05 to 3.0, was considered to account for the variations in buckling strength with changes in the column lengths, cross-sectional shapes and LDSS tube thickness viz. 5.25 mm (Class 3 section) and 2.0 mm (Class 4 section). For Class 3 sections, for  ≤ 0.5, all the sections exhibit similar structural capacities. For 0.5 <  ≤ 2.0, there is a nearly linear variation in strengths (increasing trend for +HC, similar trend for THC and decreasing trend for LHC), with stabilization beyond  > 2.0. For > 2.0, Pu for +HC showed ~ 30% higher; THC showed similar strength; and LHC showed ~20% lower than the corresponding value for SHC, indicating that +HC has an improved ultimate strength for all the ranges of  . For Class 4 sections, in contrast to Class 3 sections, cross-sectional shape becomes increasingly significant with decreasing , however, the cross-sectional shape becomes insignificant for very high  as also the case for Class 3 sections. The average strength enhancement for the Class 4 NRSs as compared to the representative square section, for all  considered, are ~ 10 %, 50 %, and 90 % higher for LHC, THC, and +HC, respectively. The deformation capacities of the LDSS hollow tubular slender columns are similar irrespective of the cross-sectional shapes and class classification. Based on the FE studies, the applicability of the currently available design standards for stainless steel, such as the European specification, EN 1993-1-4 (2006), and the American specification, ASCE 8-02 (2002), for the design of LDSS hollow tubular stub and slender columns were assessed. Both the current design specifications are generally capable of predicting the LDSS hollow tubular column strengths. However, the European specification tends to be more conservative compared to the American specification. In the next part of the investigation, the benefits in the use of NRSs LDSS hollow tubular columns over those of the representative square section is extended to concrete-filled lean duplex slender stainless steel tubular (CFDSST) columns with an objective to explore and compare the structural behaviours such as the load and deformation capacities as well as to study the effect of the LDSS tube thickness and concrete compressive strength on the CFDSST columns. Similar to the approach followed in the analyses of LDSS hollow tubular columns, the concept of equal LDSS material cross-sectional area was followed (LDSS material being expensive compared to concrete), thereby, resulting in ~ 36 % reduction of concrete core area in the NRSs CFDSST columns as compared to the representative square CFDSST columns. The study suggested that NRSs CFDSST stub columns are more effective for specimens with a normal concrete strength (≤ 40 MPa). Also, NRSs CFDSST stub columns filled with high strength concrete can also be used with the advantages of providing lighter sections (i.e. 36 % reduction in the concrete core volume) with only 18 %, 15 % and 12 % reduction of strength compared to the representative square section for L-, T-, and +-shape sections, respectively, for 100 MPa concrete strength. The axial deformation at ultimate load in CFDSST stub columns decreases with increasing concrete strengths, but increases as the cross-sectional shape changes from square → L- → T- → +-shape. In fixed-ended CFDSST slender columns, for ~ < 0.5, +-shape showed a 7 % higher strength and L- and +-shape showed similar strengths compared to the representative square section. For 0.5 ≤  ≥ 1.5, a linear increasing trend in strength is seen with increasing . However, for > 0.5, T- and +- shape showed a 10 % and 25 % more strength and L-shape showing a 10 % lower strength compared to the representative square section. Thus, in fixed-ended CFDSST slender columns, change of cross-sectional shape from square section to NRSs is significant, especially for T-shape and +-shape sections, and can promote the application of thin-wall LDSS tube. The influence of the cross-sectional shapes on the axial deformation capacity becomes less significant with increasing  , but becomes increasingly significant with decreasing . Also, NRSs shows a higher axial deformation capacity for all considered compared to the representative square section, with +-shape section showing the highest axial deformation capacity. In comparison with the European specification, EN 1994-1-1 (2004), and the American specification, ANSI/AISC 360-05 (2005), for the design of fixed-ended CFDSST columns, FE strengths over predicts the design strengths, and also European specification tends to be more conservative compared to the American specification. For CFDSST slender columns, the design standards show over conservative results for square and L-shape sections and conservative for T-shape and +-shape sections. Also, the European specification gives a conservative estimation of strengths compared to the American specification for square and L-shape sections and vice versa in case of T-shape and +-shape sections.
Supervisor: Konjengbam Darunkumar Singh