See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/229860995

The use of stainless steel in structures

Article in Progress in Structural Engineering and Materials · April 2005

DOI: 10.1002/pse.190

CITATIONS READS

358 12,248

1 author:

Leroy Gardner
Imperial College London
465 PUBLICATIONS 17,826 CITATIONS

SEE PROFILE

All content following this page was uploaded by Leroy Gardner on 14 February 2018.

The user has requested enhancement of the downloaded file.

 Gardner, L. (2005). The use of stainless steel in structures. Progress in Structural Engineering
and Materials. 7(2), 45-55.

The use of stainless steel in structures

L. Gardner BEng, MSc, PhD, DIC, CEng, MICE

Imperial College London

Summary

The past fifteen years have seen the introduction or major revision of structural stainless steel design
codes throughout the world, and at the same time, interest in the use of stainless steel in construction
has been accelerating. Historically the high initial material cost of stainless steel has limited its use
primarily to specialist and prestige applications. However, the emergence of design codes, a better
awareness of the additional benefits of stainless steel and a transition towards sustainability are
bringing more widespread use into conventional structures. Although a number of similarities
between stainless steel and ordinary carbon steel exist, there is sufficient diversity in their physical
properties to require separate treatment in structural design. In addition to the straightforward
differences in basic material properties (such as Young’s modulus and yield strength), further
fundamental differences exist, such as the nature of the stress-strain curve and the material’s
response to cold-work and elevated temperatures; these have implications at ultimate, serviceability
and fire limit states. This paper describes the use of stainless steel as a structural material, discusses
current structural design provisions, reviews recent research activities and highlights the important
findings and developments.

Keywords

Stainless steel, structural engineering, Eurocode 3, corrosion resistance, material modelling, cross-

section design, member design

Introduction

Stainless steel has traditionally been regarded as an extravagant solution to structural engineering

problems. Consequently, the use of stainless steel as a primary structural material for conventional

construction remains rather limited. Previously, coupled with the high initial material costs, there

have been a number of disincentives to adopt stainless steel, including limited structural design

guidance, restricted section availability (and no standardisation) and a lack of understanding of the

additional benefits of stainless steel amongst structural engineers. Since the last significant general

review of the use of stainless steel in structures [1], carried out in 1993, there have been considerable

advancements. These are described throughout this paper.

Historical background

For the past eighty years the longevity and aesthetic appeal of stainless steel has inspired architects

and designers alike to use the material in both practical and imaginative ways. Stainless steel was

first introduced in 1912-1913 (by Brearley in the UK and Maurer and Strauss in Germany). Brearley

referred to the material as ‘rustless steel’ and it was a cutlery manager, Ernest Stuart, who

popularised the term ‘stainless’. Stainless steel is now used as a general expression to describe

corrosion resistant iron alloys that contain a minimum of 10.5% chromium. A thorough account of

the initiation and growth of stainless steel production is given by Truman [2].

Chemical composition and material grades

As with carbon steel, there is a wide variety of grades of stainless steel, generated through variation

in chemical composition and heat treatment. These may be classified into five main groups

according to their metallurgical structure, namely austenitic, ferritic, duplex (austenitic-ferritic),

matensitic and precipitation hardening. In addition to the minimum of 10.5% chromium (Cr)

required to give stainless steel its corrosion resistance, a number of other alloying elements may be

2
present. These include carbon (C), nickel (Ni), manganese (Mn), molybdenum (Mo), copper (Cu),

silicon (Si), sulphur (S), phosphorus (P) and nitrogen (N). The chemical compositions (% by mass

of alloying elements) associated with the different grades are defined in [3].

The most common grades for structural applications are the austenitic and duplex grades. These will

be discussed in more detail. Austenitic stainless steels typically contain 17-18% chromium and 8-

11% nickel, offer very good corrosion resistance and have an austenitic microstructure. Duplex

stainless steels typically contain 22-23% chromium and 4-5% nickel and have a mixed austenitic-

ferritic microstructure. The duplex stainless steels offer higher strength, wear resistance and

generally corrosion resistance than the austenitics, but at greater expense.

In addition to the high number of grades of stainless steel, there are also a number of different

stainless steel designation systems. The system adopted in the remainder of this paper will be that

given in the European material Standard, EN 10088-1 [3]. For each standard stainless steel grade,

EN 10088-1 defines a steel name and a steel number. It is the steel number that will be referred to

herein. Equivalent designations, including those to the German (DIN) and the US (AISI) systems

are provided in [4].

Chemical compositions for three grades of stainless steel commonly adopted in structures, two

austenitic, EN 1.4301 and EN 1.4401 and one duplex, EN 1.4462 are provided in Table 1.

Equivalent designations for grades EN 1.4301, EN 1.4401, EN 1.4462 according to the popular AISI

system are AISI 304, 316 and 2205, respectively. Further information relating to stainless steel

material properties and material selection is available [1, 5, 6].

3
Structural products

Stainless steel product forms include plate, sheet, strip, tube, bar, cold-formed and hot-rolled

structural sections, castings, fasteners and fixings. For structural members, the most commonly used

products are cold-formed sections, predominantly because these are the most readily available,

require relatively low investment to achieve production capabilities, and are suitable for light

structural applications with high structural (and material) efficiency. Hot-rolled and built-up

sections are relatively scarce, though structural design guidance is available, as described later in this

paper.

Essentially cold-formed sections may be formed from flat sheet either by press-braking or roll-

forming; press-braking is generally limited to simple shapes and low production levels and is often

used for prototyping, whereas roll-forming is suitable for larger quantities. Due to the material’s

response to cold-work, the strength of cold-formed structural stainless steel sections can be

considerably enhanced during the forming process. These enhancements may arise during the

production of the flat sheet or during the formation of the final cross-section. Strength

enhancements in the sheet material may be utilised in design, with material strengths provided on the

basis of the level of cold-work that the sheet receives. Strength enhancements during the cold-

forming of the cross-sections are not included in existing design methods because there are currently

no tools to determine the level and distribution of these enhancements for the particular process

routes. Progress on the development of such tools is underway at Imperial College London.

There are also currently no standard sizes for stainless steel sections, with sections often made to

order. However, most suppliers stock commonly requested section sizes, and geometric properties

and member capacities for such sections have been tabulated based on British [7, 8] and European

[9] rules.

4
Stress-strain behaviour

The stress-strain behaviour of stainless steel is fundamentally different from that of carbon steel, as

indicated by Figure 1. Whereas carbon steel exhibits a sharply defined yield point, followed by a

flat yield plateau and then some strain hardening, stainless steel displays a rounded stress-strain

curve, with no sharp yield point, considerable strain hardening and high ductility. For the common

austenitic stainless steel grades, the ductility (strain at fracture) is approximately 40-60%, as

compared to around 20-30% for carbon steel.

The absence of a sharply defined yield point necessitates the definition of an ‘equivalent’ yield point

for design; the generally accepted approach is to adopt the stress at 0.2% plastic strain (i.e. the 0.2%

proof stress), as shown in Figure 2.

The degree of roundedness of the stress-strain curve of stainless steel depends on the grade (i.e. the

chemical composition and heat treatment) and the level of cold-work that the material has been

subjected to. Annealed material has a sharper yield point followed by approximately linear strain

hardening, whereas cold-worked material displays higher strength and more rounded stress-strain

behaviour (see Figure 3). Stainless steel also exhibits anisotropy and non-symmetry of stress-strain

behaviour in tension and compression (defining four stress-strain curves). Both of these effects

become more pronounced as the level of cold-work increases. For example, for the common

austenitic grades in the annealed condition, variation in 0.2% proof stress between the four stress-

strain curves is less than 5%, but for cold-worked material this can increase to around 20%. Most

material Standards provide tension and compression stress-strain properties in both the longitudinal

and transverse directions, thus enabling the effects of anisotropy and non-symmetry to be recognised

in design. However, numerical studies have indicated that the behaviour of common structural

5
elements (compressed plates) is relatively insensitive to material anisotropy [10] and, for simplicity

or conservatism under complex loading systems, the minimum material properties can be adopted.

For both design and numerical simulation, carbon steel material behaviour is often represented with

a bi-linear (elastic, perfectly-plastic) relationship. Although carbon steel does display some

deviation for this idealised model (e.g. strain hardening beyond approximately 1.5% strain), for the

majority of applications the model is acceptable. However, for stainless steel the bi-linear model is

less satisfactory, and indeed direct use would generally lead to overly conservative design at ultimate

limit state and under-prediction of deflections at serviceability limit state. Hence, a number of more

accurate material models have been proposed.

The most commonly adopted expression, originally proposed by Ramberg and Osgood [11] and

modified by Hill [12], is given by Equation 1.

n
   
  0.002  (1)
E0   0.2 

where  and  are engineering stress and strain, respectively, E0 is the material Young’s modulus,

0.2 is the material 0.2% proof stress and n is a strain hardening exponent. A review of other

material models and investigation into their applicability to stainless steels was made in 2002 [13].

Deviation of the basic Ramberg-Osgood expression from experimental stress-strain data at higher

strains has been observed by a number of authors [14, 15, 16]. A modification to the basic Ramberg-

Osgood expression was proposed by Mirambell and Real [14] whereby the single curve defined by

Equation 1 was replaced by a two-stage model, incorporating two Ramberg-Osgood curves. In their

proposed model, the basic Ramberg-Osgood expression was adopted up to the 0.2% proof stress.

6
Beyond the 0.2% proof stress and up to ultimate stress, a second Ramberg-Osgood curve was used.

The origin of the second curve was defined at the 0.2% proof stress, and continuity of magnitude and

gradient was ensured at the transition point. Further work on this two-stage model was conducted

[16] where the additional parameters required by the two-stage model were described in terms of the

original Ramberg-Osgood parameters (0.2, E0 and n) following analysis of stainless steel stress-

strain data.

Gardner and Nethercot [17] recognized the value of the Mirambell and Real proposal, but noted that

use of the ultimate stress, u in the second phase of the model (between 0.2 and u) has two

drawbacks. Firstly, since the strain at u is far higher than those strains concurrent with general

structural response, greater deviation between measured and modelled material behaviour results

than if a lower strain was used. Secondly, and more importantly, the model is not applicable to

compressive stress-strain behaviour, since there is no ultimate stress in compression, due to the

absence of the necking phenomenon.

Use of the 1% proof stress in place of the ultimate stress was therefore proposed [17], leading to

Equation 2. Equation 1 remains applicable for stresses up to 0.2.

n 0.2,1.0
   0.2          0.2 
   0.008  1.0 0.2      t 0.2 (  0.2) (2)
E 0.2  E 0.2   1.0   0.2 

where n’0.2, 1.0 is a strain hardening coefficient representing a curve that passes through 0.2 and 1.0.

Equation 2 provides excellent agreement with experimental stress-strain data, both in compression

and tension, up to strains of approximately 10%, and is therefore appropriate for design and

7
numerical modelling. Where higher strains are expected, the original proposal of [14] is most

appropriate.

Corner material properties

The stress-strain properties of the corner regions of cold-formed stainless steel cross-sections differ

from those of the flat regions due to the material’s response to deformation. Enhancements in 0.2%

proof strengths of between 20% and 100% have been observed, accompanied by a corresponding

loss in ductility.

A careful review of the corner properties of cold-formed stainless steel sections has been conducted

[18]. On the basis of test results from press-braked [19, 20] and roll-formed [13, 21, 22] sections,

models for the prediction of corner strength enhancements were proposed.

Corrosion resistance

An important feature of stainless steel is its corrosion resistance, enabling its application,

unprotected, in a wide range of environments. The corrosion resistance of stainless steel is primarily

attributable to its chromium content, though other alloying elements (including nickel and

molybdenum) also contribute. On exposure to oxygen, the chromium in the steel reacts to form a

thin, chromium-rich, oxide film over the surface of the material. It is the presence of this passive

film that provides the resistance to corrosion. There are a number of different forms of corrosion

including, general (uniform), abrasive, pitting, crevice, intergranular and galvanic corrosion. An

account of the susceptibility of stainless steel to each of these is given in [5]. It should be noted that

certain environments, such as those containing high concentrations of chlorides, can be extremely

detrimental to the corrosion resistance and resistance to stress-corrosion cracking of stainless steels.

In such cases, care must be taken to select suitable grades and it is recommended that specialist

advice is sought. Detailed guidance may be found in [23].

8
Factors influencing material choice

In addition to structural performance, there are a number of other issues that influence the choice of

material in construction, and although stainless steel has a high initial material cost, it possesses

additional beneficial properties that may provide cost savings elsewhere. Factors influencing

stainless steel as a material choice for use in structures is discussed in this section.

Cost

The initial cost of stainless steel structural products is about four times that of equivalent carbon

steel products, owing largely to the expense of the alloying elements and the relatively low levels of

production. Although this deficit is likely to reduce with increased production of structural stainless

steel components and more efficient design rules, a considerable disparity remains. However, there

are a number of other issues that should be included for a ‘true’ comparison of costs. These are

discussed in the following sub-sections.

Aesthetics

Historically, the aesthetics of stainless steel has been an important factor in its specification for

structural applications. Consequently, many existing examples of stainless steel structures [6]

display a high level of exposed structural members, commonly of tubular cross-section, and are often

of a prestigious or landmark nature. Its appeal is principally due to the surface finish and its ability

to retain its appearance without deterioration over time. The appearance and longevity of the

material is exemplified by the upper facade of the Chrysler Building in New York (completed in

1936), which despite the aggressive atmosphere and proximity to the ocean, remains bright and

clean.

9
Durability

The corrosion resistance of stainless steels make them one of the most durable families of

construction materials. With appropriate selection of material grade and structural design and only a

minimum of maintenance, design lives well in excess of 100 years can be achieved. Also, with no

requirement for protective coatings against corrosion, there are savings in terms of economy, weight

and reduction in environmental impact.

Elevated temperature behaviour

At elevated temperatures, stainless steel offers better retention of strength and stiffness than carbon

steel, due to the beneficial effects of the alloying elements. A comparison of the elevated

temperature performance of austenitic stainless steel and carbon steel is given in Figures 4 and 5,

where fy() and E() refer to the nominal yield strength and stiffness, respectively, at temperature ,

and fy(20°C) and E(20°C) refer to the nominal yield strength and stiffness, respectively, at room

temperature. The data for these graphs have been taken from the Euro Inox Design Manual for

Structural Stainless Steel [5], though were originally generated by [24], and are expected to be

included in EN 1993-1-2 (when it is converted from its current ENV status). The graphs

demonstrate the superior elevated temperature material properties of stainless steel, particularly in

the important temperature region (~ 600 to 800°C) that generally corresponds to approximately 30

minutes fire resistance; for example, at 800°C the strength retention, fy()/fy(20°C) of stainless steel

is almost 4 times that of carbon steel and the stiffness retention, E()/E(20°C) is 7 times that of

carbon steel. A number of studies are currently underway to further investigate the behaviour of

stainless steel structures in fire.

10
It may also be observed from Figure 4 that at low temperatures stainless steel has a strength

reduction factor greater than unity. This is possible, because at fire limit state where higher

deformations may be tolerated, ‘yield’ strength is defined at 2% strain (rather than the 0.2% proof

strain adopted at ambient temperatures), and thus, the substantial strain hardening that stainless steel

exhibits is utilised.

The resistance of structural stainless steel members in fire has been investigated experimentally [25,

26] and shown to perform better than equivalent carbon steel members. This enhanced fire

resistance may reduce or even eliminate the need for protective fire coatings to be applied to

structural members [25].

In addition to the important differences in the ability of stainless steel and carbon steel to retain

strength and stiffness at elevated temperatures, there is also variation in other thermal properties,

including specific heat, emissivity and thermal expansion. The specific heat of stainless steel is

approximately 500 J/kgK as compared to carbon steel which has a value of approximately 600J/kgK.

The lower the specific heat of a material, the more rapidly it tends to heat up. However, stainless

steel also has a lower emissivity than carbon steel (approximately 0.4 compared to 0.5). Emissivity

is a dimensionless property that ranges between zero and unity, where zero corresponds to all

radiation being reflected by a surface and unity corresponds to all radiation being absorbed, and

therefore, the lower the emissivity, the more slowly the material heats up. Studies [26] have shown

that these two factors (lower specific heat and lower emissivity) appear essentially to cancel each

other out, resulting in stainless steel heating up at a very similar rate to carbon steel. The coefficient

of thermal expansion of stainless steel is approximately 50% larger than that of carbon steel, which

may result in additional loads being induced into structural elements due to restraint from other

(stiffer) parts of the structure.

11
Ductility and impact resistance

Stainless steels, particularly the austenitic grades, offer very high ductility and impact resistance. It

is therefore particularly suited to applications where ductility and impact resistance are important,

such as offshore structures, crash barriers and structures susceptible to blast loading [27], and has

already been applied to railway carriage construction. A study into the suitability of adopting

stainless steel for structural frames in seismic regions has also been carried out [28], where it was

concluded that stainless steel is a viable alternative to carbon steel, but further investigation is

required. The level of ductility of stainless steel depends upon the material composition and heat

treatment and on the degree of cold-work that the section has been subjected to, with reduced

ductility for increasing cold-work.

Re-use and recycling

The construction industry is a major producer of waste material and a major consumer of void

(landfill) space [29]. Increasing emphasis is now being placed on the minimisation of construction

waste, with financial incentives such as the Landfill and Aggregates Levies operating in the UK.

Stainless steel possesses a combination of high residual value (due to the alloy content) and excellent

durability, lending itself to widespread re-use and recycling, bringing practical, financial and

environmental advantages. Re-melting scrap using the electric arc process is the dominant means of

production of stainless steel.

Whole-life costing

Cost comparisons made on the basis of the initial material expense of structural components do not

reflect the true cost implications of a chosen structural material. A more appropriate analysis would

12
include the additional immediate costs, such as corrosion protection and fire protection, and the

longer term costs associated with maintenance and decommissioning.

A study conducted by the Steel Construction Institute [30] compared life-cycle costs of offshore

structures made from carbon steel, aluminium and stainless steel. Consideration was given to initial

material costs, maintenance and corrosion, production losses during maintenance, fire resistance and

other potential cost savings. The study concluded that the use of both aluminium and stainless steel

result in significant life-cycle cost savings as compared to carbon steel. Furthermore the combined

economic and fire safety benefits that stainless steels offered were unequalled by the alternative

materials.

Structural design

General

Historically stainless steel design rules have been based on assumed analogies with carbon steel

behaviour, with modifications made where necessary to fit in with test results. More recently greater

recognition has been given to the particular properties that stainless steel exhibits, allowing the

generation of more efficient structural design rules. An overview of the provisions of the principal

stainless steel design codes (European, US and Australia/New Zealand) was reported by Baddoo

[31], whilst a more detailed comparison of their codified formulations with test results has also be

prepared [32].

Design standards

The earliest dedicated stainless steel structural design Standard was published by the American Iron

and Steel Institute (AISI) in 1968 as the Specification for the Design of Light Gauge Cold-formed

13
Stainless Steel Structural Members. The design rules were based primarily on the work carried out

at Cornell University [33]. With an improved understanding of the structural behaviour of stainless

steel and an increased availability of test results [34], a revised version of the Standard was

published in 1974 [35]. Further research enabled the development of the American Society of Civil

Engineers (ASCE) structural stainless steel design Standard, first published in 1991 and more

recently in 2002 [36], which effectively superseded the AISI Standard in North America.

Background and commentary to the ASCE Standards were reported [37].

In 1988, a joint industry project, managed by the Steel Construction Institute, was undertaken to

develop recommendations for the design of stainless steel structures in Europe. The design rules

were published by Euro Inox in 1994 as the Design Manual for Structural Stainless Steel [38].

Derived from the same material, the Concise Guide to the Structural Design of Stainless Steel

(aimed at the UK market and utilising BS 5950 terminology), was published by the Steel

Construction Institute [39]. In 2002, a second edition of the Euro Inox design manual was released

[5].

The current European design (pre)standard for stainless steel structures, ENV 1993-1-4 [4] was

introduced in 1996. ENV 1993-1-4 forms Part 1.4 of Eurocode 3: Design of steel structures, and

contains supplementary rules for stainless steels. Its present status is that of a European prestandard,

with conversion to a full European Standard (EN 1993-1-4) currently underway.

In 1995, the Japanese stainless steel structural design Standard was issued [40]; it is only available in

Japanese and is focussed on the design of fabricated (welded) sections. Based largely on the

Canadian design Standard for cold-formed carbon steel structures, the South African structural

stainless steel Standard was published in 1997 [41]. Most recently, in 2001, the Australia/New

14
Zealand design Standard for cold-formed stainless steel structures [42] was issued. Background to

the development of the Australia/New Zealand design Standard has also been described [43].

The following sub-sections outline and review various aspects of the structural design provisions of

the European (EN(V) 1993-1-4), US (SEI/ASCE 8-02) and Australia/New Zealand (AS/NZS 4673)

Standards, with an emphasis on the European Standard. In general, the design expressions are not

provided as these are readily accessible in the appropriate design Standards.

Material strength and stiffness

The material properties of stainless steel vary with chemical composition and heat treatment (i.e.

grade), product type, level of cold-worked, material thickness, direction of rolling (i.e. longitudinal

or transverse), and direction of loading (i.e. tension or compression). Although some material

Standards simplify matters by grouping cases with similar properties, there remains a wide range of

values. The key material properties for two commonly adopted austenitic grades (EN 1.4301 and

1.4401), as provided by the European [4, 44], US [36] and Australia/New Zealand [42] Standards,

have been summarised [31].

This paragraph gives an overview of the properties (modulus of elasticity E0 and minimum specified

yield strength fy, taken as the 0.2% proof strength 0.2) for grades EN 1.4301 and 1.4401 in the

annealed condition. For the modulus of elasticity, the European Standard EN 10088-2 [44] gives a

value of 200000 N/mm2, whilst the US and Australia/New Zealand Standards provide a lower value

of 193100 N/mm2. For minimum yield strength, the European Standard specifies 230 N/mm2 (for

cold-rolled strip of thickness less than 6 mm) and 210 N/mm2 (for hot-rolled strip of thickness less

than 12 mm) for grade EN 1.4301, and 240 N/mm2 (for cold-rolled strip of thickness less than 6 mm)

and 220 N/mm2 (for hot-rolled strip of thickness less than 12 mm) for grade EN 1.4401, with no

15
variation for anisotropy or non-symmetry of stress-strain response. The US and Australia/New

Zealand Standards do account for these effects, with minimum yield strengths for the two grades

ranging between 193.1 N/mm2 and 206.9 N/mm2, depending on the material orientation and

direction of loading.

Cross-section design

The design of stainless steel cross-sections follows the familiar carbon steel approach, utilising the

concepts of cross-section classification and, for slender elements susceptible to local buckling, the

effective width method. For the calculation of effective widths, the European guidance given in

ENV 1993-1-4 was aligned with that given in the equivalent carbon steel parts (Part 1.1 for hot-

rolled sections and Part 1.3 for cold-formed sections). The US Specification adopted the ENV 1993-

1-4 method for stiffened elements and that provided by the AISI Specification for cold-formed

carbon steel for outstand elements [45]. The Australia/New Zealand Standard follows the US

provisions.

The Euro Inox Design Manual for Structural Stainless Steel [5] recommends a more conservative

approach for the calculation of effective widths. The effective width reduction factors take account

of whether the elements are internal or external, and whether the element is welded or cold-formed.

It is envisaged that the more conservative approach of the Euro Inox Design Manual will be adopted

in Eurocode 3 Part 1.4 upon conversion from the ENV status. None of the current codified methods

make explicit account for the effect of gradual yielding in the calculation of effective widths.

16
Member design

The behaviour of stainless steel members differs from that of carbon steel members due to the

gradual yielding nature of the material stress-strain curve and variation in other characteristics such

as the level of geometric imperfections and residual stresses. Geometric imperfections are generally

lower in stainless steel products than equivalent carbon steel products because of tighter controls in

the production process to limit the adverse effects of out-of-flatness on aesthetics. Initial analyses of

measurements of geometric imperfections in structural stainless steel hollow sections [46] have

supported this assertion. Residual stresses have been measured in cold-formed sections [21] and

welded sections [47] and found to be of similar magnitude to those observed in carbon steel

members.

The European provisions for stainless steel member design mirror those for carbon steel. The basic

formulations are the same, though differences exist in the selection of the imperfection parameter

and the non-dimensional limiting slenderness  0 (the plateau length), both of which effectively

define the shape of the buckling curves. The buckling curves have been calibrated against all

available stainless steel test data to provide a suitably conservative fit for design purposes. For

simplicity (to avoid the need for iteration) and consistency with the carbon steel approach, no

explicit allowance is made for the effect of gradual material yielding in the member buckling

formulations. In addition to providing for the flexural buckling of columns and the lateral-torsional

buckling of beams, guidance is also given for design against torsional and torsional-flexural buckling

of compression members (by reference to Eurocode 3 Part 1.3). Since stainless steel structural

members are generally cold-formed (i.e. relatively thin material) and are often open-sections (i.e.

low torsional stiffness), susceptibility to these modes of failure should be checked.

17
The US provisions for stainless steel member design follow the AISI recommendations for carbon

steel [45], except, to account for the non-linear (gradual yielding) stress-strain response, the tangent

modulus Et is used in place of the usual initial modulus E0 in the buckling formulations.

Additionally, for buckling modes with torsional components (lateral torsional buckling of beams and

torsional and torsional-flexural buckling of columns), the initial shear modulus G0 is replaced by the

tangent shear modulus Gt. The non-linear stress-strain behaviour is described by the basic Ramberg-

Osgood expression (Equation 1), and gradient of this curve at any stress level (i.e. the tangent

modulus) may be determined through Equation 3.

 0.2 E 0
Et  n 1
(3)
  
 0.2  0.002nE 0  
  0.2 

where n is the Ramberg-Osgood strain hardening exponent, and E0 and Et may be replaced by G0

and Gt respectively in shear. Since the tangent modulus is dependent upon the buckling stress level,

the US member design procedure is necessarily iterative.

The Australia/New Zealand Standard adopts the US recommendations for the design of beams and

columns against member buckling. It also provides an additional method for the design of columns.

The method utilises the familiar Perry formulation and adopts sophisticated imperfection factors

derived on the basis of numerical modelling of imperfect columns.

A comparison between the flexural buckling curves of the European, US and Australia/New Zealand

design Standards is shown in Figure 6 for fy = 210 N/mm2 and E = 200000 N/mm2, where  is the

buckling reduction factor and member slenderness is given as length, L divided by radius of

gyration, i. For the US and Australia/ New Zealand buckling expressions, the Ramberg-Osgood
18
parameter n has been taken as 4.0 (which applies to grades 1.4301 and 1.4401 in longitudinal

compression). The comparison shows that the Eurocode curve for welded sections gives the most

severe reductions, but generally indicates relatively small deviation between the design Standards.

For stainless steel members subjected to combined axial load plus bending (beam-columns), each of

the Standards under consideration adopts their respective approaches for carbon steel, with the

Australia/New Zealand Standard implementing the US provisions.

Serviceability

For the determination of deflections in stainless steel flexural members, account must be taken of the

non-linear stress-strain characteristics of the material; simply assuming the initial tangent modulus

E0 will result in an under-estimation of deflections. The European, US and Australia/New Zealand

design Standards all adopt essentially the same treatment, whereby deflections are calculated based

on an effective section (for slender Class 4 cross-sections) and a reduced modulus of elasticity. In

all Standards the reduced modulus of elasticity is taken as the average of the secant moduli in

tension and compression corresponding to the maximum serviceability stresses that occur along the

member length (Equation 4).

E st  E sc
E s,ser  (4)
2

where Es,ser is the secant modulus at the serviceability limit state, and Est and Esc are the secant

moduli in tension and compression corresponding to the maximum serviceability stresses that occur

along the member length. The secant modulus for a given stress level may be approximated from

the Ramberg-Osgood expression (Equation 1) through Equation 5.

19
 E0
E s,i  n
(5)
E   ser ,i 
1  0.002 0  
 ser ,i   0.2 

where Es,i is the secant modulus, with the subscript i referring to either tension or compression, and

ser,i is the maximum serviceability stress in tension or compression. Clearly, calculating deflections

based on the secant modulus at the most highly stressed section of a member will lead to an over-

estimation of deflections, with the level of over-estimation dependent upon the distribution of

bending moments along the member length. Comparisons between measured deflections from tests

on stainless steel flexural members and predicted deflections from design Standards and other more

sophisticated techniques (that account for variation in stress along the member length) have been

made by [14] and [48].

Joints

Structural stainless steel joints may utilise welding, bolting or other mechanical fasteners. Basic

design philosophy is no different from that for carbon steel joints, and the high ductility that stainless

steel offers (particularly the austenitic grades) should be beneficial. There are however a number of

general design aspects that require consideration, in particular to minimise the risk of corrosion;

problems may be experienced where crevices exist, where there is dissimilar metal contact inducing

galvanic corrosion (e.g. joints comprising both stainless steel and carbon steel components) and in

the heat affected zone resulting from welding. General recommendations for stainless steel joint

design may be found in [5].

The common grades of stainless steel are readily weldable; the austenitic grades especially, can be

welded with ease, and since there is no change in metallurgical structure with temperature, the weld

material and surrounding zone should remain tough and ductile. As with ordinary carbon steel,
20
stainless steel grades with a low carbon (equivalent) content are desirable for welding (such as EN

1.4306 (AISI 304L) and EN 1.4404 (AISI 316L)), though these grades generally possess slightly

lower strengths. More detailed guidance on welding of stainless steels is available in [1] and [49].

The structural behaviour of stainless steel joints is similar to that of carbon steel joints, and this is

reflected in design guidance. Design rules are provided in each of the three design Standards under

consideration, with slight variations from their corresponding carbon steel recommendations. The

European guidance for stainless steel joints adopts the rules of Eurocode 3 Part 1.8 [50], subject to

some minor modifications given in Part 1.4.

Fire design

The elevated temperature physical properties of stainless steel differ significantly from those of

carbon steel, as discussed earlier; material strength and stiffness reduction factors are compared with

those of carbon steel in Figures 4 and 5, respectively. The European provisions for the design of

stainless steel members in fire largely follow the carbon steel rules, with the primary differences

being in the material properties. These properties are due to be incorporated into an Annex of EN

1993-1-2 [51] (Eurocode 3: Part 1.2 – Structural fire design) to extend the scope of this document to

the structural design of stainless steel in fire. Neither the US nor the Australia/New Zealand

structural stainless steel design Standard currently cover fire design, though some background and

guidance is provided as an Informative Annex in the Australia/New Zealand Standard.

Fatigue

The fatigue resistance of stainless steel has recently been shown to be comparable to that of carbon

steel [24], with indications of superior fatigue performance in many configurations. Hence, the

European prestandard ENV 1993-1-4 states that the design approach for carbon steel may be directly

21
applied to stainless steel. Upon completion, European guidance for the design of steel (and stainless

steel) structures subjected to fatigue loading will be contained in EN 1993-1-9 [52]. No guidance on

the assessment of fatigue strength of stainless steel structures is included in the US Standard. As

with the European Standard, the Australia/New Zealand Standard recommends use of the equivalent

carbon steel approach, given in AS 4100 [53], though with some limitations.

A new approach

An important step towards enhancing the understanding and use of stainless steel in structures has

been the development and publication of design guidance, including the European [4], US [36] and

Australia/New Zealand [42] Standards. However, although the US and Australia/New Zealand

Standards take account of the non-linear material behaviour below the 0.2% proof stress in the

determination of member buckling resistances, none of the existing design methods take account of

the non-linear material behaviour on a cross-sectional level; the consequence of this is inaccurate

prediction of cross-section behaviour where buckling occurs below the 0.2% proof stress and

inefficient prediction of cross-section behaviour where buckling occurs above the 0.2% proof stress

(due to neglection of the considerable strain hardening that stainless steel exhibits). Essentially

therefore, the existing design methods utilise a simplified elastic, perfectly-plastic material model for

the determination of cross-section resistance (upon which member resistances subsequently

impinge). This model is acceptable for carbon steel that exhibits a sharply defined yield point,

followed by a plastic yield plateau. For stainless steel, though, where there is no sharply defined

yield point and substantial strain hardening occurs, this model leads to overly conservative designs.

The existing structural stainless steel design codes place cross-sections into discrete behavioural

classes on the basis of individual element slendernesses; the well known procedure of cross-section

classification. Eurocode 3 defines four discrete behavioural classes of cross-sections. Cross-

22
sections with very high deformation capacity are classified as Class 1. They are fully effective under

pure compression and are capable of reaching and maintaining their full plastic moment in bending.

Class 2 cross-sections have a somewhat lower deformation capacity, but are also fully effective in

pure compression and capable of reaching their full plastic moment in bending. Class 3 cross-

sections are fully effective in pure compression, but local buckling prevents attainment of the full

plastic moment in bending. Bending moment resistance is therefore limited to the yield moment.

For Class 4 cross-sections, local buckling occurs in the elastic range. An effective cross-section is

defined based on the width-to-thickness ratios of individual plate elements, and this is used to

determine the cross-sectional resistance.

In view of the continuous nature of the stainless steel stress-strain curve, it seems rational that a

continuous, rather than a discretised system of design should be adopted. A new design method has

been developed [13, 32] that replaces the discrete behavioural classes by a single numerical value

that is a measure of the deformation capacity of the cross-section. The deformation capacity is based

upon the slenderness of individual plate elements and the interaction between elements within the

cross-section. The relationship between cross-section slenderness and cross-section deformation

capacity has been derived on the basis of stub column tests, including those performed as part of

recent significant testing programmes [17, 54, 55, 56].

For design purposes, cross-section resistances are determined using a local buckling strength derived

from the cross-section deformation capacity, in conjunction with an accurate material model

appropriate for stainless steels. The adopted material model is the compound Ramberg-Osgood

formulation given by Equation 2, and it enables stresses greater than the 0.2% proof stress to be

exploited in design. Member strengths are determined using the local buckling strength (raised to a

power) in combination with overall buckling curves. For bending resistance, the concept of a

23
generalised shape factor, originally devised by [57], into which material as well as geometric

properties of a cross-section are incorporated, is used. The resistances of members subjected to a

combination of axial load plus bending moments are determined through an interaction of the

component effects.

The new design method utilises accurate material modelling, accounts for element interaction within cross-

sections and provides a continuous measure of section deformation capacity (in place of the current

discretised system of section classification). The method has been shown to provide average member

strength enhancements of around 20% over existing design methods, whilst for stocky cross-sections

enhancements in member bending strengths of over 50% have been achieved [13]. Further

development to expand the scope of the design method is currently underway, and it is envisaged

that the new approach could be incorporated as an alternative method in future revisions of Eurocode

3.

Conclusions

Significant advances related to the use of stainless steel in structures have been made in recent years,

including the publication of a number of major structural stainless steel design Standards. This

paper provides of overview of the provisions of these Standards and a review of the material and

recent research activities upon which the documents have been based. Factors influencing stainless

steel as a material choice for structural engineering applications have also been discussed in the

context of recent advances. Although the initial material cost of stainless steel structural products is

about four times that of equivalent carbon steel products, it has been explained that comparison of

costs on a more holistic basis reveals a far lesser discrepancy.

24
The paper describes that, unsurprisingly, many aspects of the design of stainless steel closely follow

those of carbon steel, though there is sufficient diversity in physical properties to require separate

structural design treatment; it is the differences in structural design treatment that have been the

focus herein.

Acknowledgements

The author would like to thank Professor David Nethercot of Imperial College London, Nancy

Baddoo and Dr Bassam Burgan of the Steel Construction Institute and Dr David Dulieu of

Outokumpu for their support on recent structural stainless steel projects.

References

[1] Mann, A. P. The structural use of stainless steel. The Structural Engineer. 1993, 71:4, 60-69.

[2] Truman, J. The initiation and growth of high alloy (stainless) steel production. Journal of the Historical

Metallurgy Society. 1985, 19:1, 116-125.

[3] EN 10088-1. Stainless steels – Part 1: List of stainless steels. CEN. 1995.

[4] ENV 1993-1-4. Eurocode 3: Design of steel structures - Part 1.4: General rules - Supplementary rules for

stainless steel. CEN. 1996.

[5] Euro Inox. Design Manual for Structural Stainless Steel. Second edition. NiDI. Toronto, Canada. 2002.

[6] Baddoo, N. R., Burgan, R. & Ogden, R. Architects’ Guide to Stainless Steel. SCI-P-179. The Steel

Construction Institute, UK. 1997.

25
[7] Chung, K. F., Baddoo, N. R. and Burgan, B. A. Section Property and Member Capacity Tables for Cold-

Formed Stainless Steel. SCI-P-152. The Steel Construction Institute, UK. 1995.

[8] Baddoo, N. R. & Burgan, B. A. Structural Design of Stainless Steel (SCI-P-291). The Steel Construction

Institute, UK. 2001.

[9] The Steel Construction Institute. Stainless Steel in Construction – design software for cold-formed

stainless steel. Retrieved September 2004 from www.steel-stainless.org.

[10] Rasmussen, K. J. R., Burns, T., Bezkorovainy, P. and Bambach, M. R. Local buckling of cold-formed

stainless steel sections. Proceedings of the Stainless Steel Structures International Experts’ Seminar. Ascot,

UK, 2003. 1-19.

[11] Ramberg, W. and Osgood, W. R. Description of stress-strain curves by three parameters. Technical

Note No. 902, National Advisory Committee for Aeronautics. Washington, D.C. 1943.

[12] Hill, H. N. Determination of stress-strain relations from the offset yield strength values. Technical Note

No. 927, National Advisory Committee for Aeronautics. Washington, D.C. 1944.

[13] Gardner, L. A new approach to stainless steel structural design. PhD Thesis. Structures Section,

Department of Civil Engineering, Imperial College, London. 2002.

[14] Mirambell, E. and Real, E. On the calculation of deflections in structural stainless steel beams: an

experimental and numerical investigation. Journal of Constructional Steel Research 2000; 54, 109-133.

[15] Gardner L. and Nethercot D. A. Numerical modelling of cold-formed stainless steel sections.

Proceedings of the Ninth Nordic Steel Construction Conference – 18th-20th June 2001. Edited by Mäkeläinen

P, Kesti J, Jutila A and Kautila O. Helsinki, Finland. 2001. 781-789.

26
[16] Rasmussen, K. J. R. Full range stress-strain curves for stainless steel alloys. Journal of Constructional

Steel Research 2003; 59, 47-61.

[17] Gardner, L. and Nethercot, D. A. Experiments on stainless steel hollow sections - Part 1: Material and

cross-sectional behaviour. Journal of Constructional Steel Research 2004; 60(9), 1291-1318.

[18] Ashraf, M., Gardner, L. and Nethercot, D. A. Strength enhancement of the corner regions of stainless

steel cross-sections. Journal of Constructional Steel Research 2005; 61(1), 37-52.

[19] Coetzee, J. S., Van den Berg, G. J. and Van der Merwe, P. The Effect of Work Hardening and

Residual Stresses due to Cold-work of Forming on the Strength of Cold-formed Stainless Steel Lipped

Channel Section, Proceedings of the Tenth International Specialty Conference on Cold-Formed Steel

Structures, St. Louis, Missouri, U.S.A. 1990; 143-162.

[20] Van den Berg, G. J. and Van der Merwe, P. Prediction of Corner Mechanical Properties for Stainless

Steels due to Cold Forming, Proceedings of the Eleventh International Specialty Conference on Cold-Formed

Steel Structures, St. Louis, Missouri, U.S.A. 1992; 571-586.

[21] Rasmussen, K. J. R. and Hancock, G. J. Design of Cold-Formed Stainless Steel Tubular Members. I:

Columns, Journal of the Structural Engineering, ASCE 1993; 119(8), 2349-2367.

[22] Gardner, L and Talja A. WP2: Structural Hollow Sections, ECSC Project: Structural Design of Cold-

Worked Austenitic Stainless Steel, Contract No. 7210 PR318, The Steel Construction Institute. 2003.

[23] Sedriks, A. J. Corrosion of stainless steels. Second Edition. John Wiley and Sons, 1996.

[24] European Commission. Development of the use of stainless steel in construction. Final report,

Directorate-General for Research. European Commission, Technical Steel Research EUR 20030 EN. 2001.

27
[25] Ala-Outinen, T. and Oksanen, T. Stainless steel compression members exposed to fire. VTT Research

Notes 1864, Espoo,l Finland. 1997.

[26] Baddoo, N. R. and Gardner, L. WP5.2: Member behaviour at elevated temperatures. ECSC project –

Development of the use of stainless steel in construction. Contract No. 7210 SA/842. The Steel Construction

Institute, UK. 2000.

[27] Louca, L. A., Boh, J. W. and Choo, Y. S. Design and analysis of stainless steel profiled blast barriers.

Journal of Constructional Steel Research 2004; 60(12), 1699-1723.

[28] Di Sarno, L., Elnashai, A. S. and Nethercot, D. A. Seismic performance assessment of stainless steel

frames. Journal of Constructional Steel Research 2003; 59, 1289-1319.

[29] The Sustainable Construction Task Group. The UK Construction Industry: progress towards more

sustainable construction 2000-2003. 2003.

[30] Shuttleworth, E. P. Structural applications for stainless steel offshore - Report to The Nickel

Development Institute, Report No. SCI-RT-030. The Steel Construction Institute, 1989.

[31] Baddoo, N. R. A comparison of structural stainless steel design standards. Proceedings of the Stainless

Steel Structures International Experts’ Seminar. Ascot, UK, 2003. 131-150.

[32] Gardner, L. and Nethercot, D. A. Stainless steel structural design: A new approach. The Structural

Engineer. In press.

[33] Johnson, A. L. and Winter, G. Behaviour of stainless steel columns and beams. Journal of the

Structural Division, ASCE 1966. ST5, 97-118.

[34] Wang, S. T. and Errara, S. J. Behaviour of cold rolled stainless steel members. Proceedings of the First

Speciality Conference on Cold-formed Steel Structures. University of Missouri-Rolla, Rolla, Mo. 1971.

28
[35] AISI. Stainless Steel Cold-Formed Structural Design Manual. American Iron and Steel Institute.

Washington, D.C. 1974.

[36] ASCE. Specification for the Design of Cold-Formed Stainless Steel Structural Members, SEI/ASCE 8-02.

(Standard No. 02-008). American Society of Civil Engineers. New York, 2002.

[37] Lin, S.-H., Yu, W.-W. and Galambos, T. V. ASCE LRFD Method for Stainless Steel Structures. Journal

of Structural Engineering, ASCE 1992; 118(4), 1056-1070.

[38] Euro Inox. Design Manual for Structural Stainless Steel. First edition. NiDI. Toronto, Canada. 1994.

[39] Burgan, B. A. Concise Guide to the Structural Design of Stainless Steel. The Steel Construction

Institute, UK. 1992.

[40] SSBJA. The design and construction specifications for stainless steel structures. Stainless Steel Building

Association of Japan. (In Japanese). 1995.

[41] SABS. Structural use of steel – Part 4: The design of cold-formed stainless steel structural members.

SABS 0162-4:1997. 1997.

[42] Aust/NZS. Cold-formed stainless steel structures. Australian/ New Zealand Standard AS/NZS

4673:2001. Sydney, Australia: Standards Australia. 2001.

[43] Rasmussen, K. J. R. The Development of an Australian Standard for Stainless Steel Structures.

Proceedings of the Fifteenth International Speciality Conference on Cold-formed Steel Structures. St. Louis,

Missouri, USA. 2000; 659-671.

29
[44] EN 10088-2. Stainless steels – Part 2: Technical delivery conditions for sheet/plate and strip for general

purposes. CEN. 1995.

[45] AISI. Specification for the Design of Cold-Formed Steel Structural Members. 1986.

[46] Gardner, L and Nethercot, D. A. Numerical modelling of stainless steel structural components – A

consistent approach. Journal of Structural Engineering, ASCE 2004; 130(10). 1586-1601.

[47] Lagerqvist, O. & Olsson, A. Residual stresses in welded I-girders made of stainless steel and structural

steel. Proceedings of the Ninth Nordic Steel Construction Conference – 18th-20th June 2001. Edited by

Mäkeläinen P, Kesti J, Jutila A and Kautila O. Helsinki, Finland. 2001. 737-744.

[48] Chryssanthopoulos, M. K. and Low, Y. M. A method for predicting the flexural response of tubular

members with non-linear stress-strain characteristics. Journal of Constructional Steel Research 2001; 57.

1197-1216.

[49] Euro Inox. Welding of stainless steel. Material and Applications Series, Volume 3. Second Edition.

Lumembourg, Euro Inox. 2002.

[50] EN 1993-1-8. Eurocode 3: Design of steel structures - Part 1.8: Joints. CEN.

[51] EN 1993-1-2. Eurocode 3: Design of steel structures - Part 1.2: Structural fire design. CEN.

[52] EN 1993-1-9. Eurocode 3: Design of steel structures - Part 1.9: Fatigue strength of steel structures. CEN.

[53] AS 4100. Steel Structures. Standards Australia. 1998.

30
[54] Rasmussen, K. J. R. Recent research on stainless steel tubular structures. Journal of Constructional

Steel Research 2000; 54, 75-88.

[55] Talja, A. and Salmi, P. Design of stainless steel RHS beams, columns and beam-columns. Research

Note 1619, VTT Building Technology, Finland. 1995.

[56] Young, B. and Hartono, W. Compression tests of stainless steel tubular members. Journal of Structural

Engineering, ASCE 2002; 128(6), 754-761.

[57] Mazzolani, F. M. Aluminium Alloy Structures. 2nd Edition. E & FN Spon, An imprint of Chapman and

Hall. 1995.

31
 700

600 Stainless steel

500
Stress (N/mm2)

400 Carbon steel

300

200

100

0
0.000 0.004 0.008 0.012 0.016 0.020
Strain

Figure 1: Indicative carbon steel and stainless steel material σ−ε behaviour

1
 700

600

500
Stress (N/mm )
2

400

300

200

100

0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Strain (%)

Figure 2: Definition of ‘equivalent’ yield stress using the 0.2% proof stress

2
 600

500
Cold-worked material
Stress (N/mm )
2

400

300
Annealed material

200

100

0
0.000 0.002 0.004 0.006 0.008 0.010
Strain

Figure 3: Stress-strain behaviour of annealed and cold-work material

3
 1.4

1.2

1.0
fy( ) / fy (20 C)
o

0.8

0.6 Stainless steel

0.4
Carbon steel
0.2

0.0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
o
Temperature C

Figure 4: Strength retention of stainless steel and carbon steel at elevated

temperatures

4
 1.2

1.0

0.8
E( ) / E(20 C)

Stainless steel
o

0.6

Carbon steel
0.4

0.2

0.0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
o
Temperature C

Figure 5: Stiffness retention of stainless steel and carbon steel at elevated

temperatures

5
 1.2

1 Eurocode - cold-formed

Eurocode - welded
Reduction factor χ

0.8 US (n = 4)

Aus/NZ (n = 4)

0.6

0.4

0.2

0
0 50 100 150 200 250 300

Slenderness L/i

Figure 6: Comparison of the European, US and Australia/New Zealand flexural

buckling curves

6
 Table 1: Chemical compositions for selected stainless steel grades

Chemical composition (% by mass)

Steel Designation (Number)

Element
1.4301 (304) 1.4401 (316) 1.4462 (2205)

Carbon (C) ≤ 0.07 ≤ 0.07 ≤ 0.030

Chromium (Cr) 17.00 to 19.50 16.50 to 18.50 21.00 to 23.00

Nickel (Ni) 8.00 to 10.50 10.00 to 13.00 4.50 to 6.50

Molybdenum (Mo) - 2.00 to 2.50 2.50 to 3.50

Manganese (Mn) ≤ 2.00 ≤ 2.00 ≤ 2.00

Silicon (Si) ≤ 1.00 ≤ 1.00 ≤ 1.00

Phosphorus (P) ≤ 0.045 ≤ 0.045 ≤ 0.035

Sulphur (S) ≤ 0.015 ≤ 0.015 ≤ 0.015

Nitrogen (N) ≤ 0.11 ≤ 0.11 0.10 to 0.22

Titanium (Ti) 5×C to 0.70 5×C to 0.70 -

Tungsten (W) - - 0.50 to 1.00

V i e w p u b l i c a t i o n s t a t s