Optimising Precast Bridge Girders for Sustainability

with the use of High Performance Concrete

Doug Jenkins1, Joanne Portella2, Daksh Baweja2,3
1
Interactive Design Services, Australia
2
DMC Advisory, Melbourne
3
The University of Technology, Sydney, Australia

Abstract The standard Super-T bridge girders used in Australia were developed
to provide optimum performance at a time when the maximum concrete grade
covered by the bridge design code was 50 MPa. This paper examines the
opportunities for improved sustainability through the use of high performance
concrete, considering the use of existing standard sections, modified sections
optimised for higher strength grades, and the use of techniques such as hybrid pre-
tensioned and post-tensioned girders, and precast girders used in continuous
structures. These alternatives are compared for impact on CO2 emissions within
the context of current Australian precast and bridge construction practice. In
addition, the designs of the sections are reviewed based on a series of alternative
concrete mix designs covering a reference Portland cement concrete mix and a
series of concretes incorporating a range of supplementary cementitious materials
included at different levels of cement replacement to determine efficiencies in
design and impacts on the embodied energy required to manufacture the elements.

Introduction

The standard precast “Super-T” bridge girders used in Australia have proved to be
very popular, offering both an efficient design solution, and rapid construction. At
the time of their introduction the maximum concrete grade covered by The
Australian Bridge Design Code [1] was 50 MPa [2]. Since that time the maximum
concrete grade for use in bridges has increased to 65 MPa in AS 5100 [3], and the
latest Australian Standard Concrete Structures Code, AS 3600 [4], released late in
2009, covers concrete strengths up to 100 MPa. Use of these higher strength
concretes offers potential for reduction in quantities of concrete and/or steel, offset
by higher cement content, but the current range of standard girders are not
necessarily optimal for use with higher strength concrete, and there is little data
available on CO2 emissions associated with different alternatives.

Super-T Bridge Girders were introduced in Victoria in 1993, and were quickly
adopted by the other Australian States [2]. For the purposes of this study, open
topped girders of type T3, T4 and T5 were used as standard sections, and modified
2 Doug Jenkins, Joanne Portella, Daksh Baweja

type T2 and T3 were used for optimised designs. The Standard Type T5 Super-T
open topped section is shown in Figure 1. Table I shows overall depths and
bottom flange depths for standard sections T3 to T5 and the modified sections
used in conjunction with post-tensioning and/or continuous construction, sections
T3A, T3B and T2A.

In this paper the design of a typical two span freeway over-bridge is examined,
comparing standard strength concrete and girders with higher strength grades and
girders optimised for use with high performance concrete, post-tensioning, and
continuous structures. These alternatives are examined for their effect on life-
cycle CO2 emissions.

Details of Study

This study examines the effect of the use different high performance concrete
mixes on the life-cycle CO2 emissions of a typical 2 span freeway overbridge. The
reason for using the term performance instead of strength relates to the
mechanical, serviceability and durability requirements of the concrete necessary
for efficient design and manufacture of the structural elements. Key design
features of the section are as follows:

 Two span freeway over-bridge

 Total length; abutment to abutment - 61 m (2 x 28.5 m span + 2.5 m link
+ 1.5 m ends)
 Carriageway width - 11.0 m; Footway / verge widths - 0.75 m both sides
 5 or 6 open topped Super-T girders
 In-situ top slab of 160 mm depth.
 SM 1600 Loading
 Typical Sydney shrinkage and creep parameters
 Exposure class B1

Alternative concrete mixes selected for this study covered the following:

A. Reference case: 50 MPa characteristic compressive strength concrete

made using Portland cement without supplementary cementitious
materials (SCM’s), defined in Australian Standard AS1379 (Specification
and Supply of Concrete) [6], AS3972 (General Purpose and Blended
Cement) [7], and AS3582 Parts 1 [8], 2 [9] and 3 [10] (Supplementary
Cementitious Materials for Use with Portland and Blended Cements).
B. Typical current high strength concrete; characteristic compressive
strength = 65 MPa.
C. High strength concrete having a characteristic compressive strength of 80
MPa
Optimising Precast Bridge Girders for Sustainability 3

D. Very high strength concrete having a characteristic compressive strength

of 100 MPa
E. High SCM concrete having a characteristic compressive strength of 45
MPa.

Fig 1: Type 5 Super-T Girder

Table I. Super-T Girder Dimensions

Type O/A Depth, mm Bottom Flange, mm
Base Width Depth
T3 1200 814 260
T4 1500 757 260
T5 1800 700 325
Modified:
T2A 1000 852 150
T3A 1200 814 200
T3B 1200 814 150

Details of the five mixes and design compressive strengths are shown in Table II.
The emission data for the component materials used in the analyses are taken from
earlier published work [11], and are given in Table III. Emission calculations are
shown in Table IV. Calculations took the quantity of each component material
4 Doug Jenkins, Joanne Portella, Daksh Baweja

and obtained a total emission quantity in the mix by multiplying by the

corresponding emission factor given in Table III [11].

Table II: Mix Design Details

Concrete Property Unit Mix A, Mix B Mix C Mix D Mix E

Baseline 65 MPa HPC 80 MPa 100 MPa 45 MPa HPC
50 MPa Current HPC HPC High SCM
Total Binder kg/m3 550 490 640 680 440
1
Portland Cement kg/m3 550 350 500 540 245
Fly Ash2 kg/m3 70 80 60 85
GGBFS3 kg/m3 70 110
Amorphous Silica4 kg/m3 60 80
5
Coarse Aggregate kg/m3 1120 1050 1050 1000 1100
Sand5 kg/m3 590 675 630 650 670
Water6 kg/m3 180 180 180 180 180
Water:Binder 0.33 0.37 0.28 0.26 0.41
28 Day Strength7 MPa 60 70 90 110 50
Transfer Strength MPa 35 35 40 40 25
8
Shrinkage μstrain 700 600 550 550 650
Standard specifications: 1, AS 3972; 2, AS 3582.1; 3, AS 3582.2; 4, AS 3582.3; 5, AS 2758.1;
6, AS 1379; 7, AS 1012.9; 8, AS 1012.13

Table III: Concrete Component Emission Factors (11)

GP Cement 0.820 t CO2-e/tonne

Fly Ash 0.027 t CO2-e/tonne
GGBFS 0.143 t CO2-e/tonne
Silica Fume 0.027 t CO2-e/tonne
Basalt Coarse Aggregates 0.036 t CO2-e/tonne
Fine Aggregates 0.014 t CO2-e/tonne
Concrete Batching 0.003 t CO2-e/m3
Concrete Transport 0.009 t CO2-e/m3

Design Options

For each mix design 3 alternative structural configurations were considered:

 Type 1 - Fully Pre-tensioned Design: Typical current practice;

Standard Super-T girders, fully pre-tensioned. Simply supported
spans with in-situ top slab and link slab.
 Type 2 - Post-tensioned Design: Super-T optimised for use with High
Strength Concrete. Pre-tensioned for transport and construction loads
Optimising Precast Bridge Girders for Sustainability 5

with additional post-tensioning for live loads and long term effects.
Simply supported spans with in-situ top slab and link slab.
 Type 3 - Post-tensioned Continuous Design: As 2, but with full
structural continuity over the central support.

Table IV: Mix Emission Details (per cubic metre of concrete)

Mix A Mix B Mix C Mix D Mix E

50 MPa 65 MPa 80 MPa 100 MPa 45 MPa
Control Current HPC HPC High SCM
Portland Cement 0.4510 0.2870 0.4100 0.4428 0.2009
Flyash 0.0000 0.0019 0.0022 0.0016 0.0023
GGBFS 0.0000 0.0100 0.0000 0.0000 0.0157
Amorphous Silica 0.0000 0.0000 0.0016 0.0022 0.0000
Coarse Aggregate 0.0403 0.0378 0.0378 0.0360 0.0396
Sand 0.0083 0.0095 0.0088 0.0091 0.0094
Totals, t CO2-e/m3 0.51 0.36 0.47 0.50 0.28
% due to Portland
88% 80% 87% 88% 72%
Cement
Portland Cement
0% 36% 9% 2% 55%
Reduction

Note: As an example, for the Mix A Portland cement component, the emission derived is
550 x 0.82 / 1000 tCO2 per cubic metre of concrete

Analysis and Design Procedures

Bridge Deck Analysis

The structures were analysed with the finite element package Strand7. The
precast girders were modelled with beam elements, located on the precast section
centroid, with the in-situ top slab modelled with plate-shell elements, connected to
the beams with rigid links. A typical model is shown in Figure 2. Results of the
structural analysis are shown in Table V.

Emissions Analysis
Component emission factors used to calculate embodied energy of concrete are
presented in Table III and are taken from other studies conducted on concrete
materials [11]. Concrete mix emissions for alternative mix designs A to E are
given in Table IV and are expressed in tonnes of C02 emissions per cubic metre of
concrete (tCO2-e/m3). These values were calculated using predetermined concrete
emission factors for each of the concrete constituents [11]. An allowance of 5% of
the Portland Cement content as mineral additions and or minor additional
6 Doug Jenkins, Joanne Portella, Daksh Baweja

constituents has been made for the purpose of these calculations, though the
recently published new edition of AS3972 (General Purpose and Blended Cement)
[7] has increased this allowance to 7.5%.

Figure 2: Typical Deck Model

Bridge Deck Sections

Six type four girders were required for the base case standard mix (Mix A), and
the standard current high strength mix (Mix B). The high SCM mix (Mix E), with
a lower strength at transfer, required six Type 5 girders. The higher strength
mixes (Mix C and Mix D) allowed the number of girders to be reduced to five
Type 4 girders.

The level of prestress was controlled by the standard bottom flange depth, so
increasing the concrete strength from 80 to 100 MPa did not allow any further
reduction in girder numbers or type. Use of post-tensioning allowed higher levels
of total prestress and reduced prestress losses. This allowed the use of shallower
girders and reduced depth of bottom slab.

Providing structural continuity over the central pier allowed a further reduction in
the bottom flange depth and/or girder type, except for the Type D mix. Total
concrete, reinforcement and prestressing quantities and total CO 2 emissions are
summarised in Table VI. Emissions for the in-situ concrete were based on the
Type A mix for Deck type 1A, and the lesser of Type B mix or the girder mix for
all other deck types.
Optimising Precast Bridge Girders for Sustainability 7

0.0

-10.0

1-A, 1-B
-20.0 1-C
1-D
Deflection, mm

-30.0 1-E
2-B
2-C
-40.0
2-D
2-E
-50.0 3-B
3-C
3-D
-60.0
3-E

-70.0
0 2 4 6 8 10 12
Position, m

Fig 3: Maximum live load deflections

Bridge Deflection Results

Maximum deflections at mid span under SM 1600 loading [3] are shown for each
deck and mix combination in Figure 3. The analyses used the mean modulus of
elasticity of concrete specified in Cluse 3.1.2(a) of AS 3600 [4]. The maximum
allowable deflection (averaged across the span width) specified in AS 5100 is
Span/600 or 47.5 mm.

The smaller section depth used with the post-tensioned slabs (Decks Type 2 and 3)
has resulted in significantly increased deflections, but this is reduced by the use of
the higher strength grades, with an increased elastic modulus, and the provision of
moment continuity over the central pier for Deck Type 3. Two of the deck / mix
combinations studied in this paper were found to have deflections greater than that
permitted by AS 5100; Type 2-E exceeded the limit by 3%, and Type 2-D by
11%. In practical applications these deflections could be reduced either by using
the next deeper girder, using a higher strength concrete, or by providing moment
continuity over the central pier.
8 Doug Jenkins, Joanne Portella, Daksh Baweja

Table V: Structural Analysis Output Summary

Deck/ Composite ULS Design Actions

Mix Mid-Span Link/Continuity Slab
Type Moment Axial load Shear Moment Axial load Shear
kNm kN kN kNm kN kN
1-A/B 8,930 -1,339 1,355 45 651 292
1-C 10,080 -825 1,481 99 -1,080 353
1-D 10,080 -825 1,481 99 -1,080 353
1-E 9,459 -693 1,371 40 -569 263
2-A/B 10,148 -737 1,573 10 -797 39
2-C 10,080 -737 1,573 10 -797 39
2-D 10,125 -885 1,427 10 -1,271 39
2-E 10,148 -737 1,573 10 -797 39
3-A/B 6,730 -580 1,854 4,874 -536 2,483
3-C 6,399 494 1,847 4,878 -529 2,499
3-D 6,331 636 1,847 4,943 -1,532 3,217
3-E 6,730 -580 1,854 4,874 -536 2,483

Resource and Emission Analysis Results

In Table VI, the volume of concrete resulting for each deck option was multiplied
by the total C02 emissions per cubic metre of concrete (tCO 2-e/m3) for the
relevant mix to determine the total CO2 emissions for each deck/mix type option.
All options studied provided significant emissions savings compared with the
Base Case (Mix Type A, Deck Type 1), with the greatest savings being provided
by Mix Type E (High SCM mix). Savings were in the range of 15% to 19% for
the fully pre-tensioned deck, increasing to 24% to 32% for the post-tensioned
deck. A further 3% saving resulted from providing structural continuity at the
pier.

This result suggests that by consideration of structural design options and adopting
higher grade concrete, concrete volume can be reduced and the overall CO 2
emissions for a bridge deck, or potentially for any other structural element, can be
decreased in a broader sense.
Optimising Precast Bridge Girders for Sustainability 9

Table VI: Summary of Quantities and Emissions

Deck / Prestress; No. 15.2 Total Quantities Total Emmissions

Super-T Girders
Mix mm dia. Strands In-situ Precast Reo.
%Type
Type Type Num. Pretens Posttens m3 t t CO2-e
1A
1A T4 6 40 0 147 224 65.06 376.9 100.0%
1-B T4 6 40 0 147 224 65.06 320.1 84.9%
1-C T4 5 42 0 147 186 59.58 312.3 82.8%
1-D T4 5 46 0 147 186 58.62 315.3 83.6%
1-E T5 6 28 0 151 243 67.26 304.2 80.7%
2-B T3A 5 22 30 142 153 58.43 274.1 72.7%
2-C T3B 5 20 30 142 146 57.76 286.1 75.9%
2-D T2B 5 24 34 139 133 58.98 286.6 76.0%
2-E T3 5 22 30 142 170 58.43 255.6 67.8%
3-B T3B 5 22 10 159 146 54.25 265.5 70.4%
3-C T2B 5 24 14 154 133 54.42 274.5 72.8%
3-D T2B 5 24 14 154 133 53.15 275.0 73.0%
3-E T3B 5 22 10 159 146 54.25 241.6 64.1%
Examples:
Deck Type 3-E, derived emission = (159 + 146)m3 x 0.280 + 54.25t x 2.88 = 241.6 tCO2
Deck Type 1-C, derived emission = 147 x 0.358 + 186 x 0.472 + 59.58 x 2.88 = 312.3 tCO2

The push to replace the Portland Cement component of concrete mixes with an
increased percentage of SCM’s per se may not necessarily provide the optimum
result in the context of reduced resource use and overall CO 2 emissions. In this
investigation, it was found that the Type 2 deck (post-tensioned precast girders)
resulted in reduced overall CO2 emissions compared to Type 1 (fully pre-
tensioned design), and the Type 3 deck (post-tensioned with moment connectivity
at the central pier) resulted in a further reduction in emissions, with significantly
reduced deflections. Further, the use of the higher strength concrete with all three
deck types resulted in significantly improved efficiencies in regard to embodied
energy, compared to the Type A mix (50 MPa with no SCM’s). In addition, the
options incorporating a high SCM content concrete (mix Type E) also gave
favourable environmental results. This was not apparent when data for each
concrete mix was considered independently of the design evaluation shown in
Table V. One key factor that has been highlighted in this study is the error in
assuming that reducing Portland cement content in concrete will necessarily
achieve favourable environmental outcomes for a construction project.
10 Doug Jenkins, Joanne Portella, Daksh Baweja

Other factors that are not immediately obvious include constructability

requirements. The use of high SCM concrete such as mix E will lead to reduced
early age strengths that may not be suitable for the early application of
prestressing, as required in fully pre-tensioned products. This may be overcome
with more accurate means of measuring early age strengths [12] or with the
provision of post-tensioned reinforcement. The effect on precast productivity
therefore needs to be considered when selecting the optimum solution.

Consideration needs also to be given to how Portland cement reduction could be

achieved using other inclusions such as chemical admixtures. These materials
have a significant favourable impact on early age properties of concrete and
should also be considered in the framework of options to produce enhanced
impact on environmental outcomes.

Conclusions

The following conclusions can be drawn from this study:

 The use of SCM’s allowed significant reductions in CO2 emissions for all
the concretes studied, when compared with the standard “reference case”
concrete.
 The greatest reduction in emissions was found with the high SCM
concrete, but this was associated with a reduced compressive strength at
transfer, and increased curing period, which would increase the cost of
precast operations.
 Emissions from the 80 MPa and 100 MPa concretes were equal to or only
slightly higher than the 65 MPa concrete, and also allowed the use of a
reduced depth of girder, which would often allow significant reductions
in emissions and cost from associated works.
 The use of precast post-tensioned girders allowed significantly higher
levels of prestress, with a resulting reduction in concrete volumes and
total emissions.
 The reduced girder depth used with the post-tensioned options resulted in
increased deflections, in two cases exceeding the allowable deflection
specified in AS 5100.
 Provision of structural continuity over the central support allowed an
additional small saving in total emissions, and also significantly reduced
deck deflections.
 The overall reduction of CO2 emissions was not a simple function of the
reduction of Portland cement in the concrete, but was also based on how
the material properties of the concretes used influenced the structural
efficiency of the design.
Optimising Precast Bridge Girders for Sustainability 11

References
[1] The Australian Bridge Design Code (1992), Austroads, Sydney
[2] Merretz W (1997) Towards National Standardisation of Super-T Bridge Girders. Austroads
Bridge Conference 1997. Austroads, Sydney.
[3] Australian Standard 5100, Bridge Design (2004). SAI Global, Sydney
[4] Australian Standard 3600, Concrete Structures (2009). SAI Global, Sydney
[5] Jenkins D (2005) High Performance Concrete in Bridge Decks. Austroads Bridge
Conference, 2005. Austroads, Sydney
[6] Australian Standard AS1379, Specification and Supply of Concrete (2007). SAI Global,
Sydney
[7] Australian Standard AS3972, General Purpose and Blended Cements (2010). SAI Global,
Sydney
[8] Australian Standard AS3582.1, Supplementary Cementitious Materials for Use With
Portland and Blended Cement – Part 1: Fly Ash (1998). SAI Global, Sydney
[9] Australian Standard AS3582.2, Supplementary Cementitious Materials for Use With
Portland and Blended Cement – Part 2: Slag-Ground Granulated Iron Blast-Furnace
(2001). SAI Global, Sydney
[10] Australian Standard AS3582.3, Supplementary Cementitious Materials for Use With
Portland and Blended Cement – Part 3: Amorphous Silica (2002). SAI Global, Sydney
[11] Flower D, Sanjayan J, Baweja D (2005) Environmental Impacts of Concrete Production
and Construction, Proceedings, Concrete Institute of Australia Biennial Conference,
Concrete 2005. Concrete Institute of Australia, Sydney
[12] Sirivivatnanon V, Baweja D, Khatri RP (2009) Evaluation of In Situ Concrete Strengths
for Post-Tensioning of Concrete Slabs. Concrete Forum, Refereed Journal of the Concrete
Institute of Australia, Vol. 2, No. 1, pp 1-10.
[13] Sofi M, Mendis PA, Baweja D, Mak SL (2007), Behaviour of Post-Tensioned Anchors in
Early-Age Concrete Slabs, Proceedings, 23rd Biennial Conference, Concrete Institute of
Australia. Concrete Institute of Australia, Sydney