Whole-life costs: Concrete vs steel

01 Introduction

Which is more sustainable - concrete or steel-framed buildings? This is a question people have asked themselves time and time again and, as a result, many comparisons have been made. Currie & Brown was keen to examine the issue taking into account wider considerations and, with the help of services engineer Hoare Lea & Partners, to evaluate the energy benefits of the increased thermal inertia of concrete construction.

Increasing awareness of the importance of sustainability, and especially the latest amendments to Part L of the Building Regulations, make this even more pertinent.

This article seeks to compare a steel-framed office building using composite floor construction - steel decking supported on beams with a concrete topping - with an alternative of a concrete frame with insitu concrete floors. The evaluation will cover changes in capital costs, energy running costs, embodied energy costs and other differences in environmental impact.

The Live Options model

The whole-life costs presented here are based on Currie & Brown’s Live Options software, which is an integrated suite of geometric, engineering and energy calculation programs.

Whether the development is an office, school, hospital or warehouse, the process starts in the same way, by using a basic but faculty-specific design template. This is modified to incorporate, for example, changes in size, shape and operational performance.

The program integrates fabric and building services components and can show how costs change when a building element is changed - for example, increasing the proportion of glazing in the facade. It does this by providing a detailed breakdown of changes and shows the capital costs of physical items and the effect these have on building performance.

Any changes to heating and cooling loads will be quantified in terms of the effect on, say, chillers and the electrical supply loads.

Changes to the energy profile of the facility are also calculated. The program uses this data to assess the cost effectiveness and overall viability of alternative solutions when considered against the facility’s projected economic life.

The model also takes account of repair and maintenance costs. This includes regular inspections, planned and unplanned maintenance and replacement costs for individual building components. These costs are automatically cash-flowed, discounted and summarised and net present costs are shown against each component, which assists in making choices and in value-engineering.

02 Changing the frame from steel to concrete: Design strategy

The thermal inertia of a building determines how quickly the temperature responds to changes in heat gain or loss.

If a building’s thermal inertia can be increased, its temperature fluctuates less, so its daily response to weather conditions

or internal heat gains and losses will be more limited and slowed. This can be used by designers to control indoor temperatures without the need for mechanical cooling and can help reduce building carbon emissions.

Inevitably there will be periods during the warmer parts of the year when mechanical cooling is required. It is therefore common for buildings to be naturally cooled when temperatures allow, but to switch to mechanical cooling when ambient temperatures are high.

This mixed-mode cooling strategy will use the thermal inertia of a building to moderate the temperature swings. The mass of the building is exposed to the occupied space to enable heat transfer between the concrete and the space. During the day heat is absorbed by the slab, thus limiting the rise in temperature within the space.

The heat must be removed overnight, however, if the process is to be repeated the next day. A night cooling strategy can be used to remove this heat by allowing cooler night air to flow through the building either by opening windows automatically - security concerns allowing - or by using fans to draw the air through the building. In this way, the concrete is cooled down.

The concrete gives up its stored coolth in two ways. First, by being exposed to the room to allow a passive exchange of heat. Second, it can exchange heat with supply air in a raised floor plenum.

The model used assumes an insitu concrete floor, which will be thicker than the concrete layer in a composite floor. In the context of night-to-day cooling, only the surface 50-70 mm (opinions vary) will be of benefit and so for the top surface, the concrete system will not have an advantage over a thinner composite floor system over 24 hours. However, the increased depth can help over a longer period, such as a few day’s heat wave, when the deeper reserves of thick concrete will help.

Most importantly, above the occupied space, while the steel frame is assumed to have a conventional ceiling, the flat slab concrete floors could be partly exposed.

One option, which was considered for use in this model, was to apply just a spray-applied coating or even just paint, which would produce a big saving in ceiling costs and in further reducing the required slab-to-slab height. However, this tends to produce too much acoustic resonance in an open-plan office. Therefore, the model opts for acoustic ceiling panels over just a proportion of the area, so that the coolth would be felt from air flow between the ceiling panels.

In a modern well-sealed, well-insulated building, the slab would not cool much overnight, so it needs to be re-charged sufficiently to cool it down overnight in the summer. We have assumed that, during the summer, the air supply fans continue to push air through the floor void overnight.

In the occupied space, in addition to windows at the normal level, there would be high-level bottom-hung inward opening windows, which would be opened at night in hot summer periods, to allow the space and underside of the slab to be cooled overnight. The operation of the windows needs to be carefully controlled otherwise the building becomes too cold in the mornings and requires heating for the first hour or two, which would be counter-productive.

There are two alternative approaches to this:

• The automated approach using window actuators, which have to be linked back to the building management system for both environmental control and fire safety. For windows of more than 1 m wide, more than one actuator per window is usually needed and these have to be linked.

All this has to be very carefully balanced and controlled. Otherwise, it could result in being too cool in the mornings and require heating for the first hour or two, which would be counter-productive. For automatically controlled windows, one should allow extra money and time to commission the building during different weather conditions, but especially during a hot period.

• The manual approach, whereby a caretaker goes around and opens or shuts the high-level windows as necessary. This approach has been popular with local authorities for schools.

We have assumed the second option and assume that even if well operated, there will be some increase in the pre-heating loads and time for the concrete solution.

The exposed concrete system can be restrictive in that lighting positions and circuitry and ductwork have to be carefully co-ordinated, unless uplighters are used - and these are less energy efficient. Electrical leads may need to run through the slab unless they can be routed over the limited area of ceiling panels. This, in turn, can impair having separate tenancies on different floors.

However, a flat slab or waffle system makes the job of the mechanical subcontractor easier especially for ductwork, in not having to duck and dive around the beams needed for the composite floor on steel framed buildings. The use of the thermal inertia in this way can often alleviate the need for cooling altogether, but we have assumed that occupant density and consequential internal gains are high, so that chilled beams are used in both options - though less extensively in the concrete solution.

There are proprietary patented systems, such as Termodeck, that use the thermal inertia in a more active manner. See www.termodeck.co.uk for details.

03 Changing the frame from steel to concrete: Capital cost changes

The building uses a repetitive column grid spacing of 7.5 × 7.5 m. The structural steel model was developed in conjunction with Corus and the beam depth is assessed automatically based on a span depth ratio for the beams and an assumed floor thickness for 140 mm (although such values may be overridden).

Modelling involves:

• Switching from a steel frame with composite floors to a concrete frame with insitu concrete floors. The model automatically assesses the column sizes and floor thicknesses (based on guidance from the Concrete Centre). The overall effect on the storey module is a saving of 200 mm because downstand beams are not needed for insitu concrete floors. It allows for a deflection allowance of 25 mm, based on floor spans.

• Changing the ceiling type. For a high acoustic performance, we could use acoustic panels that have a convex curve over about two-thirds of the area. However, we have assumed a cheaper option of a spray-applied coating to the underside of the slab, with flat acoustic panels over one-third of the area.

• Increasing the margin on the heating load to account for raising the temperature after cold weekends or after the Christmas break.

• Slightly reducing of cooling load for the effect of thermal inertia.

• Increasing preliminaries just 1%, assuming that about 80% of preliminaries costs are time-related. It is worth noting that this may increase interest charges, since usually insitu concrete structures would take a bit longer to build. However, this does not necessarily mean the overall development period would increase, because the off-site construction period for steel and associated lead-in time will be greater than for insitu concrete. Overall, there is not usually a significant difference in the overall development time between the two forms of construction once the design and lead-in times have been considered. A lot depends on the particular contractors and their experiences, and opinions vary a lot.

At this point, there is a very slight saving for the concrete frame option of about £32,000. Now we want to use the thermal inertia to greater benefit, so we:

• Add opening fanlights to charge the floor void and slab from cool air at night. We added an extra-over cost for opening lights, assuming an openable area of one-20th of the gross floor area.

• Reduce cooling load. Live Options automatically advises on this and pre-heating loads according to a traditional assessment of a building’s “thermal weight” - calculated from the remodelled components’ admittance values. However, this is adjusted according to advice from Hoare Lea, so the peak cooling load is reduced by 12 W/m².

Cost comparison

• Substructure The green shows the extra area, volume and cost of bored pad foundations caused by the extra weight. We assumed a gravel base with a permissible stress of 500 KN/m².
• Structure The red is the saving for the steel frame and associated fire protection, compared with concrete columns and a few beams. It is important to note that the grid of 7.5 × 7.5 m is very suitable for two-way spanning flat slab concrete construction as it has the advantage of simplifying formwork and allowing repetitive uses. The green reflects the extra cost of a 290 mm thick concrete floor as compared with a 140 mm lightweight concrete floor on a permanent steel soffit. The yellow is a saving on stairs and associated balustrading and handrailing. This is because the stairs are resized as a result of the 200 mm reduction in the storey module.
• Envelope The blue shows the saving on the external wall height. The yellow is purely the extra-over cost for manually openable fanlights over fixed lights.
• Internal division Small savings because of 200 mm reductions in the height at each level.
• Finishes and fittings Red denotes the savings on bases and finishes to walls that have reduced in height. Blue denotes the higher cost for the spray and acoustic ceiling panels compared with a conventional flat ceiling.
• Mechanical Yellow denotes a saving on cooling load enabled by an increase in thermal inertia. Blue is the extra cost of the heat source to account for the increased thermal inertia. Red is the saving on vertical pipework for the saving in height.
• Electrical Red indicates a saving on the electrical load as a result of savings in cooling load.

There is very little in it cost-wise, although the concrete frame compares better in terms of cost/net area, because 7 m2 is saved on the area of stairs, which are shorter.

It seems that for small or intermediate spans at least, comparing structural steel with insitu concrete is often a close-run thing. The situation is more likely to favour concrete when:

• Ground conditions are good (so the extra weight doesn’t have much effect).
• The wall cost is expensive, so a saving on height is more significant.
• The building’s general design, and air-conditioning and ventilation systems particularly, make good use of the thermal inertia of the concrete. (An exposed concrete soffit will induce radiant loss as well as reducing air temperatures in hot weather.)
• The height is limited, so the saving on storey module might allow an extra floor.
• Future flexibility is not considered important, as insitu construction is less easy to modify after construction.

04 Changing the frame from steel to concrete: Energy cost changes

To explain the changes to energy consumption:

• Fabric losses generally The extra thermal inertia effects heating loads and pre-heat periods after a cold weekend or after the Christmas break. However, a saving is shown on walls (green) because of the 200 mm reduction in height at each level.
• Ventilation losses The blue is purely reduced air infiltration resulting from the reduced height.
• Cooling The benefit of thermal inertia is spread across all the sources of heat (internal, solar, lighting and so on) denoted by the various colours.
• Fans The saving on cooling load and periods leads to a reduction in fan use. (The saving here would have been greater for air-based systems such as VAV.)
• Power Red denotes the reduced power for pumps serving the cooling.

The savings on cooling dwarf the extra on heating, but this is partly because heating uses gas at 2p per kWhr, whereas cooling uses electricity at 6p per kWhr.

It is normal to use the same discount rate for all operating costs, typically, the Treasury’s recommended rate for public projects of 3.5%. However, energy costs are expected to rise more than general inflation, so here we are using a 2% discount rate. This “real” discount rate is the amount by which investment and monetary growth, such as through interest rates, exceeds the rate by which inflation erodes the purchasing power of that money.

Repairs and maintenance

Unlike previous articles in this series, although there may be small repairs to areas of concrete or fire protection and run times for some plant may be reduced, the overall effect of changes is not significant in the context of other items discussed here.

Whole-life cost savings

Over 30 years, the combined effects of a change from a steel-framed air-conditioned office to a building with insitu concrete frame and floors are:

Capital cost changes: +£42,000 +3.6%

Energy cost changes: -£3,800 a year = -£85,000 -4.1%

(saving 57 tonnes of CO₂ a year)

Total change in net present cost: -£43,000

05 Environmental issues

Steel

It is estimated that, worldwide, more than 85% of steel is recycled at the end of its life. Such a high figure might seem surprising until one realises that the process is enhanced by steel’s natural magnetism, which makes it easy to sort.

In UK construction, the re-use and recycling rates of various steel products have been estimated at 92% for rebar, 85% for hot-dip galvanized sheet and 99% for structural steel sections. Some sections and cladding are reused in agricultural and industrial buildings especially, and this is facilitated by the use of bolted sections rather than riveting and/or welding. By saving remelting, re-use is the most environmentally advantageous approach at the end of a building’s life.

The energy used in producing steel from recycled steel is roughly one-third of that for new steel. Recycling steel saves energy, CO₂ and resources by displacing the need to make more steel from virgin sources. Unfortunately though, both worldwide and in the UK, the demand for steel outstrips the supply from demolished or scrapped steel. In fact, all recovered scrap is already recycled through primary and secondary steel-making routes in one global system. As scrap is a globally traded raw material, it is impractical to distinguish for each country between primary produced steel and steel produced from scrap.

A global view is instead taken, which avoids the impracticalities of determining the precise origin of steel consumed in the UK. ISO 14041 sets the method by which the embodied energy and product life-cycle environmental impacts should be calculated. In this way, the mix of new and recycled steel and end-of-life recycling are taken into account, taking a “cradle to grave” approach to environmental consideration.

So, using UK recycled rates, the figures are 13.1 MJ/kg for steel sections and 12.1 MJ/kg for rebar. The corresponding CO₂ outputs are 0.76 kg of CO₂ per kg of steel and 0.79 kg of CO₂ per kg of reinforcement. Not included in these figures is the energy for fabrication, transport from factory to site and on-site construction, although these are relatively minor in comparison.

The argument from the concrete lobby is that although the figures reflect the worldwide situation regarding the proportion of available recycled steel versus steel from virgin resources, it plays down the impact of structural steel in the UK, which predominantly uses steel from virgin sources.

However, this new steel will in the main be reused many times, so it could be seen as unfair to account for the initial energy cost against its first life.

Most of the world’s iron ore production comes from a handful of large international mining companies and many of these have systems to minimise environmental impact.

Reinforced concrete

The embodied energy of producing concrete is about 380 kg of CO2 per m³ concrete in structural components such as floors and columns. It is about 310 kg of CO₂ per m³ concrete in pad foundations or the like. Increasingly, though, cement may be partly replaced by alternatives such as pulverised fuel ash (PFA), a by-product of coal-fired power stations, and ground granulated blast furnace slag (GGBS), a by-product of steel production.

A substitution of cement with 30% PFA saves about 20% CO₂, whereas substitution with 50% GGBS saves about 40% of CO2, but this assumes that CO₂ should be entirely accounted for in steel manufacturing figures, rather than the GGBS that flows from it.

The Concrete Centre says:

• 85% of aggregate travels less than 30 miles
• 90% of cement is sourced from the UK, whereas 10% is imported
• In the UK, almost all reinforcement is produced from recycled steel
• All the companies that produce cement have environmental management systems in place and programmes to minimise the environmental impact from mining activities.

It is estimated by BRE’s Green Guide that 50% of concrete is crushed and recycled, 40% is downcycled for use such as hardcore in substructure works or road construction and the remaining 10% is waste that goes to landfill. Down-cycling does help to reduce the use of aggregates, but does not help reduce the supply of materials for

new concrete.

Ecopoints

An article like this cannot analyse all environmental impacts. For example, there are many other types of gas emissions that should be considered, such as nitrous oxide (NO₄). The best overview of the overall impact of these materials is the Ecopoint rating developed by BRE.

• Structural steel has 11 Ecopoints per tonne
• Reinforced concrete to 35 N/mm² (including rebar at 100 kg/m³) has 5.3 Ecopoints/m³ (using a density of 2371 kg/m³), or 12.57 Ecopoints per tonne.

Environmental summary

If we ignore operational energy savings, the concrete option appears to be about 30% worse (see table overleaf), but when operational energy is accounted for, this dwarfs the embodied energy and the appraisal is reversed showing a saving of 6% for the concrete option.

This comparison has taken two common and recognised solutions but they are dissimilar in that one has a full ceiling while the other uses a predominantly exposed slab. This puts the steel option at a disadvantage energy-wise. In fact, partial or perforated ceilings are available for composite decks, which allow most of the cooling load reductions applied to the concrete building.

The above exercise demonstrates the need to work with the design team and allow them to challenge the model.This type of exercise

is not an exact science and needs to be interrogated by expert consultants, but hopefully this article has helped to put some perspective on the commercial and environmental issues.

Acknowledgments

Thanks to Nick Cullen, research and development manager of engineer Hoare Lea, and staff at both Corus and the Concrete Centre.