For many years, carburising has been amongst the most widely used processes for developing a hard, wear resistant layer on steel components. Fluidised bed furnaces have proven themselves to be very effective in carrying out the carburising process. Whilst conventional atmosphere carburising can be carried out in the fluidised bed, the boost/diffuse technique is the preferred option, with cycles carried out between 920°C and 1000°C. During the boost period, enriching hydrocarbon gas in excess of the desired carbon potential is passed through the furnace. This enriching hydrocarbon is generally natural gas or propane and is mixed with a carrier gas such as nitrogen to dilute the hydrocarbon, Following this boost period in which the surface of the material is saturated with carbon, the component then undergoes the diffuse period, whereby only nitrogen is present and the absorbed carbon diffuses towards the core of the material. The rate of carbon absorption is controlled by the carbon potential present in the furnace, as well as the surface activity and solubility limit of the steel, whilst diffusion is controlled by Ficks Law. As with vacuum and plasma carburising the surface carbon and effective case depth can be calculated by a computer program to give optimum times.

Advantages of Using a Fluidised Bed Furnace

There are a number of advantages gained by using fluidised beds. The major difference between carburising in fluid beds and conventional furnaces is the extremely high carbon potentials produced in the fluid bed during the boost period. Research has shown that within minutes of passing the hydrocarbon gas into the furnace it has broken down to produce free carbon particles causing the bed to reach a high carbon potential. For this reason treatment times in fluid beds can be significantly shorter than in conventional furnaces. Other advantages of the fluid bed include good temperature uniformity and excellent heat transfer properties which translates to better process control and uniformity.

The Importance of Controlling the Carbon Profile

However, even with the latest furnace technology, manufacturers are often unsure as to what are optimal case properties. It is clear that the carbon profile is crucial in producing optimal properties and performance on a case hardened component. In fact it is the carbon profile that generally determines the microstructure and hardness profile of the final treated component. For example, if the carbon content at the surface is too low then the resultant hardness and wear resistance will be subsequently poor. Conversely, too high levels of carbon leads to excessive retained austenite and the formation of carbide networks at grain boundaries which can lead to premature fatigue failure.

The Hardness Profile and Carbon Gradient

What is considered particularly desirable is a relatively high surface hardness that extends into the steel producing what is referred to as a flat hardness profile. This type of hardness profile can provide a number of advantages. Firstly, in comparison with standard hardness profiles, the flatter profile will have higher hardness values at similar depths, and therefore improved wear resistance over the life of the component. Furthermore it has been shown that higher carbon levels can result in small spheroidal carbides forming within grains which have been shown to improve wear resistance. Secondly, a great number of carburised components are machined or ground after case hardening in order to bring them back to within operating tolerances. It is common for up to 0.2 mm of material to be removed from the component in the final machining or grinding process. The problem with this practice is that in typical case depths of up to 1.00mm, this is removing 20% of your hardened zone, and once removed the underlying surface has considerably less carbon and therefore lower hardness and wear resistance. Considering these points it is therefore important to establish the correct ratio between boost and diffuse times to obtain a carbon gradient that will produce the desired transformation sequence and case depths upon quenching.

The Pulsed Boost Diffuse Technique

Tests have indicated that merely increasing the boost period to obtain higher carbon levels leads to excessive retained austenite and/or carbide networks. There is however, an alternative to the standard carburising cycle, which gives greater control over the final carbon profile. This technique is known as the Pulse Boost Diffuse technique (PBD). The key is to shorten the boost and diffuse periods but repeat them a number of times during the one cycle at the same temperature. This ensures that high levels of carbon are maintained at increased depths, hence producing a flat hardness profile. An additional advantage of this technique is that small spheroidal carbides can nucleate within the microstructure as a result of the higher carbon levels.

Case Study – Pulse Diffuse Boost Treating of 9310 Steel

A series of experimental trials was therefore conducted to evaluate the effectiveness of the Pulse Boost Diffuse treatment and determine if indeed a superior carbon and hardness profile could be obtained. The sample material was 9310, a commonly used carburising steel and one which can provide difficulties when attempting to achieve higher carbon levels due to the high nickel content (up to 3.55%) acting as a retained austenite stabiliser and causing excessive carbide networking.

Experimental Procedure

All tests were conducted in a fluidised bed furnace at 950°C using natural gas as the enriching gas and high purity nitrogen as the carrier gas. All samples were machined from 93 10 barstock.

Standard Boost and Diffuse Carburising Cycle

The first test (cycle 1) involved performing a standard boost and diffuse carburising cycle to obtain an effective case depth of 1.00mm. A computer program based upon the Harris Equation was used to determine the boost and diffuse times. It was calculated that a boost period of 47 minutes was required, followed by a diffuse period of 113 minutes. On completion of the carburising cycle the samples were direct oil quenched followed by a temper at 180°C. Finally, each sample was cryogenically treated in liquid nitrogen to transform any retained austenite and a second temper carried out. This standard cycle was performed not only to evaluate results but also to act as a standard against which the Pulse Boost Diffuse cycles could be compared.

Pulse Boost Diffuse Carburising Treatment Cycle

The second stage of the experiment was to carry out the Pulse Boost Diffuse treatments. This initially involved dividing up the standard cycle into shorter boost and diffuse periods and repeating these a number of times. The first Pulse Boost Diffuse cycle was 15 minutes boost and 25 minutes diffuse repeated four times, producing a total cycle time the same as the standard cycle. A second FBI) cycle reduced the boost period to 10 minutes and extended the diffuse period to 30 minutes also repeated four times. As with the standard cycle these samples underwent an initial temper followed by subzero treatment and final temper after carburising. It is important to note that the total treatment time was kept constant in order to keep the effective case depth to 1.00mm. All other variables such as quench rates were kept constant. After the trails each specimen was cut, mounted in epoxy resin, etched and the microstructure examined. Finally, each sample underwent a traverse hardness test to determine the hardness profile. This was carried out on a Vickers microhardness tester using a 100g load. A summary of the cycles carried out is given in Table 1.

Cycle	Carburising Temp (°C)	Boost Period (min)	Diffuse Period (min)	Repet-itions	Quench	Post Carb. Treat
1	950	47	113	1	Oil	Temper, Subzero, Temper
2	950	15	25	4	Oil	Temper, Subzero, Temper
3	950	10	30	4	Oil	Temper, Subzero, Temper

Results

Plotting hardness versus depth clearly demonstrates the difference between the cycles and ultimately provides an indication of the final carbon profile produced. Graphical representation of results for the various cycles is shown in Figure 1.

Figure 1. Hardness profiles of 9310 steel after the various carburising treatments.

Hardness Profile for the Boost and Diffuse Treatment

The standard carburising cycle (cycle 1) resulted in a surface hardness of 700HV that begins to taper off after only 100 microns down to a core hardness of 380HV The effective case depth was approximately 0.9mm (determined at 500Hv. The microstructure for the standard cycle was predominantly tempered martensite with a very few scattered carbides.

Hardness Profile for the Pulse Boost Diffuse Treatment

The hardness profiles produced by the two Pulse Boost Diffuse cycles were similar, however both of which were considerably different to that produced by the standard cycle. Both cycle 2 and cycle 3 had a high surface hardness (760Hv) as compared to the standard treatment. This higher hardness has been maintained to a depth of 0.3mm hence producing a flat hardness profile, At approximately 0.8mm all three profiles come together to give an effective case depth of between 0.9mm and 1.00mm. In all cases core hardness was between 360 and 400HV.

Microstructural Differences

The main difference between the two Pulse Boost Diffuse cycles was in the case microstructure. In both cases the matrix was composed of tempered martensite, but cycle 2 produced a microstructure with strong grain boundary carbide networks. Cycle 3 however, produced a marked improvement in microstructure. Although some large carbides are present, they are evenly dispersed and there is no carbide networking at grain boundaries. Figure 2 indicates the microstructures produced by each cycle.

Figure 2. Microstructures of 9310 steel after the various carburising treatments.

Discussion

Modifications to the standard carburising cycle in order to alter carbon levels and nucleate spherical carbides is not new technology. In fact, carbide dispersed carburising was first proposed by O.E. Cullen in 1961, and has been covered in a number of papers since. However, these methods have required that a second furnace be available in order to nucleate carbides at a lower temperature once the boost period is complete. This technique, whilst successful in producing higher hardness values and intergranular carbides, is not practical in the heat treatment industry. Therefore, a major advantage of the Pulse Boost Diffuse cycle is that it can be carried out in a single furnace at one temperature without interruption, and with modern programmable controllers the cycle can be carried out with a minimum of supervision.

Effect of Treatment Cycle on Hardness

In comparing the results obtained from the experiments, it is clear that the Pulse Boost Diffuse technique has the capability to produce a superior carbon profile, and subsequently an improved hardness profile. This is illustrated in Figure 1, as both cycle 2 and cycle 3 have achieved a greater surface hardness and a considerably flatter profile whereby hardness has been maintained at increased depths. In fact the Pulse Boost Diffuse cycles have both maintained 700HV or above, up to 0.3mm within the case. This type of profile would provide excellent wear resistance on the component and extend service life considerably. It would also mean that any grinding or machining after carburising would have less of a detrimental effect on the case. It should be noted that the effective case depths produced by all three cycles were very similar, and beyond 0.7mm all hardness values are approximately the same. This means the Pulse Boost Diffuse cycle can be carried out without having to extend standard treatment times, and that controlled case depths can be achieved on critical components.

The Effect of Microstructure on Materials Properties

The effect of microstructure on the final properties of 9310 is also of considerable importance. Again the microstructure is strongly dependant on carbon content and achieving an optimum gradient. Lower levels of carbon are not easily detectable, but will result in reduced hardenability and wear resistance. Excessive carbon has a much more pronounced effect. Excessive carbon can firstly lead to unacceptable levels of retained austenite. In the trials conducted all samples were subjected to a subzero treatment in order to minimise retained austenite, but the effects of retained austenite must be considered. Retained austenite is generally not considered harmful when present in amounts less than 30%, particularly if finely dispersed, and in some instances a small amount of retained austenite is desirable.

However, in other applications such as heavily loaded gears, retained austenite is limited to less than 5%, given that austenite is unstable and can cause lower hardness, reduced wear resistance and premature failure.

Excessive Carbon

Excessive carbon has the additional problem of creating grain boundary carbides. As carbon levels increase, carbides begin to outline prior austenite grain boundaries. If these levels continue to increase the carbides will not only grow in size but network along the grain boundaries. Large particle carbide precipitates at grain boundaries are detrimental to impact toughness and contact fatigue properties. This situation has occurred in cycle 2 and it appears that the longer boost periods have lead to an acceptable carbon level. It is likely that a component with this microstructure would fail in service. It should also be noted that 9310’s high concentration of nickel acts as a carbide stabiliser, and is the likely reason that carbides were present in all cycles conducted. An additional factor to consider is the concept of the corner effect. Most components have a sharp edge of some description, which can also enhance the problem of having excessive carbon levels as a result of carbon diffusing from both sides of the corner. However, in sensitive components where this may pose problems copper plating or stop off paints may be utilised to minimise the problem.

Controlled Carbide Formation

Controlled carbide formation, can however, provide benefits. The volume, size and distribution of carbides in a case microstructure all have a distinct influence on wear resistance. Generally, wear resistance improves as the volume of carbides increase at the wear surface due to the very high hardness levels obtained by alloy carbides. It is therefore important to be able to produce finely dispersed carbides within the microstructure of a case hardened component, such as those produced by cycle 3.

The single most important thing to remember when fabricating a product from stainless steel is to ensure that the fabrication processes retain and do not compromise the intrinsic properties of stainless steel.

Fabrication of all stainless steels should be done only with tools dedicated to stainless steel materials. Tooling and work surfaces must be thoroughly cleaned before use. These precautions are necessary to avoid cross contamination of stainless steel by easily corroded metals that may discolour the surface of the fabricated product.

Tools and blades must be kept sharp and clearances, for example between guillotine blades, must be tighter than for carbon steels.

Advantages

Apart from the commercial reasons for choosing stainless steel as a material, like heat and corrosion resistance, another advantage is its fabrication properties. The austenitic grades in particular can be fabricated by all standard fabrication methods. Some techniques are more suited to particular stainless grades than normal carbon steels. This is well demonstrated for severe deformation procedures such as deep drawing.

Due to stainless steel tending to have higher strength and work hardening rates than carbon steel, some alteration to tooling and equipment may be required:

· A more robust machine may be required due to the high strength of stainless steel.

· For the same reason the capacity of a machine, such as a guillotine may be only 60% of its carbon steel capacity.

· Greater deformation than used for carbon steels may often be necessary due to spring-back after forming. For example in tube making a different roll set design to that used for carbon steel will be required as a higher degree of over-bending is necessary.

Fabrication Methods

Drawing

Cold drawing can produce components in 301, 302 and 304 with tensile strengths in excess of 2000 MPa due to work hardening. Producing components at these strengths is limited to very thin sections and fine wires.

With increasing section thickness the amount of cold work required to produce these strengths also increases to the point that it cannot be done practically. This is due to the surface of stainless steel work hardening more rapidly than the interior of the material. Work hardening the entire cross section of larger diameters requires exceptional and impractical forces.

Austenitic stainless steel can be used to produce deep drawn components that require very high elongations and thus stainless steel is widely specified for the production of hollowware.

Forming Speeds

Unlike carbon steels, work hardening rates for stainless steels mean that more severe deformation is possible at slower forming speeds. For forming operations normally performed at high speed, like cold heading, it is recommended that the process is slowed.

Cutting

Most grades of stainless steel can be cut using standard cutting methods employed for other metals. The work hardening rate of stainless steels sometimes means heavier equipment and specialist blades or cutting edges are required. Consideration must also be given to changes in the heat affected zone when the cutting method generates high heats along the cut.

Common cutting methods include:

· Plasma cutting

· Laser cutting

· Water jet cutting

· Bandsawing

· Slitting

· Guillotining

· Abrasive disc cutting

Bending

Like many other fabrication methods for stainless steel, bending can be done using the same equipment used to bend other metals. A difference is that working hardening rates may mean a need for more rigid equipment and higher power levels. Equally, equipment capacity will be much lower than for Carbon steels.

Bar and Flat Bending

Round bar, flat bar, sheet and plate can be bent using a press brake, bending machine or ring-rolling. Due to work hardening, bending should be done quickly. Some over-bending will be required to counteract spring-back of the bend. The inside bend radius should not be smaller than the thickness of the material being bent.

Tube Bending

Tube bending is often done for architectural and other applications. Bending of stainless tube can be difficult unless done by persons experienced in the field. Rotary bending and hydraulic press bending can both be employed to bend stainless steel tube.

Centre line bend radii should not be smaller than twice the tube diameter.

Welding

Most grades of stainless steel can be welded by all traditional welding methods. However, the weldability of different grades can vary considerably. Some austenitic grades are considered to be the most readily welded metals. Weldability is generally low for ferritic and martensitic grades. For some grades, such as 416, welding is not recommended at all.

Recommended filler rods and electrodes vary depending on the grade being welded.

Machining

A commonly held belief is that stainless steel is difficult to machine. Machining can be enhanced by correct grade selection and using the following rules:

· Cutting edges must be kept sharp. Dull edges cause excess work hardening.

· Cuts should be light but deep enough to prevent work hardening by riding on the surface of the material.

· Chip breakers should be employed to assist in ensuring swarf remains clear of the work

· Use lubricants in large quantities.

Stainless steels like 303 and free machining grades have Sulphur included in the composition. This produces a marked increase in machinability but reduces both corrosion resistance and weldability. Over the years manufacturers of stainless steel bar have greatly improved the machinability of standard 304 and 316 grades. This has been achieved by careful control of compositions and production process variables that avoid the detrimental effect seen with Sulphur-bearing grades. This means the point has now been reached where many specifiers and machine shops have switched away from 303.

Finishing

Although stainless steels have excellent corrosion resistance, and are often selected for this property, proper finishing is required to maintain it. After any fabrication process that alters the surface condition of the material, the stainless steel needs to be degreased, cleaned and finished appropriately.

Finishing methods may include one or more of the following:

· Pickling

· Passivation

· Grinding

· Electropolishing

· Mechanical Polishing

· Blackening

· Colouring

Pickling

Pickling uses an acid or mixture of acids to remove scale produced in high temperature operations like welding, heat treatment or hot working. Acids and procedures depend upon the grade of stainless steel being treated. A great deal of care must be taken during the pickling process and with disposal of the waste as the acids used include sulphuric acid, nitric acid and hydrofluoric acid.

Pickling also removes rust due to corrosion of the stainless steel or corrosion of contaminant iron and steel particles.

The scale is removed as it retards the corrosion resistance of the underlying stainless steel.

Pickling is commonly done using baths or “Pickling Paste”. Pickling paste is a specially prepared stiff paste of strong acids. In this form it can be applied to vertical or overhanging surfaces and localised areas. Pickling paste is often employed to remove post-weld discolouration.

These are strong acids and appropriate caution must be taken when handling them. The same applies with acids used in passivation.

Passivation

Passivation is a process used to remove any free iron contamination of the stainless surface. Iron in the form of elemental iron, cast iron, carbon steel, mild steel or other non-stainless alloys can cause problems with stainless steels. This material is normally deposited from tools or work surfaces during fabrication. The iron particles promote corrosion on the surface of the stainless steel. At it’s mildest form, the discolouration caused can be unsightly. More serious corrosion is isolated pitting corrosion at the point of contamination.

Passivation involves treatment of the stainless steel with nitric acid or a nitric acid and sodium dichromate combination to remove the iron contaminants.

Grinding

Stainless steels are readily polished and ground if standard techniques are slightly modified. A build up of material on abrasive media occurs due to the high strength of stainless steels. Low thermal conductivity means that there is also a build up of heat. The result can be heat tinting of the surface of the material.

Using low grinding and feed speeds combined with specifically selected lubricants and grinding media can alleviate these complications.

Corrosion resistance of stainless steels tends to increase with the extent of surface polishing.

Electropolishing

The reverse of electroplating is electropolishing. This is an electrochemical process that removes the peaks of the rough surface of a metal. Electropolishing smooths the material surface making it brighter.

The resultant finish is very corrosion resistant, hygienic and attractive. On some stainless steels the surface finish appears frosted rather than smooth and reflective.

Mechanical Polishing

Any mechanical polishing procedures must be carefully done to ensure contamination by iron based materials doesn’t occur.

Sand blasting must be done with clean silica or garnet sand. Shot, grit and cut wire blasting must use stainless steel media of equal or greater corrosion resistance than the metal being cleaned. Barrel and vibratory finishing are often used to polish fittings and small parts.

Light heat tint can be removed by wire brushing but the brushes must be stainless steel and never be used on other materials.

Mechanically cleaned parts are not as corrosion resistant as pickled stainless steel. This is due to mechanical cleaning not removing all the chromium depleted material from the surface and the retention of some scale residue. Mechanical cleaning is often used to prepare the surface of stainless steel before pickling.

Blackening

Occasionally a highly polished surface is not wanted for stainless steel. In this case blackening is used to produce a non-reflective black oxide surface. Several methods can be used to achieve this finish.

Some treatments are proprietary but two common methods are immersion in a solution of sulphuric acid and potassium dichromate solutions or immersion in a molten salt bath of sodium dichromate.

Colouring

Stainless steels can be given a range of surface colours for architectural applications. These colours include bronze, blue, gold, red, purple, black and green. A range of shades can also be produced.

The colouring is done by a proprietary process that involves immersing the stainless steel in a hot chromic/sulphuric acid solution. This is followed by a cathodic hardening treatment in another acidic solution. The base material reacting with the hot acid produces a transparent film. Although the film is colourless, light interference imparts a colour to the layer. If the underlying metal surface is highly polished the effect will be a strong metallic lustre. For matt and satin finished stainless steel, the result will be a matt finish.

The old breakers’ yards, going for a long time, are soon to disappear. The future is now in recycling components from these vehicles and all as a consequence of a new Directive approved by the European Union. The new law came into being in the Spanish State in December 2002. From February of this year the Royal Decree for the Direction General of Traffic obliges owners of vehicles to obtain a certificate of destruction in order to be legally free of contractual ownership of the vehicle. Here, in the Basque Country, the Sestao-based Car Recycling company is the first of its kind.

The new European Union-approved Directive on non-useable vehicles prohibits forthwith the leaving of vehicles without them having previously undergone special. It is prohibited to accumulate vehicles in any place and, before breaking them up for scrap metal, any contaminant parts, liquids and gases have to be removed.

The European body has proposed the goal of recycling 85 % of vehicle components and converting 5 % into energy by the year 2006. In most European scrap yards, cars are piled up in fields with no kind of care or control. With this new legislation the idea is to avoid contamination of the Earth and the environment but also to make savings on raw materials and energy.

In the Basque Autonomous Community some 50,000 vehicles are taken out of use annually. In Europe, a total of some 150 million every year. The company Car Recycling of Sestao has already started to operate with the new legislation. Last year they processed 2,000 vehicles; this year they propose to treat some 6,000 and, from 2005 on, they hope to deal with about 9,000 vehicles a year.

Car Recycling receives automobiles from car dealers, from local council and also from individual members of the public. The cranes move some 40.000 kilos of scrap metal a year which are fed into the breaker and subsequently end up in the steelworks. Car Recycling has sited its plant next door to the Sestao ACB (compact steelworks).

Decontaminate and classify

Once the vehicles arrive at the recycling plant, the technicians begin to decontaminate and classify them, before being sent to be broken up. They are directed to one of three zones, depending on which market they are destined for. Pieces of plastic and rubber wheels are, for the time being, taken to the rubbish tip but. In the future, they will be elements in an integrated recycling process. The vehicle is weighed and this is when the process begins. First the vehicle must be decontaminated, i.e. the wheels, plastic parts and the battery must be removed; the oil is collected and the anti-freeze and brake liquids and then the fuel; finally the gases – the air conditioning and airbag gases – are extracted. All these contaminants are stored and distributed to the authorised companies for the treatment of waste.

The vehicles are subsequently taken apart, part by part, and classified for re-use. All the engine parts, gear change components, and so on, are re-bored – a practice widely used in Central Europe – and these recycled parts used in new cars.

According to Santiago Perea, Director-General of Car Recycling, every part recovered has one an end-market or another. There is a local market of mechanical parts or exterior accessories for old or antique cars. These are local workshops, the breakers’ yards; another market takes the bodywork, large parts for re-manufacturers which reconstruct them to be sold as spare parts; a third market are the third world countries which consume both second-hand cars and spare parts, starter motors, alternators, etc. of nearly all models.

When the parts are taken out, the rest is broken up for scrap. The crane takes the cars, one by one, and introduces them into the crushing machine. The vehicle is crushed again and again until it forms a compact packet. Once the volume is small enough, it is placed alongside the rest of the crushed vehicles, and from here they are taken to the steelworks as furnace scrap.

Car Recycling will soon be opening centres, not only in Alava and Gipuzkoa (the other two provinces within the Basque Autonomous Community), but also in other places within the Spanish State and Portugal – a total of 40 plants will be opened.

In order to facilitate the recycling process, in several zones in Europa, car manufacturers have begun to make parts in a diffferent way. If, up to now, some 22 types of plastic have been used, the idea is to make to with just 2 or 3 in the production of vehicles. Changes in design are also being undertaken in order to facilitate taking the vehicle apart.

Hardenability is the term used to describe the effect of various elements on the transformation characteristics of steels and the resultant microstructure after cooling from the austenitising (or hardening) temperature.

Depth of Hardening

In plain carbon steels the depth of hardening will be shallow, even for small sections given a relatively fast quench rate. As the alloying element content increases so does the depth of hardening. This means that the alloy steels can be hardened in many cases right through sections to the desired microstructures and give optimum service life.

It is at the design stage that these basic considerations should be made and the right steel grade chosen for the application.

Specifying Heat Treatments

In the final analysis, it is the heat treater who is asked to transform an expensive shape into a serviceable product. Remember, it is not the heat treaters function to guess the steel grade or required heat treatment. Full details of the steel grade must be given together with a property specification sheet. By far the most commonly specified property is hardness and it is normally given as a range in either Brinell, Rockwell or Vickers scale. It is normal for the heat treater to hardness test sample components and the area of the component which can be treated for a hardness test should be indicated.

Tensile Strength and Hardness

For heat treated steels a relationship exists between tensile strength and hardness. By referring to published figures the hardness values can be converted to tensile strength usually measured in N/mm² or MPa.

Properties That Can be Affected by Heat Treatment

When service conditions demand higher hardness areas on components to resist wear, abrasion, erosion, indentation or increase fatigue resistance, various treatments can be specified.

Surface Hardening

Case carburisation followed by quench hardening is a process which is well documented and has occupied a leadership role in surface hardening technology for decades. It is still the major surface hardening treatment with wide application in the gear industry and automotive components manufacturing.

Nitriding

Nitriding and various related processes are lower temperature solid state reactions which as a consequence are very valuable for surface hardening items requiring minimal distortion during processing.

By using special steel grades the surface hardness of nitrided components can be very high, say 1100 Vickers compared to 750-800 Vickers for a carburised and quenched structure.

It is important to note that for nitriding to be successful, the steel used should be hardened and tempered or the components should be rough machined then hardened and tempered, usually to a hardness range of HRc 30 - HRc 35. Nitriding must be carded out on a correct microstructure of tempered martensite to minimise the occurrence of spalling.

Flame Hardening

This process consists of hardening the surface of an object by heating it above the transformation temperature using a high intensity flame from a specially designed burner and then progressively quenching it in water, oil or suitable synthetic quenchant. The depth of hardened layer can range from 1 to 10 mm.

A considerable level of skill and understanding is required for the process and depending upon the complexity of the part to be hardened, a large number of variables need to be examined before the job should proceed.

Induction Hardening

This process is capable of a high degree of automation and as such can be used for high volume production. It involves heating the component by an induced eddy current to temperatures at which the rate of formation of austenite is rapid and then quenching to transform austenite to martensite. Induced currents are generated in the metal surface using specially designed inductors which suit the surface shape and using an alternating current in the inductors at frequencies from 50Hz to 1000kHz, depending upon the depth of hardening required. The advantage of induction hardening is that the process is quick, efficient and surface degradation is minimal. The disadvantage is that capital and set up costs are high.

Concrete is a very durable material. Fine examples of its first structural use by the Romans are still standing, and concrete is now probably the most widely used building material in the world. During Roman times and for many centuries after, its use was limited to compression structures, because of its poor tensile strength. But in the 19th century, the introduction of iron rods into the material led to reinforced concrete as we know it today, with its incredibly wide range of uses.

Iron and Steel Reinforcing in Concrete

Iron and steel rods cause potential corrosion and durability problems, however. Embedded steel is generally very durable, as it is protected from corrosion by the alkaline environment of the concrete. But in highly aggressive environments, the protection given by the concrete is often insufficient. The protective layer is broken down and corrosion begins, the initial signs being cracking and spalling of the concrete. Expensive remedial work is needed to repair this damage if the structure is to achieve its intended service life. Such repairs form a major part of the workload of the construction industry.

Tackling the Problem of Steel Corrosion in Reinforced Concrete

Tackling the problem of steel reinforcement corrosion has usually meant improving the quality of the concrete itself, but this approach has had only limited success. More recently, the construction industry has considered alternative steels for reinforcement, replacing carbon steel with stainless steel or using bars with an epoxy coating. In extreme cases cathodic protection is installed, although this is usually as part of a repair system and not for new structures.

Fibre Reinforced Plastic Reinforced Concrete

Now, the latest idea is to replace the steel with fibre reinforced plastics (FRPs). These materials, which consist of glass, carbon or aramid fibres set in a suitable resin to form a rod or grid, are well accepted in the aerospace and automotive industries and should provide highly durable concrete reinforcement. The durability is a function of both the resin and the fibre, while the amount and type of fibre are keys to determining the mechanical properties of FRPs. The strength of FRP reinforcement tends to be between that of high yield reinforcing steel and prestressing strand - about 1000 MNm^-2 for glass fibres and 1500 MNm^-2 for carbon fibres. However, the stiffness is generally much lower - about 45 GNm^-2 for glass fibres and 150 GNm^-2 for carbon fibres. All FRP materials have a straight line response to failure with no plasticity.

Manufacturing and Limitations of FRP Reinforcing Elements

FRP reinforcing rods are normally made by pultrusion. One limitation of this method is that thermoset resins are generally used and so once the material is fully cured, the rods cannot be bent into the range of shapes currently possible with steel. New manufacturing techniques are being developed to make such ‘specials’. Spiral reinforcement, both circular and rectangular, is being produced by several Japanese manufacturers, as are two- and three-dimensional grids. Other techniques are being developed in which resin-impregnated fibres are wound onto mandrels to produce closed shapes, such as shear links. As an alternative, thermoplastic resins are being developed that would allow the fully cured material to be warmed and bent to shape. However this is likely to give weaker reinforcement where the bar is bent due to misalignment of the fibres.

History of FRP Concrete Structures

The potential of FRP concrete reinforcement has already been shown around the world by the construction of many demonstration structures. Initially, owing to concerns about the lower stiffness of FRPs compared to steel, most structures were pre-stressed, with conventional steel being used as secondary reinforcement. A number of footbridges and highway bridges have been built, mainly in Japan and North America.

Highway Bridge

The first major European structure was built in Dusseldorf in 1987 - a highway bridge with glass FRP pre-stressing cables. Later demonstration structures formed an important part of the Eurocrete project, which was the first co-ordinated European programme of development work on FRP reinforcement. Eurocrete was a collaborative research project between partners in the UK, France, the Netherlands and Norway funded partly under the Eureka scheme. It was probably the first project of its kind in the world to bring together all the disciplines involved with FRPs, including materials suppliers, processors, research organisations and designers.

Footbridges and Non-Magnetic Fencing

Two footbridges were built during the Eurocrete project, one at Chalgrove near Oxford, and the other in Oslo. Part of a berthing facility at docks in Qatar was also constructed using FRP reinforcement, and another application was as reinforcement for the concrete fencing around a test facility for sensitive electrical equipment where conventional steel bars would have caused magnetic interference, (figure 1). Many applications were tested in the laboratory and may move into practice shortly, including retaining wall units and cladding panels. Meanwhile, other programmes are developing larger structures fully reinforced with FRPs, such as an 80 metre-long footbridge in Denmark.

Figure 1. Non-magnetic concrete security fencing erected around a test facility for sensitive electrical equipment.

FRP Concrete Standards

As FRP-reinforced concrete is being developed, design standards for its use are also being drawn up around the world. When introducing a new type of reinforcement with very different properties, there are two approaches - adapt the existing approach, or go back to square one and write completely new rules. The second is obviously more technically correct, but is a costly and time consuming process. As real applications are the only way to get good experience of the behaviour of a new material, modifying existing standards is the only feasible option.

The current standards for the design of reinforced concrete structures have developed over the last 100 years or so. They combine methods based on sound scientific principles and certain rules of thumb. For example, as reinforced concrete is a composite material, some aspects of its behaviour, such as shear, are still not well understood and so empirical approaches are used. FRP-reinforced concrete will follow similar rules to steel-reinforced concrete, but will differ in a number of ways.

Much experimental work has been carried out using FRP-reinforced concrete, mainly on simple beams and slabs, and basic design methods are being developed in a number of countries. The Japanese Ministry of Construction has published draft guidelines for design, the Canadian Bridge Code will shortly have a chapter dealing with FRPs and the American Concrete Institute is preparing guidance. Proposed modifications to British Standards covering the design of reinforced concrete structures were developed under the Eurocrete project and are now being validated by the Institution of Structural Engineers. They will provide a document for use by design engineers in the absence of a formal code of practice. These design approaches will lead to safe structures, but are unlikely to lead to the most economic use of the relatively expensive FRP materials. The cost of FRP rods is expected to be between that of epoxy-coated steel and stainless steel, two to eight times as expensive as normal steel bar. Such a high initial cost can only be justified by looking at ‘whole life’ costs for structures in aggressive environments. Potential users need to consider the total costs for their structures, including repairs, and not just the material costs. In the future, such savings should become obvious as design approaches are developed which take account of the enhanced properties of FRP-reinforced materials.

Differences between FRP and Steel Reinforced Concrete

· Because of the high strength and relatively low stiffness of FRPs, failure is likely to occur by compression of the concrete and not rupture of the reinforcement.

· Crack widths in steel-reinforced concrete are controlled to prevent aggressive substances reaching the steel, so improving durability. For FRP-reinforced concrete, aesthetics and possibly watertightness will be the only criteria for crack width control.

· Deflections are likely to be higher than for equivalent steel-reinforced units.

· FRP rods have low compressive strengths in comparison to their tensile capacities, so the traditional design approaches for columns are no longer valid. Studies looking at the effect of wrapping FRP around circular columns have found that the confinement leads to increases in the failure load and the failure strain.

· Fire will be a design consideration for some types of structures. The main concern is to limit the temperature rise at the surface of the FRP bar, so that it stays below the glass transition temperature of the resin. Above this temperature, the material stops acting as a composite, and so weakens.

Problems Associated with FRP Concrete

Durability

The major cause for concern in the use of FRPs as reinforcement is probably the durability of the material when embedded in concrete. The highly alkaline environment degrades glass fibres and some resins, and manufacturers are reluctant to disclose the details of the materials they use for commercial reasons. Work has concentrated on developing alkali-resistant glass and on using carbon and aramid fibres, but little attention has been paid to the resin. Ways of assessing the durability of the materials are urgently needed, but considerable work still needs to be done to develop acceptance criteria.

A major assessment of durability was carried out in Eurocrete, which included work on the materials themselves and on FRPs embedded in concrete. The latter samples were stored in laboratories under various environmental conditions and also on exposure sites in Europe and the Middle East. The results, which apply to the particular resin and fibre combinations studied, show that the composite rods resist the alkaline environment well, with no significant degradation during the test period.

Industry Acceptance

Despite its excellent properties and durability, FRP reinforcement is unlikely to replace steel for the vast majority of structures in the foreseeable future. Experiments and demonstration projects around the world have shown that FRP reinforcement is a viable and cost effective alternative to steel in special circumstances, for example as an alternative to stainless steel. But the construction industry is extremely conservative, and so the most likely development route is the use of the new materials in non-structural applications or in ones where the consequences of failure are not too severe. More highly loaded and critical applications will follow later as confidence in the materials grows.