This Paper is a Contribution to the International Symposium

"Sustainable Construction:
Use of Recycled Concrete Aggregate"

11-12 November 1998

University of Dundee, Concrete Technology Unit, London (UK)

Verification of the Dimensioning Values for
Concrete with Recycled Concrete Aggregates

Dipl.-Ing. Frank Roos
Research Assistant
Lehrstuhl für Massivbau
Univ.-Prof. Dr.-Ing. K. Zilch
Technische Universität München

ABSTRACT. In 1996 a new law called „Kreislaufwirtschafts- und Abfallgesetz" (circulating economy and waste material law) was introduced in Germany. It specifies that the person who produces, sells or consumes a product has to later recycle or dispose of the remaining waste material respectfull of the environment. The best way to do this for used building material is to produce new concrete. The research project „Baustoffkreislauf im Massivbau" (building material cycle in structural engineering) led to a new construction guideline called „Beton mit rezykliertem Zuschlag" (concrete with recycled concrete aggregates) published by the DAfStb (German Committee for Reinforced Concrete). This paper presents the work performed by the Technische Universität München which deals with the verification of the dimensioning values. Mainly creep, shrinkage and bond tests were performed. This project is supported by the BMFT (Bundesministerium für Forschung und Technologie, Federal Ministry for Research and Technology) and the industry and is for the first time published in the world wide web. The address is: http:\\www.b-i-m.de.

Keywords: Recycled concrete aggregates, Creep tests, Shrinkage tests, Bond tests, Dimensioning values.

Dipl.-Ing. Frank Roos was born in 1969 in Stuttgart, Germany. After his school studies and military service he studieed civil engineering at the Technische Universität München. After completing his studies he was employed by the Schöndorfer Bau- und Umwelt GmbH which is a company specialialized in demolishing buildings. Since November he is a research assistant at the Lehrstuhl für Massivbau of the Technische Universität München conducted by Univ.-Prof. Dr.-Ing. K. Zilch. One of his main research tasks is the dimensioning of concrete structures made of recycled concrete aggregates.

INTRODUCTION

On the 20^th of February 1998 the new construction guideline „concrete with recycled concrete aggregates" [1] was presented to the specialized public. Therefore one and a half years after the German law called „Kreislaufwirtschafts- und Abfallgesetz" was passed in 1996 there is a guideline available for the responsible engineer, who is conscious of the environment.
According to the law anyone who produces, sells or consumes goods is responsible for the avoidance, utilization or environmental waste management. The old principle that the industry produces goods and the public pays for waste management is no longer valid. As a consequence, the building industry is forced to think about the recycling of building materials. Allready shortly after the second world war the use of crushed clay brick concrete was introduced in the German standards [2]. With the new guideline the use of crushed concrete brick and plaster as concrete aggregates will become possible in Germany.

From the crushed material shown in figure one 55 % can be used again in concrete constructions of high quality. To reach the demanded high standard of safety by dimensioning structures, as know from concrete with natural aggregates, it is necessary to verify the dimensioning values and if necessary modify them. This verification performed by tests on creeping, shrinkage and bonding was the task of the Technische Universität München within the framework of the research project „Baustoffkreislauf im Massivbau (building material cycle in structural engineering)" which is supported by the BMFT and the building industry. The new guideline is mainly based on the results of this research project. This paper deals with the performance and first results of the tests conducted at our laboratory in 1997 and 1998.

TEST PROGRAMM

To set up an adequate testing programm it is necesarry to have an aim. In this case, the effect of recycled concrete aggregates on dimensioning values and bonding behavior should be verified. For practical and economical reasons it is very interesting to know whether the dimensioning can be done with the existing standards especially with the DIN 1045-1988 [4] (German standard for structures made of concrete). It is very important to establish whether there is a loss of safety if the verification is done in the old way. This verification is simply done by changing the percentage of recycled aggregates and comparing it with concrete made of natural aggregates. At the laboratory in Munich 45 bond tests and 48 creep tests have been performed. Nine shrinkage tests are still running and at the end of this year 18 crack interlock tests will be finished.
At the beginning of the research project there was a lack of basic knowledge regarding the production of recycling concrete. As most of the above mentioned tests run over a longer period of time, we found at an early stage a mixture for the concrete with recycled aggregates even through this task was to be done at other institutes sharing in the project. To guarantee that the results of all the Institutes involved in the project are comparable the following data for the concrete mixture were fixed:

• B35 (f_ck,cube = 32 N/mm²) , consistence KR (slump between 0,42 and 0,48 m)
• Cement content 320 kg/m³
• Water-cement ratio 0,55
• Use of granulometric gradation 0/4 - 4/8 - 8/16
• Plasticizing concrete admixtures are permitted

Four mixtures were tested before the decision for the three mixtures used for the tests was made. The main difference of the three mixtures is the percentage of recycled concrete aggregates.

• RC-N: This is a normal concrete with 100% natural aggregates to enable a comparison.
• RC-A: Its particle size 0/4 are natural and its particle size 4/8 and 8/16 are recycled concrete aggregates.
• RC-B: This mixture is completly made of recycled aggregates.

The recycled aggregates were crushed concrete with scarcely any dirt. The grading curve was fixed in such a way that it always layed between the curve A and B acording to DIN 1045 [4].
To reach the above mentioned comparison of all results, the production process for the concrete was also fixed. First the aggregates were dried in the oven. Then they were mixed without the cement but with the hydroscopic water and 20% of the water needed for the hydration for exactly two minutes. After that they were covered with a PE-foil for ten minutes. The cement and the rest of the water were added and mixed for another two minutes. After making the slump a decision was made whether a plasticizing concrete admixture should be added. If necessary, the concrete was mixed for another two minutes.

Creep and shrinkage tests

The unset concrete and the hardened concrete properties are stated in table 1. As one can see the concrete with 100% recycled aggregates does not fulfill the demands of the DIN 1045 [4] table 1 and 2 concerning compressive strength and slump. Within the scope of our tests it was not possible to produce a concrete which fulfilled the given demands reliably. No problems occurred with the concrete RC-A and RC-N. So it seems that the main reason for the above mentioned difficulties is the recycled sand with a particle less then 4 mm and the difficulty in determining the hydroscopic water.

The percentage of the recycled material was restricted in the new guideline not only because of these difficulties. The abolishment of this restriction would mean further extensive research work to ensure reliable concrete properties.

For each mixture 10 cubes with 150x150x150 mm for the standard tests and 16 respectively 25 cylinders with a diameter of 150 mm and a height of 300 mm were produced for the creep and shrinkage tests.

The storage and testing took place in a laboratory of the Technische Universität München where a constant temperature of 20°C and a relative air moisture of 65% prevailed. The tests were performed with the same equipment which was introduced and used by Nechvatal, Stöckl and Kupfer [5]. The load was applied by a rubber bubble which was connected with a compressed air bottle.

Some cylinders of each series were sealed to reduce the influence of shrinkage on the concrete deformation. After the cylinders reached the age of 28 days they were installed and loaded with 30, 50 and 70% of their compressive strength. During the first four days the deformation was measured every two minutes by a UPM 100 followed by manual measurements in increasing intervals up to 14 days before the end of the running period of three months.

Always three cylinders of one series, one of them sealed, were installed in one testing apparatus. For stability and safety reasons only one cylinder was installed when high loads were applied.

The installation for the tests with 30 and 50% loads is shown in figure 2. Figure 3 shows a cylinder loaded with 70% of its compressive strength.

The occurrence of fractures on cylinders with 70% loading (Figure 4) at the beginning of the tests was a reason to have doubts about the long term strength of concrete with recycled aggregates. A single test with an 80% load applied by a hydraulic piston was then performed. Ten days later the cylinder showed first small and then bigger scales. However the test is still running and the cylinder is now seven months old. As this is the only test up to now it will be necessary to perform further tests to confirm this excelent result.

The shrinkage tests are running parallel to the creep tests. The cylinders are from the same series and stored under the same conditions as the ones for the creep tests. They are unloaded which is to compensate the shrinkage deformation during the creep tests.

Bonding tests

The most important factor for the dimensioning concerning the bond behaviour is the length of the steel anchorage. With pull out tests a bond slip relation was established and the length of anchorage was calculated. The interpretation and the comparison with normal concrete enables a decision to be made as to whether the dimensioning of structures made of recycling concrete can be done according to the DIN 1045 [4] or if the dimensioning values will have to be changed.

The bond tests were preformed with the same mixtures as the creep tests. For each series 12 Janovic [6] and 3 Rilem [7] test specimens were used. The concrete properties were tested with 7 cubes with 150x150x150 mm. The reached concrete properties are shown in table 2.
As in the creep tests it was not possible to achieve the concrete properties of the series RC-B as specified.

The test specimens (according to Janovic = J) are cubes with eccentric placed steel. The con-crete in which the steel is placed is separated in two areas by a steel plate with a styropor layer. The steel is only bonded at the back (see figure 5). The advantage of this specimen is that a sort of bracket results and there is no compressiv vault which jams the steel anymore. This construction shows in a very good way the bond behaviour in the bending tension area of a beam.
For the tests always three identical specimens were concreted. The concrete cover and the placing of steel during concreting were varied. The specimens with the top and bottom rein-forcement were produced in one formwork and splitted in a prepared place after 28 days. The length of the bonding for all tests was 70 mm. The steel properties are stated in table 3. It was a steel BSt 500S according to German standards with a diameter of 14 mm.

The related rip area f_R can be calculated exactly according to the ENV 10080 (B. 8) [8]. During the tests with the specimens (J) the force, slip and twice the elongation of the side with the risk of cracking were measured. Figure 6 shows an installed specimen ready for the bond test. To reduce the influence of the deformation of the concrete on the slip measurement the gauges were placed as near as possible to the steel. Schmidt-Thrö [9] described an optimal placing of gauges.

The pull-out specimen according to Rilem is very useful to show the influence of certain parameters on the bond stress. The steel is placed in the middle of the cube. The concrete is subjected to compression loading which has the consequence that the steel is jammed and the deformation of the concrete in the supported area is hindered. During this tests only the force and slip were measured.

All the specimens were stored 7 days in a room with 100% relative air moisture which was followed by a storage in the test laboratory at 15°C and 50% relativ air miosture. The tests were performed exactly 28 days after concreting. The series RC-B showed small cracks caused by shrinkage. The crack pattern and width were recorded before the tests. Figure 7 shows a typical crack pattern of a specimen according to Janovic. The concrete failed along the reinforcement in the bonded area.

TEST RESULTS

Creep and shrinkage tests

Due to the long duration of the tests, there are only single results available at the moment so it is not possible to fix the final calculation values for the creep deformation. Figure 8 shows the course of total deformation of the different concrete mixtures with 30% loading. As expected, the concrete with recycled aggregates shows a larger deformation.

Bonding tests

With these tests we got a basic bond slip relation for each serie and varied parameter. Figure 9 shows exemplary the result of the Rilem pull out tests. On the abscissa the slip is in a logaritmic scale and on the ordinate the related shearing stress. According to Müller [10] the

Influence of the concrete compressive strength is linear at least in the middle area, so that the stress can be divided by the strength to enable a comparison. With the ENV 1991-1 [11] it is true that it is possible to calculate a standard deviation using only three tests but the mean variation is then higher and might not be true for recycling concrete. Due to this each of the three curves is the calculated average of three tests.
Very interesting is that all the maximum of the shearing stress are nearly the same while at the beginning the curves run clearly separated. The concrete with recycled aggregates shows a much smoother bond behaviour then normal concrete. The adhere bonding force of the normal concrete is much higher. This force is insignificant for the total bearing capacity as it disappears with very small movements. The smoother bond behaviour of the recycling concrete might be regarded as the more important factor.
To answer the question regarding the consequences for the anchorage length of steel it is useful to set this basic bond slip relation in the differential equation of the movable bond equation (1) and with numerical integration it is possible to get the different anchorage lengths.

(1)

By taking the anchorage length of the RC-N as the basis and setting it to 100%, it is possible to determine a factor for the dimensioning value for recycling concrete according to equation 2.

(2)

This factor guarantees the security standard of the DIN 1045 if this standard is used for dimensioning. For the RC-A the factor F_RC-A is in the worst case 1,25 and for the RC-B 2,00.

According to the new German guideline the percentage of recycled material is restricted to about 30%. This restriction was made after the results of the research work on other institutes participating on the research project were available. When the tests were started there was the hope that it might be possible to produce a concrete completly made of recycled aggregates. The problem with the properties of the aggregates especially with a particle size less then 4 mm showed that it is opportune to reduce the percentage of recycled aggregates. If this is done according to the guidline (Table 1-1) [1] it should be possible to set the factor F_x to 1,00. By using natural sand the concrete properties are reliable and our tests showed that the factor declines very quickly by reducing the recycled material.

OUTLOOK

The use of recycled aggregates saves natural resources, dumping space and supports the environment. The first experience with a building in Darmstadt, Germany proved that the use of recycling concrete is no problem. The new guideline now enables safe dimensioning and consequently the use of this „new" building material. It can be hoped that the acceptance will rise as fast as the knowledge resulting from the new research work.

REFERENCES