- SELF-COMPACTING CONCRETE - INFLUENCE OF THE COARSE ... · discrepancy between the clearances in the slump flow test and the minimal acceptable clearance values according to DIN

- SELF-COMPACTING CONCRETE - INFLUENCE OF THE COARSE AGGREGATES

ON THE FRESH CONCRETE PROPERTIES

Tilo Proske, Carl-Alexander Graubner

SUMMARY

In the research program the influence of the shape and size of the coarse aggregates on the workability of Self- Compacting Concrete was analysed. Changing the mortar - coarse aggregate ratio the optimal powder contend was found. Hereby the consistence of the mortar was maintained constant. The optimising of the paste content has influence on the economic efficiency as well on the hardening concrete properties like creep and shrinkage.

1. INTRODUCTION

Self-Compacting Concrete (SCC) is one of the most important innovations in the concrete technology. The SCC must fill the formwork completely, enclose the reinforcement, deaerate only by the force of gravity and must not segregate.

The following technical and economic reasons advantages can be listed for SCC, when professionally applied:

• enlargement of the architectural variety through realisation of any forms • better quality in the area of reinforcement concentration • high und very homogeneous quality of the concrete, very dense concrete-structure • optically appealing surface

Proske, T.: Influence of the coarse aggregates on the fresh concrete properties. Darmstadt Concrete 18 (2003). http://www.darmstadt-concrete.de/2003/coarse.html

http://www.massivbau.tu-darmstadt.de/user/proske/web/proske.htm

http://www.massivbau.tu-darmstadt.de/massiv/index.htm

2/15

• simplification of the casting process, increase of the moulding-performance • increase of efficiency, reduction of staff • improvement of working-conditions • advantages for the environment and the conditions at work (noise) • lower investment costs and higher lifespan of the formwork

Unfortunately the absence of regulations and standards - especially in Germany – is preventing a generally application until now. However, at least we need a SCC with reliable fresh concrete properties and moderate coast for the material.

Fig. 1: Slump-Flow Test in conjunction with the J-Ring


3/15

2. RESEARCH PROGRAM

2.1 Overview

In the research program the effect of the aggregate shape as well as the maximum diameter on the workability of SCC was analysed. In the series 1 and 2 the coarse aggregate distributions had a maximum diameter ranging from 8 mm to 32 mm (Tab. 1). Additionally different materials like gabbro, limestone, granite and river aggregate (Fig. 2) were investigated. The distribution of the coarse aggregates was equivalent to DIN 1045 [2] “Distribution No. B”. The optimal powder volume was found by changing the mortar coarse-aggregate ratio. Hereby the consistence of the mortar was maintained constant, adjusted only by the superplasticizer.

The influence of the paste design on the optimal mortar content was verified in the series 3. The types of the filler (fly ash and limestone) and the cement as well as the filler cement ratio were changed.

Tab. 1: Systematisation of the test series 1-3

Series 1 Series 2 Series 3

Crushed aggregate Rounded aggregate

Paste composition

a b c (River gravel) (Gabbro)

Limestone • Variation of the material

D = 8 mm Granite •

8 mm • •

11 mm •

16 mm • • •

Variation of the maximum diameter

(River gravel and gabbro) 22 (32) mm • •

Limestone • Variation of the material

D = 16 mm Granite •


4/15

b

a

d

c

Fig. 2: Coarse aggregate fraction 11/16 mm, Granite (a), Gabbro (b), Limestone (c) und River gravel (d)


5/15

2.2 Shape of the coarse aggregates and packing density

The shape of the coarse aggregates was classified by the SI-value according to DIN 933-4 [3] (Tab. 2). Additional the particle shape was analysed by the Computerised Particle Analyse (CPA). CPA it is a photo optical method. The used test machine was HAVER CPA 4 (Bauhaus-University Weimar) [7], [8]. The shape of the particles can be described by parameters like sphericity (SPHT), ratio xmax/xcor and the equivalent diameter of the 2D particle projection, see Fig. 3.

Projection Circle of equal area

xmax - Maximal length of the particle

xcor - Corresponding width

D - Diameter of the circle of equal area

Spericity (SPHT):

U – Perimeter of the particle projection A – Area of the particle projection

D x max

x cor

A2USPHT

⋅π=

Fig. 3: Classification of the shape of the aggregates by CPA

Tab. 2: Classification of the shape of the coarse aggregates according to DIN 933-4

SI [%] 1) Coarse aggregate

Fraction River gravel Gabbro Limestone Granite

5/8 21 16 16 17

8/11 24 11 15 16

11/16 14 10 11 15

16/22 - 6 5 - 1) SI – Volume of non cubic particles with lmax/lmin > 3; lmax, lmin – Minimal/maximal

particle length


6/15

Tab. 3: Sphericity of the coarse aggregates by the Computerised Particle Analyse (CPA)

SPHT Coarse aggregate


11/16 1,083 1,176 1,158 1,198

The packing density of the aggregates and the mixture was calculated with the Compressible Packing Model (CPM) [5]. Therefore the exactly particle size distribution is required. The particle distribution of the powders was analysed by the laser granulometrie. Also the packing density of every aggregate (Tab. 4) fraction and of the powder particles had to be determined in experimental tests before. The compressible index K is characterising the packing process. The higher K the denser the packing. In the experimental tests K = 9 was used for the aggregates.

Tab. 4: Virtual packing density of the coarse aggregates by experimental tests (CPM)

βi

(Virtual packing density of the size class i)

Coarse aggregate


5 - 8 0,7293 0,6571 0,6627 0,6385

8 - 11 0,7455 0,6674 0,6686 0,6542

11 - 16 0,7433 0,6737 0,6760 0,6554

16 - 22 0,6676

It was found, that the packing density of the aggregates is not correlating with the shape classification according to DIN 933-4, if different materials are used. A better evaluation is possible using the spericity of the aggregates (Tab. 3) and the xmax/xcor -value. Here the surface condition of the particles as well as the shape is included.

For the optimised mixtures also the hardening properties were verified. The results are not presented in this paper.


7/15

3. REQUIREMENTS ON THE MIXTURES

3.1 General requirements

Still the primary application of SCC is seen in the precast technology. Here a high early

strength and a moderate workability loss must be provided. Further the mixtures were to

optimise in terms of the economic aspects. First of all the “valuable” content of the

powders and the additives should be low. Though to avoid the blocking an adequate

mortar volume is necessary. Therefore the water-powder ratio and the sand content was

chosen relatively high.

3.2 Mixtures design

To achieve early strength cement CEM II A-S 52,5 R was used as standard. The w/c-

value of 0,53 still attains the required strength and durability. Including the addition of

the fly ash the w/cequ – value is 0,48. The particle distribution of the aggregates is

approximated to the distribution No. B after DIN 1045 [2]. Because of the altered mortar

content there is a slightly difference in the sand content.

The mixture proportion of the mortar is presented in Tab. 5.

Tab. 5: Mixture proportion of the mortar

Series 1 and 2 Series 3 (Fly ash) Mass

[kg/m³] Vol. [%]

Mass [kg/m³]

Vol. [%]

CEM II 52,5R 531 17,0 474 20,5

Fly ash 136 6,0 237 10,0

Water 278 27,5 253 25,0

SP App. 4 App. 0,5 App. 5,5 App. 0,5

Sand 0/2 1277 49,0 1277 49,0

w/c 0,53 0,54

w/c equ. 0,48 0,48


8/15

3.3 Optimisation of the mixture

The optimal mixture had to fulfil specific requirements. First of all a good flowability

without segregation and no blocking is of importance. For the evaluation the test methods

according to the German code “DafStb-Richtlinie Selbstverdichtender Beton” [1] were

used.

The “Setzfließversuch” (Slump-Flow Test) and “Trichterauslaufversuch” (V-Funnel

Test) qualified the rheological behaviour. To evaluate the blocking risk the

“Blockierring” (J-Ring) was used in conjunction within the Slump-Flow Test. The

clearance between the steel bars is depending on the maximal aggregate diameter. But the

discrepancy between the clearances in the slump flow test and the minimal acceptable

clearance values according to DIN 1045-1 [1] is seen critical (Tab. 6).

Tab. 6: Number of the bars on the J-Ring and the according clearances according to [1]

Maximal diameter of the aggregates

Number of bars

Clearance between the bars according to [1]

Minimal acceptable clearance according to DIN 1045-1 [6]

D [mm] n a [mm] alim [mm]

8 22 25 20

11 19 32 20

16 16 41 20

22 13 55 27

32 10 76 37

To evaluate the blocking behaviour, the slump in the centre of the concrete pool was

measured as well as the average diameter of the concrete. It was figured out, that the

combination of diameter and slump value testifies the blocking risk very well. A larger

diameter demands a higher slump. The limits for the blocking criteria are presented in

Fig. 4. Further can be seen that the acceptable limits are depending on the maximal

diameter. The average level of the concrete pool is approximately 25 mm high


9/15

(depending on the spread diameter). Therefore an aggregate particle of 32 or 22 mm is

normally not completely integrated in the mortar. First of all this “loss of concrete

volume” effects the reduced diameter.

Further on the separation of the mortar and paste was observed at the brim of the concrete

circle. A brim less 3 mm was accepted for the optimal mixture.

The sedimentation of the coarse aggregates was evaluated by the “Auswaschversuch”

(rinse out test). A sedimentation value |∆m| ≤ 10 % is generally considered as acceptable.

It is to remark, that sometimes sedimentation was noticed in the mixer (after a short rest)

but not in the rinse out test.

Further on more the air content and the workability loss were tested.

60

70

80

2,0 2,5 3,0 3,5 4,0 4,530 cm - Slump [cm] (J-Ring)

Gabbro 8 mmGabbro 16 mmGabbro 22 mmRiver 8 mmRiver 16 mmRiver 32 mmGranite 8 mmGranite 16 mmLimestone 8 mmLimestone 16 mmLower limit 8/11/16Lower limit 22/32Upper limit 8/11/16Upper limit 22/32Blocking criteria

Slu

mp

flow

[cm

]

Seggregation

Blocking

good workability

Blocking andno flowability

No flowability

Fig. 4: Limiting criteria for the blocking behaviour (Series 1 and 2)


10/15

4. RESULTS

It was found, that the shape and size of the coarse particles has a significant influence on

the required mortar and paste content. Though the results must be seen in relation to the

test methods. Especially the blocking criteria has highest impact. As expected the

blocking risk increased if the mortar volume was reduced. A large maximum diameter of

the aggregates reduced the blocking risk (Fig. 5). An explanation is the greater clear

space between the bars if the J-Ring test [1] is used. Also the shape of the aggregates has

an influence on the optimal mortar volume. Naturaly rounded river gravel requires mostly

less mortar or paste than gabbro and limestone. Granite requires the highest mortar

volume.

2,00

3,00

4,00

5,00

6,00

55,0 57,0 59,0 61,0 63,0 65,0 67,0 69,0 71,0 73,0 75,0Mortar volume [%]

River 8 mmRiver 16 mmRiver 32 mmGabbro 8 mmGabbro 16 mmGabbro 22 mmGranite 8 mmLimestone 8 mmLimestone 16 mmGranite 16 mmSFA

30 c

m -

Slu

mp

[cm

] (J-

Rin

g)

Fig. 5: Blocking behaviour versus mortar volume (series 1 and 2, and concrete with high

fly ash content)

To maintain a high workability without segregation the dosage of superplasticizer had to

be adjusted. The lower the mortar volume the higher the optimal SP content, Fig. 6.


11/15

Fig. 6: Optimal content of superplasticizer

1,00

1,10

1,20

1,30

1,40

50,0 55,0 60,0 65,0 70,0 75,0Mortar volume [%]

River 8 mm

River 16 mm

River 32 mm

Gabbro 8 mm

Gabbro 16 mm

Gabbro 22 mm

Limestone 8 mm

SP

/Cem

ent [

m-%

]

It can be seen that generally the rounded river aggregates require the lowest mortar

volume. Using granite aggregates the SP content was highest. The presumption, that

aggregates with a high surface have a better resistance again segregation and therefore

require less paste volume could not be proved. Though less resistance again

sedimentation was measured using the rounded aggregates, implying an equal mortar

volume.

Using smaller aggregates less risk of segregation was observed. Particularly if there was

a considerable surplus of mortar.

An interesting phenomena was the decreasing of the required mortar volume by the

increasing viscosity, first of all if large aggregates are used (Fig. 7 and Fig. 8). The low

velocity of flowing in combination with high shear forces between mortar and aggregate

is reducing the blocking risk at the steel bars. The disadvantage is the higher content of

superplasticizer in the mixture.


12/15

Fig. 7 presents the optimal mortar and paste volume for the different aggregates.

74 7268 68 66 66 64 63

6064

72 72

0

25

50

75

100

8 8 8 8 11 16 16 16 16 22 32 Fly as

Gr Ls Ga Ri Ga Gr Ls Ga Ri Ga Ri Ga 16

h

Vol

. [%

]

MortarPasteCement

Legend - Name of the mixtures: Number - Maximal diameter of the aggregates [mm], Gr - Granite (crushed), Ga - Gabbro (crushed), Ls - Limestone (crushed) Ri - River gravel (natural), Fly ash - High fly ash content (high viskos)

Fig. 7: Optimal volume of mortar, paste and cement for different types of coarse

aggregates and a mixture of modified mortar rheology (high viscous)

The packing density of the aggregates and the mixture was calculated with the

Compressible Packing Model (CPM).

A correlation between the porosity of the coarse aggregates and the optimal paste and

mortar volume was found (Fig. 8). The higher the porosity of the aggregates the higher is

the required paste and mortar volume. Though the relation is non-linear. Fig. 9 is

presenting the relation between the porosity of the compressed coarse aggregates and the

optimal mortar (respective paste) volume. Obvious is the decreasing volume of mortar if

the maximum aggregate is lower.

Considering the whole packing of aggregates the relation between porosity and paste

volume is approximately linear (Fig. 9)


13/15

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,250 0,275 0,300 0,325 0,350 0,375

Relative porosity of the compressed coarse aggregates [-]

Paste volumeMortar volumePaste volume (high viskos)Mortar volume (high viskos)

Rel

ativ

e op

timal

pas

te a

nd m

orta

r vol

ume

[-]

Fig. 8: Relation between the porosity of the compressed coarse aggregates (K = 7) and the

optimal paste and mortar volume

1,0

1,2

1,4

1,6

1,8

2,0

2,2

2,4

2,6

32-Ri 22-Ga 16-Ri 16-Li 16-Ga 16-Gr 11-Ga 08-Ri 08-Ga Sand 16-GaFA

Rat

io [-

]

Mortar Volume/Porosity coarse aggregates (K = 7)

Paste Volume/Porosity aggregates (K = 7)

Legend - Name of the mixtures: Number - Maximal diameter of the aggregates Gr - Granite (crushed), Ga - Gabbro (crushed), Ls - Limestone (crushed) Ri - River gravel (natural), FA - High fly ash content (high viskos)

Fig. 9: Relation between the porosity of the compressed aggregates (K = 7) and the

optimal mortar and paste volume.


14/15

5. CONCLUSION

The results prove that the shape and size of the coarse aggregates has a significant

influence on the required mortar and paste content. Though the results must be seen in

relation to the test methods. Especially the blocking criteria has highest impact. In the

tests a large maximum diameter of the aggregates was reducing the required mortar

volume significant. Also the shape of the aggregates as well as the viscosity of the mortar

influenced the optimal mortar content. It was found a correlation between the porosity of

the coarse aggregates and the optimal paste and mortar volume. Applying smaller

aggregates less risk of segregation was observed.

The research project was financially supported by “Deutscher Beton- und Bautechnik-

Verein E.V.” and “Arbeitsgemeinschaft industrieller Forschungsvereinigungen e.V.”.

Further we thank Dr.-Ing. J. Nicolay, Dyckerhoff AG, Dipl.-Ing. E. Kleen, MC

Bauchemie GmbH, Dr.-Ing. U. Stark, Bauhausuniversität Weimar for the kindly support.


15/15


6. LITERATURE

[1] Deutscher Ausschuss für Stahlbeton: DAfStb-Richtlinie Selbstverdichtender Beton, Ergänzungen und Änderungen zu DIN 1045-2: 2001 und DIN EN 206-1 sowie DIN 1045-3 : 2001, 9. Entwurf März 2003, Berlin.

[2] DIN 1045, Beton und Stahlbeton, Berlin, Juli, 1988.

[3] DIN 933-4, Prüfverfahren für geometrische Eigenschaften von Gesteinskörnungen – Teil 4: Bestimmung der Kornform Kornformkennzahl, Berlin, Dezember, 1999.

[4] DIN 4226-1, Gesteinskörnungen für Beton und Mörtel, Teil 1: Normale und schwere Gesteinskörnungen, Berlin, Juli, 2001.

[5] de Larrard, F.: Concrete Mixture-Proportion – Scientific Approach, E & FN SPON, London, 1999.

[6] DIN 1045-1, Tragwerke aus Beton, Stahlbeton und Spannbeton, Teil 1: Bemessung und Konstruktion, Juli, 2001.

[7] HAVER-CPA-4 Photooptisches ONLINE – Partikelanalysegerät Firmenprospekt Haver + Boecker Drahtweberei und Maschinenfabrik Oelde.

[8] Stark. U., Reinhold, A., Müller, A.: Neue Methoden zur Messung der Korngröße und Kornform von Mikro bis Makro, In: Tagungsbericht, 15. Internationale Baustofftagung, Band 1, Weimar, 2003..

Contact to the authors:

[email protected]

[email protected]

Homepage of Darmstadt Concrete:

http://www.darmstadt-concrete.de

mailto:[email protected]

mailto:[email protected]

http://www.darmstadt-concrete.de/

- SELF-COMPACTING CONCRETE - INFLUENCE OF THE COARSE ... · discrepancy between the clearances in the slump flow test and the minimal acceptable clearance values according to DIN

Documents