Impact of Crushed Aggregate in Concrete

7/29/2019 Impact of Crushed Aggregate in Concrete

1/58

TRITA-LWR Degree Project 11:14

ISSN 1651-064X

IMPACTS OF USING CRUSHED ROCKS INCONCRETE

Andr Horta

June 2011


2/58

Andr Horta TRITA LWR Degree Project 11:14

ii

Andr Horta 2011

Degree Project for the Masters program in Environmental Engineering andSustainable Infrastructure

Department of Land and Water Resources Engineering

Royal Institute of Technology (KTH)

SE-100 44 STOCKHOLM, Sweden

Reference should be written as: Horta, A (2011) Impacts of Crushed Rocks inConcrete TRITA LWR Degree Project 11:14.


3/58

Impacts of using crushed rocks in concrete

iii

DEDICATION

This Degree Project is dedicated to Nelson Amaral, my maternalgrandfather, owner of a brilliant and creative mind that has alwaysinspired me in the search of knowledge.

Nelson Pereira Amaral


4/58


iv


5/58


v

FOREWORD

Aggregates from crushed rocks are in general flakier and have a roughersurface than natural aggregates. This leads to an increased cementdemand. To avoid this aggregate producers have to learn how to identify

rocks most suitable for concrete aggregate production and how to treatthe aggregate to improve the properties. The concrete producers have tolearn how to use the crushed rocks as aggregate in concrete.

This work is part of a larger project. The aim of this is to learn how touse crushed rocks, especially granitic, rocks as aggregate in concrete.Granitic rocks as coarse aggregate is common in Sweden but theknowledge of how to use fine aggregate from crushed granites is limited.Thus the research concentrated on fine aggregate in size less than 2 mm.

The work given to Andr Horta was to make rheological test with finematerials from different crushed granitic rocks. The task was to find outthe variation to identify which granite products that was good and bad.

The project then linked this to material properties and the effect ofdifferent types of granites and the effect of different treatments.

In the later part of Andrs work he have appointed to look on the effectof resorting of aggregates and the effect of this on cement consumption.

Bjrn Lagerblad


6/58


vi


7/58


vii

SUMMARY

Concrete is a mix of cement (binder), aggregate and water, and eventuallychemical and/or mineral admixtures. In Sweden concrete hastraditionally been manufactured using natural aggregate fromglaciofluvial eskers in the sand fraction (0-8mm). The remaining eskers

need to be preserved; therefore, an alternative aggregate should be used.The only regionally and economically alternative to natural aggregate iscrushed rocks.

The replacement of natural sand by crushed rock will, however, have animpact on both the industry and the environment. Crushed rocks in thefine fraction have a different shape and the particle size distribution isdifferent. Especially more flaky and rough particles will lead to a highercontent of cement to give a certain workability and strength. This in turnwill have a negative impact on the environment. The increase of cementconsumption will lead to more CO2 emissions, since the cementproduction releases large amounts of carbon dioxide both as a result ofthe burning process and decarbonation of limestone. Cement

production is responsible for around 5% of the global release of carbondioxide and thus the consumption must be limited. This tendency ofhigher cement consumption is because crushed rocks have less favorableproperties than natural aggregate for the workability of fresh concrete.Crushed rock grains are usually flakier, elongated and rough. In additionto that, crushed rocks normally have more fine particles. With crushedgranite rocks large amounts of flaky free mica in the fine fractions iscommon and leads to problems with workability. The main problemswith replacing natural aggregate with crushed rocks lies in the finefraction, with the particles less than 2 mm. With crushed granites largeamount of free micas is commonly found in this size fraction and leadsto large problems with workability. This work is part of a larger research

program with the aim to find means introduce crushed rocks to theconcrete production without having to increase the cementconsumption. This means to find ways to characterize materialproperties and link this to workability and rheological properties. Thiswill include choosing rocks that gives aggregate with good properties andto find ways to improve the material properties of the aggregate. Thework also includes proper recommendations of how to proportionateconcrete with crushed rocks as aggregate.

Rheology tests of mortar with several crushed rocks from all overSweden were performed; viscosity and yield stress were significantlyhigher for crushed rocks than for the natural aggregate. With gradingoptimization and use of superplasticizer, some crushed rocks hadrheological performance close to or even better than natural sand.

To check if the knowledge acquired from mortar rheology could beapplied in the concrete used by the building industry, the maximumaggregate size was gradually increased. The tests confirmed that the 0-2 mm fraction is the controlling factor regarding the aggregate.

The optimum grading in concrete was the one with the minimumamount of fines required for a concrete with good cohesion. The use ofsuperplasticizer was limited by separation.

Filler addition and the replacement of cement by filler were studied. Forfiller replacing cement, two situations were assessed: constant watercontent and constant w/c. For constant water content, the slump did not

have significance variations, and, up to a replacement of 20 kg/m3, thecompressive strength was resembling. For w/c constant, the slump


8/58


viii

decreased but there was a relevant improvement in the compressivestrength.

The conclusion was that crushed aggregate can be used to replace naturalaggregate in concrete without increasing, and even decreasing, thecement consumption. This, however, demands rocks that give propercubic particles. Grading optimization and the use of superplasticizer arethe most important tools to reach this goal. However, grading andsuperplasticizer are limited by cohesion and separation, essentialproperties for a good quality of fresh concrete. Filler of good crushedrocks should not be treated as waste or byproduct; it has valuable uses inconcrete and also in other products.


9/58


ix

SUMMARY IN SWEDISH

Betong r en blandning av cement (bindemedel), ballast och vatten, samteventuellt kemiska och/eller minerala tillsatser. I Sverige har man vidbetongtillverkningen traditionellt anvnt naturlig ballast frnfluvioglaciala rullstenssar i sandfraktionen (0-8 mm). Men eftersom de

rullstenssar som fortfarande finns kvar ska bevaras, r man tvungen atthitta ett alternativ till detta material. Det enda ekonomiskt hllbaraalternativet i denna region r krossade stenblock

Att erstta den naturliga sanden med krossad sten pverkar dock bdeindustrin och miljn. Krossad sten i finfraktionen har en annan form ochpartikelstorleken r annorlunda frdelad. I synnerhet flagiga, grovapartiklar innebr att cementinnehllet mste vara hgre fr att man skakunna uppn en viss grad av brukbarhet och styrka. Detta i sin tur har ennegativ inverkan p miljn. En hgre cementfrbrukning leder till kadekoldioxidutslpp, eftersom cementtillverkningen frigr stora mngderkoldioxid som ett resultat bde av frbrnningsprocessen och avdekarboneringen av kalksten. Cementtillverkningen orsakar ca 5% av den

globala frigringen av koldioxid, vilket begrnsar anvndningen.Tendensen till en hgre cementfrbrukning beror p att krossad steninte har lika gynnsamma egenskaper som naturlig ballast vad gllerbrukbarheten hos frsk cement. Krossade stenkorn r normalt settflagigare, grvre och mer avlnga. Dessutom r partiklarna i krossad stenoftast mindre. Finfraktion gjord av krossad granit innehller vanligtvisflagiga, fria glimmerpartiklar som frsvrar bearbetningen. Det frmstaproblemet med att erstta naturlig ballast med krossad sten ligger ifinfraktionen, dvs. partiklar mindre n 2 mm. Krossad granit innehllerstora mngder fritt glimmer, vilket leder till stora brukbarhetsproblemhos finfraktion gjord av detta mineral.

Denna uppsats r en del av ett strre forskningsprogram med syfte atthitta stt att anvnda krossad sten i betongproduktionen utan att behvaka cementfrbrukningen. Arbetet bestr i att hitta metoder fr attkategorisera materialegenskaper och koppla dessa till stensorternasbrukbarhet och reologiska egenskaper. Detta innebr att man vljerstentyper som ger ballasten frdelaktiga egenskaper och att man hittarstt att frbttra materialegenskaperna hos ballasten. Uppsatsen syftarven till att ge lmpliga rekommendationer kring sammansttningen hosbetong gjord med krossad sten.

Reologiska prover utfrdes p olika sorters murbruk gjorda p stensorterfrn hela Sverige. Viskositeten och strckgrnsen fanns vara betydligthgre hos krossad sten n hos naturlig ballast. Med hjlp av gradering

och anvndandet av superplasticering var det mjligt att ta fram ngrastensorter som hade lika bra eller t.o.m. bttre reologiska egenskaper nnaturlig sand.

Fr att underska om kunskaperna hmtade frn murbruksreologinkunde tillmpas ven p den betong som anvnds inom byggindustrin,kades storleken hos sandfraktionen gradvis. Testerna bekrftade attfinfraktionen (0-2 mm) r den styrande faktorn vad gller ballasten.

Den fr betongtillverkning optimala stentypen var den som hade lgstkrav p finhet fr att nd ge en betong med god sammanhllning.Anvndandet av superplasticering begrnsades av separation.

Effekterna av att tillstta fyllmedel samt att erstta cement med fyllmedelundersktes ocks. Fr fyllmedel som ersttning fr cement studeradestv situationer: konstant vatteninnehll och konstant v/c. Fr konstantvatteninnehll frndrades inte graden av brukbarhet nmnvrt, och upp


10/58


x

till en ersttning p 20 kg/m3 frblev tryckhllfastheten praktiskt tagetdensamma. Fr konstant v/c minskade brukbarheten mentryckhllfastheten kade betydande.

Slutsatsen r att man kan erstta naturlig ballast i betong med krossadballast utan att behva anvnda mer cement, tvrtom kan man t.o.m.minska cementfrbrukningen. Detta krver emellertid sten som gerlmpliga kubiska partiklar. Optimal partikelfrdelning och anvndandetav superplasticering r de viktigaste verktygen fr att uppn detta ml.Gradering och superplasticering begrnsas dock av sammanhllning ochseparation, vilka r vsentliga parametrar fr att uppn en god kvalitethos frsk betong. Fyllmedel gjort p rtt krossad sten br inte betraktassom avfall eller biprodukt, det r vrdefullt i betong och andraprodukter.


11/58


xi

SUMMARY IN PORTUGUESE

Concreto uma mistura de cimento (aglomerante), agregado e gua, eeventualmente aditivos qumicos e/ou mineirais. Na Sucia, o concretotem sido tradicionalmente manufaturado usando agregados oriundos deeskersglaciofluviais. Os eskersremanescentes necessitam ser preservados,

logo, um agregado alternativo deve ser usado. A escolha mais plausvelpara substituir os agregados glaciofluviais so as rochas britadas.

Esta substituio de areia natural por rocha britada no deve terimpactos negativos no meio ambiente e nem na industria da construocivil. Estes impactos negativos esto aptos a acontecer especialmentedevido ao fato de que o consumo de cimento tende a crescer, pelo usode rochas britadas.

O aumento do consumo de cimento aumentaria as emisses de CO2,uma vez que a descarbonataao da rocha calcrea para produo decimento responsvel por grande parte das emisses globais. Almdisso, aumentar o consumo de cimento resulta em maiores custos para aproduo do concreto, o que afetaria a industria da construo civil.

Esta tendencia de um maior consumo de cimento porque as rochasbritadas tm propriedades menos favorveis que o agregado natural paraa trabalhabilidade do concreto fresco. Gros de rocha britada sousualmente mais laminares, alongados e rugosos. Ademais, rochasbritadas normalmente tm mais particulas finas e maior quantidade demicas. Estas caractersticas levariam a um concreto fresco menostrabalhvel.

Para se ter um concreto to trabalhvel quanto o produzido comagregado natural, mantendo a resistencia compresso sem aumentaro fator gua/cimento deve-se adicionar mais pasta mistura. Logo,para no aumentar o consumo de cimento, ou at mesmo dimimu-lo,

comparado com agregado natural, o uso de rocha britada no concretodeve ser estudado e otimizado.

Testes de reologia da argamassa com vrias rochas britadas de todaSucia foram feitos; a viscosidade e a tenso de escoamento foramsignificantemente mais altas para rochas britadas que para o agregadonatural. Com a otimizao da granulometria e uso de superplastificantes,algumas rochas britadas tiveram o desempenho reolgico perto ou atmelhor que a areia natural.

Para checar se o conhecimento adquirido da reologia da argamassapoderia ser aplicado no concreto usado pela industria da construo civil,o tamanho mximo do agregado foi aumentado gradualmente. Os testesconfirmaram que a frao de 0- 2 mm o fator controlador no que dizrespeito ao agregado.

A granulometria tima no concreto foi aquela com a mnima quantidadede finos requeridos para um concreto com boa coeso. O uso desuperplastificante foi limitado pela separao.

Adio defillere substituio de cimento por fillerforam estudados. Parao filler substituindo cimento, duas situaes foram avaliadas: quantidadede gua constante e a/c (relao gua/cimento) constante. Para aquantidade de gua constante, o slump no teve variaes significativas, e,at uma substituio de 20 kg/m3, a resistencia compresso foisemelhante. Para a/c constante, o slump diminuiu mas houve umrelevante aumento na resistencia compresso.

A concluso foi que rocha britada pode ser usada para substituir oagregado natural no concreto sem aumentar, e at mesmo diminuindo, o


12/58


xii

consumo de cimento. Otimizao da granulometria e uso desuperplastificantes so as ferramentas mais importantes para alcanareste objetivo. De qualquer forma, granulometria e superplastificantes solimitados pela coeso e separao, propriedades essenciais de umconcreto fresco de boa qualidade. O filler no deve ser tratado comoresduo ou subproduto, ele tem valiosos usos no concreto e em outros

produtos.


13/58


xiii

ACKNOWLEDGEMENTS

I wish to thank specially my supervisor Bjrn Lagerblad, SeniorResearcher at CBI and adjunct Professor at Concrete Construction atDepartment of Architectural and Civil Engineering KTH, a personwho I greatly admire and who received me with arms wide open in

Sweden; the main incentivator of this work.I would like to thank my examiner Professor Joanne Fernlund, from theDepartment of Land and Water Resources Engineering of the RoyalInstitute of Technology (KTH), for accepting the challenge and for thecomprehension.

I wish to thank KTH, my Professors and colleagues for all the benefitsproportioned due to the Master Programme of EnvironmentalEngineering and Sustainable Infrastructure.

I am grateful to the institution and to all the workmates of CBI, for thehelp and for the great work environment that was proportioned.

I would like to thank Annika, Mike, Leif and Ulrika, people that worked

with me in this project.I wish to thank Hans-Erik from Cementa AB for the support.

I wish to thank my family, example of honesty, dedication and work.Everything I am, I owe to them.

I am grateful to Marcela, for the love and fellowship.

I would like to thank Professor Estvo Bicalho, for the teaching and forthe incentive.

I would like to thank Colgio Santo Antnio, responsible for myscholastic formation and for the great friendships that were born there.

I wish to thank my friends from UFMG Federal University of MinasGerais for the companionship and for the sharing of the knowledge,includingFazendeiros.

I would like to thank Siduca, more than a group of friends.


14/58


xiv


15/58


xv

TABLE OF CONTENT

Dedication iiiForeword vSummary viiSummary in Swedish ixSummary in Portuguese xi

Acknowledgements xiiiTable of content xvList of Figures xviiList of Tables xviii

Abstract 11. Introduction 1

1.1. Impacts on the environment 21.2. Concrete: Nomenclature and descriptions 21.3. Aggregate characteristics 3

2.

Aims4

3. Methods 43.1. Mortar Rheology 5

3.1.1. Grading optimization 73.1.2. Cement paste content variations 83.1.3. Superplasticizer 93.1.4. Wind sieving and cubical-shape crushing 10

3.2. Concrete tests 103.2.1. Tests performed on concrete 113.2.2. Concrete mixes 123.2.3. Concrete with 0-4 mm aggregate 143.2.4. Concrete with 0-8 mm aggregate 153.2.5. Concrete with 0-16 mm aggregate 154. Results 17

4.1. Mortar rheology 174.1.1. Crushed rocks compared to natural aggregates 174.1.2. Grading 184.1.3. Reference and original grading 194.1.1. Cement reduction 194.1.2. Superplasticizer 204.1.3. Wind Sieving and cubical shape crushing 214.1.4. Cement savings 22

4.2. Fresh concrete tests 234.2.1. Concrete with 0-4 mm aggregate 234.2.2. Concrete with 0-8 mm aggregate 244.2.3. Concrete with 0-16 mm aggregate 25

4.3. Compressive strength 285. Discussion 29

5.1. Mortar rheology 295.2. Fresh concrete tests 325.3. Compressive strength 34

6. Conclusions 357. Recommendations and Future Research 36

7.1. Filler 367.2. Limestone 36

8. References 37


16/58


xvi

8.1. Other references 379. Appendix IGrading of natural aggregate and crushed rocks I10. Appendix IIConcrete tests to determine the most suitable 0-2 mm

grading II


17/58


xvii

LIST OF FIGURES

Fig. 1 - Bingham model. ............................. ................................. ................................ ......................... 5Fig. 2 - Viscometer. ............................................................................................................................... 6Fig. 3Mixer. ........................................................................................................................................ 7Fig. 4Gradings 1-4 for aggregate 0-2 mm. ................................ ................................ .................... 8Fig. 5 - Mortar mixes; reduction and addition of cement paste compared to the Referencemix design. .............................................................................................................................................. 9Fig. 6 - Reference grading for each maximum aggregate size. ............................................... ...... 10Fig. 7 - Slump test. ............................. ................................ ................................ ................................. . 11Fig. 8 - Strength cubes. ............................... ................................. ................................ ....................... 12Fig. 9 - Compressive strength test. ...................................... ................................. ............................ 12Fig. 10 - Concrete mixer. ................................. ................................. ................................. ................. 13Fig. 11 - Reference mix design for each maximum aggregate size. ............................................. 13Fig. 12 - Reference mix design for concrete with 0-16 mm aggregate aimed for housebuilding. ............................... ............................... ................................. ................................. ................. 14Fig. 13 - Yousiff mix design for concrete with 0-16 mm aggregate. ............................... ............ 14Fig. 14 - Mixes design for filler addition. ............................ ................................. ............................ 16Fig. 15 - Mixes design for cement replacement by filler. ....................... ................................. ...... 17Fig. 16Rheological performance of crushed rocks compared to natural aggregate. ............ 17Fig. 17 - Comparisson between Reference and original grading, and natural aggregate. ......... 18Fig. 18 - Rheology results for different gradings of K50, K49 and K43. ............................. ...... 18Fig. 19 - Effect of cement reduction for Gradings 1-4 with K50. ............................................ .. 19Fig. 20 - Effect of cement reduction for Gradings 1-4 with K43. ............................................ .. 19Fig. 21 - Effect of 2 ml of superplasticizer in mixes with aggregate according to their originalgrading. ................................ ............................... ................................. ................................ .................. 20Fig. 22 - Effect of 4 ml of superplasticizer in mixes with aggregate according to their originalgrading. ................................ ............................... ................................. ................................ .................. 20Fig. 23 - Effect of 2 ml of superplasticizer in mixes with aggregate according to theReference grading. ............................. ................................ ................................ ................................. . 21Fig. 24 - Effect of cube shaping (through VSI crusher) and wind sieving. ............................... . 21Fig. 25 - Comparisson of how much cement can be saved by using superplasticizer andgrading optimization for K50. .............................. ................................. ................................ ............ 22Fig. 26 - Comparison of how much cement can be saved by the use of superplasticizer andgrading optimization for K57. .............................. ................................. ................................ ............ 22Fig. 27 - Plastic viscosity for different proportions and aggregates of 0-2 and 2-4 mm .......... 24Fig. 28 - Yield stress for different proportions and aggregates of 0-2 and 2-4 mm. ................ 24Fig. 29 - Rheology for concrete with aggregate 0-8 mm. ............................ ................................. . 25Fig. 30 - Slump for concrete with 0-8 mm aggregate. ............................................................. ...... 25Fig. 31 - Slump and the effect of superplasticizer on fresh concrete workability. .................... 26Fig. 32 - Slump for concrete with filler addition. ............................... ................................ ............ 27Fig. 33 - Superplasticizer needed to reach a slump of approximately 200 mm for filler

addition. ............................... ............................... ................................. ................................. ................. 27Fig. 34 - Slump for cement replaced by filler. ............................... ................................. ................. 28Fig. 35 - Superplasticizer needed to reach a slump of approximately 200 mm for fillerreplacing cement. ............................... ................................ ................................ ................................. . 28Fig. 36 - Compressive strength for concrete with maximum aggregate size equal to 4 mm. .. 28Fig. 37 - Compressive strength for concrete with maximum aggregate size equal to 16 mmand with filler........................................................................................................................................ 29Fig. 38 - Rheology for N3, K50 and K57. ................................ ................................ ....................... 30Fig. 39 - Rotational volume of a particle. ................................ ................................. ....................... 31Fig. 40 - Rheological behavior of limestone compared to granitic rocks and natural aggregate................................................................................................................................................................. 35
http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615830http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615830http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615831http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615831http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615832http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615832http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615832http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615832http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615833http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615833http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615833http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615833http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615834http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615834http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615834http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615835http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615835http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615836http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615836http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615837http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615837http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615838http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615838http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615839http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615839http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615840http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615840http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615842http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615842http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615843http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615843http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615844http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615844http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615845http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615845http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615845http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615845http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615846http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615846http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615847http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615847http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615848http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615848http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615849http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615849http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615850http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615850http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615850http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615851http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615851http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615851http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615852http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615852http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615852http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615853http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615853http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615854http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615854http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615854http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615855http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615855http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615855http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615856http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615856http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615857http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615857http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615858http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615858http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615859http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615859http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615860http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615860http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615861http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615861http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615862http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615862http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615862http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615863http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615863http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615864http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615864http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615864http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615865http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615865http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615866http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615866http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615866http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615867http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615867http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615868http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615868http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615869http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615869http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615869http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615869http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615869http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615868http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615867http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615866http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615866http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615865http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615864http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615864http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615863http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615862http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615862http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615861http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615860http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615859http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615858http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615857http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615856http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615855http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615855http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615854http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615854http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615853http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615852http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615852http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615851http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615851http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615850http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615850http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615849http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615848http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615847http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615846http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615845http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615844http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615843http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615842http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615841http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615840http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615839http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615838http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615837http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615836http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615835http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615834http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615834http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615833http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615832http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615831http://c/Users/Andre/Documents/Thesis/Andre_Horta_EESI_Master_Thesis.docx%23_Toc300615830


18/58


xviii

LIST OF TABLES

Table 1 - Combinations for the 0-4 mm concrete tests. ............................. ................................. . 15Table 2Gradings for concrete up to 16 mm. ............................ ................................. ................. 26
http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190575http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190575http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190576http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190576http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190576http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190576http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190576http://c/Users/Andre/Documents/Thesis/Andre_Horta_Master_Thesis_Version4_%5bwith%20figs%5d%20(Repaired)2.docx%23_Toc292190575


19/58


ABSTRACT

Concrete in Sweden has traditionally been manufactured with natural aggregate fromglaciofluvial eskers. There is a need to preserve the remaining eskers because of their culturalvalue and importance for water filtration, thus natural aggregate has to be replaced. The mostrealistic alternative is to use crushed rocks. . The major problem with crushed rocks in

concrete production is the workability. This is because crushed rocks have less favorableproperties. The fragments are flakier and have a rougher surface than natural aggregates thathave been rounded in water. Without any amelioration of the crushed rock, to reach a certainworkability and strength, the amount of cement in the mix has to be increased.

Cement production requires large amounts of energy and the decarbonation of limestonereleases large amounts of CO2. Combined, the release of CO2, due to burning anddecarbonation of limestone, accounts for about 5% of the global CO2 emissions. An increasein cement consumptions is less desirable. Thus to replace natural aggregates, the use ofcrushed rocks has to be optimized as regard cement consumption.

Several crushed aggregates, most from granitic rocks, from all over Sweden were analysed inthis study. These crushed rocks were characterized according to their grading, specificsurface, shape and petrography and compared to natural sand.

Rheological tests that reveals the workability in detail was performed on mortars. The testsshowed that as regard workability the 0-2 mm fraction is the most important factor. Further,the maximum aggregate size was gradually increased up to 16 mm, to have a more realisticapproach to the concrete produced by the building industry.

The results showed that with grading optimization and superplasticizer, some crushed rockscan be used for concrete production without increasing, and even decreasing, the cementconsumption. This research also contemplated the use of filler. As a mineral admixture it canimprove the compressive strength. It can also be used to replace cement; a replacement up to20 kg/m3 of cement by filler can be done without significant effect on compressive strength.

Key words: Crushed rocks; Mortar rheology; Fresh concrete workability; Cement

content; Superplasticizer; Filler.

1. INTRODUCTION

Reinforced concrete is the most used construction material. Concreteconsists of a mix of cement, aggregate and water and can have chemicaland/or mineral admixtures to improve some properties. Water accountsfor 7-9% in weight in concrete, while cement weighs 10-20% andaggregate has the biggest contribution 70-80 %; admixtures are usuallynot more than 1 % (Fagerlund, 1999).

Aggregates can be natural or obtained from rocks crushed. In Swedenconcrete is usually manufactured with fine aggregate (0-8 mm) fromglaciofluvial deposits. There is a need to preserve remaining sand. Thesand that in most cases comes from glaciofluvial eskers are important towater filtration. Moreover, they have natural and cultural values. A goalto reduce its production was set by the Swedish government(Westerholm, 2006).

Crushed rock is the only type of aggregate that is regionally available forconcrete production. Therefore, it should be used to replace natural sandas fine aggregate in concrete. Due to the different characteristics of theaggregates, investigations have to be done in order to evaluate the impactof crushed rocks on the environment and on concrete.

To be able to optimally replace natural aggregate by a crushed rock

equivalent, one has to learn to characterize especially the fine fractions in


20/58


2

a correct way and learn how to use the data in proportioning concrete(Lagerblad, B., 2007, pers. comm., December 18th).

1.1. Impacts on the environmentThe climate is also affected by crushed aggregates. One of the mostimportant factors contributing to the global warming phenomena is the

carbon dioxide (CO2) emissions. Cement production, apart from energyconsuming, releases large amount of CO2 by decarbonating limestone.Due to less favourable properties for fresh concrete workability, crushedaggregate is likely to increase the cement consumption; this will bediscussed on the Impacts on concrete section. Thus from anenvironmental point of view it is advantageous to decrease theproduction of cement or at least ensure that production does notincrease. To make it without negative effect in the building industry, theuse of crushed aggregates in concrete has to be studied.

1.2. Concrete: Nomenclature and descriptionsThe properties of the fresh concrete are negatively affected by using

crushed rock aggregates compared to natural aggregate. To understandits effect, it is important to know the concepts about some parametersthat are going to be used to evaluate the performance of the freshconcrete:

Water-cement ratio (w/c)

It is the proportion between the content of water and cement. The w/cis dimensionless and it is the weight of cement divided by the weight ofwater.

Workability

A workable freshly mixed concrete or mortar can be easily mixed,placed, consolidated, and finished to a homogenous condition (ACI

Committee 116, 2000). Rheology

Rheology is the science of deformation and flow of matter, it relatesstress, strain, rate of strain and time (Barnes et al, 1989).

Yield Stress

It is the shear stress in which the fresh concrete begins to deformplastically. Its unity is Pascal (Pa). It gives information comparable toslump.

Plastic Viscosity

It measures the amount of energy needed to keep concrete in motion,

i.e., how fast the concrete flows. Its unity is Pa.s. Homogeneity

Concrete should be homogenous, i.e., its constituents need to beuniformly distributed. The absence of aggregate on top of mortar andfresh concrete mix and/or stiffer bottom characterizes separation.

Cohesion

It is the ability of fresh concrete to resist segregations of its constituentsand there is no method to measure it (Newman & Choo, 2003).Nonetheless, poor cohesion can be detected by the aspect of freshconcrete.

Strength of concrete

If the aggregates used are stronger than the cement, which is usually thecase of granitic rocks, then they do not affect compressive strength


21/58


3

directly. In this case the strength of hardened concrete depends mainlyon the strength of the cement paste (water and cement together), theweakest part, which in turn depends on the water/cement ratio (w/c).

Fresh concrete

Fresh concrete is particle slurry that eventually stiffens and hardens to

the desired product. The most significant impact of using crushedaggregate instead of natural sand is the fresh concrete workability. To beable to flow and fill a form the fresh concrete has to have certainrheological properties. The rheology in turn depends on the propertiesof the particle slurry, how the particles interact with each other. Aconcrete with good workability can be compacted without excessiveeffort and lack of homogeneity (Neville, 1973).

Desirable concrete

A concrete shall have as low yield stress and viscosity as possible. A toolow yield stress can, however, result in separation and bad cohesion.Cohesion and separation are phenomena which cannot be measured orquantified; they are usually detected by the visual and tactile aspect orconsistency of the fresh concrete.

To keep certain strength the w/c must be kept constant; thus a reducedamount of cement gives a reduced amount of water, i.e., lower pastecontent; which lowers the workability of the fresh concrete.

For crushed aggregates, the general rules for concrete proportioningbased on experience on natural aggregates will not apply. To retain theworkability and compressive strength without increasing (evendecreasing) the cement paste, the use of crushed rocks has to beoptimized.

Superplasticizer and grading optimization are important aids to achieve

this goal; nevertheless, they have to be carefully used, otherwise thequality of the fresh concrete will be compromised by separation andpoor cohesion.

1.3. Aggregate characteristicsThe water content, mix proportions, time, temperature, cement type,admixtures and aggregate properties are factors that influence theworkability (Mindess et al, 2003); as the aggregate is the largestcomponent in volume, the effect of its properties is very significant.

The properties that most influence the fresh concrete workability aregrading, shape and surface texture. These characteristics are differentwhen comparing crushed rocks with natural aggregate, and they also

differs among the different crushed rocks.The crushing process produces large amount of fines (Appendix I). Thefragments of crushed rocks are more rugged and asymmetrical. This isespecially the case with crushed granitic rocks that contains flaky micaparticles. (Westerholm et al, 2008).

Under the same conditions, fresh concrete with crushed rocks compared to natural aggregates will probably have less workableparticle slurry. Irregular particles, like rock powder, needs more space tomove than rounded particles. Moreover, crushed rocks are likely to havebigger specific surface due to higher amount of fine particles, roughertexture and irregular shape, which would demand more water to wet theaggregate surface.


22/58


4

The characteristics can also have a potential effect on the compressivestrength. Especially the texture and shape, they can affect the interactionbetween the aggregate surface and the cement paste.

2. AIMS

The aim of this project is to develop methods that can be implemented

that will result in keeping the amount of cement used for concreteproduction at the same level or even lower for the concretemanufactured with crushed rocks instead of natural aggregate.

The use of crushed rocks will be optimized in order to get a freshconcrete with good workability and the cement content as low aspossible. It is important to highlight that other quality parameters offresh concrete, such as cohesion and separation, must also yieldacceptable results.

To achieve this goal, it is needed to:

Compare rheology of crushed rocks with natural aggregate.

Evaluate the possibility of reducing cement paste.

Find out the most suitable aggregate grading for crushed aggregate.

Assess how grading improves workability by comparing the rheologyof mixes with aggregate without grading modification and withoptimized grading.

Study the use superplasticizer to improve workability and check if itcan let rheology of mixes with crushed rocks comparable to naturalaggregate.

Evaluate the effect of other processes, e.g., wind sieving and cubeshaping.

Estimate how much cement can be saved by grading optimization

and superplasticizer comparing to a situation which nothing is doneto improve rheology.

Increase gradually the maximum aggregate size up to 16 mm to havea more realistic approach about how the outcomes of this researchcan be applied building industry.

Appraise the influence of the aggregate type in each fraction.

Study the use of filler as a mineral admixture to improve compressivestrength and to replace cement.

3. METHODS

In the broader project, more than 60 crushed rock aggregates from

different quarries from all over Sweden are evaluated. In this study about25 crushed rocks were assessed. Each aggregate is represented by a code,in order to preserve the manufacturer. The code is a letter followed by anumber, where N represents the Natural aggregates and K standsfor crushed rocksKross-ballastin Swedish.

The natural aggregate used for concrete in Sweden usually has the samegeological formation, the glaciofluvial eskers. Even if they come fromdifferent parts, they have similar properties. So, when comparing naturalaggregates and crushed rocks, it was assumed that only one natural gravelsample was necessary, since the properties of natural aggregatesindependent of origin are very similar. The natural aggregate used formortar rheology tests was N3.

Mortar-rheology tests were done to evaluate the performance of eachcrushed rock on fresh concrete, always comparing with natural aggregate.


23/58


5

Having this comparison, some parameters were optimized to enhancethe workability. Different gradings were tested to find out the mostsuitable particle-size distribution. Variations in the cement content werealso assessed.

With the appropriate grading and paste content, the effect ofsuperplasticizer, wind sieving and cube shaping were evaluated. Anestimation of how much cement can be saved by grading optimizationand superplasticizer was done.

In a further instance, concrete tests were performed to check if theresults obtained from mortar can be applied to the concrete used by thebuilding industry. Tests on concrete contemplated not only fresh, butalso hardened concrete, through compressive strength tests.

3.1. Mortar RheologyTo evaluate the effect of crushed aggregate on the workability of freshconcrete, compared to natural aggregate, rheology tests on mortar wereperformed.

Mortar can be defined as fresh concrete without coarse aggregate (Banfil,2004). In this project the mortar was considered the constituents ofconcrete with maximum particle size bellow 2 mm. The work was doneon 0-2 mm mortar as this fraction is the controlling factor regardingaggregate from crushed rocks (Westerholm et al, 2008).

Fresh concrete and mortar behave as Bingham plastic fluid; which meansthat shear stress and shear rate have a linear relationship.

To start the flow, the shear stress needs to reach a certain value, 0;which is called yield stress. Plastic viscosity is obtained through the linearregression between shear stress and shear rate (Fig. 1).

The yield stress gives information similar to the ones obtained with theslump test. Plastic viscosity is a measurement of flow time, it is theamount of energy needed to keep fresh concrete (or mortar) in motion,i.e., how fast the concrete flows. The concrete shall not separate; thus tolow rheological values often lead to separation.

Rheology tests

The equipment used for the rheology test was a rotational viscometer(Contec 4SCC, ConTec Viscometers) (Fig. 2).

To run the test, the viscometers recipient is filled with mortar andplaced over a rotational plate. A stationary palette immerses into therecipient and the plate starts to rotate in different speeds. During therotation the palette measures the torque for the different speeds.

Fig. 1 - Bingham model.


24/58


6

With the data from the viscometer a linear regression of torque androtational velocity is done. The plastic viscosity and yield stress are

obtained according to the Bingham model. These calculations areperformed using software Freshwin 4.0. To minimize the errors twomeasurements are done for each mix, the considered value is the average.

Mortar mixes design

The test program was mainly performed on the concrete aimed forhouse building. The mix design used for house buildings is referred to asReference-mixdesign. It is calculated based on a concrete with 350 kg ofcement per cubic meter and a w/c ratio equal to 0.57. For mortar mixesthe weight is 30% cement, 17% water and 53% aggregate. It wascalculated using the density of 3.08 g/cm3 for cement, 2.65 g/cm3 forgranitic rocks and 1.00 g/cm3 for water.

For rheology tests of mortar the mix volume was usually four liters, i.e.,to test mortar according to the Reference-mixdesign, 2.54 kg of cement,4.52 kg of crushed granitic aggregate and 1.54 liters of water wereneeded.

Remark that if using another aggregate but granitic rocks, the weightproportion should be corrected according to the density.

The influence of the cement type was not in the scope of this study.Thus the same cement type was used for all mixes; CEM II/A-L 42.5 R,Swedish limestone-Portland cement, Byggcement delivered by CementaAB, the most used cement in Sweden.

Mortar mixing procedure

Cement and aggregate were put into a bowl and mixed before wateraddition (Fig. 3). After one minute, water was added in the bowl and

Fig. 2 - Viscometer.


25/58


7

mixed at low speed for one more minute. To avoid that particles stick inthe walls of the bowl, it was scraped off manually with a shovel; than themortar was mixed for one more minute at high speed. After getting thehomogenous mix, it was put into a recipient to be tested in the rotationalviscometer.

In the cases that superplasticizer was added, the mortar of theviscometers recipient was returned to the bowl and mixed at high speedfor 30 seconds before the addition. The walls were manually scraped offagain and superplasticizer added. Then, the mortar was mixed for more

30 seconds and put into the recipient to be tested in the viscometer. Ifmore superplasticizer is added, the same procedure is repeated.

3.1.1. Grading optimization

In concrete different particles interact in such a way that larger particleswill flow or roll on smaller particles. Thus the particle-size distribution isimportant.

The most suitable particle-size distribution is the one that provides betterworkability without negative consequences on homogeneity and cohesiveproperties. To find out the most appropriate, four grading curves werepre-defined. Grading 1-4 has the highest to the lowest fines contentrespectively (Fig. 4).

Higher amount of fines results in higher water demand to wet theaggregate, due to the higher surface area; consequently, yield stress andplastic viscosity values increase. Remark that the grading that better fits

Fig. 3

Mixer.


26/58


8

rheology can vary for different aggregates since it depends on factors likeshape and texture (Quiroga, 2003).

For the evaluation of grading effect on rheology, the aggregate wassorted out to fit each one of these four gradings. The rheology wastested according to the Reference-mixdesign. Three crushed rocks weretested according to each grading; one with properties close to naturalaggregate (K50), one intermediate (K49) and the other with unfavorableproperties (K43).

Reference and original grading

Original grading refers to the particle-size distribution of the aggregatewithout any modification. Reference Grading, as the name says, is thegrading which will be used as reference for the rheology tests. TheReference Gradingwas chosen as the one that better fits the criteria of goodworkability (low plastic viscosity and yield stress) and still good cohesionand no separation.

To evaluate how plastic viscosity and yield stress can be improved usingthe Reference Grading, a comparison between the Referenceand the originalgrading was done. Several crushed rocks were tested according to theReferenceand the original grading. The tests were done according to theReference-mixdesign and rheology was also compared to natural aggregate.

3.1.2.

Cement paste content variationsVariations in the paste content were tested. The w/c ratio was keptconstant (0.57) not to influence other concrete properties, e.g.,compressive strength. The amount of cement was reduced and increasedup to 15% compared to the Reference-mixdesign (Fig. 5).

Fig. 4

Gradings 1-4 for aggregate 0-2 mm.


27/58


9

Cement paste reduction

To meet the main goal of the project use crushed aggregates withoutincreasing, and even decreasing, the cement content rheology testswere performed in order to assess the possibility of reducing the amountof cement paste in mixes with crushed rocks. The aggregates used inthese tests were K50 and K43. Each aggregate had its rheology testedaccording to Grading1-4.

Cement paste addition

Cement paste was added on mixes with crushed rocks according to itsoriginal grading; their rheology was compared to mixes with the sameaggregate, but with superplasticizer and optimized grading. The crushedrocks used for these rheological tests were K50 and K57.

This comparison is done in order to evaluate how much cement can besaved by grading optimization and using superplasticizers.

3.1.3. Superplasticizer

Superplasticizer is a chemical admixture that can disperse flocculatedcement particles, which lowers the plastic viscosity and mainly the yieldstress, improving the workability.

The superplasticizer used in the experiments was Glenium 51 fromDegusa Construction Chemicals AB. This superplasticizer is carboxylateesther based with approximately 35% solid content, the density isapproximately 1.08 g/cm3.

On mortar mixes, superplasticizer was added 2 by 2 ml with ananalogical dosing device. As the mixes had a volume of 4 liters, theamount of cement was 2.45 kg. Thus each 2 ml addition ofsuperplasticizer corresponded to 0.8 g/kg of cement, approximately0.1%, for the Reference-mixdesign. The adding procedure of this chemicaladmixture is explained in the Mixing procedure section.

The tests with superplasticizer were performed according to the Reference-mix design. Mixes with aggregate according to its original grading hadtwo additions of 2 ml of superplasticizer.

Fig. 5 - Mortar mixes; reduction and addition of cement paste

compared to the Reference mix design.


28/58


10

For the mixes with aggregate according to the Reference Gradingonly oneaddition of 2 ml was done. Due to the lower fines content, furtheradditions would lead to separation.

Rheology tests were performed for different crushed rocks, first withoutsuperplasticizer, and then after a 2 ml addition; for the aggregateaccording to its original grading, an addition of 4 ml of superplasticizerwas also tested.

3.1.4. Wind sieving and cubical-shape crushing

Wind sieving is a process where wind is used to remove fine particles ofthe crushed aggregate. This process also removes micas, a flaky mineral.The wind-sieved aggregates studied were K40 and K60.

For the aggregates studied, the cube-shaping process was the verticalshaft impact (VSI) crusher. The VSI crusher has a high speed spinningrotor which throws the particles against the chamber walls to crush therocks into smaller pieces. This process provides more cubic grains thanother crushing methods. The cube-shaped aggregates which were testedwere K40 and K41.

To evaluate the effect of these two processes, rheology tests were doneon mixes with crushed rocks according to the ordinary-crushing processand with the same aggregate, but wind sieved and/or cube shaped. Themixes were tested according to the Reference-mix design and ReferenceGrading.

3.2. Concrete testsIn the building industry it is important to have good quality not only inthe fresh concrete, but also in its hardened form. For this reason, in thispart of the project, there will be tests to measure the workability of freshconcrete rheology and slump and also tests to measure thecompressive strength of hardened concrete.

The concrete used for buildings usually has the maximum aggregate sizeof 16 mm. To avoid a gap in the data and to get a better adjustment forthe concrete tests, the transition from mortar to concrete was gradual.This gradual transition also allowed the evaluation of the influence ofeach fraction on the quality of the fresh and hardened concrete. Themaximum aggregate size was increased first to 4 mm and then to 8 mm,before the tests on concrete with 0-16 mm aggregate.

Fig. 6 - Reference grading for each maximum aggregate size.


29/58


11

The grading used for concrete mixes, also referred to as Reference Gradingwas based on the grading of a natural aggregate, which also correspondsto the maximum value of packing (Fig. 6). The Reference Grading wasoptimized, since crushed rocks do not behave as natural aggregate.

3.2.1. Tests performed on concrete

For mortar, only rheology tests on the viscometer were performed. Forfresh concrete, rheology tests were also done; however, the viscometerwas not suitable for rheology measurements with concrete withmaximum aggregate size above 8 mm, i.e., mixes with 0-16 mmaggregate.

Thus fresh concrete with 0-4 and 0-8 mm aggregate had their rheologytested in the viscometer. For mixes with 0-8 and 0-16 mm aggregate, forthe purpose of measuring workability, slump tests were performed.

To study the influence of the aggregate type on hardened concrete,compressive strength tests were done in concrete with 0-4 mm aggregate.To evaluate the effect on hardened concrete of using filler, compressivestrength tests were performed on concrete with 0-16 mm aggregate.

Rheology

The rheology tests followed the same procedure described in the Mortarrheology section. Although the procedure is the same as for mortar, abigger viscometers recipient and stationary immerged palette is used.

Slump

A truncated cone of 300 mm is filled with fresh concrete. The truncatedcone is filled in three steps; in each step, 15 strokes are applied with ametal stick, to compact the fresh concrete. The top is leveled with themetal stick, and the truncated cone is lifted slowly. The height betweenthe highest stone and the top of the truncated cone is the slump value

(Fig. 7) Compressive strength

The fresh concrete is put into a metallic cubical oil-spayed mould; afterbeing vibrated for 10 seconds, the top is leveled and smoothed with atrowel (Fig. 8).

Fig. 7 - Slump test.


30/58


12

The cubes are labeled and wait for 28 days in a chamber of 100% relativehumidity. After these 4 weeks, the cubes are removed from the mouldand taken to the compressive strength machine, where strength isapplied till it fails (Fig. 9).

For concrete with 0-4 mm aggregate the cube has 10 cm high, for themaximum aggregate size of 16 mm, the cube has 10 cm of height. Twocubes are tested and the average value is considered.

3.2.2. Concrete mixes

To prepare the concrete mix, the aggregate is mechanically sieved to fitthe desired grading curve. The same mixing procedure is done for allmixes, to ensure that it will not affect the results of the tests. For the

Fig. 9 - Compressive strength test.

Fig. 8 - Strength cubes.


31/58


13

same reason, only one cement type (CEM II/A-L 42.5 R, Swedishlimestone-Portland cement, Byggcement delivered by Cementa AB) isused in the mixes.

Concrete mixing procedure

For the concrete with 0-4 mm aggregate, the mixing procedure is similarto the one for mortar, described in the Mortar rheology section. For 0-8and 0-16 mm aggregate concrete, a bigger mixer is used, (Fig. 10).

The cement and aggregate are mixed for one minute before the wateraddition. When water is added, the concrete is mixed for 90 seconds,then the materials adhered in the corners are manually scraped with ashovel and the mixer is turned on again for 30 seconds.

Concrete mixes design

Fig. 11 - Reference mix design for each maximum aggregate size.

Fig. 10 - Concrete mixer.


32/58


14

For the rheology tests of the mortar and for the fresh concrete tests, the

Reference-mixdesign was calculated based on a house building concrete,with a w/c of 0.57 and 350 kg of cement per cubic meter of concrete(Fig. 11).

The Reference-mixdesign for concrete with crushed rocks up to 16 mmleads to a very stiff mix (Fig. 12). Thus to study the effect of filler as amineral admixture replacing aggregate on fresh concrete, the Reference-mixdesignwas not suitable. Thus to get a more flowable fresh concrete, thecement content and w/c were changed to 310 kg/m3 and 0.68,respectively (Yousiff, 2006); referred to as Yousiff-mixdesign (Fig. 13).

The proportions of the constituents in each mix design do not have alarge difference; however, the fresh concrete workability is sensitive evento small proportion changes, therefore, fresh concrete will have higher

flowability if prepared according to the Yousiff-mix design.

3.2.3. Concrete with 0-4 mm aggregate

For concrete with 0-4 mm aggregate, the influence of the aggregate typeand the effect of different proportions between 0-2 and 2-4 mmfractions were studied on fresh and hardened concrete.

To evaluate the influence of the aggregate type, two crushed rocksK50and K57 were used in each fraction, 0-2 and 2-4 mm, with differentcombinations. To find the most favourable aggregate grading, differentproportions of these two fractions were tested; the proportion of the 2-4 mm fraction was gradually increased from 0 to 70% (Table 1).

Fig. 12 - Reference mix design for concrete with 0-16 mmaggregateaimed for house building.

Fig. 13 - Yousiff mix design for concrete with 0-16 mm aggregate.


33/58


15

The particle-size distribution in the 0-2 mm fraction was kept constant,according to the Reference Gradingfor aggregate up to 2 mm.


For concrete made with 0-8 mm aggregate, only the rheological effect ofthe aggregate type in the 0-2 mm fraction was studied. Two mixes, onewith K50 and the other with K57 in the 0-2 mm fraction, were prepared.

The concrete was mixed according to the Reference-mixdesign and theaggregate followed the Reference Grading. There were no variations in the2-8 mm fraction. The aggregate 2-4 mm was K50 and for the fraction 4-8 mm the aggregate was K33.


The investigation of the concrete with aggregate up to 16 mm had, in afirst instance, the grading optimization. After the grading is optimizedthe use of filler is going to be evaluated. Filler was used to improvecompressive strength as a mineral admixture replacing aggregate, and toreplace cement.

Optimum grading

A minimum amount of 0-2 mm particles is necessary to keep concretecohesive and without separation. This minimum content of the 0-2 mmfraction should be the optimum grading composition for the crushedaggregate, as it is known that fine particles are not favourable for aworkable fresh concrete.

Cohesion and separation are difficult parameters to be quantified,therefore, it can be observed by the aspect of the fresh concrete.Sometimes separation and loss of cohesion are not evident in mixes withaggregate with maximum particle size equal to 4 or 8 mm, however, itcan be detected with good accuracy in the concrete mixes with aggregatesize up to 16 mm.

To find the minimum content of the 0-2 mm fraction needed to getconcrete with good cohesion and no separation, different proportionsbetween 0-2 and 2-4 mm particles were assessed.

For the Reference Gradingof concrete with particles up to 16 mm, the 0-4 mm fraction corresponds to 44% of the total aggregate. Withoutchanging the amount of 0-4 mm particles, different proportions of 0-2and 2-4 mm aggregate were evaluated.

These proportions varied from the Reference Grading(80% of 0-2 and 20%of 2-4 mm) to 40% of 0-2 and 60% of 2-4 mm particles. The grading ofthe fraction 4-16 mm was kept constant, according to the ReferenceGrading.

In the 0-2 and 2-4 mm fractions K50 and K57 were tested with differentcombinations, for the 4-8 mm fraction the aggregate was K33 and forthe 8-16 mm fraction it was the K58.

Table 1 - Combinations for the 0-4 mm concrete tests.

Aggregate Fraction Proportion of the 2-4mm fraction

K50 0-4mm 0%, 20%, 25%, 30%, 50%, 70%

K50 0-2mm0%, 20%, 30%, 40%, 50%, 60%, 70%

K57 2-4mmK57 0-2mm

0%, 20%, 30%, 40%, 50%, 60%, 70%K50 2-4mm

K57 0-4mm 0%, 20%, 25%, 30%, 35%, 40%, 50%, 70%


34/58


16

The most suitable grading obtained from these tests, referred to as theOptimum grading, was used in further investigations where the use of fillerwas studied.

Filler

This very small fraction of the aggregate can be used to improve some

concrete properties, especially compressive strength, and to replacecement. Two cases were studied: filler addition, where it was used as amineral admixture replacing aggregate; and filler replacing cement, toevaluate whether it is possible or not to reduce cement content.

The mixes that were prepared for the grading optimization of the 0-16 mm concrete were too stiff. Using filler to replace aggregate, as amineral admixture, the stiffness of fresh concrete tends to be evenhigher, therefore the Yousiff-mix design was used instead. With fillerreplacing cement, the Reference-mix design was used, even if theworkability is likely to decrease when the replacement is done in constantw/c.

Filler addition

Addition of filler was performed gradually up to 160 kg/m3 of concrete(Fig. 14).The Optimum gradingwas used for the aggregate, but filler wasadded to replace part of it. Yousiff-mixdesign was used and the aggregatein the 0-4 mm fraction, including the filler, was K50. K33 was used inthe 4-8 mm fraction and the K58 was used in the 8-16 mm.

Filler replacing cement

When filler was used in replacement of cement, the tests were performedusing the Yousiff-mixdesign and the aggregate according to the ReferenceGrading. The addition of filler to replace cement was evaluated in twocases: water content constant and w/c constant. For the water contentconstant, the w/c increased; for the w/c constant the paste contentdecreased (Fig. 15).

Fig. 14 - Mixes design for filler addition.


35/58


17

4. RESULTS

In this section the results for mortar rheology and concrete tests arepresented.

4.1. Mortar rheologyThe results of the rheological tests of mortar are presented in a diagramplastic viscosity [Pa.s] versus yield stress [Pa]. In some diagrams a curveline is used to connect some points. These lines have no mathematicalcorrelation; they are used just to make the visualization easier.

4.1.1. Crushed rocks compared to natural aggregates

The rheological performance of crushed rocks was compared to natural(glaciofluvial) fine aggregate. No changes in the aggregate grading tookplace, only particles above 2 mm were removed, since mortar in thisstudy contemplates only particles below 2 mm. The tests were performedaccording to the Reference-mix design. Wind sieved materials were notincluded in this diagram.

The values of viscosity and yield stress have large variation amongcrushed rocks, and they are considerably higher compared to naturalaggregate (Fig. 16).

Fig. 15 - Mixes design for cement replacement by filler.

Fig. 16Rheological performance of crushed rocks compared tonatural aggregate.


36/58


18

4.1.2. Grading

K50, K49 and K43 crushed rocks with good, average and lowrheological performance, respectively were submitted for rheologytests with Grading1-4, according to the Reference-mixdesign. Grading2hasthe lowest plastic viscosity and the yield stress decreases from Grading1-4.

These results were still not enough to decide which one is the mostappropriate grading; other parameters have to be considered. Therefore,further investigations on fresh concrete were done (Appendix 2).

Due to better cohesion and homogeneity, Grading2was considered themost suitable particle-size distribution for fresh concrete, regarding the0-2 mm aggregate. Thus this grading will from now on be referred to asReference Grading(Fig. 17).

Fig. 17 - Comparisson between Reference and original grading,and natural aggregate.

Fig. 18 - Rheology results for different gradings of K50, K49 andK43.


37/58


19

4.1.3. Reference and original grading

Rheology tests were performed to evaluate how rheology can beimproved by grading optimization. The tests were performed on crushedrocks according to the Reference Grading. The results were compared tothe rheological behavior of crushed rocks according to their originalgrading. The proportions of aggregate, cement and water followed theReference-mixdesign in all cases. The results were also compared to naturalaggregate.

Significant rheological improvements are achieved by the gradingoptimization. However, natural aggregate still has lower plastic viscosityand yield stress (Fig. 18).

4.1.1.Cement reduction

To evaluate the effect on rheology of the paste content, tests with loweramount of cement paste were performed. K50 and K43 were used asfine aggregate and w/c was kept constant. These materials K50 andK43 have, respectively, good and average rheological performancecompared to other crushed rocks; No reduction corresponds to theReference-mixdesign.

Viscosity and yield stress increases when cement content is decreasedwithin a constant w/c ratio. Again, Grading2Reference Gradinghas thelowest plastic viscosity; the yield stress keeps decreasing from Grading1-4(Figs. 19 and 20).

Fig. 19 - Effect of cement reduction for Gradings 1-4 with K50.

Fig. 20 - Effect of cement reduction for Gradings 1-4 with K43.


38/58


20

4.1.2. Superplasticizer

Superplasticizer was added to the mixes to test how the yield stress andplastic viscosity can be lowered. The effect of superplasticizer was testedfor crushed rocks in two cases according to their original and ReferenceGrading. In both cases the rheological performance was compared tocrushed rocks and natural aggregate without superplasticizer; theReference-mixdesign was used.

Significant improvements were achieved by using superplasticizer;however, attention should be paid when using this chemical admixture,in order not to compromise other properties of fresh concrete, such ascohesion and homogeneity.

For mixes with aggregate according to the original grading, 2 ml ofsuperplasticizer were added twice. For mixes according to the Reference-

Fig. 21 - Effect of 2 ml of superplasticizer in mixes with aggregateaccording to their original grading.

Fig. 22 - Effect of 4 ml of superplasticizer in mixes with aggregateaccording to their original grading.


39/58


21

mix design, 2 ml of superplasticizer was added once, because withanother 2 ml dosage it would separate, due to the lower fines content.

A 2 ml addition of superplasticizer on mixes using crushed rocks withoriginal grading was not enough to make their rheology comparable tonatural aggregate (Fig. 21). After a 4 ml addition of superplasticizer, fewcrushed rocks had the rheology close to natural aggregate (Fig. 22). Itwas also observed that with 4 ml of superplasticizer, some crushed rocksmay have lower yield stress, but none of them had lower viscosity. Witha dosage of 2 ml, few crushed rocks according to the Reference Gradinghad the rheology as good as natural aggregate (Fig. 23).

4.1.3. Wind Sieving and cubical shape crushingWind sieving and cube shaping (through vertical shaft impactVSI), canimprove rheological properties. The aggregates were tested according tothe Reference-mix design and Reference Grading, so that only these twoprocesses could influence the rheology. The effect of cube shaping andwind sieving is considerable; however, natural aggregate still has betterrheological behavior (Fig 24).

Fig. 24 - Effect of cube shaping (through VSI crusher) and windsieving.

Fig. 23 - Effect of 2 ml of superplasticizer in mixes with aggregateaccording to the Reference grading.


40/58


22

4.1.4. Cement savings

Cement paste was added to evaluate how much cement can be saved byimproving the grading and using superplasticizer. To have thiscomparison, the rheology of the mix with optimized grading andsuperplasticizer was compared to the mix with no improvement, butwith cement paste addition. The comparison was done for K50 and K57.The Reference-mixdesign was used, and the paste addition is related to it.

K50 according to Grading1 would need an addition of 10% cement pastethe get a rheology close to natural aggregate. If compared to the ReferenceGradingwith 2 ml of superplasticizer, an addition of 15% cement pastewould be necessary to get similar rheology (Fig. 25).

For K57, grading optimization corresponds to an addition of 5% cementpaste. Even if 15% cement paste is added, natural aggregate still hasbetter rheology. Superplasticizer has more effect on yield stress than onplastic viscosity, yet rheology cannot get even close to natural aggregate(Fig. 26).

Fig. 25 - Comparisson of how much cement can be saved by usingsuperplasticizer and grading optimization for K50.

Fig. 26 - Comparison of how much cement can be saved by the useof superplasticizer and grading optimization for K57.


41/58


23

4.2. Fresh concrete testsAs it was described in the Methods section, the maximum aggregate sizewas gradually increased up to 16 mm, as the one used by the buildingindustry. These results will confirm whether the knowledge acquiredfrom mortar can be applied to concrete or not.

4.2.1. Concrete with 0-4 mm aggregateTests were performed to test how the proportion and the aggregate ineach fraction affect the rheology. K50 and K57, respectively withfavorable and unfavorable properties for good rheology, were tested indifferent proportions of 0-2 and 2-4 mm fractions and in differentcombinations.

Plastic viscosity and yield stress were plotted against the proportion of 2-4 mm particles; the Reference-mixdesign was used. Plastic viscosity andyield stress are significantly better using K50 in the 0-2 mm fraction.Yield stress is lowered when the proportion of 2-4 mm particlesincreases (Fig. 27 and 28).

Fig. 27 - Plastic viscosity for different proportions and aggregatesof 0-2 and 2-4 mm

Fig. 28 - Yield stress for different proportions and aggregates of 0-2

and 2-4 mm.


42/58


24


Again, the influence of the aggregate in the 0-2 mm fraction wasevaluated. The 2-8 mm aggregate used was the same for all mixes; for the2-4 mm fraction the K50 was used, and for the 4-8 mm fraction theK33. The tests were done according to the Reference-mixdesign, and theeffect of superplasticizer was also studied.

Rheology tests for mixes using K50 and K57 in the 0- 2 mm fraction andsuperplasticizer addition were done. The rheological performance isconsiderably better using K50 in the 0-2 fraction; with K57 in the 0-2 mm fraction, superplasticizer is not very effective to lower the plasticviscosity (Fig. 29).

The evaluation of how the 0-2 mm fraction and superplasticizer affectthe slump was also studied. It is significantly higher with K50 in the 0-2 mm fraction and superplasticizer has a considerable effect when usingK57 (Fig. 30).


Usually, 16 mm is the maximum aggregate size for the concrete used forhouse buildings. In a first instance the grading was optimized, and thenthe use of filler was assessed.

Fig. 29 - Rheology for concrete with aggregate 0-8 mm.

Fig. 30 - Slump for concrete with 0-8 mm aggregate.


43/58


25

Optimum grading

For grading optimization of the concrete with 0-16 mm aggregate, theaim was to have the minimum of 0-2 mm particles, but still goodcohesion and no separation. These parameters are measured by tactileand visual observations. Even though the mixes were not likely to

separate, the ones with 0-2 mm aggregate lower than 30% of the 0-4 mmdid not provide a cohesive fresh concrete. The low cohesion could beobserved in the slump test, where the truncated cone collapsed insteadof flowing. In addition to that, mixes with less than 30% of 0-2 mmaggregate of the total 0-4 mm could not handle large dosages ofsuperplasticizer without separation.

These tests were done according to the Reference-mixdesign; K33 andK58 were used in the 4-8 and in the 8-16 mm fractions respectively. K50and K57 were used in the 0-4 mm fraction. With K57 in the 0-4 mmfraction the slump was equal or close to zero. Using K50 in the fractionup to 0-4 mm, when the 0-2 mm fraction was higher than 30% of thetotal 0-4 mm aggregate, there was basically no slump at all; for the 0-

2 mm fraction lower than 30%, the cone collapsed due to lack ofcohesion.

Thus the Optimum gradinghas 13% of 0-2 mm particles, while the ReferenceGradinghas 9% (Table 2).

Aggregate type

For this Optimum grading, with 30% of 0-2 mm of the total 0-4 mm

Table 2Gradings for concrete up to 16 mm.

Reference grading for 0-16 mm concrete

Fraction 0-2 mm 2-4 mm 4-8 mm 8-16 mm

Proportion 9% 34% 11% 46%

Optimum grading for 0-16 mm concrete

Fraction 0-2 mm 2-4 mm 4-8 mm 8-16 mm

Proportion 13% 30% 11% 46%

Fig. 31 - Slump and the effect of superplasticizer on fresh concreteworkability.


44/58


26

aggregate, slump tests were performed. The effect on fresh concrete ofthe aggregate type used in the 0-2 and 0-4 mm fraction was studied.

Different combinations of K50 and K57 in these two fractions weretested. For these mixes, K33 was used for the fraction 4-8 mm and forthe 8-16 mm K58 was used. The mixes were according to the Yousiff-mixdesign, and the effect of superplasticizer was studied, an addition of0.3% of superplasticizer was done.

The aggregate varied only in the fraction of 0-2 and 2-4 mm. Using K50in the 0-2 mm fraction the slump is considerably higher than using K57.The influence of the 2-4 mm fraction is not so significant; especially withK50 in the 0-2 mm. Relevant improvements can be done usingsuperplasticizer (Fig. 31).

Filler addition

Slump test was used to evaluate the impact on fresh concrete wor