WR2004-45_web

8/12/2019 WR2004-45_web

1/169

Anna Kronlf

Working Report 2004-45

Injection Grout for Deep Repositories

Low-pH Cementitious Grout for Larger Fractures:

Testing Technical Performance of Materials

8/12/2019 WR2004-45_web

2/169

Anna Kron l f

VTT Bu i l d i ng and T ranspor t

Work ing Repor t 2004 -45

Injection Grout for Deep Repositories

Low-pH Cementitious Grout for Larger Fractures:Testing Technical Performance of Materials

8/12/2019 WR2004-45_web

3/169

TIIVISTELM

Tyn tavoitteena oli kehitt vhintn yksi sementtipohjainen injektointiaine ONKA-LOn sek suomalaisten ja ruotsalaisten ydinjtteen loppusijoitustilojen injektointiin,

joka tytt alla esitetyt vaatimukset:Vaaditut ominaisuudet:

1. pH 112. Tunkeutuvuus bmin80 m

3. Tunkeutuvuus bcrit 120 mToivotut ominaisuudet:4. Viskositeetti 50 mPas5. Veden erottuminen 10%6. Avoin aika 60 min7. Leikkauslujuus 6 tunnissa500 Pa8. Leikkausjnnitys 5 Pa9. Puristuslujuus 28 vrk iss 4 MPa

Lisksi esitettiin yleisi vaatimuksia, joita ei ilmaistu numeerisesti. Niihin kuuluivatsilyvyys sek, raaka-aineiden saatavuus ja tunnettuus kytnnn rakentamisessa.

Tyss tutkittiin nelj injektointi systeemi (sideaine kombinaatiota) seuraavasti:1. Tavallinen Portland sementti silika (silica fume)2. Masuunikuona

3. Supersulfaattisementti4. Matala-alkalisementti (LAC)

LAC oli japanilainen sementtituote, joka oli kehitetty matalan pH:n betoneihin. Sensoveltuvuutta injektointiin testattiin varioimatta mineraalikoostumusta. Ainoastaan vesikiintoainesuhdetta ja hidastimen annostusta modifioitiin. Niiss puitteissa LAC-tuote eisoveltunut injektointiaineeksi.

Muiden systeemien osalta vaatimukset saavutettiin laboratorio-olosuhteissa lukuun otta-matta leikkausjnnityst (reologiaan liittyv suure). Vaatimus puristuslujuudelle eikuulunut alkuperisiin vaatimuksiin. Se asetettiin, kun projekti oli jo ollut kynnissmuutaman kuukauden. Niiden massojen puristuslujuus, jotka valittiin pH testeihinennen kuin puristuslujuusvaatimus asetettiin, on hieman alhaisempi kuin vaatimus.

8/12/2019 WR2004-45_web

4/169

8/12/2019 WR2004-45_web

5/169

8/12/2019 WR2004-45_web

6/169

8/12/2019 WR2004-45_web

7/169

TABLE OF CONTENTS

TIIVISTELMABSTRACTNOTATIONS ................................................................................................................... 31 INTRODUCTION ................................................................................................... 52 OBJECTIVES ........................................................................................................ 73 BACKGROUND TO EXPERIMENTAL STUDIES ................................................. 9

3.1 Background about pH ................................................................................ 93.2 Background about injection properties of grouts with high SF content .... 12

4 INTRODUCTION TO EXPERIMENTAL STUDIES.............................................. 155 MATERIALS ........................................................................................................ 196 METHODS .......................................................................................................... 257 MIX MODIFICATION - EXPERIMENTAL STUDIES ........................................... 29

7.1 OPC - SF system: First experiments ....................................................... 297.1.1 Experiments ................................................................................. 297.1.2 Results and conclusions............................................................... 30

7.2 OPC - SF system: First penetration-ability and pH measurements ......... 33

7.3 OPC - SF system: Ettringite acceleration (ETTA) .................................... 337.3.1 Experiments ................................................................................. 337.3.2 Results and conclusions............................................................... 36

7.4 OPC - SF system: Penetration-ability and pH with ettringiteacceleration (ETTA) ................................................................................. 387.4.1 Experiments ................................................................................. 387.4.2 Results and conclusions............................................................... 38

7.5 OPC - SF system: Effect of W/DM with ettringite acceleration (ETTA) .... 39

7.5.1 Experiments ................................................................................. 397.5.2 Results and conclusions............................................................... 407.6 OPC - SF system: Low alkali white cement (WCE) ................................. 43

7.6.1 Experiments ................................................................................. 437.6.2 Results and conclusions............................................................... 43

7.7 OPC - SF system: Effect of premixing SF with cement ........................... 447.7.1 Experiments ................................................................................. 447.7.2 Results and conclusions............................................................... 44

7.8 Slag system: First experiments ................................................................ 457.8.1 Experiments ................................................................................. 457.8.2 Results and conclusions............................................................... 46

7.9 Slag and Super Sulphate Cement systems (SSC): Activation of slag ..... 507.9.1 Experiments ................................................................................. 507.9.2 Results and conclusions on Slag system and ETTA - slag

batch SL15 ................................................................................... 53

8/12/2019 WR2004-45_web

8/169

9.1.1 Compositions of mixes tested for pH............................................ 67

9.1.2 Effect of alkalis on pH................................................................... 709.1.3 Effect of Ca, Mg, Si, Fe, Al and SO3 on pH .................................. 739.1.4 Effect of curing temperature on pH .............................................. 76

9.2 Penetration-ability .................................................................................... 769.2.1 Comparison of filter pump and penetration-ability

(Bminand Bcrit) methods ................................................................. 769.2.2 Effect of water / dry materials ratio on penetration-ability

(Bminand Bcrit) ............................................................................... 79

9.2.3 Effect of rheology on penetration-ability (BminandBcrit) ................ 809.2.4 Observations on cake build-up phenomena inpenetration-ability test (Bminand Bcrit) ........................................... 81

9.3 Compressive strength .............................................................................. 839.4 Setting (shear strength at 6 h) ................................................................. 85

10 CONCLUSIONS .................................................................................................. 8710.1 Portland cement silica fume system (OPC SF) .................................. 8710.2 Super sulphate cement system (SSC) ..................................................... 8710.3 Slag system activated with OPC (Slag) ................................................... 8710.4 Low Alkali Cement system (LAC) ............................................................ 8810.5 Effect of alkalis (Na2O and K2O) on pH .................................................... 8810.6 Effect of Ca, Si, Fe, Al, SO3 on pH ........................................................... 8810.7 Penetration-ability .................................................................................... 8810.8 Compressive strength .............................................................................. 8910.9 Setting (shear strength at 6 h) ................................................................. 89

11 SUMMARY .......................................................................................................... 9112 FUTURE NEEDS ................................................................................................ 95

12.1 Requirements (open time) ....................................................................... 9512.2 Mixing order ............................................................................................. 9512.3 Superplasticizer ....................................................................................... 9612.4 Glass and alkalis ...................................................................................... 9612.5 Binder development ................................................................................. 9612.6 Ettringite acceleration (ETTA) control ...................................................... 9712.7 Sulphides and sulphates .......................................................................... 97

REFERENCES ............................................................................................................. 99

APPENDICESAppendix 1 Requirements for the low pH cementitious grout .................................... 101Appendix 2 Determination of the filtration stability ..................................................... 111Appendix 3 Short manual for the Penetrability meter ................................................ 115Appendix 4 Details of procedures followed when determining injection properties ... 121

8/12/2019 WR2004-45_web

9/169

NOTATIONS

The notations used in the report are as follows:

Bcrit = critical aperture (related to grout penetration-ability)Bmin = minimum aperture (related to grout penetration-ability)CH = calcium hydroxideCSH = calcium silicate hydrate

DM = dry materials (all dry materials including the dry content of slurries)ETTA = ettringite-accelerationG = gypsumHAC = alumina cement (High Alumina Cement)LAC = Low Alkali Cement (product name)OPC = Ordinary Portland Cement (white, grey or SR cement)SF= silica fumeSH = silicate hydrate

SL = blast furnace slagSP = superplasticizerSR = sulphate resistantSSC = super sulphate cementUF16 = Ultra fin 16 (A commercial injection cement by Cementa, Sweden)W = water (Water from all sources; pure water, water from slurries and other chemicals)WCE = Egyptian white cement (very low alkali content).

The ratios such as W/DM or SF/OPC are given as weight units of the dry materials. Inthe case of GroutAid (a silica slurry product) SF denotes only the dry material contentof GroutAid.

In the ratios G/OPC or G/ SL (often in text Gypsum/OPC or Gypsum/SL) gypsumdenotes only the gypsum added while making the mix. All gypsum that is a part of theoriginal OPC product is considered to be OPC only.

Note: OPC = Ordinary Portland cement (white, grey or sulphate resistant). In literaturealso PC is used. PC (Portland Cement) = OPC.

8/12/2019 WR2004-45_web

10/169

8/12/2019 WR2004-45_web

11/169

1 INTRODUCTION

Posiva and SKB are planning to deposit spent nuclear fuel in deep repositories. Use ofcommon construction materials, as steel and concrete, are foreseen. With respect to thelong-term safety a suitable chemical environment is vital. The use of low-pH products isnecessary in order to get leachates with sufficiently low pH (< 11).

A pre-study was carried out in 2001, followed by feasibility study in 2002 mid 2003.The present work is based on the outcome of those earlier studies as well as the outcome

of related studies, which are compiled to the summary report of the project "InjectionGrout for Deep Repositories".

The aim of the whole project "The injection grout for deep repository" was to achievesome (at least one) well-quantified, tested and approved low-pH injection grouts to beused for smaller fractures ( 100 m) in futurerepositories. The whole project, which is a joint project between Swedish Nuclear Fueland Waste Management Co (later SKB) and Nuclear Waste Management Organisationof Japan (later NUMO) and Posiva Oy (later Posiva), was further divided into four sub-

projects:1. Low-pH cementitious injection grout for larger fractures2. Non-cementitious low-pH injection grout for smaller fractures3. Field testing in Finland4. Field testing in Sweden

The subproject SP1 "Low-pH cementitious injection grouts for larger fractures" waslead by Posiva. The aim was to develop low-pH grout for sealing larger fractures (> 100m) in the bedrock of surroundings of the deep repository. It included seven tasks:

1. Setting the requirements for the grouting materials to be developed2. Selecting materials for grinding and testing3. Grinding the materials4. Testing the technical performance of the materials5. Testing pH and leaching behaviour of the most promising materials6. Evaluating the environmental acceptance of the materials7. Evaluating the long-term safety of the materials

The work report presently covers task 4. The objective of Task 4 was to developinjection grouts based on task 2 with a low-pH and injection as well as long-term

i d fi d i T k 1 d i b l (Ch 2 Obj i ) d l

8/12/2019 WR2004-45_web

12/169

penetration-ability as well as material selection restrictions based on long-term safety

aspects (Appendix 1).

According to the brief literature review and information gathered through discussionswith product suppliers, it became evident that neither are there commonly acceptedtheories about grout behaviour during grouting nor good understanding about how mixdesign affects the behaviour. Each grouting cement behaves in its own individual way,depending strongly on its chemical composition, the type and dosage of additives used,duration of the mixing period and ambient temperature. Details of cement products are

often kept confidential.

In this work VTT has modified the composition of the four candidate grouting systemse.g. material combinations and performed the necessary laboratory tests. Testedmaterials were commercially available. Only the fineness of the materials was modified

by grinding when needed.

Mixes, which gave the best overall results in the laboratory tests, were selected for pilotfield tests.

8/12/2019 WR2004-45_web

13/169

2 OBJECTIVES

The objective of this work was to design at least one mix that would meet givenrequirements, as listed below (Table 1) (Appendix 1).

Table 1. Required properties of low-pH cement based grouts.

Order of

Importance

Property Requirement Measuring method

RequiredProperties

pH 11 Leaching tests

Penetration-abilitybminPenetration-ability

bcrit

80 m120 m

Penetrability meterat 60 min

Desired

Properties

Viscosity 50 mPas Rheometry at 60 minBleed 10% Measuring glass at 2 hours

Workability time

Shear strengthYield value

60 min

500 Pa5 Pa

Determined by penetration-ability and viscosityFall cone at 6 hRheometry at 60 min

Compressive strength 4 MPa Uni-axial compressivestrength at 28 d 1)

1) The compressive strength requirement was not included to the original requirements.

It was set after the project had been proceeding for a few months.

In addition to the requirements listed in Table 1 the grout should fulfil other, moregeneral requirements that cannot so far be expressed by any numbers (Table 2)(Appendix 1).

Table 2. Other desired properties.

Material must be available in practice during the construction and operation of

the repository Material (or at least its components) must have a history of use in cement

technology (or practical engineering) Durability (chemical and physical) properties of the material needs to be

sufficient that the grouted zone maintains its required properties during thet d lif ti

8/12/2019 WR2004-45_web

14/169

2. Slag denotes an OPC activated slag based system. Alkali and water glass

activation were not examined, as low-pH was the most important requiredproperty and alkali activation was considered an unnecessary risk.3. SSC is a slag based system activated with gypsum (G) and OPC, the content

of G being higher than that of OPC. A few experiments on enhancing SSC withalkali activation were made. The systems Slag and SSC are reported here as one

parallel system.4. LAC was introduced to the project by NUMO as a product, ground to fixed

fineness by the producer. Neither were its mineral composition nor fineness

modified in the present experiments.

Originally, according to the project plan, also fly ash was to be studied. However, flyash was ruled out at an early stage of the work as there were problems foreseen in thedelivery and quality stability of the material.

The scientific modelling of the mechanisms of grout behaviour (setting, rheology,penetration-ability, compressive strength, pH) was not included in the objectives of thiswork (Task 4). All pH measurements were performed by Ulla Vuorinen at VTTProcesses and the pH and leaching tests are reported in a separate report (Vuorinen et al.2004).

8/12/2019 WR2004-45_web

15/169

3 BACKGROUND TO EXPERIMENTAL STUDIES

3.1 Background about pH

CaO

The goal of the work was to develop injection grouts with low pH (11) to be used in arepository for spent fuel. According to the work by Lagerblad partly presented inLagerblag (2001), the use of silica fume was found to be an efficient component for

lowering the pH of grout leachate. One of the reaction mechanisms of silica fume is thesilicate reaction with Ca(OH)2, which forms 20% of pure Portland cements hydrationproducts. Because the equilibrium pH of calcium hydroxide with water is as high as12.5, it is clear the all Ca(OH)2needs to be consumed in the reaction in order to lowerthe pH to 11 or lower.

Silica fume is a very reactive pozzolan. The reaction is fast enough to bind Ca(OH)2into CSH to a large extent as it forms. Yet, it is likely that a part of the silica fume

particles react slowly and remain un-reacted over long periods (years) of time.

It is possible to use also less reactive pozzolans such as blast furnace slag, glass orquartz. Those would first allow Ca(OH)2precipitation. The Ca(OH)2crystals would bein contact with pore solution, which would be seen as high pH until the pozzolanicreaction proceeds.

A complete reaction of Ca(OH)2and silica into CSH does not produce a product with asufficiently low pH. This is because also CSH with a high Ca/Si -ratio (1.8) yields high

pH (Table 3, Stronach & Glasser 1997). According to Table 3 to reach the pH of 11.03or lower the total Si content composed of all Si sources such as pozzolans, OPC or blastfurnace slag, needs to be high enough to produce CSH with the Ca/Si ratio of 1.1 orlower.

As the pozzolanic reactions proceeds the Ca/Si ratio of the CSH decreases which in turn

decreases the pH. Therefore the pH does not depend solely on chemical composition(elements) or material combination, but on the composition of reaction products formed,which depends largely on reaction rate and time.

The CSH-reactions takes several months, possible years to be completed. The reactionrate depends on the fineness and chemical composition and content of vitreous material

8/12/2019 WR2004-45_web

16/169

Table 3. Invariant points in the CaO-SiO2-H2O system at 25C (Stronach & Glasser

1997).

Solids in equilibrium [Si]aq(mg/L) pH

Amorphous SHAmorphous SH + CSH (0.8)

CSH (0.8)CSH (0.8) + CSH (1.1)

CSH (1.1)

CSH (1.1) + CSH (1.8)CH + CSH (1.8)

CH

39.5117.043.441.131.4

0.70.50

6.3810.1710.8810.9111.03

12.4312.5312.52

The amount of silica fume needed for pH < 11 depends on the cement composition.

According to the work by Lagerblad partly presented in Lagerblag (2001) approx. 30 w-% silica fume was needed with Aalborg White Cement (SF 30 w%, OPC 70 w%,SF/OPC = 0.43). According to his estimation this would give a CaO/SiO2 ratio of theCSH product lower than 1 and consequently pH below 11. A closer estimation is givenin Table 4. According to Table 4 slightly too low SF/OPC ratios was used both byLagerblad as well as in the preliminary experiments of this work to yield the CaO/SiO2ratio of 1.

Table 4. Estimation of SF/OPC ratio needed to yield Ca/Si ratio on 1 in the CSH.

CaOcem

CaOAl+Fereact.

CaOin

CSH

CaOin

CSH

SiO2inCSH

4)

SiO2inCSH

4)

SiO2in cem

5)

SF6)

SF7)

SiO2in mix

8)w% w% w% moles w% moles w% w% w% w%

100 g cem 100 g cem SF/OPC SF/OPC

OPC 1) 65 15.0 50 0.89 53 0.89 25 0.28 0.43 46

OPC 2)Aalborg

69.0 3.5 65.5 1.17 66 1.17 25 0.45 0.43 46

UF16 3) 64.6 10.2 54.4 0.97 47 0.97 22.8 0.35 0.30 401) Figures from the calculations by Lagerblad based on approximate cement composition. (Unpublishedwork, Lagerblad 2003).2) Calculations based on White OPC (from Aalborg plant) chemical composition (Table 6). CaOconsumed in Al and Fe reactions was estimated according to Equations 1 and 2.3) Calculations based on UF16 chemical composition (Table 6).CaO consumed in Al and Fe reactionswas estimated according to Equations 1 and 2

8/12/2019 WR2004-45_web

17/169

C4AF +2 CH + 10 H = C3AH6 + C3FH6 (2)

Ca(OH)2+ SiO2 = CSH (1) (3)

CSH (1) + + SiO2 = CSH (2) (4)

whereC is CaOS is SiO2

A is Al2O3Fe is Fe2O3H is H2O

C3A is tri calcium aluminateC3AH6 is tri calcium aluminate hydrateC4AF is tetra calcium aluminate ferrateC3FH6 is tri calcium ferrate hydrate

CSH (1) is calcium silicate hydrate with higher Ca/Si contentCSH (2) is calcium silicate hydrate with lower Ca/Si content

According to Cline Cau Dit Coumes et al (2004) the total SiO2 content showed goodcorrelation with the equilibrium pH regardless of the composition. The SiO2 content(originating from all sources) needed to yield the pH 11 or less is 55% or more. Thereason for this behaviour is not known. This leads to a higher demand for SF than

shown in the Table 3 above: According to the 55w% requirement the minimumSF/OPC would be as high as 0.75 and 0.80 for the white cement used by Lagerblad(2003) and UF16 used in the present work, respectively.

8/12/2019 WR2004-45_web

18/169

pH mostly controled by silica content in binding

OPC/SF/FA

OPC/MK/FA

OPC/SF

OPC/MK

y = -0.0707x + 14.912

R2= 0.91

10

10.5

11

11.5

12

12.5

13

30 35 40 45 50 55 60 65 70

%SiO2in binding

Equilibriu

m

pH

Figure1. Equilibrium pH vs. total SiO2 content in the binder material. The silica

content (total SiO2) should be 55 w% or more to ensure pH 11or lower. (Cau Dit

Coumes et al 2004).

Na2O + K2O

Another mechanism is CSHs tendency to bind alkali hydroxides, which are responsiblefor the early age pH peak value of 13.5. When cement paste ages, the early age pH peakvalue drops from 13.5 to around 12.5. This pH drop is understood to be an indication ofalkalis being bound to CSH. This is known to happen also in ordinary Portland cementwithout pozzolan addition.

3.2 Background about injection properties of grouts with high SFcontent

In Kronlfs work (2003) it was found that with fine cements some other mechanismapart from the traditional particle blocking inhibits penetration. This other mechanismwas named gel blocking. It dominated the penetration-ability of mixes made with fine

8/12/2019 WR2004-45_web

19/169

8/12/2019 WR2004-45_web

20/169

8/12/2019 WR2004-45_web

21/169

4 INTRODUCTION TO EXPERIMENTAL STUDIES

The research was planned to proceed in a stepwise manner so that the most interestingsystems were to be taken into closer research and less promising were to be ruled out.The results were planned to be examined shortly after testing and the decisions aboutdeveloping/modifying the systems were to be made continuously. The pH and leachingtests were performed by Ulla Vuorinen only to the most promising grout mixes(Vuorinen et al. 2004).

Mixes were based on the four different preliminary systems that were modifiedexperimentally.

1. Ordinary Portland Cement + Silica Fume (OPC+SF)2. Blast furnace slag (Slag)3. Super sulphate cement (SSC)4. Low Alkali Cement (LAC)

Originally there was also a fifth system based on fly ash, but that was ruled out due toexpected problems concerning stability of raw material properties (chemicalcomposition and reactivity) over long periods of time as well as the expected difficultieswith the setting time of the mixes. No other systems were ruled out during the workreported presently (Task 4).

The work proceeded through a 15-step procedure each giving necessary knowledge forthe next step as listed below. During the process a number of additions were made to the

original testing program.

Step 1 First tests on OPC+SF and Slag systems

The first systems examined were OPC+SF and Slag. SF was used in the form ofGroutAid (Chapter 5). The slag was ground by a pilot plant jet mill by CT-Group. Theamount of SF was kept to minimum due to its

- detrimental effect to penetration-ability as large quantities

- tendency to increase water demand- tendency to retard strength development, if water demand is increased due to

high SF content.- cost.

Slag was activated by OPC only and its content was also kept to minimum for pH

8/12/2019 WR2004-45_web

22/169

Step 2 First penetration-ability and pH measurements

A few most promising mixes out of the ones tested in step 1 were chosen to be tested forpenetration-ability by penetration-ability meter (Chapter 6). Based on their overallinjection properties one OPC+SF mix (12) and one slag mix (44) were later chosen forthe pH test.- The penetration-ability of OPC+SF mix (12) was within requirements but that of the

slag mix (44) was slightly too poor.- The shear strength at 6 h OPC+SF mix (12) was too low (poor) while that of slag

mix (44) was within requirements.- The preliminary pH of both mixes was too high (poor).

The low shear strength was disturbing, because the early age strength development istemperature sensitive and mix temperature in field conditions might be lower than 12C, which would lower the shear strength respectively. On site the temperature ofmaterials may be lower than 12 C and the agitation period shorter. The later would givethe reactions less time to proceed at the agitation temperature that is slightly higherthan the ambient temperature. Therefore in the laboratory a few extra kPa of shearstrength over the given requirement was preferred to avoid grouting problems on site.

Step 3 Checkpoints

- Something needed to be done to lower the pH. Increase of SF was the first obviouschoice. Also lower alkali content through material selection was considered.

- Something needed to be done to increase the shear strength at the age of 6 h. Typicalaccelerators (Cl- and NO3-) were ruled out due to long-term safety reasons.

Commercial shotcrete (spray concrete) accelerator (Meyco SA 161), which is basedon aluminium salts and produces ettringite, was suitable from the long-term safetyaspect, but it reacted too fast and the mix lost the penetration-ability altogether.

- At this stage superplasiticizers were ruled out as to simplify long-term safetyanalyses.

- The filter pump was found to be a too approximate method for penetration-abilityassessment even in the initial stages of any mix development. Therefore the

penetrability meter (determination of Bminand Bcritvalues) was introduced to the testprogram throughout the mix development.

- Fly ash system was ruled out without testing.

8/12/2019 WR2004-45_web

23/169

- large water to dry materials ratio for penetration-ability and viscosity

- ETTA for setting (6 h shear strength).Two mixes with differing SF contents (f63 and f64) were tested for pH, which waswithin the requirements in both cases.

Step 6 - WCE

Low alkali white cement from Aalborg White Sinai (Egyptian) plant (WCE) was

introduced and the two mixes were tested for pH. Alkali content of the cement wasfound not to be the pH-determining factor.

Step 7 Pre-mixing

Pre-mixing of components by jet mill was tested with and without dry SF (un-densified,type 938). The aim was to produce a single product and reduce the number of materialsto be handled on site. Premixing the un-densified SF with cement instead of using

GroutAid deteriorated the penetration-ability.

Step 8 - Low Alkali Cement (LAC)

LAC tested and completed. Low penetration-ability was achieved and therefore LACwas not suitable for grouting.

Step 9 - pH

Correlation between the chemical composition of the mixes and the results of the pHtests was examined. As modelling was outside the scope of this work, this study did

produce neither modelling nor validation of any existing model. Yet, the observationsoffered valuable guidelines for further mix modification.

Step 10 - Compressive strength

The requirement for uni-axial compressive strength was set to 4 MPa.

First the testing age was not defined because the strength development rate is notcritical. The testing age was initially 28 d. Later also 91 d was added to the program.

Setting the requirement was somewhat contradictory because unreasonably high

8/12/2019 WR2004-45_web

24/169

8/12/2019 WR2004-45_web

25/169

5 MATERIALS

The following materials were used in the experiments. The particle size distributionsexcept that of Injekipsi 1, LAC fine LAC coarse and Rheocem 900 were determined byCT-group by Coulter Counter (Figures 2 and 3 and Table 5). Injekipsi 1, LAC fine,LAC coarse and Rheocem 900 were determined by the producer. The chemicalcomposition is given according to the supplier given information in Table 6.

- Ultrafin 16 (UF16)

OPC, SR, micro cement developed for grouting purposes by Cementa- White Portland cement from Egypt (WCE)

OPC, low alkali by Aalborg White from Sinai plantWCE VTT; ground at VTT with a jet millWCE CT; ground at CT-Group with a jet mill

- White cement (WC10)

OPC, low alkali by Aalborg White from Aalborg plant- Rheocem 900

OPC micro cement developed for grouting purposes by Master Builders- High alumina cement (HAC)

Secar 71 by Lafarge aluminates (HAC)HAC; ground at VTT with a jet mill

- Rapid hardening Portland cement (RC)

OPCCEM II A 42.5 R

Ground with jet mill by CT-Group- Injekipsi 1 (G)A gypsum slurry by Kemira, CaSO42 H2O. DM content 66.5 67.5%.The two water molecules are included on the DM.

- Blast furnace slag (SL)

Ground with a jet mill by CT-Group into two finenesses.SL15; d 98%value about 15 mSL10; d 98%value about 10 m

- GroutAid (GA)Slurry made of silica fume type UN 920 developed for grouting

purposes by ElkemSolid content 48 - 52 %SiO2typically 95 % of solidsC b 2 5 % f lid

8/12/2019 WR2004-45_web

26/169

Dry mix of WCE CT, HAC and G all ground and pre-mixed at CT-

Group- LAC coarseLow Alkali Cement by NUMO, coarse grinding

- LAC fine

Low Alkali Cement by NUMO, fine grinding- SP40

Superplasiticizer, melamine based (sulphonated melamineformaldehyde condensates) Dry material content 40 w%.

- Meyco SA 161Shotcrete accelerator, aluminium salt

0

20

40

60

80

100

0.1 0.2 0.5 1 2 5 10 20

m

%

UF16WCE VTTWC10HACInjekipsi 1

0

20

40

60

80

100

0.1 0.2 0.5 1 2 5 10 20

m

%

UF16

SL15

SL10/1

SL10/2

SL10/3

RC10/1RC10/2

(a) (b)

Figure 2. Particle size distributions of materials.

20

40

60

80

100

%

UF16

WCE CT

G3

G4

20

40

60

80

100

%

UF16

LAC coarse

LAC fine

8/12/2019 WR2004-45_web

27/169

Table 5. Material fineness. The values are based on particle size measurements made

by CT-group except that of Rheocem 900.Ultrafin 16

HAC WCEVTT

SL10/1 SL15 SL10/2 SL10/3 RC10/1 RC10/2 WCECT

G3 G4

1) 2) 3) 4) 5) 6) 1) 1) 1)Groundby

Cementa VTT VTT CT CT CT CT CT CT CT CT CT

Median,m

3.8 3.6 2.4 1.9 2.5 2.2 1.9 2.4 2.8 1.9 5.5 1.8

d98, m 13.9 15.8 12.8 10.8 14.0 11.7 10.3 11.7 11.8 13.0 17.0 11.5Specific

surfacearea,cm2/g

14406 13141 17341 19023 15465 17059 17925 17740 16584 18895 11218 19862

> 63m 7)

0.3

1) Ground by the pilot plant jet mill by CT-Group.2) Batch no 1 ground by full scale jet mill by CT-Group. (Designated Ajo skki 1).3) Batch no 2 ground by full scale jet mill by CT CT-Group. (Designated Skki 11).4) Batch no 3 ground by full scale jet mill by CT CT-Group. (Designated Skki 12).5) Ground by a pilot plant jet mill by CT-Group.

6) Ground by full scale jet mill by CT-Group.7) Wash sieving.

8/12/2019 WR2004-45_web

28/169

8/12/2019 WR2004-45_web

29/169

3) Clinker composition. Gypsum taken into account an. Cement includes also ground CaCO3 7w%. It ischemically nearly inert and therefore it is not taken into account in the composition. Alkali content of

sample taken May 5.5 2003. Other values are averages in 2003.4) Typical chemical compositions based on analysis received from the manufacturer (ScancemChemicals, Norway) in the first part of 20035) Typical chemical compositions based on analysis received from the manufacturer (Cementa,Degerhamn, Sweden) in the first part of 20036) Cement composition. Alkali content of sample taken May 5.5 2003. Other values are averages in the

period Sep. 4. 2002 - 20.6.20037) Composition of cement produced in Aug. and Sep. 20038) Given as Na20 + K2O

9) LAC may contain minor traces of sulphides (Vuorinen et al. 2004)10) Grinding aid probably used. The content not given by the producer11) Grinding aid (salts of alkaloamines) used 0.05%. Organic acids used for retardation 0.5 - 1.5%.12) All organic

Table 6b.Sources of information in Table 6a.

Product Producer

Blast furnaceslag

CT/Rautaruukki Aki Kyckling. Finnsementti Oy, e-mail Mar. 29.2004

(Finnsementti uses blast furnace slag in cement production.)GroutAid (a) Elkem Product data sheet : GroutAid MultiGrout. Product P-2-1GroutAid (b) Elkem Anne-Marit Tonnesland, Elkem, e-mail Dec.15.2003HAC Secar 71 Lafarge Secar 71 Product Data Sheet

Reference FC-S71-RE-GB-LAF-10/02Injekipsi 1Gypsum slurry

Kemira Safety data sheet Oct.10.2003. Ref 813/4.0/FIN/FINCorrespondence: Hannu ijl, Kemira, e-mail Nov.17.2003

LAC NUMO/CRIEPI Harutake Imoto, Central Research Institute of Electric Power Industry(CRIEPI,) e-mail 29.1.2004

RC10(Rapid)

Finnsementti Oy Satu Kosomaa, Finnsementti Oy, e-mail Sep.2.2003.

Rheocem 900 Master Builders Steve Odell, Laferge Special Cements, e-mail Sep.8.2003SF UN 920 (a) Elkem Product data sheet: Elkem Microsilica. Product C2-01. Grade 929 for

construction. The values were considered too inaccurate. Further information(below) was obtained from producer.

SF UN 920(b)

Elkem Anne-Marit Tonnesland e-mail, Elkem, Dec.15.2003

SF UN 983 (a) Elkem Product specification. Elkem Microsilica 983. Dec.2002SF UN 983

(b)

Elkem

Anne-Marit Tonnesland e-mail, Elkem, Dec.15.2003SP40. (a) Scancem

ChemicalsNa20 content = 5.6%Product data sheet: SCANCEM SP-40. Scancem. Prod No.2152.A/S Scancem Chemicals Oslo Norway Phone +47 22 87 85 30

SP40 (b) ScancemChemicals

Terje Nilsen, Elkem, e-mail Sep.2.2003

Ultra fin 16 Cementa T j Nil Elk il S 2 2003

8/12/2019 WR2004-45_web

30/169

8/12/2019 WR2004-45_web

31/169

6 METHODS

The following experimental methods were used. Detailed mixing procedures along withsampling for each measurement are given in Chapter 7 for each set of experiments.

Temperature

Of the properties listed in the Table 1 all others except pH and compressive strengthwere determined at the ambient temperature of 12 C (approx. the temperature of

Olkiluoto bedrock at a depth of 400-500 m). The compressive strength specimens werecured for the first 24 h at the temperature of 12 C followed by curing at 20 C untiltested. For pH and leach testing samples two procedures have been followed; 1) curingat 20oC until tested and 2) the first two weeks at 20 oC followed by curing at 20 C untiltested. All materials and equipment were tempered at the ambient temperature of 12 C

prior testing.

Filtration stability with filter pump

The penetration-ability through a 100 m filter with a hand-operated pump wasmeasured at the age of 1 h (after mixing and agitation for 1 h at the temperature of 12C). Pressure difference over the filter was 1 bar (0.1 MPa) and sieve diameter 30 mm.The volume of the pump was 300 ml, which denotes the maximum penetration-abilitythrough the sieve. The larger the value up to the 300 ml the better is the penetration-ability. A detailed description of the method is given in Appendix 2 (VattenfallUtveckling AB 1996). Some details of the procedure are given also in Appendix 4.

Penetration-ability with penetrabilitymeter

The penetration-ability through filters with different sieve openings was measured at theage of 1 h (after mixing and agitation for 1 h at the temperature of 12 C) at the ambienttemperature of 20 C. The penetrabilitymeter was cooled with water (8 - 10 C) prior tomeasurement. The change was due to practical reasons: The measurement could not be

preformed in the cooled room because the space was limited and the sewerage of theroom was not suitable for the test. Pressure difference over the filter was 1 bar (0.1MPa). The penetration-ability is given as bminand bcritvalues. The smaller the value the

better is the penetration-ability. A detailed description of the method is given inAppendix 3 along with some details of the procedure in Appendix 4.

8/12/2019 WR2004-45_web

32/169

accelerator. The method is modified from the bleeding test in the Technical code BY 1(1972). Specimen height is as in BY 1, but the volume is smaller.

Early shear strength (fall cone test)

The fall cone test was used for testing setting of the mixes at the age of 6 h (after thestart of the mixing procedure) at the temperature of 12 C. The undisturbed period was5 h 55 min if all components were added during the 5 min mixing period. Theundisturbed period was 5 h if any component was added after the 1 h agitation period.

The method gives the early age shear strength. Five individual measurements weremade, the highest and lowest vales were excluded and the result given as the average ofremaining 3 values (Lojander 1985).

Rheology

The plastic viscosity and yield value was measured at the age of 1 h (after mixing andagitation) at the temperature of 12 C by co-axial rheometry. The measurement started

at 0 rpm and was done in 119 steps of increased shear rate up to 238 rpm and thedecreasing shear rate was done the same way back to 0 rpm. Every measurement took 2seconds and the total time for measuring was 9 min. The data was used to determine theyield value and viscosity applying the Bingham model (Equation 5) and Casson model(Equation 6). In the later the data points are linearized along x1/2 an y1/2axis. It is usedwith pseudo plastic materials. The measurements were done with a DIN adapter whichcomplies with DIN 53019 requirements for sample geometry. The DIN adapter has acylindrical geometry which provides defined shear rates. A measurement loop described

above was selected to collected data through the whole measuring time. The data wasprocessed with the Rheocalc software which provides modelling according to severalmodels (Bingham, Casson, Power law, IPC Paste analysis). Together Brookfield DV-III+ Rheometer and the Rheocalc software can be used to predict a materials flow,spray, or pumping behaviour by studying shear rate profiles. The measurements were

performed by Juha Hokkanen VTT PRO and data processed by Juha Hokkanen VTTPRO. The results are not published elsewhere.

= o+ p* (5)

()1/2= o+ p* ()1/2 (6)

where

8/12/2019 WR2004-45_web

33/169

Compressive strength

The compressive strength was measured by casting the specimen into 40 mm * 40 mm *160 mm moulds. The specimens were stored at the temperature of 12 C for 24 hfollowing curing at the temperature of 20 C, RH 100% until tested at the ages of 28 dand 91 d. The result given is the average of three determinations. The procedure is amodification of the standard SFS-EN196-1. According to the standard the result is givenas the average of six determinations. In this work only three determinations were usedfor practical reasons to save materials. Another difference compared to SFS-EN196-1

was the curing conditions. According to the standard the curing is 20 C for 24 h andafter de-moulding in water. In this work the specimen were kept covered with plastic orglass plate at 12 C for 24 h in order to make observations of hardening at thattemperature. After the 24 h on they were further kept in the moulds covered becauseseveral samples were too week to be de-moulded.

Outlet d = 4,8 mm

304Filling height 281 mm

50 mm

d = 151 mm

Figure 4. Schematic illustration of the Marsh cone. The volume of specimen run

through the cone was 1 litre. All dimensions in mm.

8/12/2019 WR2004-45_web

34/169

8/12/2019 WR2004-45_web

35/169

7 MIX MODIFICATION - EXPERIMENTAL STUDIES

7.1 OPC - SF system: First experiments

7.1.1 Experiments

Grouting mixes based on the SR cement UF16 and SF in the form of GroutAid weremodified to reach the required injection properties. The SF/OPC ratio was constant 0.30according to calculations shown in Table 4. Material information is given in Chapter 5,the base mix compositions in Table 7 and mixing orders 1, 2 and 4 in Table 8. The aimof testing the mixing orders 1 and 2 was to show if delayed addition of GroutAid wouldimprove the properties by minimizing the effects of gel blocking (Kronlf 2003). Thedelay was 1 h, as long as the open time requirement. The bases for this set up was agrouting procedure where GroutAid would be added in the nozzle instead of mixing it inthe mixer with other components. Mixing order 4 was applied when shotcreteaccelerator (Meyco SA 161) was tested. The results are discussed in the up-comingfigures below. The mix compositions and test results are given in detail in Appendix 5

in Tables 1, 2 and 3 as follows: The effect of W/DM (1 2.2) in the rows 1101 1104. The effect of mixing orders 1 and 2 (Table 8) with SP (1%) and without SP with

a constant W/DM of 1.26 in the rows 1107 1115. The effect of shotcrete accelerator (Meyco SA 161) with the mixing order 4

(Table 8) and a constant W/DM of 1.26 in the rows 1118 1121. Comparison of the UF16 cement to a relatively rapidly reacting OPC micro

cement (Rheocem 900) with the mixing orders 1 and 2 in the rows 1125 1130.

The accelerator (Meyco SA 161) and more rapid micro cement were applied toovercome the problem of too slow setting (shear strength at the age 6 h after mixingwater). The accelerator was added at the end of the delay period for the reasons above,to produce the best possible conditions for the accelerator performance. The testingtemperature was 12 C throughout the experiments as described earlier (Chapter 6).

Table 7. The first OPC-SF mix compositions.Binder OPC

-typeSFtype

SF/OPC

OPC/DM

SF/DM

Superpl./DM

Mixingorder

Water/DM

OPC-SF UF16 orRheocem900

GroutAid'

0.30 0.77 0.23 0%or

1%

1or2

1-2.2

8/12/2019 WR2004-45_web

36/169

Table 8. Mixing orders and testing procedures 1, 2, and 4 for the OPC+SF-mixes.

Mixing order and testingprocedure 1



1 Water + dry materials Water + dry materials Water + dry materials2 Start timing for all tests. Start timing for all tests. Start timing for all tests.3 Mix 2 min 12 000 rpm 1) Mix 2 min 12 000 rpm 1) Mix 2 min 12 000 rpm 1)4 Superpl. if any Superpl. if any Superpl. if any5 Mix GroutAid Mix 3 min 12 000 rpm 1) Mix 3 min 12 000 rpm 1)6 Mix 3 min 12 000 rpm 1) Do not test yet. Do not test yet.7 Start bleeding and

shear strength tests

Move the mix to agitator Move the mix to agitator

8 Move the mix to agitator Slow agitation for 1h Slow agitation for 1h9 Slow agitation for 1h Mix GroutAid for 0.5 min Mix GroutAid and shotcrete

accelerator for 0.5 min10 Test rheology and penetration-

ability at 1hTest rheology and penetration-ability at 1 h. Start bleedingand shear strength tests at 1 h.

Test rheology and penetration-ability at 1 h. Start bleeding andshear strength tests at 1 h.

11 Cast for pH and comp. str. Cast for pH and comp. str. Cast for pH and comp. str.12 Test bleeding at 2 h and shear

strength at 6 h

Test bleeding at 3 h and shear

strength at 6 h

Test bleeding at 3 h shear

strength at 6 h13 Test compressive. strength at

28/91 dTest compressive strength at28/91 d

Test compressive strength at28/91 d

14 Selected specimen sent to pH testto be tested at 2 months

Selected specimen sent to pHtest to be tested at 2 months

Selected specimen sent to pHtest to be tested at 2 months

1) Dispersing equipment. Diameter of the rotator-stabilator aggregate 30 mm. Batch volume 3 litre.

7.1.2 Results and conclusions

The results (given in detail in Appendix 5, Tables 1, 2 and 3, rows 1101 - 1130) were asfollows. (The bleeding of all mixes was within the requirements and it is not discussedhere):

Effect of W/DM

As W/DM was increased the mix became very clearly more fluid but the shear strength

decreased as well (Figure 5a). The three highest shear strength values (with the lowestW/DM values) were higher (better) that the requirement (0.5 kPa) (Step 2 in Chapter 4)but still lower than needed for practical grouting. The large variation of mixes barelyaffected the penetration-ability measured by the filter pump. Only the very densest mixshowed a slightly lower value (Figure 5b). This was due to the obvious insensibility ofthe method when testing mixes with good penetration ability (Appendix 5 in Tables 1

8/12/2019 WR2004-45_web

37/169

UF16, SF/OPC = 0.3

0

10

20

30

40

50

0.8 1 1.2 1.4 1.6 1.8 2 2.2

Water / dry materials

Yield, Casson, Pa

Yield, Bingh am, Pa

Shear 1st point, Pa

Visc, Casson, mPas

Shear str. 6h, kPa

Shear str. x10, 6h, kPa

1 2 3 4

UF16, SF/OPC = 0.3

0

50

100

150

200

250

300

350

0.8 1 1.2 1.4 1.6 1.8 2 2.2


Filterpump

,100m

,ml

1 2 3 4

(a) (b)

Figure 5a and 5b. Effect of W/DM on early injection properties. The mix numbers are

given in the boxes.

Effect of mixing order and SP

Delayed addition of SF (mixing order 2, Table 8) made the mix clearly more fluid. Yet,such a procedure was considered too laborious and so the mixing everything togetherduring the 5 min mixing period was preferred. The effect of 1% SP addition wasnegligible (Figure 6a). The modifications did not effect on the filter pump result

probably for the reasons explained above. (The only diverging behaviour of mix 9 wasconsidered to be due to an experimental error, not seen in the parallel test mix 13.)(Figure 6b). (Appendix 5 in, Tables 1, 2 and 3, rows 1107 1115).

UF16, SF/OPC = 0.3

5

10

15

20

25

Yield, Casson, Pa

Yield, Bingham, Pa

Shear 1st point, Pa

Shear str. 6h, kPa

Visc, Casson , mPas

UF16, SF/OPC = 0.3

50

100

150

200

250

300

350

Filterpump100m

,m

l

8/12/2019 WR2004-45_web

38/169

Effect of shotcrete accelerator

Even though the shotcrete accelerator (Meyco SA 161) was added at the end of thedelay period (mixing order 4, Table 8) it totally deteriorated the penetration-ability aswell as other properties to a smaller extent. (Figures 7a and 7b). However it did notimprove the shear strength development as intended. The accelerating reaction wasobviously far too fast for grouting purposes. It did not offer a solution to the low settingrate problem. (Appendix 5 in, Tables 1, 2 and 3, rows 1118 1121).

UF16, SF/OPC = 0,3

0

2

4

6

8

10

12

14

16

18

0 0.5 1 1.5

Ac ce le re to r, w %

Yield, Casson, Pa

Yield, Bingh am, Pa

Shear 1st poin t, Pa

Shear str. 6h, kPa

Visc, Casson , mPas

UF16, SF/OPC = 0.3

0

50

100

150

200

250

300

350

0 0.5 1 1

Acc el er at or , w %

Filterpump,

100m,

ml

.5

(a) (b)

Figure 7a and 7b. Effect of shotcrete accelerator (Meyco SA 161)

Effect of fast reacting OPC

The relatively rapidly reacting OPC micro cement (Rheocem 900) was tested in order tofind a solution to the slow setting problem, Rheocem 900 was not compatible withGroutAid. The penetration-ability was poor even with high W/DM and delayed SFaddition (Figure 8). (Appendix 5, Tables 1, 2 and 3, rows 1125 1130).

Rh eocem900, SF/OPC = 0.3

200

250

300

350

100m,

ml

Mix order 1

Mix order 2

8/12/2019 WR2004-45_web

39/169

8/12/2019 WR2004-45_web

40/169

components were high alumina cement (HAC) (ground at VTT) and gypsum (G) as aslurry (a commercial product). (Chapter 5).

The basis of the system was the accelerating effect of HAC on OPC. The options ofadding and not adding gypsum to the mix were examined. Both ETTA componentswere applied as fine particles (Chapter 5). A number of different mixing orders weretested. The formation of ettringite was not verified experimentally.

The ETTA system was first developed with a constant SF/OPC ratio of 0.26, which is

relatively low, in order to observe the early OPC reactions without too much SFinterference. W/DM was 1.62. Later higher SF/OPC ratios were examined along withhigher W/DM. The results are discussed in the up-coming figures below. The mixcompositions and test results are given in detail in Appendix 5 in Tables 1, 2 and 3 asfollows:

The effect of HAC/OPC and mixing order without gypsum in the rows 1141 1182.

The effect of HAC/OPC and mixing order with gypsum (G/OPC = 0.027) in the

rows 1184 1240. The effect of SF/OPC and W/DM. HAC/OPC= 0.075, G/OPC = 0.027 in the

rows 1242 1252. The effect of SF/OPC and W/DM. HAC/OPC= 0.1, G/OPC = 0.027 in the rows

1254 1264.

The mixing and testing procedure for mixing orders 11 (without gypsum) and 21 (withgypsum) are outlined in Table 10. The basic mixing orders 11 and 21 were furthermodified as shown below (Tables 11 and 12).

8/12/2019 WR2004-45_web

41/169

8/12/2019 WR2004-45_web

42/169


The results (Appendix 5, Tables 1, 2 and 3) were as follows:

Increasing the HAC content (HAC/OPC) without gypsum accelerated the strengthdevelopment very clearly (Figures 9a and 9b). The highest values were more than 100times as high as the requirement (0.5 kPa). Unfortunately penetration-ability measured

by the filter pump decreased as well. Out of the four mixing orders the simplest one, 11(Table 10), gave the best mixes. (Rows 1141 1182).

Similar trends were seen from the results with gypsum (Figures 10a and 10b). Theeffect of mixing order was more dominant than without gypsum. Again the simplestone, 21, was the best. Other mixing orders deteriorated especially the penetration-ability. (Rows 1184 1240).

When comparing the results with the best mixing orders 11 and 21 (Table 10), thepenetration-ability of the mixes made with gypsum (with the mixing order 21) was

better (Figures 11a and 11b).Increasing the SF content (SF/OPC) while using ETTA (HAC/OPC = 0.75, G/OPC =0.027) gave also promising results (Figure 12a and 12b). The shear strength vs. W/DMwas not quite as high as with smaller SF/OPC, which was expected since the OPCcontent decreased while SF/OPC increased. (The one poor penetration-ability result is

probably due to the large W/DM and low SF/OPC of the mix. Very lean and thinmixes are unable to carry the particles through the sieve.) (Rows 1242 1252) Similar

test series was also made with higher HAC dosage (HAC/OPC = 0.1, instead of 0.075),but the results with the lower dosage were slightly better. (Rows 1254 1264).

UF16, SF/OPC = 0,26, W/DM=1,62

10

20

30

40

50

60

70

80

Shearstr.

6h

,kPa

HAC/11

HAC/14

HAC/10

HAC/12

UF16, SF/OPC= 0,26, W/DM=1,62

50

100

150

200

250

300

350

Filterpump

,1h

,ml

HAC/11

HAC/14

HAC/10

HAC/12

8/12/2019 WR2004-45_web

43/169

UF16, SF/OPC = 0,26, W/DM=1,62

0

10

20

30

40

50

60

70

80

0.00 0.05 0.10 0.15

HAC/ OPC

Shearstr.6h,kPa

HAC+G/21

HAC+G/26HAC+G/20

HAC+G/22

UF16, SF/OPC = 0,26, W/DM=1,62

0

50

100

150

200

250

300

350

0.00 0.05 0.10 0.15

HAC/ OPC

Filterpump,

1h,

ml

HAC+G/21

HAC+G/26HAC+G/20

HAC+G/22

(a) (b)

Figure 10a and 10b. Effect of HAC/OPC and mixing order with gypsum.

UF16, SF/OPC = 0 ,26, W/DM=1,62

0

10

20

30

40

50

60

70

80

0.00 0.05 0.10 0.15HAC/ OPC

Shearstr.,

6h,

kPa

HAC/11

HAC+G/21

UF16, SF/OPC = 0,26, W/DM=1,62

0

50

100

150

200

250

300

350

0.00 0.05 0.10 0.15

HAC/ OPC

Filterpump

,1h

,ml

HAC/11

HAC+G/21

(a) (b)

Figure 11a and 11b. Comparison between ETTA mixes with and without gypsum.

Mixing orders 11 and 21.

8/12/2019 WR2004-45_web

44/169

No ETTA / ETTA

HA C/OPC = 0.075,

G/OPC = 0.027

0

2

4

6

8

10

12

14

16

18

0 0.5 1 1.5 2 2.5 3 3.5

Water / dry material s

Sh

earstr,6h,kPa

ETTA , 0.3

ETTA 0.5

ETTA , 0.94

ETTA , 1.25

No ETTA , 0.3

Poly. (ETTA , 0.3)

No ETTA / ETTA

HA C/OPC 0.075, G/ OPC = 0.027

0

50

100

150

200

250

300

350

0 0.5 1 1.5 2 2.5 3 3.5

Water / dry material s

Filterpump,

1h,

ml

ETTA, 0.3

ETTA 0.5

ETTA, 0.94

ETTA, 1.25

No ETTA , 0.3

(a) (b)

Figure 12a and 12 b. Effect of increasing SF/OPC from 0.3 to 1.25 and effect of

W/DM. HAC/OPC= 0.075, G/OPC = 0.027. Numbers in the legend box denote SF/OPCcontent.

7.4 OPC - SF system: Penetration-ability and pH with ettringiteacceleration (ETTA)

7.4.1 Experiments

The OPC+SF mixes were further modified by increasing the SF content to reach therequired pH value ( 11). The SF/OPC ratios were applied 0.69 an 0.94. The mixingorder was no 21 (Table 10). No SP was used. The results are given in the up-comingTable 13. The mix compositions and test results are given in detail in Appendix 5 inTables 1, 2 and 3 as follows:

The effect of SF/OPC and W/DM with ETTA (HAC/OPC = 0.075 and 0.10,G/OPC = 0.027) in the rows 1266 1270 and

similarly without gypsum (HAC/OPC = 0.075 and 0.10, G/OPC = 0) in the rows1272 1275.


h l i bl h d h l i h i h

8/12/2019 WR2004-45_web

45/169

Table 13. Properties of OPC-SF mixes with ETTA e72 and f63 f70. The mixing order

was no 21 (Table 10). No SP was used.

Mix G/OPC

HAC/OPC

SF/OPC

W/DM

Bleed-ing, %

Shearstr, 6h,

kPa

B minm

B crit,m

ViscosityBingham,

mPas

Yieldvalue

Bingham,Pa

e72 0.027 0.075 0.69 2.0 0 6.5 53 152f63 0.027 0.075 0.69 2.5 0 3.7 44 65 49.6 20.7f64 0.027 0.075 0.94 2.9 0 3.4 44 63 40.4 16.4f65 0.027 0.1 0.69 3.1 0 1.3 43 70 30.7 12.1

f66 0.027 0.1 0.94 2.9 0 2.8 44 68 41.3 22.6

f67 0 0.075 0.69 2.5 0 4.8 44 71 43.8 13f68 0 0.075 0.94 3.0 0 3.7 43 71 40.4 11.6f69 0 0.1 0.69 3.2 0 2.2 44 83 26.3 6.5f70 0 0.1 0.94 2.9 0 4.1 41 85 36.2 12.5

7.5 OPC - SF system: Effect of W/DM with ettringite acceleration (ETTA)

7.5.1 Experiments

When the compressive strength requirement of 4 MPa was set, the effect W/DM oncompressive strength was preliminary tested with specimen originally cast but not usedin the pH experiments (cast in plastic tubes). Testing age was about 3 months. SF/OPCwas 0.3 and no ETTA applied.

After the preliminary compressive strength tests all properties of the f63 and f64 type ofmixes were tested against the W/DM. The results are given in the up-coming Table 14.The mix compositions and test results are given in detail in Appendix 5 in Tables 1, 2and 3 as follows:

f64 type of mix (SF/OPC = 0.94 with ETTA). Mixes u1, u2 and u3 in the rows1293 - 1295.

f63 type of mix (SF/OPC = 0.69 with ETTA). Mixes u4, u5 and u6 in the rows

1297 - 1299. f63 type of mix (SF/OPC = 0.69 with ETTA) repeated with a new UF16 batch.Mixes u8, u9 and u10 in the rows 1302 - 1304.

f63 type of mix without ETTA. Mixes u11, u12 and u13 in the rows 1306 -1308.

8/12/2019 WR2004-45_web

46/169

Table 14. Mix compositions u1 u6 and u8 u13.

Mix Mix type OPC type SF type OPC/DM SF/DM Gypsum/OPC HAC/OPC SF/OPC Superpl./DM Water/DM

SF/OPC = 0,94, with ETTA, W/DM = 2 - 2,5 4

u1 f64 UF16 GroutAid 0.49 0.46 0.027 0.075 0.94 0.00 2.01

u2 f64 UF16 GroutAid 0.49 0.46 0.027 0.075 0.94 0.00 2.51u3 f64 UF16 GroutAid 0.49 0.46 0.027 0.075 0.94 0.00 4.01SF/OPC = 0,69, with ETTA, W/DM = 2 - 2,5 4

u4 f63 UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 2.00

u5 f63 UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 2.51u6 f63 UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 4.01SF/OPC = 0,69, with ETTA, W/DM = 2 - 2,5 - 4, new UF16 batch

u8 f63 UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 2.00

u9 f63 UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 2.51

u10f63

UF16 GroutAid 0.56 0.38 0.027 0.075 0.69 0.00 4.01

SF/OPC = 0,69, without ETTA, W/DM = 2 - 1,6 - 1,2, new UF16 batchu11 f63 no

ETTA

UF16 GroutAid 0.59 0.41 0.00 0.000 0.69 0.00 2.00

u12 f63 noETTA

UF16 GroutAid 0.59 0.41 0.00 0.000 0.69 0.00 1.60

u13 f63 noETTA

UF16 GroutAid 0.59 0.41 0.00 0.000 0.69 0.00 1.20


Properties of the f63 type mix vs. W/DM are given in Figure 13 and for the f64 type ofmix in Figure 14. The bleeding of all mixes was 0%. The bleeding values are not shownin the Figures.

8/12/2019 WR2004-45_web

47/169

Strength

0

2

4

6

8

10

12

14

16

18

20

1 1.5 2 2.5 3 3.5 4 4.5


Compressivestr28d

an

d91d

,MPa

Est, prev project. Experim. without SF, 1)

UF16+SF no ETTA, 3 mont hs, 2)

UF16+SF+ETTA (f63 type) 4)


UF16+SF no ETTA (f63 t ype) 5)

91 d

28 d

0

2

4

6

8

10

12

14

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Shearstr,

6h,

kPa

UF16+SF+ETTA (f 63 type) 3)



UF16+SF no ETTA (f63 type) 5)

Expon . (UF16+SF+ETTA (f 63 type) 4))

Expon. (UF16+SF no ETTA (f63 type) 5))

(a) (b)

0

50

100

150

200

250

300

350

400

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Penetrab

ility,

B

critandB

min,m





B minB cr i t

0

20

40

60

80

10 0

12 0

14 0

16 0

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Viscos

ity(mPas)and

yieldvalue(Pa)




Visc.

Yield

(c) (d)

Figure 13a, 13b, 13c and 13d. Properties of the f63 type (SF/OPC = 0.69) mix vs.W/DM.1) Curve fitted to results without SF. (Kronlf 2003)2) Tested with pH cylinders at the age of about 3 months. Experiments 1,2,3 and 4. SF/OPC 0.3, smallerthat in f63 and f64.3) First UF16 batch Former experiments Tested for pH (Chapter 7 4)

8/12/2019 WR2004-45_web

48/169

Strength

0

2

4

6

8

10

12

14

16

18

20

1 1.5 2 2.5 3 3.5 4 4.5


Compressivestr,

28dand

91d,

MPa

Est , prev project . Experim. wi th out SF, 1)

UF16+SF no ETTA , 3 months , 2)


91 d

28 d

0

2

4

6

8

10

12

14

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Shearstr

,6h

,kPa

UF16+SF+ETTA (f64 t ype) 3)

UF16+SF+ETTA (f64 t ype) 4)

Expon. (UF16+SF+ETTA (f64 type) 4))

(a) (b)

0

50

10 0

15 0

20 0

25 0

30 0

35 0

40 0

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Pene

trability,

B

minandB

crit,m



b min

b cr i t

0

20

40

60

80

100

120

140

160

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5


Visc

osity(mPas)and

y

ieldvalue(Pa)


Visc.

Yield

(c) (d)

Figure 14a, 14b, 14c and 14d. Properties of the f64 type mix (SF/OPC = 0.94) vs.

W/DM.1) Curve fitted to results without SF (Kronlf 2003).2) Tested with pH cylinders at the age of about 3 months. Experiments 1,2,3 and 4. SF/OPC 0.3, smaller

that in f63 and f64.3) First UF16 batch. Former experiments. Tested for pH (Chapter 7.4).4) First UF16 batch. Final experiments.

8/12/2019 WR2004-45_web

49/169

Note: The excellent penetration-ability results given in Table 13 could not be repeatedin the tests reported in Figures 13 and 14. The tests were run with two batches of UF16

to rule out the effect of possible ageing of the cement. Therefore the ones given in Table13 should be over looked and not be treated as the true penetration-ability behaviour ofthose mixes. The reason for the disagreement was not found.

7.6 OPC - SF system: Low alkali white cement (WCE)

7.6.1 Experiments

The very low alkali white OPC (WCE) was ground at VTT to micro cement finenessand tested for all the injection properties. SF was used in the form of GroutAid as in the

previous mixes (Chapter 5). The experiments and results are listed in Appendix 5,Tables 1, 2 and 3, Rows 1315 1319. The aim was to find answers to the followingquestions:

How are the pH and leaching affected by the low alkali content of the cement? Is it possible to achieve penetration-ability using WCE? OPCs are known to

behave very differently in the presence of SF. Does ETTA work with WCE?


The results are given in Table 15. The mixes met the requirements except that given tothe yield value of the rheology measurement. Yet, the yield values were lower (better)than in the case of similar UF16 mixes (Table 13) whereas the setting (shear strength at

6 h) was slower which was not desired. More ETTA would be needed on siteconditions. Mixes w1 and w2 were chosen to be tested for pH. W1 did not but w2 didreach required value ( 11). With both SF/OPC ratios the pH of the WCE mix washigher than that of the comparable UF16 mix, which led to an unexpected conclusionabout cement alkali content: The low alkali cement (WCE) did not produce a lower pHmix compared to the higher alkali cement (UF16) but, in fact, reverse. (pH will beexamined closer in Chapter 9.1).

Table 15. Properties of mixes w1 w4. The mixing order was no 21 (Table 10). No SP

was used.

Mix G/OPC

HAC/OPC

SF/OPC

W/DM

Bleed-ing, %

Shearstr, 6h,

B minm

B crit,m

ViscosityBingham,

Yieldvalue

8/12/2019 WR2004-45_web

50/169

7.7 OPC - SF system: Effect of premixing SF with cement

7.7.1 Experiments

The possibility of pre-mixing the ETTA components and dry un-densified SF withcement was examined. The motivations for the examinations were as follows:

A single pre-mixed dry product would be practical to handle on site. A dry SF product might react slower and produce less gel, which could

improve penetration-ability. A low alkali dry product, un-densified, SF type 983 is available (Chapter 5).

Testing its effect on pH and leaching was considered, but the tests wereexcluded eventually.

The results are given in the up-coming Table 17. The mix compositions and test resultsare given in detail in Appendix 5 in Tables 1, 2 and 3 in rows 1321 - 1333. The lowalkali white OPC (WCE) and gypsum were ground at CT-Group to micro cementfineness. Low alkali SF type 983 was used in the experiments (Chapter 5, Materials).The mixing procedure was the mixing order 211 in Table 16. The aim of the

experiments was to find answers to the following questions: Can the ETTA components be pre-mixed as dry products prior to adding water.

This necessitated the substitution of gypsum slurry with dry gypsum. How does dry un-densified SF affect the injection properties compared to

GroutAid when mixing dry un-densified SF while making the grout? How does the premixing of dry un-densified SF affect the injection properties

compared to mixing dry un-densified SF while making the grout?

Table 16. Mixing order and testing procedure 211 for pre-mixed dry products.

Mixing order and testing procedure 211

with pre-mixed products

1 Water + all dry materials2 Start timing for all tests2 Mix 2 min 12 000 rpm 1)3 Superpl. if any4 GroutAid

5 Mix 3 min 12 000 rpm 1)6 Start bleeding and shear strength tests7 Move the mix to agitator8 Slow agitation for 1h9 Test rheology and penetration-ability at 1h10 C f H d

8/12/2019 WR2004-45_web

51/169

The agglomeration is demonstrated by the particle size distributions WCE, G3 and G4in Figure 3a where G3 is coarser than G4.

When considering the possibility of mixing dry SF in the mixer instead of GroutAid, theresult was not quite so clear (mixes w5 and w6). For the behaviour of both mixes w5and w6 was somewhat obscure, the question was further re-examined with UF16 bycomparing the previously presented traditional GroutAid mix u2 with u7. The onlydifference was the use of dry un-densified SF type 920 (the raw material of GroutAid)instead of GroutAid. The SF was added directly to the mixer as GroutAid (Experimentu7. Row 1333). The result showed quite clearly that GroutAid cannot be substituted bydry un-densified SF without deteriorating the penetration-ability.

Table17. Properties of mixes w5 w and u2 - u7. HAC/OPC = 0.075,

G/OPC = 0.027, SF//OPC = 0.94, W/D = 2.5. No SP was used.

Mix Premixedmaterials

Premixeddry product

SF-type Bleed-ing, %

Shearstr, 6h,

kPa

B minm

B crit,m

ViscosityBingham,

mPas

Yield valueBingham,

Pa

w5 WCE, G,HAC G4 GroutAid 0 5.32 89 269 48.6 17

w6WCE, G,HAC

G4un-densifiedtype 983

0 3,37 145 297 21 4.8

w7WCE, G,HAC, SF

G3

un-densifiedtype 983.

pre-mixed

0 0,9 165 410 16.9 3.6

u2 None GroutAid 0 5.83 46 101 75.8 25.2

u7 Noneun-densifiedtype 920

0 6.47 163 324 49.4 21.9

7.8 Slag system: First experiments

7.8.1 Experiments

A number of preliminary tests were run to the study the basic injection properties ofslag.

The slag used was ground by CT-Group in a pilot plant jet mill. The batch wasid ifi d SL10/1 i Ch 5 Th OPC Fi i h id h d i d

8/12/2019 WR2004-45_web

52/169

The effect of SP, OPC/SL = 0.2, no SF, W/DM = 0.82.In the rows 1357 - 1358

The effect of mixing order with SP, OPC/SL = 0.2, SF/DM = 0.04,W/DM = 0.83. In the rows 1361 - 1362

The effect of mixing order without SP, OPC/SL = 0.2. SF/DM = 0.04,W/DM = 0.83. In the rows 1365 1366

The effect of mixing order without SP, OPC/SL = 0.2, SF/DM = 0.14,W/DM = 1.25, In the rows 1369 1370

The effect of W/DM, no SP, OPC/SL = 0.2. SF/DM = 0.14.

In the rows 1373 1374 The effect of OPC type, no SP, OPC/SL = 0.2. SF/DM = 0.04,

W/DM = 0.83. In the rows 1377 1379

Table 18. Mixing orders and testing procedures 1 and 2 for Slag mixes.

Mixing order and testing

procedure 1 with Slag system

Mixing order and testing

procedure 2 with Slag system

1 Water + slag + OPC+

gypsum slurry

Water + slag+ OPC+

gypsum slurry2 Start timing for all tests Start timing for all tests3 Mix 2 min 12 000 rpm 1) Mix 2 min 12 000 rpm 1)4 Superpl. if any Superpl. if any5 GroutAid Mix 3 min 12 000 rpm 1)6 Mix 3 min 12 000 rpm 1) Do not test yet7 Start bleeding and

shear str. testsMove the mix to agitator

8 Move the mix to agitator Slow agitation for 1h9 Slow agitation for 1h GroutAid. Mix 0.5 min10 Test rheology and penetration-

ability at 1hTest rheology and penetration-ability at 1 h. Start bleeding andshear strength tests at 1 h.

11 Cast for pH and comp. str. Cast for pH and comp. str.12 Test bleeding at 2 h and shear

strength at 6 hTest bleeding at 3 h and shearstrength at 6 h

13 Test compressive strength at28/91 d

Test compressive strength at 28/91d

14 Selected specimen sent to pHtest to be tested at 2 months

Selected specimen sent to pH testto be tested at 2 months

1) Dispersing equipment. Diameter of the rotator-stabilator aggregate 30 mm. Batch volume 3 litre.


8/12/2019 WR2004-45_web

53/169

8/12/2019 WR2004-45_web

54/169

OPS/SL = 0,2;

SF/OPC = 0, 0,25, 1

0

10

20

30

40

50

49 7 24 25 8 40 46 34 46 47

Visc, Bin gham, mPas

Yield, Bingh am, Pa

Visc, Casson, mPas

Yield, Casson, Pa

Shear str . 6h, kPa

Sil/cem 1

W/DM 1,25 - 1,18

MIX. ORD. 1-1No SP - No SP

Sil/cem 1

W/DM = 1,25

MIX. ORD 1-2No SP - No SP

Sil/cem 0,25

W/DM 0,83

MIX. ORD. 1-2No SP - No SP

Sil/cem 0,25

W/DM 0,83

MIX.ORD. 1-2SP - SP

Sil/cem 0

W/DM 0,83

MIX. ORD. 1-1

SP- NO SP

(a)

OPC/SL = 0,2;

SF/OPC = 0, 0 ,25, 1

0

50

100

150

200

250

300

350

49 7 24 25 8 40 46 34 46 47

Filterpump,

80and100

m,

ml

Filter pu mp 80 m, ml

Filter pu mp 100 m, ml

(b)

Figure 16a and 16b. Effect of mix design parameters on injection properties, when

OPC/SL ratio is 0 2 Mix numbers are given on the x axis

8/12/2019 WR2004-45_web

55/169

OPC//SL = 0.05

SF/OPC = 0, 1 ,4

0

10

20

30

40

50

48 27 29 31 28 30 44 36 44 45

Visc, Bingham, mPas

Yield, Bingham, Pa

Visc, Casson, mPas

Yield, Casson, Pa

Yield str. 6h, kPa

Sil/cem =1

W/DM 0,91

MIX.ORD 1-2

SP - SP

Sil/cem= 1

W/DM 0,91

MIX.ORD: 1-2

No SP - No SP

Sil/cem = 4

W/DM 1,36-1,28

MIX.ORD. 1-1

No SP - No SP

Sil/cem = 0

W/DM 0,91

MIX.ORD. 1-1

SP - No SP

Sil/cem = 4

W/DM 1,36

MIX.ORD. 1-2

No SP - No SP

(a)

OPC/SL = 0,05;

SF/OPC = 0, 1, 4

0

50

100

150

200

250

300

350

48 27 29 31 28 30 44 36 44 45

Filterpump,

80and100m,

ml

Fi l ter pump 80 u m, ml

Fi l ter pump 100 u m, ml

(b)

8/12/2019 WR2004-45_web

56/169

Table 19. Properties of slag mixes 8, 47, 27, 28 and 44. The mixing order was no 1

(Table 18). No SP was used.

Mix OPC/SL

SF/OPC

W/DM

Bleed-ing, %

Filterpump100 m

Filterpump80 m

Shearstr, 6h,

kPa

B minm

B crit,m

ViscosityBingham,

mPas

Yieldvalue

Bingham,Pa

8 0.20 0.25 0.83 0 290 - 0.8 62 224 48.6 11.347 0.20 1.00 1.18 0 310 240 1.3 60 132 40.3 12.627 0.05 0.00 0.90 4 310 210 1.3 63 147 14.8 4.2628 0.05 1.00 0.91 0.5 290 200 0.6 63 159 32.6 9.18

44 0.05 4.00 1.36 0 310 250 1.3 61 136 32 7.89

7.9 Slag and Super Sulphate Cement systems (SSC): Activation of slag

7.9.1 Experiments

Once the requirement for the compressive strength was set (Step 10 in Chapter 4) thework was focused on the activation of slag, because the strength gain and the low pH of

the end product are conflicting requirements. The former experiments on slag showedthat besides for compressive strength the mixes needed to be improved also for

penetration-ability, setting and pH. These needs are somewhat conflicting: Lowering pHcalled for more silica, which in turn was to inactivate slag even more. SF would alsoincrease water demand and thus retard setting. In the first experiments slag wasactivated with OPC (Chapter 7.8). In the up-coming experiments also super sulphatecement system (SSC) was used. SSC denotes a binder system which is based on slag

and activated with gypsum (G) and OPC. The aim of the slag experiments was to findanswers to the following questions: What would be the safe SF dosage from the pH point of view? Can slag be activated with OPC only while using dosages that would produce

the low pH required? How can ETTA be applied to the slag system in order to accelerate setting? Is the SSC needed to increase the strength development? What is the gypsum content needed in the super sulphate system? Is NaOH activation needed to start the reactions? What is the effect of W/DM on the properties?

The SF content was increased to the SF/SL ratio of 0.5 to ensure the total SiO2contentf 50 5 54 5 % th h t th i (Fi 1) Gi SF/DM ti th l

8/12/2019 WR2004-45_web

57/169

8/12/2019 WR2004-45_web

58/169

Table 20. Experiments with slag SL10/3. Compositions of mixes S16 S47. The mixing

order was no 21 (Table 21). No SP was used.

MixOPCtype

SFtype

OPC/DM

SF/DM

SL/DM

Gypsum/DM

SF/SL

OPC/SL

Gypsum/SL

Superpl./DM

Water/DM

S16 RC10 GroutAid 0.032 0.32 0.65 0.000 0.50 0.050 0.00 0.00 1.41

S17 RC10 GroutAid 0.032 0.32 0.65 0.000 0.50 0.050 0.00 0.00 1.57

S18 RC10 GroutAid 0.032 0.32 0.65 0.000 0.50 0.050 0.00 0.00 2.01

S19 RC10 GroutAid 0.063 0.31 0.63 0.000 0.50 0.100 0.00 0.00 1.40

S20c RC10 GroutAid 0.063 0.31 0.63 0.000 0.50 0.100 0.00 0.00 1.58S21 RC10 GroutAid 0.063 0.31 0.63 0.000 0.50 0.100 0.00 0.00 2.00

S22 RC10 GroutAid 0.120 0.29 0.59 0.000 0.50 0.204 0.00 0.00 1.40

S23 RC10 GroutAid 0.120 0.29 0.59 0.000 0.50 0.204 0.00 0.00 1.57

S24 RC10 GroutAid 0.120 0.29 0.59 0.000 0.50 0.204 0.00 0.00 2.01

S25 RC10 GroutAid 0.029 0.29 0.59 0.093 0.50 0.050 0.16 0.00 1.39

S26 RC10 GroutAid 0.029 0.29 0.59 0.093 0.50 0.050 0.16 0.00 1.57S27 RC10 GroutAid 0.029 0.29 0.59 0.093 0.50 0.050 0.16 0.00 2.00

S28 RC10 GroutAid 0.057 0.28 0.57 0.091 0.50 0.100 0.16 0.00 1.40

S29 RC10 GroutAid 0.057 0.28 0.57 0.091 0.50 0.100 0.16 0.00 1.57

S30 RC10 GroutAid 0.057 0.28 0.57 0.091 0.50 0.100 0.16 0.00 2.00

S31 RC10 GroutAid 0.028 0.28 0.560.130

0.50 0.050 0.23 0.00 1.40

S32 RC10 GroutAid 0.028 0.28 0.56 0.130 0.50 0.050 0.23 0.00 1.57

S33 RC10 GroutAid 0.028 0.28 0.56 0.130 0.50 0.050 0.23 0.00 2.00

S34 RC10 GroutAid 0.055 0.27 0.55 0.126 0.50 0.100 0.23 0.00 1.40

S35 RC10 GroutAid 0.055 0.27 0.55 0.126 0.50 0.100 0.23 0.00 1.57

S36 RC10 GroutAid 0.055 0.27 0.55 0.126 0.50 0.100 0.23 0.00 2.00

S37 RC10 GroutAid 0.027 0.27 0.54 0.166 0.50 0.050 0.31 0.00 1.40S38 RC10 GroutAid 0.027 0.27 0.54 0.166 0.50 0.050 0.31 0.00 1.57

S39 RC10 GroutAid 0.027 0.27 0.54 0.166 0.50 0.050 0.31 0.00 2.00

S40 RC10 GroutAid 0.052 0.26 0.52 0.161 0.50 0.100 0.31 0.00 1.40

8/12/2019 WR2004-45_web

59/169

8/12/2019 WR2004-45_web

60/169

Effect of ETTA (HAC/G = 3),

OPC/SL = 0.05, SF/DM = 0.3, W/DM = 1.6

Mix: S2, S3, S4

0

2

4

6

8

10

12

0 0.02 0.04 0.06 0.08

HAC/DM

Sh

ear.str.andcompr.str.

Shear str . 6 h,kPa

Compr.str. 28 d,

MPa

Figure 18a. Effect of ETTA components (HAC and gypsum) on setting and compressive

strength of the Slag system. Slag batch was SL15.Effect of HAC vs . OPC,

SF/DM = 0.3, W/DM = 1.6

Mix: S4, S5, S6

0

2

4

6

8

10

12

0 0.02 0.04 0.06 0.08

HAC/DM

Shearstr.and

compr.str.

Shear str. 6 h,kPa

Compr.str. 28 d,MPa

OPC

No OPC

Figure 18b. Effect of OPC compared to no OPC on setting and compressive strength ofSlag system. Slag batch was SL15.

8/12/2019 WR2004-45_web

61/169

Effect of G/DM,

OPC/SL = 0.05, SF/DM = 0.3, HAC/DM = 0.035. W/DM = 1.6

Mix: S3, S7, S8

0

2

4

6

8

10

12

0 0.02 0.04 0.06 0.08

Gypsum/slag

She

arstr.andcompr.str. Shear str. 6 h,

kPa

Compr.str. 28 d,

MP a

Figure 18c. Effect of increasing gypsum content in ETTA system on setting and

compressive strength of Slag system. Slag batch was SL15.7.9.3 Results and conclusions on activation without ETTA - slag batch SL15

The mixes without ETTA (e.g. without the presence of HAC) all showed strengthdevelopment. The tests were made with the same slag batch SL15 as the previous testsin Figures 18a 18b. The results in Figures 19a 19d are as follows:

The effect of gypsum to strength development was positive. Yet, there was anunexplained minimum at the 3 % gypsum dosage. (Figure 19a).

The same results of Figure 19a are reproduced in Figure 19b along with resultson accelerating slag activation with NaOH. The dosages used showed noacceleration, which ended the interest to the originally questionable alkaliactivation from the desired low-pH point of view.

The setting of all mixes was relatively slow. Gypsum seemed to depress it evenmore.

The Bcritvalues were generally slightly over the requirement (too high) while the Bminvalues were within the requirements. Mix S14 was selected to closer examination(circled in Figures 19a 19d) as the new finer slag batch SL10/3 arrived. The identicalmix with the finer slag batch SL10/3 was named S26 and is also circled in the upcomingfigures in Chapter 7.9.4. Mix S14 was sent to pH tests, which yielded low enough

l i d h

8/12/2019 WR2004-45_web

62/169

Effect of G/DM,

OPC/SL = 0.05, SF/DM = 0.3, HAC/DM = 0, W/DM = 1.6

Mix: S0, S1, S5, S9, S14

0

2

4

6

8

10

12

0 0.05 0.1 0.15 0.2

Gypsum/slag

Compr.str.,MPa

Mix 12

without OPC

Figure 19a. Effect of gypsum activation on compressive strength determined at the age

of 28 d. Slag batch was SL15. The circled mix S14 was re-examined (Chapter 7.9.4).

Effect of G/DM and NaOH,


Mix: S0, S1, S5, S9, S11 , S12, S14, S15,

0

2

4

6

8

10

12

0 0.05 0.1 0.15 0.2

Gypsum/slag

Compr.str

.,MPa

OPC/SL = 0, NaOH/SL = 0

OPC/SL = 0.05, NaOH/SL = 0

OPC/SL = 0.05, NaOH/SL = 0.01 - 0.02

Figure 19b. Effect of gypsum and alkali activation on compressive strength determined

at the age of 28 d. Slag batch was SL15. Gypsum activation results are the same as in

Figure 19a Notation as in Figure 19a

8/12/2019 WR2004-45_web

63/169


OPC/SL = 0.05, SF/DM = 0.3, HAC/DM = 0, W /DM = 1.6

Mix: S0, S1, S5, S9, S11 , S12, S14, S15,

0

2

4

6

8

10

12

0 0.05 0.1 0.15 0.2

Gypsum/slag

Sh

ear.s

tr.

6h,

kPa



OPC/SL = 0.05, NaOH/SL = 0.01 - 0.02

Figure 19c. Effect of gypsum and alkali activation on shear strength determined at theage of 6 h. Slag batch was SL1. Notation as in Figure 19a.



Mix: S0, S1, S5, S9, S11 , S12, S14, S15,

0

50

100

150

200

250

0 0.05 0.1 0.15 0.2

Gypsum/slag

B

crit,um



OPC/SL = 0.05, NaOH/SL = 0.01 - 0.02

Figure 19d Effect of gypsum and alkali activation on penetration ability (B values

The compressive strength was doubled compared to the mixes without gypsum

8/12/2019 WR2004-45_web

64/169

The compressive strength was doubled compared to the mixes without gypsum(Figure 20a).

On the contrary, setting was delayed by gypsum (Figure 20b). This wasobserved also in the previous results with the coarser slag batch SL15 (Figure19c).

Increasing the gypsum content over the G/SL ratio of 0.16 did not produce anybenefits (Figures 20a 20d).

Effect of W/DM and G/SL, OPC/SL=0.05

Mix: S16, S17, S18, S44, S45, S25 S26, S27,S32

0

2

4

6

8

10

12

14

16

18

20

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

Com

pressivestr.

28dand

91d,

MPa

G/SL=0

G/SL=0.16

G/SL=0.23

91 d

28 d


Mix: S16, S17, S18, S44, S45, S25, S26, S27, S32

0

1

2

3

4

5

6

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

S

hear.s

tr.

6h,

kPa

G/SL=0

G/SL=0.16

G/SL=0.23

(a) (b)

Effec t of W/DM an d G/SL, OPC/SL=0.05

Mix : S16, S17, S18, S44, S45, S25, S26, S27,

S32

0

20

40

60

80

100

120

140

160

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

B

critandB

min,m

G/SL=0

G/SL=0.16

G/SL=0.23

B min

B cr i t

Effect of W/DM an d G/SL, OPC/SL=0.05

Mix: S16, S17, S18, S44, S45, S25, S26, S27,

S32

0

20

40

60

80

100

120

140

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

Vis

c.(mPas)andYield

(Pa),Bingham

G/SL=0

G/SL=0.16

G/SL=0.23

Visc .

Yield

of view, for the strength was expected to increase Only a slight acceleration of setting

8/12/2019 WR2004-45_web

65/169

of view, for the strength was expected to increase. Only a slight acceleration of setting(shear strength) was observed, but it was not significant. Yet, even the slight

acceleration may turn out to be significant if the mix was to be used in a lowertemperature.


Mix: S43, S19, S20c, S21, S46, S47, S29, S35,

0

2

4

6

8

10

12

14

16

18

20

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

Compressivestr.

28d

and91

d,

MPa

G/SL=0

G/SL=0.16

G/SL=0.23

91 d

28 d


Mix: S43, S19, S20, S21, S46, S47, S29, S35,

0

1

2

3

4

5

6

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

Shear.s

tr.

6h,k

Pa

G/SL=0

G/SL=0.16

G/SL=0.23

(a) (b)


Mix: S43, S19, S20c, S21, S46, S47, S29, S35,

0

20

40

60

80

100

120

140

160

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry material s

B

critandB

min

,m

G/SL=0

G/SL=0.16

G/SL=0.23

B min

B cr i t


Mix: S43, S19, S20c, S21, S46, S47, S29, S35,

0

20

40

60

80

100

120

140

1.1 1.3 1.5 1.7 1.9 2.1

Water/dry materials

Visc.(mPas)andYield(Pa),

Bingham

G/SL=0

G/SL=0.16

G/SL=0.23

Yield

Visc.

(c) (d)

Figure 21a, 21b, 21 and, 21d. Effect W/DM on the compressive strength and injection

properties of OPC activated slag and OPC-gypsum activated slag (super sulphate

cement) when the OPC/SL ratio was 0 1 Slag batch was SL10/3

all mixes in this work. In those cases the readings were not recorded. Only four readings

8/12/2019 WR2004-45_web

66/169

g y gwere received and they are listed in Table 22. Three had the highest W/DM value of 2;

only mix S47 had a lover value of 1.4. The Marsh cone values were not furtherexamined.

Table 22. Properties of the mixes that passed the Marsh cone test.

Mix W/DM OPC/SL G/SL Mash cone time, s

S21 2.0 0.1 0 100

S27 2.0 0.05 0.16 71

S35 2.0 0.05 0.31 205

S47 1.4 0.1 0.16 172

7.9.5 Oversized particles

The slag batch SL10/2 was also tested, but the penetration-ability results werefluctuating in a random manner due to a material failure: There were oversized particles,

which were able to block the sieves of the penetrabilitymeter very efficiently. Thefraction of large particles (> 63 m) was 0.3%. The weight of ground slag per 1000 ml(the volume that passes the sieves in the penetrabilitymeter) is approximately 330 g.Therefore the weight of over sized particles per 1000 ml would be 0.1 g. Since the sievediameter is only 30 mm, 0,1 g is far enough to block the sieve. Problems would

probably occur in field conditions as well.

The observation above is significant, while planning the quality control of ground

grouting products: The content of oversized particle should be kept at a low level bysetting a requirement and verifying it regularly.

7.10 LAC system: Low Alkali Cement

7.10.1 Experiments

The Nuclear Waste Management Organisation of Japan (NUMO) delivered twospecimens of low alkali cement (LAC) to test the possibility

WR2004-45_web

Documents