thermal analysis in eco-concrete research

Thermal analysis in

eco-concrete research

Els Bruneel, Mieke De Schepper, Ruben Snellings, Joris Schoon, Isabel Van Driessche, Nele De Belie, Klaartje De Buysser SCRiPTS, Department of Inorganic and Physical Chemistry, UGent Magnel Laboratory for Concrete Research, UGent Sagrex N.V., Brussel

Magnel Laboratory for

Concrete Research

1

Concrete production: 10 billion ton concrete each year, 3 billion ton cement

Consumption of natural raw materials » 42% of produced aggregates is for concrete production » 1 kg cement = 1.6 kg raw materials

Production of waste » 850 million tons construction and demolition waste whereof 40-67% is concrete

Emission of CO2 » 1.6 billion tons each year which is around 5-8% of the total CO2 emissions

Introduction

~ 40% energy required for cement production ( at > 1400 °C) ~ 60% calcination of limestone (to produce cement) CaCO3 → CaO + CO2

2

www.wbcsd.org

Evolution

3

What is concrete Overview Portland Clinker

hydration

1450°C

Formation of alite, belite, aluminate, ferrite

Cement hydrates = creation of adhesive bonds

CO2

Fine and coarse granulates >60 %

H2O

CSH gel, ettringite, portlandite

Ca-source Si-source Al-source Fe-source

4

What can be done ? Portland Clinker

Use of the best technology Thermal efficiency of kiln and cooler systems

Use of alternative fuels (i.e. solvent waste)

Use of demolition waste as aggregate

Replacement of clinker and cement by industrial byproducts and waste : supplementary cementitious materials (SCMs)

fly ash, slag, silica (glass powder) Clincker production using industrial byproducts and waste. slag, fly ash, silica (glass powder)

/ alternative binders

H2O hydratation

1450°C

5

H2O hydratation

Granulates Sand >60%

1450°C

What can be done ?

6

Complete recyclable concrete Cradle to cradle Concept M. De Schepper

Design of a concrete which

could be used for making a new

clincker

What can be done ?

7

Bekaert - meeting 21/08/09

Necessary condition: Chemical composition CRC

= Chemical composition

cement raw meal

Completely Recyclable Concrete

Most important ratios: LSF: lime saturation factor Ca versus (Si, Al, Fe) SM: silica modus Si versus Al, Fe AM: alumina modus Al versus Fe

8

To make a good clincker: composition should lie between certain limits

Two raw materials:

a CRC and a CEMENT PASTE (CP)

CRC CP

CaO 64.99 62.04

SiO2 20.97 18.99

Al2O3 6.10 5.97

Fe2O3 2.50 4.23

MgO 2.53 0.96

SO3 1.02 3.18

K2O - 0.66

Na2O - 0.46

LSF 0.96 0.98

SM 2.44 1.86

AM 2.44 1.41

2. CRC: hydrated + crushed CEMI 52.5N + Components: Limestone aggregates (Gaurain, Soignies) Limestone filler Diorite (Lessines) Fly ash Copper slag

1. Cement paste: hydrated + crushed CEMI 52.5N

Good for clincker production 9


Question: How does this mixture

behave during reclinkering?

1450°C


Cement paste

CRC

10

Bekaert - meeting 21/08/09 11


Liquid phase

SiO2 Al2O3

CO2

Other oxides

C2S= beliet

CaO

CaCO3

C3S = aliet

Fe2O3 C4AF C3A

200 400 600 800 1000 1200 1400 12


Clinckering Reactions, Ruben Snellings 1. Thermogravimetry /Differential thermal analysis 2. XRD-HTXRD

13


Clinckering Reactions TGA-DTA, heating and fast cooling of a CRC

0 20 40 60 80 100 120 140 160 Time /min

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

DTA /(µV/mg)

65

70

75

80

85

90

95

100

TG /%

0

200

400

600

800

1000

1200

1400

Temp. /°C

[3]

exo

Dry air heating @10°C/min Cooling @25°C/min

Base line is not flat

- Geometry of furnace Heat transfer towards cup

Compare with blanc

- There is a sample * emissivity

* Thermal conduction

Decarb.

Dehydr.

DT

14

0 20 40 60 80 100 120 140 160Time /min

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

DTA /(µV/mg)

65

70

75

80

85

90

95

100

TG /%

0

200

400

600

800

1000

1200

1400

Temp. /°C

[3] CRC4M_Air.ngb-ss3

TG

DTA

Temp.

[4] CRC4M_Air2.ngb-ss3

TG

DTA

Temp.[3]

[3]

[3]

[4]

[4]

[4]

exo

DTA curve derived from a small mass sample as the baseline for a large mass sample using the same material. diminishes - ``apparatus effect'' asymmetric heat transfer problem attributed to the and -``sample influence'‘ improving the linearity between the DTA curve and enthalpy change.

low mass 25 mg (green) high mass 50 mg (blue)

Clinckering Reactions TGA-DTA, mass-difference baseline method Yang & Roy, Thermochimica acta, 1999

15

65

70

75

80

85

90

95

100

TG /%

0 20 40 60 80 100 120 140 160Time /min

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

DTA /(µV/mg)

0

200

400

600

800

1000

1200

1400

Temp. /°C

[1] subtr_CRC4M_Air.ngb-ss340_CRC4M_Air2.ngb-ss340.ngb-ms3

TG

Temp.

[2] subtr_CRC4M_Air.ngb-ss3230_CRC4M_Air2.ngb-ss3230.ngb-ms3

DTA

Temp.

[1]

[1]

[2]

[2]

exo

Higher temp: possible explanations: melt formation and crystallisation, polymorphic transformations of alite and belite Much better resolution on

endothermal and exothermal events

Clinckering Reactions TGA-DTA, heating and fast cooling of a CRC

exo

Melt formation α’’L to α’H

C2S transition

aluminate =C3A formation

16

cement CSH, CAH dehydration

Ca(OH)2 dehydroxylation

CRC : endoth. decarbonation

Melt formation

α’’L to α’H C2S transition

C3A formation from Ca-aluminates

C3A, C4AF crystallisation C2S polymorp. transitions

Clinckering Reactions TGA-DTA

DTA curves for CRC and Cement Paste (CP) during heating (B) and cooling (C)

Lower T of melt formation and cooling exotherms Indicative for better burnability of CP

17


In situ XRD measurements 25 – 1050 °C: • Calcite & dolomite decomposition • Decomposition of quartz • Formation of intermediate phases

CaCO3 CaO

CaMg (CO3)2

Ca-(alumino-)silicates

• Gehlenite (C2AS) • Yeelimite (C4A3S) • Belite (C2S) • Mayenite (C12A7) • Lime :CaO • Aluminate: celiet (C3A) • Ferrite (C4AF)

Clinckering Reactions HTXRD

18


Ex situ XRD measurements, 1050 – 1550 °C, dwell 1h • C2S + CaO → C3S • Decomposition of intermediate

phases to form main clinker phases End product

• Alite (C3S = Ca3SiO5 ) • Belite (C2S = Ca2 SiO4 ) • Aluminate (C3A = Ca3Al2O6 ) • Ferrite ( C4AF= Ca2 (Al,Fe) 2O5 )

ZnO intern standard for Rietveld analysis

Clinckering Reactions XRD, after calcination

19


0

10

20

30

40

50

60

70

80

90

100

20 100 250 400 500 600 700 800 900 1000 1050

T (°C)

Other

C2S

Gehlenite

C4AF_Jupe

C3A cubic

C12A7

Lime

Dolomite

Quartz low

Calcite

0

10

20

30

40

50

60

70

80

90

100

1050 1150 1250 1350 1450 1550

T (°C)

Other

Periclase

Alkali Sulphates

C3S_M3_DLT

beta-C2S

Gehlenite

Yeelimite

C4AF_Jupe

C3A_Na_cubic

C12A7

Lime

Quartz low

Rietveld analysis gives a quantitative view on the occurring reactions: Using : internal standard Using data from TGA: mass loss ,melt formation

1050 – 1150 °C: • Extensive crystallisation of C2S • C2AS, C4AŠ (yeelimite), C12A7, C • Quartz decomposition 1250 – 1450 °C • Intermediate Ca-aluminates form C3A • C2S + C → C3S (gradual increase) 1450 -1550 °C • C4AF and C3S are formed at the expense of C3A

HTXRD Furnace, 1h + XRD

Clinckering Reactions Rietveld

= aliet

20

Why?: more S => more / melt at low temperature. : faster

S in Belit: less alite

Comparison with CRC: • in cement:

• Lower aliet / beliet due to solid solution in C2S • Higher C4AF = ferrite (higher Fe2O3 content)

Clinckering Reactions CRC versus Cement paste

21

0

10

20

30

40

50

60

70

80

90

100

1050 1150 1250 1350 1450

We

igh

t p

erc

en

tage

[w

t%]

Burning temperature [°C]

Other

Magnetite

Quartz

GehlenitePericlase

Alkali sulphates

Mayenite

Ye'elimite

Aluminate

Ferrite

AliteBeliteLime

Lime

Belite

Alite

Gehlenite

0

10

20

30

40

50

60

70

80

90

100

1050 1150 1250 1350 1450

We

igh

t p

erc

en

tage

[w

t%]

Burning temperature [°C]

Lime

Belite

Alite

CRC Cement Paste


Clinckering Reactions Microscopy, CRC versus cement paste, burned at 1450°C

At 1450°C more open pores More alite

• more and earlier melt formation • earlier formation of well-formed

alite/belite crystals • better distribution of the crystals in

the melt • lower porosity cement paste clinker

has a better burnability

CRC Cement Paste Alite

Belite mainly as inclusions

in alite

Melt (mainly

grey) rich in aluminate

Melt (mainly bright/white) rich

in ferrite

Belite as individual crystals

Open porosity of about 24v%

Closed porosity of about 11v%

Traditional clinker phases are formed!

22


S: acts as flux -Lowering melitng temp. (seen in TGA) -Reduces melt viscosity -Stabilizes belite => less alite

General conclusions

In the cement paste clinker

• more and earlier melt formation

• earlier formation of well-formed alite/belite crystals

• better distribution of the crystals in the melt

• lower porosity

cement paste clinker has a better burnability

This effect is probably caused by its higher sulfur content

CRC CP

CaO 64.99 62.04

SiO2 20.97 18.99

Al2O3 6.10 5.97

Fe2O3 2.50 4.23

MgO 2.53 0.96

SO3 1.02 3.18

23

hydratation

Granulates Sand 1450°C


Cement hydrates = creation of adhesive bonds

24

Next step : hydration


Question: How does the cement

behave during hydration?

1450°C

clincker CRC + 40% H2O hydratation

+ CaSO4

+ 40% H2O hydration

Commercial Cement CEM I 52.5 N

cement

25

Alite + water: 2Ca3OSiO4 + 6H2O → 3CaO.2SiO2.3H2O + 3Ca(OH)2 (= fast) -121.1 KJ/mol Belite + water: 2Ca2SiO4 + 4H2O → 3CaO.2SiO2.3H2O + Ca(OH)2 (= slow) -45.1 KJ/mol Aluminate + water + gypsum: -263.0 KJ/mol Ca3(AlO3)2 + 3CaSO4 + 32H2O → Ca6(AlO3)2(SO4)3.32H2O Ferrite +water + gypsum: 2Ca2AlFeO5 + CaSO4 + 16H2O → Ca4(AlO3)2(SO4).12H2O + Ca(OH)2 + 2Fe(OH)3

Cement hydration

Reaction rate: formation Ca(OH)2 bound H2O heat of hydration 26


Hydration of alite, rapid crystallization of CH and C-S-H

Dormant period

1° period Ca 2+

OH-

Isothermal calorimetry

Early stage of hydration Apparatus: TAM air, TA instruments : 2 ampules, temp constant

Sample= Cement+ H2O

reference= sand

Seebeck heat flow sensor 27

Isothermal calorimetry - Shows us the reaction speed in the first couple of hours, days….

Hydration of regenerated cement is slower, but after 7 days the cumulative hydration heat is approaching the one of CEM I 52.5 N

Size of grains is of major importance..

= 7d

More aluminate

28


TGA, DTA

Cement + H2O -> react for 1h, 3 h , 6h, 9h, 1d, 2d, 3d, 7d, 28 d and stop the reaction by - Freeze drying - Solvent Exchange: soaking in dry solvent to replace capillary water

ethanol / isopropanol + drying.

Measure of hydration: - % bound water - formation of Ca(OH)2

29

Bekaert - meeting 21/08/09 60,00

65,00

70,00

75,00

80,00

85,00

90,00

95,00

100,00

0 200 400 600 800 1000

Mass [

m%

]

Temperature [°C]

28d

Ca(OH)2

Y –x : Bound water

CaCO3

free water

x

400-520 °C : decomposition of portlandite Ca(OH)2

Ca(OH)2 → CaO + H2O

TGA, DTA @ 10°C/min, N2

600-780 °C : decomposition of calcite (CaCO3)

CaCO3 → CaO + CO2

= carbonated Ca(OH)2 => correction

30

Bekaert - meeting 21/08/09 60

65

70

75

80

85

90

95

100

0 200 400 600 800 1000

Mass [

m%

]

Temperature [°C]

1h

3h

6h

9h

1d

2d

3d

7d

28d

Ca(OH)2 CaCO3 Bound water

TGA, DTA

31

TGA – Bound water content

0

2

4

6

8

10

12

14

16

18

20

0,01 0,1 1 10 100

Mass [

m%

]

Time [days]

CEM I 52.5 [SEM]

CRC2e [SEM]

32


TGA – Portlandite

0

5

10

15

20

25

0,01 0,1 1 10

Ma

ss [m

%]

Time [days]

CEM I 52.5 [SEM]

CEM I 52.5 N [SEM] corrected for CO2

0

5

10

15

20

25

0,01 0,1 1 10M

ass [m

%]

Time [days]

CRC2e [SEM]

CRC2e [SEM] corrected for CO2

CEM I 52.5 N CRC2e

33


CRC2e

Hydration of clinker minerals …

… Resulting in hydration products

XRD, Rietveld

0

5

10

15

20

25

30

0,01 0,1 1 10M

ass

[m%

]

Age [days]

Other_SEM [XRD]

Portlandite_SEM [XRD]

AFm_SEM [XRD]

Ettringite_SEM [XRD]

0

10

20

30

40

50

60

0,01 0,1 1 10

Mas

s [m

%]

Age [days]

Alite_SEM [XRD]

Belite_SEM [XRD]

Ferrite_SEM [XRD]

Aluminate_SEM [XRD]

Monosulfate aluminate

34


Portlandite: TGA vs. XRD

0

5

10

15

20

25

30

35

40

0,01 0,1 1 10

Mas

s [m

%]

Age [days]

CEM I 52.5 N [XRD]

CEM I 52.5 N [TGA]

CEM I 52.5 N [TGA not corr.]

CEM I 52.5 N CRC2e

0

5

10

15

20

25

30

35

40

0,01 0,1 1 10

Mas

s [m

%]

Age [days]

CRC2e_SEM [XRD]

CRC2e_SEM [TGA]

CRC2e_SEM [TGA not corr.] 35


Future: design of the next CRC

Use of Waste fibrecement Joris Schoon, Sagrex

Fibrecement: a non carbonate CaO source less energy consumption to heat up less C02 emission = recycling: fibrecement is available as waste

Fibrecement: composite material consisiting of portland cement, inert and or reactive mineral filler and a mixtrure of sevreal typrs of organic fibers.

Most important: LSF: lime saturation factor Ca/ (Si, Al, Fe) SM: silica modus Si / Al, Fe AM: alumina modus Al / Fe Ca : 60-65%

Again: To make a good clincker: composition should lie between certain limits

37


1. Thermogravimetry /Differential thermal analysis

- Mass loss => - amount CO3 (600-800°C)

- DTA => energy needed to heat till 1450°C - endotherm: decomp of CO3 (CaCO3 -> CaO +CO2 : 1782kJ/kg) - Endotherm: dehydration

- TGA of fillers: determine decomposition temp. and products.

Use of Waste fibrecement

38


-0,4

-0,3

-0,2

-0,1

0

0,1

0,2

0,3

0,4

0,5

-50 450 950 1450

DT

A (m V

/mg

)

temp (°C)

classiccompositionfibrecement

PVA decomp CO3 decomp

Emission of CO2 » 1.6 billion tons each year

~ 40% energy required for cement production ( at > 1400 °C) ~ 60% calcination of limestone (to produce cement) CaCO3 → CaO + CO2

Energy to decompose fibres is small enough

Use of Waste fibrecement

39


“Clinckering Reactions during firing of recyclable concrete “ R. Snellings J Am Cer Soc , 2012 “The hydration of cement regenerated from completely recyclable concrete” M. De Schepper, J Am Cer Soc, 2012 “ Waste Fibrecement: An interesting alternative raw material for a sustainable Portland clinker production, J. Schoon , Construction and building materials, 36 (2012)

Conclusions

Use of fibrecement Could lead to Re-Use of fibrecement waste Energy gain Low CO2 emisions

Information for new design

40

Use of TGA , DTA, (HT- XRD) and calorimetry Completely recyclable concrete Characterise end products and study the reactions

During clinckering And hydration Identify end products and intermediates Identify the difference in burnability Follow reactions rate

industrial tests are the next step

thermal analysis in eco-concrete research

Documents