XRD in the Cement Industry

Rainer Schmidt

October 2013 1
Topics

Quantitative Phase Analysis in the Cement Industry

• Why X-ray Powder Diffraction?

• XRD Application Areas
• Selected examples
• Instrumentation / Software / Application Support

XRD - the tool to bring together process and quality

October 2013 2
Topics

Quantitative Phase Analysis in the Cement Industry

• Why X-ray Powder Diffraction?

• XRD Application Areas
• Selected examples
• Instrumentation / Software / Application Support

October 2013 3
Why X-ray Powder Diffraction?

• The quality of the product and therefore of the production process is

mainly governed by its phase composition (mineralogy) and NOT by
its elemental composition
• The classical methods like the Bogue analysis cannot describe the
reality or are too slow for process control like optical microscopy
• XRD enables quality and process control of raw material, hotmeal,
clinker and cement, based on the knowledge of the true phase
composition, that is information not accessible by other analysis
techniques

Benefits are manifold, and depend on the size of plant,

type of products, quality of local raw materials, local
energy costs, and local environmental regulations

October 2013 4
Bogue analysis

• Bogue calculates a potential clinker phase composition based on

assumptions, not describing the real phase formations. It can
therefore only be a first estimation
• It does not take into account solid solution effects and can never
consider polymorphs
• During clinker cooling no thermal equilibrium is reached, which is
the prerequisite of such a normative phase calculation
• Neither the absolute values of Bogue can be right, nor relative
changes can be monitored correctly, in particular if secondary fuels
and secondary raw materials are used

October 2013 5
H.F.W. Taylor: Cement Cemistry (1990)
Chapter 3.1.3 The Bogue Calculation

October 2013 6
Optical Microscopy

Optical microscopy is a perfect tool for studying the microstructure of

the clinker, but shows severe limitations for quantitative analysis:
• Poor discrimination of interstitial phases (C3A / C4AF)
• Poor or no detection of polymorphs (C3S, C2S, C3A)
• Poor detection of minor phases
• Labour intensive, operator dependent
• Slow, not suited for process control

October 2013 7
Why X-ray Powder Diffraction?

X-Ray Powder Diffraction (XRD) and Rietveld analysis

• Enables direct and accurate quantitative phase analysis
• Is a standardless method - no calibration involved
• Is independent of equipment and operator
• Can be fully automated and thus operated free of human interaction

Fully automated XRD with Rietveld analysis provides

• Quantitative phase compositions
• Detection and quantification of polymorphs (C3S, C2S, C3A)
• Lattice parameters / cell volumes
• Crystallite sizes

October 2013 8
Topics

Quantitative Phase Analysis in the Cement Industry

• Why X-ray Powder Diffraction?

• XRD Application Areas

• Selected examples
• Instrumentation / Software / Application Support

October 2013 9
 Cement Production Process
XRD Applications Areas

1. Raw Materials
Cyclones 2. Cyclones - blockage
prevention by hotmeal
Quarry Cement phase analysis
Mill
3. Kiln – process control
based on real clinker
phase composition
4. C
ƒement - direct analysis
of Gypsum dehydration
Kiln during cement milling
5. Offline Quality Control
Raw Mill
and R&D

October 2013 10
XRD Application Areas
1. Raw Materials

Quality control of a wide range of raw materials

• Limestone
• Early warning in case of detrimental phases, e.g. Pyrite
• Gypsum
• Anhydrite, semihydrate, dihydrate, calcite, dolomite, quartz
• Blast Furnace Slag, Flyash, Puzzolan
• Ca(OH)2 for SO3 absorption
• High lime and calcite contents decrease absorption quality
• Iron sulfates for Cr(VI+) reduction
• XRD as a backup for XRF
• phase composition → calculation of chemical composition
• chemical composition → calculation of LSF, SR, AR

October 2013 11
XRD Application Areas
2. Hot Meal /Bypass

Minimization of energy losses and cloggings

• Degree of decarbonation
• Clinker dust determination
• Measure of energy losses due to dust cycles and inappropriate kiln
operation
• Formation of phases causing build-ups / clogging
 Improved clogging prevention

Hotmeal phase analysis does not only allow to stabilize the preheater
operation at high SO3 and Chlorine levels, but also to reduce the
bypass rate

October 2013 12
XRD Application Areas
3. Clinker

Process control based on the actual phase composition

• Check and control of kiln operation / quality control of clinker
• Freelime analysis always included
• Polymorphism C3S / C2S / C3A
• Detection of minor phases (Alkalisulphates, Quartz, Pericalse, …)
• Operation with secondary fuels / secondary raw materials
 Spectacular failures of Bogue method!
• Setting time / strength development / degree of sulphatisation
• Clinker grindability
• XRD as a backup for XRF

October 2013 13
XRD Application Areas
4. Cement

Control of cement mill operation

• Check and control of mill operation
• Dehydration of gypsum
• Cold start of the mill can be controlled
• Quality control of the finished cement
• Fillers and additions such as Limestone, Blast Furnace Slag, Flyash and
Puzzolan can be analyzed directly
 Meet requirements of norms (CEM)
• Quality management (archive XRD data as a fingerprint)

October 2013 14
XRD Application Areas
5. Offline Quality Control / R&D

Miscellaneous offline applications

• Characterization of competition products
• Analysis of materials forming build-ups and cloggings
• Customer complaints
• Cement hydration studies
• Research and development, ambient and non-ambient studies
• Characterization of any raw material deliveries - at the gate

October 2013 15
Phase Analysis in the Production Process
Benefits and Payback

Examples:

• XRD versus Bogue method

• Polymorphism C3A and Sulphatisation
• Polymorphism Alite (C3S) and Strength Development
• Hotmeal phase analysis and clogging/blockage prevention
• Seamless integrated analysis of amorphous phases

October 2013 16
 Clinker
XRD versus Bogue Method

85

Rietveld
Alite content after Rietveld and Bogue

80

75

70

65 Bogue
Deuna plant case study, 2003;
Dyckerhoff (Buzzi Unicem) - Polysius - Bruker AXS
60
92 93 94 95 96 97 98 99 100 101 102
Lime saturation factor (LSF)

October 2013 17
Clinker
XRD versus Bogue Method

 Deuna plant case study, 2003;

Dyckerhoff (Buzzi Unicem) - Polysius - Bruker AXS

• There is no linear relation between the Alite abundance and the LSF,
as indicated by Bogue
• Depending on the individual plant scenario, there is a maximum
Alite level, which cannot be exceeded by increasing the LSF
• Generally the clinker phase composition is more complex and more
variable as indicated by Bogue

A too high LSF results in waste of

energy and raw materials!

October 2013 18
Clinker
Polymorhism C3A and sulphatisation

C3A in wt.%

Degree of sulphatisation

October 2013 19
Clinker
Polymorhism C3A and sulphatisation

• High amounts of Sodium and Potassium lead to the formation of

orthorhombic instead of cubic C3A
• Both modification show differences in reactivity and hydration
• The ratio C3Acubic/C3Aorthorhombic influences the water consumption of the
cement
• Higher water consumption by C3Aorthorhombic
• If Na and K are fixed in Alkali-Sulphates these elements are not
available for the incorporation in C3A
• By changing the degree of sulphatisation the ratio of
C3Acubic/C3Aorthorhombic can be controlled

October 2013 20
Alite Polymorphism

• In cement clinkers two monoclinic

polymorphs of the main phase Alite
occur (M1 and M3)
• Both polymorphs have different
hydraulic properties, like strength
development
• We are now able to distinguish
these polymorphs
• It allows to understand and predict
the early strength development

The missing link for quality and Maki & Goto, CCR, 1982:
process control! Dependence of the polymorphic modification
of alite in production clinkers on the MgO
and SO3 contents of the clinker.

October 2013 21
Alite Polymorphism

Hydration Study
• Cement paste put in a Dome like air-tight Specimen Holder
• D4 ENDEAVOR with LYNXEYE detector
• 138 measurements every ten minutes; in total 24 hours

October 2013 22
Hydration Study 24 hours

time

Let us inspect the Alite peak group at

about 52°2Θ, which can be used to
identify the presence of the polymorphs.

October 2013 23
 Hydration Study 24 hours

Alite Peak group at 51.5 to 52°2Theta Taylor: Cement Chemistry

time

M3 M1+M3 M1

At the beginning of the hydration process a mixture of M1 & M3 can

be found; cf. Taylor: Cement Chemistry (right figure).
After 24 hours only the M3 polymorph is remaining in this sample.

October 2013 24
Clinker reactivity test
Data: HOLCIM plant Sehnde (Germany)

Correlation 2 days compressive strength

80

70
y = 0.163x + 62.878
60
R2 = 0.0231
50
Weight %

C3S
40
C3S M1
30

20 y = 0.7483x + 9.6999
R2 = 0.5761
10

0
15 20 25 30 35 40

2d compressive strength [MPa]

No correlation for total C3S. Only the M1 polymorph shows correlation to

two days compressive strength!

October 2013 25
Step by step application of
phase analysis for process
optimization.

Urs Häseli
Holcim Switzerland , Plant Siggenthal
 Kiln system of Siggenthal

 4 stage preheater without precalciner

 High thermal load (8 MW/m2)
 Frequent blockages (26 in 2007)
27
Holcim Switzerland © 2010
 Observations from hotmeal phases

 Ca-langbeinite >
7%
 Spurrite > 7%
 Sylvite > 7%

High concentrations of different hotmeal phases create different buildups

28
Holcim Switzerland © 2010
Relation between hotmeal phases
and buildups

Ca-Langbeinite formation
Spurrite formation
Material direction
Gas direction

Normal
800° 1000° Balance of
C C Ca-Langbeinite and Spurrite

Auto
Sampler

The reason for the buildups are not the measured phases
in the hotmeal, but the phases in unfavorable locations.
October 2013 29
Relation between hotmeal phases and buildups

Ca-Langbeinite formation
Spurrite formation
Material direction
Gas direction

Hot
800° 1000° High concentration of
C C Ca-Langbeinite

Auto
Sampler

The reason for the buildups are not the measured phases
in the hotmeal, but the phases in unfavorable locations.
October 2013 30
Relation between hotmeal phases and buildups

Ca-Langbeinite formation
Spurrite formation
Material direction
Gas direction

Cold
800° 1000° High concentration of
C C Spurrite

Auto
Sampler

The reason for the buildups are not the measured phases
in the hotmeal, but the phases in unfavorable locations.
October 2013 31
 Optimized fuel mix
Sylvite limit

Ca-
langbeinite
limit

The balanced ratio of chlorine and sulfur lead to higher possible

concentrations of these elements in hotmeal without buildups
Source: Holcim Group Support
32
Holcim Switzerland © 2010
 Reduction of cyclone blockages due to phase information

Clinker loss
Cyclone blockages

The kiln performance improved step by step

33
Holcim Switzerland © 2010
Hotmeal phase analysis
Lab Report 72

October 2013 34
Seamless integrated analysis of
amorphous phases by PONKCS

Examples amorphous phases

• blast furnace slag

• coal fly ash
• different puzzolan

October 2013 35
Seamless integrated analysis of
amorphous phases by PONKCS

Example: Blast Furnace Slag

October 2013 36
Seamless integrated analysis of
amorphous phases by PONKCS

Example: Coal Fly-Ash

hkl_Phase_Flyash 67.79 %
6,500
Lime 0.67 %
6,000 Periclase 0.36 %
5,500 Quartz 13.11 %
Anhydrite 0.34 %
5,000
Calcite 0.23 %
4,500 Mullite 3:2 14.45 %
4,000 Magnetite 1.32 %
Hematite 1.31 %
3,500
Thenardite 0.34 %
3,000 Rutile 0.08 %
Counts

2,500
2,000
1,500
1,000
500
0
-500
-1,000
-1,500
-2,000
10 15 20 25 30 35 40 45 50 55 60 65
2Th Degrees

October 2013 37
Seamless integrated analysis of
amorphous phases by PONKCS

Example: Distinction between Coal Fly-Ash and BFSlag

20000
Flyash BFSlag
Lin (Counts)

10000

0
10 20 30 40 50 60

2-Theta - Scale

October 2013 38
Blast Furnace Slag Cement analysis
Lab Report 65

October 2013 39
Quantification of Puzzolan in Cements
by XRD

• Various types of rocks are called Puzzolan

• Reactivity mainly depends on
• Amorphous phases
• Clays
• Zeolites
• Alkalis
• ……..
• XRD analysis might require
• stable background calculation
• hkl_Phase model for the amorphous phase
• hkl_Phase model to describe anisotropic size and strain broadening of
clays (stacking faults)

October 2013 40
Quantification of Puzzolan in Cements
by XRD

Example:
Indocement Cirebon

homogeneous quarry
mainly crystalline phases
Puzzolan is the sum of the
listed phases

October 2013 41
Quantification of Puzzolan in Cements
by XRD

Example: Indocement Citeureup

• different types of Puzzolan used for the production of the same Cement type
• no other method existing for quality control
• hkl phase model for amorphous phase and Kaolinite
• can only be solved by stable background calculation
• accuracy: +/- 2 weight %

October 2013 42
Quantification of Puzzolan in Cements
by XRD

Example: Cementos Yura

• homogeneous quarry
• amorphous phase & crys-
talline phases
• hkl phase model for
amorphous phase
• can only be solved by
stable background
calculation

October 2013 43
Quantification of Puzzolan in Cements
by XRD

Example: Cementos Yura (Peru)

Laboratory mixtures to proof

concept

Phases considered to describe

the Puzzolan

• Amorphous by hkl_phase
• Quartz
• Plagioclase
• Muscovite/Illite
• Hematite
• Cristobalite

October 2013 44
Topics

Quantitative Phase Analysis in the Cement Industry

• Why X-ray Powder Diffraction?

• XRD Application Areas
• Selected examples
• Instrumentation / Software / Application Support

October 2013 45
Instrumentation

• Bruker offers XRD instrumentation for analysing the entire process

mineralogy starting with the preheater processes, to the clinker
burning and the cement milling

D2 PHASER D8 ADVANCE
D4 ENDEAVOR

October 2013 46
The D4 ENDEAVOR
High Sample Throughput

• Various sample magazines

(66, 72, 120 samples)
• Offline or fully automated
operation
• Connections via robot or
conveyor belt
• Automation interface fully
supporting priority levels
• Manual and automated
operation simultaneously!

October 2013 47
Automation via Robot or Conveyor

• Robot interface • Conveyor belt

October 2013 48
The LYNXEYE Detector

• Active area: 14.4x16 mm

• Fluorescence suppression
eliminating the need for
secondary monochromators
• More than 150 times faster /
more intensity than a
scintillation counter
• The ideal detector for all
applications in the cement
industry

D2 PHASER / D4 ENDEAVOR / D8 ADVANCE with LYNXEYE

The world's fastest X-ray diffractometers

October 2013 49
Instrument performance
D4 ENDEAVOR equipped with LYNXEYE

Note the overall intensity

and the background at
low angles!

Figure:
Cement Clinker (scan time: 4 min 37 sec)
Intensity in counts versus °2 Theta

October 2013 50
D4 ENDEAVOR & LYNXEYE & TOPAS
Repeatability of analysis (values in wt.%)

the same sample was measured and analysed ten times

(scan time: 4 min 37 sec; details in Lab Report XRD 59)

October 2013 51
 Software: TOPAS Rietveld

TOPAS defines a new generation of Rietveld software

• Fundamental Parameters Approach (FPA)
• Numerically stable and extremely fast
• Typical calculation time is a few seconds per sample
• Easy to use and fully automatic analysis without user input
• Large convergence radius
• Seamless integrated analysis of amorphous phases (e.g. BFSlag
or Fly-ash) by PONKCS

October 2013 52
Software: TOPAS Rietveld

Due to its outstanding analytical capabilities, TOPAS has become

the scientific and industry standard for quantitative phase
analysis in the world of cement

- 0 + - 0 + - 0 + - 0 + - 0 +
Emission Profile Target Slit Width Horizontial Axial
tan(θ) Divergence Divergence
α cot(θ) SL cot(θ)
2 2

W G
- 0 +
Final Profile

- 0 + - 0 + - 0 +

Y(2θ) = (W * G) * S
Absorption Crystallite Size Strain
sin(2θ) / AB 1 / cos(2θ) CS STR tan(θ)

October 2013 53
Turn-key XRD Method Package Cement

Application Support for the complete

process mineralogy

Highlights:

• Quantification of the Alite

polymorphs M1 & M3
• Quantification of amorphous phases
by hkl_Phases (PONKCS method)
• Methods are factory assembled and
ready for use at customer site
• Three days of onsite training are
included

October 2013 54
Turn-key XRD Method Package Cement

The package covers the installation of ten methods:

• Raw Materials: Limestone, Gypsum, Iron Sulfates,

Blast Furnace Slag (BFS), Coal Flyash

• Hotmeal/Bypass

• Portland Cement Clinker

• Portland Cement (CEM I)

• Limestone Cement
(CEM II/A-L, CEM II/B-L, CEM II/A-LL, CEM II/B-LL)

• Blast Furnace Slag Cement

(CEM II/A-S, CEM II/B-S, CEM III/A, CEM III/B, CEM III/C)

• Instrument Check (alignment and tube aging)

October 2013 55
©October
6. Copyright
Mai 2015
2013Bruker Corporation. All rights reserved 56
56