CEMENT GRINDING

ADDITIVES DIVISION

CEMENT GRINDING ADDITIVES DIVISION

CEMENT GRINDING
ADDITIVES DIVISION

HEAD OFFICE
Mapei SpA
Via Cafiero, 22 - 20158 Milan
Tel. +39 02 37673.1
Fax +39 02 37673.214
Internet: www.mapei.com
E-mail: [email protected]
C.P. MK 661130 - (GB) 06/09
 INDEX

t
THE WORLD OF MAPEI 2
t

LIQUID ADMIXTURE DIVISION 6

t

CEMENT GRINDING ADDITIVES DIVISION 8

t

TECHNICAL BACKGROUND OF MAPEI GRINDING AIDS 10

t

A BRIEF OUTLINE OF AN ANALYSIS

OF THE SEPARATION PROCESS 22
t

PRODUCTS 37

MA.G.A./C Technical Data Sheet 39

Industrial Trial with MA.G.A./C1 41
Industrial Trial with MA.G.A./C2 51

MA.G.A./M Technical Data Sheet 61

MA.P.E./S Technical Data Sheet 63

Industrial Trial with MA.P.E./S 65

MA.P.E./W Technical Data Sheet 77

Industrial Trial with MA.P.E./W1 79

MA.P.E./A Technical Data Sheet 89

MA.P.E./Cr 05 Technical Data Sheet 91

Scientific Background 94
Industrial Trial with MA.P.E./Cr 05 100
The World of Mapei

Founded in Milan in 1937, Mapei is

today’s world leader in the
production of adhesives and
chemical products for building.

Starting in the 1960’s Mapei put its

strategy of internationalization into
action in order to have maximum
proximity to the needs of local
markets and reduction of shipping
costs to a minimum.

The Group now counts 57

subsidiaries with 54 production
facilities in operation in the 5
continents in 24 different countries.

Furthermore, Mapei has developed

a sales and technical service
network with offices all over the
world and offers an efficient

Technical Assistance Service that is

much appreciated by architects,
engineers, contractors and owners.

Group Headquarters
Mapei S.p.A.
viale Jenner, 4
Milan - Italy
 TO BE OUR CUSTOMERS’ BEST SUPPLIER
We offer our services as a business partner and
we are highly committed to provide our customers
solutions with a high added value.

TO BUILD A PARTNERSHIP WITH OUR 3

STRATEGIC SUPPLIERS
We are committed to ensuring that our strategic
suppliers turn meeting our needs into an opportunity
to jointly design new products and solutions which
can also cater for the latest market requirements.

TO ALWAYS KEEP ONE STEP AHEAD

We try to anticipate designers’ requirements and
interpret business and building-site needs.

TO INNOVATE CONSTANTLY
Every year we strategically invest more than 5%
of our overall turnover in Research & Development.

TO BUILD A WINNING AND COMMITTED TEAM

Our most precious resource is the value of our
team: we work together passionately and grow
professionally thanks to constant training.

TO BE DETERMINED TO ACHIEVE EXCELLENCE

THE 10 PILLARS
We strive to achieve ambitious goals so that we
are market leaders in our chosen sectors.

OF OUR SUCCESS
TO ALWAYS TELL THE TRUTH
A high-profile and transparent communication
strategy allows us to convey and share our values
with as many people as possible.

TO MAINTAIN SOLID FINANCIAL FOUNDATIONS

So that we can invest in the technology
and products of the future.

TO IMPROVE ALL OUR PROCEDURES BY MEANS

OF QUALITY CONTROLA
All our products and services conform to the
highest ISO 9001 certified standards.

TO BE AT THE CUTTING-EDGE IN TERMS OF

ENVIRONMENTAL SAFETY AND HEALTH
We attach great importance to the environmental
sustainability of our products, the eco-friendly
nature of our procedures and the safety of our
customers, workmates and entire community.
 Mapei has always devoted a lot
of effort to research. 5% of turnover
are allocated to investments in this
field of activity. 12% of Mapei
employees work in research.

Research and Development

Mapei has always placed great and also co-ordinates the work of the
emphasis on research. So much so, in other seven laboratories. In addition, it
fact, that 5% of its turnover is invested acts as a central analytical laboratory for
in Research & Development and 70% of the whole group. Mapei recruits more
its R&D efforts are directed to develop new employees in research than in any
eco-sustainable and environment other area, with priority given to
friendly products which meet LEED graduates and other qualified students.
requirements. Mapei boasts eight main The laboratories are equipped with
research centres: three in Italy (Milan, avant-garde equipment and work in
Villadossola and Treviso), one in Canada close contact both with each other and
(Laval), one in the United States (Deerfield with universities and other scientific and
Beach), one in France (Toulouse), one in industrial research institutions. They also
Norway (Sagstua) and one in Germany provide back-up technical assistance to
Our commitment for the environment (Wiesbaden). Research staff make up help solve customers’ most complex
More than 150 MAPEI products, approximately 12% of the company’s problems. There are further quality
featuring the “Green innovation” mark,
help to contribute valuable points
total workforce. The Milan Research control laboratories in all of the Group’s
toward LEED-certified projects. Centre is the biggest, in terms of staff, 54 factories.
THE WORLD OF MAPEI

2008. ROBBIANO
DI MEDIGLIA (MILAN)
In the picture, the main
manufacturing plant of the
54 facilities of the Mapei
Group. In 2008 the
expansion project of
Robbiano di Mediglia plant
is still ongoing with the new
wall finishing production line.
Liquid Admixtured Division

Bridge over
the River Danube,
Hungary
 7

Integrated solutions for the cement-concrete sector, for lower

consumption of non-renewable raw materials, to reduce CO2 emissions
and increase the service life of structures.

CEMENT GRINDING CONCRETE

ADDITIVES ADMIXTURES

more more
energy efficiency durability

less less
clinker and CO² non-renewable raw material

HPSS* UNDERGROUND
SYSTEM TECHNOLOGY TEAM

more
more durability, safety and energy
respect for the environment efficiency
and health
less
less non-renewable raw materials
waste material

* HPSS System High Performance Solidification

System for treating contaminated ground and sediments
Cement Grinding Additives Division
t

t
CEMENT GRINDING R&D ECO-SUSTAINABILITY
ADDITIVES DIVISION By investing over 5% of its Mapei grinding admixtures
Founded in 2000, D.A.M. turnover and 12% of its Human form a system of innovative
(Divisione Additivi di Resources in Research and solutions for cement works;
Macinazione) has grown an Development, the Mapei Group they allow a reduction in clinker
astonishing 30% every year in has become market leader in while offering the same
terms of turnover and volume, terms of innovation. The mechanical performance of
thanks to innovative and high- dedicated D.A.M. scientists cement, thus guaranteeing a
quality products combined with at Mapei’s Research Centres reduction of 5-10% in CO 2
technical support and not only develop new raw emissions and a saving in non-
dedicated Research and materials and grinding aid renewable raw materials.
Development. components, but are also
t

Today, supported by the active in customer support. In TAG TEAM

group’s structure and expertise, fact, Mapei’s state of the art (Technical Assistance
D.A.M. is supplying all major laboratories allow D.A.M. to Group)
cement groups worldwide, perform specific and in-depth A team of experienced process
offering new technologies and clinker and cement analysis engineers from the cement
local technical assistance. in order to optimise the industry joined D.A.M. in order
By combining high quality raw use of Grinding Aids and to provide specific technical
materials, fully computer-based to offer customized solutions assistance to D.A.M. customers.
production facilities and for cement performance By performing complete plant
specific expertise in terms of enhancement and production audits and by analysing the
product chemistry, industrial improvement. grinding circuit’s performance,
employment and grinding plant they are able to assist
technology, D.A.M. is able to D.A.M. customers with the
guarantee high levels of implementation of Grinding Aids
customer assistance and and to optimise the grinding
product quality. process in all its aspects.
 9

Our Products

Product Typical Production Strength Workability* Air Cr(VI) Typical CO2

Description
Group Dosage Increase* Increase* Entrainment* reduction* Application Reduction

Highly concentrated,
high performance
All
MA.G.A./C grinding aids, suitable 200 - 400 g/t t
t
t tt t
t
t
Cements
for grinding
of all cements.

Highly concentrated,
high performance
Minerals
grinding
MA.G.A./M 300 - 600 g/t tt t and Raw t

aids, particularly
Materials
suitable for grinding
of minerals.

Grinding aids, strength

improvers, specifically
formulated for
Blended
MA.P.E./S grinding of blended 1000 - 2000 g/t tt t
t
t t
t
t
Cements
cements (pozzolanic,
blast-furnace slag
and fly- ash cements).

Grinding aids, strengths

and workability (flow)
improvers, specifically
Blended
MA.P.E./W formulated for grinding 1000 - 2000 g/t t
t
t
t
t
t t

Cements
of blended cements
(pozzolanic, blast-furnace
slag and fly- ash cements).

Additives formulated
for grinding Masonry
MA.P.E./A 400 - 800 g/t t
t
t
t
of masonry cements Cements
(Air Entrainment).

Specific additives All

MA.P.E./Cr 50 g/t ppm t
t
t
for Cr(VI) reduction. Cements

* at typical Dosage

t
Possible t
t
Recommended t
t
t
Highly Recommended

Please contact us dircetly for specific product selection, assistance and technical documentation.
Technical
background of Mapei Grinding Aids

1. THE AGGLOMERATION PHENOMENA

During the grinding process, the increase rate in the specific surface proportionally decreases
with the fineness increase. Rittinger’s law demonstrated that there is a direct proportionality
within the grinding time and specific surface till a precise fineness (depending on the material and
the grinding system). Beyond this fineness the real curve diverges from the theoric one, and there
is no more direct proportionality between grinding time (energy) and produced specific surface.
In practice, it is not possible to exceed some fineness values, not even by extending the grinding
time. This is principally caused by the particles agglomeration that drastically reduces the pro-
cess performance. The cement particles agglomeration acts on the grinding and on the mill lining
as an abrasion resistant film, but also as fine particles, already grinded, agglomerated by electro-
static forces and local conditions of pressure and temperature. It is easy to understand how the
film and particles agglomerates can reduce the mill balls effect, by absorbing bumps and disper-
sing the energy needed for the particles comminution.

The agglomeration phenomena entity depends on:

• material type (chemical composition, crystalline structure);
• Grinding fineness;
• mill type (ball mill, vertical mill etc.);
• grinding system: open or closed circuit;
• balls and lining conditions;
• temperature, humidity, ventilation etc. inside the mill.

2. MECHANISM OF ACTION OF THE CEMENT ADDITIVES

Grinding aids have been widely used for more than 50 years. It is also well known that their main
aim is to prevent cement particle agglomeration during the milling process. As a consequence they
reduce mill retention time and improve separation efficiency, which decrease energy consumption
of the plant while maintaining constant the quality and quantity of the cement produced.
Cement additives improve mechanical strengths by producing a narrower cement particle size
distribution, which at the same time is shifted toward smaller diameters (Figure 1).

Figure 1 Particle Size Analysis

Ø (MA.GA./C 100) = 28,70 µm Ø (blank) = 31,50 µm

 11

While all that we have said up to now is evidence of cement additive use which has allowed their
diffusion, their real action mechanism in the hydration of cement is not completely understood
yet. What is not yet clear are the actual chemical-physical interactions they have with cement,
which give it all the properties and advantages already mentioned above.
The more accepted thesis is that the grinding aids act to reduce the surface energy forces gene-
rated on cement grains during comminution. They are composed of polar organic compounds,
which arrange their dipoles so that they saturate the charges on the newly formed particle surfa-
ces of the clinker, reducing agglomeration (Figure 2).

Figure 2

This theory stems from the fact that water, itself a polar molecule, has been considered to be a
grinding aid, but deemed less efficient due to its low screening effect. However, this theory does
not fully explain the fact that an additive is effective even at very low dosages (< 500 ppm) con-
sidered insufficient for a complete covering of the cement particle surface which would be neces-
sary for the “free charges” screening theory to be the only explanation.
Since no exhaustive evidence has been found that this is the unique mechanism, research in this
field is still going on, making use of the most sophisticated and modern technologies, trying to
find an appropriate answer to this problem.
Various research groups have followed distinct approaches, such as analysis of the sample addi-
tive extracted from cement after grinding, morphological analysis of the cement paste after addi-
tion, evaluation of the relationship between additive effects on mechanical strengths and cement
mineralogical composition.
Since 2000 MAPEI R&D laboratories have also concentrated their efforts on this subject, with the
objective of formulating high performance cement additives, which can satisfy client expectations.
Our approach has been as much multidisciplinary as possible with the aim of look at all aspects
of this complex subject. To do this, we have used some of the most updated analytical techni-
ques such as:
Figure 3

a b c d
(0 mins) (5 mins) (60 mins) (420 mins)
 TECHNICAL BACKGROUND
OF MAPEI GRINDING AIDS

13

Since the first gel forms within a few seconds and covers completely the cement grain, it actual-
ly represents the grain surface. Then water and any substances added to the cement, working as
an admixture, must interact with this gel.
This means that during the milling process of the cement, which is usually carried out in presen-
ce of water and grinding aids, conditions exist for the electrostatic charges dispersion discussed
before, but most of all for a preliminary hydration of the clinker.
Once the clinker powder produced is then mixed with water, we observe the formation of a gel,
whose structure is drastically different when grinding aids, such as MA.G.A. / C, are used during
the milling process (Figure 4).

Figure 4

Clinker hydration (10 mins) Clinker hydration (10 minutes)

ground with MA.G.A./C

These preliminary investigations suggest that the cement maintains a “memory” of the kind of
manufacturing process.

3. BENEFITS OF GRINDING AIDS

Grinding aids (GA) are principally employed for mill output increase at the same specific surface
(specific energy saving). The main benefits of GA are the following:

• Increase in output at the same fineness. Output increase reduces specific energy costs. The
increase in production could be used to cover market demand (increased sales), to reduce pro-
duction costs (grinding during lower energy cost periods) and to reduce maintenance costs.
• Fineness increase at equal output, or both effects. In some cases very high fineness may only
be obtained by using GA. Fineness increase at equivalent specific consumption allows improve-
ment of the final product performances (e.g. cement strengths).
• Improved granulometric particle distribution curve at equal fineness. In case of two cements
grinded at the same Blaine specific surface, the one grinded with GA has higher mechanical
strengths thanks to a more compact granulometric particle distribution curve obtained through the
elimination of the finest and coarse particles.
• Lower grinding media consumption.
• Higher separator efficiency. When agglomeration occurs a particle agglomerate would be seen
as a coarse particle and swept out by the separator. With the addition of MA.G.A./C the particles
are separated and passed as fine particles suitable for the final product.
• Improved flow characteristic of the cement during transport, silo storage and during
loading/unloading operations. Reduction of silo clogging, that usually results in reduced silos volu-
me and extraordinary repairs costs.

4. QUALITY IMPROVERS/PERFORMANCE ENHANCERS

Even if the pure grinding aids could be considered as grinding products which somehow improve
the quality of the finished material (ex. calcium carbonate with better granulometry and/or with
better flow) it is necessary to distinguish them from the pure quality improvers, that are tailor-made
formulations, and for which it is advisable to consult with the Cement Additives Division techni-
cians.

4.1 Cement additives

Cement performances are principally characterized by mechanical strengths, water demand and
setting times. MAPEI has formulated additives focused on improving each single characteristic,
while at the same time maintaining a strong grinding effect.
• Mechanical strengths. The UNI ENV 197-1 sets out the mechanical strengths required for each
cement class. The reasons why mechanical strengths (at early and ultimate ages) could be impro-
ved are the following:
- technical (insufficient strengths)
- commercial (competition on the market)
- economical (decrease in clinker additions, changing the cement composition etc.)
• Water demand. For certain types of cement (usually pozzolanic and fly-ashes cements) it is
important to control the flow values and modify them on end-users requests. There are specific
additives especially formulated to solve this problem, and that do not interfere with concrete
admixtures.
• Initial/final setting time. Special additives work on initial and final setting time, offering a cement
adapted to local temperature or to the season. Usually the capacity of the additive to modify the
setting times is an additional option, and its principal aim is as a grinding aid or strengths impro-
vement.

4.2 Air entrainers for hydraulic masonry cements

Artificial hydraulic masonry cement is obtained with the addition of 15-40% of clinker, and is cha-
racterized by good mechanical strengths, but lower workability than the natural one. The MA.P.E./A
give to artificial hydraulic masonry cements characteristics similar to the natural ones (workability
and durability). They also improve entrapped air and water retention. Air is entrained in micro-bub-
bles, homogeneously distributed, that improve cement workability, yield per surface unit and resi-
stance to freeze-thaw cycles.
Water retention improves adhesiveness by avoiding cracking.
 TECHNICAL BACKGROUND
OF MAPEI GRINDING AIDS

15

4.3 Pack-set inhibitors for minerals

Pack-set inhibitors are added during the dry minerals grinding process to obtain output increase
or high fineness. In case of calcium carbonate the additives must improve the flow characteristics
of the grinded material. For this reason specific products have been formulated, and are apprecia-
ted as much for their dispersing properties as for their grinding aid characteristics. MA.G.A./M GA
are available to satisfy various requirements, and tailor-made formulations can be studied by our
Division.

5. GRINDING AIDS EVALUATION

The evaluation of grinding aids has to include both technical and economical considerations. The
technical evaluation follows a reliable industrial trial, performed after preliminary lab tests. The eco-
nomical evaluation, to be preferably performed after the technical one, has to consider several fac-
tors that are not always easily quantified in terms of money.

5.1 Technical evaluation: industrial trials with grinding aids

Industrial evaluation of GA is necessary to verify their real obtainable performances. For this rea-
son a reliable industrial trial has to be carried out for a right evaluation of the product. A reliable
industrial trial must reduce to the minimum the quality differences of the raw materials used for
the cement production. Our suggestion is to grind for 2 weeks – 1 month, in order to compare the
average results obtained in a similar control period immediately before the industrial trial.
We hereby summarize the most important points that the Cement Additives Division suggests to
follow during an industrial evaluation of the additives, specifically referring to the cement.
• Only a reliable industrial trial performed for a long time can assure reliable results; our policy is
to carry out a long period test and to render technical assistance during the trial.
• A reliable industrial trial must be carried out under normal and regular production cycles.
• A long-period trial allows results to be obtained that are not affected by the normal variations of
the clinker, lime or other added materials.
• Each MAPEI additive has been completely tested in the lab, but the industrial trial is always
necessary to define the real additive performance on the cement.
• At the beginning of each trial, it is necessary to precisely define the plant working conditions with
or without the additive to be tested.
• It is important to define a sampling plan; it is advisable to divide the samples for the planned lab
tests to be performed both by the cement plant and Mapei laboratories.
• To optimize the results, a co-operation between plant technicians and Mapei consultant is advi-
sable. At the end of the trial, we suggest a meeting in order to work on the same results.
• At the end of the industrial trial, Mapei technicians will write a complete technical report to be
sent to the cement plant.

5.2 Economic evaluation

The most apparent effects of a grinding aids are output increase and energy saving, that could be
easily evaluated in terms of money. The other benefits are not so easily estimated, though some-
times they are more conspicuous than the energy saving.
The following formula quantifies the energy saving in terms of money S[€/t]:

Where: [€/t]

-P mill output without additive [t/h]

-α mill output increase [%]
- Ke total energy costs (absorbed kWh * cost kWh) [€]
- Ka additive cost (cost per kg * dosage) [€/t]

This formula calculates the saving per ton of ground material, in the cases where:
- Output increase at equivalent energy consumption;
- Fixed energy cost, at all times of the day;
- Product with similar quality to the previous one (without additive);
- Additive dosage calculated on the initial mill output (most common case).

When the dosage is calculated on the new mill output, the formula is modified as follows:

In practice, with the above calculation of the economical advantages, other several factors are not
considered; they have to be evaluated case by case and are always more difficult to estimate. We
hereby summarize the principal elements to be considered for a correct and complete estimation.
It has to be noted that, while the energy saving is always considered, other advantages depend on
the process conditions and should be evaluated to each specific situation.

5.3 Production benefits

• Mill output increase: more material available to increase sales.
• Grinding during low energy cost periods: lower average cost per unit of energy.
• Better management of peak market demands, due to increased production capacity. The addi-
tive could be considered as a flexible production tool, even though not constituting a structural
investment.

5.4 Quality benefits

• Cement performance improvement; greater competitive position on the market;
• Production of the same cement by decreasing the clinker percentage, with a saving in produc-
tion costs.

5.5 Maintenance benefits

• A higher production per hour allows more time for preventive maintenance interventions, avoi-
ding urgent maintenance that is always more expensive.
• Better flow characteristic of the cement during transport and in silos, with a significant reduction
of the plant stops and maintenance costs.
 TECHNICAL BACKGROUND
OF MAPEI GRINDING AIDS

17

6. ADDITIVE SELECTION
The grinding aid is liquid, and has to be dosed through a metering pump, onto the conveyor belt
or directly in the first chamber of the mill.

6.1 Dosage system

The simplest system is a manually regulated pump, installed soon after the other material measu-
rers, at the mill entrance. When the mill is operating under standard conditions, the additive is
dosed at a fixed flow rate, and does not alter in line with production variations. It is a very efficient
system, but has no flexibility and has to be strictly controlled by the operators.
In modern plants, the flow rate depends on the real mill output, and the additive is dosed by a flow
gauge that controls the pump through a computer. In this way the additive is treated as one of the
other components to be added in the mill, by indicating its percentage in the cement composition.

6.2 Metering pumps

Plunger or diaphragm metering pumps can be used. Plunger metering pumps (with steel plunger
and pump head) are largely used, due to their low cost. Diaphragm pumps are more sophisticated
and their benefit is that the plunger is not in direct contact with the additive. They can also suck
viscous additives and are not damaged by impurities sucked from the bottom of the tank. In order
to optimize the additive dispersion (especially for low dosage additives) our suggestion is to install
a double headed pump: the first head for the GA, and the second one for water, with an
additive/water ratio of 1:2 or 1:3. It is advisable to install a cartridge filter between the additive tank
and the pump, in order to prolong its functional lifetime.

6.3 Additive dosage onto the conveyor belt

This is the simplest system and allows a direct visual control of the flow. The additive must not wet
the conveyor, to avoid crusts whose thickness may enlarge on the return and support rolls. The
most frequent problems are caused by high dosage additives (0.2-0.3%) and on quick conveyor
belt (with a low material layer): in this case is necessary to use a rake to better distribute the addi-
tive and avoid drops falling onto the bare belt.
 Mill output

Regulation

30 m3 30 m3

6.4 Additive dosage into the first chamber of the mill

This is the “cleanest” dosage system, but it has to be carefully controlled to avoid problems. It is
advisable to install a metal lance with two coaxial tubes: the additive will flow inside, and the out-
side support air will be supplied by a fan. The additive tube must extend 1 cm more than the air
tube, to allow any possible drops to fall on the material. The fan has to work for a certain period
after each mill stop, to avoid the obstruction of the additive nozzle by the powder in suspension.
It is also advisable to position a T tap before the mill entrance, in order to manually measure the
additive flow.

Mill output

Regulation

30 m3 30 m3
 TECHNICAL BACKGROUND
OF MAPEI GRINDING AIDS

19

ADDITIVE LANCE SCHEME

1. Position of the lance relative to the mill

ST

2. Lance profile and section

6.5 Pump installation

Installation
1. Secure the pump to a horizontal support
2. Control that the oil level for the lubrification of the gears is ok
3. Protect the pump from liquids and powder
4. Leave enough free space around the pump for maintenance operations and adjustments.

150 mm

80 mm

100 mm

Fig. 1

100 mm

Typical scheme of installation in the plant

1. Low positive pressure in suction, hsuc
2. Higher positive pressure in discharge, h.>hdisc
3. Short and linear suction circuit
4. Suction and discharge circuits section pump connection section

hdisc

hsuc

Fig. 2
 TECHNICAL BACKGROUND
OF MAPEI GRINDING AIDS

21

In the cement plants the simple scheme below is used to dose the additives, considering that they
are dosed at low pressure and room temperature. For a more complete dosage system the follo-
wing components could be installed:

Suction circuit
a. Realise a short and linear circuit, with tube diameter of about 1.5 times the pump nozzles. The
max flow rate per single piston pump is 3,14 times more than the average one.
b. Install a permanent filter (1) in front of the pump in an easily accessible place, with 150 µm
meshes.
c. In case of long tubing, install a damper (2) beside the nozzle, in order to avoid the phenomena
of cavitation.
d. Avoid connecting the pump to the tank bottom, to avoid the suction of potential impurities.
e. Install a shut-off valve (3) for the disassembly of the pump when the lines are full.

Discharge circuit
f. Realize the circuit with tubes with diameter wider than or similar to the one of the pump nozzle.
g. It is advisable to maintain between suction and discharge a positive difference of pressure (40
– 100 KPa), as per fig. 2. If this is not possible due to the position of the tank and of the dischar-
ge circuit, (ex. tank positioned higher than the suction circuit) it is possible to install a back pres-
sure valve (5).
h. Soon after the pump it is advisable to install a calibrated safety valve (4), with a visible and open
discharge.
i. To regulate the liquid flow it is advisable to install a pulsation damper (with direct contact or with
a dry diaphragm), soon after the pump discharge valve. In case of installation of a back pressure
valve, it is better to position the damper after the valve.
j. In case of installation of a flow regulator (mechanical or inductive) (6), it is advisable to install a fil-
ter before the flow regulator.
k. A check valve should be installed for the possible disassembly of the pump when the lines are full.
A brief outline of an analysis
of the separation process

Amongst the large range of installations schemes of closed grinding circuits, which could inclu-
de one or more separators, we will consider and study the classic disposition represented in Fig.
1 where one mill and one dynamic separator are used.

Fig. 1

In the above scheme the machines operational systems are well defined.
The mill has a comminution function that, from a starting particle size distribution, creates new
and smaller cement particles and subsequently new specific surface.
From the particle size distribution supplied by the mill it is possible to produce a new product with
requested specific surface through a separator: the residue is automatically sent back to the mill
for further grinding.
In a closed circuit the separator increases the working value of the principal system: the mill.

1. EVALUATION OF A CLOSED CIRCUIT GRINDING SYSTEM

1.1 Circulation coefficient
In practice there are several systems for the evaluation of a plant working in a closed circuit; the
most immediate one seems to be the “circulation coefficient”, expressed by the ratio F/P (sep.
feed material/final product). It is a general opinion that a high circulation rate means a “poor fun-
ction” of the separation and vice versa; this could be true but the concept should not be genera-
lised.
In fact the final product could only be evaluated from the particle size distribution obtained with
the mill; if the final product is “big” the separator could only extract a fine product by refusing a
significant amount of coarse material, resulting then in a higher circulating load.
The circulation coefficient could then be considered as a quality index of the functioning of the
complete system, and not only of the separator.
It is useful to precise that the circulation coefficient A/F (circulating load) is also based on the mill
 23

dimensions, and depends on the L/D ratio. It is then logical that with short mills, the circulating
charge F/P has to be proportionally higher. In practice the circulating load represents an artificial
extension of the mill, that is to say a uniformity of permanence time of the material in the mill, with
respect to other L/D ratios.

1.2 Separator precision

Several methods have been proposed for the evaluation of the precision or performance of a
separation process.
In case of separation in two granulometric classes, having decided a separation dimension, and
called a, g and f the fractions (% in weight) contained respectively in the feeding material, coar-
se and final product, the “separator performance” can be defined as the ratio between the “fine
material” contained in the final product and that contained in the feed material.
Tab. 2 summarizes the methods normally used in accordance with the VDZ and MT28 standards.
The above assumptions have already been criticized, by highlighting the inadequacy of the eva-
luation of the functioning of an air separator, due to the influence of the granulometry of the feed
material on the results.
Furthermore, the choice of a particular separation dimensions, obtained by the above formulas,
can be considered as arbitrary, or at least, unjustified.
For this reason it is advisable to apply the Tromp curve (or T curve). This function represents the
trend of the separations coefficient rates of the feed material on the coarse and on the final pro-
duct, for the several granulometric intervals as a frequency curve on which the separation pro-
cess takes place. This should allow for greater precision, irrespective of the feed material’s com-
position and of the choice of the separation dimensions.
The quantitative evaluation of the above precision depends on some curve parameters; the
“separation precision” is defined by the rule VDZ as the rate d35/d65 (being d35 e d65 respecti-
vely the granulometries in µm corresponding to a separation degree of 35 % and 65 %); the same
rule defined the “separation limit” as the particle dimension of 50% between “large” and “fine”.
In practice these parameters are not always applicable due to the fact that, as the curve T is
dependent on the A/F rate, the separation curve of many plants is above these distinctive points,
and it is thus quite impossible to make a reliable calculation.

Recently other evaluation methods have been studied and proposed by C.E.T.I.C., always deri-
ved from the Tromp curve that, though its value is not intrinsic, allows the location of the curve
with precise coefficients, hence allowing calculations to be carried out.

• In practice, following CETIC (that proposes to use the Gaussian coordinates to draw the dia-
gram), the curve T is resolved into two lines:

1. the horizontal line, that is the mean of the separation degrees below the ordinate at the abscis-
sa point 1 µm (soutirage or by pass);
2. a regression line related to the succeeding points on the ascending part of the curve.
We can state the following:
a. the “soutirage” or “by pass”, measured from the minimum ordinate (1), represents the material
passed through the separator without being classified.
b. The quality of the “cutting”, defined by the inclination of the ascending part of the
T curve; the inclination has also been measured by the “imperfection” value.

TAB. 3 shows in details the drawing system of the T curve following CETIC.

2. COMMENTS
We tried to summarise the different methods applied for the evaluation of the separator functio-
ning.
We consider it opportune, at this point, to express our considerations on the argument:

• We already criticized the limits of the so-called “performance” (arbitrary choice of the particle
dimensions, influence of the granulometry of the feeding material, etc.); it is necessary to remem-
ber that the above formulas derives from the mineral preparation technique where, generally, it is
necessary to separate the components of a granulometric mix into two fractions. These fractions
could be different one from the other for particle dimensions or chemical composition; in this case
a product can contain a material that has not been treated and that has to be separated or recy-
cled, or a fine component mixed in a inert mass, that has to be extracted for a further treatment.
The formulas represent a concept bound to the results (performance) of a classification; they
measure the efficiency of the operation in “open cycle” in relation, for example, to the loss of “fine
material” (final product) in the coarse recycled product.
In our case, separation in “closed cycle”, the coarse material is re-introduced in the mill for a fur-
ther refining and consequently, from the point of view of the preparation of the granulometric
curve for that type of cement, the material to be ground in the mill is already “fine”, with a con-
sequent energy saving when compared to a “bigger” granulometry.

• The Tromp curve and related parameters, even though they precisely represents the separation
process, are affected by the ratio between the loading charge (A) and productivity (F) in such a
way that is not possible to evaluate the separator, independently from the kind of mill in use. The
separation coefficients confirms the theoretical deduction that the preciseness of the separation,
appearing from the T curve, is higher when the circulation coefficient A/F is lower. Following this
assumption, the performance of the separator is optimal if combined with a long mill, and on the
contrary is worse if combined with a short or very-short mill.

3. CALCULATION EXAMPLES
Taking into consideration the limits defined by our assumptions concerning the possibility of
comparison between two different separators, we hereby report the T curves of some separator
plants, useful for a clear understanding of the above mentioned concepts.
 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

25

The table below summarises the representative parameters (rounded off to significant values):

Mill Loading Separator Soutirage Line Circuit

L/D charge inclination
ratio A/F type % degrees N°
3.6 1.3 Wedag ZUB36 12 31 1
3.1 1.4 Polysius TSU 19 24 2
3.3 1.5 Wedag ZUB60 9 46 3
3.0 3.9 Smidth CV short 66 22 4

Some comments:
• The separators are of the 1st (Polysius e Smidth) and of the 2nd generation (Wedag): the
Polysius TSU has double shafts while the Smidth separator has only one shaft.
• The mills L/D ratios ranges between 3 e 3.6, and so are not different one from the other
• The loading charges, except from circuit no. 4, are coherent to the mills L/D ratios, even if they
are quite low.

As already anticipated, the "soutirage" value quantifies the material passed through the separa-
tor, but that has NOT been classified. It could be defined as a “separation” error.

At this point, the performance of the Wedag separator was good.

On the contrary, the performance of the Polysius separator of plant [1] was worse. In fact, with a
loading charge of 1.4 (that is quite low, and it is already known the direct correlation within loa-
ding charge and soutirage), a by-pass error of 19% has been noted. It has to be taken into
account that this is a 1° generation separator.

Always considering the “soutirage”, the separation results of plant [4] would appear really nega-
tive; the soutirage error was 66%; however knowing the plant and the separator in use, we can
define that these poor performances are essentially caused by the high fineness of the product,
and by the limits of the fineness regulation system of the separator itself.

Concerning the separation “quality”, represented by the inclination of the second branch of the
T curve, a substantial equality between all the separators is revealed. The inclinations range bet-
ween 20 and 30 degrees. The only exception is the separator of plant [3] which has a higher incli-
nation, and hence precision, with a line angle of 46°.

In conclusion, the last observation:

Taking into account the limits of our assumption, and also the difficulty of comparison, we deem
it useful to highlight the good performance of the Wedag separator which may be judged as relia-
ble and competitive even when compared to the latest 3° generation separators.
 Separation Analysis (Tromp Curve)

Cement Plant Cement Type CEM II/A-M 32.5 R

Mill Smidth 2.9 x 10.4 Production t/h 42.5
Separator Wedag ZUB 36 Spec.surf. M2/kg
Test no. blaine
Date laser

dim.µm f a g df da dg df*F/A dg*G/F A G/A T (%)

2 1.0 16.0 15.0 9.6 16.0 15.0 9.6 12.1 2.32 14.45 0.16 16.05
4 3.0 27.7 25.0 16.8 11.7 10.0 7.2 8.9 1.74 10.61 0.16 16.40
8 6.0 37.4 32.8 21.0 9.7 7.8 4.2 7.4 1.02 8.37 0.12 12.13
16 12.0 49.2 42.3 23.8 11.8 9.5 2.8 8.9 0.68 9.62 0.07 7.03
32 24.0 70.6 60.0 28.5 21.4 17.7 4.7 16.2 1.14 17.36 0.07 6.54
64 48.0 93.9 83.1 50.9 23.3 23.1 22.4 17.7 5.41 23.08 0.23 23.46
128 96.0 99.9 97.9 85.6 6.0 14.8 34.7 4.5 8.39 12.94 0.65 64.83
190 159.0 100.0 100.0 98.8 0.1 2.1 13.2 0.1 3.19 3.27 0.98 97.68

total 494.7 456.1 335.0 Materials in cycle residue [determ. laser / µm]
t/h 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.7583 final 42.5
G/F 0.2417 feeding 56.0
coarse 13.5
Loading Charge A/F 1.3 Sep. Line angle Deg. 31
Soutirage % 11.6
Ord.Sep.diam % 56

T separation curve

T curve soutirage regression ord.sep.diam. Linear (regression)

 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

27

Separation Analysis (Tromp Curve)

Cement Plant Cement Type CPC 30

Mill Polysius 4.6 x 14.5 Production t/h 126.0
Separator Polysius TSU 6.5 (n.ro 2) Spec.surf. M2/kg
Test no. blaine 325
Date laser 328

dim.µm f a g df da dg df*F/A dg*G/F A G/A T (%)

2 1.0 11.7 10.3 8.1 11.7 10.3 8.1 8.2 2.42 10.62 0.23 22.81
4 3.0 19.5 17.7 13.9 7.8 7.4 5.8 5.5 1.73 7.20 0.24 24.09
8 6.0 29.4 26.9 20.9 9.9 9.2 7.0 6.9 2.09 9.03 0.23 23.18
12 10.0 36.2 32.8 24.9 6.8 5.9 4.0 4.8 1.20 5.96 0.20 20.06
21 16.5 54.2 47.8 33.0 18.0 15.0 8.1 12.6 2.42 15.04 0.16 16.11
30 25.5 64.1 56.1 37.5 9.9 8.3 4.5 6.9 1.35 8.28 0.16 16.25
60 45.0 84.0 74.8 53.0 19.9 18.7 15.5 13.9 4.64 18.58 0.25 24.95
90 75.0 94.3 87.5 71.2 10.3 12.7 18.2 7.2 5.44 12.66 0.43 42.99
128 109.0 98.4 96.0 89.9 4.1 8.5 18.7 2.9 5.59 8.47 0.66 66.06
192 160.0 99.8 98.9 96.1 1.4 2.9 6.2 1.0 1.85 2.84 0.65 65.40

total 591.6 548.8 448.5 Materials in cycle residue [determ. ALPINE]

t/h 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.7009 final 126.0
G/F 0.2991 feeding 179.8
coarse 53.8
Loading Charge A/F 1.4 Sep. Line angle Deg. 24
Soutirage % 19.0
Sep.diam µm 131

T separation curve

T curve soutirage regression ord.sep.diam. Linear (regression)

 Separation Analysis (Tromp Curve)

Cement Plant Cement Type 42.5 PTL

Mill Tosi 4.6 x 15.0 Production t/h 115.0
Separator Wedag ZUB 60 Spec.surf. M2/kg
Test no. blaine 397
Date laser 332

dim.µm f a g df da dg df*F/A dg*G/F A G/A T (%)

2 1.0 9.8 6.6 3.6 9.8 6.6 3.6 6.6 1.18 7.76 0.15 15.25
4 3.0 16.8 11.9 6.3 7.0 5.3 2.7 4.7 0.89 5.59 0.16 15.89
8 6.0 29.4 22.9 9.2 12.6 11.0 2.9 8.5 0.95 9.41 0.10 10.13
12 10.0 39.6 31.3 10.3 10.2 8.4 1.1 6.8 0.36 7.21 0.05 5.02
21 16.5 62.1 46.9 11.4 22.5 15.6 1.1 15.1 0.36 15.46 0.02 2.34
30 25.5 73.7 55.1 13.0 11.6 8.2 1.6 7.8 0.53 8.31 0.06 6.33
60 45.0 93.8 74.8 35.1 20.1 19.7 22.1 13.5 7.27 20.76 0.35 35.01
90 75.0 99.4 85.9 59.8 5.6 11.1 24.7 3.8 8.12 11.88 0.68 68.36
128 109.0 99.9 93.8 83.5 0.5 7.9 23.7 0.3 7.79 8.13 0.96 95.87
192 160.0 100.0 96.4 91.5 0.1 2.6 8.0 0.1 2.63 2.70 0.98 97.51

total 624.5 525.6 323.7 Materials in cycle residue [determ. ALPINE]

t/h 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.6712 final 115.0 26.3 11.4 6.2 0.6 0.0
G/F 0.3288 feeding 171.3 44.9 32.6 25.2 14.1 3.6
coarse 56.3 87.0 76.9 64.9 40.2 8.5
Loading Charge A/F 1.5 Sep. Line angle Deg. 46
Soutirage % 9.2
Sep.diam µm 77

T separation curve

T curve soutirage regression ord.sep.diam. Linear (regression)

 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

29

Separation Analysis (Tromp Curve)

Cement Plant Cement Type CEM II/A-S 42.5 R

Mill Smidth 4.0 x 12 (2100 kW) Production t/h 85.0
Separator Smidth CV 6m (short) Spec.surf. M2/kg
Test no. blaine
Date laser

dim.µm f a g df da dg df*F/A dg*G/F A G/A T (%)

2 1.0 16.9 12.0 11.8 16.9 12.0 11.8 4.3 8.82 13.09 0.67 67.42
4 3.0 30.2 22.9 21.4 13.3 10.9 9.6 3.4 7.18 10.53 0.68 68.15
8 6.0 42.5 32.9 30.4 12.3 10.0 9.0 3.1 6.73 9.83 0.68 68.44
12 10.0 54.1 41.2 37.0 11.6 8.3 6.6 2.9 4.94 7.86 0.63 62.78
21 16.5 69.8 52.3 45.9 15.7 11.1 8.9 4.0 6.65 10.62 0.63 62.69
30 25.5 79.5 60.6 53.3 9.7 8.3 7.4 2.4 5.53 7.98 0.69 69.34
60 45.0 97.2 82.9 77.5 17.7 22.3 24.2 4.5 18.10 22.56 0.80 80.21
90 75.0 100.0 94.4 91.2 2.8 11.5 13.7 0.7 10.24 10.95 0.94 93.55

total 490.2 399.2 368.5 Materials in cycle residue [determ. ALPINE]

t/h 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.2523 final 85.0
G/F 0.7477 feeding 337.0
coarse 252.0
Loading Charge A/F 4.0 Sep. Line angle Deg. 22
Soutirage % 65.9
Sep.diam µm 65

T separation curve

T curve soutirage regression ord.sep.diam. Linear (regression)

4. CONCLUSIONS
On the separator performance problem, it is important to remember the assumption of
B. Beke (Principles of comminution, Akademiai Kiado, Budapest, 1964):
“…luckily the separators do not have a 100% performance, otherwise it would not be possible
to create the granulometries necessary for the development of the cement characteristics…”.

In our opinion it may be added that all the formulas, results and evaluations obtained with the
available methods, are NOT transferable; that means that an objective analysis of the functioning
of the separator is not possible if separated from the analysis of the combined mill.
Indisputable is the fact that, inside a grinding system, each single variation of the separation
parameters can affect the mill output (and vice versa), and this is revealed by the specific energy
consumption. The successive separation analysis will only confirm the variation (positive or nega-
tive) of the parameters of evaluation.

With the above assumptions, and taking into consideration the usefulness of the analysis of a
separation process, we would like to emphasise our belief that it is impossible to have objective
comparisons of other and different separators when installed on different circuits.
 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

31

SEPARATION ANALYSIS
SYMBOLOGY BASED ON VDZ MT 28 STANDARD

CAPACITY

M t/h “fresh” material feed = F

A t/h separator feed
F t/h fine material
G t/h coarse material

GRANULOMETRIES with a one reference particle dimension

a % passing of A
f % passing of F
g % passing of G

FUNDAMENTAL EQUATION

A = F+ G
A*a = F*F + G*G

LOADING CHARGE u = A/F Ratio between the separator feed A and final product F.
u = (f-g) / (a-g)

FINENESS PERFORMANCE Vf = F/A*100 Ratio between the final product F and the feed material A.
Vf = 1/u
Vf = (a-g)/(f-g) * 100

The performance is referred to a granulomoeric dimension:

PERFORMANCE Ew = F/A * f/a it indicates the feeding material that passes in the final pro-
Ew = (a-g)/(f-g) * f/a duct after the separation.

SEPARATION EFFICIENCY Ek = (a-g)/(f-g) * 100 * (f-a)/a * (100-a)

Always referred to a granulometric dimension, indicates the
yield of the passing material in the product, reduced by the
yield of the residue in the same.
1. SYMBOLOGY

M = fresh feed (t/h)

A = separator feed (t/h)
F = separator final product (t/h)
G = separator coarse material (t/h)
a = passing of A (%)
f = passing of F (%)
g = passing of G (%)

We can define the “passing” of the material as the fraction (% in weight)

of particles which have equal or lower dimensions to the granulometry of reference.

Fundamental equations

A=F+G [i]
A*a = F*f + G*g [ii]

a = part in weight of a specific granulometric class of A (%)

f = part in weight of a specific granulometric class of F (%)
g = part in weight of a specific granulometric class of G (%)

We can define granulometric class the interval between the two granulometric measures.

A*a = F*f + G*g [iii]

 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

33

2. GRANULOMETRIC CHARACTERISTICS THE SEPARATOR FLOWS

Granulometric Final product Coarse Feeding

intervals f % df g % dg a % da
0-2 13.3 13.3 7.4 7.4 10.3 10.3
2-4 22.2 8.9 12.0 4.6 17.1 6.8
4-8 34.6 12.4 17.2 5.2 25.3 8.2
8-16 51.8 17.2 23.5 6.3 37.5 12.2
16-32 73.2 21.4 32.8 9.3 52.5 15.0
32-64 92.0 18.8 54.3 21.5 72.1 19.6
64-128 99.4 7.4 75.1 20.8 87.6 15.5
> 128
Σi 386.5 222.3 302.4

3. CIRCULATING LOAD CALCULATION

It’s expression is Tc = A/F

It is defined by the granulometric analysis of the 3 separator flows (see table – point 2) using the Koulen
formula.

Being fi ai gi the fine product, feeding material and coarse passing on the mesh i (for the following dimen-
sions: 2,4,8,16,32,64 e 128 µm).

Example
4. SEPARATION CURVE DRAWING - “T CURVE”
[with recomposed separator feed]

We calculate:
- performance in fine material Vf = 1/Tc
- performance on “big” material Vg = 1-Vf
- the feeding fraction between 2 granulometric measures.

a = df*Vf + dg*Vg
a = df*Vf + dg(1-f)

- the separation degree T

Example

We calculate the T separation degrees, with the data from the table - point 2.

Tc = 2,04 Vf = 1/Tc = 0.4902 Vg = 1-Vf = 0,5098

Granulometric Final Coarse f g Feed Separation Mean

interval product degree diameter
m df[%] dg[%] Df*Vf Dg(1-Vf) f+ g T[%] m
0-2 13.3 7.4 6.52 3.77 10.29 36.7 1
2-4 8.9 4.6 4.36 2.34 6.70 34.9 3
4-8 12.4 5.2 6.08 2.65 8.73 30.3 6
8-16 17.2 6.3 8.43 3.21 11.64 27.6 12
16-32 21.4 9.3 10.49 4.74 15.23 31.1 24
32-64 18.8 21.5 9.21 10.96 20.17 54.3 48
64-128 7.4 20.8 3.63 10.60 14.23 74.5 96
> 128
 A B R I E F O U T L I N E O F A N A N A LY S I S
O F T H E S E PA R AT I O N P R O C E S S

35

5. T CURVE DRAWING

The curve is resolved in two lines:

- Horizontal: the ordinate is the mean of the separation degrees value, represented by the corre-
spondent points. The points to be considered for the calculation of the mean value (Soutirage)
are below the ordinate at the abscissa point 1 µm;
- a regression line related to the succeeding points on the ascending part of the curve;
- [in practice it is better to cancel one or more points on the right that curve the line to the bot-
tom. Only the 2 points situated more or less straddling the normalized separation diameter will be
often considered. For this reason it is useful to consider the point on the right of the horizontal
line.

6. SEPARATION CHARACTERISTIC PARAMETERES (NORMALIZED)

6.1 Soutirages
Material passed through the separator without being classified. It is measured by the minimum
ordinate (horizontal) of the T curve, calculated as per point 5.

100

d
6.2 Separation diameter Φp
This diameter corresponds to a probability of 50% to be distributed within coarse and fine mate-
rial. This per centage is not applicable to all the feeding material, but only to the one that has real-
ly been separated; that means 100% - soutirage s. The ordinate is calculated as follows: s +
(0,5*(100-s)) and we calculate, on the pending part of the T curve, the corresponding abscissa.

100
S+(0.5*(100-s))

Φp d

6.3 Imperfection I
Represents the T curve inclination (perfect separation = 0). It is represented by the difference bet-
ween the diameters correspondent to the separation coefficients 25% e 75% for the material that
has really been separated. That means 100%- soutirage s. On the separation chart the following
ordinates have been reported:
S+(0,25*(100-s))
S+(0,75*(100-s))
On the pending part of the curve are defined the correspondent diameters Φ25 e Φ75;
The calculation is the following:

Φ25 -Φ75
I= 100
2 Φp
S+(0.25*(100-s))

S+(0.25*(100-s))

Φ25 Φ75 d
Products
 ./ C
.A
A.G
M
MApei Grinding Aid/C
Pack set inhibitors
for cement

DESCRIPTION APPLICATION PROCEDURE

MA.G.A./C are high performance grinding aids MA.G.A./C may therefore be successfully utilized in
generally used to increase mill production and all cases of pack-set phenomena (not due to
to improve the cement quality. humidity) inside the mill, particularly in the grinding
of Portland and limestone cements. Productive
They are highly concentrated additives formulated increases generally may vary between 10% and
with only selected raw materials, to guarantee 30%, depending on the fineness of the cement,
absolute constancy of quality and superior on the grinding system available, on clinker
performance. mineralogical composition, on the additive dosage
etc.
TECHNICAL CHARACTERISTICS
MA.G.A./C, thanks to their polar nature, notably When the right conditions are present, specific
reduce the attraction forces of cement particles formulations for limestone cements allow the
which are the main cause of agglomeration inside increase of even the ultimate strengths; it is possible
tubular mills (pack-set). They are also able to modify to decrease the percentage of clinker in the cement
the hydrates’ structure (see pictures on the back recipe, with no loss in cement quality.
page).
CHEMICAL-PHYSICAL DATA
The disappearance or the remarkable reduction of Please refer to the appropriate safety data sheets.
pack-set phenomena favourably modify the
granulometric curve of the finished product with a DOSAGE
consequent beneficial effect on the strengths and • 0.1-0.5 kg/t.
efficiency of theseparation.
• Ordinary Portland cements (2600-3200 cm2/g):
It is therefore possible to obtain important production 100-200 g/t.
increases (savings in kWh/t) or, with the same
• Rapid Hardening Portland cements
production, improvements of the specific surface of
(3200-4600 cm2/g): 200-500 g/t.
the finished cement.
• Limestone cements: 250-500 g/t.
Specific formulations allow the modification of the
hydration products of the cement increasing the early We recommend the higher dosage threshold in
and/or ultimate strengths. presence of high percentages of limestone and when
 41

INDUSTRIAL TRIAL WITH MA.G.A./C1

Report of an industrial trial performed
with MA.G.A./C1 during the production
of high fineness CEM I type cements
t

1. TRIAL DESCRIPTION
a. Plant description
b. Trial objectives and additive selection

2. TRIAL RESULTS
t

a. Immediate results during the Trial

b. Post-trial results
c. Technical analysis
t

3. CONCLUSIONS
42 1. TRIAL DESCRIPTION
a) Plant Description

The cement grinding facility consists of two mills side by side, one dedicated to the grinding of
cements while the second is dedicated to the grinding of slag. Our trials were performed on the
cement mill which has the following characteristics:

Cement Mill
UMS Mill 5.0 x 14.0, two chambers; installed power 5750 kW.

The cement grinding circuit is closed with a 3° generation Sepax 450M222 Separator. Separator
and mill ventilation is realised with sleeve filters. (Fig. 1 – Flow sheet diagram).

Fig. 1 Flow sheet diagram

1. Fresh material feed

2. Cement mill
3. Elevator
4. Separator
5. Mill bag filter
6. Bag filter

b) Trial objectives and additive selection

An industrial trial to evaluate the performance of the product MA.G.A./C1, a high performance grin-
ding additive formulated as a concentrated grinding aid able to increase early strengths, on the
production of CEM I (400 m2/kg Blaine) and CEM I (550 m2/kg Blaine) cements. The plant’s objec-
tives was to increase the hourly production (t/h) of the mill as the cement plant had significantly
increased their quota of the local market and needed to maximise their production output.
2. TRIAL RESULTS 43

Table N°1.
Details Units Blank MA.G.A./ % Blank MA.G.A./ %
C1 Difference C1 Difference
Cement Type CEM I CEM I CEM I CEM I
(400 m2/kg) (400 m2/kg) (550 m2/kg) (550 m2/kg)
Additive g/t --- 300 --- 300
Dosage
Production t/h 133.0* 144.3 +8.5 74.0 83.5 + 12.8
Specific S.A. m2/kg 411 562 539 - 4.1
Specific S.A. km2/h 59.3 41.6 45.0 + 8.2
production
Specific Mill kWh/t 36.2 70.7 63.0 - 10.9
Consumption
Specific
consumption kWh/km2 125.8 116.9 - 7.1
S.A. production
R1 MPa 20.7 27.2 30.5 +12.1
R2 MPa 35.2 41.3 44.3 +7.3
R28 MPa 59.9 62.4 62.8 +0.6
* Average Plant Data

a) Immediate results during the trial

Table N°1 in paragraph 3.1 shows the average hourly production obtained during the trial. These
results confirm the validity of the MA.G.A./C1 product as a powerful grinding aid.

CEM I (400 m2/kg)

In the case of the CEM I (400) trial the average production obtained with Mapei MA.G.A./C1 was
144.3 t/hr. This corresponds to an increase of +8.5% with respect to the blank, i.e. without an addi-
tive, which was 133.0 t/hr. This blank value is an average value for the month previous to the trial.
Grinding without the additive was not possible on this cement due to silo space restrictions. The
value of Blaine measured on the mixed sample is 411 m2/kg. This is higher than the desired value
(+11 m2/kg) and may indeed be translated into a further increase in hourly production.
 CEM I (550 m2/kg)
44
In the case of the CEM I (550) trial the average production obtained with Mapei MA.G.A./C1 was
83.5 t/hr. This corresponds to an increase of +12.8% with respect to the blank, i.e. without an
additive, which was 74.0 t/hr. This is a significant increase for such high Blaine grinding and is a
positive result which allows a reduction of the specific mill consumption from 70.7 to 63.0 kWh/t.
Another result is a higher surface production: +8.2% (45.0 km2/h versus 41.6 km2/h).

Additives 133.
133.0 144
144.
.3
Mill Production CEM I (400) Mill Production CEM I (550)
Blank MAGA/C 1 Blank MAGA/C 1
150.0
150. 85.0
85.
83.5
83.
144
144..3
145.0
145.
80.0
80.

140.0
140.
74.
74 .0
133.
133 .0 75.
75.0
135.0
135.

130
130..0 70.0
70.
Additives Additives

b) Post-Trial results

The results regarding cement strengths are available in Table N°1. All testing is carried out in
accordance with the European Standard EN 196. The cement without additive where available is
compared to that produced with Mapei MA.G.A/C1.

CEM I (550 m2/kg)

As regards the strengths at 1 day there is an increase of +12.1% with MA.G.A/C1 from 27.2 to
30.5 MPa, while at 2 days the increase is +7.3% from 41.3 to 44.3 MPa. The results at 28 days
are practically the same.

CEM I (550) - Strengths CEM I (550) - Strengths

Blank MAGA/C 1 Blank MAGA/C 1
33.
33.0 45.0
45.
44.3
44.
30.5
30.
31.
31.0 44.0
44.

29.0
29.
27.
27 .2 43.0
43.
27.0
27.
42.0
42.
25.0
25. 41.
41 .3
41.0
41.
23.0
23.

21.0
21. 40.0
40.
1 Day 2 Days
These results are positive when one considers the increase in production obtained at the same
45
time. The values of Blaine reported indicates that the cement produced without the additive is sli-
ghtly finer than that produced with MA.G.A/C1. Hence here one can appreciate the positive
influence of the product’s formulation on the early strengths.

c) Technical analysis

The following samples were taken from the separation circuit consisting of a FLS SEPAX 450 type
separator during the industrial trial:

• finished product
• separatore feed
• return or recycle – material returning to the mill

On these samples granulometric particle size analysis was carried out with the Alpine sieve
equipment and the complete particle size distribution curve was defined using a Coulter LS laser
equipment.
Successively the following were determined: Separation performance according to DIN and
Tromp.

Separation Performance According to DIN:

The calculated values according to the DIN are given in Table N°2 determined from the “Dry’’ resi-
dues of the materials tested; when one considers the very high specific surface area of the ‘‘fini-
shed’’ product: > 400 m2/kg, the calculated characteristic parameters: F/A and ETA w, lie within
the “normal’’ range.
The ETAw parameter is interesting, as it expresses the “ability” of the separator to “clean” the
material being fed; in our case, on the 40 µm control mesh, the “recovered” fraction of less than
-40 µm amounts to 67 % of the total present in the feed material. This can be defined as a “good
result”.
46
Separation Performance Analysis according to DIN

Material: CEMENT
Mill: FLS UMS Cement Mill 5.0 x 14.0

Separator Brand: FLS

Type: SEPAX 450

Separation Performance According to DIN, Calcolated with Alpine

TABLE 2 [passing values] Output calculation

µm final p. sep. feed return from TABLE 2 a (D) µm 40
(D) (f) (a) (g)
20 F= t/h 144.3 144.3
30 87.5 44.0 19.0 A= t/h 389.1 389.1
40 93.5 52.5 28.0 G= t/h 244.8 244.8
50
60 99.5 75.0 59.5 A= t/h
70 F= t/h 0.0 0.0
90 100.0 86.0 78.5 G= t/h 0.0 0.0
200
Total 380.5 257.5 185.0

(f) (a) (g) control mesh [D]

PERF a(D) 93.5 52.5 28.0 40.0 µm eta w 0.67
format 0.00
mean a (D) µm
values µm 40
CIRCULATING LOAD A/F 2.70 2.67
EFFICIENCY F/A 0.37 0.37
PERF.ACC. DIN ETA w 0.55 0.67 (§) (effective yield/possible)
CLASS.EFFICIENCY ETA k 0.61 (granulometric variation)
DIVISION PERFORM. ETA s 0.65 (fraction separated/theorical)-(NON separated fraction)

NOTES (Rosin-Rammler) (§) intended as "ability" of the

1 U A/F f-g/a-g separator to recover the finished
2 Yield F/A (1/U) product (F) the material (D)
3 Performance ETA w 1/U*f/a present in the feed A.
4 EFFIC.CLASS. ETA k (a-g)/(f-g)*100*(f-a)/a*(100-a)
5 PERF.DIVIS. ETA s a-g/a*100/100-g
According to Tromp:
47

From the laser particle size curves of the material in the separator circuit, the corresponding T
curve (of Tromp) was calculated which represents for the various granulometric fractions the
development of the separation process; Table 3 shows the T curve drawn, on the co-ordinates of
a semi-logarithmic graph and the calculation of the characteristic parameters elaborated accor-
ding to CETIC.

Sepax
Circulating load dimensionless 3.1
Soutirage % 23.8
Diameter of separation µm 42
Inclination angle of regression line (2^ branch separation curve) °G 47

The profiles of the curves and the parameters deducted from them allow the operation of the
separator to be judged as ‘‘discreet’’. In relation to the not very high circulating load: approx. 3,
the error of “soutirage” (by pass) is NOT very good; this may depend perhaps on an inadequate
ventilation.
Also as regards the profile of the 2^ branch of the T curve, which represents the type of separa-
tion, an “inclination” > 40°G can be considered as “acceptable”.
48 Separation Analysis (Tromp Curve)

Cement Plant Cement Type CEM I 400

Mill FLS UMS Cement Mill 5.0 x 14.0 Production t/h 144.3
Separator FLS SEPAX 450 Spec.surf. M2/kg
Test no. blaine 410
Description Grinding with MA.G.A./C 1 laser 362

dim.µm f a g df da dg df*F/A dg*G/F a (§) T (%)

2 1.0 9.7 4.5 2.2 9.7 4.5 2.2 3.1 1.47 4.56 0.32 32.2
4 2.0 17.1 8.1 3.6 7.4 3.5 1.4 2.4 0.96 3.34 0.29 28.8
6 4.0 23.5 11.0 4.6 6.4 2.9 1.0 2.0 0.69 2.73 0.25 25.1
8 6.0 29.5 13.5 5.3 6.0 2.5 0.7 1.9 0.49 2.41 0.20 20.5
10 8.0 34.9 15.8 5.9 5.4 2.3 0.5 1.7 0.37 2.10 0.18 17.6
15 11.5 46.8 20.7 7.1 11.9 4.9 1.3 3.8 0.86 4.67 0.18 18.5
20 15.0 56.4 24.8 8.6 9.6 4.1 1.4 3.1 0.98 4.05 0.24 24.2
30 22.5 71.0 31.6 11.9 14.6 6.9 3.4 4.7 2.28 6.95 0.33 32.8
40 30.0 83.7 39.1 17.4 12.7 7.4 5.5 4.1 3.76 7.82 0.48 48.0
60 45.0 94.7 54.9 36.6 11.0 15.9 19.2 3.5 13.05 16.76 0.79 78.8
90 65.0 97.4 70.3 60.5 2.7 15.4 23.8 0.9 16.21 17.08 0.95 94.9
100 80.0 98.6 74.5 66.5 1.2 4.2 6.0 0.4 4.09 4.47 0.91 91.4

total 663.3 368.8 230.2 Materials in cycle spec. Surface area m2/kg residue [determ. ALPINE]
t/h laser blaine 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.32 final 144.3 362 410 12.5 6.5 0.5 0.0 0.0
G/F 0.68 feeding 451.0 181 206 56.0 47.5 25.0 14.0 3.0
coarse 307.0 94 122 81.0 72.0 40.5 21.5 4.0
Loading Charge A/F 3.1 Sep. Line angle Deg. 47
Soutirage % 23.8 Efficiency Evaluation Average =(Sep. Line angle)
Sep.diam µm 42 Imperfection ad 0.43

T separation curve (acc. CETIC) - CEM I 400 with MA.G.A./C 1

3. CONCLUSIONS 49

The results obtained can be considered to be very positive both in terms of production (t/hr) and
cement quality (MPa at 1 & 2 days).
The most important objective of the plant at the time was to increase hourly production and the
results obtained are very interesting when one considers the low dosage (300 g/t) of the
MA.G.A./C1 product.

Cement Blank MA.G.A./C1 %Increase

CEM I (400 m2/Kg) 133.0 144.3 +8.5
CEM I (550 m2/Kg) 74.0 83.5 +12.8

MA.G.A./C1 guarantees significant production increases and increases early strengths. Hourly
cement production capacity has reached the levels desired by plant management, at the same
time reducing production energy costs.
 51

INDUSTRIAL TRIAL WITH MA.G.A./C2

Report of an industrial trial performed
with MA.G.A./C2 during the production
of CEM I 42.5 R
t

1. TRIAL DESCRIPTION
a. Plant description
b. Trial objectives and additive selection

2. TRIAL RESULTS
t

a. Immediate results during the Trial

b. Post-trial results
t

3. CONCLUSIONS
52 1. TRIAL DESCRIPTION
a) Plant Description

The cement grinding facility has a group of four mills side by side and the trial was performed on
mill number 3 which has the following characteristics:

FLS Mill of dimensions 4.6 m x 16.0 m, two chambers of which the first chamber is 3.0 m and the
second is 13 m long; installed power is 4500 kW.
The grinding circuit is closed with a high efficiency SEPOL SVZ separator, with an ESP filter on
the mill. (Fig.1 - Flow sheet diagram)

Fig. 1 Flow sheet diagram

1. Fresh material feed

2. Cement mill
3. Elevator
4. Separator - SEPOL SVZ
5. ESP Filter

B) Trial objectives and additive selection

An industrial trial to evaluate the performance of the additive MA.G.A./C2, a high performance
grinding additive formulated to increase 28 day strengths, on the production of the cement CEM
I 42.5 R (95.6% Clinker and 4.4% Gypsum). The plant’s objective was to increase the 28 day
strengths in order to maintain it’s market leader position without having to alter the quality of their
clinker.
MA.G.A./C2 was proposed given it’s ability to meet the client’s objectives of increased 28 days
53
mechanical strengths while at the same improving the cement mill’s hourly production. The per-
formance of the product in terms of it’s excellent grinding aid characteristics and strength impro-
vement are directly related to the product formulation and the high quality of the raw materials
utilised.

2. TRIAL RESULTS
a) Immediate results during the trial

The MA.G.A./C2 product was added onto the clinker transported on the conveyor belt just befo-
re the entrance to the mill at a dosage of 450 g/t. The diagram in Figure 2, related to the circula-
ting load (elevator cycle absorption), clearly shows the effect of the additive on the grinding cycle.
The yellow line represents the elevator consumption while the red line represents the fresh mate-
rial feed. Subsequently the circulating load has been returned to previous levels by the following
operations:

• Mill output increase

• Cement fineness improvement
54 Table N°1.
Details Units Blank MA.G.A./C2 % Difference
Cement Type CEM I 42.5 R CEM I 42.5 R ---
Additive Type Blank MA.G.A./C2 ---
Additive Dosage g/t -- 450 ---
Production t/h 82.1 90.2 + 9.9 %
Specific surface m2/kg 408 410 + 0.5 %
area
Specific surface km2/h 33.5 37.9 + 13.0 %
area production
Specific kWh/t 51.7 47.3 - 8.5 %
mill consumption
Specific kWh*t-1/ km2 126.6 112.6 - 11.1 %
consumption
surface area
production
Flow mm 73 83 + 13.7 %
Initial Setting min. 95 80 - 16.7 %
Time
Strengths 1 days MPa 15.6 20.1 + 29.4 %
2 days MPa 25.2 28.4 + 13.0 %
28 days MPa 49.8 55.6 + 11.7 %
Table N°1 in paragraph 1.4 shows the average hourly production obtained during the trial. These 55
results confirm the validity of the MA.G.A./C2 product. The mill output increase t/h: + 9,9%, has
been realised with a slightly higher Blaine specific surface: 410 m2/kg versus 408 m2/kg; a fur-
ther result is a higher surface production: + 13% (37.9 km2/h versus 33.5 km2/h).

The results show a reduction of the average specific mill consumption from 51,7 to 47,3 kWh/t.

b) Post-Trial analysis

The results are available in Table N°1. All testing was carried out in accordance with the European
Standard EN 196.

Table N°1 shows information regarding cement strengths, Blaine, setting times and cement flow.
The cement produced without any additive, i.e. Blank, is compared to that produced with
MA.G.A/C2.

As regards the strengths there are increases at all ages with MA.G.A/C2. At 1 and 2 days the
increases are respectively +29,4% and 13,0%. There is also a reduction in the setting time from
95 minutes to 80 minutes, particularly appreciated by the cement plant as this cement is for the
most part employed by precast clients.
Meanwhile at 28 days the strengths have increased from 49,8 MPa to 55,6 MPa corresponding
to an increase of +11,7%.

Compressive Strengths
Blank MAGA/C 2
60.0
60
60..0
55.6
49.8
50.0
50
50..0

40.0
40
40..0

30.0 28.4
30
30..0
25.2
20.1
20.0
20
20..0 15.6

10.0
10.
10 .0
1 day 2 days 28 days
Days

This increase at 28 days fully meets the clients objectives while at the same time obtaining a
significant production increase.
56 Description Blank MA.G.A./C2 %Variation
Strengths at 28 days 49.8 55.6 +11.7%
(MPa)
Production 82.1 90.2 +9.9%
(t/h)
Energy Consumption 51.7 47.3 -8.5%
(kWh/t)

The comparison between the energy consumption indexes, with reference to the strength cha-
racteristics of the cement produced (kWh*t-1/MPa), clearly demonstrates the improvements which
may be obtained using a high performance product such as MA.G.A/C2.

Description Blank MA.G.A./C2

kWh*t-1/MPa kWh*t-1/MPa
28 days 0.96 0.85

Laser particle size analysis of the cements are available in Fig. 3 (Blank cement) and Fig. 4
(MA.G.A./C2). The laser distribution curves demonstrate the superior fineness (higher laser spe-
cific surface) of the cement with the additive even though at the same time significant increases
in production are achieved. The increases in strengths at all ages can be principally accredited to
the positive effect of the MA.G.A./C2 additive.

In Fig. 5 and Fig. 6 the granulometric curves for the two cements produced have been drawn on
the RRB grid using data from the laser analysis, based on the correspondent regression equa-
tion, calculating also the characteristic distribution parameters.

The two distribution lines are quite similar, having the same uniformity coefficient(n) of 0,66 which
also is not particularly high. This could possibly be interpreted as being related to a low or per-
haps insufficient level of air for separation.

3. CONCLUSIONS
The principal objective of the client to increase 28 day strengths on the CEM I 42.5 R has been
fully achieved through the use of the MA.G.A./C2 product , increasing strengths from 49.8 MPa
to 55.5 MPa equivalent to an increase of +11.7%. At the same time a production increase of
+9.9% has been obtained further improving the client’s grinding performance while reducing
energy costs.

The MA.G.A./C2 product is an effective solution where ultimate strength increases are specifically
required on Portland and on blended cements where there is a significant limestone addition.
Fig. 3 Blank cement 57
58 Fig. 4 MA.G.A./C2
Fig. 5 59

Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type CEM I 42.5 R

Mill Cement Mill Production t/h
Separator Sepol SVZ Spec.surf. M2/kg
Test no. blaine 408
Description Blank, June 2004 laser 305

dim.µm pass. (%)

1 3.4 30 µm 40 µm 60 µm 90 µm 200 µm
2 7.7 Residue (%) 34.5 24.6 11.1 5.7 0.0
4 14.8
6 21.4 Linear Regression Y=a+nx
8 27.4
10 32.8 Uniformity Coe (n) 0.6606 Line coordinates
15 44.1 Cross (a) -2.2878 X(µm) Y(% R)
20 52.8 Line Angle (Gr) 33.5 2 85.2 14.8
30 65.5 Charact. Diam. (Xo) Res. 36,8% 32 µm 90 13.8 86.2
40 75.4 Corrsp. Diam. P80 66 µm 32 36.8 63.2
60 88.9
90 94.3

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

60 Fig. 6
Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type CEM I 42.5 R

Mill Cement Mill Production t/h
Separator Sepol SVZ Spec.surf. M2/kg
Test no. blaine 410
Description MA.G.A./C 2, June 2004 laser 344

dim.µm pass. (%)

1 4.1 30 µm 40 µm 60 µm 90 µm 200 µm
2 9.1 Residue (%) 30.2 20.0 8.3 4.1 0.0
4 17.3
6 24.4 Linear Regression Y=a+nx
8 30.6
10 36.1 Uniformity Coe (n) 0.6595 Line coordinates
15 47.3 Cross (a) -2.1600 X(µm) Y(% R)
20 56.2 Line Angle (Gr) 33.4 2 83.3 16.7
30 69.8 Charact. Diam. (Xo) Res. 36,8% 26 µm 90 10.6 89.4
40 80.0 Corrsp. Diam. P80 54 µm 26 36.8 63.2
60 91.7
90 95.9

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

 . / M
.A
A.G
M
MApei Grinding Aid/M
Pack set inhibitors
for minerals

DESCRIPTION Limestone
MA.G.A./M are high performance grinding aids for Grinding limestone is always difficult due to the
minerals (limestone, quartz, feldspar, zirconium, pack-set phenomena inside the mill. For this reason
sands) and raw materials and are suitable for a highly effective grinding aid is recommended.
obtaining a higher level of fineness.
MA.G.A./M offers specific additives for:
They increase mill output, and modify the particle • limestone for the food industry;
size distribution, improving the dry flow
characteristics of the grinded materials. • limestone for applications with controlled dielectric
constant (power line fillers, etc.).
They are used in dry grinding processes, preferably
in tubular mills, but are also effective in vertical mills, Other minerals
hammers mills, etc. Other minerals such as quartz, feldspar, sands etc.
are usually used in the ceramic industry (tiles,
TECHNICAL CHARACTERISTICS sanitary ware), glass and refractory products
MA.G.A./M, thanks to their polar nature, notably industry. For these minerals with a low agglomeration
reduce the attraction forces of cement particles tendency, specific formulations of MA.G.A./M are
which are the main cause of agglomeration inside available.
the mill. MA.G.A./M act by coating the particles
which normally cause agglomeration, with a CHEMICAL-PHYSICAL DATA
mono-molecular film which neutralizes the surface Please refer to the appropriate safety data sheets.
electrical charges, improving dry flow characteristics
during transport, storage and handling. DOSAGE
0.3-0.8 kg/t.
APPLICATION PROCEDURE The optimum dosage depends on the grinding
MA.G.A./M has been conceived taking into account system and material type and fineness. In any case it
the characteristics of the different minerals and their has to be found through a reliable industrial trial,
final use. A first distinction can be made between preferably with the help of MAPEI Cement Additives
additives for the grinding of limestone and additives Division technicians, that are also available to select
for the grinding of other minerals. the most suitable additive for each specific need.
 . / S
P.E
A.
M
MApei Performance
Enhancer/S
Pack set inhibitors and
strength enhancers

DESCRIPTION goals placed by the cement factory. The obtainable

MA.P.E./S are highly concentrated grinding aids strength increases may be used to improve the
formulated with only selected raw materials, to binder quality, reaching early strengths quite similar
guarantee absolute constancy of quality and superior to those of Portland cements.
performance. Alternatively, keeping the cement quality unaltered,
it is possible to substitute in the mixture up to 4-6%
They are additives formulated to improve cement of clinker with blended material.
quality (early and/or ultimate strengths) and to aid
cement grinding by increasing mill production. In normal conditions the strength increases are in
the range of 20-50% after one day and 5-15% after
TECHNICAL CHARACTERISTICS 28 days.
MA.P.E./S also guarantee, in addition to all the MA.P.E./S can also be used successfully when
advantages which come from the usage of the grinding Portland and limestone cements to increase
grinding aids (refer to MA.G.A./C), remarkable significantly early strengths.
increases to early and ultimate strengths.
CHEMICAL-PHYSICAL DATA
At equal cement fineness, MA.P.E./S are able to Please refer to the appropriate safety data sheets.
increase mechanical strengths thanks to a better
granulometric redistribution of the finished cement, DOSAGE
to a higher fineness and to a higher rate of the 1.0-3.0 kg/t.
calcium silicates hydration, which can be significantly We suggest the higher dosage threshold, if the aim
modified. is the substitution of clinker points with blended
material (i.e. blast-furnace slag, fly ash, pozzolan).
APPLICATION PROCEDURE
MA.P.E./S may be successfully utilized in the The optimum dosage, in any case, has to be
grinding of blended cements (i.e. pozzolanic, blast found through a reliable industrial trial, preferably
furnace slag and fly ash cements) and in all cases with the help of MAPEI Cement Additives Division
where a significant increase in early strengths is technicians.
needed.
MA.P.E./S should be added to the clinker on the
MA.P.E./S are generally formulated to reach the mill feed conveyor belt or sprayed in the first mill
 65

INDUSTRIAL TRIAL WITH MA.P.E./S

Report of an industrial trial performed
with MA.P.E./S during the preparation
of CEM I 52.5
t

1. TRIAL DESCRIPTION
a. Plant description
b. Trial objectives and additive selection

2. TRIAL RESULTS
t

a. Immediate results during the Trial

b. Post-trial results
c. Technical analysis
t

3. CONCLUSIONS
66 1. TRIAL DESCRIPTION
a) Plant Description

In the cement plant of reference, the clinker is produced in a rotary kiln with cyclone preheaters
produced by PSP Eng., with a potentiality of 2500 t/d. The cements produced in the plant are:
CEM I 52.5 R, CEM II/AL 42.5 R, CEM IV/B 32.5 ARS.

The industrial trial has been performed on a grinding system Wyss 3.2 x 14 (see Fig. 1) during the
production of the CEM I 52.5 R.

Fig. 1
GRINDING PLANT CHARACTERISTICS
MILL INSTALLED POWER 1850 Kw
CHAMBER DIMENSIONS Øut (m) lg ut (m) Vut (m3)
Drying chamber REDUCTION GEAR brand FLS
1^chamber 3.00 4.20 30.40 type Symetro
2^chamber 3.00 9.10 65.30 Installed power kW 1850
3^chamber rpm (n1/n2) n/1' 990/16.3
Total values 6.0 13.3 95.7 rotation speed n/1' 16.3
BALL CHARGE 1^cham 2^cham 3^cham Critical speed % 67.1
100 ø t Loading device type channel
90 ø t diameters [in/out] m
80 ø t 15 CYCLE ELEVATOR type
70 ø t 10 capacity t/h 150
60 ø t 10 STATIC SEPARATOR type
50 ø t 5 10 diametre m
40 ø t 20 SEPARATOR brand Wedag
30 ø t 50 type ZUB 32
25 ø t diametre m 3.2
20 ø t Q air Em
/h 50,000
25 ø x 25 Lg t FILTERS mill
22 ø x 22 Lg t type sleeve/tissue
20 ø x 20 Lg t model Pulse Jet
16 ø x 16 Lg t chambers no.
1^cham 2^cham 3^cham filtering surface m
Total t 40 80 FILTER FAN mill
Filling Level % 29.9 26.6 capacity Em
/h 37,600
Mill Total t 120 pressure mbar 36
LININGS head 1^cham 2^cham 3^cham temperature °C 80
type wave classifying FEEDING MEASURES No.
blocking system type
material REGULATION SYSTEM
cycle elevator kW measuring yes
DIAPHRAGM contr.gravim.sep.refusal
type double discharge electronic ear yes
slots mm 8 10 CEMENT TRANSPORT
material tipo type pneumatic/conveyor
ø central hole mm capacity t/h
regulable no
position
 Fig. 1 bis Flow sheet “Cement grinding” - cement mill no. 2 Wyss 67

cycle
clinker limestone gypsum pozzolan fly ash elevator

channel

pre-separator
dosing compressor
system dynamic
separator
filter

additive
cotto-2
VISS

cement
sampling

b) Trial Objectives and additive selection

A “strength enhancer” has been added to the CEM I 52.5 R type cement, at a dosage of about
2.5-3.0 kg/t. When formulating our offer, we also proposed an enhancer, the MA.P.E./S, that
offers, in addition, good grinding aid characteristics.
We added our MA.P.E./S in the plant during a normal production cycle with the addition of the com-
petitor’s product, without intervals, at the same dosage. The diagram in Figure 2, related to the cir-
culating charge (cycle elevator absorption), clearly shows the effect of the additive on the grinding
cycle. Subsequently the circulating charge has been re-settled by the following operations:

• Cement fineness improvement

• Mill output increase

Fig. 2 Loading charge

Introduction of MA.P.E./S
68 2. TRIAL RESULTS
Additive Type COMPETITOR MA.P.E./S DIFFERENCE
Dosage Kg/t 3.0 2.8 0.93
Production t/h 20.0 22.2 1.11
Specific surface m2/kg 490 504 1.03
Spec. Surf. Produc. km2/h 9.8 11.1 1.13
Mill consumption kW 1346 1350 1.00
Specific mill cons. kWh/t 67.3 60.8 0.90
Specif. surface kWh/km2 137.3 120.6 0.87
production cons.

Flow mm 72 71 0.99
Mixing water % 28.5 28.5 1.00
Start hardening min 270 250 0.92

Strengths 1d MPa 29.5 31.7 1.07

2d 41.7 44.3 1.06
28 d 52.2 55.1 1.05

a) Immediate results during the trial

The above results confirm the validity of our MA.P.E./S in comparison to the competitor additive.
The mill output increase t/h: + 11%, has been realised with a higher Blaine specific surface: 504
m2/kg versus 490 m2/kg; a further result is a higher surface production: + 13% (11.8 km2/h ver-
sus 9.8 km2/h); in practice this means that, at the same specific surface of the competitor’s addi-
tive: 490 m2/kg, a production of 24 t/h could have been realised, with a final increase of 20%.

b) Post-Trial Analysis

The physical-mechanical characteristics of the cement improved thanks to Mapei additive. At

higher strengths value at 1 and 2 days (+7 e + 6%) respectively, corresponds also better harde-
ning times if compared to the competitor’s product: 250/1’ vs. 270/1’.
It is also to be noted that the problem of the cement hardening is, in this case, particularly per-
ceived, and the competitor's additive has been previously modified due to its tendency to exces-
sively protract the hardening times.

c) Technical Analysis

During the trial with the MAPEI additive, some samples of the materials in the system have been
collected:
• separator final product;
• mill exit (separator feeding);
• separator coarse.
On the above samples we performed the granulometric analysis with the Alpine Siever, and the
complete distribution curve has been determined by the laser Coulter LS.
Wedag separator
69
The separator working, in the plant closed circuit, has been evaluated with the usual analysis
methods:
• Performance based on the DIN rule;
• Tromp separation curve.

Tables 7 e 8 report the performance calculation based on DIN, related to the granulometric deter-
minations with the Alpine siever; Table 8, in particular, analyses the “cleaning” function of the
separator: quantity of fine material extracted from the feeding, referred to a 40 µm separation dia-
meter.
From the granulometric laser analysis of the loading charge, the T separation curve (foll. Tromp)
has been traced on a scheme with linear coordinates: table 9. Then, the characteristic parame-
ters of the curve following CETIC have been calculated: soutirage and curve inclination.

3. CONCLUSIONS
The strengths improvements obtained with the MA.P.E./S, if compared to the competitor’s pro-
duct, did not change at different aging (1-28 days).
Obviously also the comparison between the energy consumption indexes, referred to the
strengths characteristic of the cement produced (kWh/Mpa), demonstrates the better performan-
ce of MAPEI additive:

COMPETITOR (kWh/MPa) MA.P.E./S (kWh/MPa)

1 day 2.28 1.91
1, 3, 28 days 0.54 0.46

The results of the Separator Performance Analysis have been summarised in TAB. 3.
The granulometric curves of the 3 samples have been drawn on the RRB reticule
(Fig. 4,5,6) based on the correspondent regression equation, calculating also the characteristic
distribution parameters:
• uniformity coefficient n (curve inclination)
• characteristic diameter Xo (diameter correspondent to R 36.8).
For the final product the uniformity coefficient was 0.96 (line pending of 44 degrees); this is a
mean/high value, that is characteristic of the 2nd generation separators, when producing cements
that need water additions on “normal” values.
70 Tab. 4 Regression line equation

Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type CEM I 52.5 R

Mill Wyss 3.2 x 14.0 Sample Separator Final Product
Separator Wedag ZUB 32 Spec.surf. M2/kg
Test no. blaine n.d.
Description Additive MA.P.E./S 5 laser 492

dim.µm pass. (%)

2 15.1 30 µm 40 µm 60 µm 90 µm 200 µm
4 23.3 Residue (%) 17.7 8.8 2.3 0.3 0.0
6 29.7
8 35.4 Linear Regression Y=a+nx
10 40.9
15 53.9 Uniformity Coe (n) 0.9653 Line coordinates
20 65.4 Cross (a) -2.7215 X(µm) Y(% R)
30 82.3 Line Angle (Gr) 44.0 2 87.9 12.1
40 91.2 Charact. Diam. (Xo) Res. 36,8% 17 µm 90 0.6 99.4
60 97.7 Corrsp. Diam. P80 27 µm 17 36.8 63.2
90 99.7

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

Tab. 5 Regression line equation 71

Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type CEM I 52.5 R

Mill Wyss 3.2 x 14.0 Sample Separator Coarse Product
Separator Wedag ZUB 32 Spec.surf. M2/kg
Test no. blaine n.d.
Description Additive MA.P.E./S 5 laser 246

dim.µm pass. (%)

2 7.1 30 µm 40 µm 60 µm 90 µm 200 µm
4 10.9 Residue (%) 57.7 46.1 30.0 16.2
6 13.8
8 16.3 Linear Regression Y=a+nx
10 18.7
15 24.5 Uniformity Coe (n) 0.8521 Line coordinates
20 30.6 Cross (a) -3.4147 X(µm) Y(% R)
30 42.9 Line Angle (Gr) 40.5 2 94.2 5.8
40 53.9 Charact. Diam. (Xo) Res. 36,8% 55 µm 90 21.8 78.2
60 70.0 Corrsp. Diam. P80 96 µm 55 36.8 63.2
90 83.8

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

72 Tab. 6 Regression line equation

Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type CEM I 52.5 R

Mill Wyss 3.2 x 14.0 Sample Separator Feeding
Separator Wedag ZUB 32 Spec.surf. M2/kg
Test no. blaine n.d.
Description Additive MA.P.E./S 5 laser 346

dim.µm pass. (%)

2 10.3 30 µm 40 µm 60 µm 90 µm 200 µm
4 16.1 Residue (%) 40.9 30.9 19.5 9.7
6 20.5
8 24.6 Linear Regression Y=a+nx
10 28.4
15 37.3 Uniformity Coe (n) 0.8218 Line coordinates
20 45.4 Cross (a) -2.9117 X(µm) Y(% R)
30 59.1 Line Angle (Gr) 39.4 2 90.8 9.2
40 69.1 Charact. Diam. (Xo) Res. 36,8% 35 µm 90 11.1 88.9
60 80.5 Corrsp. Diam. P80 62 µm 35 36.8 63.2
90 90.3

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

Tab. 5 Performance based on DIN 73

Separation Performance Analysis according to DIN

Material: CEMENT
Mill: Wyss 3.2 x 14.0

Separator Brand: Wedag

Type: ZUB 32

Separation Performance According to DIN

TABLE 2 [passing values] Output calculation

µm final p. sep. feed return from TABLE 2 a (D) µm
(D) (f) (a) (g)
20 F= t/h 22.2
30 A= t/h 70.4 0.0
40 98.8 86.1 80.4 G= t/h 48.2 0.0
50
60 99.7 88.2 82.8 A= t/h
70 F= t/h 0.0 0.0
90 99.9 95.3 93.1 G= t/h 0.0 0.0
200 100.0 99.7 99.6
Total 398.4 369.3 355.9

(f) (a) (g) control mesh [D]

PERF a(D) 98.8 86.1 80.4 40.0 µm eta w 0.36
format 0.00
mean a (D) µm
values µm 40
CIRCULATING LOAD A/F 3.17 3.23
EFFICIENCY F/A 0.32 0.31
PERF.ACC. DIN ETA w 0.34 0.36 (§) (effective yield/possible)
CLASS.EFFICIENCY ETA k 0.33 (granulometric variation)
DIVISION PERFORM. ETA s 0.34 (fraction separated/theorical)-(NON separated fraction)

NOTES (Rosin-Rammler) (§) intended as "ability" of the

1 U A/F f-g/a-g separator to recover the finished
2 Yield F/A (1/U) product (F) the material (D)
3 Performance ETA w 1/U*f/a present in the feed A.
4 EFFIC.CLASS. ETA k (a-g)/(f-g)*100*(f-a)/a*(100-a)
5 PERF.DIVIS. ETA s a-g/a*100/100-g
74 Tab. 8 Performance based on DIN

Separation Performance Analysis according to DIN

Material: CEMENT
Mill: Wyss 3.2 x 14.0

Separator Brand: Wedag

Type: ZUB 32

Separation Performance According to DIN

Calculation of the values related to [D] 40µm

Output calculation
(f) (a) (g) Dµm
98.8 86.1 80.4 40.0 F= t/h 22.2
A= t/h 71.7
passing residue G= t/h 49.5
t/h t/h
A 61.70 9.96 71.66 A= t/h
F 21.93 0.27 22.20 F= t/h 0.0
G 39.77 9.69 49.46 G= t/h 0.0

% t/h
separated 35.5 21.93 presence in final prod. (F) of [-D], [3]
in the feeding A.
not separated 64.5 39.77 material of [-D], present in A,
that remains in the separator coarse

Separation characteristic parameters Dµm 40

CIRCULATING LOAD A/F 3.23
EFFICIENCY F/A 0.31
PERF.ACC. DIN ETA w 0.36 (effective yield/possible)
CLASS.EFFICIENCY ETA k 0.33 (granulometric variation)
DIVISION PERFORM. ETA s 0.34 (fraction separated/theorical)-(NON
separated fraction)

NOTES (Rosin-Rammler)
1 U A/F f-g/a-g
2 Yield F/A (1/U)
3 Performance ETA w 1/U*f/a
4 EFFIC.CLASS. ETA k (a-g)/(f-g)*100*(f-a)/a*(100-a)
5 PERF.DIVIS. ETA s a-g/a*100/100-g
Tab. 9 Separation analysis 75

Separation Analysis (Tromp Curve)

Cement Plant Cement Type CEM I 52.5 R

Mill Wyss 3.2 x 14.0 Production t/h 22.2
Separator Wedag ZUB 32 Spec.surf. M2/kg
Test no. with MA.P.E./S 5 blaine
Date laser 492

dim.µm f a g df da dg df*F/A dg*G/F A G/A T (%)

2 1.0 15.1 10.3 7.1 15.1 10.3 7.1 6.3 4.14 10.42 0.40 39.71
4 3.0 23.3 16.1 10.9 8.2 5.8 3.8 3.4 2.23 5.64 0.40 39.55
6 5.0 29.7 20.5 13.8 6.4 4.4 2.9 2.7 1.69 4.36 0.39 38.89
8 7.0 35.4 24.6 16.3 5.7 4.1 2.5 2.4 1.46 3.83 0.38 38.12
10 9.0 40.9 28.4 18.7 5.5 3.8 2.4 2.3 1.40 3.69 0.38 38.00
15 12.5 53.9 37.3 24.5 13.0 8.9 5.8 5.4 3.39 8.79 0.39 38.52
20 17.5 65.4 45.4 30.6 11.5 8.1 6.1 4.8 3.56 8.35 0.43 42.69
30 25.0 82.3 59.1 42.9 16.9 13.7 12.3 7.0 7.18 14.21 0.51 50.55
40 35.0 91.2 69.1 53.9 8.9 10.0 11.0 3.7 6.43 10.13 0.63 63.45
60 50.0 97.7 80.5 70.0 6.5 11.4 16.1 2.7 9.40 12.11 0.78 77.67
90 75.0 99.7 90.3 83.8 2.0 9.8 13.8 0.8 8.06 8.89 0.91 90.65
100 95.0 99.8 92.1 86.8 0.1 1.8 3.0 0.0 1.75 1.79 0.98 97.68

total 734.4 573.7 459.3 Materials in cycle residue [determ. ALPINE]

t/h 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.4159 final 22.2 1.2 0.3 0.0
G/F 0.5841 feeding 53.4 13.9 11.8 4.7 0.3
coarse 31.2 19.6 17.2 6.9 0.4
Loading Charge A/F 2.4 Sep. Line angle Deg. 35
Soutirage % 38.8
Ord.Sep.diam % 48

T separation curve

T curve soutirage regression ord.sep.diam. Linear (regression)

 . / W
P.E
A.
M
MApei Performance
Enhancer/W
Pack set inhibitors,
strengths and
flow enhancers

DESCRIPTION goals placed by the cement factory. The achievable

MA.P.E./W are highly concentrated grinding aids strength increment may be used to improve the
formulated with only selected raw materials, to quality of the binder or, keeping cement quality
guarantee absolute constancy of quality and superior unaltered, it is possible to substitute in the mixture
performance. up to 4-6% of clinker with blended material.

They are additives formulated to improve cement In normal conditions the strength increases are in
quality (early and/or ultimate strengths, flow) and to the range of 20-50% after one day and 5-15% after
aid cement grinding, increasing mill production. 28 days.

TECHNICAL CHARACTERISTICS The formulations, which are specifically oriented

MA.P.E./W also guarantee, besides all the towards improving the water demand of the cement,
advantages which come from the usage of the especially used in the production of pozzolanic
grinding aids (refer to MA.G.A./C), remarkable
cements, allow increases to the flow of up to even
increments in early and ultimate strengths and
20 points (UNI standard 7044-72).
an improvement of the workability (flow) of the
cement paste.
CHEMICAL-PHYSICAL DATA
At equal fineness MA.P.E./W are able to increase Please refer to the appropriate safety data sheets.
mechanical strengths thank to a better granulometric
redistribution of the finished cement and to a higher DOSAGE
rate of calcium silicates hydration, which can be 1.0-3.0 kg/t.
significantly modified. We suggest the higher dosage threshold, if the aim
is the substitution of clinker points with blended
APPLICATION PROCEDURE material (i.e. blast-furnace slag, fly ash, pozzolan) or
MA.P.E./W may be successfully utilized in the the reduction of the water demand of the cement.
grinding of blended cements, (i.e. pozzolanic, blast
furnace slag and fly ash cements) and in all cases The optimum dosage has, in any case, to be
where it is necessary to improve the cement flow. found through a reliable industrial trial, preferably
with the help of MAPEI Cement Additives Division
MA.P.E./W are generally formulated to reach the technicians.
 79

INDUSTRIAL TRIAL WITH MA.P.E./W1

Industrial Trial Report on the production
of CEM IV/B-P 32.5 R using the Mapei
grinding additive MA.P.E./W1
t

1. TRIAL DESCRIPTION
a. Plant description
b. Trial objectives and additive selection

2. TRIAL RESULTS
t

a. Immediate results during the Trial

b. Post-trial results
c. Technical analysis
t

3. CONCLUSIONS
80 1. TRIAL DESCRIPTION
a) Plant Description

The grinding circuit is particularly interesting in that it consists of two mills and a mixer.
Pozzolana is dried and ground in a Löesche LM24 vertical mill and is then stored in an interme-
diate silo.

Clinker and gypsum are ground in a Fema tubular ball mill, Ø 3.8 x 13.75m in a closed circuit con-
sisting of a 3° generation Humboldt separator, Ø 3.5 m with 4 cyclones. The gases exiting from
the mill are filtered by means of a bag filter. The grinding additive is added to the first chamber of
the ball mill.

A mixer receives and homogenises the cement (clinker & gypsum) coming from the ball mill and
the pozzolana coming from the intermediate silo, thus producing the finished CEM IV/B-P 32.5 R
cement.

The power absorption of the principle machines are as follows:

Löesche vertical mill: 465 kW

Fema tubular ball mill: 1.990 kW
Humboldt separator: 98 kW

b) Trial Description and Additive Selection

An industrial evaluation of the grinding additive MA.P.E./W1, a liquid product for the production
of Pozzolanic cements specifically formulated with raw materials of the highest quality.

MA.P.E./W1 is an additive with a triple action, developed to increase the workability and mecha-
nical strengths of the ground cement as well as ensuring high mill productivity.

The trial consisted of grinding a CEM IV/B-P 32.5 R cement at the same level of fineness, with
and without the addition of MA.P.E./W1, verifying immediately the effects on mill productivity and
successively on workability, mechanical strengths and separation performance.
2. TRIAL RESULTS 81

Table A

Details Units Blank MA.P.E./W1

Cement Type CEM IV/B-P 32.5 R CEM IV/B-P 32. 5 R
Additive dosage g/t --- 2.700
Production t/h 83.9 93.8
Passing material at 40 µm. % 80 81.5
Passing material at 63 µm. % 94 95
Specific Mill & Separator Consumption kWh/t 30.4 27.2
Workability Flow 36 48
Strengths at 24 hours MPa 8.0 10.6
Strengths at 2 days MPa 16.3 18.5
Strengths at 28 days MPa 34.9 40.3

a) Immediate results during the trial

On examination of Table A, it is possible to see that the addition of MA.P.E./W1 has produced an
increase in productivity of 11.8% with relevant energy savings. This confirms the product’s valid
grinding aid characteristics.

Industrial T
Trial
rial Data
Blank MAPE/W 1
100
100..0 93.8
83.9
80
80..0

60
60..0

40
40..0
30.4
27.2
20
20..0

0.
0.0
Productivity (t/h) Energy Consumption
(kWh/t)
 b) Post Trial Analysis
82
Strengths

Strength testing was conducted in accordance with the European standard EN 196-1.
The utilisation of the additive increases the strengths of the cement as follows:
+ 32.5% at 24 hours, + 13.5% at 2 days and + 15.5% at 28 days.

Compressive Strengths
Blank MAPE/W 1
45
45..0
40.3
40.
40 .0
34.9
35
35..0
30
30..0
25
25..0
18.5
20
20..0 16.3
15
15..0
8.0 10.6
10
10..0
5.0
5.
0.
0.0
1 day 2 days 28 days

Workability

The utilisation of the additive increases the workability by 12 points equivalent to 33.3%.

Cement Flow
Blank MAPE/W 1
60.0
60.

48.0
50.0
50.

40.0
40.
36.0

30.0
30.

20.
20.0
Flow
c) Technical Analysis
83

During the industrial trial samples of the following materials were taken:

Separator feed
Separator finished product
Recycle from separator
Finished cement

The samples were subjected to laser and Alpine granulometric particle distribution analysis.
In Table B the results of the separator yield, the Tromp curve and the R.R.B. regression line equa-
tion are reported.

Table B

Details Units Blank MA.P.E./W1

Cement Type CEM IV/A 32.5 R CEM IV/A 32. 5 R
Additive dosage g/t --- 2,700
Circulating load A/F 4.1 3.0
Soutirage (by-pass) % 41.5 18.8
Diameter of separation µm 45 37
Inclination of the 2° branch of the curve ° 56 56
Efficiency evaluation Good Good
Angle R.R.B line ° 45.2 45.7
Characteristic diameter (res. 36,8%) µm 29.4 28.4

The results indicate the separator’s (3° generation) good level of separation efficiency in both con-
ditions of operation.
It is interesting to observe how the utilisation of the additive leads to a significant reduction of the
circulating load from 4.1 to 3.0. Mill grinding has improved and the cement particles are less
agglomerated. In addition there is a substantial reduction in the quantity of the material which by-
pass the separator without being classified, passing from 41.5% without an additive (blank) to
18.8% with the additive.
As regards the investigations conducted on the cement, tracing the R.R.B. regression line we do
not find substantial differences except a minimal increase in the slope of the line and a minimal
reduction of the particle distribution curve when the additive is utilised.
 Separation Analysis (Tromp Curve)
84
Cement Plant Cement Type Intermediate
Mill FEMA 3.8 x 13.75 Production t/h 93.8
Separator Humboldt Spec.surf. M2/kg
Test no. blaine 411
Description Blank Grinding laser 342

dim.µm f a g df da dg df*F/A dg*G/F a (§) T (%) T (%) corr

2 1.0 6.9 3.1 2.0 6.9 3.1 2.0 1.69 1.51 3.20 47.2 16.9
4 2.0 13.8 6.2 3.8 6.9 3.1 1.8 1.69 1.36 3.05 44.6 12.8
8 5.0 27.1 11.6 6.7 13.3 5.4 2.9 3.26 2.19 5.45 40.2 5.9
10 7.0 33.4 14.1 7.9 6.3 2.5 1.2 1.54 0.91 2.45 37.0 0.8
15 11.5 47.9 20.0 10.6 14.5 5.9 2.7 3.55 2.04 5.59 36.4 0.0
20 15.0 59.4 25.2 13.5 11.5 5.2 2.9 2.82 2.19 5.01 43.7 11.4
30 22.5 74.1 33.0 19.1 14.7 7.8 5.6 3.60 4.23 7.83 54.0 27.6
40 30.0 85.7 41.6 26.6 11.6 8.6 7.5 2.84 5.66 8.50 66.6 47.4
60 45.0 93.0 55.9 43.6 7.3 14.3 17.0 1.79 12.83 14.62 87.8 80.7
90 65.0 96.8 69.0 62.0 3.8 13.1 18.4 0.93 13.89 14.82 93.7 90.1

total 538.1 279.7 195.8 Materials in cycle spec. Surface area m2/kg residue [determ. ALPINE]
t/h laser blaine 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.245 final 94.0 300 14.5 7.5 1.5 0.0 0.0
G/F 0.755 feeding 383.0 157 58.3 49.2 21.3 10.1 1.0
coarse 289.0 104 78.4 66.7 31.0 14.2 1.0
Loading Charge A/F 4.1 Sep. Line angle Deg. 56
Mayer's Min. Ord. (S) % 36.4 Efficiency Evaluation Good =(Sep. Line angle)
Soutirage % 41.5 Imperfection ad 0.24
Sep.diam µm 45

Tromp Curve according to Mayer

 Separation Analysis (Tromp Curve)
85
Cement Plant Cement Type Intermediate
Mill FEMA 3.8 x 13.75 Production t/h 93.8
Separator Humboldt Spec.surf. M2/kg
Test no. blaine 411
Description Grinding with MA.P.E./W 1 (0.27%) laser 342

dim.µm f a g df da dg df*F/A dg*G/F a (§) T (%) T (%) corr

2 1.0 6.8 2.9 1.1 6.8 2.9 1.1 2.23 0.74 2.97 24.8 12.6
4 2.0 13.3 5.7 1.9 6.5 2.8 0.8 2.14 0.54 2.67 20.1 7.0
8 5.0 24.7 10.1 2.9 11.4 4.4 1.0 3.75 0.67 4.42 15.2 1.3
10 7.0 29.7 11.9 3.3 5.0 1.8 0.4 1.64 0.27 1.91 14.0 0.0
15 11.5 41.3 16.4 4.5 11.6 4.5 1.2 3.81 0.81 4.62 17.4 4.0
20 15.0 51.8 21.0 5.9 10.5 4.6 1.4 3.45 0.94 4.39 21.4 8.6
30 22.5 68.1 29.8 10.7 16.3 8.8 4.8 5.36 3.22 8.58 37.6 27.4
40 30.0 82.4 40.4 19.9 14.3 10.6 9.2 4.70 6.18 10.88 56.8 49.7
60 45.0 93.1 60.7 43.7 10.7 20.3 23.8 3.52 15.98 19.50 82.0 79.0
90 65.0 97.2 77.2 68.5 4.1 16.5 24.8 1.35 16.65 18.00 92.5 91.3

total 508.4 276.1 162.4 Materials in cycle spec. Surface area m2/kg residue [determ. ALPINE]
t/h laser blaine 30 µm 40 µm 60 µm 90 µm 200 µm
calc. F/A 0.329 final 94.0 300 14.5 7.5 1.5 0.0 0.0
G/F 0.671 feeding 285.0 157 58.3 49.2 21.3 10.1 1.0
coarse 192.0 104 78.4 66.7 31.0 14.2 1.0
Loading Charge A/F 3.0 Sep. Line angle Deg. 56
Mayer's Min. Ord. (S) % 14.0 Efficiency Evaluation Good =(Sep. Line angle)
Soutirage % 18.8 Imperfection ad 0.39
Sep.diam µm 37

Tromp Curve according to Mayer

86 Equation of the Regression Curve (RRB Curve) according to Laser analysis

Cement Plant Cement Type Pozzolanic CEM IV/B-P 32.5 R

Mill FEMA 3.8x13.75m; Loesche Ø 2.4m Sample
Separator Humboldt Ø 3.5m Spec.surf. M2/kg
Test no. blaine 419.7
Description Blank Grinding laser 296.7

dim.µm pass. (%) res. (%)

1 2.8 97.2 30 µm 40 µm 60 µm 90 µm 200 µm
2 6.6 93.4 Residue (%) 26.5 20.0 6.0 1.0 0.0
4 13.0 87.0
6 18.9 81.1 Linear Regression Y=a+nx
8 24.3 75.7
10 29.2 70.8 Uniformity Coe (n) 1.0075 Line coordinates
15 40.6 59.4 Cross (a) -3.4078 X(µm) Y(% R) 100-R
20 50.2 49.8 Line Angle (Gr) 45.2 1 96.7 3.3
30 62.3 37.7 Charact. Diam. (Xo) 29.4 µm 29 36.8 63.2
40 73.1 26.9 Res. 36,8% 90 4.6 95.4
60 84.0 16.0
90 94.4 5.6
100 97.3 2.7

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

 Equation of the Regression Curve (RRB Curve) according to Laser analysis 87
Cement Plant Cement Type Pozzolanic CEM IV/B-P 32.5 R
Mill FEMA 3.8x13.75m; Loesche Ø 2.4m Sample
Separator Humboldt Ø 3.5m Spec.surf. M2/kg
Test no. blaine 419.7
Description Grinding with MA.P.E./W 1 (0.27%) laser 296.7

dim.µm pass. (%) res. (%)

1 2.8 97.2 30 µm 40 µm 60 µm 90 µm 200 µm
2 6.7 93.3 Residue (%) 25.0 18.5 5.0 1.0 0.0
4 13.0 87.0
6 18.9 81.1 Linear Regression Y=a+nx
8 24.5 75.5
10 29.6 70.4 Uniformity Coe (n) 1.0240 Line coordinates
15 41.5 58.5 Cross (a) -3.4357 X(µm) Y(% R) 100-R
20 51.5 48.5 Line Angle (Gr) 45.7 1 96.8 3.2
30 64.7 35.3 Charact. Diam. (Xo) 28.4 µm 28 36.8 63.2
40 76.0 24.0 Res. 36,8% 90 3.8 96.2
60 86.5 13.5
90 95.4 4.6
100 97.5 2.5

Notes

GRANULOMETRIC CURVE ACCORDING TO RRB

88 3. CONCLUSIONS
The results obtained from the industrial trial have confirmed the triple action of the additive
MA.P.E./W1 and continuous utilisation of this product gives the following benefits:

• Increases in productivity, especially in the periods when energy costs are lower ensuring signi-
ficant energy savings.

• Cement mechanical strength increases.

• An increase in the workability of the cement. The inferior water demand reduces the consum-
ption of cement and helps the synergy with plasticizer additives employed in the production of
concrete.
 . /A
P.E
A.
M
MApei Performance
Enhancer/A
Grinding aids for
masonry cement

DESCRIPTION improving the workability of the product. Water

MA.P.E./A are air-entraining agents formulated for retention can easily exceed 95% (based on ASTM
grinding of artificial and natural hydraulic masonry standards).
cements.
Air is entrained in micro-bubbles, homogeneously
Artificial hydraulic masonry cements are obtained distributed, improving workability, yield per surface
by the grinding of clinker (15/40%) together with unit and resistance to freeze-thaw cycles. Micro-
one or more inert materials (limestone or siliceous). bubbles, with controlled diameter and high stability,
Natural hydraulic masonry cements are obtained by act as a lubricant between mortar layers, improving
firing natural marls at +900°C, and are characterized flow and workability.
by low mechanical strengths, but have a pasty
consistency, high workability and adhesiveness. Water retention prevents the mortar mixing water
from migrating towards the external substrate,
MA.P.E./A confer to artificial hydraulic masonry improve adhesiveness and helps avoid crack
cements characteristics similar to those of the formation. Regular water content in the mortar allows
natural ones (workability, adhesiveness, resistance a homogeneous and controlled hardening.
to frost-thaw cycles, etc). MA.P.E./A can also
further improve the performance of natural hydraulic APPLICATION PROCEDURE
masonry cements. Specific formulations of MA.P.E./A are formulated to improve the
MA.P.E./A also allow mill output increases. characteristics of natural and artificial hydraulic
masonry cements. They are to be added to the mill
Masonry cement evaluation is principally based on during the grinding phase for a correct dispersion
air entrapment (>10%) and water retention (> 85%) and to maximize the performance.
potential.
CHEMICAL-PHYSICAL DATA
These characteristics guarantee better workability, Please refer to the appropriate safety data sheets.
yield and durability of the cement.
DOSAGE
TECHNICAL CHARACTERISTICS 0.6-1.0 kg/t.
MA.P.E./A have both air entraining and water The optimum dosage depends on the type and
retention actions. Air entraining can reach 15 to 18% fineness of the masonry cement.
 0 5
/Cr
.E.
A.P
M
MApei Performance
Enhancer/Cr 05
Cr(VI) Reducing Additive

DESCRIPTION The additive must be used only during grinding,

MA.P.E./Cr 05 is a chromate reducing agent, added on to the clinker conveyor belt or directly in
formulated with selected raw materials, ensuring first chamber. In any case it should be added as
an elevated and constant quality. close as possible to the mill entrance, avoiding long
and complicated distances which could cause a loss
MA.P.E./Cr 05 is a liquid additive which has to of additive before mill entrance.
be added to the mill during the cement grinding
process. It ensures the production of cements It is recommended to avoid mixing MA.P.E./Cr 05
without Cr(VI), according to European obligation with grinding aids normally employed in cement
2003/53/EC. factories.

TECHNICAL CHARACTERISTICS Under normal cement storage conditions and

MA.P.E./Cr 05 guarantees, besides advantages applying a dosage based on grinding plant
coming from the usage of a liquid product, an conditions, the reducing action shall remain
efficient and stable reduction of soluble chromates unchanged for at least six months.
in cement and hydraulic binders.
CHEMICAL-PHYSICAL DATA
It is an absolutely innovative liquid product, based Please refer to the appropriate safety data sheets.
on antimony*, dispersed on to clinker and other raw The dosage depends on Cr(VI) cement content and
materials during the grinding of cement. During plant type. Generally speaking we suggest a dosage
cement hydration the reducing agent acts on soluble between 50 and 55 g/t for each ppm of Cr(VI)
chromates, transforming Cr(VI) into Cr(III). existing in cement.
The optimum dosage, in any case, has to be found
In comparison with traditional Cr(VI) reducing through a reliable industrial trial, preferably with the
additives, MA.P.E./Cr 05 has an alkaline pH. help of MAPEI Cement Additives Division
technicians.
APPLICATIONS
MA.P.E./Cr 05 can be successfully employed in DOSAGE
cement or hydraulic binders and in general wherever MA.P.E./Cr 05 can be dosed by means
it is necessary to reduce Cr(VI). of peristaltic or membrane pumps.
 93

MA.P.E./Cr 05
The liquid revolution
t

SCIENTIFIC BACKGROUND
(Previously published in ZKG-
Magazine)

MA.P.E./Cr 05: Industrial case studies

t

and economical considerations

 Thanks to intensive Research, D.A.M. developed a new, liquid and now patented technology for
94
the reduction of Cr(VI) in cement, with numerous advantages in terms of technical functionality,
practical application and economic impact.

In fact, MA.P.E./Cr 05 allows cement manufacturers to guarantee their customers extremely long
lasting Cr(VI) reduction (up to 12 months), even in extreme conditions (like cement transport by
sea) and without any negative effects in concrete applications. The advantages for the cement
manufacturers themselves, however, are confirmed in terms of material handling (the product has
a neutral pH), application (dosing a liquid product is much easier than powder compounds) and
costs (the product is to be dosed stoichiometrically which allows it to be competitive in certain
cases, even with Ferrous Sulfate).

SCIENTIFIC BACKGROUND
(Previously published in ZKG-Magazine)
Hexavalent chromium in cement
The raw material for grey Portland cement manufacturing may contain chromium. Due to the highly
oxidizing and alkaline conditions of the kiln, during clinker production chromium is partially con-
verted to hexavalent chromium and probably fixed as alkaline or calcium chromate (Na2CrO4,
K2CrO4, CaCrO4). As a result, Portland clinkers and cements contain soluble chromates (usually in
the range of 5 – 20 ppm or mg/kg, while the total chromium may reach 200 ppm) which are repor-
ted to cause skin irritation (allergic contact dermatitis). This is the reason why the European
Community has introduced the obligation (Directive 2003/53/EC) to maintain the level of soluble
chromates below 2 ppm. This has a significant economical impact on the cement industry.

The reduction of soluble chromates: iron(II) and tin(II) salts

While Cr(VI) is a strong oxidising agent in acid solution, in an alkaline media (such as the cement
mixing water) the situation is completely different and it is impossible to reduce the Cr(VI) with
most of the reducing agents which usually work at pH lower than 7. The reason lies in the fact
that the red-ox potential of the couple Cr(VI)/Cr(III) changes with pH. Using the Nernst equation
it is possible to calculate the value of red-ox potential at different pH and evaluate, from a ther-
modynamic point of view, which red-ox couple can reduce Cr(VI) to Cr(III) [1].
The reduction of soluble chromates is usually obtained with the addition of ferrous or stannous
salts (in powder or in the form of liquid additives) during cement grinding. Both iron and tin form
poorly soluble hydroxides in alkaline media, and this lowers the red-ox potential of the couples
Fe(III)/Fe(II) and Sn(IV)/Sn(II), allowing the reduction of Cr(VI) to Cr(III) according to the following
equations:

1) CrO42- + 3Fe(OH)2 + 4H2O = Cr(OH)3 + 3Fe(OH)3 + 2OH-

2) 2CrO42- + 3Sn(OH)2 + 2OH- + 8H2O = 2Cr(OH)3 + 3Sn(OH)62-

The mechanism of action of ferrous and stannous salts can be considered as follows:
• as soon as the cement (ground with the reducing agent) is mixed with water, chromates and
ferrous/stannous salts are released in solution, while the pH quickly increases following the
hydration of cement;
• Fe2+ and Sn2+ ions form insoluble hydroxides, their red-ox potential drop (in particular as the pH
increases, their red-ox potential drop faster than the red-ox potential of Cr6+) and the Fe(OH)2 and
95
Sn(OH)2 become strong reducing agents;
• soluble chromates are reduced to Cr(OH)3.

In Table 1 the red-ox potential of Cr(VI)/Cr(III), Fe(III)/Fe(II) and Sn(IV)/Sn(II) in alkaline solution are
reported [2]: the lower is the red-ox potential, the higher is the reducing power of the couple.
Both ferrous and stannous salts present advantages and disadvantages.
Ferrous sulphate (available in different hydrated forms) is very cheap, but presents serious pro-
blems related to the durability of the reducing properties: it is very sensitive to moisture and tem-
perature and tends to loose efficacy after grinding and during cement storage. This requires the
use of very high dosages of ferrous sulphate, with costs higher than expected and undesired
effects (spots on the concrete surfaces, due to the red colour of Fe3+ compounds, are reported).
In our opinion, the poor durability of iron(II) reducing agents can be explained considering the acid
character of iron and the presence of water of crystallisation. We can suppose that ferrous sul-
phate may react during cement grinding (or storage) with very alkaline free lime, being partially
converted to ferrous hydroxide according to the following acid-base reaction:

3) FeSO4·7H2O + CaO = Fe(OH)2 + CaSO4 + 6H2O

If ferrous sulphate is converted before having soluble chromates available for reduction, due to
the strong reducing properties, ferrous hydroxide is readily oxidised by oxygen and looses it’s
reduction ability.

Red-ox couple Half-reaction Red-ox potential

in alkaline solution E (Volt)
Cr(VI)/Cr(III) CrO42- + 3e- + 4H2O = Cr(OH)3 + 5OH- -0.12
Fe(III)/Fe(II) Fe(OH)3 + e- = Fe(OH)2 + OH- -0.56
Sn(IV)/Sn(II) Sn(OH)62- + 2e- = Sn(OH)2 + 4OH- -0.96
Table 1

Stannous sulphate has a superior reduction capacity (allowing very low dosages) and no undesi-
red effects, but is very expensive. Recently it has been reported [3] that in presence of high
amounts of free lime and moisture, stannous compounds partially loose their reduction ability. This
is more evident with liquid additives based on tin(II) compounds (stannous chloride or sulphate), in
particular with stannous sulphate aqueous solutions, as reported elsewhere [4]. The reason, as
explained for iron(II), can lie in the fact that tin(II) has strong acid properties, and during cement
grinding it can react with free lime and water being partially converted to stannous hydroxide:

1) SnSO4 + CaO + H2O = Sn(OH)2 + CaSO4

2) SnCl2 + CaO + H2O = Sn(OH)2 + CaCl2

The stannous hydroxide, due to its very low red-ox potential, is unstable and, if no soluble chro-
mates are present, is immediately oxidised by oxygen or water or spontaneously converts to
 tin(IV) and metallic tin. This behaviour is summarized in the Pourbaix diagram of tin [5], that
96
reports the stability range of several tin compounds in function of pH and red-ox potential (Figure
1). It is possible to see that, as the pH increases, the stability area of Sn(OH)2 is reduced and,
above pH=12, only Sn4+ and metallic tin are stable. As a result, in some cases higher tin dosages
are required, with negative effects on costs and durability of reducing properties. The reactions 4
and 5, in order to proceed, require the presence of water. This may be the reason why stannous
sulphate in powder is more efficient than water-based liquid formulations containing tin salts.

E (volt) 2.0

1.0

0.0

-1.0

-2.0

Figure 1: Pourbaix diagram of tin. The red area represents the pH range of cement
mixing water. The variation of red-ox potential of oxygen and hydrogen are pointed out

A brand new technology: antimony(III) compounds

A very promising and innovative class of reducing agents (at the moment object of an internatio-
nal patent application by Mapei SpA) has recently been studied and developed in Mapei R&D lab.
The efficacy and the superior performances of this novel class is based on the red-ox properties
of antimony(III). The couple Sb(V)/Sb(III), has a red-ox potential in alkaline solution E=-0,59 volt [2].
From a thermodynamic point of view, this means that Sb(III) is a strong reducing agent at high pH
and can reduce Cr(VI) present in the cement mixing water, according to the following equation:

6) 2CrO42- + 3H2SbO3- + 2H2O = 2Cr(OH)3 + 3SbO3- + 4OH-

The Pourbaix diagram of antimony [5], reported in Figure 2, shows that the Sb(III) is stable at alka-
line pH. In comparison to ferrous and stannous salts, Sb(III) compounds have weaker acid pro-
perties. This is an interesting advantage, because the reaction with free lime does not proceed,
avoiding any efficacy loss during cement grinding or storage, even in case of high free lime con-
97
tent and high level of humidity.
As a result, the reduction performance of antimony(III) is unaffected by moisture and high grin-
ding or storing temperatures. The reducing properties of antimony(III) remain unchanged even
after more than one year.

E (volt) 1.0

0.5

0.0

-0.5

-1.0

Figure 2: Pourbaix diagram of antimony. The red area represents the pH range of
cement mixing water. The variation of red-ox potential of oxygen and hydrogen are
pointed out

Antimony(III) in a liquid additive: MA.P.E./Cr 05

The formulation of a liquid additive based on antimony(III) for the reduction of hexavalent chro-
mium requires the selection of the most appropriate Sb(III) compound.
This should be:

• easy to incorporate in a water-based formulation;

• with an economical impact inferior to tin-based liquid additives;
• no effect on the properties and quality of the cement.

Most of all, in order to maintain it’s reduction ability safe for as long as possible, the reducing
agent should remain unaltered during the storage of the additive and, after being added in the
mill, during the storage of cement. Strong reducing agents (like ferrous or stannous salts) are rea-
dily oxidised by oxygen (or directly by water, as described for stannous hydroxide), and high tem-
peratures and high levels of humidity (e.g. presence of coordination water, as in the case of fer-
 rous sulphate) speed up the reaction. If it was possible to activate the reducing agent only when
98
cement is mixed with water (when soluble chromates are available for reduction), we would obtain
an excellent improvement of shelf life and durability after prolonged storage.
We have found that it is possible to reach all the targets using a liquid additive based on antimo-
ny trioxide. This compound is amphoteric: it is soluble only at very low or very high pH and is
completely insoluble at medium pH. A liquid additive with high load of insoluble particles of anti-
mony trioxide can be easily prepared by using the well-known technology of solid liquid disper-
sion, widely used in several industrial sectors (ceramic, polymers, textile, paints, paper, cosme-
tics and pharmaceutical, detergents) [6].
The principle of action is now clear: the antimony trioxide is dispersed in to the cement by sim-
ply dosing the liquid additive on to the clinker conveyor belt and ground in the mill. Thanks to its
insolubility and low acidity, it is not modified by water or free lime and it remains unaltered until
cement is mixed with water: at pH higher than 12 the antimony trioxide is dissolved in water and
is fully available to reduce the Cr(VI) released in water.

The advantages of MA.P.E./Cr 05 are the following:

• neutral (or alkaline) pH, while other liquid additives based on tin have strong acid pH and are
highly corrosive.
• No recrystallisation of partially solubilised salts (the active component is completely insoluble)
and consequently no formation of precipitate and difficulties in pumping.
• No reducing agent lost, in any mill conditions (high amount of cooling water, high temperature).
This allows the cement plant to avoid any extra dosage, as usually happens with ferrous sulpha-
te and sometimes with tin-based liquid reducing agents.
• No reducing agent lost during storage: this allows to maintain constant the Cr(VI) content for a
very long time, without requiring an over dosage.

Case study
In order to check the reduction performance of MA.P.E./Cr 05, the following laboratory test has
been performed. The performances of ferrous sulphate, stannous sulphate and MA.P.E./Cr05 (a
liquid suspension of antimony trioxide, with 20% active matter), have been compared. A cement
has been reproduced in a laboratory mill by grinding clinker and gypsum. A clinker with a very
high level of free lime has been chosen (free CaO = 1.78%). The amount of soluble Cr(VI) relea-
sed in water (without reducing agent) is 10 ppm. The same cement has been reproduced by grin-
ding with the reducing agents reported in the table 2.

Reducing agent Dosage (weight % over Dosage (g/t) for each ppm
cement weight) of Cr(VI)
Ferrous sulphate 0.200 % 200 g/t·ppm
(FeSO4·7H2O)
Stannous sulphate 0.020 % 20 g/t·ppm
(SnSO4)
MA.P.E./Cr 05
(20% water suspension 0.045 % 45 g/t·ppm
of Sb2O3)

The samples of cement have been stored in the same conditions and the soluble Cr(VI) content
has been evaluated for a period of six months, using the following method: 100 g of cement are
added to 100 g of water. After magnetic stirring for 15 minutes, the water is filtered off and ana-
99
lyzed by ionic chromatography (see [7] for details). The results are summarized in the graph 1.
It can be clearly seen that with this cement (characterised by a high content of free lime) the stan-
nous sulphate is effective only for a limited period of time: after two months the soluble Cr(VI)
content exceeds the limit of 2 ppm. The ferrous sulphate at a dosage commonly used (0,2%) is
unable to eliminate Cr(VI). The performances of antimony trioxide (MA.P.E./Cr 05) are clearly
superior: the Cr(VI) level is close to zero even after several months.
On an industrial scale, the results are similar. In addition it is possible to see that the economical
incidence of such an additive is lower than the incidence of stannous sulphate and sometimes
can be similar to that of ferrous sulphate.

Graph 1: results of lab test about reducing agents

0.2% Ferrous sulphate

Stannous sulphate

day days days days days month months months months

Conclusions
The use of antimony(III) compounds for the reduction of hexavalent chromium in cement and
cement based materials presents interesting advantages:
1. Due to the high stability and low acidity of antimony(III), these reducing agents are insensitive
to temperature and humidity and are not affected by the presence of high levels of free lime. This
allows superior performances to be obtained in comparison to the usual reducing agents based
on ferrous or stannous salts.
2. Antimony(III) compounds can also be used for the formulation of liquid additives. If antimony
trioxide is dispersed in water (using well known technologies), it is possible to obtain a stable
suspension that can be dosed during cement grinding as a liquid additive. Thanks to its ampho-
teric properties, the antimony trioxide remains unchanged on the surface of cement and is acti-
vated only when the cement is mixed with water and the pH rises above 12. This theoretically
allows to maintain the efficacy of the reducing agent for an infinite time.
3. The MA.P.E./Cr 05, a liquid additive formulated according to this new technology, shows supe-
rior performances in comparison to iron and tin based reducing agents.
100 MA.P.E./Cr 05: Industrial case studies and economical
considerations
Industrial use
In lab test the dosage of MA.P.E./Cr 05 is about 50 g/t for each ppm of Cr(VI) to be reduced.
During industrial trials dosages sometimes are lower, in the range 40 - 45 g/t for each ppm of
Cr(VI). This difference is probably related to a better dispersion of the product in cement, allowed
by industrial milling.
MA.P.E./Cr 05 should be dosed directly on the clinker/feed conveyor belt using a peristaltic or a
piston pump, with a dedicated pipeline. MA.P.E./Cr 05 should be stored avoiding contamination
with different products (e.g. grinding aids, reducing agent based on tin).
MA.P.E./Cr 05 is a very stable suspension, so it doesn’t need additional mixing system to avoid
solid particles sedimentation.
The picture below shows a typical industrial application:

How to use MA.P.E./Cr 05

It is important to point out that, thanks to the long term stability, it is not necessary to reduce
completely the hexavalent chromium present in cement. MA.P.E./Cr 05 can be dosed in order to
reach a Cr(VI) level of about 1 ppm. This value will usually be maintained even after several
months, which allows interesting cost savings.
MA.P.E./Cr 05 has a great Cr(VI) reducing effect on any type of cement, as shown in Table 2.

Table 2: efficacy of MA.P.E./Cr 05 on different cements

Cement type Cr(VI) Dosage Cr(VI) Cr(VI) after 6 months
CEM I 52.5 R 22.3 ppm 1.02 kg/t 1.1 1.1
CEM II/A-LL 42.5 R 10.3 ppm 0.45 kg/t 1.0 1.0
CEM II/B-S 32.5 R 7.6 ppm 0.29 kg/t 0.9 0.9
CEM II/B-M (LL-P) 32.5 R 6.0 ppm 0.23 kg/t 0.8 0.8
CEM III/A 32.5 N 4.4 ppm 0.17 kg/t 1.0 1.0
CEM IV/B-P 32.5 R 5.7 ppm 0.22 kg/t 1.2 1.2
Comparison with stannous sulphate
101
Stannous sulphate is a widespread Cr(VI) reducing agent, available in both powder and liquid
form. Usually, the powder is more difficult to handle and to dose accurately than the liquid pro-
duct. The average dosage of stannous sulphate is 15 g/t for each ppm of Cr(VI) and guarantees
a Cr(VI) reduction for several months. Stannous sulphate prices have increased very much in the
last few months. At the moment we can consider a price of around 8 - 9 €/kg.

Industrial case n°1

A CEM I 52.5 R is produced with 5000 cm2/g Blaine fineness, with a mill productivity of 48 t/h.
The hexavalent chromium content of this cement is about 13 ppm. The reducing agent used is
stannous sulphate (powder form) at a dosage of 15 g/t·ppm. MA.P.E./Cr 05 was tested in com-
parison to stannous sulphate. As summarized in Table 3, it has been possible to dose MA.P.E./Cr 05
at 48 g/t*ppm, allowing a 17% cost saving.
Compared to stannous sulphate MA.P.E./Cr 05 presents the following advantages:

• Technical advantages: easy dosing and pumping

• Economical advantage: cost savings in the cement production process (-17,3%)

Considering the use of stannous sulphate-based liquid additives, it is possible to overcome the
difficulties in handling a powder, but the economical impact is obviously higher, sometimes even
40 – 50% higher than MA.P.E./Cr 05.

Table 3: industrial case - comparison with stannous sulphate

Stannous sulphate MA.P.E./Cr 05
Price (€/kg) 8.5 2.2
Dosage (g/t*ppm of Cr(VI) 15 48
Dosage (g/t) 195 624
Economical impact (€/t) 1.66 1.37
Differences (%) // -17.3%

Comparison with iron sulphate

Iron sulphate is available only in powder form. It is commonly employed in cement plants in spite
of the following technical problems:

• Unstable product: over dosage is necessary in order to guarantee Cr(VI) reduction for at least 3
months.
• Dosing system: difficult and expensive handling.
• Side effects: possible red/dark spots on concrete surface.

The very low price is the only advantage of iron sulphate, even if this is not always true: someti-
mes the economical impact of MA.P.E./Cr 05 has been found lower (or comparable) than iron sul-
phate. The following example describes one of our practical experiences.

Industrial case n°2

An Italian cement plant produces a CEM II/A-LL 42,5 R (80% clinker, 15% limestone, 5% gyp-
sum) with Cr(VI) content of 5 ppm. To reduce and maintain the level of Cr(VI) constant for three
102
months, iron sulphate is used at a dosage of 1.5 kg/t (0.15%) of cement produced. The price of
iron sulphate is 0.14 €/kg (including delivery costs) so its economical impact is 0.21 €/t of cement.
Several dosages of MA.P.E./Cr 05 were tested in order to clarify the economical impact.
Considering that no overdosing is needed, it is not necessary to reduce Cr(VI) completely: the
industrial trial was performed in order to have a final Cr(VI) content of 1 ppm. Graph 1 shows that
MA.P.E./Cr 05 dosed at about 100 g/t has the same economical impact as iron sulphate. Any
dosage below this value means lower economical impact and allows to maintain the level of Cr(VI)
below 2 ppm even for several months.

Graph 1: industrial case - economical impact of MA.P.E./Cr 05 in comparison to iron

sulphate.
Pink line: ppm of Cr(VI) residual.
Blue line: economical impact of MA.P.E./Cr 05 (€/t)
Red line: economical impact of iron sulphate

1.6 0.8

1.4 0.7

1.2 0.6

1.0 0.5

0.8 0.4

0.6 0.3

0.4 0.2

0.2 0.1

0.0 0.0

Conclusions
1. MA.P.E./Cr 05 is a new liquid additive for Cr(VI) reduction in cement and cement based materials.
2. Thanks to the properties of its active ingredient (antimony trioxide), MA.P.E./Cr 05 shows supe-
rior performances in comparison to iron and tin based products.
3. In most cases, the use of MA.P.E./Cr 05 allows lower economical impact (in terms of euro for
each ton of cement produced), with technical performances comparable (or superior) to other
reducing agents.
NOTES

103
104
Cement Grinding Additives Division

Since 2008 DAM is part of Mapei’s Liquid Admixtures Division.

This division has been formed in order to optimise the synergy
between the various stages of the building process of any type
of structure: Cement Manufacturing - Concrete Application -
Terrain Decontamination - Underground Technology. Through
joint projects in terms of Research & Development, Technical
Customer Assistance and Product Development, knowledge is
shared and products and services are improved.

Within the Liquid Admixture Division special attention is paid to

the reduction of CO2 emissions and the limitation of the use of
non-renewable resources. In this specific field of interest, DAM
is offering various specific solutions to the cement industry for
the reduction of both clinker content and the specific energy
consumption during the production of cement.

Mapei Spa, since 1995 has enforced the EMAS is the EU environmental In the year 2000 the Robbiano di
Quality System certified according to management system Mediglia plant obtained the certification
UNI EN ISO 9001. The program also in accordance to of its occupational health and safety
entailed the ISO 9001 certification of many the CE 761/01 regulation management system in accordane
other subsidiaries in the Group with the OHSAS 18001 standard, as
well as the certificate of excellence
Mapei Group’s main productions that attests to its compliance with the
and distribution centres enforce requirements of the ISO 9001,
the Environmental Management System ISO 14001, OHSAS 18001 standards
In accordance with the ISO 14001 standard and of CE 761/01 (EMAS) Regulation
Mapei
Cement Grinding Additives Division Worldwide

Group Headquarters Mapei Italy (2 plants) Mapei Vietnam (1 plant)

Mapei / DAM Italy Strada Provinciale 159 No. 162° Nguyen Chi Thanh street
20060 Robbiano di Mediglia (MI) - Italy Hai Chau District - Da Nang - Vietnam
Via Cafiero 22 and 84100 Salerno - Italy Tel. +84 511 356 5001 / 02 / 03 / 04
20158 Milano - Italy Tel. +39 02 376 73 760
Mapei Construction Materials China (1 plant)
Tel. +39 02 376 73 760
Ibermapei Spain (1 plant) 6, Lane 8999, Hunan Road
Plaza Cataluña 20 - 5a Pianta Nanhui District 201314 - Shanghai - China
Internet: 08002 Barcellona - Spain Tel. +86 21 5818 0808
www.mapei.it/dam Tel. +34 93 343 5050
Mapei Far East Singapore (1 plant)
E-mail: 28 Tuas West Road - 638383 Singapore
[email protected] Lusomapei S.A. Portugal (1 plant)
Tel. +65 686 23 488
Business Parque Tejo XXI
Estrada Nacional 1 - Km 19,65, Gelfas Vinavil Egypt for Chemicals (1 plant)
2600-659 Castanheira do Ribatejo Attaqa Industrial Zone - Piece 175 - Suez - Egypt
Tel. +351 263 860 360 Tel. +20 62 230 501 / 02 / 03

Rescon Mapei Norway (1 plant) I.B.S. (Innovative Building Solutions) L.L.C. (1 plant)
Vallsetvegen 6 Unit 15B - Al Serkal Warehouse
2120 Sagstua - Norway 8th Street, Al Quoz Industrial 1, Dubai
Tel. +47 62 97 2000 P.O. Box 73869, Dubai, UAE
Tel. +971-04-3233167
CEMENT GRINDING
ADDITIVES DIVISION

CEMENT GRINDING ADDITIVES DIVISION

CEMENT GRINDING
ADDITIVES DIVISION

HEAD OFFICE
Mapei SpA
Via Cafiero, 22 - 20158 Milan
Tel. +39 02 37673.1
Fax +39 02 37673.214
Internet: www.mapei.com
E-mail: [email protected]
C.P. MK 661130 - (GB) 06/09