WHITE PAPER

An efficient approach to
dispersing pigments

Nikolina Milanovic and Brian O. Casey,

Evonik Resource Efficiency
 WHITE PAPER:
An efficient approach to
dispersing pigments

ABSTRACT
As problems related to pigment dispersions become more frequent and as requirements regard-
ing color reproducibility become increasingly rigorous, new demands for pigment wetting agents
and dispersants are required to resolve these pigment stabilization issues. Over the past 20 years,
advances in polymeric dispersants have made significant improvements in the process of wetting
and stabilizing pigments. As a result, many dispersant technologies that better stabilize small
pigment particles and that better prevent flocculation are emerging in the marketplace. This paper
will explain pigment and dispersant interfaces and how harmonizing and optimizing these inter-
actions will lead to a better pigment dispersion.

INTRODUCTION
Until the development of synthetic pigments, pigments of natural origin – materials such as clay,
earth colors, and minerals – were used to color paints. Vegetable oils with high lecithin content
were used to wet out natural pigments. Since then, technology has progressed immensely.

Dispersing pigments is sometimes viewed as an art instead of a technical process. This is mainly
due to the multiple steps that occur during the dispersion process. These processes can seem
confusing, but breaking the steps down into a structured manner makes them easier to under-
stand and helps find solutions faster.

The following research explains the science behind dispersing pigments and fillers, in hopes
to reach a better understanding of how to stabilize them more efficiently. This study discusses
different types of wetting and dispersing additives that can be used to stabilize two notoriously
difficult pigments.

PIGMENT AND ADDITIVE INTERACTIONS

Pigments are tiny solid particles that have the ability to refract light [1]. Pigments serve two main
functions: an optical function, providing color, opacity, and gloss; and a protective function, for
the surface beneath the coating.

Structures of pigments
Pigments come in three forms: primary particles, aggregates, and agglomerates. Primary parti-
cles are single, small particles that form during synthesis  [2]. These are the smallest components
in pigments and fillers, and they mostly consist of cuboid, rod, and spherical-shaped particles
(Figures 1–2). During calcination, primary particles may form together to create larger particles
through chemical bonding. These particles are called aggregates, and they are organized lattice
regions connected from face to face of the primary particle. Alternatively, particles that form via
physical bonds instead of chemical bonds are called agglomerates. These particles are connected
by edges, resulting in a smaller surface area and more difficulty in wetting out the pigment.

page 2
 WHITE PAPER:
An efficient approach to
dispersing pigments

Figures 1-2: Transmission electron microscope images of iron oxide pigments

Organic and inorganic pigments

Pigments are widely classified as either organic or inorganic. Inorganic pigments are used for
coloristic properties as well as other properties such as anti-static and anti-corrosion character-
istics. Inorganic pigments normally have a high refractive index, meaning they have a greater
ability to scatter light [3]. Therefore, these pigments are very good at “hiding” the surface
under the coating layer. Oxide compounds such as titanium dioxide and iron oxides are typical
examples of inorganic pigments. Carbon blacks are technically classified as inorganic pigments,
but they require a different anchor group to adsorb onto the pigment surface [4].

Organic pigments are intensely colored, therefore they are incorporated solely for their coloristic
properties. Organic pigments are classified as azopigments, polycyclic pigments, and anthraqui-
none pigments [4]. They tend to have a smaller particle size than inorganic pigments, thus making
them more transparent. They have a tendency to dissolve when moisture is present, which causes
them to migrate and chalk to the surface. Because of their smaller particle size, organic pigments
are normally more difficult to disperse compared to inorganic pigments. Because most inorganic
pigments have polar surfaces, they are much easier to wet out.

Dispersing technology
In order to understand wetting and dispersing agents, a fundamental knowledge of the dispersing
process is required. Dispersing agents stabilize deflocculated pigment particles. For stabilization
of these particles to occur, the dispersant must be able to overcome van der Waals attractions
that are constantly moving pigment particles back together [2]. The pigment dispersion process
can be broken down into three steps: wetting, de-agglomeration, and stabilization.

page 3
 WHITE PAPER:
An efficient approach to
dispersing pigments

Step one: Wetting

The first step of the dispersion process consists of wetting the pigments by a liquid. The liquid
spreads over the pigment surface and fills the voids and pores of the pigment, displacing any
remaining air pockets [2]. For a pigment to be wetted by a liquid, the surface tension of the liquid
must be lower than the surface energy of the pigment [1]. This interaction between pigment and
liquid is described below by the Young equation. A liquid with a low surface tension typically
wets pigments better than one with a higher surface tension.

Ys = Ysl + Yl • cosΘ Equation 11

Equation

Step two: De-agglomeration

After the pigments are wetted, they are broken down to achieve small particle sizes with a
large surface area. This yields a higher color strength, which is more cost-efficient for paint
manufacturers.

In order to grind down to smaller particle sizes, more energy is required. To break up agglomerates
and increase the surface area (ΔA), an increased energy input (ΔW) is required (Equation 2) [1].
This energy is proportional to the surface tension (Y) of the dispersion. The smaller the surface
tension, the greater the surface area will be for a certain amount of energy [3].

ΔW = Y • ΔA Equation 2

When dispersing, agglomerates are broken down into primary particles and small aggregates.
When breaking down agglomerates, only physical bonds are being interrupted.

Figure 3: Typical ranges of energy contents of one mole (6.0231 x 10 ) 23

associated with the different types of chemical and physical bonds [4]

Interactions

Hydrogen
Physical Bonds Bonds Chemical Bonds
<50 kJ/mol

Permanent Induced Dispersive Covalent Ionic Bonds

Dipoles Dipoles Forces Bonds 600 –1000
<30 kJ/mol <2 kJ/mol <40 kJ/mol 60 –700 kJ/mol kJ/mol

These energies range from 40 –50 kJ/mol, meaning 40,000 to 50,000 joules are required to
break these physical bonds [4]. If aggregates are milled, then approximately 600 to 1000 kJ/mol
are required to break these chemical bonds [4]. This is about ten times the energy it takes to
grind agglomerates.

page 4
 WHITE PAPER:
An efficient approach to
dispersing pigments

Step three: Stabilization

Consequently, when there are large surface areas and small pigment particles, the energy is very
high and thermodynamically unstable. Solid particles will always gravitate towards each other in
Brownian motion to minimize their surface area and return to a lower energy state that is more
stable [2] (Figure 4).

Figure 4: Stability of pigment dispersions If these particles are not well stabilized,
then they will flocculate back together.
To achieve good pigment stabilization,
the dispersing agent must be able to
Energy

Influence of
dispersing additive adsorb onto the surface of the pigment.
Therefore, the additive must have anchor
groups with high affinity for the pigment
surface.

The stabilization of pigments can be

Reaction coordinate achieved by electrostatic, steric, or
electrosteric stabilization (Figure 5).

Figure 5: Mechanisms of pigment stabilization [1]

Electrostatic stabilization is most suitable for use in waterborne formulations with high dielectric
charges. The dispersing additive adsorbs onto the pigment surface and dissociates into an anionic
and cationic part. This creates an electric double layer that prevents pigments from flocculating
together via an electrostatic repulsion of like-charges [3].

In contrast to electrostatic stabilization, steric stabilization uses polymeric side chains to keep
pigment particles stable in a dispersion. When pigment particles come closer together, the poly-
meric side chains restrict their movement and decrease entropy [4]. The outcome is a repulsive
force between the two particles. These interactions also restrict the movement of the particles
that give rise to the viscosity.

In some cases, using only electrostatic or only steric stabilization is not enough. Pigment disper-
sions have complex demands, therefore it is sometimes necessary to combine the two stabiliza-
tions to create electrosteric stabilization.

page 5
 WHITE PAPER:
An efficient approach to
dispersing pigments

WETTING AND DISPERSING AGENTS

The terms wetting agent, wetting additive, dispersant, or wetting and dispersing additive are
often used without full understanding of their exact definitions. There are very important differ-
ences in the chemical structures and functions of each.

Wetting agents are low molecular weight, amphiphilic molecules with a hydrophilic and hydro-
phobic segment. They aid in reducing surface tension and wetting out surfaces, but most of the
time, stabilization of pigments is not achieved. Dispersing agents are oligomers or polymers that
aid in the stabilization of pigments and fillers. At the core of every dispersing agent, there must
be a wetting agent in order to facilitate the first step in the dispersion process – wetting out the
pigment. The major difference between these two technologies, is that dispersing agents utilize
anchor groups and polymeric side chains to stabilize pigments.

Polymeric side chains

Polymeric side chains aid in the adsorption of the dispersing agent onto the pigment surface. The
side chains must be soluble in the medium. If they are not soluble, they could collapse onto the
pigment surface causing flocculation.

Additionally, polymers of medium molecular weight are optimal. If the mass is too low, in the
case of wetting agents, then there might not be enough of an effect to stabilize pigments. On the
other hand, if the mass is too large, then it could be incompatible and cause viscosity to rise.

Anchor groups
Anchor groups are placed at the end of the polymeric chains to attach to the surface of the
pigment. Without these anchor groups, the polymeric side chains would not be useful. Specific
chemical groups are used as anchors for certain pigment types.

Dispersants with aromatic rings have an affinity for surfaces of organic pigments. They adsorb
onto the surface by van der Waals forces. Dispersants with hydroxyl-, carbonyl-, or carboxyl-
groups have a high affinity for the surface of inorganic pigments. These adsorb onto the surface
by hydrogen bonding or induced dipole interactions. Amine groups have high affinity for carbon
black surfaces. Without nitrogen, there is not much suitability for carbon black.

EXPERIMENTAL DESIGN
For this study, two of the most difficult pigments were chosen to prepare pigment concentrates:
yellow iron oxide and organic violet (Tables 1-2). Four 100 %-active additives were chosen to be
tested, ranging from 10 % to 30 % additive solids on pigment.

• Additive A: a low molecular weight alkoxylate

• Additive B: a medium-weight polyether with aromatic groups
• Additive C: a medium-weight polyether phosphate with acid groups
• Additive D: a high molecular weight polymer with hyper-branched polyester chains with
aromatic and acid groups

All samples were made in 8-ounce glass jars with 100 grams of material to work with. Glass
beads of size 2.4 –2.9 mm were added as grinding media in a 1:1 ratio. The formulations were
processed on a Skandex shaker for one hour. After dispersing was completed, samples were
cooled to room temperature and filtered through a mesh cone filter.

page 6
 WHITE PAPER:
An efficient approach to
dispersing pigments

Yellow iron oxide formulations

The yellow iron oxide formulations consist of 55 % pigment loading, with 10 % additive solid on
pigment. Because of increasing environmental demands, an exempt solvent (Oxsol 100) was
chosen for this study.

Table 1: Yellow iron oxide formulations

Products Blank Sample A Sample B Sample C Sample D

Oxsol 100 20.0 20.0 20.0 20.0 20.0

Grinding Resin 25.0 19.5 19.5 19.5 19.5

Additive A – 5.5 – – –

Additive B – – 5.5 – –

Additive C – – – 5.5 –

Additive D – – – – 5.5

PY42 55.0 55.0 55.0 55.0 55.0

TOTAL 100.0 100.0 100.0 100.0 100.0

ASOP 0 10 % 10 % 10 % 10 %

Pigment violet formulations

Price is a big driving factor in the pigment concentrate market. Because organic pigments are
normally more expensive, titanium dioxide was coupled in this formulation to reduce price. The
pigment loading for violet was 6 % and titanium dioxide was 30 %, with 30 % additive solid on
pigment.

Table 2: Pigment violet formulations

Products Blank Sample A Sample B Sample C Sample D

Water 43.0 32.2 32.2 32.2 32.2

Grinding Resin 20.0 20.0 20.0 20.0 20.0

Defoamer 1.0 1.0 1.0 1.0 1.0

Additive A – 10.8 – – –

Additive B – – 10.8 – –

Additive C – – – 10.8 –

Additive D – – – – 10.8

PW6 30.0 30.0 30.0 30.0 30.0

PV23 6.0 6.0 6.0 6.0 6.0

TOTAL 100.0 100.0 100.0 100.0 100.0

ASOP 0 30 % 30 % 30 % 30 %

page 7
 WHITE PAPER:
An efficient approach to
dispersing pigments

YELLOW IRON OXIDE RESULTS

Viscosity was measured on a Haake Rheostress 1 Rheometer equipped with PP35 Ti L03 089
plate and tested under a 0.20 mm gap at room temperature. Initial viscosity studies showed that
the blank and Sample A had very shear thickening behavior, meaning their viscosities rose along
with the shear rate. This is not ideal for pigment concentrates because it could clog the dosing
machine when tinting white base paints. Sample B had a shear thinning viscosity until 1,000/s,
where there was a rise in viscosity. This indicates that there was not enough dispersant to wet
out the remaining pigment. Sample C and Sample D had shear thinning curves, with viscosity
being the lowest overall for Sample D. This is attractive to pigment concentrate suppliers because
a low viscosity means more pigment loading is possible. Viscosity will also be tested after a storage
period at elevated temperatures to confirm the performance of the additives.

Figure 6: Initial room temperature viscosity for yellow iron oxide formulations

INITIAL VISCOSITY
Viscosity [mPas]

5,000
4,500
4,000
3,500
3,000
2,500
2,000
1,500
1,000
500
0
1/s 100/s 1,000/s

Shear Rate

Blank Sample A Sample B Sample C Sample D

The blank, Sample A, and Sample B did not have a suitable grind after one hour of processing. A
suitable grind for most pigment concentrate suppliers would be a 6 Hegman or above. Sample C
and Sample D both had grinds over 7 on the Hegman scale. Sample C had a higher viscosity but
good grind after one hour; this indicates that the dispersant loading was not optimized. Further
work will continue to create an optimized formulation with this dispersant. For Sample D, there
was a very low viscosity and high grind, indicating that this was a suitable dispersant loading for
this formulation. Color acceptance and color strength research will be done additionally to
confirm the performance of the additives.

page 8
 WHITE PAPER:
An efficient approach to
dispersing pigments

Figure 7: Hegman grind readings after 1 hour of processing for yellow iron
oxide formulations

GRIND AFTER 1 HOUR PROCESSING

Sample D

Sample C

Sample B

Sample A

Blank

0 1 2 3 4 5 6 7 8
Hegman Grind

VIOLET-TITANIUM DIOXIDE RESULTSGEN

Viscosity parameters for the violet samples were the same as previously mentioned for the yellow
iron oxide formulations. Initial viscosity was very low for the blank and Sample D. Sample A had
a very high viscosity at a low shear rate, but as the shear rate increased, the viscosity lowered.
Sample B and Sample C both had shear thinning curves.

Figure 8: Initial viscosities of violet-titanium dioxide formulations

[mPas] INITIAL VISCOSITY

100
90
80
70
60
50
40
30
20
10
0
1/s 100/s 1,000/s
Shear Rate
Blank 13.99 2.67 1.12
Sample A 1957 61 22.6
Sample B 893 40.71 16.7
Sample C 956 110 68
Sample D 37 3.5 1.8

Blank Sample A Sample B Sample C Sample D

page 9
 WHITE PAPER:
An efficient approach to
dispersing pigments

Samples were tested for particle size using a Beckman Coulter LS 13-320 Laser Diffraction
Particle Size Analyzer. The blank, Sample B, and Sample C showed the lowest median particle
size after one hour of grinding. However, the results are still too high for an organic pigment.
(Recall: There is also titanium dioxide in this formulation. It is possible that this could be skewing
the particle size measurements.) These formulations likely require more energy to break up the
remaining agglomerates and aggregates. Additional measurements will be taken over longer
periods of grind time to see if this will lower the particle size.

Figure 9: Median particle size of violet-titanium dioxide formulations

d50 PARTICLE SIZE

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
particle size in micrometers

Sample D Sample C Sample B Sample A Blank

Pigment dispersions were evaluated for color acceptance and color strength using a commercial
grade architectural paint at 4 % by volume. Drawdowns were made on Leneta 3B cards at 3 mils
wet and allowed to air dry before color measurements were taken using an X-Rite 962 spectro-
photometer.

Figure 10: Color strength values for violet-titanium dioxide formulations

COLOR STRENGTH
500
450
400
350
Color Strength [F]

300
250
200
150
100
50
0

Blank Sample A Sample B Sample C Sample D

page 10
 WHITE PAPER:
An efficient approach to
dispersing pigments

Figure 11: Color acceptance values for violet-titanium dioxide formulations

COLOR ACCEPTANCE
6

5
Color Difference [ΔE]

Blank Sample A Sample B Sample C Sample D

The blank has a high color strength but also a very high ΔE. This indicates that while the blank
may look stable, after the pigment is disrupted, it falls out of stabilization. Sample A and Sample
D both had lower color strengths and higher ΔE values. This would be unacceptable to paint
manufacturers because not only would the pigment not be utilized for tint strength, it would also
change color when applied. Sample B and Sample C had the highest color strength and a ΔE less
than one. These would be the most stable and most attractive formulations in this case.

For this formulation, overall Sample B and Sample C had the best performance. More research
will be done to optimize these formulations. Additionally, all dispersion samples will undergo
elevated temperature testing for 1 week at 50 °C. Viscosity, particle size, color acceptance, and
color strength will then again be tested for confirmation of each additive’s performance.

CONCLUSION
The results for yellow iron oxide convey that Sample D was the best dispersant for that formula-
tion. The higher molecular structure in combination with the hyper-branched polymeric chains
was found to be especially useful with the viscosity of the formulation. This allows for a higher
pigment loading to be achieved at a fraction of the grinding time. Less machine time, less energy
consumption, and fewer labor costs drastically reduce the complexity of processing, simply by
utilizing the most suitable additive. This contributes to a more cost-efficient way to formulate
pigment concentrates and coatings.

On the other hand, Sample D did not perform well in the violet-titanium dioxide formulation.
Most likely, the molecular weight was too high and the polymeric side chains were not soluble
and caused incompatibilities with this system. Here, Sample B and Sample C performed well in
regards to color strength and color acceptance, and they produced the most economic grinds by
getting the most out of the pigment. These two additives kept the violet and titanium dioxide
stable using a “controlled flocculation” matrix. This allowed for the two pigments to be bound
together to prevent further flocculation by using the dispersant as the bridge between the two.

page 11
 WHITE PAPER:
An efficient approach to
dispersing pigments

Figure 12: Controlled flocculation matrix between organic and inorganic pigment

Pigment and additive interactions can be difficult to determine because there are so many vari-
ables at play. However, the ability to understand the dispersant and pigment structure and the
forces acting on them leads to a better understanding of how to harmonize the two when devel-
oping paints and pigment concentrates.

REFERENCES
[1] Evonik Corporation, TEGO Chemie Service GmbH. TEGO Journal, 4 ed.; 2012. pp 79–89.
th

[2] Mueller, Bodo. Understanding Additives. European Coatings Literature,

Vincentz Network GmbH & Co.: Hannover, Germany 2010
[3] Heilen, Wernfried. Additives for Waterborne Coatings. European Coatings Literature,
Vincentz Network GmbH & Co.: Hannover, Germany, 2009.
[4] Winkler, Jochen. Dispersing Pigments and Fillers. European Coatings Literature,
Vincentz Network GmbH & Co.: Hannover, Germany, 2012.

CONTACT
Brian O. Casey, Phone +1 804 727-0667
brian.casey@evonik.com

page 12
This information and all further AEROSIL®/ACEMATT® ACEMATT®, ADDID®,
technical advice are based on Evonik Resource Efficiency AEROSIL®, AIRASE®,
our present knowledge and GmbH ALBIDUR®, CARBOWET®,
experience. However, it Rodenbacher Chaussee 4 DYNOL , NANOCRYL®,
TM

implies no liability or other 63457 Hanau SILIKOFTAL®, SILIKOPHEN®,

legal responsibility on our part, Germany SILIKOPON®, SILIKOPUR®,
including with regard to exist- Phone +49 6181 59-12532 SILIKOTOP®, SURFYNOL®,
ing third party intellectual Fax +49 6181 59-7112532 TEGO®, TEGOMER®,
property rights, especially ask-si@evonik.com ZETASPERSE® are registered
patent rights. In particular, no trade marks of EVONIK
warranty, whether express or TEGO® INDUSTRIES AG
implied, or guarantee of prod- Evonik Resource Efficiency or one of its subsidiary
uct properties in the legal GmbH companies and are written in
sense is intended or implied. Goldschmidtstraße 100 capital letters.
We reserve the right to make 45127 Essen
any changes according to tech- Germany
nological progress or further Phone +49 201 173-2222
developments. Fax +49 201 173-1939
coating-additives@evonik.com
The customer is not released
from the obligation to conduct SURFYNOL®/DYNOL / ™

careful inspection and testing AIRASE®

of incoming goods. Perfor- Evonik Materials
mance of the product described Netherlands BV
herein should be verified by Kanaalweg 15
testing, which should be 3502 GD Utrecht
carried out only by qualified Netherlands
experts in the sole responsibil- Phone +31-30-2857100
ity of a customer. Reference to apcse@evonik.com
trade names used by other
companies is neither a recom-
mendation, nor does it imply
that similar products could not
be used.

03_2018