See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/310505408

Use of Optical density and TiO2 light scattering

Article · September 2016

CITATION READS

1 1,171

3 authors, including:

Ricardo Tadeu Abrahão Michael Diebold

King's College London Retired from The Chemours Company
7 PUBLICATIONS 4 CITATIONS 90 PUBLICATIONS 973 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

TiO2 dispersion analisis View project

CaCO3 Prec View project

All content following this page was uploaded by Ricardo Tadeu Abrahão on 29 March 2018.

The user has requested enhancement of the downloaded file.

 Use of optical density and TiO2 light scattering to identify
optimization potential in architectural coatings
Abrahao. R.T., Umemura. R., DuPont do Brasil S.A., Brazil

Abstract

Optical density is a technique that evaluates the amount of light scattered by particles in a suspended
paint sample, after developing an improvement in it, is now possible to understand the titanium
dioxide (TiO2) light scattering efficiency without knowing the complete paint formulation. Spreading
rate analyses take into consideration the TiO 2 amount to define this material scattering coefficient in
the dry paint film.

With these analyses it was possible to identify how well the TiO 2 was dispersed in the liquid paint and
how evenly distributed in the dry paint film it was (dispersion/flocculation problems or crowding effect
during drying process) which we called TiO 2 dispersion efficiency and formulation efficiency. Some
paints presented an opportunity to improve the TiO2 dispersion in almost 20% and the formulation
(extenders/fillers, resin, etc.) in 55%.
After defining the TiO2 usage efficiency it was possible to compare paints from different countries in
Latin America.

Keywords

Properties optimization; architectural coatings; titanium dioxide; artificial neural networks

Introduction

Titanium dioxide (TiO2) is the most important white pigment in several industries due to its ability to
conciliate light scattering and white color (1-4). However, to achieve adequate performance, coatings
that use TiO2 pigments must consist of individual particles (5).

In this study, a direct measurement of TiO2 dispersion in dry film was studied using electron
microscopy. The image was broken into sub-images, and then the sub-images were compared to one
another (6). Spatial point statistics methodologies were used, such as those developed by Clark-
Evans and Ripley, as well as the neighborhood distribution function (7).

Not all paint producers have access to equipment like electron microscopes, so the purpose of this
work was to understand final coatings’ formula TiO2 dispersion and how it stays dispersed in dry film
once it will directly affect its final properties using a combination of theoretical and empirical
methodologies.

TiO2 dispersion in dry film

The light scattering provided by TiO2 is strongly influenced by several factors, including particle size,
crystal phase, refractive index of the binder or surrounding medium, and degree of isolation from
other TiO2 particles. This degree of isolation is also referred to as the degree of dispersion or lack of
crowding (8-11).

To achieve the best possible performance, TiO2 particles must be separated; the distance from
particle to particle has to be at least 400 nm (8). For instance, agglomerated pairs of particles exhibit
a 20% decrease in the scattering parameter associated with the hiding power of a paint film relative to
a single particle.

Another way of looking at this is demonstrated in Figure 1, which depicts the impact of particle-to-
particle spacing on the efficiency of a paint’s hiding power; the pictures shown are from electron
microscopy of three different paints with different particle dispersions. It can be seen that the spacing
is responsible for a 30% increase in the hiding power performance of the paint.
 Agglomerated Random Spaced

Relative Hiding Power:

100.0 121.5 130.0

PVC = 20% for all three paints

Figure 1 – Impact of particle-to-particle distance on the paint’s hiding power.

To achieve the condition in which the particles are separated or isolated from each other, the
dispersion process and stability of the suspension must be carried out such that when the coating film
is drying or curing, the particles are already dispersed and isolated from each other.

A proper TiO2 dispersion in the dry film can be achieved by combining a good dispersion of the TiO2
with a proper formula for keeping the TiO2 apart after the film is formed (12). On the one hand,
previous studies have shown that the de-agglomeration process is influenced by wetting pigmentary
TiO2 (13-14) and the mechanical forces applied by the equipment design (15). On the other hand,
Diebold (16) demonstrated that TiO2 spacing is effective for scattering light and that its crowding is
influenced by other coatings of raw materials.

There are several theories for explaining the interaction of a light beam with a particle (see, for
example, the work of Mie, Rayleigh, and Fraunhofer - 17). Mie developed a basic theory for spherical
particles that is useful when particles with more complex shapes are under evaluation. Rayleigh’s
scattering theory applies when particles are shorter than the wavelength of the incident beam, and
Fraunhofer’s theory can be applied when the particles are larger than the wavelength of the incident
beam.

In the case of pigmentary TiO2, the particles are about half the size of the visible light wavelengths
(400 nm to 750 nm). Therefore, the particles range between 200 nm and 350 nm. Bigger particles are
sometimes found due to the particular requirements of the application. Therefore, Rayleigh’s and
Mie’s solutions for Maxwell’s equation were used in this study to calculate the particle size based on
light scattering.
According to Rayleigh’s and Mie’s theories, the scattering pattern (I) is independent of the particles’
shape, but it takes into account the distance from each particle to the detector (r), the incident light
intensity (I0), and the wavelength ( )in Equation 1:

Equation 1

where is the light polarization or the parameter of the equation that contains the relative refractive
index (m) and the volume of the particle (V) as presented in Equation 2:
Equation 2

.
Optical density

Optical density originated from the Lambert-Beer law, which was derived from the first-order Taylor
expansion of Maxwell’s equations and is depicted by Equation 3 (18).

𝐼(𝑡)
𝑂𝐷 = −𝑙𝑜𝑔 ( ), Equation 3
𝐼0
where I(t) is the light transmitted through the media of interest and I0 is the incident light; e.g., the
optical density is the absorbance of light when passing through the media under investigation. If the
TiO2’s main function is to scatter light, measuring the light absorbance is equivalent to measuring the
light scattering efficiency of this TiO2. Measuring particle size could be an alternative option, but the
different refractive index of each particle makes it harder to understand and interpret the output data.

The problem of optical density is to define a reference of the optimized dispersion; by dividing the
sample into two and sonicating one one of these, creates the reference we are looking for. Sonication
might break some extenders and resins into smaller particles, but the TiO2 particles will remain intact,
only breaking down the agglomerates remaining from the poor dispersion process. The ratio between
the sonicated and nonsonicated samples is the dispersion efficiency of TiO2 (𝜖𝐷𝑖𝑠𝑝 ).

Spreading rate

The weighed hiding test method is a form of ASTM D2805-11, hiding the Power of Paints by
Reflectometry. The ISO equivalent is the ISO 6504-1:1983 Kubelka-Munk hiding power method for
white and light-colored paints. Both test methods include the means to perform the calculations
presented by the Kubelka-Munk theory (Equation 4 to Equation 11).

1  Rg (a  b ctgh bSX ) and

R  f ( SX , Rg , R )  Equation 4
a  Rg  b ctgh bSX
1 1  aRo ,
SX  f ( Ro , R )  Ar ctgh Equation 5
b bRo
where
1 Ro  R  Rg  ,
a R   Equation 6
2  Ro Rg 
1
b  (a 2  1) 2 , Equation 7
R  a  b , Equation 8
Equation 9
, and
K
1 R  ,2

 Equation 10
S R

where R is the reflectance of film applied to the substrate, Rg is the substrate reflectance, R is the
reflectance of a film so thick that increasing its thickness has no effect on R, Ro is the reflectance of
the film applied to an ideal black substrate (Rg = 0), X is the thickness of the film, S is the scattering
coefficient of the film, and K is the absorption coefficient of the film.

Basically, the ASTM test method determines dry film thickness and back-calculates wet film thickness
using an additional measurement (i.e., nonvolatile content). The ISO test method and the ASTM test
methods are similar in this regard.
Using the scattering coefficient of the film and the TiO2 amount, it is possible to calculate the
scattering coefficient of TiO2 (STiO2). The TiO2 concentration determines the average refractive index
of the coating film once resin, fillers, and extenders have about the same refractive index (1.56 to
1.6). The average refractive index will determine the TiO2 scattering efficiency in dry film.

If a single particle of TiO2 is surrounded by resin (with a refractive index of 1.5), the refractive index
ratio would be of 1.83. This is the highest value that this ratio can reach for a TiO2 resin mix; when
you add TiO2 to the resin, the average refractive index rises and reduces the TiO2 scattering
efficiency. If arbitrary values are given to the scattering value, then the curve of light scattering as a
function of TiO2 concentration would be described as in Figure 2.
 Figure 2 – Scattering efficiency of TiO2 as a function of the average
refractive index of dry film coatings.

The curve in Figure 2 assumes that the TiO2 is fully dispersed into the resin or dry film, meaning that
this would be the maximum value possible for achieving each TiO2 concentration. It is possible to
compare paints with different TiO2 concentrations by simply looking at them. However, it is better to
compare them by examining their theoretical curves; e.g., in Figure 3, the TiO2 efficiency of Paint 1 is
85%, while it is 90% for Paint 2.

Figure 3 – Comparison of paints with different TiO2

concentrations.

As previously mentioned, the lack of TiO2 dispersion efficiency in paints may be a result of dispersion
or formulation, and the STiO2 efficiency is an intensive property and basically ignores formulas
differences, looking to the TiO2 scattering efficiency alone. To calculate the formulation efficiency of a
paint (𝜖𝐹𝑜𝑟𝑚 ), the dispersion efficiency (𝜖𝐷𝑖𝑠𝑝 ) must be excluded from the overall efficiency and
calculated from the STiO2 (𝜖𝑆𝑇𝑖𝑂2 ), as shown in Equation 11.

𝜖𝐹𝑜𝑟𝑚 = 1 − (𝜖𝐷𝑖𝑠𝑝 − 𝜖𝑆𝑇𝑖𝑂2 ). Equation 11

Samples

For this study, 208 paint samples from different Latin American countries were collected. Colombia,
Argentina, Brazil, Chile, Ecuador, Peru, Venezuela, and Mexico were the most dominant countries.
The samples were collected from producers that controlled at least 50% of their markets to assure
representativeness. As many of these countries do not have regulations regarding paint quality, this
was determined according to the perception of our representatives in these countries.
Results and discussions

An average dispersion efficiency boxplot for each of the paints from the different countries is
presented Figure 4; it can be seen that most of the countries have a dispersion efficiency above 90%,
meaning that the dispersion is primarily efficient. Once this analysis was made using the final paint,
there was no way to increase the dispersion before the paint is used, thus directly affecting the paint
quality.

1 00.00%
Average of Dispersion efficienc

90.00%

80.00%

70.00%

60.00%

50.00%
at at at at at at in in in in at at at at at in a t i n at in a t in at i n at a t i n at at at
F l Fl Fl F l Fl Fl at at at at Fl Fl F l Fl Fl at F l at F l at Fl a t F l at F l F l at F l F l Fl
w id p w id p S S S S w id p w id S p S w S id S p S w id S w id p
Lo M To l Lo l M l To To p Lo w Mid To p Lo M T o Lo r M Mi d r To Top Lo Low M Mi d To Top Lo M Top Lo M To
na n na zi zi z i il e e e ia ia ia or o r o r co
a c o o co o r u r u u la la la
t i t i t i r a r a r a az il hil hil b mb mb a d a d do ad d o x i ico x i ic x i ic Pe Pe er z ue zue zu e
gen gen gen B B B Br C h C C lo m lo lo q u qu ua qu ua M é éx Mé éx M é éx P e e e
n en en
r r r o o o E E q E q M M M
A A A C C C E E Ve V V

Figure 4 – Average dispersion efficiency of paints from different

countries.

In Figure 4, it is also possible to see which paints have poor dispersion, such as the low-quality paints
from Mexico whose dispersion efficiency is about 60%. Our investigation showed that this poor
dispersion rate was a mix of poor TiO2 quality and direct use of it, not pre-dispersing it.

In Figure 5, it is possible to see the formula’s efficiency for each country; this efficiency is primarily
determined by the mix of resin and extenders, if particles are pushing the TiO2 into a crowding
system in the middle of everything, it reduces the TiO2 efficiency in the dry paint film; if the TiO2
accommodates evenly separated from each other, it will have a better efficiency.

1 00.00%
Average of Fromula Efficiency

90.00%

80.00%

70.00%

60.00%

50.00%

40.00%
at at at at at at in in in in at at at at at in a t i n at in a t in at i n at a t i n at at at
F l Fl Fl F l Fl Fl at at at at Fl Fl F l Fl Fl at F l at F l at Fl a t F l at F l F l at F l F l Fl
w id p w id p S S S S w id p w id S p S w S id S p S w id S w id p
Lo M To l Lo l M l To To p Lo w Mid To p Lo M T o Lo r M Mi d r To Top Lo Low M Mi d To Top Lo M Top Lo M To
na na na zi zi z i il e e e ia ia ia or o r o r co c o o co o r u r u u la la la
t i t i t i r a r a r a az il hil hil b mb mb a d a d do ad d o x i ico x i ic x i ic Pe Pe er z ue zue zu e
en en en B B B Br C h C C lo m lo lo q u qu ua qu ua M é éx Mé éx M é éx P e e e
rg rg rg Co Co Co E E E q E Eq M M M n en en
A A A Ve V V

Figure 5 – Average formula efficiency of paints from different

countries.

The reason that the Mexican low-quality paint appears as one of the top performers, is because of the
crowding effect due to a lack of dispersion. It does not allow TiO2 to be pushed to the available space
between the other particles; meaning that it does not matter the extenders and resins in use, the TiO2
dispersion cannot be improved (or in other words, a bad dispersion cannot be worsen by a bad choice
of extenders and resins).

A different way of looking at the same data above is the graph in Figure 6 that combines the
dispersion and formulation efficiencies with TiO2 content (i.e., bubble size). A different perspective is
then presented, showing that the Latin American region’s average dispersion efficiency is 93% and
that its average formulation efficiency is about 70%.

When looking to specific cases (in Figure 6), the Brazilian Coating 18 shows good performance
regarding formulating the paint, using the proper fillers and extenders, and making a good dispersion
of the TiO2 this specific customer is currently using.

Bubble Plot of Average of Dispersion ef vs Average of Fromula Effic

Bubble siz e: Average of TiO2 %(Paint w/w)
70. 00%
1 00. 00%
BRA 18
B RA 7
Average of Dispersion efficienc

CO L 1CBORLA34 B RA 5 BRA 12 BRA 15

BRA 11
95. 00% BRA 17
B RA 2
B RA 8
ECBURA1 19
CO LB2RA 10 B RA 6
B RA 1 BRAAR3G5
ARG 3 VC
EN
HI 11 BRA 13
A R G
BRA C9HI 21 BRA 20 91
90. 00% ARG 2 VEN 2
ARG 4 MEX 4 ECU 3
P ER 2 BRA 14
BRA 16
A rgent i na
85. 00% Br az i l MEX 2 ECU 2
Chi l e
Col ombi a
Equador MEX 1
80. 00% Méxi co
Peru MEX 3
CHI 1
Venez uel a P ER 1

55. 00% 60. 00% 65. 00% 70. 00% 75. 00% 80. 00% 85. 00% 90. 00%
Average of Fromula Effic iency

Figure 6 – Formula and dispersion efficiency combined with TiO 2 amount (i.e., bubble
size).

When looking at the individual qualities, like the high-end coatings (Figure 7), the first thing you might
notice is that Brazilian coatings have lower TiO2 amounts (i.e., smaller bubble sizes), but most of them
are well dispersed (with dispersion efficiency of more than 90%). The range of formulation efficiency
spreads throughout all the existing ranges, meaning that good dispersion is achieved, but not used
well by the combination of ingredients, like in Brazilian Paints 2 and 3. On the other hand, the
Mexican coatings have an impressive formulation efficiency but lack dispersion efficiency. At this
point, one can only assume that this is probably caused by Mexico’s habit of not pre-dispersing the
TiO2.
 Top paints of Latin America Region
Bubble siz e: g TiO2 / L
1 00. 00% PCEO
R2 B RA 7
P ERL21 CO L P1EBRR2 A 5 B RA 7 BBRRAA
B1R1A 4
B1BRH
RAIA
I 252 6 RR
ECU 1 COCL H3I CC H 15BA AG6 1 ACROGL 12BRAB6RAARG 4 1 B RA 8
CHIB2RA 2 BRA 3
VEN 1 ARG 2
B RA 8 B RA 7 B RA 8
ARG 2 BRA 3
90. 00% CO L 2
ARG 2
Dispersion effic ienc y

ECU 2
Co u n t r y ( Flat/Satin )
80. 00% Argen tin a Flat
Argen tin a Satin P ER 1 P ER 1
Braz il Flat MEX 2 MEX 3
Braz il Satin
Ch ile Flat MEX 3
70. 00% Ch ile Satin MEX 3
Colombia Flat
Colombia Satin
Equ ador Flat
Equ ador Satin CHI 1
60. 00% México Flat
México Satin MEX 1
P eru Flat MEX 1 MEX 2
MEX 1
P eru Satin
MEX 2
Ven ez u ela Flat
50. 00%
0. 5 0. 6 0. 7 0. 8 0. 9 1 .0
Fromula Efficienc y

Figure 7 – Top-quality coatings’ dispersion and formula

efficiency.

Figure 8 adopts the perspective of a single customer to determine if he is facing batch-to-batch

variations and to allow the targeting of specific paint lines and grades. For example, this customer has
selected a particular paint as being top quality (Top 1); it uses a good amount of TiO2, but lacks
dispersion. This coating’s dispersion average is below the other coatings average from this same
customer (that ranges about 96%). It is also possible to see variations in the same paint lines, such as
Top 3, where a difference of almost 20% is found between two samples, or in the Mid paint, where no
sample is close to any of the others (which is the same for the Top 2 paint), suggesting formula and
process variations.

Figure 8 – Customer-specific perspective.

Conclusions

Through the analyses described in this paper, it was possible to identify how TiO2 is dispersed in
liquid paint and if it is evenly distributed in the dry paint film (i.e., if there are dispersion/flocculation
problems or a crowding effect during the drying process, which are referred to as TiO2 dispersion
efficiency and formulation efficiency in this study). Almost 20% of the paints showed a need to
improve the TiO2 dispersion, and 55% of the paints demonstrated a need for improved formulation
 (extenders/fillers, resin, etc.). When combined, the optical density and the scattering efficiency of the
dry film can also help guide improvement projects where the most benefit can be extracted from the
TiO2 particles.
It is also possible to use the same basis to compare TiO2 use and performance in paints produced by
different companies, even those from different countries.

Acknowledgments

The authors extend their gratitude to DuPont for the opportunity to work on this project. Thanks is also
given to Dr. Paulinia’s laboratory team for the work they performed and to Pablo Aragon for his
continuous support during this study.

References

1. Ahmed, M, Coloring of plastics: theory and practice. VanNostrand Reinhold, New York(1979),
page 52
2. Auer, G, Woditsch, P, Westerhaus, A, Kischkewitz, J, Griebler O.D, and Liedekerke, M,
Pigments, Inorganic, 2. White Pigments, Standard Article, Ullmann’s Encyclopedia of
Industrial Chemistry, Published Online: 15 OCT 2009, DOI: 10.1002/14356007.n20_n01 page
257.
3. Farrokhpay, S, “A review of polymeric dispersant stabilisation of titania pigment” Advances in
Colloid and Interface Science, 151, (1–2)24–32 (2009)
4. Murphy, J, Additives for Plastics Handbook (Second Edition) Chapter 7 – Modifying Specific
Properties: Appearance-Black and White Pigmentation (2001), page 73
5. Thiele, ES, French, RH, ‘‘Light-Scattering Properties of Representative Morphological Rutile
Titania Particles Studied Using a Finite-Element Method.’’ J. Am. Ceram. Soc., 81 (3) 469–
479 (1998)
6. Diebold, M.P. “Quantifying HO2 pigment dispersion” Polymers Paint Colour Journal, 199
(4540), pp. 32-34. (2009)
7. Elton, N.J., Legrix, A. “Spatial point statistics for quantifying TiO2 distribution in paint” Journal
of Coatings Technology Research, 11 (3), pp. 443-454. (2014).
8. Thiele, E. S., and French, R. H., 2005. Light-scattering properties of representative,
morphological rutile titania particles Studied Using a Finite-Element Method. Journal of the
American Ceramic Society, vol. 81(3), pp. 469-479.
9. Bruehlman, J, Thomas, L. W., Gonick, E., 1961. Effect of particle size and pigment volume
concentration on hiding power of titanium dioxide. Off. Dig. 33(433), pp. 252-267
10. Tunstall, D. F., Hird, H. J., 1974. Effect of Particle Crowding on Scattering Power of TiO2
Pigments. J. Paint Technol., 46(588), pp. 33–40.
11. Auger, J-C, Martinez, VA, Stout, B., 2009. Theoretical Study of the Scattering Efficiency of
Rutile Titanium Dioxide Pigments as a Function of Their Spatial Dispersion. J. Coat. Technol.
Res., 6(1) pp. 89–97.
12. Auger, J.-C., McLoughlin, D. “Theoretical analysis of light scattering properties of
encapsulated rutile titanium dioxide pigments in dependent light scattering regime” Progress
in Organic Coatings, 77 (11), pp. 1619-1628 (2014).
13. Abrahao, RT, Postal, V, Paiva, JL, Guardani, R., “Study on Required Energy to
Deagglomerate Pigmentary Titanium Dioxide in Water.” J. Coating. Technol. Res., 11 (2) 239-
253 (2014)
14. Abrahao, RT, , Postal, V, Paiva, JL, Guardani, R, “Wettability Study for Pigmentary Titanium
Dioxide.” J. Coating. Technol. Res., 10 (6) 829-840 (2013)
15. Patton, T., 1979, Paint flow and pigment dispersion, a rheological approach to coating and ink
technology, Second edition, John Wiley & Sons, Inc, ISBN 0-471-03272-7.
16. Diebold, M.P. “A Monte Carlo determination of the effectiveness of nanoparticles as spacers
for optimizing TiO 2 opacity” Journal of Coatings Technology Research, 8 (5), pp. 541-552.
(2011).
17. Gouesbet, G. “Generalized Lorenz-Mie Theory and Applications”, Particles and Particle
System Characterization, I1 (1994) 22-34.
18. Combessis, A., Mazel, C., Maugin, M., Flandin, L., “Optical Density as a Probe of Carbon
Nanotubes Dispersion in Polymers”, Journal of Applied Polymer Science, volume 130 (3),
1778-1786, Willey, (2013).

View publication stats