Progress in Organic Coatings 72 (2011) 541552

Contents lists available at ScienceDirect

Progress in Organic Coatings

journal homepage: www.elsevier.com/locate/porgcoat

A new approach to investigate scratch morphology and appearance of an

automotive coating containing nano-SiO2 and polysiloxane additives
B. Ramezanzadeh a,b, , S. Moradian a , A. Khosravi a , N. Tahmasebi c
a
b
c

Department of Polymer and Color Eng, Amirkabir University of Technology, P.O. Box 15875-4413, Tehran, Iran
Department of Surface Coatings and Corrosion, Institute for Color Science and Technology, No. 59, Vafamanesh St., Hosainabad Sq., Lavizan, Tehran, Iran
Department of Polymer Engineering, Mahshahr Branch, Islamic Azad University, Postal Code 63519, Mahshahr, Iran

a r t i c l e

i n f o

Article history:
Received 18 February 2011
Received in revised form 29 May 2011
Accepted 21 June 2011
Keywords:
Acrylicmelamine
Gonio-spectrophotometry
Fracture
SEM
Scratch testing

a b s t r a c t
Attempts have been carried out to verify whether gonio-spectrophotometry could be utilized as a
new scratch testing approach to investigate scratch morphology and its effect on appearance of an
acrylicmelamine clearcoat separately containing nano-silica or polysiloxane additives. The results of
gonio-spectrophotometry/colorimetry illustrated that this new approach is capable of differentiating
between plastic and fracture types scratches. Furthermore, this approach was found to be suitable for
analyzing the self healing abilities of such coatings. The accuracy and reproducibility of such results were
checked against the results of scanning electron microscope (SEM) and illustrated great potential as a
new approach for such studies.
2011 Elsevier B.V. All rights reserved.

1. Introduction
Outdoor durability of an automotive coating especially the
clearcoat layer is one of the main concerns of automobile manufacturers. This concern arises from the fact which for most customers,
retained appearance during service life is the most important criterion. Since 80s, the basecoat/clearcoat system of coating has
been in use in the automotive industries. Generally, basecoat is the
color producing layer, whilst the clearcoat improves physical and
chemical properties of the automotive coating system as well as
enhancing its appearance [13].
When an automotive clearcoat is exposed to various outdoor
conditions such as sunlight, humidity, acid rain and bird droppings,
certain photo, biological and mechanical degradations seemingly
affect appearance more and tend to be investigated most by various researches. The most investigated mechanical degradation
is scratch and mar which are created by various sharp objects
(scratch) or nger nails (mar) respectively. Depending on the force,
different types of deformation with varying dimensions could
be produced. Scratch has depth and width of 0.20.8 m and
1020 m, respectively, whilst the depth and width of mar are

Corresponding author at: Department of Polymer and Color Eng, Amirkabir University of Technology, P.O. Box 15875-4413, Tehran, Iran. Tel.: +98 2164542425;
fax: +98 2188080871.
E-mail address: ramezanzadeh@aut.ac.ir (B. Ramezanzadeh).
0300-9440/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.porgcoat.2011.06.014

much less pronounced being 0.050.6 m and 610 m, respectively [4,5].

Different factors such as temperature, intensity rate of applied
external load, chemical composition and mechanical properties of
a clearcoat layer can cause different types of deformation. Elastic deformation, due to its returnable properties, does not affect
appearance of a clearcoat, immensely plastic and fracture deformations can both cause visible damage to a clearcoats appearance
but to a varying extent. Plastic deformation usually has a more regular shape with a smoother surface and edge having a lower ability
to scatter light. In addition, this type of deformation has inherent
ability to self heal with time or with increased temperature. On the
other hand, fracture deformation has an irregular shape with sharp
edges and an unsmooth surface which can scatter the light intensively, causing irreparable damage to the appearance. Therefore,
fracture type scratches affect clearcoats appearance more signicantly than the plastic type [615].
Various parameters such as hardness, toughness, glass transition (Tg ) and cross-linking density of the clearcoat may affect the
scratch or mar resistance of an automotive clearcoat. In order to
control such parameters various procedures are at hand such as
varying curing procedures (e.g. IR, UV, EB), changing chemical compositions of resin (e.g., acrylic/melamine, polyurethane) or the use
of various additives (i.e. polysiloxane, nano silica) in the clearcoat
formulation. In this study, the effect of different nano and micro size
additives on scratch resistance of an acrylic/melamine clearcoat
was studied. Therefore, the effect of using three micro polysiloxane

542

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

Table 1
Clearcoats specications and samples coding.
Sample

Chemical structure

Si-1
Si-2
Nano

Polydimethylsiloxane (polyether modied)

Polydimethylsiloxane (uorocarbon modied)
Hydrophobic surface treated nano silica particle

(polyether or uorocarbon modied) additives as well as a nano

silica additive on scratch resistance and scratch morphology of the
clearcoat was compared [1618].
In order to investigate the scratch or mar resistance of an
automotive clearcoat, different methods can be used. A conventional method to investigate the effect of scratch or mar on
appearance is to obtain gloss retention of a clearcoat after car
washing using a laboratory carwash simulator. Hosseinpour et al.
[19] measured weight loss after abrasion test to investigate the
scratch resistance of an automotive topcoat. Rossi et al. [2022]
studied the anticorrosion performance of coating after abrasion
test and mechanical damage. The same experiment was carried
out by Reyes-Mercado et al. [23] and Rutherford et al. [24]. In

Additive content (wt.%)

No additive

0
0
0

0.1
0.18
1.25

0.2
0.26
2.5

0.25
0.34
3.75

0.4
0.5
6.0

addition, a nano indentation test can be also useful in describing the scratch resistance of a clearcoat at nano scale based
on the minimum required critical force for fracture deformation
[2532].
Different factors such as the numbers of scratches per square
area of a clearcoat, size (depth or width of scratch) and morphology of a scratch can affect gloss retention. Therefore, gloss
retention just illustrate the total nal effect and is unable to
clarify whether number, size or morphology of the scratches were
responsible for decreased gloss retention. Many of researchers
have used microscopic analysis in order to provide better description of scratch dimension and/or scratch morphology. The most
common microscopic analyzes are SEM and AFM micrographs of

Fig. 1. SEM micrographs after scratch test of the coatings containing different loadings of nanoparticles; (A1 ) 0 wt.%, (A2 ) 1.25 wt.%; (A3 ) 2.5 wt.% and (A4 ) 3.75 wt.%.

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

543

Fig. 2. SEM micrographs after scratch test of the coatings containing different loadings of Si-2 (A1 A3 ) and Si-1 (B1 B3 ); (A1 ) 0.18, (A2 ) 0.26 and (A3 ) 0.34 wt.%, (B1 ) 0.10 wt.%,
(B2 ) 0.20 wt.%; (B3 ) 0.25 wt.%.

scratches demonstrating exact size and morphology of scratches.

Researchers have claimed that using optical microscope with or
without polarizing lters, various parameters such as the number
of scratches per square area of the coating and scratch morphology
can be distinguished. This technique relies higher scattering ability
of fracture type scratches due to their sharp edges and unsmooth

surfaces compared to plastic types. Plastic type scratches tend to

disappear at certain angles. Therefore, using polarizing lter some
plastic scratches which were invisible in non-polarized light can be
suddenly observed. However, this technique is unable to describe
scratch morphology and like SEM or AFM micrographs are therefore
only suitable for predicting scratch or mar morphology [3336].

544

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

Fig. 3. Schematic illustration of the effects of two types scratches with fracture morphology of: (a) fracture type second and (b) fracture type rst on coating appearance.

It is the objective of the present study to illustrate how colorimetric parameters may be used as a simple and accurate method
to describe the effect of scratch or mar morphology on appearance
of an automotive clearcoat.
2. Experimental
2.1. Sample preparation
An acrylic/melamine clearcoat containing no additive was provided by a local automotive coating supplier (Rangafarin Paint Co.).
Three polysiloxane additives with various modications coded as
Si-1, Si-2 and a pre-dispersed surface treated nano silica dispersion
coded as nano was also provided. The polysiloxane additives and
nano-SiO2 were supplied by EFKA and BYK Companies respectively.
The additives contents are illustrated in Table 1.
The polysiloxane additives were separately dispersed in the
acrylic/melamine clearcoat formulation using a 15002000 rpm
mixer for 20 min. The nano silica, supplied as predispersed
nano silica in mixture solvents of methoxypropylacetate and
methoxypropanol (6:1 mixture ratio) and claimed to have a mean
particle size of 2025 nm and non volatile content of 32% was also
similarly dispersed in clearcoat.
Steel plates with 10 cm 20 cm 0.1 cm dimensions were used
as substrate. The elemental composition of the steel sample is studied by EDS (energy dispersive spectroscopy) (Table 2).
The steel sheets were provided by Flooad Mobarake (Iran).
Steel sheets were degreased using alkaline cleaner and detergents.
Samples were then washed by acid solution. Two main purposes
Table 2
The elemental composition of steel substrate.
Elements

Weight (%)

S
P
Al
Si
Mn
C
Fe

0.04
0.06
0.05
0.29
0.30
0.22
99.04

were followed during acid washing process. First, the steel samples were cleaned from metal oxides and hydroxides. Second, the
metal surface was activated prior phosphating process. Finally, acid
washed samples were rinsed in distilled water. Samples were dried
in an oven prier phosphating process. A three-cationic phosphating coating was applied on each steel sheet after acid washing.
The phosphated substrates were then coated by cationic waterborne electrodeposition coating (ED) (with dry lm thickness (DFT)
of 1520 m), followed by curing at 140 C for 20 min. Saturated
polyester/melamine coating was used as surfacer on ED coated
samples (with a DFT of 3045 m). Samples were then followed
by curing at 140 C for 20 min. A black metallic acrylic melamine
basecoat (to facilitate easier morphologic studies) was subsequently applied. Finally, clearcoats formulations were applied in
triplicates (for error estimation purposes) by a wet on wet method.
Samples were then cured for 20 min at 140 C.
2.2. Characterization
An Elcometer carwash simulator having polyethylene brush
type, 318320 nm brush length, washing powder of Al2 O3 (suspension), ow rate of 2 drop/min and car wash rotating speed
of 150 rpm was used to produce scratches on the complete
coating system applied on the steel plates. A Gretag Macbeth gonio-spectrophotometer was used in order to obtain
spectrophotometric/colorimetric data of various clearcoats applied
on a black metallic basecoat. The values of reectance and color
attributes were measured after carwash test on 0.785 cm2 area
of each sample. To distinguish scratch and mar morphology
of car wash damages, a Philips scanning electron microscope
(SEM) was used in SE (secondary electron) mode. SEM micrographs were recorded at voltage and magnication of 20 kV and
2000.
3. Results and discussion
3.1. Investigating coating scratch resistance using SEM
The most common method to investigate the scratch size and
morphology is SEM (scanning electron microscope). The SEM

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

545

Fig. 4. Variations of particle size of detached coating particulates versus particle frequency of scratched coatings including different contents nano and polysiloxane additives.

micrographs after scratch test of the clearcoats containing various loads of polysiloxane additives and nano-SiO2 are shown in
Figs. 1 and 2.
The morphology of scratch can be studied in SEM micrographs
of the clearcoats after scratch test. Although, literature generally
divides scratch morphology into plastic or fracture types but, this
study will demonstrate that a fracture type scratch can itself be
subdivided into two types: the rst fracture type scratch has sharp
edges but the second fracture type scratch contains sharp edges
and detached pieces of the coating simultaneously. Both fracture
types scratches can scatter visible light intensively. However, it
will be shown that scratches with detached pieces of coating can
scatter visible light even more intensively than the other types
which would be distinguishable by a gonio-spectrophotometry.
The effects of two types fracture scratches on the light scattering
are schematically shown in Fig. 3.

Scratches created on the surface of blank clearcoat (sample

without any additive) are completely in form of fracture type
scratches. They include lots of big detached pieces of the clearcoat.
Because of this, this kind of scratch can be categorized as second type fracture scratch. Incorporation of nano-SiO2 up to nano-a
lead to the creation of harsher (with sharper edges) fracture type
scratches (fracture type rst) (Fig. 2). Increase in nano-SiO2 loading up to nano-b can cause an increase in the number and a
decrease in the size of detached pieces of the coating. Therefore,
scratches produced at nano-b are in form of both the rst and
the second types of fracture scratch. Fig. 2 shows that, although,
scratch morphology of the clearcoat containing nano-b is mainly
in form of fracture type but, the morphology of the scratches produced at nano-c is clearly in form of plastic type with a smoother
surface compared to all other scratches. Similar behaviors can be
observed on the samples containing various loads of different types

546

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

behavior will be mostly in form of fracture behavior. This can lead

to fracture scratch creation.

Fig. 5. Schematic illustration of the mechanism which gonio-spectrophotometer

measures the color coordinates. (For interpretation of the references to color in this
gure legend, the reader is referred to the web version of this article.)

of polysiloxane additives. As it can be seen in Fig. 2, harsh fracture

type scratches are created on the surface of samples containing low
loadings of all polysiloxane additives (Si-a and Si-b). Plastic type
scratches with smooth surfaces are produced using polysiloxane
additives at high loadings. The fracture type scratches produced
on the clearcoats containing uorocarbon modied polysiloxane
additive has sharper edges compared to the clearcoat containing polyether modied polysiloxane additive. Therefore, it can be
generally concluded that additive with higher surface activity
shows severer fracture behavior [15]. Additive with higher surface
activity (Si-2) can migrate to the surface of the clearcoat greater
than Si-1. As a result of this, additive can increase the surface
hardness of the clearcoat. This leads to more elastic behavior of
the clearcoat in exposure to external stresses. However, when the
applied stress exceeds a certain value (critical load) the coating

3.1.1. Investigating the size of detached pieces of coating

SEM micrographs of the clearcoats (Figs. 2 and 3) can reveal that
the rigid parts of the clearcoat will be detached in form of coating
particles with different sizes (in fracture type scratch). Depending
on the size of these particles, they can scatter visible light to various
amounts. In order to quantify the effects of such particles on the
appearance of the clearcoat precisely, determination of particles
sizes seems necessary. The particles sizes were determined using
particle size analyzer. The software used to determine particles
sizes belongs SEM. Variations of particle frequency versus particle
size of the clearcoats containing various contents of polysiloxane
additives and nano-SiO2 are shown in Fig. 4.
Fig. 4 shows that irrespective of the type of additives (i.e.,
nano-SiO2 or polysiloxane additives) increase of the loads up to
nano-b and Si-1-a can induce signicant increases in the number of
detached pieces of the clearcoats. However, decrease in number of
detached coating particles are obtained using polysiloxane additive
and nano-SiO2 at higher loads. For the sample containing nano-c, no
particle and for the sample containing Si-1-c only a small number
of coating particles are observed. This behavior can be attributed
to the morphology of scratches. Plastic type scratches usually contain lowest number of detached coating particles. In fact, coating
cannot be detached in form of particles when it behaves like a plastic material. On the other hand, fracture type scratch includes large
number of detached coating particles (Figs. 2 and 3).
Results show that addition of nano-SiO2 at low loadings (nano-a
and nano-b) causes signicant increase of the number of detached

45

75

110

1
0

-1

-1

20

-2

-2

-3
-4

20

45

-3

-5

75

110

-4

-6

-5
0

Nano silica content (wt%)

3

2
1

20

45

75

110

20

45

75

110

0
-1

Nano silica content (wt%)

-2

3
2

-3
1

-4
-5

Nano silica content (wt%)

Nano silica content (wt%)

Fig. 6. Variations of the color attributes of L, b, a and E at four goniometric angles of observation of clearcoats containing nano-SiO2 . (For interpretation of the
references to color in this gure legend, the reader is referred to the web version of this article.)

-1

-1

-2
-3

20

45

-2

75

110

-3

-4
0

0.1

0.2

0.3

0.4

0.5

20

45

75

110

-4
0

0.1

Additive content (wt%)

0.2

0.3

0.4

0.5

Additive content (wt%)

3.5

2.5

547

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

-1

20

45

75

110

1.5

-2

-3

20

45

75

110

-4

0.5
0

0.1

0.2

0.3

0.4

0.5

Additive content (wt %)

0.1

0.2

0.3

0.4

0.5

Additive content (wt %)

Fig. 7. Variations of the color attributes of L, b, a and E at four goniometric angles of observation of clearcoats containing Si-1. (For interpretation of the references
to color in this gure legend, the reader is referred to the web version of this article.)

pieces of the coating. Results show that, the size of detached

coating particles is signicantly decreased (compared to blank sample) using nano-SiO2 . The mean particle sizes of 6.61, 0.669 and
0.433 m are obtained for the blank sample, nano-a and nano-b,
respectively. These observations can reveal that increase in nanoSiO2 loading can cause decrease in detached coating particle size.
However, no detached coating particles are observed at nano-c,
indicating plastic morphology of the scratch. The same results
are obtained for the clearcoats containing polysiloxane additive.
The particle size of the coating is decreased (compared to blank
sample) using lowest loading of Si-1 (Si-1-a). However, decrease
in number of the particles as well as the increase in size of the
particles is obtained at higher loadings of the additive. These
observations can demonstrate that using low loading of additive
causes fracture deformation of the coating together with coating
particles detachment. On the other hand, the morphology of the
scratch will be mostly in form of plastic at higher loadings of the
additive. Clearcoat containing nanoparticles shows larger number of detached coating particles (nano-b than the one containing
polysiloxane additive). This can demonstrate that harsher type of
fracture scratch can be created on the surface of the clearcoat containing nano-SiO2 compared to the one containing polysiloxane
additive.
The shape of scratch as well as the size and number of
detached coating particles can inuence the light scattering ability of the scratches. This presumption could be put to test by
measuring appearance parameters of the scratch by a goniospectrophotometry in order to elucidate prediction of scratch
morphology.

3.1.2. Investigating the scratch morphology using

spectrophotometry technique
Gonio-spectrophotometer has been used for color measurement of diffusely reecting materials. Both appearance and color
attributes can be investigated in this technique [3739]. To this
end, light at different wavelengths will be exposed to the surface
of sample at incident angle of 45 . The reected light will be studied at four observation angles of 20 , 45 , 75 and 110 . Measuring
reected light at different wavelengths and near specular (20 and
45 ) angles, the appearance of the sample can be studied (Fig. 5).
On the other hand, measuring diffusely reected light at unspecular angles, the color attributes can be obtained. Measurement in
this technique will be carried out in a color space of CIELab (CIE is
international commission on illuminations) [4043]. The concept
of this technique is based on which colors can be considered as
combinations of red and yellow, red and blue, green and yellow,
and green and blue. To determine the exact combination of colors of a product, coordinates of a three dimensional color space are
assigned (L*: the lightness coordinate, a*: the red/green coordinate,
b*: the yellow/blue coordinate). The L* coordinate of an object is the
lightness intensity as measured on a scale from 0 to 100, where 0
represents black and 100 represents white. The a* coordinate of
an object represents the position of the objects color on a pure
green and pure red scale, where 127 represents pure green and
+127 represents pure red. The b* coordinate represents the position of the objects color on a pure blue and pure yellow scale,
where 127 represents pure blue and +127 represents pure yellow. Measuring L*, a* and b* at near specular angles (20 and 45 )
and unspecular angles (75 and 110 ), the color and appearance

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

548

-1

45

75

110

-1
-2

-2
-3

20

45

75

110

-3
-4

-4
0

0.1

0.2

0.3

0.4

0.5

0.6

0.1

Additive content (wt%)

0.2

0.3

0.4

0.5

0.6

Additive content (wt%)

3.5

3
2

20

45

75

110

2.5

20

-1
-2

20

45

75

110

1.5
1

-3

0.5

-4
0

0.1

0.2

0.3

0.4

0.5

0
0

0.1

0.2

0.3

0.4

0.5

Additive content (wt %)

Additive content (wt %)

Fig. 8. Variations of the color attributes of L, b, a and E at four goniometric angles of observation of clearcoats containing Si-2. (For interpretation of the references
to color in this gure legend, the reader is referred to the web version of this article.)

parameters of the sample will be studied. The difference between

color coordinates of two samples can be calculated according to
Eqs. (1)(3).
a = (a2 a1 )

(1)

where a2 and a1 are the red/green coordinates of samples 1 and 2.

a = (b2 b1 )

(2)

where b2 and b1 are the red/green coordinates of samples 1 and 2.

a = (L2 L1 )

(3)

where L2 and L1 are the red/green coordinates of samples 1 and 2.

The difference between two color samples is often expressed as
delta E (E). E displays the difference as a single value for color
and lightness (Eq. (4)).
E = (a2 + b2 + L2 )

0.5

(4)

According to the above explanations, surface geometry as

well as color characteristics can affect color coordinates (L*, a*
and b*). Because of this, any changes on the surface of sample which may inuence light scattering can be studied by
gonio-spectrophotometer. Scratch and other kinds of mechanical
damages can signicantly affect the surface geometry of the sample. The effects of such kinds of damages on the surface geometry
of the sample can be studied by the determination of a, b and
L.

It is the objective of this study to show that goniospectrophotometry can be introduced as an alternative sound
novel technique for analyzing scratch resistance and morphology
and predict to what extent they affect appearance of a clearcoat.
In this novel technique, effects of scratch on appearance of a
clearcoat was investigated by various difference in spectrophotometric/colorimetric parameters such as differences in reectances
(R), total color differences (E), difference lightness (L) and
differences in colorimetric (a and b). It has been previously
mentioned (in SEM micrographs) that fracture type scratches can
be subdivided into two types. Both types of fracture scratches have
sharp edges which can scatter incident light and affect the appearance parameters such as R, L and E intensively. Parameters
such as scratch depth and width, sharpness of scratchs edges and
number and size of detached coating particles can all affect the
appearance parameters. In general, scratches with sharper edges
and higher number of coating particles enhance the scattering ability of such scratches.
Color attributes including a (a1 (before scratch test) a2
(after scratch test)), b (b1 (before scratch test) b2 (after scratch
test)) and L (L1 (before scratch test) L2 (after scratch test))
were calculated after carwash test. Variations of a, b, L
and E versus additives and nano-SiO2 loadings are shown
in Figs. 68.
Scratch depending on its morphology and dimension can inuence light scattering ability of the clearcoat surface differently.
Plastic type scratches due to their smoother surface compared to
the fracture type can scatter visible light less than fractured one.

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

1.5
1

Nano-0-20

Nano-a-20

Nano-b-20

Nano-c-20

Nano-d-20

0.5
0

-0.5
-1
-1.5
300

400

500

600

700

800

Wavelength (nm)
1.5
1

Nano-0-110

Nano-a-110

Nano-b-110

Nano-c-110

Nano-d-110

0.5

Fracture scratches can scatter visible light (at specular angles) diffusely, causing decrease in coating lightness. As a result of this,
the value of lightness (L2 ) of the scratched sample will be lower
than unscratched one (L1 ). Therefore, the value of L of the sample including fracture type scratches becomes negative. Variations
of b and a at specular (20 ) and unspecular (110 ) angles can
have different meanings. b and a values changes at 110 can be
mostly attributed to color changes of the clearcoat after carwash
test. However, b and a values changes at 20 can be attributed
to the light scattering at blue and red wavelength regions of visible light. Scratches can scatter blue and/or red lights depending on
their morphology and size of detached coating particles. Therefore,
increase in b and a values can illustrate light scattering at blue
and/or red wavelength regions.
As it can be seen in Fig. 6, addition of nano-SiO2 up to nano-b
causes decrease in b. Increase in value of b and E is obtained
using nano-SiO2 up to nano-b. Results show that addition of further
nano-SiO2 causes decrease in b and E and increase of the values of L. Results can also show that addition of nano-SiO2 cannot
inuence a signicantly. Same results are obtained for the samples containing polysiloxane additives (Figs. 7 and 8). The value
of L is decreased (compared to blank sample) using Si-1 up to
0.2 wt.%. On the other hand, values of b and E are increased
using Si-1 up to 0.2 wt.%. Results show that higher concentrations
of Si-1 causes decrease of the values of E and b. The same as
clearcoats containing nano-SiO2 , Si-1 cannot affect a after scratch
test. Fig. 8 shows that addition of 0.18 wt.% Si-2 can cause decrease
in L and increase in b and E. Addition of further additives
causes increase in value of L and decrease of b. The changes of
the color attributes are more signicant at observation angle near
to specular angle (20 ). This can show that scratches mainly inuence clearcoat appearance at near specular angles. Decrease in L
or negative values of this parameter are obtained using low loadings of polysiloxane additives and nano-SiO2 . The negative values
of L of the clearcoats containing low loadings of nanoparticles
and polysiloxane additives can reveal fracture morphology of the
scratches produced on these samples. Results show that using Si1, Si-2 and nano-SiO2 at higher loadings lead to increase in L of
the samples. Based on the previous explanations, positive value
of L of the clearcoats containing high loadings of nanoparticles and polysiloxane additives can reveal plastic morphology of
the scratches produced on these samples. These deductions are
in good agreement with the corresponding SEM micrographs. The
scratches produced on the surface of the clearcoats containing low
loadings of additives and nano-SiO2 have sharp edges including
small detached coating particles. The sharp edges and detached
coating particles can cause severe light scattering and therefore
decrease in L.
SEM micrographs in Fig. 1 illustrate that increase in nano-SiO2
content causes increase of the sharpness of scratches edges and
number of detached coating particles. It was found that coatings
containing nano-b and Si-1-a include high number of detached
coating particles. The average size of detached coating particles
(during scratch test) was lower than 300 nm (Fig. 4). Therefore,
they have high capability of scattering of visible light in blue wavelength region. Moreover, with respected to the particle size of
detached coatings, they are not able to scatter visible light in red
wavelength region. Because of this, the value of b was significantly affected after scratch test whilst the value of a almost
remains constant. Therefore, increase in values of b (positive values) of the clearcoats containing low loadings of nanoparticles and
polysiloxane additives can show fragile behavior of the scratches
produced on these samples. On the other hand, decrease in the
values of b at higher loadings of the additives can reveal lower
capability of the scratches to scatter visible light in blue region. It
has been previously shown that coatings containing high loadings

549

0
-0.5
-1
-1.5
300

400

500

600

700

800

Wavelength (nm)
Fig. 9. Variations of R (reectance difference) versus wavelength of the clearscoats
containing nano-SiO2 (at different contents of ad) at four goniometric angles of
observation.

of polysiloxane additives and nanoparticles include lower number of detached coating particles together with larger particle size.
The scratches produced on these samples have smooth surface
and include low number of detached coating particles (with large
average size) which could not scatter visible light at blue region.
Therefore, decrease in b can illustrate the plastic behavior of
the scratches. These nding is again in good agreement with the
observation made in Fig. 2. Results can show greater variations
of the values of L and b of the clearcoats containing nanoSiO2 compared to the clearcoats containing polysiloxane additives.
This can be attributed to the differences between the fracture and
also plastic morphologies of the scratches produced on the surface
of the clearcoats containing nanoparticles and polysiloxane additives. Scratches produced on the sample containing nano-b have
harsher edges as well as larger number of detached coating particles. Because of this, scratches produced on the surface of the
clearcoats containing nanoparticles inuence coating appearance
greater than other samples (specially the ones containing polysiloxane additives).
From the above explanations, it can be clearly understood that
gonio-spectrophotometry can be used to distinguish two types of
fracture scratches. Moreover, gonio-spectrophotometry technique
has the added advantages of being capable of easily quantifying
variations between these scratches of in term of the magnitude of
L and b values.
3.1.3. Reectance measurement
Scratch morphology can extremely inuence the appearance
properties of the clearcoats. It seems the morphology of the
scratches may signicantly affect the reective properties of the

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

1.5

1.5

0.5

0.5

550

-0.5

-0.5

-1

Si-2-a-20

Si-2-b-20

Si-2-c-20

Si-2-d-20

Si-2-b-110

Si-2-c-110

Si-2-d-110

-1
Blank-110

Blank-20

-1.5

-1.5

300

400

500

600

700

800

300

400

Wavelength (nm)

500

600

700

800

Wavelength (nm)
1.5

0.5

0.5

1.5

Si-2-a-110

0
-0.5

-0.5
-1

Si-1-a-20

Si-1-b-20

Si-1-c-20

Si-1-d-20

-1

Si-1-a-110

Si-1-b-110

Si-1-c-110

Si-1-d-110

Blank-75

Blank-20
-1.5

-1.5
300

400

500

600

700

300

800

400

500

600

700

800

Wavelength (nm)

Wavelength (nm)

Fig. 10. Variations of R (reectance difference) versus wavelength of clearscoats containing Si-2 (at different contents of ad) at four goniometric angles of observation.

clearcoat. Variations of R (R2 (reectance before scratch) R1

(reectance after scratch)) versus wavelength of the clearcoats
containing polysiloxane additives and nano-SiO2 are shown in
Figs. 910.
As it can be clearly seen in Figs. 9 and 10, decrease in wavelength causes increase in values of R (at 20 ) of the clearcoats
containing nano-SiO2 and polysiloxane additives. Maximum variations of the values of R can be seen at nano-b, Si-1-a and Si-2-a
(at 20 ). Variations of R of the clearcoats containing nano-b, Si1-a and Si-2-a at 110 (non-specular angle) is not as signicant
as 20 (specular angle). This indicates that variations of R after
scratch test cannot be attributed to the color changes. Light scattering in visible region is responsible for the signicant changes of
the value of R at low wavelengths (specially at blue zone of visible light). Dependency of R on wavelength for nano-b, Si-2-a and
Si-2-a demonstrates a signicant increase in scattering of light in
the blue zone of the visible spectrum which is indicative of fracture
morphology of scratches at these loadings of additives as well as

increase in number of detached coating particles. In contrast, nondependency of R on wavelength demonstrates plastic behavior of
scratches containing lower number of detached coating particles.
In general, there is a good agreement between the reectance properties of the clearcoats and other techniques in estimating scratch
morphology and its effect on appearance [37,41].
3.1.3.1. Self-healing measurement. It has been shown that using
gonio-spectrophotometer, the morphology of scratch can be easily studied. Using this technique, fracture and plastic scratches
were detected on the samples containing low and high loadings
of nano-SiO2 and polysiloxane additives respectively. It is well
known that plastic type scratches have higher capability of selfhealing behavior compared to fracture type when they are exposed
to the temperatures near to Tg (transition glass temperature) of the
clearcoat. At the temperatures close to clearcoat Tg , the resin chains
can easily dissipate internal stresses produced during scratch test.
This can be responsible for signicant decrease in number and size

Table 3
Self healing ability of scratches in terms of E, L, a and b at two goniometric angles of observation of 20 and 110 for coatings containing various contents of nano silica.
Sample

Nano-0
Nano-a
Nano-b
Nano-c
Nano-d

Parameter
a 20

a 110

b 20

b 110

L 20

L 110

E 20

E 110

0.17
0.297
0.033
0.083
0.180

0.153
0.208
0.055
0.234
0.266

0.491
0.588
0.415
0.244
0.40

0.002
0.10
0.331
0.042
0.04

0.612
0.629
0.565
1.992
0.165

0.06
0.076
0.323
0.154
0.064

0.802
0.910
0.701
2.008
0.468

0.164
0.242
0.465
0.283
0.272

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

4. Conclusion

1
0.8

Nano-c

0.6
0.4

Blank

0.2

551

0
-0.2

Nano-b

-0.4
-0.6

Blank
Nano-b

-0.8

Nano-a
Nano-c

Attempts have been carried out to use new approach to investigate the scratch resistance as well as the morphology of the
scratches of the automotive clearcoats containing nano-SiO2 and
polysiloxane additives. To this end, the effects of scratches on the
appearance of the coating and the morphology of scratches were
studied according to spectrophotometric/colorimetric parameters.
The results obtained in this study are listed below:

Nano-d

-1
300

400

500

600

700

800

Wavelength (nm)
Fig. 11. Variations of R versus wavelength of the scratched clearcoats (containing
nano-SiO2 ) after self healing.

of plastic type scratches. As a result of this, the surface morphology and therefore appearance of the clearcoat can be signicantly
affected after self-healing. The appearance changes can be easily
studied by gonio-spectrophotometer. To this end, the scratched
samples were kept at 60 C for 4 h. Variations in E, L, a and
b values (i.e. differences in L*, a* and b* after and before self healing) for various loads of nano silica and at two goniometric angles
of observation of 20 and 110 are shown in Table 3.
As it can be seen in Table 3, the values of a and b are
not signicantly changed after self-healing process. However, the
values of E and L of samples are signicantly changed after
self-healing. Based on the results shown in Table 3, the values of
E 20 and L 20 of the clearcoats containing nano-a and
nano-b loadings of nano-SiO2 are reduced after self-healing. However, an increase in values of these parameters is obtained at
nano-c loading of nano-SiO2 . As it has been previously discussed,
the number and size of scratches with high self-healing ability
will be decreased after self-healing process. This can also result in
an increase in clearcoat lightness. Therefore, increase in values of
E 20 and L 20 of the clearcoat containing nano-c can reveal
high self healing ability of scratches produced on this sample. This
observation is in agreement with the results shown in Fig. 7 which
scratches produced on the sample containing nano-c are mostly
in form of plastic type scratches. Results can show lower values
of E 20 and L 20 of the clearcoats containing nano-a and
nano-b loadings of nano-SiO2 compared to nano-c. This observation can reveal lower self healing ability of the scratches produced
on nano-a and nano-b samples compared to nano-c. In fact, the
scratches produced on nano-a and nano-b samples are mostly in
form of fracture type with low capability of self-healing. According
to the above explanations, gonio-spectrophotometry can be considered as a sound technique for studying self healing properties of
the scratches.
The self-healing properties of scratches can be also studied by
measuring the variations of reectance of coatings after and before
self healing process. Variations in R (R2 (reectance after selfhealing) R1 (reectance before self healing)) of the clearcoats are
shown in Fig. 11.
Results shown in Fig. 11 can reveal higher values of R of
the clearcoat containing nano-c load of nano silica than nano-a
and nano-b. This can show greater self healing capability of the
scratches produced at nano-c loading. Additionally, the low values of R of the coating containing nano-b load of nano silica
can demonstrate low self healing ability of this sample. These
deductions completely conrm the results obtained from SEM
micrographs. Therefore, reectance is another favorable parameter like L, E, a and b derived from gonio-spectrophotometry
which can easily quantify scratch morphology as well as self healing
properties of the scratches.

Results showed that scratch resistance of the clearcoat was signicantly improved using certain value of each polysiloxane
additive or nano-SiO2 . The fracture deformation at low loadings
and plastic deformation at high loadings of each additive were
obtained. It was also found that, coating containing nano particles
had greater tendency to form harsh fracture scratches.
It was found that, using gonio-spectrophotometry, not only
the scratch morphology but also the differences between two
types fracture morphologies can be studied. Two different types
of fracture deformations (one only including detached coating
particles and another including detached coating particles and
sharp edges) were differentiated according to spectrophotometric/colorimetric parameters (b and L).
There was good agreement between the results obtained
by gonio-spectrophotometry and microscopic techniques of
scratched coatings in the case of using different kinds of additives.
Using gonio-spectrophotometry, the self healing behavior of different types of scratches of the clearcoats containing nano and
polysiloxane additives were investigated. The greater capability
of healing of the scratch was observed when the morphology of
scratches was in form of plastic. This was observed by the increase
in the value of the reectance of the coating after self-healing.
References
[1] A. Karbasi, S. Moradian, N. Tahmassebi, P. Ghodsi, Prog. Org. Coat. 57 (2006)
175182.
[2] N. Tahmassebi, S. Moradian, Polym. Degrad. Stab. 83 (2004) 405410.
[3] U. Schulz, V. Wachtendorf, T. Klimmasch, P. Alers, Prog. Org. Coat. 42 (2001)
3848.
[4] M. Kutschera, R. Sander, P. Herrmann, U. Weckenmann, JCT Res. 3 (2) (2006)
9197.
[5] V. Jardret, P. Morel, Prog. Org. Coat. 48 (2003) 322331.
[6] Y. Hara, T. Mori, T. Fujitani, Prog. Org. Coat. 40 (2000) 3947.
[7] E. Amerio, P. Fabbri, G. Malucelli, M. Messori, M. Sangermano, R. Taurino, Prog.
Org. Coat. 62 (2008) 129133.
[8] Y. Bautista, M.P. Gmez, C. Ribes, V. Sanz, Prog. Org. Coat. 70 (4) (2011) 358364.
[9] V. Jardret, R. Ryntz, JCT Res. 2 (8) (2005) 591598.
[10] B.J. Briscoe, P.D. Evans, E. Pelillo, S.K. Sinha, Wear 200 (1996) 137147.
[11] Y. Bautista, J. Gonzalez, J. Gilabert, M.J. Ibanez, V. Sanz, Prog. Org. Coat. 70 (4)
(2011) 178185.
[12] Y. Bautista, M.P. Gmez, C. Ribes, V. Sanz, Prog. Org. Coat. 70 (4) (2011) 358364.
[13] M. Barletta, D. Bellisario, G. Rubino, N. Ucciardello, Prog. Org. Coat. 67 (2010)
209219.
[14] N. Tahmassebi, S. Moradian, B. Ramezanzadeh, A. Khosravi, S. Behdad, Tribol.
Int. 43 (3) (2010) 685693.
[15] B. Ramezanzadeh, S. Moradian, A. Khosravi, N. Tahmassebi, JCT Res.,
doi:10.1007/s11998-010-9239-4.
[16] G. Chen, S. Zhou, G. Gu, H. Yang, L. Wu, J. Colloid Interface Sci. 281 (2005)
339350.
[17] S. Zhou, L. Wu, W. Shen, G. Gu, J. Mater. Sci. 39 (2004) 15931600.
[18] S.J. Hinder, C. Lowe, J.T. Maxted, J.F. Watts, Prog. Org. Coat. 54 (2005) 104112.
[19] D. Hosseinpour, J.T. Guthrie, J.C. Berg, Prog. Org. Coat. 62 (2008) 214218.
[20] S. Rossi, F. Deorian, L. Fontanari, A. Cambruzzi, P.L. Bonora, Prog. Org. Coat. 52
(2005) 288297.
[21] S. Rossi, F. Deorian, M. Risatti, Surf. Coat. Technol. 201 (2006) 11731179.
[22] S. Rossi, F. Deorian, M. Risatti, Wear 261 (2006) 922931.
[23] Y. Reyes-Mercado, S. Rossi, F. Deorian, M. Fedel, Wear 265 (2008) 18201825.
[24] K.L. Rutherford, R.I. Trezona, A.C. Ramamurthy, I.M. Hutchings, Wear 203204
(1997) 325334.
[25] Z. Ranjbar, S. Rastegar, Prog. Org. Coat. 64 (2009) 387391.
[26] K. Yusoh, J. Jin, M. Song, Prog. Org. Coat. 67 (2) (2010) 220224.
[27] H. Pelletier, C. Mendibide, A. Riche, Prog. Org. Coat. 62 (2) (2010) 162178.
[28] H. Pelletier, C. Gauthier, R. Schirrer, Tribol. Int. 43 (2010) 796809.
[29] L. Lin, G.S. Blackman, R.R. Matheson, Mater. Sci. Eng. A 317 (2001) 163170.

552
[30]
[31]
[32]
[33]
[34]
[35]
[36]

B. Ramezanzadeh et al. / Progress in Organic Coatings 72 (2011) 541552

G. Wagner, M. Osterhold, Mater.-Wiss. U. Werkstofftech 30 (1999) 617622.

S. Rossi, F. Deorian, E. Scrinzi, Wear 267 (2009) 15741580.
A. Cambruzzi, S. Rossi, F. Deorian, Wear 258 (2005) 16961705.
L. Lin, G.S. Blackman, R.R. Matheson, Prog. Org. Coat. 40 (2000) 8591.
W. Shen, L. Mi, B. Jiang, Tribol. Int. 39 (2006) 146158.
W. Shen, B. Jiang, S.M. Gasworth, H. Mukamal, Tribol. Int. 34 (2001) 135142.
P. Bertrand-Lambotte, J.L. Laubet, C. Verpy, S. Pavan, Thin Solid Films 398399
(2001) 306312.
[37] M. Klanjsek Gunde, M. Kunaver, A. Hrovat, U. Cvelbar, Prog. Org. Coat. 54 (2005)
113119.

[38] G. Baba, K. Suzuki, Anal. Chim. Acta 380 (1999) 173182.

[39] R. Besold, Metallic effect characterization, parameter, and methods for instrumentally determination, Die Farbe 37 (1990) 7985.
[40] Colorimetry, CIE Publication No. 15.2, 2nd ed., CIE, Vienna, Austria, 1986.
[41] P.H. Gedam, P.S. Sampathkumaran, Prog. Org. Coat. 11 (1983) 313333.
[42] B. Ramezanzadeh, M. Mohseni, H. Yari, J. Coat. Technol. Res. 7 (6) (2011)
677689.
[43] Y. Hosoya, T. Shiraishi, M. Oshiro, S. Ando, M. Miyazaki, J.M. Powers, J. Dent. 37
(2009) 585590.