ThreeBond Technical News

Issued July 1, 2006

67

Surface Crack Detection Coating Agent

Introduction ________________________________________________
The present state of the art for identifying surface cracks is less established than is generally
believed. During on-site inspections, it is virtually impossible to detect surface cracks on the surfaces
of components having complex shapes, such as weld areas. Given the constant possibility of
overlooking surface cracks, a new method must be devised that can effectively assist unaided visual
tests and improve detection rates.
We propose surface crack detection coatings, ThreeBond 2050 (black capsules) and ThreeBond
2050B (orange capsules), jointly developed for practical use by the Independent Administrative
Agency National Maritime Research Institute and ThreeBond Co., Ltd. Figure 1 illustrates the scheme
of the surface crack detection coating system.
This issue will describe the mechanism underlying surface fatigue crack detection coatings and their
applications and introduce the results of basic performance and durability evaluation tests performed
on flat plates and welded joint test specimens.
Mechanism of the surface
[for ThreeBond 2050B (orange)]
crack detection agent

Crack detected!

Epoxy resin coat layer

Dyed oil
seeping out

Protective layer

Crack detection layer

Metal substrate

Fatigue
crack

When cracks form, capsules within the crack detection layer applied
to the substrate break, causing paint to seep out to the surface.
Microcapsules containing
dyed oils
shell material: gelatin;
core material: orange/black pigment dissolved oil
(average diameter: 80-120 m)

Figure 1. Schematic diagram of the surface crack detection coating system

Contents

Introduction .......................................................1
1 Demand for Surface Crack Detection
Agents ..........................................................2
2 Coating Method ............................................2
2.1 Surface Crack Detection Coating
Agent ....................................................2
2.2 Coating and Coating Method................3
3 Experiment ...................................................4

3.1 Performance Test Using Flat-Plate

Specimen ............................................. 4
3.2 Performance Test Using Welded
Joint Specimen..................................... 5
3.3 Performance Test using a large
L-Shaped Welded Specimen................ 5
3.4 Weathering Test ................................... 6
4 Research Themes and Future Prospects..... 7
Conclusion........................................................ 8

Demand for Surface Crack Detection

Agents

Although several non-destructive testing methods

have been proposed for identifying surface fatigue
cracks in various structures including magnetic
particle, penetrant, and eddy-current inspection
methods, most approaches continue to rely on
visual test. But the number of cracks that escape
visual test is much greater than commonly thought,
creating a demand for improved detection rates in
visual tests.
Also needed is a test method that would enable
inspectors to continuously and easily observe the
crack growth to determine the appropriate time of
repair work after crack detection. In addition, the
method is rquired to veridate the repair work
performed.
This issue presents a crack detection agent that
meets these needs. Figure 2 illustrates its underlying
mechanism.
Full-fledged
testing

Visual
tests

cracklike
formation
detected!

Magnetic
particle
inspection
Eddy-current
inspection
Penetrant
inspection

Facilitate verification by
visual test
Verification
of repair

Difficult

Problem
present!!

Repair
work
Reinforcing
Left
plate
untreated:
Stop hole
ObservaDifficulty
in
Failure in
tion
of
Gouging
detection
status of determining crack tip
growth
Improved detection rates
Enables easy observation of
in visual tests
the status of crack growth.

Figure 2. Intended Use of Crack Detection Agent

In practice, detecting surface cracks that have
formed on components with complex shapes such
as welded joints is quite difficult. To give one
example, such cracks are commonly encountered in
hull inspections. Based on the results of a survey of
maintenance and inspections of fatigue damage to
bulk carriers, current test methods detect
approximately 60% of existing cracks, rest of 40%
unidentified. The results also indicate that
approximately 30% of the surface cracks generated
exceed 200 mm in length, while 10% grow to
lengths exceeding 500 mm.1) One would expect
similar crack detection failures in marine vessels
other than bulk carriers, including crude oil carriers
and passenger liners. So we should devise a new
method that would provide effective assistance with
on-site visual tests and improve detection rates.
In the 10 years subsequent to 1990, researchers at
Lehigh University in the United States have studied
crack detection technologies using a special coating
agent called Smart Paint.2)
Smart
Paint
incorporates
microcapsules
containing pigments mixed into a normal epoxy

resin-based maintenance coating. When this coating

is applied to a base metal, the formation and growth
of cracks breaks the capsules, releasing the pigment
and subsequently coloring the coating surface to
facilitate detection of cracks. We have successfully
developed just such a surface crack detection
coating system and plan to market it as ThreeBond
2050 (black capsules) and ThreeBond 2050B
(orange capsules), discussed in this issue.
Field tests on steel bridges and marine vessels
were performed to validate the effectiveness of this
coating, confirming coloring at the surface cracks.3)
Results from our laboratory experiment provided
the effectiveness of the present surface crack
detection coating. To verify the scope of
applicability, we plan to perform testing to evaluate
its conformity to actual structures.4)

Coating Method

2.1 Surface Crack Detection Coating Agent

The surface crack detection coatings, ThreeBond
2050 (black capsules) and ThreeBond 2050B
(orange capsules), used in this study are epoxy
resin-based anticorrosion coatings that incorporate
pigment-filled microcapsules. The method for
preparing the microcapsules, their coating
characteristics, and coating procedures are
described below.
2.1.1 Preparing the Microcapsules

The core material is microfabricated by

dissolving an orange organic pigment in spindle oil
and encapsulating solution into the gelatin-forming
shell material. The microcapsule is prepared by the
coacervation method,5) then washed and dried.
Figure 3 shows the finished microcapsule image
observed by SEM (scanning electron microscope).
Particle size distribution analysis shows average
diameters of around 100 m, with most particles
plotting within the range of 80-120m.

(3) Add capsules and stir.

Figure 3. SEM image of the pigment-filled

microcapsules
2.2 Coating and Coating Method
The coating used was a two-part epoxy
resin-based anticorrosion coating. The coating is
prepared by first mixing the primary agent and the
curing agent at the prescribed ratio, then adding the
microcapsules and stirring thoroughly to achieve a
uniform mixture. The coating is then applied using
a brush at the center and near the weld bead tip of a
flat-plate specimen pretreated using a wire brush.
After the capsule-containing coating layer has fully
cured, a top coat of white epoxy resin-based coating
prior to mixing ThreeBond 2050 or ThreeBond
2050B capsules is applied to protect the capsules
distributed at the surface and to enhance contrast
upon coloring.

(4) Crack-detection layer.

Application process

Dry & curing time until protective layer is formed

Minimum

Maximum

5C

36 hours

5 days

10C

24 hours

4 days

20C

12 hours

3 days

30C

6 hours

2 days

(5) Leave to dry and cure.

12-24 hours at room temperature

(1) Stir primary agent.

(6) Application of protective coating

Add curing agent and stir.

(2) Crack-detection layer coating

Add curing agent and stir.

(7) Protective layer

Apply with brush.

Note: masking tape

If masking tape was sticked to the surface, peel off
the tape within around 30 minutes after the
application of the crack detection coating and
protective layer.

Experiment

3.1

Performance Test Using Flat-Plate

Specimen

3.1.1 Materials tested and test specimen

The materials tested for the flat-plate specimen

are a high-tensile steel SM490A (yield strength: 393
MPa; tensile strength 543 MPa) specimen and an
aluminum alloy A5083P-O. Figure 4 shows their
shapes and dimensions. A 10-mm long notch was
created at the center of the specimen by electrical
discharge machining.

observations using a CCD microscope (x25 to x175

magnification).
3.1.3 Test results

As shown in Figure 5, destruction of the capsule

and subsequent coloring were observed with crack
growth in the test specimen (high-tensile steel
SM490A).
Clear coloring was observed with the crack
growth, and we confirmed the absence of problems
with the crack detection mechanism. Average
coating thickness was approximately 200 m,
accounted for by a 100-m capsule-mixed coating
layer and a 100-m top coat layer.
The tests for the aluminum alloy (A5083P-O)
specimen produced similarly excellent results.

After drying of top coat

N = 480,000

Figure 4. Flat-plate test specimen with notch

(Photograph courtesy of the Independent Administrative Agency
National Maritime Research Institute)

3.1.2 Test method

An electronic-hydraulic servo fatigue tester with a

dynamic capacity of 10 tonf was used in the fatigue
test, and the fatigue crack growth test was
performed under a complete pulsating repeated
tensile load at a frequency of 4 Hz and zero load
ratio. With each preset repetition cycle, the load
amplitude was reduced while maintaining the
maximum load, and beach marks (striations
indicating the shape of the crack tip) were formed
on the fracture surface while we performed in situ

N = 600,000
Figure 5. Observation of capsule destruction
and coloring accompanying crack
growth
(Photograph courtesy of the Independent Administrative Agency
National Maritime Research Institute)

3.2

Performance Test Using Welded Joint

Specimen

3.2.1 Materials tested and test specimen

The material tested for the welded joint specimen

was KA 32 TMCP steel conforming to NK
standards (plate thickness: 10 mm). On both
surfaces of the 50-mm-wide main plate, 100-mm
stiffener plates with the same thickness were
attached by corner boxing fillet welding. Figure 6
shows the shape of the test specimen.

3.3

Performance Test using a large

L-Shaped Welded Specimen
The data was provided by the group under
Associate Professor Kainuma at Gifu University
(currently at Kyushu University).
3.3.1 Materials tested and test specimen

The material tested for the large L-shaped welded

specimen was steel (yield strength: 393 MPa;
tensile strength: 555 MPa, elongation: 32%). Figure
8 shows its shape, dimensions, and welded
dimensions.
Table 1. Mechanical properties and chemical composition (%) of the steel material
Yield
strength
(MPa)

Tensile
strength
(MPa)

398

555

Impact
value
(J)

Si

Mn

Cu

Ni

Cr

(%)
32

281

0.15

0.27

1.33

0.09

0.01

0.02

Elongation

Figure 6. Welded joint test specimen

(Photograph courtesy of the Independent Administrative Agency
National Maritime Research Institute)

3.2.2 Test method

For
the
fatigue
test,
we
used
an
electronic-hydraulic servo fatigue tester with a
dynamic capacity of 10 tonf, performing the fatigue
crack growth test under a complete pulsating
repeated tensile load at a frequency of 4 Hz and a
load of 0-58 kN. With each preset repetition cycle,
the load amplitude was decreased while maintaining
the maximum load. Beach marks (striations
indicating the shape of the crack tip) formed on the
fracture surface while we performed in situ
observations using a CCD microscope (x25 to x175
magnification).
3.2.3 Test results

Figure 8. Detailed dimensions of the test

specimen shape and welded joint
3.3.2 Test results

As shown in Figure 9, we observed the test

specimen for capsule destruction and coloring
accompanying surface crack growth. Table 2
presents the results of the test. The difference in the
length of the cracks and that of the colored region is
less than 2 mm.

As shown in Figure 7, the capsule destruction and

subsequent coloring were observed with crack
growth in the test specimen.

6mm

Figure 9. Crack observations after 50,000

repeated stress cycles
Table 2.

Figure 7. Observation of capsule destruction and

coloring in the welded joint specimen
(Photograph courtesy of the Independent Administrative Agency
National Maritime Research Institute)

Observation of crack length and

length of colored region

Test
specimen

Stress repetition
(x10,000 cycles)

Crack length

A-1
A-2
A-3
B-1
B-2
B-3

5
3
5
11
8
26

68
11
7
18
6
32

Crack length (mm)

Colored region
length
67
11
6
18
4
34

3.4

Weathering Test

3.4.1 Materials tested and test specimen

The material tested was an aluminum plate

(A1050P) with dimensions of 1x60x100 mm
(Figure 10). Both surface crack detection coatings,
ThreeBond 2050 and ThreeBond 2050B, were used
in the test.

Orange (after one month)

Cut made by a small blade

Cut made by the back of a small blade (approx. 0.5 mm in width)

Figure 10. Shape and dimension of the test

specimen for the weathering test
3.4.2 Test method

The test specimen was exposed to an outdoor

environment for the time specified by JIS
K-5600-7-6 or placed in a weather meter for a
preset duration. The surface was then cut with a
small blade for visual confirmation of coloring.
Exposure test condition: facing south at a 45
angle

Black (after 26 months and still ongoing)

Figure 11. Confirmation of coloring in outdoor
exposure test specimen by visual
test
Table 3. Results of durability test
Test items

Test conditions

3.4.3 Test results

As shown in Figure 11, the coloring was observed

in the outdoor exposure specimen by visual test. We
confirmed that the orange pigment remained
effective for about one month and the black for
more than 26 months.
Coloring was also confirmed in the test specimen
for the weather meter after 1,000 hours of exposure.
Table 3 provides the results in various durability
performance tests.

Weather meter

1,000 hours

Heat and cold

resistance test

-40C to 80C 1, 000

hours

Salt spray test

1,000 hours

(Salt water spray 35C

4 hours hot-air
Combined cycle drying; 60C 2 hours
test
humid conditions
50C 95% RH 2
hours) 100 cycles

Outdoor
exposure test

Choshi Exposure Laboratory, Chiba prefecture

JIS K-5600-7-6

Results
(1) Fading of orange
coloring
(2) Black coloring: OK
(3) Coloring capacity
of uncolored
region: OK
Colored region:
identifiable

Coloring capacity of
uncolored region: OK

(1) Fading of orange

coloring after one
month
(2) Black coloring
effective after 26
months
(evaluation still
underway)
(3) Coloring capacity
of uncolored
region: OK

Research Themes and Future

Prospects

Our study confirmed that the two types of

coatings used in this study, ThreeBond 2050 and
ThreeBond 2050B, offer the basic performance
required for a surface fatigue detection coating
capable of assisting in visual tests by facilitating
crack detection and potentially improving detection
rates significantly. In future studies, we plan to
challenge the following themes:
(1) Pigment weather resistance
(2) Application to non-metal systems
(3) Detection of micro surface fatigue cracks
(4) Detection of stress distribution

The surface crack detection agent set contains the

following materials:
Epoxy resin (Primary agent 680 g 2 , curing
agent 120 g 2)
Microcapsules (200 g; black or orange)
Solvent (300 g; for dilution)
Brush (2; one for crack detection coating and
another for protective layer application)
Stirrer (2)
* Total area of application: 15,000-20,000 cm2
(120-140 centimeters in
height and width)

Conclusion
Acknowledgements
The research described here was undertaken as a joint development project with the Independent Administrative
Agency National Maritime Research Institute. We gratefully acknowledge the institute staff members as well as
everyone who participated in the study.
In particular, we wish to thank Dr. Ichihiko Takahashi for his guidance and cooperation.
We also wish to express our gratitude to Associate Professor Shigenobu Kainuma at Kyushu University for his
invaluable support and assistance as we performed our evaluations of surface crack detection coatings.
Hideki Takahashi
Industrial Materials Development Division
Development Department
Tomio Onoguchi
Research Planning Division
Research Department
Research Center
ThreeBond Co., Ltd.
Mitsuhiko Uchida
R&D
ThreeBond International, Inc. (Cincinnati)

1)
2)
3)

4)

5)

References
The 226th meeting of the Shipbuilding, Research Association of Japan: Investigation of Rational Hull
Structure Design for Bulk CarriersFinal Report (in Japanese) (1996)
Roy R. Miron, Wayne Bilder and William Lademan, SmartPaint: A Structural Crack Monitoring Method,
NACE (National Association of Corrosion Engineers), Northern Area Eastern Conference, Ottawa (1999)
Ichihiko Takahashi, Michio Ushijima, Shigeru Akiyama, Atsushi Takada, and Noriyuki Kotani: Detection of
Fatigue Cracks with a Paint System for Inspections (in Japanese), Proceedings of the FY 2002 (2nd) Research
Presentation Meeting of the Independent Administrative Agency National Maritime Research Institute, pp.
137-140 (2002)
Ichihiko Takahashi, Michio Ushijima, Shigeru Akiyama, Atsushi Takada, Tomio Onoguchi, and Mitsuhiko
Uchida: Prototype and performance evaluations of a paint system for fatigue crack detection (in Japanese),
Techno Marine Journal of the Society of Naval Architects of Japan, vol. 871, pp. 70-73 (2003)
Tamotsu Kondo: Microcapsules (in Japanese) (Kyoritsu Shuppan Co., Ltd., Chemical One Point 13)

1456 Hazama-cho, Hachioji-shi, Tokyo 193-8533, Japan

Tel: 81-426-61-1333