Content uploaded by Marta Klanjšek Gunde
Author content
All content in this area was uploaded by Marta Klanjšek Gunde on Jan 25, 2019
Content may be subject to copyright.
1
Colour changes of UV-curable thermochromic inks
Rahela Kulčar1, Mojca Friškovec2, Nina Knešaurek1, Barbara Sušin2, Marta Klanjšek Gunde3
1University of Zagreb, Faculty of Graphic Arts, Getaldićeva 2, Zagreb, Croatia
E-mail: rkulcar@grf.hr; nknesaurek@grf.hr
2Cetis, Graphic and Documentation Services, d.d., Čopova 24, Celje, Slovenia
E-mail: mojca.friskovec@cetis.si; barbara.susin@cetis.si
3National Institute of Chemistry, Hajdrihova 19, Ljubljana, Slovenia
E-mail: marta.k.gunde@ki.si
Abstract
UV-curable leucodye-based thermochromic inks with activation temperature of 31 °C in blue, red and
black shades were analysed. The colour of samples does not depend only on their temperature but also on
thermal history – they show colour hysteresis. Similar hysteresis loops were obtained for the red and blue
samples and narrower for the black sample, as measured by cycling in 14–40–14 °C region. The blue and
black samples have approximately closed loops but a small colour difference was obtained even after one
cycle for the red one thus the loop remains slightly opened. The colour of all samples also depends on the
UV dose applied at curing. This difference was followed over the whole hysteresis loop. The largest
values were measured for the red sample where the CIEDE2000 total colour difference rises above 4 and
remains well observable all over the temperature region applied for cycling. One of the possible reasons
for this effect might be different degree of polymerisation of the polymer matrix achieved by different
UV curing.
Keywords
Thermochromic inks, UV curing, screen printing, colorimetry, dynamic colour, hysteresis
1. Introduction
Colour-changing inks become increasingly important for various applications in graphic art such as smart
packaging, security printing and marketing. To achieve precise appearance and reasonable durability of
prints, much knowledge about the dynamic properties is required. Extended research of dynamic
appearance of the final product is presently covered only very approximately or not at all.
Liquid crystals and leuco dyes are thermochromic material types which are currently the most frequently
used in printing inks. We have studied here leucodye-based printing inks. They consist of two-phases: a
thermochromic phase is dispersed in a non-thermochromic polymer matrix, a resin. A thermochromic
phase is realised in microcapsules which are mostly named simply as pigments. Both components build
separate phases and it is supposed that they don't influence the other component. Leucodye pigments
consist of a mixture of leuco dye, developer and solvent, which are microencapsulated in a protective
polymer coating. Each microcapsule contains the entire system required to reproduce the dynamic colour.
They are more than 10-times larger than pigment particles in conventional printing inks. They are not
completely inert and insoluble and show practically negligible scattering effect. The activation
temperature is determined by the temperature where the solvent changes from solid to liquid state. This
phase change inhibits the colour-forming reaction between dye and developer below the transition
temperature and protects it well above this temperature. The colouration/discolouration process requires
phase changes of the entire solvent in each microcapsule1. Therefore, a rather broad temperature range is
needed for the process to be completed in all microcapsules. The colouration/discolouration process is
regarded to be reversible and it is believed that it can be repeated a few thousand times2. Leucodye-based
thermochromic inks with various activation temperatures are available, from -15 C up to 65 C.
Nevertheless, most applications are limited to three standard temperature ranges, cold (10 C), human-
2
touch (31 C) and warm (43 C)3. All major ink types such as water-based and photocuring for paper,
plastics and textile are available.
Only few scientific studies applying leucodye thermochromic inks were reported until now. The most
important is the work of Linda Johansson from University of Linköping, Sweden4-8. She investigated
printed dynamic images with completely different images in different temperature ranges. For this
purpose, the leucodye thermochromic inks were combined with conventional process ink layers to obtain
multicoloured printing dynamic images. Knowledge about the properties of leucodye-based
thermochromic ink has to be enhanced for several areas of commercial applications1. The most important
are the temperature-dependent properties of this complex system, the degree of its reversibility and the
possible factors that may influence on it.
We have studied dynamic colorimetric properties of leucodye-based thermochromic inks. UV-curable
thermochromic inks with activation temperature of 31 C in three hues were applied. Special attention
was devoted to characterize the colour hysteresis and the meaning of the activation temperature. The
influence of UV curing parameters was also analysed.
2. Experimental
UV-curable thermochromic screen printing inks UV TCX (Coates Screen Inks GmbH, Germany) in red
(UV TCX R-31), blue (UV TCX B-31) and black (UV TCX N-31) shades were applied. They consist of
acrylate UV-curable vehicle and leucodye thermochromic pigments with activation temperature (TA) of
approximately 31 °C. The inks were screen printed applying SEFAR PET 1500 high-modulus
monofilament polyester mesh 120/34Y on gloss-coated paper (120 g/m2). The off-prints were cured
applying UV dryer Aktiprint L by UV energy of approximately 0,83 W/cm2 at two different speed of
conveyor belt, 22 and 8 m/min. At these speeds the maximum curing energy falling on the sample are
approximately 150 and 400 mJ/cm2, respectively. The UV integrator (Technigraf, Germany) was applied
for these measurements.
Spectral reflectance of all prepared samples were measured by UV-VIS-NIR spectrophotometer Lambda
950 (Perkin-Elmer) with 150 mm integrating sphere applying (8°:di) measuring geometry (diffuse
geometry, specular component included)9. The temperature of the sample was varied by the help of
copper plate, which was heated/cooled by circulation of thermostatically controlled water in tubes inside
the plate (Figure 1). The reflectance spectra were measured every 2 °C at heating from 14 °C up to 40 °C
and then at cooling down to 14 °C. Colorimetric calculations were done in CIELAB space applying D50
illuminant and 2° standard observer. The total colour differences were calculated according to the
CIEDE2000 colour difference equation.
Figure 1. Thermostatic circulator and the front view of the copper plate with sample.
3
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
400 450 500 550 600 650 700
Wavelength [nm]
Reflectance
40°C
14°C
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
400 450 500 550 600 650 700
Wavelength [nm]
Reflectance
14°C
40°C
Figure 2. Spectral reflectance curves for the black sample measured at heating (top) and cooling
(bottom). The spectra were measured in 2°C interval.
The degree of polymerization of the acrylic vehicle was analysed applying infrared spectroscopy. For this
reason a thin layer of thermochromic ink was applied on a ZnSe (Irtran-4) optical window. The as-
deposited layer and layers cured by UV doses of 150 and 400 mJ/cm2 were measured in the transmittance
mode. The degree of polymerisation was followed by the intensity reduction of the C=C bond of the
unsaturated acrylate group located at 1409 cm-1. When the vehicle is fully polymerised, these bonds
dissappear10,11.
4
3. Results ad discussion
3.1. Changes of colour in dependence on temperature
The CIELAB values L*, a*, and b* in dependence on temperature of printed samples are shown in
Figures 3 and 4. These figures show that the colour of samples does not depend only on temperature, but
also on the thermal history, i.e. whether this temperature was reached during heating or during cooling.
This phenomenon is known as hysteresis. Our samples become discoloured during heating and coloured
again during cooling. The process is illustrated by change of lightness L* as a function of temperature
(Figure 3). When a sample is heated, L* becomes higher. At TA it rises above 80 whereas at somewhat
larger temperatures the sample seems to be completely discoloured and L* values remain approximately
the same with further increasing of temperature. When the same sample is cooled, the reverse process (i.e.
colouration) occurs at much lower temperatures. The entire L*(T) curve has a form of a loop, known as
hysteresis loop. Similar loops were obtained also for C*(T) but in opposite direction. During heating the
C* diminishes and rises again at cooling. The samples studied here do not show equal colour hysteresis:
they differ in the temperatures where the loop starts and finishes, in its steepness and area.
If a completely reversible process is assumed, a thermochromic sample should return to the same colour
after the whole heating/cooling cycle. Hysteresis loop of such a sample is closed. We have obtained
approximately closed loops for black and blue samples. The loop of the red sample remains opened,
where the final L* is somewhat lower than the initial.
14 16 18 20 22 24 26 28 30 32 34 36 38 40
40
50
60
70
80
90
L*
Temperature ( C)
TA
Figure 3. CIELAB lightness L* of red (triangles), blue (squares) and black (circles) thermochromic
samples in dependence on temperature at heating (solid signs) and cooling (open signs). The samples
were cured applying UV dose of 400 mJ/cm2. The activation temperature is denoted by TA.
During heating/cooling cycle the a*, b*, L*, and C* values of a thermochromic sample describe a path on
the (a*,b*) and (L*,C*) planes (Figure 4). These paths are in general not the same at heating and cooling,
but differences are small. It can also easily be recognized that the discoloration is not complete – all
systems shown here retain similar yellow shade even at the highest temperature applied here (40 °C).
5
-5 0 5 10 15 20 25 30 35 40
-20
-10
0
10
20 400C
140C
140C
140C
b*
a*
0 5 10 15 20 25 30 35 40 45 50
30
40
50
60
70
80
90 400C
140C
140C
140C
L*
C*
Figure 4. Changing of CIELAB values of red (triangles), blue (squares) and black (circles)
thermochromic samples in (a*,b*) plane (top) and (C*,L*) plane (bottom) at heating (solid signs) and
cooling (open signs). All samples shown here were cured applying UV dose of 400 mJ/cm2.
The obtained hysteresis has to be shown for all three values of the colour space simultaneously, thus by a
4-dymensional graph. This was simplified by the colour gradient, i.e. the rate of colour change (expressed
by CIEDE2000 colour difference) between the two adjacent measurements. It was evaluated in
dependence on temperature. Such a graph represents the first derivative of the corresponding three-
dimensional hysteresis. These graphs are shown in Figure 5. The largest colour gradient vas obtained for
black samples, where it is smaller at heating and larger at cooling. The corresponding colour gradients for
red and blue samples are smaller and approximately similar for heating and cooling. This figure may help
6
in explaining the meaning of TA at heating. For the black sample the TA corresponds to the temperature
where the colour gradient is maximal but for the red and blue samples it is located at the nearest
neighbour measuring point after the peak of colour gradient. In the reverse direction (when the sample is
cooled) the colour gradient peaks much more apart form the TA, at smaller temperatures.
0,00
5,00
10,00
15,00
20,00
25,00
30,00
14H-16H
16H-18H
18H-20H
20H-22H
22H-24H
24H-26H
26H-28H
28H-30H
30H-32H
32H-34H
34H-36H
26H-38H
38H-40H
40H-38C
38C-36C
36C-34C
34C-32C
32C-30C
30C-28C
28C-26C
26C-24C
24C-22C
22C-20C
20C-18C
18C-16C
16C-14C
Temperature (°C)
CIEDE2000
just perceptible
perceptible
large
TA
TA
Figure 5. The total colour change (CIEDE2000) of the red, blue and black samples as obtained between
two adjacent temperatures at heating (H) and cooling (C) (e.g. 26H-28H denotes the total colour
difference measured on the same sample when it is heated from 26 to 28 C). All samples were cured
applying UV dose of 400 mJ/cm2.
The samples that were cured at different UV doses have somewhat different colour. The effect has to be
analysed by the help of colour hysteresis. The colour differences between samples shown above and that
cured at UV dose of 150 mJ/cm2 in dependence on temperature are shown in Figure 6. This difference is
the largest in temperature regions where big colour changes occur. The largest differences were obtained
for the red ink where it is above the observable limit all over the entire measurement region and rises
above 4,0 at the largest rate of colour change.
7
16H20H24H28H32H 36H 40H 36C 32C 28C 24C 20C 16C
0
1
2
3
4TA
CIEDE2000
Temperature ( C)
TA
Figure 6. CIEDE2000 colour difference between red (triangles), blue (squares) and black (circles)
thermochromic samples that were cured with UV dose of 400 and 150 mJ/cm2 in dependence on
temperature at heating and cooling. The temperature at heating is denoted by H and at cooling by C (e.g.
28C denotes 28 C on cooling).
3.2. Degree of polymerisation
The samples cured at different UV dose do not exhibit the same colour. The CIEDE2000 colour
difference between weakly (150 mJ/cm2) and hardly (400 mJ/cm2) cured samples with the same printed
ink is shown on Figure 6. The possible reason for the effect was analysed by infrared spectroscopy.
Infrared spectra of the thermochromic inks depend on chemical structure of the ink. Absorption of UV
light applied at curing process causes crosslinking of monomers and oligomers in the liquid resin whereby
a solid polymer is formed. Therefore in the infrared spectra some peaks diminish and some new may
appear. The effect on our samples is shown in Figure 7. The peak at 1409 cm-1 due to double-bond
vibrations (C=C) of the acrylic group was applied here to measure
, the degree to which the
unpolymerized sample was conversed to the polymerized state:9
0
1AAE
where AE and A0 are the infrared absorbance of the exposed and unexposed samples at the representative
peak, respectively. This gives
= 0,87 for the lower UV dose and
= 0,91 for the higher UV dose.
8
2000 1800 1600 1400 1200 1000 800 600
Absorbance
Wavenumbers (cm-1)
0,2
liquid
150 mJ/cm2
400 mJ/cm2
1409
1500 1450 1400 1350
=0
=0,91
Absorbance
Wavenumbers (cm-1)
0,2
liquid
150 mJ/cm2
400 mJ/cm2
1409
=0,87
Figure 7. Infrared spectra of differently cured samples. After UV curing the peak at 1409 cm-1
diminishes. The effect is larger for higher UV dose, representing higher degree of polymerisation of the
acrylic vehicle.
4. Conclusions
The screen-printed UV-curable leucodye thermochromic inks with activation temperature of 31 °C were
investigated. The variation of colour as a function of temperature was measured and analysed.
No complete discoloration was obtained well above the activation temperature. This is not necessarily a
consequence of imperfect discoloration, but could also be a combined effect of the vehicle, the properties
of the thermochromic system in microcapsules and of the resin applied for microencapsulation. Similar
yellow colour was obtained for all samples.
All the analysed samples show colour hysteresis: their colour does not depend only on temperature but
also on the way this temperature was reached. A system with hysteresis has memory, i.e. it exhibits a
path-dependence. The colour of samples prepared by a thermochromic ink is different when the same
temperature is reached by heating or by cooling. Different colour hystereses were obtained. The steepest
changes and the largest colour gradient were measured for the black sample and similar for the blue and
red samples.
The activation temperature describes the temperature where a heated sample undergoes the largest colour
change. This is well obeyed for black sample and more loosely for red and blue ones. When a sample is
cooled, the activation temperature fails any meaning – the colouration process occurs at considerably
lower temperatures. More-or-less closed loops were obtained in the measured temperature cycle, with
observable opening for the red sample only. This shows that a small colour shift may retain after the
entire temperature cycle.
When the layers are prepared by the same ink and cured by different UV dose they may have different
colour. The effect is the largest for the red samples. A possible reason was investigated by infrared
spectroscopy. The infrared spectra of thermochromic ink depend on the amount of UV energy applied to a
sample at curing. Our results show that the degree of polymerization of the polymer matrix increases in
samples cured with higher UV dose. This could be one among possible reasons for the colour difference
obtained in differently cured samples.
Further research of the complex two-phase system is needed. It is very likely that the properties of colour
hysteresis characterize the details of the microstructure inside thermochromic layers and is therefore the
most important property of such samples. Special attention should be devoted to the span of temperatures
applied for cycling and to durability of the colouration/discolouration processes. The influence of the UV
energy applied for curing on the microstructure has to be analysed carefully. It is very likely that the
change in crosslinking of the host polymer is only one among several reasons for the observed effect.
9
5. Acknowledgements
Mojca Friškovec acknowledges the Slovenian Technology Agency for young researcher support,
operation part financed by the European Union, European Social Fund. Operation implemented in the
framework of the Operational Programme for Human Resources Development for the period 2007-2013,
priority axis 1: promoting entrepreneurship and adaptability, Main type activity 1.1.: Experts and
researchers for competitive enterprises.
6. References
1. Arno Seeboth and Detlef Lötzsch: “Thermochromic Phenomena in Polymers”, Smithers Rapra
technology Limited, Shawbury, UK, 2008.
2. J. Homola: “Color-changing inks”, Color Change Corporation, 2003;
http://www.screenweb.com/index.php/channel/6/id/1425/, http://www.mhest.com/supparticles/Color-
changing-inks.pdf .
3. “Thermochromatic effects”, Printcolor high performance inks,
http://www.apcis.fr/ft/printcolor/documents/Thermochromatic_Effects.pdf.
4. L. Johansson: “Chromatic properties of thermochromic inks”, Proceedings of TAGA Conference,
San Antonio, 2004.
5. L. Johansson, B. Kruse: “Colour separation with dynamically changeable inks”, Proceedings of the
TAGA Conference, Toronto, 2005.
6. L. Johansson, B. Kruse, “The influence of paper properties on colour reproduction with dynamics
inks”, Proceedings of the CSIST/IS&T 2005 International Conference on Imaging: Technology &
Applications for the 21st Century, Beijing, 2005.
7. L. Johansson, B. Kruse: “A colour separation strategy for reproduction of printed dynamic image for
paper substrate”, Proceedings of the 32nd International iarigai Research Conference, Porvoo, 2005.
8. L. Johansson, “Creation of printed dynamic images”, Linköping Studies in Science and Technology,
Thesis No. 1234, Norköpping 2006.
9. “Colorimetry”, 3rd edition, CIE 15:2004. Vienna (2004).
10. V. Landry, B. Riedl, P. Blanchet: „Nanoclay dispersion effects on UV curing coatings“, Prog. Org.
Coat. 62 (2008) 400-408.
11. N.B. Colthup, L.H. Daly, S.E. Wiberley: Introduction to Infrared and Raman Spectroscopy,
Academic Press, Boston, San Diego, New York (1990).
