Laser welding of polymers using unsymmetrical squaraine dyes

Article abstract of DOI:10.1002/pola.29196

Transmission laser welding of plastic is typically achieved by incorporating a laser absorbing dye into one of the two pieces of plastic one wishes to weld together and mediating the welding process using a laser emitting light at a suitable wavelength. Desirable properties of laser absorbing dyes include: (1) the dye is colorless in the visible region but absorbs strongly in the region at the wavelength where the laser emits, (2) the dye has a good conversion of the energy from the laser to heat, (3) the dye is stable enough to be incorporated into the molten plastic, (4) the dye is soluble in the plastic of choice to give an even distribution in the plastic, and (5) the dye is nontoxic. Here, we explore an alternative approach to achieve laser welding of colorless plastics using a laser dye that is colored, but that can be bleached after the welding process. Nonsymmetrical squaraine dyes were prepared by condensation of squaric acid and two equivalents of various 3,5-dimethoxy-N,N-dialkylanilines. After the incorporation into polyethylene and laser welding, the plastic was bleached either by heat treatment or by irradiation with a high pressure Xe-lamp giving colorless polyethylene.

Full text of DOI:10.1002/pola.29196

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Laser Welding of Polymers using Unsymmetrical Squaraine Dyes
Helle Østergren Bak, Bjarne Enrico Nielsen, Anne Jeppesen, Theis Brock-Nannestad,
Christian Benedikt Orea Nielsen, Michael Pittelkow
Department of Chemistry, University of Copenhagen, DK-2100, Copenhagen Ø, Denmark
Correspondence to: M. Pittelkow (E-mail: pittel@chem.ku.dk)
Received 17 April 2018; Accepted 15 July 2018; published online 28 September 2018
DOI: 10.1002/pola.29196
ABSTRACT: Transmission laser welding of plastic is typically achieved
by incorporating a laser absorbing dye into one of the two pieces
of plastic one wishes to weld together and mediating the welding
process using a laser emitting light at a suitable wavelength.
Desirable properties of laser absorbing dyes include: (1) the dye is
colorless in the visible region but absorbs strongly in the region at
the wavelength where the laser emits, (2) the dye has a good con-
version of the energy from the laser to heat, (3) the dye is sta-
ble enough to be incorporated into the molten plastic, (4) the
dye is soluble in the plastic of choice to give an even distribu-
tion in the plastic, and (5) the dye is nontoxic. Here, we explore
an alternative approach to achieve laser welding of colorless
plastics using a laser dye that is colored, but that can be
bleached after the welding process. Nonsymmetrical squaraine
dyes were prepared by condensation of squaric acid and two
equivalents of various 3,5-dimethoxy-N,N-dialkylanilines. After
the incorporation into polyethylene and laser welding, the
plastic was bleached either by heat treatment or by irradiation
with a high pressure Xe-lamp giving colorless polyethylene.
© 2018 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym.
Chem. 2018, 56, 2245–2254
KEYWORDS: laser welding of plastic; photophysics; polyethylene;
squaraine dyes; squaric acid
shift is the difference between the energy of the absorbed and
emitted photon, which is converted into vibrational energy
thus heat.24 If the fluorescence quantum yield is smaller than
one, then additional photophysical processes can take place
such as intersystem crossing. Decay from the excited state
back to the ground state through release of vibrational energy
is also possible and for the efficiency of a laser absorber for
welding more desirable. To a first approximation, it is desir-
able for the laser absorber to have large Stoke’s shift and a
small fluorescence and triplet quantum yield.
INTRODUCTION The interdiffusion of polymer layers across
their boundaries due to the influence of a heat source is of
immense importance for the processing of polymers and is
more commonly referred to as welding of polymers.1 Insight
into the chemical and physical processes that are occurring
when the melts interact and mix into the interdiffusion zone
or “welding zone” has been the subject of many reports and
has been investigated with a wide variety of techniques such
as neutron scattering and microscopy techniques.2–22 Energy
has to be transferred from a source to the polymer layers in
order to cause the change of phase from solid to liquid for the
interdiffusion or welding process to take place. The heat
sources could be something as simple as heated metal that
transfers heat to the polymer upon contact or more sophisti-
cated techniques such as ultrasound, where an ultrasound
horn is placed in direct mechanical contact with the polymer
part.23 Heat is then transferred upon bringing the horn to
vibrate with a frequency in the ultrasound region and heat is
then generated due to friction. A novel technique is to induce
heat by conversion of light8,11,17: One of the polymer parts
contains an absorbent with absorption properties that are
optimized for the wavelength of the laser used in the process.
By traversing the laser in the pattern for the desired weld,
heat is generated as a result of a photophysical process thus
melting the polymers. Conversion of the incident laser light
into heat is in part explained by the Stokes shift: The Stokes
The coupling between the absorption and fluorescence spectra
and the phonons involved can be identified with a Franck–
Condon/Huang–Rhys analysis24,25 in order to determine some
of the phonons that are active in the generation of heat. There
are several references, mainly in commercial literature, detail-
ing which semicrystalline polymers are suitable for welding
with a laser absorbing dye.26–34 As an example, polyethylene
is normally not considered suitable to be welded with this
technique, whereas it can readily be welded with other tech-
niques such as ultrasonic welding, which leads to the specula-
tion that the transfer of energy from the absorbing dye to
polyethylene is not very efficient. Another reason could be the
dispersibility or solubility of the dye in the polymer. In the
case of poor heat transfer, this reflects back to the active pho-
nons available for generating heat both in the laser absorber
© 2018 Wiley Periodicals, Inc.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254
2245

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
dithranol as the matrix. Column chromatography was per-
formed on Merck Kieselgel 60 (0.015–0.040 mm) by using the
DCVC technique or on Merck Kieselgel 60 (0.040–0.063 mm)
for flash chromatography. Ultraviolet/visible (UV/Vis) spectra
were recorded on a CARY 5E UV–VIS–near-infrared region
(NIR) spectrophotometer. Fourier transform infrared spectra
were recorded on a Bio-Rad Excalibur series instrument. The
electrochemical experiments were recorded using a CH Instru-
ments electrochemical workstation. Fluorescence spectra were
recorded on a Perkin Elmer LS50B fluorimeter with a Hama-
matsu 928 PMT. GC–MS data were collected on a Agilent
6890N/5793 system. Temperature-dependent UV/Vis was
recorded in a custom made apparatus inserted into a CARY
5E UV/Vis/NIR spectrophotometer.35
FIGURE 1 Schematic illustration of the laser welding process.
The red arrow illustrates the laser that is used to weld together
the two items of polymer parts. The lower polymer part contains
a
laser absorber that is responsible for the photophysical
process, where light is converted into heat. [Color figure can be
viewed at wileyonlinelibrary.com]
Synthesis
N,N-dimethyl-3,5-dimethoxyaniline (1a)
Under a nitrogen atmosphere, 1-chloro-3,5-dimethoxybenzene
(27.7 g, 0.16 mol) was dissolved in dry THF (170 ml) at 0 C.
itself and in the polymers—due to the “chemical simplicity” of
polyethylene there are relatively few vibrational modes when
comparing to, for example, polycarbonate, which is considered
very compatible with, for example, perylene bisimides as the
laser absorbing dyes. Perylene bisimides have limited solubil-
ity in almost every solvent unless long alkyl substituents are
placed on the dye which makes the argument of heat transfer
the more likely for determining the compatibility of this class
of dyes in a polymer for welding purposes.
ꢀ
A solution of dimethyl amine (100 ml 2 M, 0.2 mol) was
added. Butyl lithium (2.5 M, 80 ml, 0.2 mol) was added gradu-
ally. The reaction was stirred over night at room temperature.
Water (150 ml) was added and THF was evaporated in vacuo.
The water was extracted with CH2Cl2 (4 x 150 ml). The
organic phases were combined and the solvent was removed
in vacuo. The isolated crude product was recrystallized from
ethanol. Yield: 16.44 g, (90 mmol) 57% as pale yellow crys-
tals; mp 100–101 ꢀC; 1H-NMR (500 MHz, CDCl3) δ 5.84 (s,
3H), 3.71 (s, 6H), 2.85 (s, 6H); 13C-NMR (500 MHz, CDCl3) δ
161.57, 152.47, 91.85, 88.74, 55.20, 40.63. HRMS (MALDI-
A recent report20 has detailed the importance of light scatter-
ing in laser welding, where having light scattering particles
such as TiO2 present can decrease the amount of energy
needed for achieving a sufficient melting and subsequent
interdiffusion. Having this finding in mind, it becomes impor-
tant to choose the right substitution pattern on a novel dye
for laser welding applications in order to make it disperse suf-
ficiently but also make sure that the dye is not dissolved in
the polymer.
+
TOF) m/z 182.1194 [M + H] + (Calcd for C10H16NO2 m/z
182.1176); Anal. Calcd for C10H15NO2: C, 66.27, H 8.34, N
7.73; Found: C, 66.33; H, 8.45; N, 7.56.
N,N-diisobutyl-3,5-dimethoxyaniline (1b)
Diisobutylamine (4.1 g, 31 mmol) and 1-chloro-3,5 dimethoxy-
benzene (5.2 g 30.3 mmol) were dissolved in dry THF
(150 ml) at 0 ꢀC under a dry nitrogen atmosphere. n-Butyl
lithium (14 ml 2.5 M in ether, 35 mmol) was added dropwise.
After 2 h, all 1-chloro-3,5-dimethoxy-benzene had reacted, as
seen with GC–MS. Water (50 ml) was added to the reaction
mixture and THF was removed in vacuo. The remaining water
phase was extracted with CH2Cl2 (4 × 50 ml). The solvent was
removed in vacuo from the combined organic phases leaving a
crude product which was recrystallized from ethanol. Yield
5.46 g (20 mmol) 68% as yellow crystals; mp 62–65 ꢀC; 1H-
NMR (500 MHz, CDCl3) δ 5.76 (m, 3H), 3.70 (s, 6H), 3.04 (d,
4H, J = 7.2 Hz), 2.10–1.96 (m, 2H), 0.81 (d, 12H, J = 6.7 Hz);
13C-NMR (126 MHz, CDCl3) δ 161.51, 150.03, 91.93, 86.96,
60.52, 55.09, 26.54, 20.43; HRMS (MALDI-TOF) m/z 266.2114
Here, we report the synthesis of a type of laser-absorbing dye
molecule based on a squaric acid motif that is suitable for
joining two polymer parts. Characterization of this dye is car-
ried out with the purpose of understanding the interplay of
physical–chemical properties with the polymer in which the
dye is dispersed.
EXPERIMENTAL
General Data
Nuclear magnetic resonance (NMR) spectra were measured
with a 500 MHz CryoProbe instrument from Bruker. All chem-
1
ical shift values in the H and 13C NMR spectra are referenced
to the solvent (δH = 7.26 ppm; δC = 77.00 ppm). Thin-layer
chromatography was carried out on commercially available
precoated plates (silica 60) with fluorescence indicator. Melt-
ing points were measured with a Büchi B-140 melting point
apparatus and are uncorrected. Elemental analyses were
performed with a Thermo Scientific Flash 2000 Organic Ele-
mental Analyzer. matrix-assisted laser desorption ionization
time-of-flight (MALDI-TOF) mass spectra were recorded with
a Bruker Autoflex spectrometer in the positive ion mode using
+
[M + H]+ (Calcd for C16H28NO2 m/z 266.2115); Anal. Calcd
for C16H28NO2: C, 72.41; H, 10.25; N, 5.28; Found: C, 72.24; H,
10.48; N, 5.13.
N,N-bis-decyl-3,5-dimethoxyaniline (1c)
Bis-decyl amine (25.0
g
84 mmol) and 1-chloro-3,-
5-dimethoxybenzene (13.2 g 76 mmol) were dissolved in dry
THF (100 ml) at 0 ꢀC under a dry nitrogen atmosphere. n-
2246
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Butyl lithium (52 ml 1.6 M, 83 mmol) was added dropwise.
After 1 h, the reaction mixture still contained 5-chloro-1,-
3-dimethoxybenzene (as determined by GC–MS) and addi-
tional n-butyl lithium was added (15 ml 1.6 M, 24 mmol).
Water (70 ml) was added to the mixture and THF was
removed in vacuo. The aqueous phase was extracted with
CH2Cl2 (4 × 80 ml) and the organic phases were combined.
Celite was added and the solvents were removed in vacuo.
The crude product was purified by dry column vacuum chro-
matography (column Ø = 10 cm, fractions 100 ml) with
CH2Cl2 in n-heptane with 5% increments. Yield 22.38 g
(51 mmol) 68% as a yellow oil. 1H-NMR (500 MHz, CDCl3) δ
5.82–5.73 (m, 3H), 3.70 (s, 6H), 3.18–3.09 (m, 4H), 1.56–1.45
(m, 4H), 1.29–1.10 (m, 28H), 0.81 (t, 6H, J = 6.8 Hz); 13C-NMR
(126 MHz, CDCl3) δ 161.69, 150.00, 91.11, 87.12, 55.09, 51.22,
31.91, 29.68, 29.59, 29.55, 29.33, 27.35, 27.20, 22.69, 14.13;
HRMS (MALDI-TOF) m/z 434.3995 [M + H]+ (Calcd for
100.35, 94.16, 88.40, 87.68, 55.97, 55.57, 40.56, 40.39; HRMS
+
(MALDI-TOF) m/z 427.1862 [M + H]+ (Calcd for C23H27N2O6
:
m/z 427.1864). Anal. Calcd for C23H26N2O6: C, 64.78; H,
6.15; N, 6.57; Found: C, 64.14; H, 6.10; N, 6.30.
Isobutyl Trimethoxy Hydroxy Squaraine Dye (2b)
N,N-diisobutyl-3,5-dimethoxy aniline (35 g, 132 mmol) and
squaric acid (8.6 g, 75 mmol) were dissolved in n-butanol
(100 ml) and toluene (300 ml). The reaction mixture was
heated to with removal of water by azeotropic distillation
overnight. Celite was added to the reaction mixture and the
solvents were removed in vacuo. The resulting crude product
was purified by dry column chromatography on silica (column
Ø = 12 cm, fractions 200 ml). First with heptane:ethyl acetate
with 10% increments, followed by ethyl acetate:methanol 1%
increments. The product was recrystallized from ethanol.
Yield: 1.49 g (2.5 mmol) 4% as blue/green crystals; mp
1
C
28H52NO2+ m/z 434.3993).
(decomp): 160 C; H-NMR (500 MHz, CDCl3) δ 15.78 (s, 1H),
5.79 (s, 2H), 5.77 (d, 1H, J = 2.3 Hz), 5.59 (d, 1H, J = 2.3 Hz),
3.91 (s, 3H), 3.88 (s, 6H), 3.27 (d, 4H, J = 7.4 Hz), 3.23 (d, 4H,
J = 7.2 Hz), 2.20–2.07 (m, 4H), 0.93 (t, 24H, J = 6.7 Hz); 13C-
NMR (126 MHz, CDCl3) δ 188.84, 185.12, 177.01, 173.98,
168.72, 164.36, 161.71, 158.99, 153.71, 106.81, 100.08, 94.75,
89.15, 88.47, 60.40, 55.81, 55.43, 27.79, 27.18, 20.43, 20.29;
HRMS (MALDI-TOF) m/z 595.3736 [M + H]+ (Calcd for
ꢀ
N,N-bis(2-ethylhexyl)-3,5-dimethoxyaniline (1d)
Bis(2-ethylhexyl)amine (1.13 g, 6.1 mmol) and 1-chloro-3,-
5-dimethoxybenzene (0.73 g, 4.2 mmol) were dissolved in dry
THF (40 ml) under a dry nitrogen atmosphere and stirred at
0
ꢀC. n-Butyl lithium (1.8 ml 2.5 M, 4.5 mmol) was added
gradually. After 30 min all 5-chloro-1,3- dimethoxybenzene
had reacted, as seen by GC–MS. Water (75 ml) was added to
the mixture and the THF was removed in vacuo. The remain-
ing water phase was extracted with CH2Cl2 (4 × 50 ml) and
the organic phases were combined. Celite was then added and
the solvents removed in vacuo. The crude product was subse-
quently purified by dry column vacuum chromatography (col-
umn Ø = 5 cm, fractions 50 ml), CH2Cl2 in n-heptane with 3%
increments. Yield 1.14 g (3.0 mmol) 72% as a yellow oil. 1H-
NMR (500 MHz, CDCl3) δ 5.78 (d, 2H, J = 2.1 Hz), 5.75 (t, 1H,
J = 2.1 Hz), 3.69 (s, 6H), 3.20–3.02 (m, 4H), 1.74 (h, 2H,
J = 6.1 Hz), 1.34–1.10 (m, 16H), 0.88–0.76 (m, 12H); 13C-NMR
(126 MHz, CDCl3) δ 161.43, 150.12, 91.85, 87.47, 56.58, 55.06,
36.96, 30.75, 28.78, 23.98, 23.25, 14.10, 10.75; HRMS (MALDI-
C
35H51N2O6+ m/z 595.3742).
Decyl Trimethoxy Hydroxy Squaraine Dye (2c)
N,N-bis-decyl-3,5-dimethoxy aniline (19.3 g, 44 mmol) and
squaric acid (2.3 g, 20 mmol) were dissolved in n-butanol
(100 ml) and toluene (300 ml). The reaction mixture was
heated to reflux with removal of water by azeotropic distilla-
tion over night. Celite was added to the reaction mixture and
the solvents were removed in vacuo. The resulting crude
product was purified with dry column chromatography (col-
umn ø = 10 cm, fractions 150 ml), with heptane:ethyl acetate
with 10% increments, on two consecutive columns with the
same gradient increments. A blue semisolid was isolated.
Yield: 900 mg (4%); 1H-NMR (500 MHz, CDCl3) δ 15.85 (s,
1H), 5.79 (s, 2H), 5.77 (d, 1H, J = 2.2 Hz), 5.58 (d, 1H,
J = 2.2 Hz), 3.93 (s, 3H), 3.91 (s, 6H), 3.46–3.31 (m, 8H),
1.75–1.60 (m, 8H), 1.43–1.21 (m, 56H), 0.98–0.83 (m, 12H);
13C-NMR (126 MHz, CDCl3) δ 188.61, 185.11, 176.98, 173.62,
168.86, 164.62, 161.89, 158.61, 153.41, 106.83, 100.01, 93.98,
88.36, 87.75, 55.88, 55.47, 51.71, 51.42, 31.90, 31.88, 29.60,
29.56, 29.52, 29.45, 29.37, 29.30, 29.29, 27.97, 27.53, 27.12,
27.02, 22.68, 14.12; HRMS (MALDI-TOF) m/z 931.7467
[M + H]+ (Calcd for C35H51N2O6+ m/z 931.7498).
TOF) m/z 378.3367 [M + H]+ (Calcd for C24H44NO2 m/z
+
378.3367).
Methyl Trimethoxy Hydroxy Squaraine Dye (2a)
N,N-Dimethyl-3,5-dimethoxy aniline (15.0 g, 83 mmol) and
squaric acid (4.3 g, 37 mmol) were dissolved in n-butanol
(100 ml) and toluene (300 ml). The reaction mixture was
heated to reflux with removal of water by azeotropic distilla-
tion over night. Celite was added to the reaction mixture and
the solvents were removed in vacuo. The resulting crude
product was purified by dry column chromatography on silica
(column Ø = 12 cm, fractions 200 ml). First with heptane:ethyl
acetate with 10% increments, followed by ethyl acetate:meth-
anol with 1% increments. The product was recrystallized from
96% ethanol. Yield: 100 mg (0.23 mmol) 6% as blue crystals;
mp (decomp) 257 ꢀC; 1H-NMR (500 MHz, CDCl3) δ 15.88 (s,
1H), 5.79 (s, 2H), 5.77 (d, 1H, J = 2.3 Hz), 5.57 (d, 1H,
J = 2.3 Hz), 3.93 (s, 3H), 3.91 (s, 6H), 3.17 (s, 6H), 3.09 (s,
6H); 13C-NMR (126 MHz, CDCl3) δ 189.83, 185.13, 176.85,
174.10, 168.96, 164.71, 161.74, 160.32, 155.35, 107.19,
2-Ethyl Hexyl Trimethoxy Hydroxy Squaraine Dye (2d)
N,N-Bis(2-ethylhexyl)-3,5-dimethoxy
aniline
(45.1
g,
129 mmol) and squaric acid (6.3 g, 55 mmol) were dissolved
in n-butanol (100 ml) and toluene (300 ml). The reaction mix-
ture was heated to reflux with removal of water by azeotropic
distillation over night. Celite was added to the reaction mix-
ture and the solvents were removed in vacuo. The resulting
crude product was purified with dry column chromatography
(column Ø = 12 cm, fractions 200 ml), with heptane:ethyl
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254
2247

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
acetate with 10% increments. The isolated product was fur-
ther purified with another chromatography step with heptane:
ethyl acetate in 5% increments. A blue semisolid was isolated.
Yield 1.45 g (3.4%); 1H-NMR (500 MHz, CDCl3) δ 15.74 (s,
1H), 5.80 (s, 2H), 5.78 (d, 1H, J = 3 Hz), 5.61 (t, 1H, J = 3 Hz),
3.91 (s, 3H), 3.87 (s, 6H), 3.41–3.20 (m, 8H), 1.89–1.75 (m,
4H), 1.42–1.13 (m, 32H), 0.96–0.78 (m, 24H); 13C-NMR
(126 MHz, CDCl3) δ 188.65, 185.22, 177.13, 174.00, 168.61,
164.28, 161.63, 159.00, 153.41, 106.81, 99.98, 94.93, 89.31,
88.70, 56.93, 56.84, 55.82, 55.47, 38.31, 37.70, 30.97, 30.93,
30.63, 28.98, 28.65, 23.13, 23.07, 14.04, 10.95, 10.92, 10.74;
HRMS (MALDI-TOF) m/z 819.6236 [M + H]+ (Calcd for
The condensation reactions were performed in a mixture of tol-
uene and n-butanol with removal of water using a Dean–Stark
setup. Although the yields are not high, the procedure is repro-
ducible and consistently affords the trimethoxy-mono-hydroxy-
4-(N,N-dialkylamino)-phenyl]-squarines (2a-d). Methods to
prepare nonsymmetrical squarine absorbers are scarce, and
normally require several synthetic steps. Synthetic protocols
for unsymmetrically substituted squarine dyes (two different
anilines) have been described via a condensation of the acid
chloride of squaric acid and an N,N-dialkylaniline followed by
hydrolysis and another condensation reaction with a different
N,N-dialkylaniline. This is the first nonsymmetrical squarine
absorber to be prepared in a one-step procedure.
C
47H74N2O6+ m/z 819.6246).
UV–Vis and Fluorescence Spectroscopy
Solvent Effects
Voltammmetry
In the square wave voltammetric experiments, a three-
electrode setup was employed with a freshly polished glassy
carbon electrode (o.d. = 1 mm) serving as the working elec-
trode, a platinum wire counter electrode, and a Ag/AgCl refer-
ence electrode. All experiments were performed at a 1 mM
concentration of dye in a 100 mM solution of tetrabutylammo-
nium hexafluorophosphate in CH2Cl2. All potentials are refer-
enced to the ferrocenium/ferrocene (Fc+/Fc) redox couple, by
measurement of said redox couple immediately after collec-
tion of data. All experiments utilized the instrumental option
of correcting for the ohmic drop of the system as incorporated
in the CHI electrochemical workstation.
The coupling between absorption of light and the distribution
of heat into the embedding polyethylene is suggested above to
occur through vibrational redistribution of energy. As such, it
is important to look in more details into the absorption and
fluorescence properties of the squaric absorbers. Squarine
dyes of the type shown in Scheme 1 have absorption bands in
the region 623–670 nm depending on the substitution pattern
on the squarine motif. The exact position of the band is
strongly solvent dependent and varies from 613 nm in hexane
to 640 nm in NMP for 2b (Fig. 2).
Chromophores with significant intramolecular charge transfer
(CT) have UV–VIS spectra that strongly depend on the polarity of
the solvent which is ascribed to how well the solvent stabilizes
the ground and exited states. Squaraine dyes show a negative sol-
vatochromism (Fig. 3, top), where a redshift of the absorption
maximum is observed with increasing polarity of the solvent:
The ET(30) values are used as the measure of solvent polarity
and for hexane the ET(30) value 31 kcal mol−1, 40.7 kcal mol−1
for dichloromethane and 42.2 kcal mol−1 for NMP.
Computational Details
Geometry optimizations for determining the molecular struc-
tures in the present work were performed at the
B3LYP/6-31G(d) level using the Gaussian 09 suite of pro-
grams.37 Vibrational frequency calculations were carried out
on the optimized structures in order to verify that the struc-
tures represent minima on the potential energy surface. These
geometry optimizations and frequency calculations were per-
formed for the molecules in isolation.
The solvent effects of the absorption properties of the squara-
nine absorbers shows that the extinction coefficients are also
RESULTS AND DISCUSSION
Synthesis
Squarine absorbers have been reported in other contexts,
where electronic properties have been achieved by carefully
selecting the substituents on the squaric acid motif.38–41 The
use of 3,5-dimethoxy-N,N-dialkylanilines (1) has been chosen
in the present work as a convenient way of introducing vari-
ous alkyl-substituents and the methodology of condensing
3,5-dimethoxy-N,N-dialkylaniline with squaric acid has not
previously been described. The ability to introduce various
alkyl substituents in a convenient way comes in the present
work at the price of not having absorption properties in the
NIR but rather in the visible region. The aniline is isolated in a
distillation process.
A
convenient method to prepare
3,5-dimethoxy-N,N-dialkylanilines in high yields via a reaction
between 3,5-dimethoxy-chlorobenzene and a secondary amine
has been developed (see Scheme 1).
SCHEME
1
Synthesis of trimethoxy-mono-hydroxy-4-(N,N-
dialkylamino-phenyl)-squarines. (a) THF, n-BuLi, HNR2, 80–90%,
(b) toluene, n-BuOH.
2248
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
solvents (alcohols) fall on the same line as the rest of the sol-
vents, thereby strongly suggesting that despite the presence
of OH-functionalities on the squaric acid motif there are no
hydrogen bonding to an H-donor.
15
Hexane
CH2Cl2
NMP
10
5
Electronic Structure
To gain further insights to the electronic structure of the
squaraine dyes, calculations were carried out at the
B3LYP/6-31G(d) level of theory. A geometry optimization was
performed out and the geometry was confirmed by calculating
the frequencies.
0
500
600
700
Wavelength/nm
The highest occupied molecular orbital (HOMO) and lowest
unoccupied molecular orbital (LUMO) plots support the the-
ory that there is CT occurring during an electronic transition
(assuming that the HOMO–LUMO transition dominates the
electronic transition) inasmuch as there seems to be a change
in charge located at the central oxygen atoms when compar-
ing the HOMO and LUMO orbitals (Fig. 5).
FIGURE 2 Absorption spectra of 2b in hexane, dichloromethane
and N-methyl-pyrolidene. [Color figure can be viewed at
wileyonlinelibrary.com]
dependent of the polarity of the solvent—and to a quite signif-
icant extent. Figure 3 (bottom) shows the extinction coeffi-
cients as a function of the ET(30) for a series of solvents. A
maximum seems to be observed at a value of ~40 kcal mol−1
which corresponds to an apolar solvent (hexane) and thus
mimics rather well the polarity of a polyolefin such as poly-
ethylene. Previously, it has been reported that for CT dyes, the
strongest absorbance occurs in solvents with polarity corre-
sponding to an equal distribution between a polar and apolar
resonance form.42 Suitable resonance forms that displays
polar and apolar charge distributions can be drawn for squar-
aine absorbers (Fig. 4).
Electrochemistry
Redox potentials of the squaraine dye have been obtained in
0.1 mM CH2Cl2 with 0.1 M Bu4NPF6 as the carrier electrolyte, a
freshly mirror-polished platinum disc electrode, platinum wire
as the counter electrode. A Ag/AgCl electrode was used as a ref-
erence electrode. Ferrocene was used as a reference to deter-
mine the oxidation potential. Similar to previously reports on
electrochemical data for squaraine dyes, where two oxidation
potentials obtained. The first oxidation is partially reversible
while the second oxidation is reversible (Fig. 6).38
The first two oxidation potentials for 2b were determined to
be 0.16 and 0.68 V.
A linear dependence between the Stokes shift and the
ET(30) values of a series of solvent is expected.24 Chemical
interactions can be inferred from the ET(30) plots: If for
example, there was a linear dependence of the Stokes shift on
the ET(30) values for alcohols with a different slope than for
similar plots including a wide variety of solvents with differ-
ent chemical functionalities, there is most likely hydrogen
bonding between the alcohols and the solute. The dependence
of the Stokes shifts on the ET(30) values for the range of sol-
vents displays a single linear dependence and thus there are
no apparent outliers or groups of solvents that shows a sepa-
rate trend than the rest. In particular, hydrogen-bonding
Generation of Heat after Excitation
The release of heat after excitation to a higher excited elec-
tronic state is described as nonradiative processes, where
vibrational modes in the molecule and surroundings are acti-
vated. A typical Jablonski diagram showing the different path-
ways after excitation is shown in Figure 7.
Excitation can occur to, for example, a S2 state but emission
almost always occurs from the S1 state as rapid relaxation
1400
1200
1000
800
120000
110000
100000
90000
600
80000
400
30
35
40
45
50
55
30 35 40 45 50 55 60 65 70
ET(30)/kcal·mol-1
ET(30)
FIGURE 3 Top: Stokes shift as a function of ET(30) values for 2b. Bottom: Solvent polarity as a function of ET(30) values for 2b
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254
2249

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 4 More polar and less polar resonance forms of the
charge-transfer squaraine dye.
occurs. Indeed, Kasha’s rule states that most photochemical
processes occur from either the S1 or the T1 state.24 Thus,
excitation from S0 to a higher singlet state usually results in
nonradiative decay to the S1 state—in the nonradiative pro-
cess heat is generated as the vibrational levels are activated.
Intersystem crossing to the triplet manifold can also occur
and from here other radiative and nonradiative processes can
also occur. Usually, emission of a photon from the S1 results
in fluorescence.
FIGURE 5 LUMO (top) and HOMO (bottom) of 2a calculated at
the B3LYP/6-31G(d) level of theory. [Color figure can be viewed
at wileyonlinelibrary.com]
e−SSn
FC0!f
=
n!
where the initial state is set as the ground state.
The vibrational mode that is active in the absorption of light
by a squarine dye can be identified by employing a Franck–
Condon analysis of the absorption spectrum using the expres-
sion. Including a more general treatment of the Franck–
Condon analysis than that described above results in the fol-
lowing expression for analyzing electronic transitions,
A recent study has quantified the triplet quantum yield of a
series of squaraine dyes.43 Low quantum yields were
observed in both air and N2 saturated toluene indicating that
intersystem crossing from the singlet to the triplet manifold is
not a significant pathway. Another interesting finding from
this study is also that a linear dependence between the fluo-
rescence lifetime and the fluorescence quantum yield was
observed for the different compounds, indicating that the radi-
ative rate of decay is independent of the substitution pattern
of the squaraine compounds studied. The fluorescence quan-
tum yields of the series of compounds studies were in many
cases very low (<0.1) but in some cases significant quantum
yields were observed.43 In the present study, there was no
detected fluorescence from the squaraine embedded in the
polymer matrix indicating that radiative relaxation from S1 to
S0 does not occur. As it is also not likely that the triplet mani-
fold is populated based on the findings,43 it appears most
likely that nonradiative decay from S1 to S0 takes place.
!
ꢁ
ꢂ
P
2
ni
XY
ω−ω0 ꢁ i niωp,i
Si expð−SiÞ
IðωÞ /
exp
−
,
ni!
2σ2
ni
i
where ω is the frequency, ω0 is electronic origin frequency, ωp
is the phonon frequency, Si is the Huang–Rhys factor, i is a
normal mode in the final state with frequency ωp, σ is the
width of the broadened line at half height for the electronic
transition at ω0, and ni is the progression of the ith normal
mode in the Poisson distribution. The “+” sign is used for ana-
lyzing absorption spectra and the “–” sign for emission spec-
tra. The Franck–Condon analysis is carried out for compound
6b but only for i = 1. The resulting fitted curves along with
experimental data are shown in Figure 8.
The rate at which the nonradiative decay from S1 to S0 occurs
can be described with time-dependent perturbation theory
There is a very good agreement with the experimental data
and the fitted Franck–Condon models. The parameters used in
the data modeling are presented in Table 1.
2π
2
^
knr
=
jhψijvjψfij ρðEÞ,
ℏ
where knr is the rate constant for the nonradiative decay, ψi and
ψf are the wave functions for the initial (S0-state) and final state
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
^
(S1-state), respectively, v is the kinetic energy operator and
ρ(E) is the density of states, at the energy E. By invoking the
Born–Oppenheimer approximation, the wave functions can be
factored into electronic (ψie and ψfe) and nuclear (ψin and ψfn)
wave functions, which results in the following:
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
2
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
D
E
D
E
D
E
2
2
2π
2π
ꢀ ꢀ
e
e
n
n
e
e
^
^
^
knr
=
ψi jvjψf
ψi jvjψf
ρðEÞ =
ψi jvjψf
ꢀ ꢀ
-1.0
ℏ
ℏ
FCi!f ρðEÞ
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Potential/V
where FC is the Franck–Condon factor between the two states.
The Franck–Condon factors can be found the absorption spec-
trum and is further related to the Huang–Rhys factor (S)
FIGURE 6 Cyclic voltammogram of 2b (R = isobutyl) recorded at
0.1 mM CH2Cl2 using PF6Bu4N as the carrier electrolyte.
2250
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Thermal Stability
When incorporating the laser absorbing dye into plastic for
use in the laser welding experiments, the plastic is typically
heated to a temperature where it is soft or melted. Then, the
dye is added and the plastic is mixed and the soft plastic injec-
tion molded and cooled to give the laser absorbing plastic
product. It is important that the dye does not decompose to
any significant degree during this treatment.
The simplest method to quantify the stability of a dye is by
exposing it to heat at one temperature, and then follow the
degradation process by UV–Vis spectroscopy. To this end, the
dye was heated in a thermostated cell in a UV–Vis spectropho-
tometer in a dodecane solution and the disappearance of the
main absorbance was monitored. In a parallel experiment, this
was done in a degassed dodecane solution. Dodecane was cho-
sen to mimic polyethylene (Fig. 9). It is apparent that oxygen
does not play a major role in the decomposition.
FIGURE 7 Jablonski diagram showing excitation from the S0
state to a higher vibrational state of the electronic S2 state.
Subsequent internal conversion leads to the population of the
S1 state after activation of vibrational modes (“curved arrow”).
From the S1-state, several pathways can result in either
population of vibrational modes or radiative decay.
The decomposition rate is strongly solvent dependent. In an
aromatic solvent like chlorobenzene no decomposition is
observed after 10 h at 90 ꢀC. At reflux in chlorobenzene
ꢀ
The analysis reveals that there is
a
vibration at
(132 C), the dye is less than 90% decomposed after 5 h, but
~1442–1594 cm−1, depending on the solvent, that is coupled
in dodecane, the same dye is more than 90% decomposed
ꢀ
to the absorption of a photon.
after 3 h at 115 C, which could reflect different solubilities of
oxygen in the various solvents.
A similar analysis was carried out for the commercial laser
dyes Lumogen 788 and 765 and the results are presented in
Table 1.
The stability to light exposure was examined in a chloroben-
zene solution and after incorporated into polyethylene. The
stability to light is important to estimate stability of the dye
against both laser light and to gauge the stability at storage.
The solution is exposed to light from a Xe-lamp in a cuvette
For molecules where the fluorescence and triplet quantum
yields are negligible, it is apparent that the nonradiative decay
is dependent on the transition elements between the two elec-
tronic states and the Franck–Condon factor. When, for exam-
ple, compound 6b is embedded in a polymer matrix there is
no fluorescence observed. Lumogen 765 and Lumogen
788 are commercially available pigments specifically for laser
welding and they are chemically most likely rylenes, which
means they have very high extinction coefficients
(>100.000 M−1ꢂcm−1) and broad spectral features. When
comparing compound 6b with these two commercial dyes it is
evident that the Lumogens should perform better than, for
example, compound 2b as they have higher extinction coeffi-
cients and higher Franck–Condon factors.
Experimental data
Franck-Condon model
0,3
0,2
NMP
0,1
0,0
0,3
0,2
Hexane
0,1
0,0
TABLE 1 Parameters from the Franck–Condon analysis
hexane NMP
CH2Cl2
0,3
2a
2a
2a
Lumogen
765
Lumogen
788
0,2
0,1
0,0
CH2Cl2
S
0.21
0.066
2.02
0.20
0.24
0.18
0.49
0.45
0.054
1.59
0.17
σ
0.068 0.066 0.052
1,6
1,8
2,0
2,2
2,4
ω0/eV
1.94
0.18
1442
1.96
0.18
1490
1.62
0.16
Energy/eV
ωp/eV
FIGURE 8 Franck–Condon model fitted to experimental data for
different solvents (2b).
ωp/cm−1 1594
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254
2251

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Degassed
Air saturated
1
0.1
0
20 40 60 80 100 120 140 160
Exposure time/s
FIGURE 9 Decomposition of dye 2b in 90 ꢀC dodecane, under
ambient atmosphere and with a nitrogen atmosphere.
with stirring, the concentration is determined via UV–Vis
absorbance measurements. Under oxygen-free conditions in a
degassed solution the degradation rate is about 25% lower
than in the presence of O2, which also lend support to differ-
ences in stability when the dye is solubilized in either chloro-
benzene or dodecane when exposed to light.
FIGURE 11 Transmission spectrum of PE with 2b. [Color figure
can be viewed at wileyonlinelibrary.com]
electro spray of squaraine dyes and they observed alkyl trans-
fer from nitrogen to oxygen in the squaric acid ring. It is not
clear exactly what the decomposition products are, but the
majority of products appear to result from a decomposition of
the squaric acid ring.
Another useful technique to characterize thermal stability is to
examine the neat material in a thin film. We have built an oven
insert for a UV–Vis spectrophotometer that allows us to moni-
tor the absorbance as a function of temperature of thin films.
We monitored the disappearance of the main absorbance band
of the laser absorber as a function of increasing temperatures.
Absorbance spectra are shown in Figure 9. The dye decom-
poses gradually. At 150 ꢀC, 50% of the dye had decomposed.
Laser Welding
Laser absorbing dyes can be incorporated in polymers by dis-
tribution on polymer pellets followed by compounding—a
loading of 0.1 wt% of 2b in a linear low density PE matrix
(Flexirene MS 20 A from Versalis) was used. This was done at
Decomposition Products
To our satisfaction, we observed that the squarine dyes
decomposed to colorless products. This is significant for the
future use of this type of dye in plastic. To investigate the
composition of the decomposition products obtained after
thermal decomposition, we heated a solid sample to 150 ꢀC
for 30 min and analyzed the resulting mixture using LC–MS
analysis. Law et al. have analyzed the mass spectra from
24 o
34 o
53 o
73 o
92 o
111 o
131 o
150 o
170 o
C
C
C
C
C
100
80
60
40
20
0
0.08
0.06
0.04
0.02
0.00
25 50 75 100 125 150 175
C
C
C
C
o
Temperature/
C
400
500
600
700
800
Wavelength/nm
FIGURE 10 Decomposition of dye 2b by exposure to light from a
Xenon lamp. Degradation is about 25% slower in a degassed
solution. Solvent chlorobenzene. The decomposition process is
first order, approaching zero order at high concentrations.
[Color figure can be viewed at wileyonlinelibrary.com]
FIGURE 12 Decomposition of the isobutyl trimethoxy hydroxyl
dye (2b) in PE by exposure to light from a Xenon lamp. The dye
clearly decomposes to yield the PE as a colorless material. The
last picture shows the PE exposed to light for 1 h (bottom part).
[Color figure can be viewed at wileyonlinelibrary.com]
2252
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Laser welding experiments were performed using a home
build setup as shown in Figure 13. The welding process can
be performed using PE with the dye incorporated or with the
dye applied to the surface of the PE by simple drop casting.
The laser welding was performed using 1.7 W optical effect
with a 650 nm diode laser. This is equivalent to a current of
1.4 A or a voltage of 6.25 V. The spot size was estimated to be
2 × 2 mm, which gives an output per area of ~0.4 W mm−2
.
The laser was traversed a rate of 40 mm s−1. To our satisfac-
tion, the strength of the welding zone as evaluated by tensile
strength (each end of the specimen is fastened into clamps
and the force required to move the clamps away from each
other is recorded and noted upon breaking the specimen) and
the strength of the welding zone was comparable to that of
the plastic itself.
CONCLUSIONS
A series of squaraine compounds have been synthesized and
characterized. It is also discussed how to describe potential
new laser welding absorbers with respect to efficiency and it
suggested that such an analysis can be carried out by applying
a
Huang–Rhys/Franck–Condon analysis and deriving the
Huang–Rhys factors. This value together with the extinction
coefficient can be used as a guideline for the characterizing
the efficiency of laser dyes. However, if the potential laser dye
intersystem crosses to a significant extent to the triplet mani-
fold and/or fluorescence, these pathways must also be taken
into account, making the analysis more subtle.
FIGURE 13 Top: laser welding setup. Bottom: picture of laser
welding zone when using the drop casting method to apply the
dye (2b). [Color figure can be viewed at wileyonlinelibrary.com]
ACKNOWLEDGMENT
The authors would like to thank the University of Copenhagen,
the Danish foundations “Højteknologifonden” and “Det strate-
giske forskningsråd” for their financial support.
ꢀ
140 C and the plastic is extruded (twin screw extruder) and
divided into pellets. In the compounding process, the polymer
was melted and admixture with the dye was obtained. After
the compounding process, the polymer can be molded in dif-
ferent shapes. After these processes, polymers with laser
absorbing dyes can be welded to laser transparent polymers.
The successful incorporation was verified by recording simple
transmission spectra (Fig. 11) that showed the absorbance in
the region where the dye absorbs. Visually, it is also apparent
that the dye is incorporated in the PE.
REFERENCES
1 R. Klein, Laser Welding of Plastics, Wiley, Wiley-VCH Verlag
GmbH & Co. KGaA, 2011.
2 B. Acherjee, A. S. Kuar, S. Mitra, D. Misra, Opt. Laser Technol.
2012, 44, 1281.
3 A. C. A. Asseko, B. Cosson, M. Deleglise, F. Schmidt, Y. Le
Maoult, E. Lafranche, Int. J. Mat. Form. 2015, 8, 127.
4
A. C. A. Asseko, B. Cosson, F. Schmidt, Y. Le Maoult,
R. Gilblas, E. Lafranche, Infrared Phys. Technol. 2015, 73, 304.
5 F. Becker, H. Potente, Polym. Eng. Sci. 2002, 42, 365.
The typical available wavelength used for laser welding is
808 nm, but high energy lasers with various wavelength are
now readily available—presently, lasers with wavelength
lower than 700 nm are less commonly used due to costs.
6 A. Boglea, A. Olowinsky, A. Gillner, J. Laser Micro Nanoen.
2010, 5, 138.
7 G. Casalino, E. Ghorbel, J. Mater. Process. Technol. 2008,
207, 63.
8 M. Chen, G. Zak, P. J. Bates, Weld. World 2013, 57, 171.
The polymers can be bleached after incorporation/welding
using a Xenon lamp (450 W), in Figure 12, a polymer contain-
ing a squaraine dye incorporated with gradual decomposition
and bleaching as a function of light exposure. It is significant
that the dye bleaches completely into colorless products leav-
ing the PE colorless.
9 B. Cosson, M. Deleglise, W. Knapp, Compos. Part B–Eng. 2015,
68, 85.
10 M. Ilie, D. Grevey, S. Mattei, E. Cicala, V. Stoica, Opt. Laser
Technol. 2009, 41, 608.
11 M. Ilie, J.-C. Kneip, S. Mattei, T. Girasole, Infrared Phys.
Techn. 2007, 51, 73.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254
2253

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
12 T. B. Juhl, D. Bach, R. G. Larson, J. D. Christiansen,
E. A. Jensen, Polymer 2013, 54, 3891.
32 A. M. Visco, N. Campo, L. Torrisi, A. Borrielli, Radiat. Eff.
Defect. S. 2010, 165, 601.
13 G. N. Labeas, G. A. Moraitis, C. V. Katsiropoulos, J. Compos.
Mater. 2010, 44, 113.
33 A. M. Visco, N. Campo, L. Torrisi, F. Caridi, Appl. Phys. A
2011, 103, 439.
14 H. X. Liu, H. Jiang, D. Guo, G. Chen, Z. Yan, P. Li, H. Zhu,
J. Chen, X. Wang, Materials 2015, 8, 4961.
34 Y. Y. Wang, A. H. Wang, Z. K. Weng, H. B. Xia, Opt. Laser
Technol. 2016, 80, 153.
15 H. X. Liu, W. Liu, X. Zhong, B. Liu, D. Guo, X. Wang, Mater.
Design 2016, 99, 83.
35 T. Brock-Nannestad, C. B. Nielsen, H. Ø. Bak, M. Pittelkow,
Rev. Sci. Instrum. 2013, 84, 046106.
16 H. X. Liu, Z. Yan, P. Li, X. Wang, Sci. Technol. Weld. Joi.
2014, 19, 487.
36 W. R. Ware, W. Rothman, Chem. Phys. Lett. 1976, 39, 449.
37 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr. , J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich,
A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian 09, Gaussian, Inc.,
Wallingford, CT, 2009.
17 C. J. Nonhof, Polym. Eng. Sci. 1994, 34, 1547.
18 G. Overton, Laser Focus World 2014, 50, 19.
19 B. Rooks, Int. J. Intell. Robot Appl. 2004, 31, 338.
20 J. F. von Bulow, K. Bager, C. Thirstrup, Appl. Surf. Sci. 2009,
256, 900.
21 K. Yokomizo, Y. Banno, T. Yoshikawa, M. Kotaki, Poly. Eng.
Sci. 2013, 53, 2336.
22 A. Yousefpour, M. Hojjati, J. P. Immarigeon, J. Thermoplast.
Compos. Mater. 2004, 17, 303.
23 M. J. Troughton, Handbook of Plastics Joining: A Practical
Guide, Elsevier Science, Cambridge, UK, 2008.
24 M. Klessinger, J. Michl, Excited States and Photochemistry of
Organic Molecules, VCH, New York, New York 10010, 1995.
25 E. C. Lim, Advances in Photochemistry, John Wiley & Sons,
Inc., New York, 2007, p. 165.
26 M. Bashir, M. S. Mehmood, M. A. Choudary, T. Yasin,
I. Ahmed, M. Tariq, M. Nomin-ul-Haq, A. Asif, M. Ikram, J. Russ.
Laser Res. 2015, 36, 258.
38 K. Y. Law, J. S. Facci, F. C. Balley, J. F. Yanus, J. Imaging Sci.
1990, 34, 31.
39 A. J. McKerrow, E. Buncel, P. M. Kazmaier, Can. J. Chem.
1995, 73, 1605.
27 G. Galtieri, A. Visco, D. Nocita, L. Torrisi, G. Ceccio,
C. Scolaro, J. Instrum. 2016, 11, CO4013.
40 S. Wang, L. Hall, V. V. Diev, R. Haiges, G. Wei, X. Xiao,
P. I. Djurovich, S. R. Forrest, M. E. Thompson, Chem. Mater.
2011, 23, 4789.
28 T. B. Juhl, J. D. Christiansen, E. A. Jensen, J. Appl. Polym.
Sci. 2013, 129, 2679.
29 L. Torrisi, F. Caridi, A. M. Vasco, N. Campo, Appl. Surf. Sci.
2011, 257, 2567.
41 G. J. Ashwell, P. C. Williamson, A. Green, C. R. Brown, Austr.
J. Chem. 1998, 51, 599.
30 L. Torrisi, A. M. Visco, N. Campo, F. Caridi, Nucl. Instrum.
Methods Phys. Res. B 2010, 268, 3117.
42 A. I. Kiprianov, Russ. Chem. Rev. 1960, 29, 618.
43 P. Salice, J. Arnbjerg, B. W. Pedersen, R. Toftegaard,
L. Beverina, G. A. Pagani, P. R. Ogilby, Chem. A Eur. J. 2010,
114, 2518.
31 A. M. Visco, V. Brancato, L. Torrisi, M. Gutroneo, Int. J. Poly.
Anal. Ch. 2014, 19, 489.
2254
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2018, 56, 2245–2254