MIR-Photorheology: A Versatile ... - ACS Publications

Subscriber access provided by University of Newcastle, Australia

Article

Real time-NIR/MIR-photorheology: A versatile tool for the in situ characterization of photopolymerization reactions Christian Gorsche, Harikrishna Reghunathan, Stefan Baudis, Patrick Knaack, Branislav Husar, Joerg Laeuger, Helmuth Hoffmann, and Robert Liska Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00272 • Publication Date (Web): 06 Apr 2017 Downloaded from http://pubs.acs.org on April 11, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

RT-NIR-Photorheology study of IBoA (▬) and formulations with 10 wt% (-.-) and 50 wt% (---) HDDA; a) storage modulus G’; b) normal force FN; c) double bond conversion DBC. 81x184mm (300 x 300 DPI)

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Detection of shrinkage stress using the photocuring of formulation IBoA/10 wt% HDDA as example (liquid state: grid structure; solid network: check pattern). 81x129mm (300 x 300 DPI)


Page 2 of 38

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RT-IR-Photorheology study of HDDA (▬) and formulations with 20 rg% (-.-) and 40 rg% (---) TMPMP; a) storage modulus G’; b) normal force FN; c) double bond conversion DBC and thiol conversion TC. 81x184mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

RT-NIR-Photorheology study of BADGE at 50 °C (▬) and 70 °C (---); a) storage modulus G’; b) normal force FN; c) epoxy conversion EC. 81x184mm (300 x 300 DPI)


Page 4 of 38

Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RT-NIR-Photorheology study of PEGDA hydrogel precursor formulations with 50 (▬), 40 (-.-) and 30 wt% (--) PEGDA and 0.2 wt% photoinitiator (Irgacure 2959 or LiTPO); a) storage modulus G’ and b) double bond conversion DBC for formulations with Irgacure 2959; c) G’ and d) DBC for formulations with LiTPO. 162x122mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Results of RT-NIR-Photorheology of PEGDA based hydrogel precursor formulations with two different photoinitiators, Irgacure 2959 and LiTPO; a) max. storage modulus G’max, b) double bond conversion DBC, c) and d) the slopes, ∆G’ ∆t 1 and ∆DBC ∆t 1, of the linear regions of the curves. 162x122mm (300 x 300 DPI)


Page 6 of 38

Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Real time-NIR/MIR-photorheology: A versatile tool for the in situ characterization of photopolymerization reactions Christian Gorsche,†,‡ Reghunathan Harikrishna,†,‡ Stefan Baudis,† Patrick Knaack,† Branislav Husar,†,║ Joerg Laeuger,§ Helmuth Hoffmann,† and Robert Liska*,†,‡ †

Institute of Applied Synthetic Chemistry, Technische Universität Wien, Getreidemarkt 9/163

MC, 1060 Vienna, Austria ‡

Christian-Doppler-Laboratory for Photopolymers in Digital and Restorative Dentistry,

Getreidemarkt 9, 1060 Vienna, Austria §

Anton Paar Germany GmbH, Helmuth-Hirth-Strasse 6, D-73760 Ostfildern, Germany

║

Unfortunately deceased on Sept 17th 2014

*corresponding author: [email protected]

Keywords (meth)acrylate, thiol-ene, hydrogel, cationic photopolymerization, epoxy resins

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

Abstract In photopolymerization reactions, mostly multifunctional monomers are employed as they ensure fast reaction times and good final mechanical properties of the cured materials. Drawing conclusions about the influence of the components and curing conditions on the mechanical properties of the subsequently formed insoluble networks is challenging. Therefore, an in situ observation of chemical and mechanical characteristics during the photopolymerization reaction is desired. By coupling of an infrared spectrometer with a photorheometer, a broad spectrum of different photopolymerizable formulations can be analyzed during the curing reaction. The rheological information (i.e. time to gelation, final modulus, shrinkage force) can be derived from a parallel plate rheometer equipped with a UV- and IR-translucent window (glass for NIR and CaF2 window for MIR). Chemical information (i.e. conversion at the gel point and final conversion) is gained by monitoring the decrease of the corresponding IR-peak for the reactive monomer unit (e.g. C=C double bond peak for (meth)acrylates, H-S thiol and C=C double bond peak in thiol-ene systems, C-O epoxy peak for epoxy resins). Depending on the relative concentration of reactive functional groups in the sample volume and the intensity of the IR signal, the conversion can be monitored in the near-infrared region (e.g. acrylate double bonds, epoxy groups) or the MIR region (e.g. thiol signal). Moreover, an integrated Peltier element and external heating hood enable the characterization of photopolymerization reactions at elevated temperatures, which also widens the window of application to resins that are waxy or solid at ambient conditions. By switching from water to heavy water, the chemical conversion during photopolymerization of hydrogel precursor formulations can also be examined. Moreover, this device could also represent an analytical tool for a variety of thermally and redox initiated systems.


Page 9 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Thermosetting resins consist of multifunctional monomers that form crosslinked polymer networks. These crosslinking monomers ensure fast curing due to early gelation and give suitable mechanical properties (e.g. high hardness, stiffness, and heat deflection temperature). Moreover, multifunctional monomers are preferred for toxicological reasons as they are less prone to migrate out of the coating or bulk material, which generally decreases health concern. Due to the crosslinking nature, in situ characterization for the curing reaction of such thermosetting resins is a difficult task.1,2 Combinations of multiple analytical methods such as IR analysis for the characterization of chemical conversion with mechanical (e.g. DMTA, TMA) and thermal characterization (e.g. DSC, TGA) are desired. For the design of such an analytical device the initiation of the curing reaction (e.g. thermal, redox or light-triggered) has also a major influence. While heating of the specimen can often easily be realized, a more advanced set-up is necessary for light-triggered reactions (e.g. photopolymerizations). Homogeneous irradiation of the sample with simultaneous real time analysis of reaction kinetics is challenging. Photopolymerization is a well-established processing technique in the fields of protective and decorative coatings3 for wood, cardboard, paper, plastic, metal, or ceramics, but has recently gained popularity in more advanced areas such as micro-electronics,4,5 optical materials,6 biomedical applications,7,8 and 3D-printing.9 As photopolymerization has progressed in the past decades, researchers are dealing with a number of different analytical techniques when performing kinetic studies on a UV-curable coating.10 Different light sources have been coupled to appropriate analytical devices such as in photo-DSC (Hicks et al. in 1977),11 photodilatometry12,13 or real time (RT)-FTIR-photocuring.14 The desire for more advanced and especially hyphenated analytical methods remains up until today as new insights can be expected from their application. This was demonstrated recently 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

by the coupling of NIR and ultrasonic reflectometry,15 representing another set-up for the in situ monitoring of photopolymerization reactions, and a hyphenated UV-Vis/FT-NIR spectroscopy set-up16 paving the way to simultaneous recording of photoinitiator bleaching and monomer conversion. Moreover, an in situ monitoring of photopolymerization induced phase separation via NIR coupled turbidity measurements, using a UV/VIS spectrometer, has been shown by the research group of Stansbury.17 Ultimately, in-line process controlling for various applications (e.g. coatings) could be achieved.18 Hyphenated RT-IR-photorheology devices serve as the most common combination for an in situ characterization of photopolymerization reactions with crosslinking resins. In 1992 Khan et al. published the first report on a photorheometer as in situ technique for the monitoring of photopolymerization reactions with respect to their rheological curing characteristics (e.g. gelation, increase in moduli).19 In the following, photorheology and RTIR measurements were conducted independently from each other.20 Here, only relative correlations can be drawn as identical irradiation conditions, layer thickness, and curing temperature are difficult to realize. Botella et al.21 have first reported a direct coupling of near infrared (NIR) spectroscopy and photorheology enabling the in situ monitoring of chemical and rheological information. With a similar set-up, the research group of Scherzer22 has recently published a method for the simultaneous monitoring of conversion (via NIR) and polymerization induced shrinkage stress (via rheometry).23 NIR analysis in photorheology is rather simple from the experimental set-up, but requires larger sample volumes and therefore a layer thickness ≥ 200 µm due to the generally weak NIR absorptions. To allow the analysis of different layer thicknesses (~ 50 – 500 µm) and various concentration of functional groups within the sample volume, a new combination of photorheology and near-IR (NIR) as well as mid-IR (MIR) is desired for application oriented experiments. While NIR measurement set-ups are easily realized by using light guiding, the implementation of 4 ACS Paragon Plus Environment

Page 11 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


MIR measurements requires sophisticated optical pathways with mirrors, translucent windows (e.g. NaCl, CaF2), and dry conditions. Consequently, advanced resins such as thiol-ene or epoxy-based formulations could be monitored more thoroughly. With the additional integration of a Peltier-based heating system, photopolymerization at elevated temperatures could be conducted, which would enable the photocuring of waxy (at ambient conditions) resins with higher molecular weight. Within our research group, we have developed this unique real time-NIR/MIRphotorheometer prototype. Some of its potential has already been exploited in recent studies by characterizing photopolymerizations for toughening of photopolymer networks using chain transfer agents24 or thiol-yne resins25 and by studying hydrogel precursor formulations.26,27 Additionally, cluster-reinforced systems,28 vinylcyclopropanes as radical ring opening monomers29 and degradable polyphosphazene photopolymers for tissue regeneration30 have also been under investigation. In this work, we demonstrate the potential of our real time hyphenated IR-photorheology (RT-IR-photorheology)

prototype

instrument

for

in situ

characterization

of

photopolymerizable resins, which are easily accessible to any researcher and relevant for industry. Combined mechanical information about curing (e.g. observation of gelation, polymerization induced shrinkage stress, final modulus) and chemical conversion via IR spectroscopy (e.g. conversion at gel point, final conversion) helps to give a more complete understanding of the investigated photopolymerization reactions. First, the experimental set-up of the RT-IR-photorheometer is described and the limitations and general considerations for the presented set-up are discussed. Then, the photocuring of a monofunctional acrylate (i.e. isobornyl acrylate, IBoA) with different amounts of a multifunctional acrylate (i.e. 1,6-hexanediol diacrylate, HDDA) was characterized using the NIR-set-up. Another system consisting of a multifunctional acrylate (i.e. HDDA) containing 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

different amounts of a trithiol (i.e. trimethyolol propane tris (3-mercaptopropionate), TMPMP) was studied by analysis of both NIR and MIR spectra ranging from 2000 7000 cm-1. A combination of MIR and NIR enables the in situ characterization of thiol-ene systems by following both, the H-S thiol peak (at MIR) and the C=C double bond peak (at NIR). This enables the analysis of thicker layers (~ 50 - 500 µm) ensuring reliable rheological data and a thiol signal with high absorbance which would be a limiting factor for characterizations solely in the MIR region.31 More advanced photopolymer precursors such as epoxy resins and hydrogels have been investigated as well. The option to measure wax-like monomers or oligomers at higher temperatures using a Peltier-heated glass plate and an integrated heating hood, is shown by studying the cationic photopolymerization of an epoxybased resin (bisphenol-A-diglycidyl ether, BADGE). Finally, the in situ monitoring for the photopolymerization of hydrogel precursor formulations is demonstrated using heavy water as matrix.

Materials and Methods Chemicals. The photoinitiators ethyl (2,4,6-trimethylbenzoyl) phenylphosphinate (TPO-L, Lambson),

4-iso-propyl-4'-methyldiphenyliodonium

tetrakis(pentafluorophenyl)-borate

(BARF, TCI), 2-hydroxy-4’-(2-hydroxyethoxy)-2-methyl-propiophenone (Irgacure 2959, Ciba), and the solvent deuterium oxide (D2O, Eurisotop) were used as received. The photoinitiator lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LiTPO) was synthesized as descripted recently.32 The monomers isobornyl acrylate (IBoA, Sigma Aldrich), 1,6-hexanediol diacrylate (HDDA, Alfa Aesar), trimethylolpropane tris(3-mercaptopropionate) (TMPMP,

Bruno

Bock),

bisphenol-A-diglycidyl

ether (BADGE,

Huntsman),

and

poly(ethylene glycol)-diacrylate (average Mn ~ 700 g mol-1, PEGDA700, Sigma Aldrich) were also used as received.


Page 13 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Preparation of resin formulations. The resin formulations were always freshly prepared and stored under exclusion of light. 0.5 mol% TPO-L (for BADGE 0.5 mol% BARF was used as photo acid generator) were added to all resin formulations and then mixed with a vortex mixer. The vials containing the formulations were then submerged into an ultrasonic bath for approximately 30 min. Hydrogel precursor formulations were prepared by weighing an appropriate amount of Irgacure 2959 or LiTPO (for a final content of 0.2 wt% photoinitiator in the formulation), respectively, into vials. Thereafter, the calculated amounts of PEGDA700 and D2O were weighed into the vials to obtain formulations with 30, 40, and 50 wt% monomer content for each photoinitiator series. RT-IR-photorheology set-up. Together with the company Anton Paar GmbH in Graz we have developed a unique real-time-IR-photorheology prototype for the in situ characterization of chemical and mechanical information during photopolymerization reactions of crosslinking resins (Figure 1a and S1). The coupling of a Bruker Vertex 80 FTIR spectrometer, equipped with interchangeable MIR/NIR optics, a rapid scan module, and an Anton Paar MCR302 WESP rheometer is depicted in Figure 1. An optical channel from the left exit of the IR spectrometer to a temperature-controlled, IR transparent window is incorporated at the bottom of the rheometer. The IR light exits the spectrometer through a KBr window as a parallel beam and is focused by an off axis gold coated parabolic mirror through a window onto the sample. IR and UV/VIS transparent windows made from KBr, NaCl, or CaF2 for MIR and borosilicate glass for NIR can be mounted, respectively. The parallelism of the window and the rheometer plate was confirmed using a dial gauge (Figure S2). The window can be temperature controlled by Peltier elements located around the window holder in the temperature control system (Anton 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

Paar P-PTD200/GL). Additional heating from above by an external Peltier-controlled hood was employed for experiments at elevated temperatures (H-PTD 200 from Anton Paar), allowing a maximum temperature of 200 °C (Figure S2). A fixed temperature was selected and kept constant over the course of each measurement (20 °C for acrylic and thiol-ene, 50 or 70 °C for epoxy-based formulations). In case of hydrogel precursor formulations, the temperature was set to 25 °C. The temperature was monitored during the experiment and no significant increase in temperature caused by irradiation or the exothermic polymerization reactions was observed. Disposable aluminum, steel or gold-coated steel plates with diameters of 8-25 mm can be used as rheometer plates. Smaller geometries are preferred for stiffer samples, since the influence of instrumental compliance is reduced in comparison to larger plate diameters, thus allowing the measurement of samples having larger elastic moduli. The reflected light is collected and focused onto an external mercury cadmium telluride (MCT) IR detector (Bruker DigiTect) by a second off axis gold coated parabolic mirror. IR spectra were measured at 8 cm-1 resolution (4 cm-1 for hydrogel precursor formulations) and a scanner velocity of 40 kHz (up to 320 kHz possible) by dividing sample spectra with a preset sample thickness of typically 50-500 µm (distance between optical window and rheometer plate) against background spectra of the optical window in direct contact with the rheometer plate. Due to the almost perpendicular incidence of the IR beam with respect to the reflective measurement plate, the resulting path length in the MIR/NIR is twice the preset measurement gap of the rheometer (typically between 100 µm and 1 mm). Potential artefacts arising from this configuration with respect to a quantitative spectra evaluation are described in the supporting information (Figure S3 and S4). One important feature of this set-up is that sample layers up to 500 µm thick can be probed in transmission. A technically much simpler ATR set-up with the sample in contact with an ATR crystal and probed by internal reflection would provide IR information only from a very thin 8 ACS Paragon Plus Environment

Page 15 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


surface layer of typically a few µm defined by the penetration depth of the evanescent wave. The rheological properties of such a thin film, however, may not be representative for the bulk sample. For initiating photopolymerization reactions UV-light is projected via a waveguide into the sample through the bottom window using an Exfo OmniCureTM 2000 light source with a broadband Hg-lamp (320-500 nm). In general, monochromatic irradiation from a coupled LED light source could also be realized. The UV-light was directed towards the sample by using a double-headed light guide with diffusers on each end to ensure homogeneous irradiation conditions throughout the sample area. (Figure S5 and S6).

Figure 1. a) RT-IR-Photorheology set-up: Bruker Vertex 80 FTIR spectrometer coupled with an Anton Paar MCR302 WESP rheometer using an optical channel with integrated parabolic mirrors, an external MCT detector, and an Omnicure UV-light source; b) Schematic illustration of the IR beam path.

Measurement procedure for RT-IR-photorheology. The rheometer was equipped with a parallel plate steel measurement system (Ø = 25 mm). A specific sample volume (150 µl) was placed at the center of the optical plate and the measurements were conducted at a defined temperature and sample thickness (200 µm for acrylic, thiol-ene, and epoxy-based formulations). In case of hydrogel precursor formulations, a fixed sample volume of 270 µl was injected into a 500 µm gap. Before UV irradiation, the respective samples were measured 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

via IR and analyzed by rheology. The formulations were sheared with a strain of 1% (0.1% for hydrogel precursor formulations) and a frequency of 1 Hz. CaF2 windows for MIR and borosilicate windows for NIR have been used, respectively. In general, measurements directly on the IR transparent window are possible (also for highly crosslinked resins24) and have been conducted for hydrogel precursor formulations, but such measurements potentially cause damage to the IR transparent windows when glassy-type materials with high adhesion are formed. This called for an adjusted measurement method in the case of samples based on highly crosslinking monomers especially when the CaF2 window was used. For protection of the optical windows adhesive polyethylene tape from TESA, 4668 MDPE, was used. The irradiation intensity was determined on the surface of the PE tape at the position of the sample. The IR signals of the PE tape do not interfere with the relevant IR signals from the respective reactive units and are also cancelled out by including the tape in both, sample and background measurements. The moderate settings for the rheological measurements (strain 1%, 1 Hz) also ensure reproducible rheological information with the tape present.29 In order to initiate photoreaction the samples were illuminated by UV-light with a wavelength of 320-500 nm for a period of 300 s (600 s for epoxy-based formulations). The UV-light intensity was approximately 10 mW cm-2 on the surface of the sample, except for the hydrogel precursor formulations where an intensity of ~ 30 mW cm-2 was applied. An Ocean Optics USB 2000+ spectrometer was used to measure the total irradiation intensities at the position of the samples. The whole sample was illuminated at once and the respective illuminated volume (Vi) can be calculated from the radius of the upper measurement plate (r = 12.5 mm) and the used measurement gap.


Page 17 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


During the measurements, the storage modulus and loss modulus were recorded. The chemical conversion (double bond conversion DBC; thiol conversion TC; epoxy conversion EC) was determined by recording a set of single spectra (time interval ~ 0.25 s) and then integrating the respective peaks. The relevant peaks for the reactive groups of each tested formulation (i.e. 6080 – 6250 cm-1 for acrylic, 6100 – 6250 cm-1 and 2440 - 2635 cm-1 for thiol-ene, 4493 – 4585 cm-1 for epoxy-based and 6125 – 6250 cm-1 for hydrogel precursor formulations) were then integrated over the reaction time. The ratio of the relevant peak area from the start to the end of the measurement gave the DBC, TC, and EC, respectively. Considerations and limitations for the presented RT-IR-photorheology set-up. The measurement of thermosetting resins provides a number of challenges. For once, the resulting high hardness and stiffness of the cured resins and their resulting good adhesion to the transparent (for UV/VIS and MIR/NIR radiation) glass or CaF2 plate make the removal of the cured sample difficult. Therefore, a protective PE tape can be used to avoid destruction of the windows. Additionally, a disposable upper plate measurement system with exchangeable aluminum plates is available.

Sample thickness. The optimum sample thickness for the rheological characterization lies in the range of 50-500 µm. At a sample thickness < 50 µm the rheological information becomes unreliable and at a sample thickness of > 500 µm a uniform photocuring reaction cannot be ensured (potential gradient throughout the sample due to limited light penetration). The maximum sample thickness for sufficient UV-light irradiation (extinction < 1) can be estimated using Lambert-Beer’s law and taking into consideration the concentration and extinction coefficient of the respective PI used (typical PI concentrations are < 0.2 mol L-1 and extinction coefficients are usually < 100 L cm-1 mol-1). Moreover, the applied resin formulations should show limited to no absorption in the relevant wavelength spectrum for photoinitiation (dependent on the PI, but in general between 320-460 nm). 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

This new hyphenated RT-IR-photorheology set-up was specifically designed to analyze samples within the MIR or NIR spectroscopical region, which allows for more degrees of freedom in IR analysis. The options of analyzing weak MIR signals with an increased sample thickness (SH peak) or characterization of overtones of MIR bands, if their intensity is too high in the MIR region (acrylate at ~ 6200 cm-1), become available. Depending on the respective IR signal being analyzed the spectral region for analysis is chosen accordingly. When thinking about reactive groups, which are analyzed via MIR or NIR, the amount of chromophore depends on the extinction coefficient of IR bands, which often vary by 4 orders of magnitude. With the option of choosing MIR or NIR for analysis, a sample thickness of as small as 10 µm and as large as 1 mm is generally possible. The absorbance of the relevant peaks was aimed to be in the range of 0.05-1 and becomes a deciding factor for MIR or NIR characterization. Additionally, the concentration of reactive functional units (e.g. acrylic double bonds) within the sample volume is crucial. For instance, highly diluted systems with less than ~ 20% monomer and monomers with high molecular weight (> 5000 g mol-1) can only be measured in the MIR region. Here, the absorption of the solvent is crucial (measurements with D2O are not possible due to its high absorptivity in the MIR region, but CCl4 as solvent with little IR signals might be an option).

Rheological settings. The oscillation protocol for the curing reactions involves moderate frequencies (1 Hz) and amplitudes < 1%, to minimize the chance for slippage and delamination of the sample from the measurement system. The gel point during photocuring is determined by the intersection of G’ and G’’, which corresponds to tan(δ) = 1. A more correct way of determining the gel point and ensuring that the gel point is not a function of frequency can be achieved by measurements at different frequencies. The retrieved tan(δ) plots would intersect at a single point, which corresponds to the critical gel point.19,20,33 12 ACS Paragon Plus Environment

Page 19 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


However, within this study we have assessed the intersection of G’ and G’’ at a single frequency to determine the gel point. In case of rapid, radical curing reactions gelation is often observed at times < 5 s. The set frequency of 1 Hz has been evaluated as optimal value and frequencies > 10 Hz are already unsuitable for the in situ characterization of most thermosetting resins. For the investigated hydrogel precursor formulations such an intersection was generally not observed, thus an alternative evaluation of gel time needed to be conducted (see section for hydrogel precursor formulations).

Reproducibility and error of measurements. Measurements are usually performed in triplicate and show good reproducibility (see Table 1). For good reproducibility the applied amount of sample (ensured via pipetted sample volume), irradiation conditions, and temperature need to be uniform for all measurements. Parameters such as time of gelation (tg) and conversion (Cfinal) can be determined with errors of usually less than 5%, while parameters such as final modulus (G’max) or shrinkage force (FN) are more challenging to evaluate and contain errors of sometimes up to 20%. This can be explained by the influence of adhesion and friction between the rheometer plate and the sample on G’max and FN. Concerning MIR/NIR spectral evaluation, the reproducibility for all resin systems is very good (error < 3%, except for thiol conversion, where the error is < 10% due to a lower signal to noise ratio in the MIR region). Previously, a good correlation between conversion values assessed by photo-DSC and the presented RT-NIR-photorheology set-up was found.24 Additionally, the evaluated conversion with IBoA was assessed via 1H-NMR spectroscopy and the derived DBC values are in good agreement (87% via NIR vs 85% via 1H NMR, Figure S7).

Results and Discussion Acrylic formulations. As first example, the monofunctional acrylate IBoA was studied neat and with 2 different concentrations of crosslinker, 10 and 50 wt% HDDA, respectively. Low 13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 38

molecular weight monomers have a relatively high concentration of double bonds within their respective sample volume, thus making measurements in the NIR-region most suitable. The MIR-region at around 1640 cm-1, where the interesting double bond signals of acrylates appear, would be over saturated already at a layer thickness of > 10 µm. This low film thickness does not allow reliable rheological measurements, where a gap of at least 50 µm is required. In our experiments, 150 µl of resin were placed on the glass window of the rheometer for each measurement, which was observed to impart a sample thickness of 200 µm. The temperature was set to 20 °C and the irradiation parameters selected as described in the experimental section.


Page 21 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. RT-NIR-Photorheology study of IBoA (▬) and formulations with 10 wt% (-.-) and 50 wt% (---) HDDA; a) storage modulus G’; b) normal force FN; c) double bond conversion DBC.

The conducted experiments enable an extensive characterization for the photocuring reaction of a monofunctional acrylate with various crosslinker concentrations. The curing of the resin during UV irradiation is monitored with the rheometer giving storage modulus G’, loss modulus G’’, and normal force FN plots over time and the chemical information is retrieved 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 38

via the decreasing NIR signal of the acrylate double bond at ~ 6160 cm-1 during irradiation (Figure 2 and S8). The intersection of G’ and G’’ is defined as the gel point and resembles the transition from liquid resin to solid polymer.34 This intersection corresponds well with the onset of measured normal force, which is an indication of the polymerization induced shrinkage and shrinkage stress evolution (Figure 3). During the measurements, the gap is kept constant, thus enabling direct correlation of FN with shrinkage stress. However, physical effects such as adhesion to the glass window or the steel plate of the measurement system, glass transition of the cured material and measurement temperature have to be taken into consideration, when drawing conclusions on the developed shrinkage stress.

Figure 3. Detection of shrinkage stress using the photocuring of formulation IBoA/10 wt% HDDA as example (liquid state: grid structure; solid network: check pattern).


Page 23 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


When interpreting the accumulated results from the RT-IR-photorheology experiments the time to reach gelation (tg) is the first result that needs to be considered. Naturally, the faster a system gels, the more reactive a photopolymerization will be viewed. Simultaneously to the tg, the chemical conversion at the gel point (DBCg) is reported via the monitoring of the NIR double bond signal giving a crucial value for the characterization of the photopolymerization reactions. As expected, it can be assumed from the G’ plots in Figure 2a and clearly seen in Table 1, that resins with more HDDA as crosslinker gel faster, thus represent a faster reaction speed, compared to non-crosslinking IBoA (tg = 11.4 s for IBoA versus tg = 3.8 s for IBoA/50 wt% HDDA). Moreover, the DBCg for IBoA/10 wt% or 50 wt% HDDA is with < 20% vastly reduced compared to ~ 60% for neat monofunctional IBoA. As described in Figure 3, the polymerization induced shrinkage stress is a direct consequence of DBCg and the conversion after gelation. The observed FN plots confirm this statement and give much higher FN values for resins cured with higher crosslinker content (FN ~ -5.5 N for IBoA versus FN ~ -8.6 N for IBoA/50 wt% HDDA, Figure 2b). For the cured material, the value of G’max showed an increase with increase in concentration of HDDA pointing towards a higher extent of crosslinking. Another important conclusion from the conducted measurements is the final conversion (DBCfinal) of the cured resins, which describes the final extent of photopolymerization. It is interesting to note that the final conversion in all cases (Figure 2c and Table 1) did not vary considerably (~ 87 - 88%) even though gel point conversion decreased with increase in concentration of HDDA. It has been reported that the final stages of a photoinitiated polymerization is dependent on mobility restrictions arising with increase in viscosity of the system with conversion.35 Faster conversion and increased modulus can result in enhanced shrinkage stress (lower FN value) with respect to chemical conversion towards the final stages of photopolymerization. Hence, the faster system (IBoA/50 wt% HDDA) will encounter faster mobility restrictions which 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 38

resulted in a similar extent of DBCfinal for all formulations. The mobility restriction, which prevented higher conversion with increase in double bond concentration, can thus be correlated to the evolved values of G’max (Table 1). A higher value of DBCg and monofunctionality of IBoA resulted in higher values of t95% (time taken to attain 95% of final observed conversion) than formulations including crosslinker HDDA. This is further proof for a faster reaction for formulations with added crosslinking monomers. Table 1 RT-IR-Photorheology results (tg … time to gel point, Cg … conversion at gel point, Cfinal … final conversion, t95% … time to reach 95% of final conversion, G’max … maximum storage modulus, FN … shrinkage force). resin IBoA IBoA/10 wt% HDDA IBoA/50 wt% HDDA HDDA HDDA/20 rg% TMPMP HDDA/40 rg% TMPMP BADGE (50 °C) BADGE (70 °C) a values derived from DBC plot b values derived from TC plot

tg (s) 11.4 ± 0.1 5.4 ± 0.1 3.8 ± 0.1 1.8 ± 0.1 2.8 ± 0.1 3.4 ± 0.1 198 ± 2 90 ± 1

Cg (%) 60 ± 2 17 ± 2 12 ± 1 28 ± 2 a 41 ± 3 / b27 ± 2 a 62 ± 4 / b32 ± 3 24 ± 2 26 ± 2

Cfinal (%) 87 ± 1 87 ± 1 88 ± 0 88 ± 1 a 94 ± 1 / b63 ± 2 a 97 ± 0 / b60 ± 5 64 ± 1 82 ± 2

t95% (s) 50 ± 7 33 ± 2 35 ± 3 47 ± 8 a 21 ± 6 a 18 ± 3 528 ± 14 441 ± 43

G’max (kPa) 721 ± 67 924 ± 37 999 ± 53 859 ± 9 818 ± 11 461 ± 11 476 ± 17 405 ± 14

FN (N) -5.5 ± 0.4 -7.5 ± 1.7 -8.6 ± 0.5 -12.7 ± 0.7 -8.2 ± 0.3 -2.4 ± 0.1 -8.9 ± 0.3 -7.7 ± 0.3

Thiol-ene formulations. For the evaluation of thiol-ene formulations a simultaneous determination of conversions at MIR and NIR regions (i.e. from 2000-7000 cm-1, thiol signal ~ 2570 cm-1, acrylate signal ~ 6160 cm-1, Figure S9) is necessary. With the expansion to the MIR region, thiol-ene formulations can be characterized also with respect to their SHconversion, which is challenging when only thin layers of resin are cured, as the SH signal is rather weak.31 As optical window, a CaF2 plate was used to ensure penetration of IR radiation at wave numbers > 1000 cm-1. For this study, the difunctional acrylate HDDA was selected as classical ene-compound and the trifunctional TMPMP as suitable thiol compound. The concentrations of thiol in the compositions were 0, 20, and 40 rg% (rg% … % reactive groups in the formulation; reactive groups are SH and acrylate double bonds). Radical photopolymerizations conventionally undergo a chain growth mechanism, which results in an increase in modulus and shrinkage stress with an increase in concentration of crosslinking monomer. However, in the presence of thiol, the photopolymerization 18 ACS Paragon Plus Environment

Page 25 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


mechanism involves a mixed chain growth / step growth process, which in turn is strongly dependent on the reactivity of the constituents.36 Furthermore, it is well known that a higher concentration of thiol-ene linkages reduces the rigidity of the formed network and hence its modulus, as the thioether linkages are flexible. In addition, the thioether linkages can impart enhanced network toughness. This makes the material more resilient or rubbery and a reduction in shrinkage stress is expected. The characteristic values for the gelation process, tg and DBCg, of the measured thiol-ene resins showed a delayed gelation and enhanced gel point conversion, with an increase in concentration of thiol (Table 1). The delayed gelation occurring due to the presence of thiols results from the introduced chain transfer reactions and the decreased G’max can be explained by the network flexibility introduced by the formed thioether bridges (Figure 4a). As expected, the increased DBCg results in reduced shrinkage stress with higher concentration of thiols (FN ~ -12.7 N for HDDA versus FN ~ -2.4 N for HDDA/40 rg% TMPMP, Figure 4b). The evaluation of thiol conversion in the MIR region simultaneously with acrylate double bond conversion in NIR showed increased conversion of double bonds in the presence of thiol (Figure 4c and S9). Addition of 20 rg% TMPMP itself resulted in around 94% DBC, which increased marginally to 97% at 40 rg% concentration of TMPMP. However, with reference HDDA only 88% DBC in the absence of thiol were reached. The thiol conversion (TC) profiles show nearly equivalent conversion for both concentrations of thiols (~ 60%), but lower than the respective DBC of acrylates. If the reaction of thiol to acrylate would follow an ideal alternating mechanism and suppress the propagation of acrylates, then the system would show nearly full conversion of thiol. In general, the reactivity of thiol to any ene-compound is found to vary with its nature.37 If there is a 1:1 reactivity, then the concentration of thiyl radical is equal to the carbon centered radical so that the rate constant for chain transfer (ktr) is equal to the rate constant of propagation reaction (kp). At the same time, if the thiol is much less reactive, then ktr

MIR-Photorheology: A Versatile ... - ACS Publications

Recommend Documents