Printer Friendly

Kinetic elucidation of a novel photoproduct via irradiation of isopropylthioxanthone with maleic anhydride.

Abstract Isopropylthioxanthone is a popular cosynergist for Type II and three-component photoinitiators used to photopolymerize ultraviolet (UV) curable coatings and can be used in the presence of maleic anhydride derivatives to efficiently cure coatings used in numerous applications. Isopropylthioxanthone [ITX] was shown via ultraviolet-visible (UV-vis)spectroscopy to yield a single photoproduct with maleic anhydride (MA)when irradiated ([lambda] [greater than or equal to] 350nm]). Following irradiation of ITX and MA at set time intervals, high-performance liquid chromatography (HPLC) was used to observe the formation of the photoproduct by monitoring the UV-vis absorbance spectra of the photoproduct and the reactants. Based on the consumption of ITX shown in the HPLC plots, the complete rate law calculated based on the initial rates. Finally, a possible mechanism is proposed for the formation of the photoproduct which may provide evidence for the reduced initiation efficiency of photoinitiators incorporating ITX and a MA derivative when compared to a similar photoinitiator substituting an N-substituted maleimide for the MA derivative.

Keywords Isopropylthioxanthone, Maleic anhydride, Photoproduct, Kinetics, Photoinitiator


All industries look for more rapid, less expensive. and more reliable ways to produce a product. The same is true in ultraviolet (UV) radiation curing; therefore, photoinitiating systems continue to be researched and developed. The UV curable coatings industry is predicted to grow about 7% in the next few years,much of which is due to research in Types I and II photoinitiators. (1) Type I, or cleavage, photoinitiators are efficient and often use acetophenone derivatives to cleave a photoinitiator and produce two initiating radicals. Type II, or abstraction, photoinitiators are relatively inexpensive and often use UV absorbent ketones (e.g.,benzophenone) and tertiary amines that, when combined, undergo an electron/proton transfer to produce an initiating radical and a terminating radical. Developing a third type of initiator that combines efficiency and low cost would be ideal as both are desired features (2). This third type of photoinitiator has been developed for some systems and is referred to as a three-component photoinitiating system. Some of these systems use maleic anhydride (MA) derivations or maleimides (made from maleic anhydrides), tertiary amines, and an UV absorbent cosynergist (3). A three-component photoinitiating system, also offers two initiating radicals, similar to Type I photoinitiators, rather than one initiating and one terminating radical produced with Type II photoinitiators. (4)

One of these systems includes isopropylthioxanthone (ITX), 2,3-dimethyl maleic anhydride (DMMA), and N-methyl-N,N-diethanolamine (MDEA) (4). A similar system that replaces DMMA with a maleimide gives a high rate of polymerization (2). However, DMMA, which is a better electron acceptor, should produce an even faster polymerization, but it does not (2). In this system, two mechanistic pathways have been reported, chemical sensitization (Scheme 1) and Photosensitization (Scheme 2) (2). When considering the excited triplet state energies of both ITX and maleic anhydride derivatives (64 and 72 kcal/mol, respectively), photosensitization is highly unlikely since an energy transfer fronot m a lower energy to a higher energy is not thermodynamically favored (2). However, laser flash photolysis does show that a reaction is occuring on the order of 3.5 X [10 sup 9][L.mol sup -1].[sup -1] (2) Fither energy transfer occurs despite thermodynamics or an electron transfer occurs. yielding a side reaction which might explain the lack of rapidity with the overall polymerization as compared to the maleimide-based three-component photoinitiators. If this side reaction occurs, a photoproduct should be present.


MA was used as a model compound instead of DMMA since their photochemical properties should be very similar. UV--vis was used to determine. first, if a possible side reaction was occurring between the excited state triplet of ITX and MA. The consumption of ITX was monitored via HPLC, indicating ITX was reacting from its excited state with MA to produce the photoproduct. The UV--vis absorbances of ITX, MA. and the photoproduct were shown in the HPLC plots. The kinetic rate law was determined for the reaction based on the disappearance of ITX. Finally, a possible mechanism for the formation of the photoproduct was proposed.



Acetonitrile (ACN) of 99.5% purity was obtained from Sigma-Aldrich. MA of 99% purity was obtained from Across Organics. ITX of an unknown purity was a donation from the Albemarle Corporation.



Procedure 1

Approximately 44 mg of MA and 38 mg of ITX (3:1 molar ratio, respectively) were weighed and diluted into 100 mL of ACN. This mixture was then transferred into a Pyrex brand 250 mL round bottom flask. Pyrex glassware was chosen as it absorbs UV radiation having wavelengths below 350 nm, therefore preventing MA from absorbing any UV radiation during irradiation. A nitrogen blanket was placed on the system to minimize the effects of oxygen quenching the excited state triplet of ITX. The solution was then placed under a Sylvania medium pressure mercury arc lamp (HPL80MDX(R) 80 Watt (R9) 0303--outer casing removed) with stirring. A 1 mL aliquot was taken at 1 min intervals for 15 min with a final sample taken at 60 min. Each sample was diluted to 10 mL with ACN.

Procedure 2

Approximately 380 mg of MA and 38 mg of ITX (25:1 molar ratio, respectively) were used in a similar irradiation as described in Procedure 1.

Procedure 3

Individual irradiations of ITX and MA were performed to verify that neither was reacting with itself. Samples were prepared similarly as in Procedure 2, but samples were taken every 2 min between 0 and 14 min with a 60 min sample also taken.

UV--vis spectroscopy

Absorption was measured at all wavelengths (250-450nm) simultaneously with a Hewlett Packard Ultraviolet-Visible 8453 Photodiode Array. The desired solutions used ACN as the solvent.


This was performed with a system by Shimadzu USA containing these components: SCL-10A VP System Controller, SPD-M10A VP Diode Array Detector, LC-10AT VP Liquid Chromatograph, and a FCV-10AL VP Pump. The system uses a Beckman Coulter Ultrasphere C18 High Performance Column. The column measures 250 x 4.6 mm and was purchased through Alltech. Absorption was again measured at all wavelengths (250-450 nm) simultaneously with this device. Also, the HPLC chromatographically separated each species in the system, providing a retention time and UV spectrum for each.

Results and discussion


UV--vis is a simple and convenient method to monitor the absorption of ITX as it decreases with increasing irradiation time. Figure 1 shows the peak of interest for ITX at 383 nm. Figure2 shows no significant absorbance above 350 nm for MA. Once the absorbances have been measured, Beer's law can be applied to determine the extinction coefficient since the path length and concentrations are already known. Overlapping the absorbance spectra for mixtures of ITX and MA having differing irradiation times between 0 and 60 min showed that a single isosbestic point was produced at approximately 350 nm (Fig. 3). Also significant is the decrease in the height of the 383 nm ITX peak. Since an isosbestic point means the entire system has an equivalent extinction coefficient at that wavelength, a single photoproduct can then be assumed. ITX and ITX/MA samples were taken and diluted into 10 different concentrations and run on the UV--vis to calculate the extinction coefficient at 383 nm. The two runs gave extinction coefficients of 6357.7 and 6346.1 L. [mol.sup.-1] [cm.sup.-1], respectively, giving an average of 6351.9 L. [mol.sup.-1] c [m.sup.-1] and demonstrating that ITX is the only significantly absorbing species at 383 nm. Furthermore, the similar extinction coefficients of ITX and the ITX/MA mixture indicate that there is no charge transfer complexation.





Once an isosbestic point was elucidated. It became apparent that further UV--vis work would not lead to any conclusion as a photoproduct was not visible and could be competitively absorbing. HPLC was then considered as an option as it separates all components out on its column and is nondestructive. Two separate series were run on the HPLC where the molar ratios of ITX to MA were 3:1 and 25:1 and performed using Procedures 1 and 2, respectively. A photoproduct developed steadily over the first 7 min of irradiation whereupon the system reached equilibrium. The plots of the irradiated ITX/MA solution (Fig. 4) show a marked change between 0 and 7 min where the ITX peak diminishes and a photoproduct forms having a retention time of approximately 35 min. The weak absorbance near the 30 min retention time might be a second photoproduct. After 7 min of irradiation, the system reaches equilibrium and very little change is observed. Possible reactant (ITX or MA) degradation or reaction of a reactant with itself was also considered when exposed at length to UV radiation. HPLC could possibly confer this with longer UV exposure times; however, no significant change in ITX or MA is observed over a 14 min irradiation indicating no degradation or reaction with themselves.



The rate law for the observed reaction was determined in full by applying the method of initial rates to the HPLC data for the MA to ITX ratios of both 3:1 and 25:1. Initially, the rate law for the formation of the photoproduct can be expressed as shown in equation (1) where R is the rate of reaction, k is the rate constant, [ITX] is the concentration of ITX, m is the reaction order in ITX, [MA] is the concentration of MA, n is the reaction order in MA, and t is time in seconds.

- d[ITX]/dt = R = k * [[ITX].sup.m] * [[MA].sup.n] (1)

The reaction order of ITX can easily be determined by doing a best curve It for zeroth, first, and second order rate equations. This, assumes a whole number order of reaction in ITX and is reasonable as most two component photochemical reactions are first order in the excited state species. After performing the best curve fit for zeroth, first, and second orders, ITX was determined to be first order having the best correlation ([R.sup.2]). With the new rate law having the form of equation (2), the reaction order in MA can be determined using the HPLC data obtained for both the 25:1 and 3:1 MA to ITX mixtures.

R = k * [ITX] [dot [[MA].sup.n] (2)

To determine the reaction order in MA, the first order plots of both mixtures must be acquired utilizing the linear equations for each plot (Fig. 5). The slope of each linear fit is equivalent to k-[[MA].sup.n]. By dividing one equation by the other, the result is equation (3).


[m.sub.25]/[m.sub.3] = k * [[[MA.sub.25]].sup.n]/k * [[[MA.sub.3].sup.n] (3)

The rate constant is identical for both the 25:1 and the 3:1 MA to ITX mixtures. The concentration of MA for both mixtures can be defined in terms of the concentration of ITX which is initially identical for both mixtures resulting in equation (4) where the reaction order in MA, n, may be determined.

[m.sub.25]/[m.sub.3] = [25.sup.n]/[3.sup.n] (4)

Thus, the reaction was found to have a reaction order in MA of 0.5 thereby yielding the rate law in equation (5).

R = k * [ITX] * [[MA].sup.0.5] (5)

As stated previously, the slope of each linear fit from Fig. 5 is equivalent to k [[MA].sup.0.5]; therefore, the rate constant, k, for each plot can be calculated. After averaging the two calculated rate constants and converting to standard form, the complete rate law may be expressed as given in equation (6).

R = (0.0269 * [M.sup. - 0.5] * [S.sup. - 1) * [ITX] * [MA].sup.0.5] (6)

From the rate law, two observations can be made. First, the diminutive rate constant indicates a reaction that is not kinetically favored. Second, ITX is consumed faster in this reaction than MA. Thus, the effect on a three-component pholoinitiator incorporating both ITX and MA would be significant based on the consumption of ITX in the reaction to form the photoproduct even though the reaction is not kinetically favored.

Possible mechanism

The reported rate of reaction of ITX with MA is 3.5 x [10.sup.9] [M.sup.-1] [S.sup.-1] indicating an energy or electron transfer. (2) Most other possible reactions generally occur having a maximum rate constant of [10.sup.5] [M.sup.-1] [S.sup.-1]. Based on the rate constant data, a new mechanism was considered (Scheme 3). Following the excitation of ITX and subsequent intersystem crossing to produce the excited stale triplet of ITX, an electron transfer probably occurs from ITX to the electron acceptor, MA. The two species may then react to couple the radicals and produce a Zwitterion. The Zwitterion is stabilized via resonance through both the MA component and the ITX component. The Zwitterion may then neutralize itself to form a cyclization product similar to a [2 + 2] photocyclization reaction. Though kinetically unfavorable, if it were to occur, this mechanism would slow down the overall polymerization reaction, even with low quantum yields, by removing both ITX and MA from the photoinitiator system. With the low-optimized ITX concentration (1.49 x 10-4 M). the effect on pholoinitiation efficiency could indeed be dramatic, especially for extended periods of irradiation. This effect could explain the reduced polymerization rate of maleic anhydride-based three-component photoinitiators relative to similar maleimide-based photoinitiators.



Elucidation of a photoproduct was shown through an isosbestic point on UV-vis and through the absorhance of a photoproduct peak on the HPLC plots. ITX was also found to be unreactive with itself (as was MA) when irradiated with radiation in excess of 350 nm. The excited state triplet of ITX was determined to be the reactive species that reacts with MA. Based on the HPLC data, the complete rate law for the formation of the photoproduct was determined via die method of initial rates. From the calculated rate law for photoproduct formation, the effect on the rate of photo-initiation could be significant enough to explain the observed rate reduction of maleic anhydride-based three-component photoinitiators relative la similar maleimide-based photoinitiators which contradicts theory saying that better electron acceptors (e.g., maleic anhydride derivatives) should increase the polymerization rate. This behavior was explained based on the reaction of the excited state triplet of ITX with ground state MA thereby reducing the effective concentration of the photoinitiator components in the system. Finally, a possible mechanism consistent with reported rate constants was presented.

Acknowledgements The authors would like to thank the Albemarle Corporation for providing the ITX and would also like to acknowledge the financial support of the Robert A. Welch Foundation and the Abilence Christian University Office of Research and Sponsored Programs.


(1.) Cavitt, TB, Nguyen, CK, Hoyle, CE, Phillips, B, Daniels, C, Kalyanaraman, V, Jonsson, S, "Mechanisitic Elucidation of Both Maleic Anhydride and Phthalimide Photoinitiating Systems." In: RadTech 2002 Proceedings, p. 111 (2002)

(2.) Cavitt, TB, Photoinitiation of Free-Radical Polymerization by Aryl Disulfide, Cyclic Anhydride, and Phthalimide Derrivatives: ProQuest Information and Learning Company, Ann Arbor, MI (2002)

(3.) Cavitt, TB, Hoyle, CE, Nguyen, CK, Kalyanaraman, V, Jonsson, S, "Sensitized Photopolymerization of Maleic Anhydride/Acrylate Systems". In: RadTech 2000 Proceedings, p. 785 (2000)

(4.) Cavitt, TB, Hoyle, CE, Kalyanaraman, V, Jonsson, S, "Initiation of 1,6-Hexanedioldiacrylate Polymerization by Three Component Photoinitiators Incorporating 2,3-Dimethylmaleic Anhydride". Polymer, 45 (4) 1119 (2004)

Presented at e|5:UV and EB Technical Expo and Conference, sponsored by RadTech International, North America, in Chicago, IL on April 25, 2006

M.E. Mullings, C.D. Walker, A.McDonough, T.B.Cavitt (

Department of Chemistry and Biochemistry, Abilene Christian University, ACU Box 28132, Abilene, TX 79699-8132,USA e-mail: cavitt@chemistry.acu.edus
COPYRIGHT 2007 Federation of Societies for Coatings Technology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Mullings, Matthew E.; Walker, Corry D.; MeDonough, Adam; Cavitt, T. Brain
Publication:JCT Research
Date:Sep 1, 2007
Previous Article:Paints and coatings monitored by laser-induced breakdown spectroscopy.
Next Article:A new class of silicone resins for coatings.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters