Acs Symposium Series 531

ACS SYMPOSIUM SERIES 531

Photochemistry of Lignocellulosic Materials

Cyril Heitner, editor

Pulp and Paper Research Institute of Canada

J. C. Scaiano, editor

University of Ottawa

American Chemical Society. Washington, DC 1993

Chapter 3

Action Spectra in the UV and Visible Region of Light-Induced Changes of Various Refiner

Pulps

Ingegerd Forsskahl1 and Henrik Tylli2

1The Finnish Pulp and Paper Research Institute, Paper Science Centre,

P.O. Box 70, SF-02151 Espoo, Finland

2Department of Chemistry, University of Helsinki, E. Hesperiankatu 4, SF-00100 Helsinki, Finland

Various chemimechanical refiner pulps (unbleached, peroxide-bleached and pre-yellowed) were irradiated with monochromatic light at selected wavelengths. The change in reflectance at 457 and 557 nm was monitored using UV-VIS reflectance spectroscopy and post-color values were calculated from the reflectance changes. Photoyellowing and photobleaching were observed. Unit yellowing was subsequently obtained using experimentally derived kinetic curves for the reflectance versus exposure dose at a certain wavelength. The action spectra for the photoyellowing were obtained by plotting the reciprocal of the exposure dose necessary to produce a certain change versus wavelength. A different set of action spectra for both photoyellowing and photobleaching was constructed by keeping the exposure time constant and plotting the ratio of the photoyellowing or photobleaching to the light intensity at a certain wavelength against wavelength. The action spectra obtained with the two methods are similar in shape, suggesting that the observed changes are linearly dependent on the light intensity Photoyellowing was found to be most extensive with light of wavelength 310-320 nm. For strongly pre-yellowed pulp, photo-bleaching with a maximum effect at 430-450 nm was observed to be the major process on irradiation. The implications of the action spectra for the different pulps are discussed.

In spite of numerous efforts and considerable progress the detailed mechanism of the photoyellowing of high-yield pulps is still unsettled. The inherent chromophores in the wood and fresh pulps, their modification during manufacturing and subsequent reactions during aging or/and photo-oxidation are indeed difficult to establish and further work is required to reach a full understanding of the process.

The biological effects of ultraviolet exposure of human skin have been studied extensively because of the growing general awareness of the risks of solar ultraviolet exposure [1]. A further spur to research has come from the benefits of ultraviolet radiation, the ultimate goal being to developed new phototherapy and other methods. In that context the limits of safe exposure to long wavelength ultraviolet radiation have been estimated and different techniques for their assessment have evolved. High-

intensity sources of monochromatic light of various wavelengths are now available for basic research and ingenious detection systems have been developed to measure the human skin response. The knowledge obtained in these studies should be equally applicable to radiation studies of other biological materials such as wood and wood

products

The light-induced yellowing or color reversion of high-yield pulps can in some respects be compared to the photochemical ultraviolet-induced formation of erythema redness) of the skin. One important advantage for the pulp researcher, however, is the act that delayed biochemical reactions, often referred to as sunburn in case of skin, are absent. Only in fresh wood material is there any biochemical activity that might cause discoloration on storage [2]. Enzymes are readily inactivated by the heat used in manufacturing pulps from wood logs or chips.

Nevertheless, delayed photochemical reactions after the end of the exposure are

likely in both systems. For wood pulps, the outcome is strongly influenced by factors

much as the increased lifetime and much retarded transport rate due to the trapping of

intermediates and photo products in the solid matrices. This has to be taken into account

when designing experiments. The main color changes caused by light in mechanical

pulps are fairly rapid, which enables accurate measurements to be made either directly

after the irradiation or after a certain delay. Thermal reactions, e.g. hydrolysis and

oxidation, which also occur in the pulps, can be prevented by keeping the pulps in a

freezer before analysis.

Most photo-biological responses follow the reciprocity law, which states that a given exposure dose yields a constant biological response. This means that constant quantities of some photochemical reaction products are produced per absorbed photon >t given wavelengths and that these products yield a constant biological response [1]. This should certainly be true also for light-induced changes in pulps.

Action Spectra

The method used in the present work to obtain action spectra of wood pulps is analogous to that developed to construct the erythema action spectrum for human skin /I]. In order to evaluate the action spectra for the photochemical discoloration of the >pulps, the spectral radiance at the sample position must be measured and the response of the exposure, e.g. in terms of the reflectance at 457 nm, has to be analyzed. Thus, in action spectrum takes into account both the exposure dose and the character of the chromophores that are photoactive in the pulp. The action spectrum for the yellowing is hen obtained by plotting the reciprocal of the exposure dose necessary to produce a •certain predetermined degree of yellowness versus wavelength.

In principle, a carefully constructed action spectrum may be used to identify the

absorbing chromophore. This is possible at least when the action spectrum corresponds

losely to the absorption spectrum of a molecule that can be shown to react

Photochemically in the wavelength region studied [3). The success of this approach

lecreases rapidly as the number of different reacting chromophores increases. In such

ases the action spectrum becomes complex and difficult to interpret. When a large and

aried number of chromophores are involved, some of them may only act as

intermediates in conveying the incident energy to important sites in the material [3],

naking the interpretation of the action spectrum even more complicated. Further

difficulties are shifts in the absorption spectra caused by solvent changes and different

states of aggregation.

The situation is also complicated by the fact that the response, e.g. the yellowing, in reality is not a function of a single radiation wavelength, a fact that has ed to the construction of polychromatic action spectra [3). Nevertheless, the polychromatic action spectra become very complex and may obscure the individual chromophores.

The wavelength dependence of the photoyellowing and photobleaching of

ground wood was studied by Nolan et al. \4] and van den Akker et al. [5] more than forty years ago. The latter authors presented a relative spectral sensitivity curve over the wavelength region 250-385 nm for eastern spruce ground wood together with spectral absorption coefficient curves of native lignin and lignin derivatives. Based on their results, the authors stated that lignin is mainly responsible for the yellowing and darkening of the color caused by light [5]. Later Leary [6] and Claesson et al. [7] studied the yellowing and bleaching of newsprint and mechanical pulps using light at different wavelengths. Recently, the same phenomena have been investigated either with filter combination systems [8] or using monochromatic illumination [9-11]. Summarizing the results, yellowing or discoloration was found to occur at 360-395 nm [6], at 290-390 nm [8], at 255-346 nm with the maximum effect at 310-328 nm [9], at 340 nm (maximum effect) [10] or at 310-320 nm (maximum effect) [11] and bleaching or brightening, often of a pre-yellowed sample, at 410-520 nm [6], 396-420 nm and 420-470 nm [8], at 420-500 nm [9] and at 420-430 nm (maximum effect) [11]. Considerable variations with tree species were found.

A study of the wavelength sensitivity of the light-induced yellowing of newsprint containing thermomechanical pulp (TMP) made from unbleached loblolly pine has recently been published [72] in which some action spectral data in the region 280-600 nm are presented. As far as we are aware, no action spectra for the light-induced changes of untreated and treated (bleached or pre-yellowed) chemimechanical and chemithermomechanical pulps made from spruce have been published.

Experimental

Outline of Methods Used to Obtain Action Spectra. One problem encountered in wood photochemistry is that it is difficult to irradiate pulp sheets to a certain predetermined level of yellowness with the equipment usually available for response detection, the reflectance spectrometer. The exposure dose required for unit yellowing, which for a given area equals the product of radiance and exposure time (H=E t, where H is the exposure dose, often expressed in J/cm2, E is the irradiance, expressed in W/cm2, and t is the radiation time in seconds), is different for different radiation wavelengths and has to be evaluated. This problem was solved in an empirical way. In method 1 experimentally derived kinetic curves for the various pulps were obtained by plotting the yellowing versus the exposure dose. The measurements were made at certain wavelength intervals. From these curves the exposure dose required for unit yellowing was obtained graphically. The kinetic curves change slowly with the radiation wavelength and no sudden discontinuity could be found. In the vicinity of the most effective wavelengths the kinetic curves were measured at shorter wavelength intervals.

In part of this work, method 2. the radiation time was kept constant for all selected wavelengths, 2h for unbleached and bleached chemimechanical pulps (CMP and CMPB), and 7h for an untreated (CTMP) and a pre-yellowed (CTMP-Y) chemithermomechanical pulp. The action spectra of the pulps were then constructed by plotting the ratio of the degree of yellowing in PC units to the number of quanta/cm2 provided by the light source - monochromator combination at the various wavelengths as a function of the radiation wavelength.

Pulps. Chemimechanical pulp (CMP) was made from fresh spruce chips on a laboratory scale and the pulp was bleached with 4% hydrogen peroxide as previously reported [11]. Chemithermomechanical pulp (CTMP) was made in a full-scale refiner (at KCL) after pretreatment of industrial spruce chips (Picea abies) with sodium sulfite solution (30 g/I) at 130°C for 5-10 min, followed by pressurized refining.

Thick sheets (ca. 400g/m2) were made from all pulps. The sheets were stored in the freezer before being irradiated

Irradiation experiment. Pre-irradiation of the CTMP pulp for 90 h was performed at 23°C and 50% RH in a Xenotest 150 S (Hereaus Hanau) apparatus equipped with • 1.3 kW xenon lamp and with IR and glass filters (UV cut-off at 310 nm). The irradiation caused strong yellowing of the pulp and the reflectance measured at 457 nm decreased from 64.3% to 29.9% and that at 557 nm from 77.2% id 58.4%.

Irradiations with monochromatic light were carried out with an Applied Photophysics Model 5350 photo-irradiator, equipped with a 900 W short arc high pressure xenon lamp, a f/3.4 monochromator and an exit lens of quartz providing illumination over an area of 1.3 by 2.4 cm at a distance of 16 cm from the lens. The entrance slit of die monochromator was kept at 5 nm and the exit slit at 10 nm to secure a sufficiently high output energy. Irradiations were performed at ambient temperatures (ca. 23°C) in air. The fight source - monochromator combination was calibrated with an Applied Photophysics precalibrated thermopile coupled to a voltmeter. The active area of the thermopile was 10 mm2. Several points over the illuminated area were measured and a small decrease towards the edges was found. However, the central area which was analyzed using reflectance spectroscopy was much smaller (0.8 by 1.7 cm) and over this area the illumination was uniform. The spectral irradiance measured for the xenon arc - monochromator assembly is shown in Figure I.

UV-VIS reflectance spectra of the pulps were recorded directly after irradiation in the wavelength range 250-750 nm on a Perkin-Elmer Lambda 15 spectrophotometer equipped with an integrating sphere. The reflectance values (R.., strictly speaking the reflectivity of an infinitely thick specimen) at 457 and 557 nm were taken from the reflectance curves. Difference spectra were calculated by subtracting the spectrum of the irradiated pulp from the spectrum of the unirradiated one. The post-color values (PC) were calculated as previously [13] according to Giertz [14] using the equation:

(present work) also display a deep minimum for radiation wavelengths of 300-320 nm and maximum for wavelengths of 450-500 nm.

wavelengths of 280, 320 350, 370, 390,410 and 480 nm for two hours are shown in

CTMP pulp 60 mmol/kg

For pre-yollowed CTMP, irradiation at long wavelengths (450-550 nm) caused

The amount of photoyellowing of the pre-yellowed CTMP caused by radiation

wavelengths below 410 nm was of the same magnitude as the amount of photobleaching. The overall changes in intensity for the pre-yellowed pulp caused by iradiation at selected wavelengths were small (about 6 units, Figure 6) compared with hose for the untreated CTMP (about 18 units).

Post-Color (PC) Values of Irradiated Pulps. The PC values at 457 nm and

557 nm were calculated for the irradiated pulps and plotted as a function of the radiation

wavelength. The results are shown for the untreated CTMP in Figure 7and for the pre-

yellowed CTMP in Figure 8. Very similar curves were obtained for CTMP (Figure 7)

for CMP and CMPB (figures not shown), whereas the curve for pre-yellowed

pulp, CTMP-Y (Figure 8) was very different. For the first three pulps the most

koefficient wavelength causing yellowing was around 310-320 nm. This region was

somewhat broader when the response of the irradiation was measured at 557 nm. In

terms of the PC values, bleaching the pulps did not seem to change the situation at all:

he unbleached and the bleached pulps yellow to about the same extent

For CTMP-Y, there occurs a small decrease in reflectance, expressed as PC at 157 nm, at short radiation wavelengths (around 270 nm). an increase in reflectance at 290-340, a sudden decrease in reflectance at 350 nm and a pronounced brightening of he pulp by wavelengths in the region from 360 nm to at least 550 nm. The somewhat surprising variation in positive and negative PC values in this curve for different adiation wavelengths is supported by the curve generated using the measurements at 657 nm. In the latter curve the increase in reflectance at around 350 nm extends over a

Figure 3. UV-VIS difference reflectance spectra (unirradiated minus irradiated) of untreated chemirnechanical pulp (CMP) after irradiation with light at selected wavelengths.