- Email: [email protected]
Temperature effect on an ultrasound-assisted paper de-inking process
Ultrasonics Sonochemistry 16 (2009) 698–703
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultsonch
Temperature effect on an ultrasound-assisted paper de-inking process Anne C. Gaquere-Parker *, Ayan Ahmed, Temitayo Isola, Bintu Marong, Christopher Shacklady, Phoebe Tchoua Chemistry Department, University of West Georgia, 1601 Maple Street, Carrollton, GA 30118, United States
a r t i c l e
i n f o
Article history: Received 10 May 2007 Received in revised form 29 December 2008 Accepted 6 January 2009 Available online 24 January 2009 Keywords: Sonication Degradation Ink Paper recycling Cavitation Ultrasound
a b s t r a c t The influence of temperature on an ultrasound-assisted ink removal process has been investigated. White copy paper was evenly soaked in black writing ink. After drying the paper to constant weight at 75 °C, ink removal was attempted under varying conditions. Results were assessed by monitoring the UV–vis absorbance of the aqueous phase and measuring the brightness of the paper. Sonication was observed to improve the brightness of the paper in the temperature range of 15–45 °C with an optimum effect at 35 °C. Monitoring UV–vis spectra of the aqueous phase provided evidence that modification of the chemical structure of the ink desorbed from the paper occured. Further investigation under the same conditions showed that ink, when not absorbed on paper, did not undergo the same chemical change. This supports the hypothesis that only the compound released from the ink absorbed onto the paper is sensitive to sonodegradation. One possible explanation is that the metal binding component of the ink stays absorbed on the paper, releasing the organic part, whose chemical structure can be altered by the effect of sonication. Inductively coupled plasma analysis was used to confirm that during the de-inking process of the paper, the metal binding component stays absorbed on the paper and only the organic part is released in the aqueous phase. Published by Elsevier B.V.
1. Introduction In recent years, interest in paper recycling has increased for a number of environmental reasons. A major problem associated with paper recycling is the whitening process, with emphasis being placed on the production of brighter de-inked pulps. Traditionally, oxidizing agents are used to improve the de-inking process and whitening of the paper pulp. Among them are hydrogen peroxide [1], ozone, chlorine dioxide, hypochlorite, dithionite and formamidine sulfinic acid [2]. Ultrasound has been utilized in the paper processing industry at different stages; of which enhancement of pulping, bleaching, depolymerisation of cellulose and treatment of wastepaper have been reported [3–6]. During the process of bleaching, ultrasound has been used to enhance the de-inking of recovered paper when traditional alkaline treatments were not satisfactory [7–12]. These studies report the successful use of ultrasound with temperatures up to 90 °C, in some cases in the presence of additives. In addition, ultrasound has been shown to improve the quality of paper fibers recovered in the recycling process [13]. Other studies report that when varnish is present, optimum results have been attained at temperatures of 65 °C [14]; 70 °C [15] or 80 °C [8], as heat is needed to soften the varnish present. However, excluding one paper that discusses the influence of tem* Corresponding author. Tel.: +1 404 895 0100. E-mail address: [email protected] (A.C. Gaquere-Parker). 1350-4177/$ - see front matter Published by Elsevier B.V. doi:10.1016/j.ultsonch.2009.01.004
perature (20 °C, 40 °C and 60 °C) on the removal of indigo dye from paper in the presence of ultrasound [6], no previous study has investigated the effect of sonication at lower temperatures on unvarnished paper. Furthermore, the articles cited above describe the influence of ultrasounds on the ink particle size but not on its chemical structure. The main advantage of using sonication resides in the fact that experiments can be carried out at close to ambient temperature under atmospheric pressure contrary to other advanced oxidation processes. The propagation of the acoustic waves in a liquid produces the phenomenon of cavitation [16]. The non-linear expansion and sudden collapse of microcavities result in tiny areas of very high temperature (up to 5000 K) and high pressure (500– 1000 atm) [17]. Under these highly localized but harsh conditions, formation of radicals may occur [18,19]. For instance, sonolysis of water generates hydroxyl radicals and hydrogen atoms [20]. These particles may recombine to form hydrogen and hydrogen peroxide or may react with the species present in the solution. Moreover, the high temperature and pressure conditions may contribute to the pyrolysis of volatile and hydrophilic chemical species. The effectiveness of ultrasound-assisted processes is often affected by temperature [21]. With an increasing temperature of the solvent, dissolved gases and solvent vapors enter the cavitation bubbles. This reduces the collapse and the effectiveness of the cavitation process. However, cavitation may be reduced when working at lower temperatures, since a more viscous medium hinders
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
699
acoustic wave propagation. Given such an influence the effect of temperature on the de-inking process of unvarnished inked paper has been investigated. 2. Experimental set up 2.1. Materials The 20 kHz sonication probe with a 750 W output, was manufactured by Sonics and Materials. A Suslick reaction vessel from Sonics (250 ml, three necks, flat bottom glass sonoreactor with a chamber height of 162 mm) was used. The amplitude was set at 70%. A 9 s on and 9 s off duty cycle was used for the experiments. The experiments were carried out in a temperature control water bath from Forma Scientific (model 2067). Xerox acid free paper (Business 4200) and Waterman black writing ink were used. Analytical grade nitric acid 15.8 M HNO3 from Acros Organics was used and diluted with deionized water to 7.9 M HNO3 as needed. Anhydrous ethanol was purchased from Fisher Scientific. Whatman silica gel TLC plates were used. A UVG11 UV lamp (254 nm) from UVP Company was used for TLC analysis. Ultraviolet–visible spectra were recorded using a Varian Carey IE spectrophotometer. The brightness of the paper residues was measured with a Technidyne Brightimeter S-5, with a 45° illumination and reading at 0°. The orifice was taped such that a 3 2 mm area was illuminated, providing a relative input allowing for a comparison within the samples. Except for the brightness measurements, each experiment was duplicated under identical conditions, thus the data are expressed as the mean value with the standard deviation. Photographs were taken with a Canon 20D camera, with an exposure of 1/160 s at f/11 and a lens focal length of 50 mm. The inductively coupled plasma (ICP) data was collected on a Perkin–Elmer Plasma 4000 emission spectrometer. The 1.0 103 ppm metal standards for the ICP analysis were purchased from SPEX CertiPrep Group. 2.2. Ultrasonic power measurement Due to heat loss, the actual acoustic power dissipated in solution, Pdiss, is lower than the maximum irradiated power, Pmax [22]. In order to compare data and ensure reproducibility, Pdiss has to be calculated. Calorimetry allows the calculation of Pdiss, [23,24], with the assumption that all power entering the reaction solution is dissipated as heat, using the following equation:
Pdiss ¼ ðdT=dtÞt¼0 mwater C water
ð1Þ
where mwater is the mass of the water, Cwater the heat capacity of the water and (dT/dt)t = 0 the change of temperature with respect with time. The value for (dT/dt)t = 0 is calculated from the initial slope of the increase of temperature of the water as a function of time when sonicated without any temperature control. Ultrasonic power measurement data were obtained when 100.0 ml of deionized water at 35 °C was sonicated without any temperature control. The temperature change, dT/dt, was recorded every nine seconds for eight minutes and then every four minutes for a total time of twenty five minutes. After thirteen minutes of sonication, the temperature reached a plateau at 81.9 ± 0.1 °C. The results are plotted in Fig. 1. Under these conditions, Pdiss was calculated for the first 27 s to be 83.7 ± 0.1 Watts. 2.3. Paper inking Two centimeter-edge squares of white paper were dried to constant weight overnight in a 75 °C oven. The paper was then soaked
Fig. 1. Ultrasonic power measurement: Temperature increase as a function of time.
evenly in ink for 15 min, dried to constant weight overnight in a 75 °C oven and weighed. The average ink loading was (4.7 ± 0.1) 104 g/cm2. The paper squares were stored in an air tight plastic bag in a refrigerator. 2.4. Sonodegradation of inked paper The sonoreactor containing 100.0 ml of deionized water was placed in a temperature regulated water bath (15–45 °C). When the water in the sonoreactor reached the desired temperature, two squares of inked paper were added. The sonoreactor was then covered with aluminum foil to avoid any interaction between light and the samples. Sonication was immediately started for the required amount of time (5–30 min). At the end of the experiment, the reaction mixture was filtered, diluted by a factor of 10 with deionized water and the filtrate was analyzed by UV–vis spectroscopy. The paper residue was dried at room temperature overnight, photographed and the brightness was measured. 2.5. Ink sonodegradation In the sonoreactor, 4.0 mg of ink was diluted into 100.0 ml of deionized water. This solution was sonicated for 60 min at 35 °C. Sampling was performed every minute for the first 5 min, then every 5 min for the next 25 min and at the 60 min mark. Samples were filtered, diluted by a factor of 10 with deionized water and the solution was analyzed by UV–vis spectroscopy. 2.6. Sonodegradation of the aqueous solution of desorbed ink Two squares of inked paper were stirred at 35 °C for 30 min in 100.0 ml of deionized water. After removal of the paper by filtration, the filtrate was sonicated at 35 °C for 60 min. Samples were diluted with deionized water by a factor of 10. Monitoring by UV–vis spectroscopy was performed every 5 min in 6 consecutive intervals (30 min total) and then at the 60 min mark. 2.7. Thin layer chromatography A solution consisting of 4.0 mg of ink diluted in 100.0 ml of deionized water, the filtrate obtained in part 2.6 before sonication and after sonication, were spotted on a TLC plate. The TLC plate was eluted with ethanol and analyzed at 254 nm with a UV lamp. 2.8. Iron, manganese and copper analysis by ICP spectrophotometry 2.8.1. Metal identification The most likely binding metal components in the ink samples are iron, manganese and copper [25,26]. Commercially available standardized aqueous solutions of iron, manganese and copper
700
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
were analyzed by ICP and specific wavelengths for each metal were determined. Iron has an emission line at 259.94 nm, manganese at 257.61 nm and copper at 224.70 nm. An ink solution (solution A) prepared by dissolving 4.0 mg of ink in 50.0 ml of 15.8 M HNO3 and 50.0 ml of deionized water was analyzed at the specific wavelength for each metal. 2.8.2. ICP spectrophotometer calibration Only iron and manganese aqueous solutions were prepared since no copper was detected in solution A. The amount of iron in the ink sample was far greater than the amount of manganese, therefore iron and manganese solutions with different concentration ranges were prepared. Iron solutions with a concentration of 50.0 ppm, 25.0 ppm and 5.0 ppm were prepared by diluting 5.0 ml of 1.0 103 ppm commercially available iron aqueous solution in 95.0 ml, 195.0 ml and 995.0 ml of 7.9 M HNO3. The 1.0 103 ppm commercially available manganese solution was diluted with deionized water by a thousand fold to prepare a 1.0 ppm stock solution. Manganese aqueous solutions of 500.0 ppb, 250.0 ppb and 50.0 ppb were prepared by diluting 5.0 ml of 1.0 ppm manganese stock solution in 5.0 ml, 15.0 ml and 95.0 ml of 7.9 M HNO3. Calibration curves were obtained from the emission data collected after injection of the metal solutions in the ICP spectrophotometer. 2.8.3. Metal concentration in ink sample Solution A was injected into the ICP spectrophotometer and the emission data were recorded. Metal concentrations in the solution were determined from the calibration curves. 2.8.4. Paper digestion for ICP spectrophotometry analysis Two centimeter-edge squares of white paper were heated in 50.0 ml of 15.8 M HNO3 for 30 min. After cooling to room temperature, 50.0 ml of deionized water were added drop wise. The resulting solution of the digested paper was filtered and analyzed by ICP spectrophotometry. Identical treatment was performed on the paper obtained after sonication of two squares of inked paper at 35 °C for 30 min. 3. Results and discussion 3.1. Evaluation methods In the paper industry, the measurement of whiteness and brightness as well as visual ratings are used to assess the effectiveness of the de-inking process. In this study, quantitative brightness measurements have been performed. Brightness is the reflectance of blue light with an effective wavelength at 457 nm. A change of 0.5 unit in brightness is the average value most people can notice [27]. Square pieces (2 cm 2 cm) of inked paper were sonicated under different conditions of time (5–30 min) and temperature (15– 45 °C). A photograph of typical samples obtained is shown below (Photograph 1).
Photograph 1. Pictures of selected samples. (a) Aqueous phase after sonication of the paper at 35 °C for 5 min; (b) from left to right: paper before procedure, paper after de-inking without sonication at 35 °C for 5 min, paper after de-inking with sonication at 35 °C for 5 min.
Spectrum 1. UV–vis spectrum of the aqueous solutions obtained sonication of inked paper at various temperatures (15–45 °C with increments of 5 after 5 min of 5 °C).
In each case, ink was removed leaving a grey hue on the paper. A yellow color of the aqueous phase indicated that a chemical compound had been transferred to the aqueous phase. UV–vis spectra of the aqueous phase were recorded (Spectrum 1) showing a kmax at 310 nm and at 475 nm. The small peak at 475 nm reflects the yellow color observed in the samples (Photograph 1a). Many black inks similar to the type used in this work contain metals, iron being the main one [25,26]. ICP analysis of the ink described in Section 3.4 below showed the presence of iron and at a lesser extent manganese, whereas the aqueous phase obtained after sonication does not show any trace of these metals. Their absence in the aqueous phase indicates the organic nature of the compound dissolved in the aqueous phase. As a consequence, the peak at 310 nm can be attributed to the transition of electrons in non-bonding or p bonding orbitals to p orbitals. Since the chemical structure of the ink or any of its components is unknown, the exact nature of these transitions cannot be more detailed. The change in absorbance at 310 nm, the strongest peak, was monitored as another assessment method of the de-inking process. 3.2. Influence of the sonication duration The first set of experiments was to test the influence of sonication and its duration on the de-inking process. Experiments were
Fig. 2. Influence of the reaction time on the absorbance at 310 nm: (O) at 15 °C with sonication; (X) at 15 °C without sonication; (d) at 35 °C with sonication; (4) at 35 °C without sonication.
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Fig. 3. Influence of reaction time on the brightness. (s): at 15 °C with sonication; (4): at 15 °C without sonication; (+): at 35 °C without sonication; (h): at 35 °C with sonication.
701
Fig. 4. Influence of the temperature on the absorbance at 310 nm. (-): 5 min with sonication; (h): 5 min without sonication; (X): 30 min with sonication; (}): 30 min without sonication.
performed at 15 °C for 5, 10, 20 and 30 min in the presence of ultrasounds and for 30 min without ultrasounds. UV–vis absorbance and brightness measurements are summarized in Figs. 2 and 3. 3.2.1. Analysis of UV–vis spectroscopic data When the silent reaction is carried out for 30 min, better results were obtained at 35 °C than 15 °C. When comparing the silent reaction with the sonicated one for 30 min at 15 °C, no noticeable difference in the results was observed. However the data obtained from the silent and the sonicated reaction differ greatly when the experiments were conducted at 35 °C for 30 min: A greater absorbance is recorded for the sonicated experiment. Regarding the series of sonicated experiments, a higher temperature (35 °C versus 15 °C) for 30 min or a longer reaction time (30 min versus 5 min) at 15 °C produces a larger absorbance value for the filtrate. Finally absorbance is greater when the experiment is carried out for 5 min at 35 °C rather than 30 min at 35 °C, suggesting a change in the chemical composition of the aqueous phase overtime, which will be discussed later in this paper. 3.2.2. Analysis of the brightness data The silent and the sonicated experiment conducted for 30 min at 15 °C show similar brightness results. When the silent and the sonicated experiments were conducted at 35 °C a greater brightness was observed for the sonicated experiment. Increasing the sonication time from 5 min to 30 min only slightly improves the brightness at a temperature of 15 °C or 35 °C. However, increasing the temperature from 15 °C to 35 °C on the sonicated experiment conducted for 30 min greatly improves the brightness. 3.2.3. Conclusion UV–vis data and brightness results are in fairly good agreement with one another. A slight difference is noted for the sonicated experiments carried out at 35 °C. UV–vis data show a greater absorbance at 5 min, whereas brightness is greater at 30 min. 3.3. Temperature study In the temperature study, experiments were run for 5 min, since sonication for a longer period of time results in paper disintegration, which makes brightness measurements more difficult to record. The influence of the temperature on the de-inking was determined by carrying out the experiments with the temperature ranging from 15 °C to 45 °C. The solution temperature rose during sonication. The average temperature change recorded was +2 °C after 5 min of ultrasound treatment. UV–vis absorbance and
Fig. 5. Influence of the temperature on the brightness. (+): 5 min with sonication, (4): 30 min without sonication, (}): 30 min with sonication.
brightness are plotted in Figs. 4 and 5. Results of the sonication duration study were also included to facilitate comparisons. UV–vis absorbance and brightness data both indicate that the optimum temperature is between 30 °C and 35 °C. This result is consistent with previously reported findings done in water [28]. However, because of instrumentation constraints, individual temperatures between those two points were not tested. It is noteworthy to mention that increasing the reaction time from 5 to 30 min at 35 °C barely improves the brightness, indicating that most of the ink is removed within the first few minutes. However, the same increase in reaction time at 35 °C gives a decreased absorbance. This result suggests that once the ink has been removed from the paper, it undergoes further sonodegradation at 35 °C, a phenomenon not observed at 15 °C. This hypothesis has been tested and the results are reported later in this article. Finally results obtained with silent experiments at 15 °C and 35 °C show that the temperature in the absence of ultrasound, has little influence on the de-inking process. 3.4. Ink sonodegradation An aqueous solution of ‘‘neat” ink similar in concentration as the one obtained after sonication of the paper was prepared and sonicated at 35 °C for 60 min with regular sampling. No noticeable change in absorbance was observed under these conditions. This clearly suggests that the ink is not affected by ultrasound. The aqueous phase resulting from the de-inking process is yellow, the paper after de-inking is grey and the aqueous phase resulting from the mixing of ‘‘neat” ink and water is black. This indicates that the chemical in solution after de-inking is different from the ink dissolved in water. The compound left on the paper has to be
702
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Fig. 6. Change in absorbance at 310 nm of the aqueous solution of desorbed ink with sonication time.
dissociated from the ink allowing for the yellow compound to be degraded by the ultrasound. To test this hypothesis, inked paper was immersed in water at 35 °C and stirred for 30 min. The paper was then filtered off and the aqueous phase was sonicated at 35 °C for 60 min. The absorbance results are shown in Fig. 6. This data shows a decrease in the UV–vis absorbance at 310 nm, consistent with a chemical modification of the yellow compound. After 30 min, no additional change is observed, suggesting that the molecule cannot be further modified under these conditions. Thin layer chromatography was performed on the aqueous solution of the ink, the aqueous solution obtained after stirring the inked paper in water before sonication and on the aqueous solution obtained after sonication of inked paper at 35 °C for 60 min. The ink shows one spot with a 0.89 Rf; the aqueous solution before sonication one spot with a 0.82 Rf; and the aqueous solution after sonication one spot with a 0.71 Rf. This chromatography data support the production of a new compound during the sonication of the aqueous phase, which structure remains unknown. Even though the chemical structure of the ink is unknown, it seems reasonable to speculate that its black color comes from an organometallic compound; the metal being most likely iron [25,26]. This would mean that the organic component of the ink has to be dissociated from the metallic part and dissolved in the solution in order to be degraded. The hypothesis that the metallic part stays absorbed onto the paper and the organic part is transferred to the aqueous solution has been tested. Three samples were analyzed by ICP: An aqueous solution of the ink, an aqueous solution resulting from the digestion of the paper recovered after
Fig. 8. ICP detection of manganese at 257.61 nm. (}): Calibration curve for 0– 0.5 ppm, (X): data obtained from the paper residue.
de-inking and the aqueous phase obtained after sonication of the inked paper. The samples were tested for the presence of iron (259.94 nm), manganese (257.61 nm) and copper (224.70 nm). White paper was tested under the same experimental conditions by ICP and no trace of the tested metals was found. The analysis of the aqueous solution of ink showed no trace of copper. The results obtained for iron and manganese are summarized in Figs. 7 and 8. Iron in the ink sample, with a concentration of 5.0 ± 0.1 ppm, is the most abundant metal among the ones tested. Iron was not detected in the aqueous phase in the concentration range studied and almost all of it was found in the solution of the digested paper with a solution concentration of 4.8 ± 0.1 ppm. When analyzing the digested paper residue and the aqueous phase, levels of manganese were too low to be detected, although the ink sample had a manganese concentration of 6.0 102 ppm. The absence of iron in the aqueous phase validates the hypothesis that the iron stays absorbed on the paper. 4. Conclusion The experiments proved the effectiveness of ultrasound for the removal of ink from paper at any temperature tested (15–45 °C). Brightness data and UV–vis absorbance have been performed, showing that the optimum temperature was between 30 °C and 35 °C. In industry, paper bleaching takes place at higher temperatures in order to remove the varnishes. This study could lead to a process during which the varnish would be first washed away at high temperature and the residual paper transferred to a sonication bath that operates at a lower and optimum temperature. Finally, it was shown that the ink residue in the aqueous phase underwent further degradation, whereas an aqueous solution of ink could not, likely indicating that the dissociation of the metallic and organic part of the ink seems to be a crucial step for its sonodegradation. Acknowledgements
Fig. 7. ICP detection of iron at 259.94 nm. (}): Calibration curve for 0–50 ppm, (X): data obtained from the paper residue.
The authors wish to thank Dr. James Bu from Clark Atlanta University for lending the sonication probe, Dr. Roman Popil at the Institute of Paper Science and Technology in Atlanta for the brightness measurements and Dr. Julie Bartley from the University of West Georgia for technical support with the ICP spectrophotometry analysis. Financial support from the University of West Georgia (Chemistry Department, Faculty Research Grant and Student Research Assistant Program) is gratefully acknowledged. Finally Dr. Sharmistha Basu-Dutt, Dr. Megumi Fujita, Dr. Spencer Slattery,
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Mrs. Jill Stallings and Mr. Steven Broome are sincerely thanked for their friendly support. References [1] R. Seccombe, K. Brackenbury, D. Vandeberg, Appita J. 56 (2003) 184. [2] T. Riippa, Asia Pacific Appita 56th Annual Conference Proceedings, 2002, p. 139. [3] R.C. Thompson, A. Manning, Prog. Pap. Recycl. 14 (2005) 26. [4] R.C. Rogelio, R.Q. Juan, T.S. Jose, Prog. Pap. Recycl. 13 (2004) 29. [5] A. Fricker, R.C. Thompson, A. Manning, Pigm. Resin. T. 36 (2007) 141. [6] A. Fricker, A. Manning, R.C. Thompson, Surf. Coat. Int., Part B: Coat. Trans. 89 (2006) 145. [7] R.C. Thompson, A. Manning, Prog. Pap. Recycl. 14 (2005) 38. [8] W.E. Scott, P. Gerber, Tappi J. 78 (1995) 125. [9] J.C. Norman, N. Sell, M. Danelski, Tappi J. 77 (1994) 151. [10] J. Laine, J. MacLeod, H. Bolker, D. Goring, Pap. Tim. 59 (1977) 235. [11] A. Manning, R.C. Thompson, Surf. Coat. Int., Part B: Coat. Trans. 87 (2004) 21. [12] (a) J. Saari, Paper Technology 46 (2005) 19; (b) J. Saari, P. Qvintus-Leino, S. Isäninen, Proceedings – PulPaper 2004 Conferences, Efficiency, 2004, p. 161.
703
[13] D. Tatsumi, T. Highashihara, S. Kawamura, T. Matsumoto, J. Wood Sci. 46 (2000) 405. [14] L. Turai, C.-H. Teng, Tappi J. 62 (1979) 45. [15] R.C. Thompson, A. Manning, J. Lane, Surf. Coat. Int. 83 (2000) 322. [16] V.A. Shutilov, Fundamental Physics of Ultrasounds, Gordon and Breach Science, New York, 1988. [17] K.S. Suslick, Scientific Am. 260 (1989) 80. [18] D. Peters, Ultrason. Sonochem. 8 (2001) 221. [19] D. Drijvers, H.V. Langenhove, L.N.T. Kim, L. Bray, Ultrason. Sonochem. 6 (1999) 115. [20] I. Hua, M.R. Hoffmann, Environ. Sci. Technol. 30 (1996) 864. [21] B. Niemczewski, Ultrason. Sonochem. 14 (2007) 13. [22] J.P. Lorimer, T.J. Mason, K. Fiddy, Ultrason. 23 (1991) 157. [23] L.C. Hagenson, L.K. Doraiswamy, Chem. Eng. Sci. 53 (1998) 131. [24] S. Wang, B. Huang, Y. Wang, L. Liao, Ultrason. Sonochem. 13 (2005) 506. [25] E. Kedesdy, Papier-Zeitung 33 (1908) 58. [26] J. Martin-Gil, M.C. Ramos-Sanchez, F.J. Martin-Gil, M. Jose-Yacaman, J. Chem. Ed. 83 (2006) 1476. [27] Discussion with Dr. R. Pogil, Institute of Paper Science and Technology in Atlanta. [28] J.L. Luche, Synthetic Organic Sonochemistry, Plenum Press, 1998, p. 55.
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultsonch
Temperature effect on an ultrasound-assisted paper de-inking process Anne C. Gaquere-Parker *, Ayan Ahmed, Temitayo Isola, Bintu Marong, Christopher Shacklady, Phoebe Tchoua Chemistry Department, University of West Georgia, 1601 Maple Street, Carrollton, GA 30118, United States
a r t i c l e
i n f o
Article history: Received 10 May 2007 Received in revised form 29 December 2008 Accepted 6 January 2009 Available online 24 January 2009 Keywords: Sonication Degradation Ink Paper recycling Cavitation Ultrasound
a b s t r a c t The influence of temperature on an ultrasound-assisted ink removal process has been investigated. White copy paper was evenly soaked in black writing ink. After drying the paper to constant weight at 75 °C, ink removal was attempted under varying conditions. Results were assessed by monitoring the UV–vis absorbance of the aqueous phase and measuring the brightness of the paper. Sonication was observed to improve the brightness of the paper in the temperature range of 15–45 °C with an optimum effect at 35 °C. Monitoring UV–vis spectra of the aqueous phase provided evidence that modification of the chemical structure of the ink desorbed from the paper occured. Further investigation under the same conditions showed that ink, when not absorbed on paper, did not undergo the same chemical change. This supports the hypothesis that only the compound released from the ink absorbed onto the paper is sensitive to sonodegradation. One possible explanation is that the metal binding component of the ink stays absorbed on the paper, releasing the organic part, whose chemical structure can be altered by the effect of sonication. Inductively coupled plasma analysis was used to confirm that during the de-inking process of the paper, the metal binding component stays absorbed on the paper and only the organic part is released in the aqueous phase. Published by Elsevier B.V.
1. Introduction In recent years, interest in paper recycling has increased for a number of environmental reasons. A major problem associated with paper recycling is the whitening process, with emphasis being placed on the production of brighter de-inked pulps. Traditionally, oxidizing agents are used to improve the de-inking process and whitening of the paper pulp. Among them are hydrogen peroxide [1], ozone, chlorine dioxide, hypochlorite, dithionite and formamidine sulfinic acid [2]. Ultrasound has been utilized in the paper processing industry at different stages; of which enhancement of pulping, bleaching, depolymerisation of cellulose and treatment of wastepaper have been reported [3–6]. During the process of bleaching, ultrasound has been used to enhance the de-inking of recovered paper when traditional alkaline treatments were not satisfactory [7–12]. These studies report the successful use of ultrasound with temperatures up to 90 °C, in some cases in the presence of additives. In addition, ultrasound has been shown to improve the quality of paper fibers recovered in the recycling process [13]. Other studies report that when varnish is present, optimum results have been attained at temperatures of 65 °C [14]; 70 °C [15] or 80 °C [8], as heat is needed to soften the varnish present. However, excluding one paper that discusses the influence of tem* Corresponding author. Tel.: +1 404 895 0100. E-mail address: [email protected] (A.C. Gaquere-Parker). 1350-4177/$ - see front matter Published by Elsevier B.V. doi:10.1016/j.ultsonch.2009.01.004
perature (20 °C, 40 °C and 60 °C) on the removal of indigo dye from paper in the presence of ultrasound [6], no previous study has investigated the effect of sonication at lower temperatures on unvarnished paper. Furthermore, the articles cited above describe the influence of ultrasounds on the ink particle size but not on its chemical structure. The main advantage of using sonication resides in the fact that experiments can be carried out at close to ambient temperature under atmospheric pressure contrary to other advanced oxidation processes. The propagation of the acoustic waves in a liquid produces the phenomenon of cavitation [16]. The non-linear expansion and sudden collapse of microcavities result in tiny areas of very high temperature (up to 5000 K) and high pressure (500– 1000 atm) [17]. Under these highly localized but harsh conditions, formation of radicals may occur [18,19]. For instance, sonolysis of water generates hydroxyl radicals and hydrogen atoms [20]. These particles may recombine to form hydrogen and hydrogen peroxide or may react with the species present in the solution. Moreover, the high temperature and pressure conditions may contribute to the pyrolysis of volatile and hydrophilic chemical species. The effectiveness of ultrasound-assisted processes is often affected by temperature [21]. With an increasing temperature of the solvent, dissolved gases and solvent vapors enter the cavitation bubbles. This reduces the collapse and the effectiveness of the cavitation process. However, cavitation may be reduced when working at lower temperatures, since a more viscous medium hinders
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
699
acoustic wave propagation. Given such an influence the effect of temperature on the de-inking process of unvarnished inked paper has been investigated. 2. Experimental set up 2.1. Materials The 20 kHz sonication probe with a 750 W output, was manufactured by Sonics and Materials. A Suslick reaction vessel from Sonics (250 ml, three necks, flat bottom glass sonoreactor with a chamber height of 162 mm) was used. The amplitude was set at 70%. A 9 s on and 9 s off duty cycle was used for the experiments. The experiments were carried out in a temperature control water bath from Forma Scientific (model 2067). Xerox acid free paper (Business 4200) and Waterman black writing ink were used. Analytical grade nitric acid 15.8 M HNO3 from Acros Organics was used and diluted with deionized water to 7.9 M HNO3 as needed. Anhydrous ethanol was purchased from Fisher Scientific. Whatman silica gel TLC plates were used. A UVG11 UV lamp (254 nm) from UVP Company was used for TLC analysis. Ultraviolet–visible spectra were recorded using a Varian Carey IE spectrophotometer. The brightness of the paper residues was measured with a Technidyne Brightimeter S-5, with a 45° illumination and reading at 0°. The orifice was taped such that a 3 2 mm area was illuminated, providing a relative input allowing for a comparison within the samples. Except for the brightness measurements, each experiment was duplicated under identical conditions, thus the data are expressed as the mean value with the standard deviation. Photographs were taken with a Canon 20D camera, with an exposure of 1/160 s at f/11 and a lens focal length of 50 mm. The inductively coupled plasma (ICP) data was collected on a Perkin–Elmer Plasma 4000 emission spectrometer. The 1.0 103 ppm metal standards for the ICP analysis were purchased from SPEX CertiPrep Group. 2.2. Ultrasonic power measurement Due to heat loss, the actual acoustic power dissipated in solution, Pdiss, is lower than the maximum irradiated power, Pmax [22]. In order to compare data and ensure reproducibility, Pdiss has to be calculated. Calorimetry allows the calculation of Pdiss, [23,24], with the assumption that all power entering the reaction solution is dissipated as heat, using the following equation:
Pdiss ¼ ðdT=dtÞt¼0 mwater C water
ð1Þ
where mwater is the mass of the water, Cwater the heat capacity of the water and (dT/dt)t = 0 the change of temperature with respect with time. The value for (dT/dt)t = 0 is calculated from the initial slope of the increase of temperature of the water as a function of time when sonicated without any temperature control. Ultrasonic power measurement data were obtained when 100.0 ml of deionized water at 35 °C was sonicated without any temperature control. The temperature change, dT/dt, was recorded every nine seconds for eight minutes and then every four minutes for a total time of twenty five minutes. After thirteen minutes of sonication, the temperature reached a plateau at 81.9 ± 0.1 °C. The results are plotted in Fig. 1. Under these conditions, Pdiss was calculated for the first 27 s to be 83.7 ± 0.1 Watts. 2.3. Paper inking Two centimeter-edge squares of white paper were dried to constant weight overnight in a 75 °C oven. The paper was then soaked
Fig. 1. Ultrasonic power measurement: Temperature increase as a function of time.
evenly in ink for 15 min, dried to constant weight overnight in a 75 °C oven and weighed. The average ink loading was (4.7 ± 0.1) 104 g/cm2. The paper squares were stored in an air tight plastic bag in a refrigerator. 2.4. Sonodegradation of inked paper The sonoreactor containing 100.0 ml of deionized water was placed in a temperature regulated water bath (15–45 °C). When the water in the sonoreactor reached the desired temperature, two squares of inked paper were added. The sonoreactor was then covered with aluminum foil to avoid any interaction between light and the samples. Sonication was immediately started for the required amount of time (5–30 min). At the end of the experiment, the reaction mixture was filtered, diluted by a factor of 10 with deionized water and the filtrate was analyzed by UV–vis spectroscopy. The paper residue was dried at room temperature overnight, photographed and the brightness was measured. 2.5. Ink sonodegradation In the sonoreactor, 4.0 mg of ink was diluted into 100.0 ml of deionized water. This solution was sonicated for 60 min at 35 °C. Sampling was performed every minute for the first 5 min, then every 5 min for the next 25 min and at the 60 min mark. Samples were filtered, diluted by a factor of 10 with deionized water and the solution was analyzed by UV–vis spectroscopy. 2.6. Sonodegradation of the aqueous solution of desorbed ink Two squares of inked paper were stirred at 35 °C for 30 min in 100.0 ml of deionized water. After removal of the paper by filtration, the filtrate was sonicated at 35 °C for 60 min. Samples were diluted with deionized water by a factor of 10. Monitoring by UV–vis spectroscopy was performed every 5 min in 6 consecutive intervals (30 min total) and then at the 60 min mark. 2.7. Thin layer chromatography A solution consisting of 4.0 mg of ink diluted in 100.0 ml of deionized water, the filtrate obtained in part 2.6 before sonication and after sonication, were spotted on a TLC plate. The TLC plate was eluted with ethanol and analyzed at 254 nm with a UV lamp. 2.8. Iron, manganese and copper analysis by ICP spectrophotometry 2.8.1. Metal identification The most likely binding metal components in the ink samples are iron, manganese and copper [25,26]. Commercially available standardized aqueous solutions of iron, manganese and copper
700
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
were analyzed by ICP and specific wavelengths for each metal were determined. Iron has an emission line at 259.94 nm, manganese at 257.61 nm and copper at 224.70 nm. An ink solution (solution A) prepared by dissolving 4.0 mg of ink in 50.0 ml of 15.8 M HNO3 and 50.0 ml of deionized water was analyzed at the specific wavelength for each metal. 2.8.2. ICP spectrophotometer calibration Only iron and manganese aqueous solutions were prepared since no copper was detected in solution A. The amount of iron in the ink sample was far greater than the amount of manganese, therefore iron and manganese solutions with different concentration ranges were prepared. Iron solutions with a concentration of 50.0 ppm, 25.0 ppm and 5.0 ppm were prepared by diluting 5.0 ml of 1.0 103 ppm commercially available iron aqueous solution in 95.0 ml, 195.0 ml and 995.0 ml of 7.9 M HNO3. The 1.0 103 ppm commercially available manganese solution was diluted with deionized water by a thousand fold to prepare a 1.0 ppm stock solution. Manganese aqueous solutions of 500.0 ppb, 250.0 ppb and 50.0 ppb were prepared by diluting 5.0 ml of 1.0 ppm manganese stock solution in 5.0 ml, 15.0 ml and 95.0 ml of 7.9 M HNO3. Calibration curves were obtained from the emission data collected after injection of the metal solutions in the ICP spectrophotometer. 2.8.3. Metal concentration in ink sample Solution A was injected into the ICP spectrophotometer and the emission data were recorded. Metal concentrations in the solution were determined from the calibration curves. 2.8.4. Paper digestion for ICP spectrophotometry analysis Two centimeter-edge squares of white paper were heated in 50.0 ml of 15.8 M HNO3 for 30 min. After cooling to room temperature, 50.0 ml of deionized water were added drop wise. The resulting solution of the digested paper was filtered and analyzed by ICP spectrophotometry. Identical treatment was performed on the paper obtained after sonication of two squares of inked paper at 35 °C for 30 min. 3. Results and discussion 3.1. Evaluation methods In the paper industry, the measurement of whiteness and brightness as well as visual ratings are used to assess the effectiveness of the de-inking process. In this study, quantitative brightness measurements have been performed. Brightness is the reflectance of blue light with an effective wavelength at 457 nm. A change of 0.5 unit in brightness is the average value most people can notice [27]. Square pieces (2 cm 2 cm) of inked paper were sonicated under different conditions of time (5–30 min) and temperature (15– 45 °C). A photograph of typical samples obtained is shown below (Photograph 1).
Photograph 1. Pictures of selected samples. (a) Aqueous phase after sonication of the paper at 35 °C for 5 min; (b) from left to right: paper before procedure, paper after de-inking without sonication at 35 °C for 5 min, paper after de-inking with sonication at 35 °C for 5 min.
Spectrum 1. UV–vis spectrum of the aqueous solutions obtained sonication of inked paper at various temperatures (15–45 °C with increments of 5 after 5 min of 5 °C).
In each case, ink was removed leaving a grey hue on the paper. A yellow color of the aqueous phase indicated that a chemical compound had been transferred to the aqueous phase. UV–vis spectra of the aqueous phase were recorded (Spectrum 1) showing a kmax at 310 nm and at 475 nm. The small peak at 475 nm reflects the yellow color observed in the samples (Photograph 1a). Many black inks similar to the type used in this work contain metals, iron being the main one [25,26]. ICP analysis of the ink described in Section 3.4 below showed the presence of iron and at a lesser extent manganese, whereas the aqueous phase obtained after sonication does not show any trace of these metals. Their absence in the aqueous phase indicates the organic nature of the compound dissolved in the aqueous phase. As a consequence, the peak at 310 nm can be attributed to the transition of electrons in non-bonding or p bonding orbitals to p orbitals. Since the chemical structure of the ink or any of its components is unknown, the exact nature of these transitions cannot be more detailed. The change in absorbance at 310 nm, the strongest peak, was monitored as another assessment method of the de-inking process. 3.2. Influence of the sonication duration The first set of experiments was to test the influence of sonication and its duration on the de-inking process. Experiments were
Fig. 2. Influence of the reaction time on the absorbance at 310 nm: (O) at 15 °C with sonication; (X) at 15 °C without sonication; (d) at 35 °C with sonication; (4) at 35 °C without sonication.
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Fig. 3. Influence of reaction time on the brightness. (s): at 15 °C with sonication; (4): at 15 °C without sonication; (+): at 35 °C without sonication; (h): at 35 °C with sonication.
701
Fig. 4. Influence of the temperature on the absorbance at 310 nm. (-): 5 min with sonication; (h): 5 min without sonication; (X): 30 min with sonication; (}): 30 min without sonication.
performed at 15 °C for 5, 10, 20 and 30 min in the presence of ultrasounds and for 30 min without ultrasounds. UV–vis absorbance and brightness measurements are summarized in Figs. 2 and 3. 3.2.1. Analysis of UV–vis spectroscopic data When the silent reaction is carried out for 30 min, better results were obtained at 35 °C than 15 °C. When comparing the silent reaction with the sonicated one for 30 min at 15 °C, no noticeable difference in the results was observed. However the data obtained from the silent and the sonicated reaction differ greatly when the experiments were conducted at 35 °C for 30 min: A greater absorbance is recorded for the sonicated experiment. Regarding the series of sonicated experiments, a higher temperature (35 °C versus 15 °C) for 30 min or a longer reaction time (30 min versus 5 min) at 15 °C produces a larger absorbance value for the filtrate. Finally absorbance is greater when the experiment is carried out for 5 min at 35 °C rather than 30 min at 35 °C, suggesting a change in the chemical composition of the aqueous phase overtime, which will be discussed later in this paper. 3.2.2. Analysis of the brightness data The silent and the sonicated experiment conducted for 30 min at 15 °C show similar brightness results. When the silent and the sonicated experiments were conducted at 35 °C a greater brightness was observed for the sonicated experiment. Increasing the sonication time from 5 min to 30 min only slightly improves the brightness at a temperature of 15 °C or 35 °C. However, increasing the temperature from 15 °C to 35 °C on the sonicated experiment conducted for 30 min greatly improves the brightness. 3.2.3. Conclusion UV–vis data and brightness results are in fairly good agreement with one another. A slight difference is noted for the sonicated experiments carried out at 35 °C. UV–vis data show a greater absorbance at 5 min, whereas brightness is greater at 30 min. 3.3. Temperature study In the temperature study, experiments were run for 5 min, since sonication for a longer period of time results in paper disintegration, which makes brightness measurements more difficult to record. The influence of the temperature on the de-inking was determined by carrying out the experiments with the temperature ranging from 15 °C to 45 °C. The solution temperature rose during sonication. The average temperature change recorded was +2 °C after 5 min of ultrasound treatment. UV–vis absorbance and
Fig. 5. Influence of the temperature on the brightness. (+): 5 min with sonication, (4): 30 min without sonication, (}): 30 min with sonication.
brightness are plotted in Figs. 4 and 5. Results of the sonication duration study were also included to facilitate comparisons. UV–vis absorbance and brightness data both indicate that the optimum temperature is between 30 °C and 35 °C. This result is consistent with previously reported findings done in water [28]. However, because of instrumentation constraints, individual temperatures between those two points were not tested. It is noteworthy to mention that increasing the reaction time from 5 to 30 min at 35 °C barely improves the brightness, indicating that most of the ink is removed within the first few minutes. However, the same increase in reaction time at 35 °C gives a decreased absorbance. This result suggests that once the ink has been removed from the paper, it undergoes further sonodegradation at 35 °C, a phenomenon not observed at 15 °C. This hypothesis has been tested and the results are reported later in this article. Finally results obtained with silent experiments at 15 °C and 35 °C show that the temperature in the absence of ultrasound, has little influence on the de-inking process. 3.4. Ink sonodegradation An aqueous solution of ‘‘neat” ink similar in concentration as the one obtained after sonication of the paper was prepared and sonicated at 35 °C for 60 min with regular sampling. No noticeable change in absorbance was observed under these conditions. This clearly suggests that the ink is not affected by ultrasound. The aqueous phase resulting from the de-inking process is yellow, the paper after de-inking is grey and the aqueous phase resulting from the mixing of ‘‘neat” ink and water is black. This indicates that the chemical in solution after de-inking is different from the ink dissolved in water. The compound left on the paper has to be
702
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Fig. 6. Change in absorbance at 310 nm of the aqueous solution of desorbed ink with sonication time.
dissociated from the ink allowing for the yellow compound to be degraded by the ultrasound. To test this hypothesis, inked paper was immersed in water at 35 °C and stirred for 30 min. The paper was then filtered off and the aqueous phase was sonicated at 35 °C for 60 min. The absorbance results are shown in Fig. 6. This data shows a decrease in the UV–vis absorbance at 310 nm, consistent with a chemical modification of the yellow compound. After 30 min, no additional change is observed, suggesting that the molecule cannot be further modified under these conditions. Thin layer chromatography was performed on the aqueous solution of the ink, the aqueous solution obtained after stirring the inked paper in water before sonication and on the aqueous solution obtained after sonication of inked paper at 35 °C for 60 min. The ink shows one spot with a 0.89 Rf; the aqueous solution before sonication one spot with a 0.82 Rf; and the aqueous solution after sonication one spot with a 0.71 Rf. This chromatography data support the production of a new compound during the sonication of the aqueous phase, which structure remains unknown. Even though the chemical structure of the ink is unknown, it seems reasonable to speculate that its black color comes from an organometallic compound; the metal being most likely iron [25,26]. This would mean that the organic component of the ink has to be dissociated from the metallic part and dissolved in the solution in order to be degraded. The hypothesis that the metallic part stays absorbed onto the paper and the organic part is transferred to the aqueous solution has been tested. Three samples were analyzed by ICP: An aqueous solution of the ink, an aqueous solution resulting from the digestion of the paper recovered after
Fig. 8. ICP detection of manganese at 257.61 nm. (}): Calibration curve for 0– 0.5 ppm, (X): data obtained from the paper residue.
de-inking and the aqueous phase obtained after sonication of the inked paper. The samples were tested for the presence of iron (259.94 nm), manganese (257.61 nm) and copper (224.70 nm). White paper was tested under the same experimental conditions by ICP and no trace of the tested metals was found. The analysis of the aqueous solution of ink showed no trace of copper. The results obtained for iron and manganese are summarized in Figs. 7 and 8. Iron in the ink sample, with a concentration of 5.0 ± 0.1 ppm, is the most abundant metal among the ones tested. Iron was not detected in the aqueous phase in the concentration range studied and almost all of it was found in the solution of the digested paper with a solution concentration of 4.8 ± 0.1 ppm. When analyzing the digested paper residue and the aqueous phase, levels of manganese were too low to be detected, although the ink sample had a manganese concentration of 6.0 102 ppm. The absence of iron in the aqueous phase validates the hypothesis that the iron stays absorbed on the paper. 4. Conclusion The experiments proved the effectiveness of ultrasound for the removal of ink from paper at any temperature tested (15–45 °C). Brightness data and UV–vis absorbance have been performed, showing that the optimum temperature was between 30 °C and 35 °C. In industry, paper bleaching takes place at higher temperatures in order to remove the varnishes. This study could lead to a process during which the varnish would be first washed away at high temperature and the residual paper transferred to a sonication bath that operates at a lower and optimum temperature. Finally, it was shown that the ink residue in the aqueous phase underwent further degradation, whereas an aqueous solution of ink could not, likely indicating that the dissociation of the metallic and organic part of the ink seems to be a crucial step for its sonodegradation. Acknowledgements
Fig. 7. ICP detection of iron at 259.94 nm. (}): Calibration curve for 0–50 ppm, (X): data obtained from the paper residue.
The authors wish to thank Dr. James Bu from Clark Atlanta University for lending the sonication probe, Dr. Roman Popil at the Institute of Paper Science and Technology in Atlanta for the brightness measurements and Dr. Julie Bartley from the University of West Georgia for technical support with the ICP spectrophotometry analysis. Financial support from the University of West Georgia (Chemistry Department, Faculty Research Grant and Student Research Assistant Program) is gratefully acknowledged. Finally Dr. Sharmistha Basu-Dutt, Dr. Megumi Fujita, Dr. Spencer Slattery,
A.C. Gaquere-Parker et al. / Ultrasonics Sonochemistry 16 (2009) 698–703
Mrs. Jill Stallings and Mr. Steven Broome are sincerely thanked for their friendly support. References [1] R. Seccombe, K. Brackenbury, D. Vandeberg, Appita J. 56 (2003) 184. [2] T. Riippa, Asia Pacific Appita 56th Annual Conference Proceedings, 2002, p. 139. [3] R.C. Thompson, A. Manning, Prog. Pap. Recycl. 14 (2005) 26. [4] R.C. Rogelio, R.Q. Juan, T.S. Jose, Prog. Pap. Recycl. 13 (2004) 29. [5] A. Fricker, R.C. Thompson, A. Manning, Pigm. Resin. T. 36 (2007) 141. [6] A. Fricker, A. Manning, R.C. Thompson, Surf. Coat. Int., Part B: Coat. Trans. 89 (2006) 145. [7] R.C. Thompson, A. Manning, Prog. Pap. Recycl. 14 (2005) 38. [8] W.E. Scott, P. Gerber, Tappi J. 78 (1995) 125. [9] J.C. Norman, N. Sell, M. Danelski, Tappi J. 77 (1994) 151. [10] J. Laine, J. MacLeod, H. Bolker, D. Goring, Pap. Tim. 59 (1977) 235. [11] A. Manning, R.C. Thompson, Surf. Coat. Int., Part B: Coat. Trans. 87 (2004) 21. [12] (a) J. Saari, Paper Technology 46 (2005) 19; (b) J. Saari, P. Qvintus-Leino, S. Isäninen, Proceedings – PulPaper 2004 Conferences, Efficiency, 2004, p. 161.
703
[13] D. Tatsumi, T. Highashihara, S. Kawamura, T. Matsumoto, J. Wood Sci. 46 (2000) 405. [14] L. Turai, C.-H. Teng, Tappi J. 62 (1979) 45. [15] R.C. Thompson, A. Manning, J. Lane, Surf. Coat. Int. 83 (2000) 322. [16] V.A. Shutilov, Fundamental Physics of Ultrasounds, Gordon and Breach Science, New York, 1988. [17] K.S. Suslick, Scientific Am. 260 (1989) 80. [18] D. Peters, Ultrason. Sonochem. 8 (2001) 221. [19] D. Drijvers, H.V. Langenhove, L.N.T. Kim, L. Bray, Ultrason. Sonochem. 6 (1999) 115. [20] I. Hua, M.R. Hoffmann, Environ. Sci. Technol. 30 (1996) 864. [21] B. Niemczewski, Ultrason. Sonochem. 14 (2007) 13. [22] J.P. Lorimer, T.J. Mason, K. Fiddy, Ultrason. 23 (1991) 157. [23] L.C. Hagenson, L.K. Doraiswamy, Chem. Eng. Sci. 53 (1998) 131. [24] S. Wang, B. Huang, Y. Wang, L. Liao, Ultrason. Sonochem. 13 (2005) 506. [25] E. Kedesdy, Papier-Zeitung 33 (1908) 58. [26] J. Martin-Gil, M.C. Ramos-Sanchez, F.J. Martin-Gil, M. Jose-Yacaman, J. Chem. Ed. 83 (2006) 1476. [27] Discussion with Dr. R. Pogil, Institute of Paper Science and Technology in Atlanta. [28] J.L. Luche, Synthetic Organic Sonochemistry, Plenum Press, 1998, p. 55.