determination of catalytic reaction with copper during

DETERMINATION OF CATALYTIC REACTION WITH COPPER DURING DIOXIN FORMATION Takaoka Masaki1, Fujimori Takashi1, Oshita Kazuyuki1 , Shiota Kenji4, Tanida Hajime3, Morisawa Shinsuke1 1 Department of Urban & Environmental Engineering, Kyoto University, Kyotodaigakukatsura, Nishikyo-ku, Kyoto, 615-8540, Japan; 2 Japan Synchrotron Radiation Institute (JASRI), 1-1-1 Kouto, Mikazuki-cho, Sayo-gun, Hyogo, 679-5198, Japan; Introduction Understanding the formation mechanisms of dioxins is necessary in order to reduce dioxin emissions from municipal solid waste incinerators. Many researchers have indicated that de novo synthesis is the key mechanism in the formation of dioxins during combustion. 1-3 The de novo reaction is described as the breakdown reaction of a carbon matrix in fly ash. Copper is thought to be the key catalyst of de novo reactions in fly ash3-5, but there is limited direct evidence to support this theory. In a previous study, we investigated the behavior of copper at temperatures suitable for de novo synthesis by performing in situ X-ray absorption near edge structure (XANES) experiments with real fly ash and with various types of model fly ash.6,7 On heating the fly ash (using CuCl2·2H2O), dynamic changes were observed in the copper, in both the real and model fly ash. The reduction of cupric compounds began at below 200°C, which corresponds to the lowest temperature in the range within which de novo synthesis has been observed via experiments and incinerators.8 The cupric compounds reoxidized to cuprous compounds at 400°C. However, increasing the temperature from room temperature to 400°C required 1 h, because the heating device, a conventional resistance type electric furnace, was unable to raise the temperature over a shorter period of time. Thus, we have been able to observe only the limiting step of the change in copper during dioxin formation. Temperature clearly acts as a trigger for the formation of dioxins. However, the heating method used in the previous study is unsatisfactory for creating temperature changes, as it cannot quickly stop and restart. If we were able to accomplish a controlled temperature increase over a shorter time period, we would be able to confirm the rapid and reversible reactions between cuprous compounds and cupric compounds and, in turn, to show that the de novo synthesis of dioxins involves a catalytic reaction with copper. Therefore, in this study, we prepared a new in situ cell using a high-speed infrared heating system and conducted in situ XANES experiments to investigate the dynamic changes in copper in a model fly ash during heating. Materials and Methods We prepared the model fly ash by mixing CuCl2·2H2O, activated carbon (AC), and boron nitride. The fly ash contained 1.86% Cu, 2.0% Cl, and 5.0% AC. After grinding the fly ash using a mortar for 10 min, the fly ash was pressed into a disk (13 mm in diameter). The direct speciation of copper in the fly ash was examined using XANES with a new in situ cell, shown in Fig. 1. The high-speed infrared heating system (GA298; Thermo Riko Co. Ltd.) is composed of a roughly spherical infrared lamp, a spheroid reflection mirror, and an infrared guide rod made of transparent quartz.9 Infrared rays from the infrared lamp are condensed by the spheroid mirror and transmit heat by total reflection, and the process is repeated within the transparent quartz rod. With this system, a sample can be heated to 400°C in 1 min. While a disk of model fly ash was heated under 10% O2/90% N2 in the in situ cell, XANES was performed using beamline BL01B1 in SPring-8. The heating profile is shown in Fig. 1. Heating and cooling were performed over a temperature range centered on 300°C, which is suitable for the de novo synthesis of dioxins4,10, and were repeated over a short period of time. The Cu-K edge XANES spectra were collected in fluorescence mode using a Lytle detector with a Si(111) monochromator. A time-resolved Quick XAFS

Organohalogen Compounds, Volume 70 (2008) page 002260

(QXAFS) mode was applied to rapidly measure XAFS spectra. QXAFS produces data of a superior quality at 1-min intervals. The data were analyzed using REX 2000 ver. 2.5 (Rigaku Co. Ltd.).

Sample holder

I0

Inlet gas

I0

IF Heat-gun

Sample holder

Outlet gas

Sample disk

IF Heat-gun

Figure 1 A high speed infrared heating system developed for the in situ observation of the dynamic change of copper during de novo reaction Time (min) 100

OFF

ON

80

OFF OFF 60

ON 400

ON 300

ON

40

200

OFF ON

20

OFF ON

0

Temperature (degree C)

Figure 2 Time and temperature profile for the heating and cooling procedure.

Results and Discussion The changes in the Cu-K XANES spectra on heating the model fly ash are shown in Fig. 3(a). The spectrum at each constant temperature, except room temperature, is displayed as a heavy black line. The broken lines indicate spectra at increasing temperatures. The spectra displayed as gray lines were recorded during cooling from a certain temperature. At room temperature, the copper was present in the form CuCl2·2H2O. With the first increase in temperature to 300°C, the spectrum changed dramatically; however, the spectrum changed only slightly afterward. Owing to the subtle changes, it was difficult to identify the chemical form of copper using linear combination fitting of reference spectra.6,7 Therefore, the chemical state of the copper was evaluated using the valence. It is well known that the X-ray absorption edge position is sensitive to the valence of copper11,12. Thus, the valence can be determined by analyzing the shift of the edge position. In this study, the absorption edge position was taken as the energy point of the first maximum in the derivative of the Cu-K XANES spectrum. The changes in the derivative spectra are shown in Fig. 3(b). The edge positions of the spectra in this experiment ranged from 8980 to 8982.5 eV, except for that of the spectrum before heating. The shift of the edge position appears to correspond to the change in temperature. The changes in the shift of the edge position on heating the model fly ash are shown in Fig. 4. It appears that the chemical state of copper in the model fly ash was changed by instant heating and cooling, and this change was repeated after the second heating and cooling. The edge position indicates that the copper in the model fly ash was reduced by heating and was oxidized by


cooling. The reduction and oxidation are considered to be dependent on carbon breakdown in the model fly ash and oxygen in the atmosphere. The edge positions were 8981.5 eV at 200°C, 8981.0 eV at 300°C, and 8980.5 eV at 400°C. The edge of elemental copper (Cu) was taken as 8977.9 eV. The edge shift, calculated as the relative edge difference from elemental copper, was between 1.5 and 6.0 eV (average, 3.75 eV),12 which indicates that most of the copper was present as monovalent Cu(I) compounds. When the time period between heating and cooling was within 10 min, the edge position at 300°C was replicated instead of that related to the temperature history, which suggests that the reaction was repeated because the reactant had not been consumed completely. These results indicate that this redox reaction occurred and that it was reversible.

300 ºC

300 ºC 400 ºC

200 ºC

300 ºC

dA /dE (eV-1)

Normalized absorption (-)

300 ºC

300 ºC

400 ºC 300 ºC 200 ºC 8975

8985

8995

Photon energy (eV)

(a) XANES

8970

8975

8980

8985

8990

Photon energy (eV)

(b) derivative of XANES

Figure 3 The changes in (a) the Cu-K XANES spectra and (b) the derivative spectra on heating the model fly ash. The chemical state of copper was not entirely dependent on temperature. When the model fly ash was cooled instantly to room temperature, the chemical form of copper did not return to CuCl2·2H2O. In addition, when the temperature cooled to room temperature after an incremental increase from 200°C to 400°C via 300°C, the edge did not increase to 8982.0 eV (the position after the first and second instant cooling) but remained at 8981.5 eV (the position at 200°C). This phenomenon may be related to a severe chemical alteration at 400°C as well as to the experimental method, which involved a decrease in reactant over time. In the post-combustion zone of a municipal solid waste incinerator, the temperature of the fly ash on the wall of the gas cooling tower is not kept constant but changes within a certain range as the fly ash is collected by a dust collector. Thus, given the temperature sensitivity of the chemical state of copper, the redox reaction at around 300°C will form a catalytic cycle in the fly ash of the incinerator.


Many researchers regard copper chloride as an influential catalyst;3–5 however, copper chloride does not necessarily act as a catalyst but may be a chlorine source13. Direct chlorination by CuCl2·2H2O is one route of the de novo synthesis of dioxin, but direct chlorination is not a reversible reaction. In general, it has been difficult to distinguish direct chlorination from other reactions involved in de novo synthesis in model fly ash using CuCl2·2H2O. In the present study, we found evidence that de novo synthesis with copper included a reversible reaction, which suggests that the copper in fly ash is acting as a catalyst as opposed to a chlorine source. Accordingly, we conclude that the de novo synthesis of dioxins in fly ash involves a catalytic reaction with copper. Acknowledgements The synchrotron radiation experiments were performed at the SPring-8 with approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No.2007B1540-NX-np).

100

80

References

400 ºC Time (min)

1. Huang H, Buekens, Chemosphere 1995; 31:4099 2. Stanmore B.R, Combustion and Flame 2004; 136:398 3. Addink R, Olie K, Environ. Sci. Technol.1995; 29: 1425. 4. Stieglitz L, Zwick G, Beck J, Roth W, Vogg H, Chemosphere 1989; 18: 1219. 5. Addink R, Altwicker ER, Environ. Eng. Sci.1998;15: 19. 6. Takaoka M, Shiono A, Nishimura K, Yamamoto T, Uruga T, Takeda N, Tanaka T, Oshita K, Matsumoto T, Harada H, Environ. Sci. Technol 2005; 39: 5878. 7. Takaoka M, Shiono A, Yamamoto T, Uruga T, Takeda N, Tanaka T, Oshita K, Matsumoto T, Harada H, Chemosphere 2008; in press. 8. Addink R, Drijver J, Olie K, Chemosphere 1991; 23:1205. 9. Awaji N, Sugita Y, Horii Y, Takahashi I,Applied Physics Letters 1999; 74: 2669. 10. Vogg H, Stieglitz L, Chemosphere 1986;15: 1373. 11. Lamberti C, Prestipino C, Bonino F, Capello L, Bordiga S, Spoto G, Zecchina A, Diaz Moreno S, Cremaschi B, Garilli M, Marsella A, Carmello D, Vidotto S, Leofanti G, Angew. Chem. Int. Ed. 2002, 41, 2341. 12.Fujimori T, Takaoka M, Kato K, Oshita K, Takeda N, X-ray Spectrometry 2008;37: 210. 13.Jay K, Stieglitz L, Chemosphere 1991;22: 987.

300 ºC

60

200 ºC 40

20

0 8980

8981

8982

Cu-K XANES edge (eV)

Figure 4 The change in the edge position on heating the model fly ash
