Tetramethyl ammonium hydroxide (TMAH) is an anisotropic etchant used in the wet etching process of the semiconductor industry and is hard to degrade by biotreatments when it exists in wastewater. This study evaluated the performance of a system combined with ultraviolet, magnetic catalyst (/) and , denoted as UV/, to TMAH in an aqueous solution. The mineralization efficiency of TMAH under various conditions follows the sequence: UV/ > UV// > /// > / > // > > UV/. The results suggest that UV/ process provides the best condition for the mineralization of TMAH (40 mg/L), resulting in 87.6% mineralization, at 60 min reaction time. Furthermore, the mineralization efficiency of /// was significantly higher than that of , /, and UV/. More than 90% of the magnetic catalyst was recovered and easily redispersed in a solution for reuse.

1. Introduction

The semiconductor industry is an important component of the electronics industry, whose global market yield has already exceeded that of the automobile industry. Anisotropic chemical wet etching is widely used in the semiconductor industry to fabricate microstructures on single crystal silicon wafers [1]. Of all the anisotropic etchants, the inorganic KOH (potassium hydroxide) and organic TMAH (tetramethyl ammonium hydroxide) solutions are the most commonly used [1, 2]. Moreover, TMAH solution has also attracted attention because it is clean room compatible, nontoxic, and easy to handle. It also exhibits excellent selectivity to silicon oxide and silicon nitride masks [3, 4]. It has been estimated that, for an 8 in wafer manufacturing facility with a monthly production of 20,000 wafer units, TMAH is by far the most concentrated chemical in wastewater [5]. Biological processes are the most widely accepted treatment for organic wastewater of both domestic and industrial origins; however, available information on the biodegradability of TMAH is scarce. Chemical oxidation involving various forms of advanced oxidation processes (AOPs) can be employed as preliminary treatment to convert the potentially biorefractory compounds into intermediate products that are more amenable to biodegradation [6].

Ozone () is a chemical agent widely used for the mineralization (i.e., transformation into and inorganic ions) of herbicides and related biorecalcitrant organic contaminants in water [7]. Disadvantages of ozonation alone ( system) for water treatment are the high energy cost required for its generation and very limited mineralization of refractory COD in industrial effluents. Indeed, hydroxyl radical is a less selective and more powerful oxidant than molecular ozone. A common objective of AOPs is to produce a large amount of radicals (especially) to oxidize the organic matter. Alternative procedures involving ozonation catalyzed with [5], UV light [6], catalysts [1], and [8, 9] allow a quicker removal of organic pollutants, because such catalysts improve the oxidizing power of yielding a significant reduction of its economic cost.

The present study assessed the function of UV light, magnetic catalyst (), and on the enhancing to mineralize TMAH. A concentration of total organic carbon (TOC) was chosen as a mineralization index of decomposition of TMAH. The effects of pH value of aqueous solution and ionic strength, , on the mineralization of TMAH were examined in this study.

2. Material and Methods

The batch experiments of mineralization reaction were conducted in a 2.3 L glass flask reactor as illustrated in Figure 1. The UV irradiation source was two 8 W lamps encased in a quartz tube with wavelengths of 254 nm. A UVX Radiometer (UVP Inc., USA) was employed for the determination of UV light intensity. The UV intensity of one 8 W UV lamp at 254 nm is 18.6 . The ozone generator is from Triogen with the capacity of 10 g/hr. The flow rate of ozone air stream was 4 L/min directed into the photoreactor, and the inlet ozone concentration was 26 mg/L. The concentration of ozone was analyzed online by an ozone analyzer (Anseros, Ozamat GM-6000-PRO). was added into the reactor by a syringe pump at constant dosage rate. The pH value of the solution was controlled by the addition of 0.01 N during the whole reaction time. The effects of pH values, ionic strength, and the initial concentration of on mineralization efficiency of the process were examined by varying one factor while keeping the other parameters fixed.

2.1. Catalysts Preparation

Magnetite () was purchased from Sigma-Aldrich (St. Louis, MO, USA) and used without any further purification. A total of 1.08 L of aqueous solution containing 20 g of particles was held in a 2 L beaker at 90°C; the pH was maintained at 9.5 with 0.1 N NaOH, while being stirred by a mechanic stirrer. An appropriate amount of 20 g was dissolved in 100 mL of deionized water; the aqueous solution was then mixed with the aqueous solution containing magnetite () with a mechanic stirrer for 30 min. At last, the magnetic catalysts (i.e., ) were dried at 105°C, after the pH value of the slurry solution was maintained at 8 with 5 N .

2.2. Instrumental Analysis

Tetramethyl ammonium hydroxide, TMAH (, ), supplied by Aldrich (USA) was used as the target compound in this study. Hydrogen peroxide () of 35 wt.% supplied by Shimakyu Co. (Japan) was injected into the reactor at a constant feed rate by syringe pump. All chemicals from several suppliers were reagent grade. The mineralization efficiency of TMAH by this advanced oxidation process was determined by the analytical results of a TOC analyzer (Tekmar, Dohrmann Phoenix 8000). Na2SO3 solution (1.0 g/L) and Spectroquant Picco colorimeter test kit (Merck, Germany) was used to quench and measure residual dissolved ozone in samples for TOC analysis, respectively. This instrument utilizes the UV-persulfate technique to convert organic carbon into carbon dioxide (), analyzed by an infrared analyzer and calibrated with potassium hydrogen phthalate. The magnetic properties and the isoelectric point (IEP) of the catalysts were determined by a vibrating sample magnetometer (Lake Shore, 7407) and a Zetasizer (Nano ZS ZEN 3600), respectively.

3. Results and Discussion

3.1. Surface Characteristics of Magnetic Photocatalyst

Figure 2 showed the surface zeta potential versus pH for the magnetic catalyst , and this same process was repeated two times. The pH of the isoelectric point (IEP) for the magnetic catalyst was 2.8. The IEP for particles and particles was previously determined by other procedures and ranged from 2 to 3 and from 6.5 to 6.8, respectively [10]. Therefore, the results observed that (core) was almost covered by (shell) because the IEP of was close to particles.

The magnetic properties of the and core were measured with a vibrating sample magnetometer (VSM), as shown in Figure 3. The M-H plots showed the change in Ms of the particles, after the incorporation of a shell. The saturation magnetization (Ms) was 39.2 emu  and was observed in . The results indicated that the prepared samples exhibited paramagnetic behaviors at room temperature [11].

3.2. Mineralization Efficiency of TMAH under Various Conditions

To confirm the roles of , UV, and in the mineralization reaction of TMAH, five sets of experiments were performed to compare the mineralization efficiency of TMAH under various conditions as a function of time, which is given as , and the results are shown in Figure 4. Condition (a) denotes a reaction system with UV 254 nm, power density of 37.2 , and of  mol/min (called UV/). The obtained was 38.5% after 60 min of reaction time. We also attempted to mineralize TMAH by alone, and the result indicated that does not possess the ability to mineralize TMAH. As is the case with UV, with , theoretically, the photolysis of one mole leads to two moles according to (1), and the resulting with higher oxidation potential could mineralize TMAH (38.5%) as follows:

Condition (b) represents a reaction system with only , and the resulting percentage of was approximately 49.2% after 60 min of reaction time. There are two mechanisms through which can degrade organic pollutants, namely, (i) direct attack and (ii) indirect attack through the formation of hydroxyl radicals [12]. The observed is better than that for UV/. Under condition (c), using and (denoted here as /), a better (approximately 58.1% at ) was achieved as compared to the values under conditions (a) and (b), thereby confirming the strong oxidation ability of . The oxidation potential of / is based on the fact that the conjugate base of can catalyze ozone into the formation of (Gottschalk et al., 2000). An optimum dose ratio of has often been shown to be in the molar range of 0.5–1 depending on the presence of promoters and scavengers. itself can act as a scavenger as well as an initiator, and therefore determining the optimum dose ratio of is important [7]. With the stoichiometric molar ratio of being 0.5, the overall reaction of catalyzes to produce as shown in (2). In this study, the dosage rates of and are  mol/min and  mol/min, respectively; consider the following

Condition (d), in which UV and (denoted as ) were employed, yielded a value of approximately 87.6% for after 60 min of reaction time, which is far better than conditions (a), (b), and (c). Due to the strong photolysis of ozone in combination with UV radiation ( for ), the decay rate of ozone for decay resulting from the UV is higher than that resulting from by a factor of approximately [7]; this results in the production of more hydroxyl radicals according to (3) and (4), which in turn results in higher mineralization efficiency, as follows:

Condition (e), involving UV, , and (denoted as ), resulted in a poor efficiency (% at ) as compared to condition (d). In the process, the amount of residual in the aqueous solution was small due to the high decay rate of by UV irradiation, and subsequently , when added, will react with instead of with as indicated in (5). Under this condition, behaves as a scavenger of and a small value of results. In this study, we also evaluated the effect of the addition of on the mineralization efficiency of , and the experimental results (data not shown) indicated that even the relatively low dosage of  mol/min causes the molar ratio of to have a value considerably below 0.5. The mineralization efficiency of with the addition of is poorer than that without the addition of . This result led to the conclusion that adding into the process will result in a negative effect on as follows:

Condition (f), involving , , and (denoted as ), resulted in a high efficiency ( at ) as compared to condition (a). To evaluate the ability of catalyst adsorption, the adsorption experiment at  g and initial concentration of  mg/L revealed that the adsorption of TMAH was less than 10% within 60 minutes. The mineralization efficiency of was significantly higher than that of pure , possibly because and could react with the dissolved molecules to generate reactive oxidative species (,,, and ) via free-radical chain reactions. The initiator () could induce the formation of superoxide ions ( from molecules, and formed in the chain reaction was used for the mineralization of organic compounds [13]. Furthermore, the paramagnetic behaviors of the prepared gave rise to the magnetic catalyst , which could be separated more easily through the application of a magnetic field. According to the experimental results, >90% of the magnetic catalyst was recovered and easily redispersed in a solution for reuse.

Condition (h), involving and (denoted as ), resulted in a high efficiency (% at  min) as compared to condition (a). The mineralization efficiency of was significantly higher than that of pure possibly because of the enhancement of reactive oxidative species with the chain reaction of and the dissolved .

As a result, the mineralization efficiency of TMAH under various conditions follows the sequence: . Figure 4 presents the variations of TOC and pH of the TMAH solution under the process as a function of time. As shown in Figure 4, the removal of TOC is close to 87.6% at min, indicating that the process could mineralize TMAH efficiently. Furthermore, the pH value of the reaction solution without a buffer system decreased considerably from 10 to 4.5 during the entire reaction time, thereby revealing that acid intermediates are formed before TMAH is converted into .

3.3. Effect of pH on the Mineralization Efficiency of UV/O3

The direct attack on organic pollutants by molecular ozone (commonly known as ozonolysis) occurs under acidic or neutral conditions. At a high pH value, ozone decomposes to nonselective hydroxyl radicals according to (6), which in turn oxidizes the organic pollutants. Many researches [6, 12] found that an increasing pH accelerates ozone decomposition to generate hydroxyl radicals, which destroy organic compounds more effectively than ozone. Therefore, the pH of aqueous solution is an important factor that determines the efficiency of ozonation since it can alter the degradation pathways as well as kinetics. One has

For the combined oxidation process, , the effect of pH on the mineralization efficiency is more complex. The study [14] found that neither low pH values nor high pH values of the could provide a degradation rate better than that obtained by the simultaneous application of with neutral pH values.

The pH values of the aqueous solution were controlled to stay between 3 and 10 to evaluate the pH effect on the mineralization efficiency of TMAH by the process, as shown in Figure 5. As the pseudo-first-order kinetic hypothesized, Table 1 reveals that the influence of the pH value on the reaction rate is negligible at pH values in the range from 3 to 10. It is clear from (6) that more hydroxyl radicals were produced at high pH values, thus enhancing the mineralization rate of TMAH. However, the production of hydroxyl radicals by the process also proceeds according to (3) and (4), and it was not influenced by the pH value of the aqueous solution. Furthermore, the high ion content of the system may trap the mineralization generated in the solution and, as a result, bicarbonates and carbonates are formed in the alkaline system. Both bicarbonates and carbonates are efficient scavengers of hydroxyl radicals due to their very high reaction rate constants with the hydroxyl radicals ( for bicarbonates and for carbonates). Thus, due to the influence of the increase in hydroxyl radicals and the formation of scavengers, the comprised results causing the pH effect on the mineralization of TMAH by are negligible for pH values of the solution in the range from 3 to 10. Note that a buffer system was not introduced in the later experiments, except for the experiment relating to the pH effect.

3.4. Effect of Chloride Ion and Ionic Strength on the Mineralization Efficiency of UV/O3

The effect of chloride ions, which are frequently present in industrial wastewater, on the mineralization efficiency of TMAH with was evaluated, as shown in Table 2. The mineralization rate of the TMAH solution containing chloride ion by the process could not be expressed by the pseudo-first-order kinetic. So, the mineralization efficiency shown in Table 2 was illustrated by at the reaction time of 60 min. The experimental results indicate that decreased with an increase in the chloride ion concentration. Chloride ions are likely to retard the efficiency of the mineralization of TMAH by competing for the oxidizing hydroxyl radicals and ozone molecules. Chloride ions can be oxidized by ozone as per (7) and (8) [15], and they can be converted into and . Thus, the effective concentration of ozone was decreased by chloride ions, and the oxidation potential of the resulting products, and , was lower than that of ozone. Furthermore, chloride ions may also act as scavenger with regard to the hydroxyl radical as per (9) [15]. One has

Ionic strength may affect the effective concentration of a compound in a solution and this becomes more significant in the presence of polar compounds [16]. As noted, TMAH is a quaternary ammonium compound. Its ammonium ions are surrounded by anions in solutions, resulting in shielding off oxidizing agents such as and hydroxyl radicals. Presumably, it decreases the mineralization efficiency of TMAH by the process. The experimental study [16] examined the effect of ionic strength on the solubility of for various types of inorganic solutions. These researchers concluded that there is no significant effect on the solubility of in sulfate solutions. Additionally, sulfates are not oxidized by ozone molecules and hydroxyl radicals. In the present study, attempts have been made to evaluate the effect of ionic strength on the mineralization of TMAH in sulfate solutions by the process, and the results are shown in Table 3. As can be seen in the table, the variation of the ionic strength neither facilitates nor suppresses the mineralization of TMAH. This indicates that, in the process, the inhibition of the mineralization of polar organic compounds in an aqueous solution is not significant at high ionic strengths.

The reuse experiments were carried out by evaluating the stability of catalyst activity. In this experiment, 0.2 g L−1 of magnetic catalyst () was used at an initial concentration of 40 mg/L of TMAH. After the ozonation process, the magnetic catalyst () was collected by magnetic force. The clear solution was used for analytical determination, and the magnetic catalyst was used directly in the subsequent catalytic ozonation process. This same process was repeated four times and the removal of TMAH in the reuse experiment is shown in Figure 6. The catalytic activity of remained constant and no obvious deactivation (<15%) was observed after being used four times. From the results of the reuse and recovery experiments, the magnetic catalyst is considered to show considerable promise in water treatment use.

4. Conclusions

The major results of applying the advanced oxidation process to mineralize TMAH can be summarized as follows.

The rank of treatment conditions based on the mineralization efficiency of TMAH has the sequence: .

The experimental results of this study suggest that UV irradiation 37.2  UV (254 nm) and flow rate of  mol/min provide the best condition for the mineralization of TMAH (20 mg/L), resulting in 95% mineralization, at 60 min reaction time. Adding into the process will suppress the mineralization efficiency.

The mineralization efficiency of was significantly higher than that of , . More than 90% of the magnetic catalyst was recovered and easily redispersed in a solution for reuse.

The addition of chloride ions in reaction solution will suppress the mineralization efficiency of the process. Ionic strength and variable pH values (from 3 to 10) in reaction solution show no effect on the mineralization efficiency of the process.


This research was supported by the National Science Council, Taiwan, under Grant no. NSC 100-2221-E-562-001-MY3.