Abstract

The experiments were performed in a lab-scale countercurrent spraying reactor to study the NOx removal from simulated gas stream by cyclic scrubbing using NaClO solution. The effects of NaClO concentration, initial solution pH, coexisting gases (5% CO2 and 13% O2), NOx concentration, SO2 concentration, and absorbent temperature on NOx removal efficiency were investigated in regard to marine exhaust gas. When NaClO concentration was higher than 0.05 M and initial solution pH was below 8, NOx removal efficiency was relatively stable and it was higher than 60%. The coexisting CO2 (5%) had little effect on NOx removal efficiency, but the outlet CO2 concentration decreased slowly with the initial pH increasing from 6 to 8. A complete removal of SO2 and NO could be achieved simultaneously at 293 K, initial pH of 6, and NaClO concentration of 0.05 M, while the outlet NO2 concentration increased slightly with the increase of inlet SO2 concentration. NOx removal efficiency increased slightly with the increase of absorbent temperature. The relevant reaction mechanisms for the oxidation and absorption of NO with NaClO were also discussed. The results indicated that it was of great potential for NOx removal from marine exhaust gas by wet scrubbing using NaClO solution.

1. Introduction

The exhaust gas emitted from marine diesel engines contains a large number of atmospheric pollutants, such as sulphur oxides (SOx), nitrogen oxides (NOx), and particulate matters (PMs), which has caused serious damage to the ecological environment [1]. A direct way to reduce SOx emission is to adopt low sulphur fuel oil (LSFO), but this will increase the transportation cost greatly. An alternative way is to install an exhaust gas cleaning system (EGCS) on board to achieve the abatement of SOx emission equivalently. But it is very difficult for SOx scrubbers to remove NOx effectively at the same time. Generally, the methods for reducing marine NOx emission can be divided into combustion control and postcombustion control techniques. The combustion control techniques include exhaust gas recirculation (EGR) [2], fuel-water emulsion (FWE) [3], and direct water injection (DWI) [4]. They aim at reducing the formation of NOx during the combustion process, but it will result in the decrease of total heat efficiency. At present, the typical postcombustion control approach for ocean-going ships is selective catalytic reduction (SCR), which can remove NOx with an efficiency of 80–95%. As a relatively mature denitrification technology, SCR has been extensively applied in power plants and mobile vehicles. But there are still some problems that limit the application of SCR in marine industry. It requires large additional space and high investment/operation cost. The ash content or sulfate salts may result in the inactivation of SCR catalyst. A complex control system is also required to reduce the ammonia slip [57]. Therefore, ship owners around the world are still seeking a better way to reduce NOx emission. It is of great importance to develop an efficient denitrification technology that can cater for the needs of ocean-going ships.

During the past decades, wet scrubbing technique becomes more and more attractive for the simultaneous removal of SOx, NOx, and other pollutants. As to NOx removal, a preoxidation process is required to oxidize NO into NO2 or other nitrogen oxides of higher values. That is due to the low solubility (1.93 × 10−3 mol·L−1·atm−1 at 25°C) of NO in water. The nonthermal plasma (NTP) can be used to oxidize NO effectively, and then an absorption process is followed for NO2 absorption. However, this method requires a high energy consumption [8, 9]. Similarly, ozone oxidation method encounters the same limitation [1012]. Another feasible approach is to add the oxidants into the absorbent. For this purpose, various oxidants such as hydrogen peroxide (H2O2) [1315], potassium permanganate (KMnO4) [16], sodium chlorite (NaClO2) [17, 18], calcium hypochlorite (Ca(ClO)2) [19], and sodium hypochlorite (NaClO) [2024] have been investigated to enhance the NO removal efficiency of the scrubbing solution. Compared with other oxidants, NaClO has some distinct advantages, such as low cost, easy availability, strong oxidative ability, easy to storage, good stability, and low toxicity. So it is attractive for researchers to investigate the simultaneous removal of NO and SO2 by wet scrubbing using NaClO solution [2528]. A previous study implied that NO could be effectively removed by wet scrubbing using NaClO solution in a cyclic mode, and the utilization of NaClO oxidant in solution was extremely high [29]. But the effect of the operating parameters (such as NaClO concentration, solution pH, absorbent temperature, NO, and SO2 concentrations) on NO removal efficiency by cyclic scrubbing using NaClO solution had not been investigated. In this paper, marine exhaust gas was chosen as the treatment objective. Though the components of marine exhaust gas might vary with the engine load and fuel type, the typical compositions of marine exhaust gas contained ~13% O2 and ~5% CO2. The effect of coexisting gases on NOx removal efficiency had also been studied preliminarily, and the relevant reaction mechanism was discussed.

2. Experimental Section

2.1. Materials

As mentioned in [11], exhaust gas scrubbers are designed in accordance with the maximum power and exhaust amount of target engine for the practical application in marine industry. The exhaust gas compositions are also measured at maximum load of engine. In this study, the concern is mainly focused on the NOx removal efficiency, and the concentrations of gas components of a typical marine slow-speed 2-stroke diesel engine are considered as the reference. The fuel type is heavy fuel oil with 2.4% sulphur content. Thus, the concentrations of O2, CO2, SO2, and NOx are ~13%, ~5%, ~600 ppm, and ~1000 ppm, respectively. Here the simulated exhaust gas was prepared by blending various kinds of synthetic gases. Five kinds of gases, N2 (99.999%), O2 (99.995%), CO2 (99.999%), NO (10.04% NO with N2 as the balance gas), and SO2 (10.1% SO2 with N2 as the balance gas) (Dalian Date Gas Co., Ltd), were used to make the simulated flue gas. As NO accounted for more than 95% of NOx in marine exhaust gas, only NO span gas was used to prepare the NOx components in simulated gas stream.

The NaClO solutions were prepared using the commercial NaClO solution (5% available chlorine, Shanghai Aladdin Bio-Chem Technology Co., Ltd) and the deionized water. The volume of NaClO solution was 1 L for each test. The pH value was adjusted by adding 0.5 M H2SO4 solution and determinated using an acidimeter (Mettler-Toledo International Trading Co., Ltd).

2.2. Experimental Apparatus

A schematic diagram of the experimental apparatus is shown in Figure 1. It consists of a gas distributing system, a gas-liquid countercurrent scrubbing reactor, and a gas analyzer.

N2, O2, CO2, NO, and SO2 were provided from separate air bottles and metered through mass flow controllers (MFC, Beijing Sevenstar Electronics Co., Ltd). The simulated flue gas was obtained from the feed gases by blending with an on-line mixer, and then it was introduced into the spraying column. The height and inner diameter of the column were 25 cm and 5 cm, respectively. A spraying nozzle (B1/4TT-SS+TG-SS0.4, Spraying System Co., Ltd) was located at the top of the column. The size of liquid droplet sprayed from the nozzle was in the range of 80–100 μm. The flow rate of the simulated flue gas was fixed at 1.25 L/min. The calculated residence time of flue gas in the column was ~23 s.

When the initial gas concentrations were adjusted to the required level, the simulated flue gas was introduced into the scrubber from the bottom of the column. The NaClO absorbent was sprayed from top to bottom. A peristaltic pump was used to pump the scrubbing solution cyclically. The flow rate of the scrubbing solution was ~0.27 L/min. Each run of the test was 20 min. The outlet concentrations of flue gas were measured at an interval of 10 s. The solution temperatures were adjusted by the constant water bath (F34-MA, Julabo Labortechnik GmbH) and measured with a mercury thermometer. A MRU MGA-5 gas analyzer was used to determinate the gas concentrations of O2, CO2, NO, and NO2 in flue gas.

2.3. Data Process

The gas concentrations measured by the bypass are taken as the inlet concentrations. The average concentrations within 20 min measured by the gas outlet are considered as the outlet concentrations. The removal efficiencies of NOx and SO2 are calculated by the following equation:in which is the removal efficiency and and are the inlet and outlet concentrations, respectively. Here NOx refers to the mixture of NO and NO2 in flue gas.

3. Results and Discussion

3.1. Effect of Initial pH and NaClO Concentration

The active components in NaClO solution were available chlorine, which mainly includes HClO, ClO-, and Cl2. The compositions of NaClO solution depended greatly on the solution pH. Firstly, it was necessary to investigate the effect of initial solution pH on the NOx removal efficiency. Figure 2 showed that NOx removal efficiency was very low when the solution pH was higher than 10. That was because little HClO existed in the solution when solution pH was higher than 10, but HClO was considered to be the main component in NaClO solution to oxidize NO [22]. It demonstrated that NaClO solution without optimizing the initial pH was not suitable for removing NOx. With the decrease of initial pH from 10 to 8, NOx removal efficiency increased quickly, which was due to the increase of fractional composition of HClO in solution. HClO oxidized NO into NO2, N2O3, N2O4, and in chain reactions of (2)–(9) [30, 31]. The possible reaction pathways were summarized as shown in Figure 3. As shown in Figure 2, with the initial pH decreasing from 8 to 7, a slight drop of NOx removal efficiency appeared. On further reducing the initial pH, the changes of NOx removal efficiency depended on NaClO concentration. When initial NaClO concentration was below 0.05 M, NOx removal efficiency continued to decrease with initial pH decreasing from 7 to 6. This was because the oxidation power of was stronger than that of HClO in neutral or weak acidic medium. As shown in Figure 4, HClO concentration increased while concentration decreased with solution pH decreasing from 7 to 6. However, when NaClO concentration was higher than 0.05 M, the oxidation power of active chlorine was high enough to oxidize NO effectively at pH 6 and 7 [32]. At the moment, the effect of the change of active chlorine concentration was not obvious enough. The result indicated that a high NaClO concentration might be appropriate for practical application for it was easy to obtain a high and stable NOx removal efficiency in a relatively wide range of solution pH.

Figure 5 showed the effect of initial solution pH on NOx concentration in flue gas. When NaClO concentration was higher than 0.05 M and solution pH was in the range of 4–8, outlet NO concentration was lower than 124 ppm. It suggested that the majority of NO had been oxidized into NO2 by NaClO. When NaClO concentration was 0.05 M, outlet NO concentration decreased sharply to 35 ppm with initial pH decreasing from 10 to 8. At the same time, NO2 concentration in outlet gas increased from 14 ppm to 353 ppm. On further decreasing the initial pH down to 6, a complete removal of NO could be achieved with 0.05 M NaClO while outlet NO2 concentration reached about 400 ppm. With the initial pH decreasing from 6 to 4, NO in outlet gas increased to 124 ppm while NO2 in outlet gas decreased to 270 ppm. NOx removal efficiency changed little in the range of pH 4–8. The results implied that, to a certain extent, NOx removal efficiency for wet scrubbing using NaClO solution was mainly limited to the absorption of NO2 rather than the oxidation of NO.

3.2. Effect of Coexisting CO2

During wet scrubbing process, CO2 in flue gas would react with absorbent thus influencing the oxidation and absorption of targeted pollutants. As an acidic oxide, CO2 was sensitive to the solution pH. The effect of coexisting CO2 on NOx removal efficiency was investigated, and results were shown in Figure 6. It can be seen that, with initial solution pH increasing from 4 to 8, NOx removal efficiency was relatively stable. A complete removal of NO had been achieved and outlet NO2 concentration was ~350 ppm. The coexistence of CO2 had not affected the NOx removal efficiency so much. But the average CO2 concentration in outlet gas changed obviously with initial solution pH. When initial solution pH was below 6, the average CO2 concentration kept stable at 5%. However, it began to decrease quickly with initial pH increasing from 6 to 8.

The change of outlet CO2 concentration during the cyclic scrubbing process was shown in Figure 7. When initial pH was in the range of 4–6, CO2 concentration decreased to 4.4–4.5% at the start of the scrubbing process. Then it recovered to the initial level due to the hydrolysis equilibrium between CO2 and absorbent solution. The hydrolysis reactions of CO2 were described in (10) and (11). Furthermore, a certain amount of might be produced during the hydrolysis process, which would affect the chlorine hydrolysis equilibrium reactions as shown in (12) and (13). Since and had buffering ability to some extent, the hydrolysis of CO2 would not influence the solution pH obviously when initial pH was in the range of 4–6. But with initial pH increasing from 6 to 8, the absorption of CO2 might reduce the solution alkalinity, resulting in the decrease of the solution pH. Thus it was necessary to keep the solution pH at 6 in order to reduce the consumption of solution alkalinity when NaClO solution was adopted to remove NOx from flue gas. When initial pH was in the range of 7-8, CO2 concentration decreased largely at the very beginning of the cyclic scrubbing process. It suggested that much more CO2 had been absorbed by the scrubbing solution due to the weak alkaline medium. With the proceeding of the scrubbing process, CO2 concentration began to increase slowly. Although no evidence showed that CO2 would react with NaClO directly, the hydrolyzation and absorption of CO2 would increase the consumption of solution alkalinity obviously. It meant that extra alkaline solution was required to maintain the solution pH during cyclic scrubbing process, which would largely increase the operation complexity and cost at the same time. Thus initial pH of 6 might be appropriate for practical application.

Figure 8 showed the change of NaClO solution pH after scrubbing for 20 min. With the proceeding of cyclic scrubbing process, the solution pH would decrease slowly due to the absorption of NOx and CO2. It was worth noting that the solution pH increased from 4 to 4.63 for NaClO solution with initial pH 4. That was because active chlorine species in NaClO solution were mainly HClO and Cl2 at pH 4, as shown in Figure 4. During the scrubbing process, Cl2 would be purged out easily from the solution and reacted with NO effectively in gas phase as (14) and (15) [25]. The decrease of Cl2 in NaClO solution would lead to a left shift of the hydrolysis equilibrium of active chlorine as shown in (16), resulting in the increase of solution pH for NaClO solution with initial pH 4.It was possible that excessive Cl2 escaped from NaClO solution might result in secondary pollution. In addition, acidic mist formed in the flue gas might result in the corrosion of operation system. In view of this, NaClO solution with pH 6 was appropriate for NOx removal in cyclic scrubbing mode.

3.3. Effect of Coexisting O2

For marine diesel engines, there was typical ~13% O2 in exhaust gas. O2 could partially oxidize NO under certain conditions, so it was necessary to investigate the effect of coexisting O2 on NOx removal efficiency. In the experiments, only NO standard gas was used to prepare NOx in the initial simulated flue gas. The introduction of O2 has oxidized a little of NO into NO2 in the gas mixer. As shown in Figure 9, the initial NO2 concentration in inlet gas increased almost linearly with the increase of NO concentration. When inlet NOx concentration was 1000 ppm, the initial NO2 concentration reached 112 ppm.With the NOx increasing from 250 to 700 ppm, the outlet NO concentration decreased gradually to 0. When inlet NOx concentration was higher than 700 ppm, NO could be removed completely. As expected, the outlet NO2 concentration increased almost linearly with the increase of inlet NOx concentration. It indicated that NOx removal efficiency depended to a great extent on the absorption of NO2 during the scrubbing process.

Figure 10 presented the change of NOx removal efficiency and outlet O2 concentration with the initial NOx concentration. Since O2 concentration was relatively high, it changed little during the scrubbing process. With the increase of NOx concentration from 250 ppm to 700 ppm, NOx removal efficiency increased from 47% to 74%. It could be ascribed to the improvement of the mass transfer at the gas-liquid interface. When O2 was present in flue gas, NOx removal efficiency was a little higher than that without O2 in flue gas. The existence of O2 improved the NOx removal efficiency in way of partly oxidizing NO in initial flue gas.

3.4. Effect of NOx Concentration

The effects of NOx concentration on NOx removal are investigated, and the results are shown in Figure 11. It can be seen that, with initial NO concentration increasing from 250 ppm to 500 ppm, NOx removal efficiency increased from 43% to ~63%. NO could be removed completely when inlet NO concentration was higher than 500 ppm. Outlet NO2 concentration increased quickly with the increase of inlet NO concentration, resulting in a relatively stable NOx removal efficiency.

3.5. Simultaneous Removal of NOx and SO2

Marine diesel engines usually burn heavy fuel oil (HFO) in order to save the operating cost. At present, the mass concentration of sulphur (S) content in HFO was about 2.5% at average. The combustion of HFO fuel would produce a large amount of SO2 in exhaust gas. Experiments were conducted to investigate the effect of coexisting SO2 on NOx removal efficiency, and the results are shown in Figures 12 and 13.

Figure 12 depicted the change of NOx removal efficiency and NOx concentration with inlet NOx concentration. A complete removal of SO2 and NO had been achieved simultaneously. Due to its high solubility, SO2 could be absorbed effectively by scrubbing solution. Then SO2 was removed through the hydrolysis reaction as (18) and (19). The hydrolysis products of would be oxidized by active chlorine species into quickly. Thus the removal of SO2 would consume the solution alkalinity and oxidants at the same time. The result demonstrated that NaClO solution could be used to remove NO and SO2 simultaneously from marine exhaust gas. However, outlet NO2 concentration increased gradually with the increase of inlet NOx concentration. Figure 13 exhibited the effect of SO2 concentration on NOx removal efficiency and NOx concentration in flue gas. The result showed that SO2 together with NO was completely removed. But with the increase of inlet SO2 concentration, NOx removal efficiency exhibited a slightly downside trend. This phenomenon could be ascribed to the competition reactions between NOx and SO2. Some oxidants in the absorbent would be consumed through the hydrolyzation and absorption of SO2. With inlet SO2 concentration increasing from 200 ppm to 600 ppm, the pH of the scrubbed NaClO solution decreased from 5.25 to 4.52. The decrease of solution pH was negative for the absorption of NO2 through (5) and (6).

3.6. Effect of Absorbent Temperature

The reaction temperature could greatly influence the diffusion, dissolution, and reaction characteristics of molecules or ions in liquid phase [19]. The effect of absorbent temperature on NOx removal was investigated, and the results were shown in Figure 14. Generally, the absorbent temperature was kept below 333 K in industrial application in order to reduce the supply of make-up water. So the absorbent temperature was chosen to be in the range of 293–333 K in the experiments. Figure 14 showed that NOx removal efficiency increased gradually with the increase of absorbent temperature. The highest NOx removal efficiency of 69% was achieved at 333 K. As NO could be removed by 100% easily, the increase of NOx removal efficiency resulted from the improvement of NO2 absorption. This could be explained by the Arrhenius equation of the reaction rate constant. The increase of temperature could enhance the mass transfer of NO2 to the absorbent, thus improving the absorption of NO2 accordingly.

4. Conclusion

NOx removal by wet scrubbing using NaClO solution was studied based on a spraying reactor in a cyclic mode. The results showed that when NaClO concentration was higher than 0.05 M and initial solution pH was below 8, NOx removal efficiency was relatively stable, which was higher than 60%. The coexisting CO2 (5%) had little effect on NOx removal efficiency, but the solution pH began to decrease with the proceeding of cyclic scrubbing process when initial pH was higher than 6. The coexisting O2 (13%) could oxidize NO partially in the gas mixer, resulting in a little improvement of NOx removal efficiency. When initial NOx concentration was higher than 500 ppm, NO could be removed completely while outlet NO2 concentration increased almost linearly with the increase of initial NOx concentration, resulting in a relative stable NOx removal efficiency. A complete removal of SO2 and NO could be achieved simultaneously at 293 K, initial NaClO solution pH 6, and 0.05 M NaClO concentration. With the increase of inlet SO2 concentration, the outlet NO2 concentration increased slightly due to the decrease of solution pH. NOx removal efficiency increased with the increase of absorbent temperature. The relevant reaction mechanism for the removal of NOx and SO2 by wet scrubbing using NaClO solution was also discussed. The results demonstrated that it was of great potential for NaClO solution to remove NOx from marine exhaust gas.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

This study has been financially supported by the National Natural Science Foundation of China (Grant nos. 51402033 and 51479020), the Science and Technology Plan Project of China’s Ministry of Transport (Grant no. 2015328225150), the Doctoral Scientific Research Staring Foundation of Liaoning Province (Grant no. 201601073), and the Fundamental Research Funds for the Central Universities (Grant nos. 3132017003 and 3132016326).