Improvement of the Standard Chimney Electrostatic Precipitator by Dividing the Flue Gas Stream into a Larger Number of Pipes

Abstract

Combustion of biomass-based solid fuels is becoming increasingly popular, especially in small heat sources. A major problem in the combustion of biomass is the increased production of emissions and especially the solid component of PM particles. Currently, the most used solution to this problem is the application of electrostatic chimney separators, which innovations are discussed in our article. Two models of electrostatic precipitators were constructed in this work. The aim of this work was to compare the use of a standard single-pipe chimney electrostatic precipitator with a newer four-pipe variant. Eight measurements were performed on both devices with and without the use of an electrostatic precipitator, on the basis of which the separation efficiency was evaluated for both variants. The results of the measurements showed the initial value of the average PM production in the one-pipe variant decreased from 1012 to 416 $\mathrm{mg}.{\mathrm{m}}^{-3}$ when using the separator, while in the use of the four-pipe variant it decreased from the starting value 342 to only 152 $\mathrm{mg}.{\mathrm{m}}^{-3}$. These results show that the improvement of the classic single-pipe separator by increasing the number of tubes significantly reduced the production of PM emissions and increased the separation efficiency from 66 to 85%.

1. Introduction

Particulate matter emissions released into the environment from thermal decomposition of fuel because of human activity are currently one of the most serious environmental burdens. PM particles from combustion, reaching a size of only a few micrometers, are easily carried into the environment together with the flue gases, where they can be inhaled by living organisms with risks of development of lung diseases [1].

The main source of PM emissions is the combustion of solid fuels in small heat sources, the number of which is constantly growing. This trend is also increased by the promotion of biomass as a renewable resource, which is becoming the preferred alternative to cleaner but more expensive fossil natural gas. In addition to promoting green energy from biomass, the European Union’s policy also aims to reduce emissions from combustion devices that are constantly innovated to reach new standards. This effort of design modifications on the combustion devices significantly reduces their emissivity while guaranteeing optimal combustion conditions. Evidence of this is one Chinese research report that shows a 76% reduction in PM emissions by design improvements to traditional coal stoves. There was a reduction of up to 95% while using a modern gasification boiler [2]. Further research has shown a significant reduction in emissions by ensuring optimal combustion conditions using separation baffles and combustion air distribution [3].

However, these design modifications in combustion devices reach the limits of technical possibilities and it is therefore necessary to focus on technologies of secondary separation of already formed particles by using mechanical, electrostatic, or filtration separators as it is in large heat sources. Initial research was focused on the use of mechanical chamber separators using centrifugal, gravitational, or inertial force to reduce the velocity of the carried particle below the rate of fall [4,5].

One study examined the use of a chamber separator with adjustable separation blades. By rotating the blades, it was possible to influence the magnitude of the forces affecting the velocity of the particle. Measurements showed that the largest particles did not pass through the first blade section. Of the total number of particles, 55.17% were captured. However, despite the entrapment of larger particles, smaller particles escaped in the flue gas stream [6].

Despite the widespread use of mechanical separators, the main part of research is currently focused on the use of electrostatic precipitators, which are preferred due to their simple installation with low space requirements, high efficiency, and low cost. The wider use of electrostatic precipitators has been facilitated by the reduction in the cost of manufacturing high-voltage transformers, which are among the more expensive components of the separator. The separation effect is achieved by creating a high voltage potential between the positive and negative electrodes between which the entrained flue particles pass. The particles are charged by ion current from the discharge electrode and are attracted to collection grounded electrode. Most electrostatic precipitators use chimney pipes directly as collection electrodes, with a charging wire electrode stretched in the middle. Much research has already been applied to conventional chimney electrostatic precipitators and some models are currently available on the market [7,8,9,10,11].

Most research confirms the low efficiency of standard chimney separators and therefore some focus on their design improvement is necessary to increase the efficiency of particle separation [12,13,14,15].

From a survey of the conducted research, the most inspiring models for us were multiple tubular separators, such as the Thai model, which achieved an efficiency of up to 70% [16,17,18].

The aim of this work was to continue this research and build an improved model of the standard chimney separator by multiplying the number of pipes. Our goal was to develop a multiple-tubular precipitator that would perform the function of a standard chimney with the highest possible efficiency with low space requirements and preservation of esthetics. Increasing the number of tubes in standard chimney pipe resulted in an increase in the separation area and a reduction in the long distance between the electrodes, which led to an increase in efficiency.

2. Materials and Methods

2.1. Experimental Combustion Device

In this work, a combustion device with a suitable chimney shape was used to which the electrodes of the electrostatic precipitator were connected. For the purposes of the experiment, a Venus 850 fireplace insert with a nominal power of 7.4 KW and an efficiency of 80.5% was used for measurements. Beech wood with a bark and calorific value of 16.8 MJ/kg was used as fuel. The measurements were performed during the nominal power of the fireplace at a load of approximately 2.7 kg of wood per 30 min measurement. During the measurement, the basic parameters of the combustion process, such as the chimney temperature and the chimney draught at the level of 12 ± 2 Pa, were recorded and maintained under standardized conditions by means of a measuring control panel. PM emissions were measured gravimetrically, each time 5 min after loading, until the flame and chimney draught stabilized. Initially, it was necessary to create a continuous layer of embers, followed by other phases of combustion such as loading, burning of volatile combustibles, and solid carbon burnout. The whole process was repeated for each measurement, as shown in Figure 1.

2.2. Model of Electrostatic Precipitator with HV Power Supply

Chimney electrostatic precipitators were connected directly beyond the combustion device on chimney pipe. The first classical model of the separator shown in Figure 2a consisted of a 2 m standard chimney pipe with a diameter of 200 mm, in the middle of which was a charging electrode in the form of a copper wire stretched by a weight. The positive electrode was separated from the negative electrode by a 200 mm ceramic fireclay chimney through which a supporting threaded rod for the negative electrode was passed.

The second model, shown in Figure 2b, consisted of a 2 m chimney pipe with a diameter of 300 mm in the center of which were four 120 mm pipes through the center of which passed four charging electrodes. The charging electrodes were formed of rigid threaded rods, which were firmly connected at the top and bottom by crosses of iron strips.

Both models of the electrostatic precipitator were connected to their own high voltage (HV) power supply with input voltage 220 V connected to the public network. For the first standard model of a chimney electrostatic precipitator, a CX-600A high-voltage power supply with an electrode spacing of up to 10 cm was used. The power supply had a maximum power of 300 W with adjustable high voltage parameters from 5 to 60 kV.

For a smaller electrode distance of up to 6 mm, the second improved model required a power supply with only 100 W and adjustable high voltage parameters from 10 to 20 kV. Therefore, we could use the lower-cost model CX-120A. Both sources shown in Figure 3 were released at full power during the measurements, and had built-in safety features against damage and skipping of the short circuit electrostatic arc between the electrodes.

2.3. Graviometric Method for Measuring PM Emissions

The basic quantity measured during all measurements was the value of PM emissions produced, which we determined using the gravimetric method. The measurement consists of taking a representative sample with a cascade cylindrical impactor from the flue gas in the flow rate measurement section, which lies directly behind the electrostatic precipitator. The impactor consists of a suction nozzle and a three-chamber space, where the flue gas flow rate is directed by conical inserts to a set of three sampling filters composed of glass fibers. The three-filter probe allows for simultaneous measurement of all three particle sizes: PM > 10, PM 10–2.5, and PM < 2.5. Along with the particle separation, the basic parameters such as chimney draft and temperature required for further calculations are also measured using the ANHOLM measuring control panel during the combustion process. It is also necessary to measure the velocity to ensure an isokinetic condition, which is determined by the same sampling rate as the flue gas flow rate in the chimney. The measuring probes that were used are shown in Figure 4.

Measurements were performed using a TCR Tecora Isostack Basic device, which consists of probes for measuring chimney draught and temperature, and a sampling probe with filter columns for capturing PM from flue gases. The flue gases were then cooled in a cooling and condensing unit and dried in a drying tower containing silica gel. The measurement itself consists of several basic steps. After setting the basic measurement parameters on the TCR Tecora Isostack Basic analyzer, it is necessary to prepare a set of collection filters. The filters must be dried and weighed dry. A Radwag WPS 50X drying balance was used for drying. Prior to each measurement, the impactor must be thoroughly cleaned of settled impurities and heated to operating temperature to prevent condensation in the impactor cylinder. After a 30 min sampling of the particles, the measuring device notified us of the end of the measurement by writing out the evaluation tape with the basic measurement parameters, which are necessary for further calculation of the PM concentration in the flue gas. The filters must be dried again and weighed after the measurement. From the difference in the weight of the filters before and after the measurement, we determined the mass of collected particles. To calculate the final concentration of PM in the gas, it is necessary to insert the mass of particles together with the data from the output tape into the following relation [19]:

$$C=\frac{(m_2-m_1)}{V_{gn}}·1000$$

(1)

where C is the concentration converted to normal conditions in dry gas ($\mathrm{mg}.{\mathrm{m}}^{-3})$, $V_{gn}$ is flue gas sample volume $\left({\mathrm{m}}^3\right)$, $m_2$ is weight of filter after measurement (g), and $m_1$ is weight of filter before measurement $\left(\mathrm{g}\right)$.

From the value of the concentration of PM particles in the flue gas under standardized conditions C, it is also possible to calculate the value of the concentration under reference conditions according to the legal regulation C^r, which is defined by the relation:

$$C^r=\frac{20.95-O_2^{ref}}{20.95-O_2^{priem}}$$

(2)

where $C^r$ is concentration for reference conditions according to the legislative regulation ($\mathrm{mg}.{\mathrm{m}}^{-3})$, $O_2^{priem}$ is measured average oxygen content in the gas volume ($\%)$, and $O_2^{ref}$ is reference oxygen content ($\%)$.

3. Results

3.1. Analysis of Measurements Results

During the measurements, the basic quantities influencing the separation process were recorded, as shown in

Table 1. Average value of basic measured parameters of the precipitation process.

ESP 1	Temperature	Draft	Current	Voltage	Power
One pipe	256 ± 3 °C	12.9 ± 0.7 Pa	0.8 ± 0.1 mA	60 ± 0.2 kV	50 ± 2 W
Four pipes	240 ± 3 °C	12.2 ± 0.7 Pa	1.2 ± 0.1 mA	20 ± 0.2 kV	25 ± 2 W

¹ Electrostatic precipitator model.

. The average values of the measured quantities mainly indicate the electricity consumption and the efficiency of the combustion process. This can be attributed to the larger cross-sectional area of the chimney, which resulted in a slight reduction in the flue gas flow rate and in the chimney draft. Closer electrode spacing also reduced the level of high voltage required to achieve the sufficient corona effect, which led to a reduction in energy consumption.

The gravimetric method of measuring the mass of particles during the measurements helped us to determine the efficiency of the collection process. During the experiments, eight measurements of PM emission production were performed on both models of separators. For each model, four measurements were made with the separator on and four with the separator off. The results of the measurements were samples of PM particles deposited on the collection filters from which we could obtain not only the mass of the particles but also visually evaluate the combustion process from an emission point of view. The individual sets of collection filters for each measurement are shown in Figure 5.

Based on the color of the filters capturing larger particles, with increasing time the chimney becomes clogged and the emissions gradually increase to a point when an electrostatic precipitator was used, which slowed this process slightly. From the masses of the particles deposited on the filters, a much more accurate value of emissions was calculated and given in milligrams per cubic meter; the results are shown graphically in Figure 6.

These results show that using a standard chimney electrostatic precipitator was able to reduce the average value of PM emissions from 1011.81 to 342.03 $\mathrm{mg}.{\mathrm{m}}^{-3}$, which represents a separation efficiency of 66.2%. Upon closer examination of the results, we notice a remarkable tendency of increasing efficiency together with decreasing particle size. In the largest PM 10 fraction, it was only 10.2%, in the case of particles with the PM 10–2.5 fraction, it had already increased to 27.5%, and in the smallest particles the PM < 2.5 fraction had already jumped up to 68.42%. The results obtained from first measurements are satisfactory for a standard chimney separator, but despite the relatively high efficiency, we can see relatively high initial values of PM emissions. However, by using the new four-pipe model, we can see a significant decrease in the input value of PM emissions, which was further reduced by using the electrostatic precipitating effect, as shown in Figure 7.

The results of the measurements when using the improved separator show the average reduction of PM emissions from 416.03 to 152.83 $\mathrm{mg}.{\mathrm{m}}^{-3}$, at an efficiency of 63.5%. More detailed results in this case show good results for separation of larger particles while maintaining the separation efficiency of smaller particles, namely 74.2% for PM 10 and 48.4% for PM 10–2.5, and 63.9% for PM < 2.5. Compared to the first model, the average separation efficiency value was lower but considering the passive separation of the technical design of the new model itself, we can see a significant reduction in emissions. Considering the initial value of 1011.81 $\mathrm{mg}.{\mathrm{m}}^{-3}$ PM emissions, the introduction of the newer model decreased the value to 152.83 $\mathrm{mg}.{\mathrm{m}}^{-3}$, which represents an increase in overall efficiency of up to 85%.

In addition to graphical data, we can also evaluate the combustion process using collection filters and observe the same tendency to increase emissions from clogging of the separation surfaces. However, occasional anomalies in the color of the filters may also be due to the higher bark content in the individual measurements. The individual measuring filters are again shown in Figure 8.

3.2. The Analysis of Detected Dependencies

These results show that the reduction of PM emissions by using the new four-pipe separator model was mainly due to the increasing number of pipes and decreasing pipe diameter. This led to better containment of particles by the collection surface [20,21], which is also confirmed by the mathematical model results shown in Figure 9.

The standard stainless-steel chimney with a diameter of 200 mm and collection surface of 0.628 m² per meter compared to the new improved model with four 120 mm pipes with a collection surface of 1.508 m² represents almost 2.4 times magnification. The distribution of the flue gas stream into smaller pipes also leads to a more efficient and stable effect of the electrostatic field on the particle [22,23], which is also confirmed by the mathematical model results shown in Figure 10.

Smaller pipe diameters allow the use of lower high voltage, which allows the use of smaller and less expensive high voltage sources with reduced energy consumption. This was also proved by our measurements in which the first stronger source during operation at a voltage level of 60 kV consumed approximately 50 W during the measurements. The weaker model at a voltage level of 20 kV consumed only 25 W. At the same time, redistributing the flue gas flow into a larger number of pipes further increased the cross-sectional area from 0.0314 m² to 0.0452 m², which further reduced the particle velocity rate. The turbulent flow caused by changing the cross-section of the pipe also helped to slow the velocity of the flowing particles [24].

4. Discussion and Conclusions

The issue of reducing PM emissions by using electrostatic precipitators in small heat sources is currently one promising area of research. The first prototypes of commercially available chimney electrostatic precipitators are slowly beginning to appear on the market. Despite their simple construction and low construction costs, these devices do not achieve sufficient efficiencies. The solution to this problem is to improve the chimney separators with elements from the industrial practice of large heat sources [25,26,27,28].

This work focused on improving the separation efficiency of a standard chimney separator by increasing the number of pipes. During the work, eight measurements of PM emissions were performed on a conventional chimney separator and eight on its improved four-pipe version. Four measurements with and without electrostatic precipitator were performed for each model. The results of the measurements showed a reduction of PM emissions from an initial value of 1011.81 $\mathrm{mg}.{\mathrm{m}}^{-3}$ to a final value of 152.83 $\mathrm{mg}.{\mathrm{m}}^{-3}$, which represents separation efficiency at 85%. We achieved this reduction by improving the effect of electrostatic precipitation processes and by improving the precipitation ability of the chimney precipitator itself. The measurements confirmed that the increase in separation efficiency was achieved by reducing the distance between the electrodes and by increasing the collection surface and cross-section of the chimney. Because of the increase in the number of pipes, the collection area increased 2.4 times and the cross-sectional area increased by 30%, which led to slowing the flue gas flow rate. From the electrostatic point of view, we also managed to achieve 50% energy savings by reducing the high voltage of the electrostatic field.

The use of electrostatic precipitators to reduce PM emissions is shown to be effective. Improving the standard chimney model by increasing the number of pipes brings significant benefits to the issue; combined with further measurements, reduction of PM emissions can be explored further by adding pipes to create a more compact and efficient model.