The vegetation bio-filter technology, which is an environmentally friendly air purification method using plants, can be divided into special soil-based type and vertical hydroponic type. Two representative commercial products are Fujita's Earth Air Purification System (EAP) and Nedlaw’s Living Wall bio-filter in Canada (FACT, 2012). EAP, a special soil-based bio-filter system by “Fujita”, a Japanese general construction company, is used to decompose air pollutants generated from roads such as CO and NOx through biochemical action of clay minerals and various microorganisms in the soil, which limits filtering speed to 2~4 cm･sec^-1 to secure response time. On the other hand, the “Living Wall bio-filter” of the Canadian “Nedlaw”’s vertical hydroponic type sets the water and rhizosphere mechanisms, not the soil, as the main reaction factor to purify the required air volume of AHU system in the general building. This will ensure a filtration wind speed of 1~2 cm･sec^-1 , which is 50 to 100 times higher than that of a special soil-based type, and provide five occupants within 100 m² in radius with fresh air through a 1 m² -vegetation bio-filter on average. Those two different types of vegetation bio-filters have different characteristics in terms of removal efficiency and air-conditioning volume, and a complex application plans reflecting both characteristics is required for sound AHU linkage.

In general, the purifying filters applied to the AHU are classified into pre-filter, medium filter, HEPA (high-efficiency particulate air) filter and ULPA (ultra-low particulate air) filter. The vegetation bio-filter in this study has a combination of functions; plant growth base and air purifying filter. However, in order to maximize air conditioning volume that can replace a fresh air apparatus within the building, it is required to connect with the AHU. KITECH (Korea Institute of Industrial Technology) summarizes the reliability evaluation criteria of the air purifying filter as shown in Table 1. The performance index consists of applied dust concentration, pressure loss, and capture efficiency, among which the pressure loss is a key index representing the physical performance of the filter reflected in the air conditioning planning and designing. The air filter, in which pressure drop takes place in AHU system, has a direct impact on the energy consumption as the system operates, therefore it is the standard for air conditioning performance (Hashim Kabrein et al., 2016). Generally, in the case of pre-filter made of fiber material, the range of pressure loss is 3~20 mmAq, and it is used as external air processor or pre-filter for a mid-performance filter.

Here, the pressure loss can be interpreted as a performance index, a standard of the physical performance and energy consumption of the air conditioning system, and can be set as a main evaluation index of the bio-filter design performance associated with the AHU.

A suitable test method is required to evaluate the mechanical performance of the air purifying filter and in this study, “AMCA Standard 210” and “KS B 6311: Test Methods for Turbo-Fans and Blowers” were referenced. Here, total airflow, wind speed, and pressure loss can be interpreted as important performance indicators, and total airflow and wind speed can be summarized as follows, which are determined by variables such as expansion coefficient, nozzle differential pressure, density, discharge coefficient, nozzle discharge station area, etc.

Generally, in the case of differential pressure type flow measurement, the reduced end area by the nozzles installed for the generation of the differential pressure generates axial flow, which is fluidly changed depending on the fluid density or the shape of the mechanism, so that actual measurement is limited. As the effluent coefficient is applied to the nozzle cross section to supplement this limitation, the correction coefficient is expressed as follows by applying the Reynolds number, a major dimensionless number of fluid dynamics, to determine dynamic similitude.

This study systematically summarizes the experimental procedure to evaluate the objective air- conditioning performance of the vegetation bio-filter as follows. First of all, the experiment was designed according to the relevant test standard, and an instrument and monitoring system capable of measuring the differential pressure type flow were constructed. In this study, the driving characteristics of the supply and return fans attached to the AHU system for general building were analyzed and the operating conditions enabling the differentiation of pressure loss by material were configured. Based on this, the reference pressure losses of the AHU and the additional experimental system were measured and the pressure loss by vegetation bio-filter material was monitored. Lastly, the economics of AHU energy consumption based on potential materials was reviewed in connection with the monitoring results.

In order to evaluate the quantitative air-conditioning performance of the vegetation bio-filter, the experimental infrastructure by the differential pressure type flow measurement was built as shown in Fig. 1. The infrastructure consists of a vegetation bio-filter chamber, AHU, and wind tunnel. First, the experimental chamber was designed to facilitate testing per filter material and system type (KSSN, 2014) by referring to “KS I ISO 16000-9: Indoor air — Part 9: Determination of the emission of volatile organic compounds from building products and furnishing — Emission test chamber method”. The structure was constructed with a unit volume of 1,000 mm × 1,000 mm × 1,000 mm based on an aluminum profile of 40 mm and PC t = 5 mm.

The AHU system in the experiment was for general building but excluded cooling and heating facilities, and had the same electric motor of 3,425 rpm in the power section divided into a supply fan and a return fan. In the wind tunnel as shown in Fig. 2, flow nozzles were applied to prevent fluid turbulence and to minimize pressure and friction losses. Here, 3-inch and 7-inch nozzles were selectively used to secure a monitoring band for the rated output of the AHU and in order to ensure data reliability, the nozzle wind speed was maintained at 35 m･sec^-1 or less according to the nozzle diameter to measure differential pressure and pressure loss.

Since the vegetation bio-filter should be suitable for both air purification and vegetation growth, the materials that can grow plants were chosen among the materials usable for filter. First, as shown in Table 2, the properties of four different fiber filters - the pre-filters, non-woven fabric, fiber-based vegetation mat, agricultural felt - were summarized.

10 mm pre-filter is applied, made up of 100% PET, a material applied to the pre-filter of general AHU. The non-woven fabric is 20 mm with structural stability, which improves the tensile strength of general pre-filter. The fabric-based vegetation mat is made of 3 mm low melting fiber, combination of bottom ash-based artificial soil, and zeolite, solidified by thermosetting at 110 to 210°C and 330 mm × 330 mm × 60 mm in size. The agricultural felt was wasted fiber material with t = 5 mm. In addition, as shown in Table 3, long-term exposure monitoring confirms the growth suitability of the selected bio-filter material. Here, the fabric-based vegetation mat is confirmed to have stable growth even in outdoor exposure of 20 months or more. The felt for agriculture was exposed for two months, the pre filter was two months, and the non-woven fabric is exposed for one month or more. The objective growth stability results for the long-term exposure test will be presented in subsequent studies.

In this study, the pressure loss of the reference material and the subject materials was measured, and the differential pressure gauge and the temperature and humidity sensor of Table 4 were installed in the experimental area to monitor the air flow environment. First, the fan condition sensitive to the differential pressure change according to the driving characteristics of the supply and return fans attached to AHU was created and the experimental environment of 400, 600, 800, and 1,000 CMH in the total air volume was established accordingly. Pressure loss measurements by material were conducted in accordance with “AMCA Standard 210” and “KS B 6311: Test Methods for Turbo-Fans and Blowers” (AMCA, 2016; KSSN, 2014).

The pressure loss by material of vegetation bio-filter affects the energy consumption of the indoor air conditioning system linked with AHU. To evaluate this, Life Cycle Costing evaluation study (Arnold et al., 2005) of the air filter, published by the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) was referred to. The study shows the energy consumption of the air conditioning system, covering the total air flow, average pressure loss, time in operation, fan efficiency, as in the following formula.

In this study, the possible energy consumption of the bio-filter per material was measured under the condition of 400, 600, 800 and 1,000 CMH in total air flow if it is linked with AHU. As the fan efficiency was set to 58% with driving condition for one week based on the operation of 10 hours a day, the economic efficiency by material was predicted.

In order to obtain accurate results of pressure loss by material, the condition was the same as the condition of fan driving condition in general AHU system. Experimental conditions were divided into independent and simultaneous operation of supply and return fans. The pressure loss was measured under the variations of non-filter, two pre-filters and four pre-filters with the basic condition of a flow nozzle of 7 inches and a total air flow rate of 800 ± 0.9 CMH. The pressure loss by material is shown in Fig. 3, which shows time series analysis and box plot. In the case of single driving of return fan, the average of -41.5 mmAq, -42.4 mmAq, and -43.0 mmAq were found in the non-filter, two pre-filters and four pre-filters respectively, confirming that higher pressure loss per material is discerned in comparison with other conditions.

The reference pressure loss was measured to calculate material-specific pressure loss against pressure loss of non-filter condition. The experiment area is made up of experimental chamber, AHU, and wind tunnel. Pressure loss due to static pressure was measured according to total air flow conditions. The total airflow was 400 ± 1.5, 600 ± 2.6, 800 ± 11.7, and 1,000 ± 17.9 CMH, respectively and pressure loss of non-filter condition was analyzed as shown in Fig. 4 and Table 5.

The highest standard deviation was recorded at a total air flow of 1,000 ± 17.9 CMH, and a relatively stable pressure loss was recorded at a total air flow of 600 ± 2.6 CMH or less. As shown in the above analysis, the pressure loss can be summarized on non-filter basis, and the average pressure loss per total air volume was reflected in the calculation of the pressure loss by material.

The total pressure loss of the pre-filter, the non-woven fabric, the fabric-based vegetation mat, and the agricultural felt selected as vegetation bio-filter materials are shown in Fig. 5. Here, the average total air volume per material was set to 400 ± 0.6, 600 ± 1.2, 800 ± 3.3, and 1,000 ± 11.3 CMH. The pre-filters recorded a low pressure loss of -3.0~-37.3 Pa, so it was selected as the reference material. Here, the reversal phenomenon is confirmed on the basis of the total air volume 800 CMH, which is a compensation phenomenon due to low nozzle air velocity and pressure loss after replacing three inch flow nozzle, which recorded more than 35 m･sec^-1 of nozzle wind speed, with seven inch nozzle. On the other hand, other vegetation bio-filter materials recorded significant pressure loss differences compared to the reference material. First, the pressure loss of the non-woven fabric against the pre-filter was about 21~57%. The other pressure loss levels were 5~23% of the fabric-based vegetation mat and 9~39% of the agricultural felt.

The energy consumption of material-specific bio-filter when linking with AHU is calculated as the following Table 6, based on the total wind volume of the pre-filter as the reference material. Here, the average total air volume was set in accordance with the pressure loss evaluation for each material. Similar to the pressure loss ratio relative to the reference material, the energy consumption of fiber-based vegetation mat, agricultural felt, and non-woven fabric were recorded. First, in the case of non-woven fabric, an additional energy consumption of 1,475.5 Wh is required in comparison with the reference material under the condition of 1,000 CMH. For agricultural felts, an additional energy consumption of 3,488.4 Wh compared to the reference material was recorded under the same air volume. Lastly filter-based vegetation mat showed the maximum additional energy consumption of 7,754.1 Wh compared to the reference material under the same air volume condition.

The energy consumption of vegetation bio-filters by material were different according to the total air volume as indicated in Fig. 6. First, the energy consumption tendency most similar to the reference material was shown in the condition of 600 CMH, and the additional energy consumption of 174~438% was calculated by material. On the other hand, the maximum energy consumption compared to the reference material showed differences by material. The non-woven fabric was 800 CMH, the agricultural felt and filter-based vegetation mat showed the highest energy consumption difference under the condition of 400 CMH. Under the maximum air volume condition, which is 1,000 CMH, the energy consumption compared to the reference material tended to decrease by material.