Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

Journal of Bionic Engineering 14 (2017) 390–400

Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources Jin Tong1,2, Xin Liu1,2, Ronaldo Maghirang3, Kaiqi Wei1,2, Linna Liu4, Chun Wang1, Yunhai Ma1,2, Donghui Chen1,2, Hongjia Yan5, Li Guo1,2 1. Key Laboratory of Bionic Engineering, (Ministry of Education of China), Jilin University, Changchun 130022, China 2. College of Biological and Agricultural Engineering, Jilin University, Changchun 130022, China 3. Department of Biological and Agricultural Engineering, Kansas State University, Manhattan, KS, 66506, USA 4. Institute of Military Veterinary, Academy of Military Medical Sciences, Changchun 130122, China 5. Department of Planning and Construction, Jilin Agricultural University, Changchun 130118, China

Abstract Vegetative Barriers (VB) have the potential to mitigate air pollutants emitted from area sources, including concentrated Animal Feeding Operations (AFOs). However, the mechanism has not been fully investigated, thereby limiting the application of vegetation systems in practice. An experimental method with repeatable and controllable conditions was developed to measure the change of Particulate Matter (PM) concentrations at upwind and downwind of VB in the wind tunnel and observe accumulated PM on leaves with Scanning Electron Microscope (SEM), thus evaluating the ability of VB in mitigating PM emitted from AFOs. Branch-scale vegetation, clove (syzygium aromaticum) was selected because its leaves are one of the major factors affecting PM dispersion. The results show that the branch-scale barriers, as porous medium have the ability to interfere with airflow and reduce PM, which could be influenced by wind speed, particle size fraction and surface area density of clove. Moreover, clove elements could adjust to the wind and the micro structure of clove (such as the hierarchical structures of leaves) affected on the PM deposition. These results indicate that the methods developed in this study may be used to evaluate the potential of vegetation in mitigating PM from stationary sources, and some characteristics of vegetation can be further studied as bionic prototype for exploring engineering application of reducing particulates. Keywords: vegetative barrier, particulate matter, wind tunnel, clove, animal feeding operation Copyright © 2017, Jilin University. Published by Elsevier Limited and Science Press. All rights reserved. doi: 10.1016/S1672-6529(16)60407-9

1 Introduction Particulate Matter (PM) consisting of the mixture of solid and liquid particles suspended in the air is among the first air pollutants that drew attention of researchers worldwide[1,2], especially in China, which recently has been experiencing extremely severe and persistent haze pollution in several regions[3]. Emissions of PM from intensive Animal Feeding Operations (AFOs) are considered harmful to human health (such as respiratory, heart and lung diseases) and affect the local and regional air quality[4–7]. Among existing mitigation strategies, animal producers show great interests in using vegetation systems to prevent and reduce the spread of pollutants to neighboring residents[2,8,9]. Corresponding author: Li Guo E-mail: [email protected]

Vegetative Barriers (VB) are also known as vegetative system, buffer, shelterbelt and windbreak, which typically include trees and shrubs arranged in row or group configurations[8]. Some special branch-scale structures and micro-structures of vegetation have formed in the evolution and contribute to their adapting to environment[10]. Based on limited number of research in urban PM reduction by vegetation, VB can collect PM primarily by physical mechanisms, such as impaction[11,12]. Leaves, porosity, canopy morphology and elements, micro-structure of trees and shrubs play major roles in trapping PM[13–16], since some unique structures of plant leaves have been formed in the evolutionary process[10,17]. Even the same species of vegetation might significantly present different capability of capturing

Tong et al.: Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

PM in different polluted regions due to variation in particle size, PM components, and meteorological conditions. Therefore the research on the removal of airborne particles emitted from specific or stationary source area by VB is important and urgent for barrier design and management to get maximum effectiveness[12,18,19]. A few studies have been conducted to evaluate the potential of vegetation in trapping PM and other air pollutants from AFOs. Adrizal et al.[9] indicated that plant foliage has the capacity to capture all size categories of PM from poultry exhaust fans and showed unique species differences in their capacity to hold PM. Laird[20] reported that downwind particles transmission could be reduced by 35% – 56% using VB. Current studies on the efficiency of vegetation in reducing PM emissions from stationary source of AFOs are very limited and more researches are needed to evaluate their effectiveness and to establish abatement strategies for engineering application. This study aimed at exploring reliable methods for evaluating the effectiveness of VB (clove) from two aspects, branch-scale structure of VB and micro-structure of the vegetation elements under comparable conditions. Considering that wind tunnel studies can provide many insights into pollutant dispersion with controllable experimental conditions and can be further linked to on-site measurements[2,12], a wind tunnel was developed for providing reproducible and consistent experimental conditions. The objectives of this study were to develop a laboratory experimental method for rapidly ascertaining the ability of vegetation barriers in controlling PM emissions with controllable wind speed, stationary PM sources from AFOs, and comparable configuration of vegetation barriers; and to quantitively and qualitatively evaluate the effectiveness of branch-scale barriers of clove and the micro-structure of clove leaves in mitigating the dispersion of small parti1.4 m

cles, which may provide bionic investigation to inspire the improvement of current PM reduction strategies.

2 Materials and methods In this study, a low-speed wind tunnel was designed based on the wind-speed statistics of Jilin Province in China with annual average of 2.6 m·s−1 from 1975 to 2012[21]. For the experimental tests, different treatments were considered which included three levels of wind speed (v1, v2, v3), two levels of size fraction (PI and PII) of particles generated for particle dispersion, four levels of VB (without barrier and three levels of Surface Area Density (SAD) with barriers, i.e., control, S1, S2, S3). A total of 24 sets of tests were replicated 3 times as described in the following sections. 2.1 Experimental setup The wind tunnel (Fig. 1) was designed according to the principles given in Refs. [22, 23], with the test section of 2.0 m in length and highest wind speed of 6 m·s−1. To simulate the emissions of PM from feedlot facilities, powder particles were continuously dispersed at the airflow inlet of the contraction section using customized aerosol diffuser and injection system (BT-901, Bettersize Instruments Ltd., China) with flow rate of 1 L·min−1. The VB were set up 0.6 m away from the inlet of the test section and perpendicular to the incoming flow direction. The bottom surface of the test section was assumed as plane of z = 0; the lateral side of barriers facing airflow was assumed as x = 0; and the vertical plane at the center of plane x = 0 was assumed as y = 0. Monitoring points 1–7 were velocity monitoring points, meanwhile monitoring points 1 and 4 were also PM concentration monitoring points, as shown in Fig. 1. During all tests, average temperature and relative humidity in the laboratory were 23 ˚C and 18%, respectively.

1.0 m 2.0 m Contraction section

1.2 m

Fan 1.5 m

3 1 2 0.6 m Monitoring point of particulate matter concentration

0.5 m 0.4 m

Test section

Airflow

z

391

4 5

0 Monitoring point of Airflow velocity

6 7

0.5 m Vegetative barrier

Injection location of generated particles

x

Fig. 1 Diagram showing the wind tunnel experimental setup (not drawn to scale).

Journal of Bionic Engineering (2017) Vol.14 No.2

392

erated and used for all sets of tests. Component analysis of these powders was performed using Energy Dispersive Spectrometer (EDS, ISIS300, Oxford Instrument, Britain).

2.2 Source of particles for dispersion An approach was explored to simulate PM source from a cattle feedlot in northeast China, where bulk of mixed samples consisting of dust, soil, manure and feeds from the feedlot surface[24] were collected. These samples were processed with the following procedure as shown in Fig. 2. Heated in an oven at 110 ˚C[25], samples were dried and then pulverized using a planetary ball mill to obtain two levels of powder particles (PI and PII: powder particles had 5.24% and 9.81% mean accumulative percentages of fine particles with diameter less than 2 μm, respectively) with different particle size distribution, tested by Laser Particle Size Analyzer (LPSA, BT-2003, Bettersize Instruments Ltd., China). The milling times were pre-determined based on fine particle sizes (≤ 2 μm), considering greater risks of adverse health effects resulted from small particles (< 3 μm )[7,26]. Then prepared powder was dispersed in the wind tunnel as stated above. Totally, 10 specimens of PI powder and 4 specimens of PII powder were gen-

2.3 Vegetative barriers Clove (syzygium aromaticum), belonging to the family Myrtaceae[27], was selected as abundant species in Asia[28] with potential of capturing particles[29]. Mature branches of clove in the campus of Jilin University were cut and rinsed, then fixed into two floral foam bricks as shown in Fig. 3. Three levels of VB were established with total leaves of 198, 132 and 66 on 16, 12 and 6 stems, respectively, considering that leaves are the primary contribution on particle collection of vegetation[30,31]. The SAD of VB was calculated based on the method in Ref. [31], as indicated in Eq. (1):

SAD = S total / V ,

(1)

where, V is the volume of vegetative barriers (m3), Stotal

Fig. 2 Diagram showing the processing of powder particles used for the source of particulate matter emissions obtained from solid mixtures from a cattle feedlot. Table 1 Mean surface area of foliage elements of clove obtained from samples of 100 leaves and twigs and 20 stems Foliage elements of clove

Length (×10−2 m)

Diameter (×10−2 m)

Surface area of clove elements (×10−4 m2) Average Range

Average

Range

Average

Range

Stem

65.0

60.0 – 70.0

0.8

0.6 – 1.0

16.3

Twig

2.3

1.8 – 2.8

0.1

0.1 – 0.2

0.9

0.5 – 1.4

Leaf

–

–

–

–

57.3

26.0 – 113.0

11.3 – 22.0

Tong et al.: Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

393

is the total surface area (m2) of VB as shown in Eq. (2): S total = Sleaf + S twig + Sstem ,

(2)

where Sleaf, Stwig and Sstem are the surface areas (m2) of leaves, twigs and stems, respectively, which can be calculated by the method in Ref. [31]. Thus, a total of 100 leaves with twigs and stems were chosen randomly from clove hedges. Then the surface area of twigs and stems could be calculated as cylinders[31], and leaves were calculated through their profiles in a coordinate paper. The mean surface areas of three vegetative elements are listed in Table 1. Consequently, the SAD of three levels applied in this study were 6.56 ×10−4 m−1 (S1), 12.81×10−4 m−1 (S2) and 19.05 ×10−4 m−1 (S3). 2.4 Monitoring of air velocity As the reduction of air velocity could impact the potential of PM deposition in this area[32], the velocity distribution around the barriers was measured using hot-wire anemometers (D8880, CEM Machinery Industry Co., Lt., China) at 0.9 m·s−1, 1.8 m·s−1 and 2.7 m·s−1. Seven monitoring points were selected in plane of y = 0, indicated as points 1–7 in Fig. 1. 2.5 Monitoring of PM concentration The mass concentrations of PM with equivalent aerodynamic diameter of 1 μm, 2.5 μm, and 10 μm or less (indicated as PM1, PM2.5 and PM10, respectively) were simultaneously monitored at upwind points 1 and downwind point 4 (Fig. 1) using two aerosol monitors (DustTrak DRX 8533, TSI, USA) which are suitable for relative concentration measuring. The flow rate of isokinetic sampling was 3 L·min−1. The sampling duration for each test was 3 min with data acquisition per second. During experiments, DustTraks were regularly checked and calibrated with its zero filters[34]. 2.6 Analysis of particles on leaves In this study, some clove leaves closed to the upwind monitoring location and with more accumulated particles were selected after tests for the analysis of Scanning Electron Microscope (SEM, EVO18, Carl Zeiss, German). All of the samples for SEM analysis were processed by a vacuum freeze dryer (LGJ-10C, Four-Ring Science Instrument Plant Beijing, China).

Fig. 3 Pictures of (a) branch-scale barrier of clove, (b) branch with one stem, and (c) freshly washed leaf.

2.7 Data analysis The percentage reduction (η, %) of PM concentration by the VB was evaluated based on the relative changes of particle concentrations at points 1 and 4 as:

η = ( C1 − C4 ) C1 × 100,

(3)

where C1 and C4 are the mass concentrations (μg·m−3) of PM1, PM2.5, and PM10 at upwind (point 1) and downwind (point 4) of VB, respectively. The mass concentrations of PM1, PM2.5 and PM10 were the mean values of three replications obtained for each test with background concentrations subtracted. The particle size distribution of the powder was reported with average values of Geometric Mean Diameter (GMD) and the cumulative percentage of particle size less than 2 μm (fine particles) and 10 μm (coarse particles) for all powder specimens generated. Statistical analysis was also conducted with 5% level of significance[8], based on Shapiro-Wilk test (W-test)[35] and two-sample Kolmogorov-Smirnov tests[36].

3 Results and discussion 3.1 Physical and chemical properties of dispersed particles The wind tunnel provided basic air source system with consistent, uniform and stable air flow. The VB with comparable SADs were established. Therefore, the properties of dispersed particles became the key controllable factor. Based on the test results of LPSA, the generated powder particles of PI and PII had GMD of 41.5 μm ± 1.17 μm and 12.3 μm ± 0.22 μm, respectively, corresponding to mean cumulative percentages of fine particles with diameter less than 2 μm of 5.24%

Journal of Bionic Engineering (2017) Vol.14 No.2

Elemental concentration (wt.%)

3.2 Effect of clove barriers on the airflow The wind speeds at monitoring points 1–7 for three levels of barriers are shown in Fig. 5. Generally, as the flow approached the barriers, the stream-wise velocities at points 1 and 2 were decreased. The flow over the top

Fig. 4 Elemental concentration (wt.%) of particles generated and used in this study tested by Energy Dispersive Spectrometer (Elements are arranged by rising atomic number). Wind speed (m·s−1) 0.9 0.9 1.8 1.8 2.7 2.7

4 4

-1

Wind speed (m⋅s )

3 3

-1

Velocity(m·s (m⋅s −1 )) Velocity

(a)

2 2

1 1

0 0

11

22

44

(c) (c)

5 6 7 33 44 5 6 7 Monitoring points Monitoring points (a) Wind speed (m⋅s ) Wind0.9speed (m·s−1) 1.8 0.9 2.7 1.8 2.7 -1

33

-1

Velocity (m ⋅s )−1) Velocity (m·s

(b)

-1

(± 0.08%) and 9.81% (± 0.07%), and coarse particles with diameter less than 10 μm of 20.1% (± 0.06%) and 43.6% (± 0.05%). The results indicated that the two levels of particle source for dispersion had great differences in particle size distribution. As for the measured particle concentrations in the air without the presence of VB (control), the downwind concentrations (monitoring point 4) were slightly less than that at upwind (monitoring point 1) (p-value < 0.05) for all PM1, PM2.5 and PM10 values, which was reasonable because of particle dispersion and deposition. The percentage reductions of PM concentration, then, were used as control to compare with that when VB was present. Their ratios (PM1/PM2.5/PM10) were nearly equal for all levels of tests at both downwind and upwind monitoring sites. These results indicated that this experimental system could provide consistent and stable particle emission source with powder particles generated. In addition, 12 different elements (C, N, O, Na, Mg, Al, Si, P, S, Cl, K, Ca) were identified through EDS tests in the particle powder specimens as shown in Fig. 4. The Si should mainly come from (or exist in) SiO2 in the soil; C, O, N were mainly from organic matter of manure; and Ca, Na, Mg and Al were mainly from inorganic salts[38]. Furthermore, C, O, N, Si and Ca were the major components with mean mass percentages of 30.4%, 36.5%, 20.5%, 4.0% and 3.1%, respectively. Compared with other researches conducted in the feedlot[24,39,40], the particles used in this study were rich in C, N and O instead of Si. This may be because the samples were collected mainly from the pen area and road dust was not included for this study, since organic dust from animal excrement was considered as the major contribution that lead to toxic syndrome, such as respiratory and lung disease[7,40]. Furthermore, this method could only consider the primary PM and the secondary PM emitted from the AFOs had to be ignored, since the collected samples were processed with higher temperature, in dry condition and lack of ammonia, hydrogen sulfide, and Volatile Organic Compounds (VOCs)[41]. In general, this experimental protocol could provide consistent particle source for all tests. It was not intended for predicting actual particle emissions from AFOs. Rather, this experimental method can be used to rapidly and inexpensively evaluate the potential of vegetation in reducing PM emitted from AFOs under a variety of conditions.

Velocity(m·s (m⋅s−1)) Velocity

394

22

11

00

1 1

2

44

3 4 5 Monitoring points

6 7 Monitoring points Wind speed ( m⋅s ) (b) Wind speed (m·s−1) -1

0.9 1.8 2.7

33

0.9 1.8 2.7

22

11

00

1 1

2 2

3 4 5 6 Monitoring Monitoringpoints points

7

(c)

Fig. 5 Air flow velocities at 7 monitoring points around vegetative barriers at various specific area density (a) S1; (b) S2; (c) S3.

of the barrier at point 3 was accelerated and the bleeding

Tong et al.: Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

flow passed through the barriers at point 5 was greatly reduced, while it fluctuated with the wind speed and SAD of the barriers at point 4. Eventually, flow at points 6 and 7 appeared to be restored. These flow characteristics were consistent with the results in the literatures related to the shelter function of vegetation barriers[42,43]. For the downwind monitoring point 5, which was further away from the top of the barrier compared with point 4, the velocity was greatly reduced by the barrier compared with the velocity of approaching flow at monitoring point 2 for all cases. The velocity at downwind monitoring point 4 had higher values compared with that at upwind monitoring point 1 for the cases when the barrier had the least SAD as shown in Fig. 5a and for the case it had medium SAD under the highest wind speed as shown in Fig. 5b. These results indicated that for sparse barrier the sheltering effect was quite limited especially when the wind became stronger because that the sheltering area was reduced. One of the reasons was that the sparse barrier had higher porosity and much air could flow through the barrier[44]. Another reason is that the orientation of leaves could easily vary with the wind when the barrier was sparse so that the vegetation density changed with winds[12,31]. In addition, considerable sway of the branch stem with the increase in wind speed was detected in this study, causing the variation in the barrier height and great increase in air velocity at monitoring point 4 especially for the sparse

395

barrier. Therefore, the adjustment of leaves and stems with wind should be further investigated for its contribution to sheltering effects of vegetation barrier, from which bionic information may be derived for the improvement of practical filter or barriers as well. Another event observed from Fig. 5 is that great reduction in velocity at monitoring point 5 occurred when using medium dense barriers. This might be caused by the random arrangement of branches and leaves on the stems, which resulted in more leaves and/or better combination of leaves standing in front of the monitoring location to disturb the air flow. 3.3 Effects of clove barriers on PM concentration reduction The VB changed the airflow as mentioned above, which might affect the dispersion of PM by reducing their transmission and dilution when they were lifted up to bypass the VB. The VB reduced PM concentration by filtering the particles through deposition[12,42]. The effect of these two ways that the clove barriers reduce PM may be reflected by the percentages of PM reduction obtained from upwind monitoring point 1 and downwind point 4, as listed in Table 2. In general, the reductions of PM concentration by VB were significantly different with that from control tests without the presence of barriers (p-value < 0.05).

Table 2 The percentage reductions* of mass concentrations (PM1, PM2.5 and PM10) obtained for all experiments under different conditions (η, %) Surface area density

Particle size PM1

S1

PM2.5

S2

S3

Control

Wind speed 1.8 (m·s−1)

Wind speed 0.9 (m·s−1) PI

Wind speed 2.7 (m·s−1)

PII

PⅠ

PⅡ

PⅠ

PⅡ

30.3 ± 10.8

35.9 ± 2.4

−0.9 ± 8.4

27.2 ± 1.7

−8.6 ± 6.7

17.5 ±11.2

33.1 ± 10.4

33.0 ± 2.5

3.5 ± 8.0

23.2 ± 2.0

−11.1 ± 4.4

12.8 ± 11.8

PM10

31.3 ± 9.8

31.8 ± 2.7

0.6 ± 8.8

23.4 ± 3.0

−23.3 ± 7.3

13.6 ± 11.7

PM1

20.4 ± 11.7

48.0 ± 6.0

17.4 ± 11.3

32.3 ± 7.4

26.8 ± 13.0

21.6 ± 7.5

PM2.5

17.5 ± 12.6

44.9 ± 6.2

13.1 ± 11.9

28.9 ± 7.5

22.2 ± 14.0

17.6 ± 7.9

PM10

21.6 ± 12.1

45.0 ± 6.6

14.0 ± 10.9

28.0 ± 7.5

20.9 ± 14.5

18.5 ± 8.2

PM1

43.3 ± 5.9

47.2 ± 8.9

28.7 ± 5.8

40.7 ± 5.7

24.9 ± 8.9

27.8 ± 5.3

PM2.5

40.1 ± 6.2

44.4 ± 9.3

24.8 ± 6.1

37.0 ± 5.6

20.7 ± 9.5

23.7 ± 5.8

PM10

42.5 ± 5.9

46.7 ± 9.3

23.8 ± 5.7

39.6 ± 5.5

19.9 ± 10.4

23.3 ± 6.0

PM1

7.2 ± 0.6

6.4 ± 2.8

9.7 ± 0.4

2.9 ± 2.7

9.3 ± 1.1

2.8 ± 2.2

PM2.5

3.1 ± 0.5

9.8 ± 1.1

4.9 ± 0.6

5.5 ± 1.1

4.8 ± 1.0

5.8 ± 1.1

1.8 ± 1.1

8.4 ± 2.3

4.9 ± 1.0

4.0 ± 2.4

4.5 ± 1.5

3.4 ± 3.0

PM10 *

All of mean values were reported with Standard Error (SE), i.e. mean ± SE, n = 3

396

Journal of Bionic Engineering (2017) Vol.14 No.2

Generally, the percentage reductions of PM centrations obtained from the tests using powder PI were significantly different from those of powder PII (p-value <0.05). When the percent is about 89%, mean PM reduction in percentages using powder PI were less than that of powder PII. This indicated that these barriers were more effective in mitigating PM sources with more fine particles. The PM reductions using powder PI also showed large variances, especially for the cases with sparse barriers and at higher wind speed. These might be because that less fine particles could easily pass through high porosity barriers and had little chance to contact with the elements of barriers when the air flowed fast. Thus, the dispersion and deposition of fine particulates appeared to be very arbitrary, even some negative reduction events occurred with the increasing wind speed when the barriers had the least SAD. Consequently, for experimental investigation of fine and coarse PM reduction, PM sources with more fine particles may provide better insight of the potential of vegetation barriers for PM mitigation. The reductions in the percentage of PM concentrations for all tests using powder PII are illustrated in Fig. 6. Concentrations reductions of PM1, PM2.5 and PM10 ranged from 17.5% to 48.0%, 12.8% to 44.9%, and 13.6% to 46.7%, respectively. The air velocity had great effect on the PM reduction for all small particles and the ability of the barriers to mitigate PM was increased with the decrease in wind speed. This may be because the broad leaves and flexible stems of clove adjusted their orientation with wind, thereby changing the porosity and structure of the barriers at high wind speed so that those fine particles could quickly pass through the barriers[31,42]. Therefore, when the clove is used in practice for capturing small particles, it may be considered to set it at low speed areas. The three levels of SAD in this study also presented significant differences in the percentage reduction (η) of PM (p-values < 0.05). At low wind speed, the PM reduction using barriers with S2 were similar with those using S3, and greater than those using S1. With the increase in wind speed, the denser the barriers were, the better the effectiveness of barriers became. This indicated that denser clove barriers may be designed and applied at high speed areas. However, low porosity vegetation may force air stream to pass over it and affect the deposition of air pollutants on vegetation[12,31].

Therefore, the optimal SAD for clove should be investigated besides the three levels of barriers. PM reduction by these barriers were not significant (p > 0.05) for all monitored particle sizes though PM1 showed slightly higher reduction than those of PM2.5 and PM10, as shown in Fig. 6. This result might be caused by the structural characteristics of clove elements, as described in the following section. 3.4 Effects of clove element characteristics on particulate mitigation In this study, the particles deposited on the surface of leaves can be also investigated to better understand PM1.0 PM2.5 PM10

50 40 30 20 10 0

0.9

1.8 Wind speed (m·s−1) (a)

2.7

60

PM1.0 PM2.5 PM10

50 40 30 20 10 0

0.9

60

1.8 Wind speed (m·s−1) (b)

2.7

PM1.0 PM2.5 PM10

50 40 30 20 10 0

0.9

1.8 Wind speed (m·s−1) (c)

2.7

Fig. 6 Percentage reduction of particulate matter concentrations for the tests using powder PII having more fine particles and barriers with surface area density of (a) S1, (b) S2, (c) S3.

Tong et al.: Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

the effects of micro-structure of leaves in trapping particles and to explore potential inspiration and feasible methods for bionic studies. The elements of vegetation determined the porosity of barriers which influenced the air flow characteristics and particulate dispersion trajectory as mentioned. In this study, leaves were one of the major factors for vegetation barriers affecting the dispersion and deposition of PM in the air. The clove used in this study had broad leaves with reniform or orbicular-ovate shape, the arrangement of simple leaf on petiole, opposite leaves on the stem, and adjacent pair of leaves had less overlap as shown in Fig. 3. These characteristics of clove leaves might cause big adjustment or porosity change of sparse barriers, which resulted in significant changes of velocity and the decrease in PM values with the increase in wind speed in this study. The clove leaves had pinnate venation crossing the non-smooth surface with midrib in the center and many “ridge”-shape structures distributed on the surface as shown in Fig. 3c and Fig. 7. More irregular “ridges” and apparent ditches of stripes distributed around or closed to midrib compared with that near margin area as shown in Fig. 7. This formed a type of tree-like hierarchical structure, which is common in biological investigation (a)

(b)

(c)

(d)

(e)

(f)

Fig. 7 Representative scanning electron micrographs showing the particles on the adaxial lamina of clove leaves selected from the tested barriers, taken from the areas (a) close to the midrib in the center of the lamina, (c) close to the apex and margin entire of the lamina and (e) between (a) and (c). The images of (b), (d), and (f) are corresponding to yellow circle areas on (a), (c), and (e), respectively, with higher magnification. In images above, scale bar = 10 μm, WD = 9.5 mm, EHT = 20 kV.

397

and lead to a large contact area with environment[45,46]. Through SEM tests, a lot of very fine particles deposited on the area near the midrib were observed (Figs. 7a and 7b). Relatively less particles was observed on the area closing to the margin of leaf (Figs. 7c and 7d). Many coarse particles appeared on the area between the midrib and margin (Figs. 7e and 7f). These indicated that the micro-structure of clove leaves greatly affected the amount and sizes of particles captured by the leaves which agreed with previous studies[48], and the non-smooth surface with ditches, ridges and other structure may increase the possibility of PM deposition. The intensive hierarchical structure contributed to the trapping of particles, provided a large contact area and roughness acting on the resistance to airflow. Thus future work may focus on reproducing biological surface of barriers simulating the micro-structure of clove and/or other plant leaves, for further exploring bio-mechanism of engineering area.

4 Conclusion In this study an experimental method was developed and explored to evaluate the ability of VB in mitigating PM emitted from AFOs and measure the changes of PM concentrations at upwind and downwind of VB in the wind tunnel with repeatable and controllable conditions as well as observe accumulated particles on leaves with SEM. This new method may be applied for conducting laboratory-scale experiments to evaluate the potential of vegetation in mitigating PM and provide references for bionic studies. The results indicated that the effects of branch-scale vegetation barriers of clove on interfering with the airflow and reducing PM concentrations in varying degrees were influenced by the wind speed, PM size fraction, surface area density and arrangement of clove, with concentration reductions of PM1, PM2.5 and PM10 ranged from 17.5% to 48.0%, 12.8% to 44.9%, and 13.6% to 46.7% (using powder PII ). The phenomena of more stereo venations and higher intensive of hierarchical structure clove leaves contributed to trapping more fine particles were observed, indicating that the surface structure of barriers elements may have a great effect on its ability to mitigate PM, especially fine PM. Above all, these mechanisms need to be further studied to find most effective vegetation species and to develop potential bionic prototypes from vegetation for the improvement of filter or screen barrier

Journal of Bionic Engineering (2017) Vol.14 No.2

398

system to increase their effectiveness.

antibiotic resistance genes: Aerial transport from cattle feed

Acknowledgment

yards via particulate matter. Environmental Health Perspectives, 2015, 123, 337–343.

This study was supported by the National Natural Science Foundation of China (No. 51575228 and 41501510), the China Postdoctoral Science Foundation (No. 2013M540252), the Scientific Research Foundation for the Returned Overseas Chinese Scholars sponsored by State Education Ministry, and the “13th Five-Year Plan” Scientific Research Foundation of the Education Department of Jilin Province. We would like to thank Mr. Xiangli Song for his technical support in the establishment of wind tunnel and Ziyang Wang, Peng Gao, Honglie Song, Zhihui Gao, Dongguang Zhang for their helpful assistance in experiments and data collection. We are also grateful to Dr. Siyan Zhao from the Institute of Military Veterinary at the Academy of Military Medical Sciences for their technical support in aerosol diffuser and injection system.

References [1]

Pascal M, Chanel P D C, Wagner V, Corso M, Tillier C, Goria S, Tertre A L, Chatignoux E, Ung A, Beaudeau P, Medina S. The mortality impacts of fine particles in France. Science of the Total Environment, 2016, 571, 416–425. Ni J Q. Research and demonstration to improve air quality for the U.S. animal feeding operations in the 21st century-a critical review. Environmental Pollution, 2015, 200, 105–119.

[3]

Huang R, Zhang Y, Bozzetti C, Cao J, Han Y, Dallenbach K R, Slowik J G, Platt S M, Canonaci F, Zotter P, Wolf R, Pieber S M, Bruns E A, Crippa M, Ciarelli G, Piazzalunga A, Schwikowsk

M,

Abbaszade

G,

Schnelle-Kreis

J,

Zimmermann R, An Z, Szidat S, Baltensperge U, El Haddad I, Prevot A. S. H. High secondary aerosol contribution to particulate pollution during haze events in China. Nature, 2014, 54, 218–222. [4]

Bonifacio H F, Maghirang R G, Trabue S L, McConnell L L, Prueger J H, Razote E B. TSP, PM10, and PM2.5 emissions from a beef cattle feedlot using the flux-gradient technique. Atmospheric Environment, 2015, 101, 49–57.

[5]

Xu W, Zheng K, Meng L, Liu X, Hartung E, Roelcke M, Zhang F. Concentrations and emissions of particulate matter from intensive pig production at a large farm in north China. Aerosol and Air Quality Research, 2016, 16, 79–90.

[6]

Cambra-lópez M, Aarnink A J A, Zhao Y, Calvet S, Torres A G. Airborne particulate matter from livestock production systems: A review of an air pollution problem. Environmental Pollution, 2010, 158, 1–17.

[8]

Parker D B, Malone W, Walter D. Vegetative environmental buffers and exhaust fan deflectors for reducing downwind odor and VOCs from tunnel-ventilated swine barns. Transactions of the ASABE, 2012, 55, 227–240.

[9]

Adrizal A, Pattersion P H, Hulet R M, Bates R M, Despot D A, Wheeler E F, Topper P A, Anderson D A, Thompson J R. The potential for plants to trap emissions from farms with laying hens: 2. Ammonia and dust. Journal of Applied Poultry Research, 2008, 17, 398–411.

[10] Ren L, Wang S, Tian X, Han Z, Yan L, Qiu Z. Non-smooth morphologies of typical plant leaf surfaces and their anti-adhesion effects. Journal of Bionic Engineering, 2007, 4, 33–40. [11] Tong Z, Baldauf R W, Isakov V, Deshmukh P, Zhang K.

Bentayeb M, Blanchard M, Cochet A, Pascal L, Host S,

[2]

[7]

McEachran A D, Blackwell B R, Hanson J D, Wooten K J, Mayer G D, Cox S B, Smith P N. Antibiotics, bacteria, and

Roadside vegetation barrier designs to mitigate near-road air pollution impacts. Science of the Total Environment, 2015, 541, 920–927. [12] Janhall S. Review on urban vegetation and particle air pollution-deposition and dispersion. Atmospheric Environment, 2015, 105, 130–137. [13] Liu J, Zhu L, Wang H, Yang Y, Liu J, Qiu D, Ma W, Zhang Z, Liu J. Dry deposition of particulate matter at an urban forest, wetland and lake surface in Beijing. Atmospheric Environment, 2016, 125, 178–187. [14] Sgrigna G, Baldacchini C, Esposito R, Calandrelli R, Tiwary A, Calfapietra C. Characterization of leaf-level particulate matter for an industrial city using electron microscopy and X-ray microanalysis. Science of the Total Environment, 2016, 548–549, 91–99. [15] Hofman J, Bartholomeus H, Calders K, Wittenberghe S V, Wuyts K, Samson R. On the relation between tree crown morphology and particulate matter deposition on urban tree leaves: A groung-based LiDAR approach. Atmospheric Environment, 2014, 99, 130–139. [16] Sæbø A, Popek R, Nawrot B, Hanslin H M, Gawronska H, Gawronski S W. Plant species differences in particulate matter accumulation on leaf surfaces. Science of the Total Environment, 2012, 2015, 347–354. [17] Wang S, Ren L, Liu Y, Han Z, Yang Y. Mechanical characteristics of typical plant leaves. Journal of Bionic Engineering, 2010, 7, 294–300.

Tong et al.: Investigation of the Potential and Mechanism of Clove for Mitigating Airborne Particulate Matter Emission from Stationary Sources

399

[18] Zhang W, Wang B, Niu X. Study on the adsorption capaci-

[31] Tiwary A, Morvan H P, Colls J J. Modelling the

ties for airborne particulates of landscape plants in different

size-dependent collection efficiency of hedgerows for am-

polluted regions in Beijing. International Journal of Envi-

bient aerosols. Journal of Aerosol Science, 2005, 37,

ronmental Research and Public Health, 2015, 12, 9623–9638. [19] Przybysz A, Sæbø A, Hanslin H M, Gawronski S W. Accumulation of particulate matter and trace elements on

990–1015. [32] Badr T, Harion J L. Numerical modelling of flow over stockpiles: Implications on dust emissions. Atmospheric Environment, 2005, 39, 5576–5584.

vegetation as affected by pollution level, rainfall and the

[33] Wang X, Chancellor G, Evenstad J, Farnsworth J E, Hase A,

passage of time. Science of the Total Environment, 2014,

Olson G M, Sreenath A, Agarwal J K. A novel opticla in-

481, 360–369.

strument for estimating size segregated aerosol mass con-

[20] Laird D J. Wind Tunnel Testing of Shelterbelt Effects on Dust Emission from Swine Production Facilities, MS Thesis, Iowa State University, Ames, USA, 1997. [21] Xie J, Liu T, Li T. Variations and potential causes of surfaces and free atmospheric wind velocities in Jilin. Plateau Meteorology, 2015, 34, 1424–1434. (in Chinese) [22] Liu Z. Structure Design of Wind Tunnel, China Astronautic Publishing House, Beijing, 2005. (in Chinese) [23] Wu R. Principle of Wind Tunnel Design. Beihang University Press, Beijing, 1985. (In Chinese) [24] Uzoma K C, Inoue M, Andry H, Fujimaki H, Zahoor A,

centration in real time. Aerosol Science and Technology, 2009, 43, 939–950. [34] TSI. Operation and Service Manual for DustTrak DRX Aerosol Monitor Model 8533/8534/8533EP. TSI Incorporated, USA, 2014. [35] Villasenor

A,

Estrada

E

G.

A

generalization

of

shapiro-wilk's test for multivariate normality. Communication in Statistics- Theory and Methods, 2009, 38, 1870–1883. [36] Lilliefors H W. On the kolmogorov-smirnov test for normality with mean and variance unknown. Journal of the

Nishihara E. Effect of cow manure biochar on maize pro-

American Statistical Association, 1967, 62, 399–402.

ductivity under sandy soil condition. Soil Use and Man-

[37] Hiranuma N, Brooks S D, Gramann J, Auvermann B W.

agement, 2011, 27, 205–212. [25] Ashbaugh L L, Carvacho O F, Brown M S, Chow J C, Watson J G, Magliano K C. Soil sample collection and analysis for the fugitive dust characterization study. Atmospheric Environment, 2003, 37, 1163–1173. [26] Lee S, Adhikari A, Grinshpun S A, McKay R, Shukla R,

High concentrations of coarse particles emitted from a cattle feeding operation. Atmospheric Chemistry & Physics & Discussions, 2011, 11, 8809–8823. [38] Rogge W F. Organic marker compounds for surface soil and fugitive dust from open lot dairies and cattle feedlots. Atmospheric Environment, 2006, 40, 27–49.

Reponen T. Personal exposure to airborne dust and micro-

[39] Suriadikusumah A, Nugraha W, Nurlaeny N, Devnita R.

organisms in agricultural environments. Journal of Occu-

Effect of different mixed media (merapi volcanic ash, cow

pational & Environmental hygiene, 2007, 3, 118–130.

manure and mineral soil) on chemical properties of soil and

[27] Alizadeh M R, Mahdavi A H, Rahmani H R, Jahanian E.

growth of maize (zea mays l.). Journal of Agricultural Sci-

Effects of different levels of clove bud (syzygium aromati-

ence, 2013, 5,188–196.

cum) on yolk biochemical parameters and fatty acids profile,

[40] Purdy C W, Clark R N, Straus D C. Analysis of aerosolized

yolk oxidative stability, and ovarian follicle numbers of

particulates of feedyards located in the southern high plains

laying hens receiving different n-6 to n-3 ratios. Animal

of Texas. Aerosol Science & Technology, 2007, 41,

Feed Science & Technology, 2015, 206, 67–75.

497–509.

[28] Cortés-Rojas D F, Souza C R F D, Oliveira W P. Clove

[41] Ying Q, Kleeman M J. Source contributions to the regional

(syzygium aromaticum): A precious spice. Asian Pacific

distribution of secondary particulate matter in California.

Journal of Tropical Biomedicine, 2014, 4, 90–96.

Atmospheric Environment, 2006, 40, 736–752.

[29] Wang F, Xiong S, Li H, Li L, Zhang Q, He M. Study on

[42] Raupach M R, Woods N, Dorr G, Leys J F, Cleugh H A. The

dust-retention ability of major afforestation tree species in

entrapment of particles by windbreaks. Atmospheric Envi-

Tianjin airport economic area. Journal of Arid Land Resources and Environment, 2015, 29, 101–104. (in Chinese)

ronment, 2001, 35, 3373–3383. [43] Judd M J, Raupach M R, Finnigan J J. A wind tunnel study

[30] Wang L, Gong H, Liao W, Wang Z. Accumulation of parti-

of turbulent flow around single and multiple windbreaks,

cles on the surface of leaves during leaf expansion. Science

part I: Velocity fields. Boundary-Layer Meteorology, 1996,

of the Total Environment, 2015, 532, 420–434.

80, 127–165.

400

Journal of Bionic Engineering (2017) Vol.14 No.2

[44] Středová H, Podhrázská J, Litschmann T, Středa T,

[47] Yan J, Lin L, Zhou W, Ma K, Pickett S T A, Steward T A. A

Rožnovský J. Aerodynamic parameters of windbreak based

novel approach for quantifying particulate matter distribu-

on its optical porosity. Contributions to Geophysics and

tion on leaf surface by combining SEM and object-based

Geodesy, 2013, 42, 213–226.

image analysis. Remote Sensing of Environment, 2016, 173,

[45] Li Y, Krahn J, Menon C. Bioinspired dry adhesive materials and their application in robotics: A review. Journal of Bionic Engineering, 2016, 13, 181–199. [46] Barthlott W, Neinhuis C. Purity of the sacred lotus, or escape from contamination in biological surfaces. Planta, 1997, 202, 1–8.

156–161. [48] Tsutsui O, Sakamoto R, Obayashi M, Yamakawa S, Handa T, Nishio-Hamane D, Matsuda I. Light and SEM observation of opal phytoliths in the mulberry leaf. Flora, 2015, 218, 44–50.