See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/318318534
Development of a Two-stage Impaction Based Bioaerosol-cum-particulate
Matter Sampler and its Use for Ambient Aerosol Profile
Article · July 2017
CITATIONS
READS
0
519
2 authors:
Amit Singh Chauhan
Tarun Gupta
Indian Institute of Technology Kanpur
Indian Institute of Technology Kanpur
6 PUBLICATIONS 5 CITATIONS
240 PUBLICATIONS 2,666 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Ambient Particle Health Effects View project
Bio-aerosols View project
All content following this page was uploaded by Tarun Gupta on 10 July 2017.
The user has requested enhancement of the downloaded file.
SEE PROFILE
Chauhan
/ Journal
of Energy and Environmental
Journal&ofGupta
Energy
and Environmental
Sustainability,Sustainability,
3 (2017) 1-9 3 (2017) 1-9
1
Journal of Energy and
Environmental Sustainability
Journal homepage : www.jees.in
Development of a Two-stage Impaction Based Bioaerosol-cum-particulate
Matter Sampler and its Use for Ambient Aerosol Profile
Amit Singh Chauhan1, Tarun Gupta1, 2*
1
2
Department of Civil Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India
Atmospheric Particle Technology Laboratory, Centre for Environment Science and Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India
ARTICLE
INFO
Received : 08 February 2017
Revised
: 07 April 2017
Accepted : 15 May 2017
Keywords:
Air sampling device, Impactor,
Bioaerosols, Particulate matter
ABSTRACT
Air contains a mixture of various biological and non-biological components. During breathing, human
body does not segregate between these components. Studies have provided sufficient proof about the
deteriorating effects of degrading air quality on human health. The deterioration in ambient air quality and
the emergence of new pollutants in developing countries is also giving rise to the requirement of simple,
customized and economical air sampling devices. Relatively simple design, and low cost as compared to
real time optical instruments have led to the development of various kinds of impactors. The present study
presents a low volume (Q = 12 LPM), two stage (PM10 and PM0.6), impaction based device that can be used
for bioaerosol and particulate matter sampling. The device is fabricated of brass and chrome plating is done
over it. This device eliminates the need for two different sampling devices to be bought for bioaerosol and
particulate matter sampling separately, which is of huge economical advantage for developing nations.
Developed device is tested in laboratory conditions and field conditions as well. Laboratory testing was
carried out for parametric characterization and optimization of the device. Yearlong sampling for PM0.6
and bioaerosols as Gram positive bacteria, Gram negative bacteria and fungi was carried out within the IIT
Kanpur campus and the results were observed in reference to similar studies in the region.For the first time,
yearlong bioaerosol inventory was generated for the IIT Kanpur campus.
© 2017 ISEES, All rights reserved
1. Introduction
Deterioration in air quality and its effects on human health are one of
the main concerns of the present time. Poor ambient air quality results in
the degradation of health and is significantly related to increase in mortality
around the globe (Brunekreef and Holgate, 2002; Evans et al., 1984;
Hall, 1996; Kampa and Castanas, 2008; Özkaynak and Thurston, 1987).
Traditionally, air quality parameters include the gaseous contents, heavy
metals, persistent organic pollutants and particulate matter present in the
ambient air (Kampa and Castanas, 2008). The air human breath-in is a
mixture of various pollutants, gases and several other species. The dose
and duration of various pollutants present in inhaled air may vary
substantially. This exposure may result in breathing discomfort, various
respiratory disorders and other related health issues. The problems due to
air pollution may range from allergies, chronic discomfort, disabilities,
and even cancerous end. (Kampa and Castanas, 2008). Air pollution
affects not only respiratory system but it may also lead to problems in
cardiovascular, digestive or urinary system (Järup, 2003; Kuo et al.,
2006; Nawrot et al., 2006; Riediker et al., 2004; Vermylen et al., 2005).
Traditional air quality measurement relies more on sampling of respirable
suspended particulate matters and compositions of these in terms of heavy
metals, ions, carbon contents and related parameters. However, fine
particles are thought to be more toxic to human beings than comparatively
coarse particles due to their higher surface area per unit mass (Harrison
and Yin, 2000; Kumar and Gupta, 2015). Further, fine particulate matters
can penetrate deep into the human body as compared to coarse particulate
*
Corresponding Author: tarun@iitk.ac.in
© 2017 ISEES All rights reserved
matters. Particulate matter of aerodynamic diameter less than 0.6 µm
may penetrate deep into alveolar region. Sub PM0.6 particles can enter
deeper into the body by diffusion into blood. This study focuses on the
fact that the air we breathed in contains a mixture of gases, organic
species, adsorbed transition metals and bioaerosols present in the form of
particulate matter. The traditional parameters of PM measurement lack
measurement of bioaerosols. The term bioaerosol is derived from two
words biological and aerosols (Nazaroff, 2016). Thus, bioaerosols may
be defined as aerosols with biological origin. They include microorganisms
like bacteria, fungi, protozoa or viruses. Bioaerosol may be cultivable,
non-cultivable, living or dead. Bioaerosols may also include by-products
of living beings like allergens and endotoxins etc. (Cox and Wathes
1995). They may be plant products like pollens, plant debris or fragments
from avian or mammalian bodies (Macher and Macher, 1997).
Bioaerosols are responsible for various diseases like aspergillosis,
pneumonia, severe acute respiratory syndrome (SARS), and whooping
cough etc. (Castellani Pastoris et al., 1997; Fernandez, 2012). They may
spread by voluntary activities like aeration of contaminated water or by
involuntary processes like sneezing, wheezing, coughing etc. If a
bioaerosol is a pathogenic microorganism it can lead to death of the
exposed organism and if it contain toxins like aflatoxin it may lead to
cancer in the exposed individual (Douwes, 2003). Many qualitative as
well as quantitative studies have reported about bioaerosols and their
effects in the different occupational and indoor environments. Bioaerosol
itself is a subset of the aerosol. Therefore, inhalation path and settling of
bioaerosols into the human respiratory tract depends on their size.
2
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
Thus, there is a need to study the concentrations and composition of
both fine particulate matter and bioaerosols present in the ambient air.
This needs an efficient low cost air-sampling device that can be used to
collect both fine particulate matter as well as bioaerosols present in the
ambient air. Relatively simple design, and low cost as compared to real
time optical instruments have led to the development of various kinds of
impactors. The deterioration in ambient air quality and the emergence of
new pollutants in developing countries is also giving rise to the requirement
of customized impactors (Marple, 2004; Singh et al., 2010).
This study was carried out with a goal to develop a two-stage impaction
based air sampling device that can be used for both PM sampling as well
as bioaerosol sampling. Stage I of the device has cut off diameter of 10
µm and stage II has cut off diameter of 0.6 µm. The device developed
during the study was tested rigorously under laboratory and field
conditions. Device was tested using Teflon filter paper for PM 0.6 and
PM10 under laboratory conditions. PM0.6 sampling was also done in the
field conditions using the device. Along with this, the device was tested
for bioaerosol sampling using different types of agar media as per the objective.
Different agar media were used as impaction substrate during the laboratory
testing as well as field sampling using this air sampling device. During the
study nutrient agar, Mac Conkey agar, Mannitol salt agar and Sabouraud
Dextrose Agar with Chloramphenicol were used for collecting bioaerosols.
All the agar media were purchased from Himedia Labs.
2. Development of two-stage air sampling device
2.1 Design of an impactor
An impactor is a simple device through which air passes around the
impaction plate such that a sharp change in the trajectory of airflow
around it leads to impaction of the higher inertia particles on the impaction
plate. Finer particles with less inertia follow the path of sharply bending
air, streamlines downstream and eventually get collected on the backup
filter. Design equations for inertial impactors have been worked out of
the empirical solutions of Naviers-Stokes equations for various air flow
conditions around the different forms of impaction plates (Marple and
Willeke, 1976; Mcfarland et al., 1978).
Dimensionless Stoke’s number is used to calculate the cut point of the
round nozzle for impactor.
The Stoke’s number is defined as:
(1)
Where ρp is the particle density (kg/m3), dp stands for diameter of the
particle (µm), U stands for the air jet velocity through the impactor nozzle
(m/s), Cc is Cunningham slip correction factor, η is the dynamic viscosity
of air (Pa.s) and Dn is the diameter of the round nozzle in the nozzle plate
(Baron and Willeke, 2001).
Cunningham slip correction factor is calculated by the following equation:
(2)
Here, P is the absolute pressure above the impaction nozzle (Baron
and Willeke, 2001).
Numerical analysis of Navier-Stokes equation gives critical design
parameters for designing of the round nozzle impactors (Marple and Liu,
1974).Theoretical cut point for collecting particles with 50% efficiency
can be calculated using equation 1 for
= 0.5 for the desired flow
rate through the specified nozzle diameter.
Equations used for calculating the theoretical d50 and Reynold’s number
are (Hinds, 1982):
(3)
(4)
Where ρ air is the density of air in g/m3
d50
particle diameter for 50% collection efficiency
Q
flow rate in m3/s
Five important concerns for designing an impaction plate and choosing
the impaction substrate for a bioaerosol sampler are:
i. Particle bounce-off from the impaction substrate.
ii. Overloading of bioaerosols on the impaction substrate.
iii. Losses of bioaerosols in the flow path within the instrument.
iv. Viability and cultivability of sampled microorganisms and
v. Stability of impaction media during sampling.
The impactor presented in this study has been tested using a thick
(3.5 mm) and smooth coating of high vacuum silicone grease as well as
different types of culture media. The culture media is used as impaction
substrate for capturing and also for cultivating different microorganisms
(Demokritou et. al., 2001; Kumar & Gupta, 2015).
Equations 1 to 4 were used for designing the round impactor nozzles
for PM10 and PM0.6 stages. Stable flow of polydisperse talcum powder
aerosols was generated and maintained with the help of a dry aerosol
generator working in the form of a venturi (aspirator, Bernouli’s principle)
(Gupta et. al., 2011). An Aerodynamic Particle Sizer (APS, Model 3021,
TSI Inc.), which is based on the time-of-flight principle was used for
characterizing the impactor nozzles as per the protocols well established
by previous researchers (Demokritou et. al., 2002; Gupta et. al. 2004;
Gupta et al., 2011). Aerosol Particle Sizer was used due to its advantage
of directly giving the aerodynamic diameter. Thus, the variation in the
density of the talcum powder can be ignored. For each stage, several
parametric experiments were done to assess the best flow rate at desired
cut off point for the respective stages.
2.1.1
Laboratory performance evaluation of the designed two stage
impactor based sampler
Impactor nozzles for each of the two stages were tested at different
flow rates ranging from 7.5 (litres per minute) LPM to 15 LPM. Identical,
isokinetic sampling probes were used to sample the particles for five
minutes from upstream of the impaction stage and alternatively with five
minute sampling from down-stream of the respective stage (figure 1).
Each experiment continued with at least four such cycles for each stage.
The efficiency of each stage was calculated by comparing the measured
particle concentrations at up-stream and down-stream of the respective
stage.
The impaction plate in each stage was characterized by using high
vacuum silicone grease as well as alternatively using different types of
agar media (Nutrient agar, Sabouraud Dextrose Agar with
Chloramphenicol, MacConkey agar, Mannitol salt agar). The selected
agar media were to be used for bioaerosol sampling experiments.
2.1.2
Optimization of sampling duration for various substrates used for
bioaerosol sampling using the developed air sampling device
Experiments were carried out to optimize the sampling duration on
different substrate that is media in the Air Sampling Device. Sampling was
carried out on each of these selected media for 1, 2, 3, 5, 7 and 10
minutes respectively in different sets of experiments. Each experiment
was repeated at least 7 times for each media and selected time duration.
After sampling, the plates were incubated as per the protocol described in
section 3.3. After incubation, the number of colony forming units per
cubic meter of the sampled air was calculated for each sample.
3. Field performance evaluation of the sampler
3.1. Study area
Field performance evaluation was carried out inside the main campus
of Indian Institute of Technology, Kanpur (IIT Kanpur or IITK). The IIT
Kanpur (26.4°N, 80.2°E) is a residential academic campus located 15 km
north of the industrial town Kanpur. Kanpur is a major industrial hub in
the Indo-Gangetic region. Kanpur city is 142 m above the mean sea level.
The Indo-Gangetic region is one of the most polluted regions in the
world. The region has high air pollution especially high particulate matter
concentrations (Tripathi et al., 2006). Aerosol load in the region is
contributed by not only the anthropogenic sources but also natural sources
such as dust. Dust storms are quite prevalent in the region, especially
during summers (Ghosh et al., 2014).
The study site is located at a distance of 1.5 km from a national
highway (N.H. 91). IIT Kanpur campus is at the upwind direction of the
city and has no major emission source nearby (Sharma and Maloo,
2005). In general, outdoor environment at IIT Kanpur has rich and
flourishing vegetation in the form of trees, plants and grasses.
The climate of Kanpur can be divided into five distinctly marked
seasons. They are pre-summer (March to April), summer (May to June),
monsoon (July to August), post monsoon (September to October) and
winter (November to February). The average temperature in Kanpur
varies from nearly 8 ºC in winter to ~ 40 ºC in summer. It may dip to
below zero degrees Celsius during peak of winter. In summer, this may
elevate to even greater than 45 ºC on several days. Relative humidity at
Kanpur also varies from <40% to > 80% on several days. This results in
two extreme weather conditions- dense fog during winter and dry dust
storms during summer.
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
!
"
#
$
%
&
'
(
)
&
*
+
,
-
.
&
+
/
0
%
1
2
'
3
4
3.2. Aerosol sampling
Particulate matter sampling was conducted on 47 mm diameter
Whatman® PTFE Membrane filter paper. Pore size of the filter paper
was ≤ 0.20 µm. Filter papers were handled with the help of clean forceps.
Each filter paper was preconditioned at 25 °C and 45% – 55% humidity
for 12 hours (Chakraborty and Gupta, 2010). Each filter paper was
weighed before the sampling and the pre-weight of the filter paper was
recorded. The filter paper weighing was carried out on a microbalance
(Mettler, Toledo) (Chakraborty and Gupta, 2010). Filter papers were
kept in clean custom-made plastic cassettes. All the cassettes were put in
a vacuum sealed container. This vacuum sealed container was stored in a
refrigerator at -4 °C temperature.
10% of the filter papers were retained as field blank. For this, every
eleventh filter paper was taken out as a blank filter paper. These blank
filter papers were also analysed in the same way as the other sampled filter
papers.
The same process of weighing the filter papers was repeated after air
sampling and the post sampling weights of the filter papers were recorded.
Pre-sampling weight of the filter paper was subtracted from post sampling
weight of the filter paper so as to calculate PM0.6 mass deposited over it.
The same procedure was followed for the field blanks. The mass difference
of field blanks were subtracted from the mass difference of sampled filter
papers so as to get the net mass of PM0.6 deposited over the filter paper.
This was divided by the volume of air sampled through the device so as
to obtain airborne concentration of PM0.6. Post sampling, these filter
papers were stored as described earlier and were used for further analysis
at a later time.
During air sampling campaign, two co-located air sampling devices
were used for ambient air sampling. One of the devices was used to
collect PM 0.6. The other device was used to collect viable bioaerosol
samples on different media. The flow rate through the Air Sampling
Device was maintained at 12 LPM by the use of vacuum pump, needle
control valve and a rotameter. Rotameter was pre-calibrated using a digital
mass flow meter. Sampling for PM0.6 was done for 6 hours and total 4.32
m3 of air was sampled at each location. Air flow rates through the device
were checked continuously at an interval of 30 minutes (Gupta and
Mandariya, 2013). Sampling for bioaerosol was done for a period of 3 to
4 minutes as decided by experiment described in section 2.2.2.
3.3. Bioaerosol sampling
In parallel to PM0.6 sampling, two sets of bioaerosol samplings were
performed at the same location. One set was performed at the start of the
,
'
,
3
+
&
'
(
5
,
+
(
2
*
2
1
+
4
&
+
6
2
(
)
%
,
3
+
(
2
*
.
+
3
,
7
&
.
sampling by the device housing filter paper and another was performed
after 5 hours of the start of sampling. Deferring of ± 15 minutes from the
sampling schedule was considered normal. Sampling at each of the
microenvironment was repeated after 15 days using the same protocol.
Major steps involved in the bioaerosol sampling were:
i. Disinfection of Air Sampling device and other glassware
ii. Preparation of Sampling Cassettes
iii. Preparation of sampling media (Nutrient agar, Sabouraud
Chloramphenicol Agar, MacConkey agar, Mannitol salt agar)
iv. Ambient air sampling at the site
v. Incubation of sampled cassettes
vi. Cell colony-counting and microscopic identification
Before each sampling, air-sampling device was properly cleaned. It
was disinfected by using 70% ethanol, absolute ethanol and methyl alcohol
(Merck KGaA, Germany). All the processes were carried out inside a
bio-safety cabinet (Biogen Scientific). After this, the sampler and all its
components were exposed to U.V. radiations for 20 minutes inside the
bio-safety cabinet. All the non-consumable glassware used in bioaerosol
sampling and subsequent processes was autoclaved at 15 lbs. and 121 ºC.
35 mm- radiation sterile Petri dishes were used as sampling cassettes for
bioaerosol sampling. Proper care was taken in preparing culture media as
per protocols described in Microbiology Laboratory Manual (Cappuccino,
2005). Previously described culture media were prepared as per the
instructions on the label of each media package. Sampling cassettes were
prepared by pouring media into Petri dishes in a bio-safety cabinet and
allowing the media to solidify by cooling at room temperature.
Bioaerosol sampling was done for 3 or 4 minutes per sample at 12
LPM of airflow. After sampling, the Petri dishes were taken out of the
sampler and stored in sterilized transportation cassettes meant for
transporting the sampling cassettes to and fro the sampling site and
laboratory. All the sampling cassettes were then transferred to the incubator
(Esco Technologies, Inc.).
The Petri dishes were kept for incubation at 35°C for 24 to 48 h for
incubation of bacterial colonies. For incubation of fungi, the plates were
incubated at 35°C for 5 days. After cultivation of the plates, the total
bacterial and fungal colonies were counted as per the protocol mentioned
in microbiology laboratory manual (Cappuccino, 2005). Cell colony
counting was done with the help of magnifying glass as per the established
methods in Microbiology Laboratory Manual (Cappuccino, 2005). Whole
colony appearance of each bacteria colony was recorded as one of the
following categories -circular, irregular, biconvex (embedded in agar),
filamentous or rhizoid (long projections). Along with this, the margin
(entire, undulate, lobate or curled) and elevation (flat, raised, convex or
4
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
umbonate) of the colony were also recorded. An inoculums loop was
sterilized until red hot on a Bunsen flame. After cooling, this was used to
prepare a smear of colony on a slide by using distilled water. The inoculums
from each colony was subjected to negative staining first and then Gram
staining as per the protocols described in Microbiology Laboratory
Manual (Cappuccino, 2005). Stained bacterial cells were viewed at 4x,
10x, 40x and oil immersion 100x magnification. Slides were viewed with
Nikon E 200 upright microscope mounted with a CCD camera make
DS-Fi2. The live view and images were analysed with the help of imaging
software NIS Elements D from Nikon Instruments Inc.
Inoculums from each fungus colony were picked with the help of
sterilized and cooled inoculums loop, and placed in cavity of depression
slide. Sample was stained with lacto-phenol cotton blue. The stained slide
was observed in an inverted microscope (Trinocular Inverted microscope
XDS 2, Optika SRL, Italy) with 4x, 10x, 20x and 40x objectives. The
live images were viewed with the camera Optika pro 5 and the software
Optika Vision Pro 4.3 Pro5. Morphological feature as conidia,
conidiophores and hyphae were used for identification of fungal growths
in the culture.
Concentration of bioaerosols was reported in terms of the Colony
Forming Unit per cubic meter of air in the concerned microenvironment.
The equation 5 was used for this purpose.
Colony Forming Units, CFU/m3 = CFUPetri X (1000/Vs)
device in the given microenvironment and CFUPetri is the number of
Colony Forming Units counted in corresponding Petri-dishes.
4 Developed Air-Sampling Device
A good air-sampling device is the foremost requirement in sampling
studies. The developed air-sampling device and its internal components
are shown in figure 2. Brass was used for the fabrication of the developed
air-sampling device, due to its advantage in machining, long-term durability,
stability and inert characteristics. It was further treated with chrome plating
techniques to cover it up with an inert and corrosion-free layer.
The device has three zones. Top of the sampling device is provided
with the rain cover.
Zone 1 is for collection of large particles and spores that have settled
down due to gravity.
Zone 2 has two stages with cut off points of PM10 and PM0.6 achieved
at an airflow rate of 12 LPM. The first stage is PM10 stage. It consists of a
nozzle plate containing four round nozzles and one impaction plate.
Impaction plate can accommodate four 35 mm Petri dishes in the
respective slots below the nozzles. Second stage is PM0.6 stage. It consists
of one nozzle plate with single round nozzle and one corresponding
impaction plate. This impaction plate can house one unit 35 mm Petri
dishes.
Zone 3 has provision to house a 47mm filter. This zone can collect
PM0.6 particulate matter on filter paper that can be subjected to further
gravimetric and chemical analysis.
(5)
Here, Vs is the total volume of air sampled through the sampling
8
9
(
'
:
,
)
%
-
(
*
7
;
&
Impactor characterization results show that stages 1 and 2 in zone two
of the Air Sampling Device have cut off points at PM10 and PM0.6 respectively (figure 3). PM losses in the nozzle were checked without the use of
impaction plates behind the nozzles and it was always found to be below
5%. The sharp cut off points at 0.6 µm and 10 µm make this device suitable
for collecting the particles in the range of 0.6 µm and 10 µm (table 1).
Impaction plate at first stage collects the particle with size greater than
10 µm. Particles of size less than 10 µm travel down towards the second
stage. The impaction plate on the second stage collects the particles of size
0.6 µm to 10 µm.
Next step was to test the performance of this device for collecting
bioaerosols. For this, it was desired to find out the optimum duration of
air sampling to collect maximum number of bioaerosols on selected
nutrient media.
4.1. Sampling duration for various substrates used for bioaerosol sampling
using the developed Air Sampling Device
The main criteria for selecting the optimum duration was the occurrence
of maximum number of colony forming units per cubic meter of sampled
<
(
3
&
,
*
=
(
+
.
(
*
+
&
'
*
,
-
3
2
)
%
2
*
&
*
+
.
air (CFU/m3) with the minimum standard deviation in repeated set of
experiments. It was decided that Nutrient agar and Sabouraud Dextrose
Agar with Chloramphenicol would be subjected to 3 minutes of sampling
per sample. MacConkey agar and Mannitol salt agar were giving best
results at 4 minutes of sampling duration. After this time, overcrowding
of colonies was observed in the incubated plates and this overlapping of
growing colonies inhibits the accurate colony counting.
5 Aerosol profile for IIT Kanpur
5.1. PM0.6 profile for IITKanpur
Maximum concentration of PM0.6 was observed during winter season
(204 ± 56.2 µg/m3). Meteorological conditions during winter resulted in
settling of the PM near the ground. The concentrations of PM started to
decrease with the approach of summer (84.7 ± 27.8 µg/m3). The monsoon
rainfall resulted in washing out of the PM from atmosphere and thus the
concentration was found to be lowest during the season (33.01 ± 9.27 µg/
m3). Post monsoon the PM concentrations again started to increase in the
ambient air (73.7 ± 2.82 µg/m 3). The PM 0.6concentrations obtained
during the campaign were compared to that with PM1 concentrations ob-
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
>
J
K
%
+
(
)
(
5
,
+
(
2
>
*
2
?
1
@
+
4
4
&
,
'
.
,
,
)
3
+
%
&
'
-
(
(
*
5
,
7
+
(
=
2
0
*
2
'
,
+
1
(
.
2
*
+
,
1
7
2
&
'
served at IITK during previous studies (Chakraborty and Gupta, 2010)
(table 2). PM0.6 and PM1 concentrations are almost similar in winter. The
pre-summer concentrations of PM0.6 (102.24 ± 37.97 µg/m3) was higher
than the PM1 concentration observed during the previous study (77.1 ±
31.0 µg/m3) whereas the PM0.6 concentration during summer (84.7 ± 27.8
µg/m3) was lower than the PM1 concentration observed during previous
study (142.3 ± 45.0 µg/m3). The previous study was done at a height of 12
m from the ground whereas the present study is conducted at a height of
A
L
B
(
2
C
D
,
&
E
'
2
F
G
.
H
,
2
-
*
=
.
,
.
)
+
%
,
-
7
(
&
*
A
7
A
2
B
*
C
D
<
I
,
'
(
E
H
2
(
0
.
)
%
(
)
,
3
%
+
,
5
2
'
3
+
.
(
2
*
.
0
L
.
+
'
,
+
&
.
0
.
&
=
only 1.5 m from the ground. Summer is a season of suspension of dry soil
and dust by the wind. Pre summer resulted in the suspension of fine
particles that were loosely bound to the surface whereas summer resulted in
relatively lesser concentration of fine PM at lower heights due to their
escape to higher levels from the ground surface. Similar concentrations for
PM0.95 were obtained by another study at IITK during winter. In this study
during last decade, the concentrations were found to be 203 µg/m3 (Tare et
al., 2006).
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
6
M
?
J
N
!
@
4
,
'
,
3
+
&
'
(
.
+
(
3
.
2
1
=
&
<
&
-
2
%
&
=
,
(
'
.
,
)
%
-
(
*
7
=
&
<
(
3
&
5.2 Bioaerosol profile for IIT Kanpur
Parameters
No. of Zones
Zone 1
Flow rate
Zone 2: Stage I
Number of nozzles
Nozzle diameter
Zone 2: Stage II
Number of nozzles
Nozzle diameter
Zone 3
Height
Diameter
GSD
∆P (Pressure drop) in Impactor with Teflon
paper (47 mm diameter and 0.2 m pore size)
M
?
J
N
8
3
2
:
*
3
&
&
,
*
.
+
'
2
*
,
,
+
(
-
2
,
*
<
&
,
+
'
,
A
7
A
&
O
C
P
D
E
,
F
*
%
G
,
0
Features
3
Settling Zone
12 LPM
10 µm and above
4
5 mm
0.6 - 10 µm
1
3.2 mm
PM sampling
5 inches
3 inches
1.28
5.2.1
18.79 cm of H2O
*
=
'
C
D
7
B
Q
I
)
)
R
,
.
.
H
µ
Season
PM0.6 concentration
PM1 Concentration
Winter
Pre Summer
Summer
Monsoon
Post Monsoon
204 ± 56.2
102.24 ± 37.97
84.7 ± 27.8
33.01±9.27
73.7±2.82
199.0 ± 66.0
77.1 ± 31.0
142.3 ± 45.0
30.1 ± 13.7
63.8 ± 23.7
B
U
C
V
/
U
S
'
,
?
)
T
C
&
2
.
-
(
+
,
+
(
<
(
<
&
&
V
(
,
*
=
3
+
2
&
2
'
'
(
3
,
The maximum concentration of fungi was recorded in the month of
September (370 ± 57 CFU/m3, n=6) followed by April (315 ± 16 CFU/
m 3 , n=6) and October (306 ± 28 CFU/m 3 , n=6). This shows that
maximum fungi concentration was during post monsoon and pre-summer
seasons. The concentration of fungi in the ambient air decreased with the
onset of winter and summer and the concentration was minimum during
the months of July (111 ± 0 CFU/m3, n=6) followed by that in January
(148 ± 16 CFU/m3, n=6). The total bioaerosol load in the ambient air of
WL 116 was highest (884 CFU/m3, n=6) during the post-monsoon season
Indoor bioaerosol profile
Indoor bioaerosol sampling was done at Environmental Engineering
Laboratory at IIT Kanpur. The concentrations of both Gram positive and
Gram-negative bacteria were lowest in the month of December. During
December, the GPB concentration was 83 ± 21 CFU/m3 (n=6) and GNB
concentration was reported as 28 ± 12 CFU/m3 (n=6). The concentrations
of GPB increased until April (271 ± 21 CFU/m3, n=6). The GNB
population reached a peak concentration in the month of March (97 ±
12 CFU/m3, n=6). This shows that concentrations of both GPB and
GNB bacteria were lowest during winter and increased until the onset of
summer. With the drastic temperature rise during summer the GPB (167
± 21 CFU/m3, n=6) and GNB concentration (63 ± 20 CFU/m3, n=6)
decreased during the summer months. The onset of monsoon triggered
the bacterial growth and the GPB load at WL 116 increased to 373 ± 18
CFU/m3 (n=6) whereas, GNB load increased to 186 ± 19 CFU/m 3
(n=6) during the month of October. The onset of winter again resulted in
the decrease of bacterial concentration from November onwards. The
concentration of GNB with respect to GPB was highest during summer
end (June) and the onset of monsoon (July). During this period, the
actual bioaerosol load was also on the lower side and it was only higher
than the bioaerosol load during the winter. During the winter, actual
concentrations of GPB (83 CFU/m3 ) and GNB (28 CFU/m3) were
lowest in the sampling duration and the relative population of GNB was
minimum (18% to 23%) during the entire sampling duration. This was
just opposite to what was observed during the summer where the bioaerosol
load was low, but GNB to GPB ratio was highest of the whole year (30%
to 37%) (figure 4a).
,
2
*
*
3
=
&
U
*
+
W
'
,
V
+
(
/
2
U
*
'
.
,
2
)
1
W
U
C
&
7
V
,
,
+
(
<
*
&
=
U
V
,
W
3
V
+
&
'
(
,
H
and was minimum (278 CFU/m 3 , n=6) during the winter season.
September (884 CFU/m3, n=6) and October (865 CFU/m3, n=6) had
maximum bioaerosol levels followed by March (613 CFU/m3, n=6) and
April (683 CFU/m3 , n=6). The months of December (278 CFU/m3,
n=6), January (294 CFU/m3, n=6) and July (313 CFU/m3, n=6) had a
lowest bioaerosol load in the indoor ambient air at WL 116. It is important
to note that despite similar total bioaerosol load, the composition of
bioaerosols was quite different during those months.
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
5.2.2
S
J
A
*
=
2
2
'
L
(
2
,
&
'
2
.
2
-
%
'
2
1
(
-
&
B
U
C
V
/
U
'
,
)
C
2
.
(
+
(
<
&
V
Outdoor bioaerosol sampling was done in backyard of Environmental
Engineering laboratory. The lowest concentration of GPB was observed
during the winter with concentrations of (70 ±12 CFU/m3, n=6) during
January and (70 ±12 CFU/m3, n=6) during December. The lowest
concentration for GNB was observed in the month of December (14 ±12
CFU/m 3 , n=6). Outdoor GPB concentration increased until early
summer (April- 264 ±12 CFU/m3, n=6 and May, 264 ±32 CFU/m3 ,
n=6). Although the concentration of GPB in May was equal to that in
April, the standard deviation was slightly higher in May. This was due to
a sharp decline in GPB concentration during early May (292 CFU/m3)
to 229 CFU/m3 during late May. This decline in concentration continued
until early monsoon (July 132 ± 32 CFU/m3, n=6). GPB concentration
started to increase with the onset of monsoon reaching a maximum
concentration of (299 ± 12 CFU/m3, n=6) during late September and
early October. With the onset of winter, decrease in temperature lead to
a decrease in GPB concentration reaching a minimum during December.
Concentration of GNB was minimum during the winter (December 21 ±
21 CFU/m 3, n=6 and January, 14 ± 12 CFU/m3, n=6). Favourable
conditions with tree sheds, gardening and other anthropogenic activities;
and absence of direct hot winds lead to the increase in concentration of
Z
?
T
&
-
,
+
(
<
&
2
0
+
=
2
2
'
3
2
*
3
&
*
+
'
,
+
(
2
*
.
2
1
U
C
V
,
*
=
U
W
V
3
+
&
'
(
,
X
U
W
V
/
U
'
,
)
W
&
7
,
+
(
<
&
V
,
3
+
&
'
(
,
,
*
=
Y
0
*
7
(
H
GNB after winter until May (111 ± 12 CFU/m3, n=6). After this there
was a decline in GNB concentration due to the intense summer (July- 49
± 12 CFU/m3, n=6). The concentration of GNB started to increase with
the monsoon season and the favourable conditions resulted in a higher
value of 132 ± 24 CFU/m3 (n=6) during October.
Although, June was the month of low bacterial load, the percentage of
GNB was maximum (37%) during this month. This might be because of
watering, gardening, manure applications and environmental engineering
laboratory related activities at backyard. Similar results were observed in
the month of October (36%) when the GNB concentration was maximum
during the year. GNB vs. GPB ratio was lowest during the winter (and
early pre-summer) as GPB population increased rapidly during this period
as compared to GNB at backyard during these months (figure 5a).
The outdoor fungi concentration was low as compared to indoor
microenvironment of Environmental Engineering Laboratory. This
suggests towards the presence of indoor fungal sources at WL116. Further,
the fungi concentration in September (306 ± 28 CFU/m3, n=6) is higher
than GPB concentration (292 ± 21 CFU/m3, n=6) at backyard. The
fungi concentration at backyard was higher than GPB from the onset of
monsoon and until winter. This might be due to fungal growth on trees,
woody scrap materials, gardening practices and other materials lying
there.
Outdoor bioaerosol profile
,
7
B
U
C
V
/
U
'
,
)
C
2
.
(
+
(
<
&
V
,
3
+
&
'
(
,
,
*
=
U
W
V
/
U
'
,
)
W
&
7
,
+
(
<
&
V
,
3
+
&
'
(
,
H
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
8
B
U
C
V
/
U
'
,
)
C
2
.
(
+
(
<
&
Z
V
,
3
J
+
K
&
'
(
0
,
+
X
=
U
2
2
W
'
L
V
/
(
U
2
'
Bioaerosol based indoor air quality assessment
Ratio of indoor bioaerosol concentrations to that of outdoor bioaerosol
concentrations was calculated to assess the indoor air quality during the
study period (figure 6). Values above 1 indicate towards the indoor source
of concerned bioaerosol in the ambient air. The controlled indoor
microenvironment provides ample habitat for bioaerosols. The controlled
temperature and thus humidity allows bioaerosols to flourish indoor. A
B
U
C
V
/
U
'
,
)
C
2
.
(
+
(
[
<
&
V
V
(
,
3
2
+
,
&
&
'
'
(
,
2
X
.
2
U
-
W
&
'
)
2
.
W
2
&
-
7
%
,
+
'
(
2
<
1
&
(
-
&
V
,
3
+
&
'
(
,
,
*
=
Y
0
*
7
(
H
low indoor bioaerosol concentration as compared to outdoor concentration
suggests that indoor has a cleaner and safe microenvironment. However
an opposite trend suggests that bioaerosols source is present indoor.
Extreme environmental conditions may lead to elimination of such
sources. The figure shows that GPB source was persistent throughout the
year at Environmental Engineering Laboratory except for the month of
May, June and August. The high ratio for GNB during January, February
and March depicts the presence of rich source of GNB in the indoor
environment. This source was persistent except for the peak summer
months of May and June. Similarly, the indoor source of fungi was
evident in the post monsoon season and continued to survive until the
pre-summer period.
The total outdoor bioaerosol load at Environmental Engineering
Laboratory backyard shows that despite the varying composition, the
bioaerosol load was maximum in the month of September (708 CFU/
m3) and April (685 CFU/m3) and winter has minimum bioaerosol load
(Jan, 149 CFU/m3 and Dec 248 CFU/m3) (figure 5b).
5.2.3
,
,
L
V
,
.
/
U
&
=
'
(
,
)
*
=
W
2
2
&
'
7
9
,
(
+
(
<
'
/
&
\
0
V
,
,
-
3
(
+
+
&
]
'
(
,
,
*
=
Y
0
*
7
(
H
Chauhan & Gupta / Journal of Energy and Environmental Sustainability, 3 (2017) 1-9
Very few studies on the bioaerosol concentrations in the ambient air
are available from central India. One of such study about the concentration
of Fungi in a library and a bakery at Gwalior city situated in central India
is done using an Andersen two-stage sampler. The authors used Sabouraud
Agar Media for measuring the fungi concentration in the ambient air.
Indoor concentration of < 300 CFU/m3 was obtained in the library and
> 200 CFU/m3 was obtained in the bakery (Jain, 2000). In one of the
rare study on bioaerosols done in Indo-Gangetic region, the authors
reported the bioaerosol concentration at Agra (27°102 N, 78°052 E). In
this study, researchers used Sabouraud dextrose agar media for fungi and
nutrient agar media for bacteria sample collection. They reported total
microbial count in the range of 125 CFU/m3 to 725 CFU/m3. Total
bacteria concentration and fungi concentration in PM10 sampling during
the study was 72.5 to 625 CFU/m 3 , and 62.5 to 104.2 CFU/m 3
respectively (Mamta et al., 2015).
n
o
i
b
u
x
v
_
u
x
v
_
u
x
v
_
c
_
j
~
_
`
j
c
i
e
s
i
b
g
s
_
i
b
c
Ambient air quality has direct as well as indirect effect on health of
living organism as well as environment. There is a need to understand
various biological as well as non-biological components present in the
ambient air. For a developing nation like India, availability of inexpensive
multi purpose air sampling device can help in saving money and also in
serving science and society. Present study fulfils the need for such sampler.
The developed sampler can be used for PM 0.6, PM 10 and bioaerosol
sampling.
The results of samples from developed sampler for PM0.6correlate
well with those obtained during previous studies at IIT Kanpur. Bioaerosol
concentrations of the samples collected from this sampler were also in
comparable range as obtained in previous studies done in Central India
and Indo-Gangetic Plain (IGP) region. Further, this is the first of its kind
of study that reports bioaerosol profile for IGP region. This study also
reports for the PM0.6 and bioaerosol profile at breathing zone of individuals.
This study presents bioaerosol as well as PM0.6 inventory and seasonal
trends for a relatively clean academic campussituated within a highly
polluted city in IGP.
¨
`
m
`
x
m
u
b
c
_
c
a
c
_
y
z
_
k
t
`
v
c
_
_
`
x
k
_
`
x
j
_
`
y
x
}
j
_
`
a
v
a
b
c
d
e
f
e
_
g
h
i
j
j
k
l
k
c
m
e
n
o
p
p
q
r
f
k
`
a
s
a
j
t
k
_
s
u
`
k
t
k
b
v
w
x
`
i
b
y
i
x
j
k
s
c
v
k
y
z
b
i
{
u
k
s
c
_
b
x
x
j
i
y
_
v
i
a
b
s
c
h
i
j
k
|
d
b
s
b
j
~
l
u
k
a
y
k
}
}
_
t
i
k
v
k
z
`
f
b
}
`
s
v
i
v
z
t
k
`
e
t
r
i
v
s
a
j
u
x
_
j
y
t
_
b
a
s
v
k
b
i
k
y
b
`
i
v
s
t
_
i
s
v
s
m
i
_
a
b
b
s
x
c
u
`
f
w
k
`
b
a
s
s
i
a
j
z
v
f
i
`
f
y
t
_
i
j
v
_
b
z
k
v
}
i
}
¦
c
l
i
d
^
s
_
a
g
a
t
i
b
_
v
k
g
e
b
a
j
g
d
`
b
}
i
k
a
i
`
`
a
}
a
`
b
e
a
p
c
w
t
z
_
_
j
p
c
c
i
o
k
_
b
_
y
v
i
k
a
b
`
j
x
c
u
z
s
k
|
e
y
_
b
z
e
t
}
p
b
g
a
b
k
`
b
a
`
i
b
v
c
`
k
_
_
z
b
c
u
p
x
`
e
}
_
v
s
j
a
b
u
`
a
}
_
k
`
v
k
e
y
q
t
i
j
k
p
b
q
v
e
a
}
e
`
y
_
i
o
_
c
a
v
`
e
e
t
a
_
a
i
v
¡
b
y
e
k
b
s
v
b
v
v
k
a
a
_
z
c
t
a
k
i
k
`
t
k
y
a
g
_
k
l
a
v
z
b
b
i
e
y
`
`
q
k
k
o
i
u
t
y
b
g
x
b
j
c
a
_
b
f
v
c
`
g
c
c
s
j
t
a
p
a
b
§
a
b
`
b
i
k
¦
v
a
k
j
s
o
q
a
a
c
_
`
e
k
}
v
i
z
i
`
`
v
v
j
a
k
v
k
y
c
v
_
b
i
k
k
j
`
b
r
v
u
v
_
k
b
_
k
j
a
e
v
k
k
v
z
e
j
x
k
k
a
b
z
`
x
i
`
s
q
g
a
a
`
j
_
¡
k
p
_
y
x
b
c
}
u
`
`
k
}
k
y
k
c
t
a
y
i
p
d
k
x
b
g
m
u
v
w
k
b
c
k
~
v
|
o
i
i
_
m
c
i
e
k
`
t
v
¢
_
j
y
i
r
k
y
j
_
|
}
b
r
}
u
u
n
i
g
u
a
y
a
q
x
u
j
e
}
v
q
b
s
j
d
_
s
i
}
a
c
`
p
v
j
a
s
v
_
v
a
d
i
j
y
k
k
b
l
o
`
a
s
e
m
m
_
s
_
i
k
j
i
k
`
z
g
x
b
s
e
a
j
i
k
`
^
j
`
s
u
u
v
a
t
q
j
i
b
c
j
a
e
i
q
b
y
q
c
b
s
o
v
o
c
b
c
c
b
k
i
o
`
k
o
a
b
e
d
`
_
j
e
q
b
v
o
a
j
j
u
v
e
c
v
i
k
k
e
_
b
w
~
a
g
_
x
s
g
a
x
t
k
b
b
v
k
_
`
v
i
b
_
g
j
i
t
k
x
j
_
g
y
v
a
_
`
j
f
u
s
_
s
v
k
i
a
b
t
a
j
}
|
_
n
~
u
j
f
v
r
i
x
c
j
f
k
k
`
j
a
i
s
v
a
j
_
b
j
g
u
b
k
k
y
c
_
~
b
e
y
k
n
o
`
p
x
p
_
v
i
r
k
f
b
v
y
s
y
i
u
b
t
u
_
i
j
_
v
_
i
b
a
b
c
a
e
}
y
a
z
`
i
a
y
t
a
i
j
e
u
t
_
b
i
b
`
g
a
b
e
n
q
a
s
_
y
`
`
y
_
i
`
k
`
i
v
ª
a
`
a
s
c
^
i
y
a
_
k
`
`
p
i
a
s
j
a
_
r
a
j
i
q
q
f
j
t
v
d
s
q
e
}
i
o
y
a
c
n
s
k
e
e
i
c
z
b
v
|
}
_
t
x
j
k
`
d
k
`
}
a
`
t
_
b
y
k
c
e
t
k
i
b
v
y
s
a
u
b
_
a
a
i
m
v
s
s
b
v
v
e
k
x
c
e
_
e
x
v
e
b
i
j
s
u
b
v
_
s
k
e
s
`
©
s
t
k
k
k
s
v
e
i
b
c
v
o
t
a
o
x
_
}
c
y
v
i
p
a
`
a
p
s
_
k
`
c
a
b
s
p
a
p
i
j
`
e
a
b
e
e
p
¦
`
`
b
k
j
b
i
v
y
_
s
`
s
c
b
b
f
v
_
k
j
`
i
©
a
s
a
k
e
a
j
q
y
j
i
`
k
b
v
b
q
b
`
y
k
_
e
`
b
g
i
s
v
k
u
y
j
o
g
e
c
_
q
s
t
q
e
i
c
i
s
t
k
b
g
c
v
a
_
j
q
y
c
`
y
a
_
o
a
y
_
c
_
k
c
z
e
v
p
a
e
u
c
b
b
s
y
b
i
z
c
a
b
a
v
y
s
x
_
`
y
k
z
k
i
k
j
k
j
c
b
a
y
e
|
~
g
e
k
k
c
u
s
j
i
k
s
a
i
t
g
t
v
e
k
a
k
v
_
a
i
s
f
d
`
e
_
c
r
f
b
b
i
b
a
s
~
r
i
k
q
c
c
_
g
n
`
s
p
j
`
e
k
j
p
u
a
j
k
o
x
v
e
a
n
a
y
h
t
f
x
_
c
t
z
s
v
c
i
y
u
`
k
r
`
a
a
y
k
t
k
x
_
v
c
j
_
g
b
z
_
u
p
b
i
j
_
_
y
_
e
j
i
t
^
`
k
d
r
t
g
k
e
c
x
k
o
y
r
q
e
v
x
a
e
e
c
i
}
a
_
a
j
t
c
i
v
z
e
v
`
q
b
z
e
o
_
y
k
o
a
k
a
m
i
b
e
q
|
c
e
`
n
o
f
`
k
e
_
b
`
}
n
b
e
a
c
|
s
_
l
e
k
v
k
c
c
y
j
_
b
c
b
j
i
s
o
v
y
i
s
e
~
i
_
`
~
_
e
`
i
c
i
v
¦
e
e
k
d
e
_
e
~
e
k
a
b
a
k
r
t
t
a
e
g
^
r
u
c
`
u
e
`
¡
b
q
q
v
i
t
u
_
t
k
i
¢
n
s
v
a
b
p
j
¦
k
c
h
{
y
v
b
`
_
p
e
e
u
`
l
}
i
y
¢
d
z
a
p
a
z
f
c
g
s
k
q
c
b
b
c
c
y
o
b
u
i
q
c
e
c
j
j
`
p
|
e
u
e
b
g
o
e
g
¡
_
e
_
i
i
_
p
b
k
j
o
y
`
e
k
`
_
v
r
y
`
k
z
k
|
z
n
z
i
k
j
_
e
b
e
f
k
a
n
}
b
_
c
p
a
a
}
v
u
t
v
p
}
^
e
e
q
c
o
a
_
a
a
n
c
i
b
e
g
n
j
_
s
s
c
b
e
i
x
_
b
z
a
f
c
_
c
`
`
f
g
}
_
g
c
e
u
v
e
s
a
y
b
e
b
a
_
e
z
_
s
i
j
v
i
i
u
e
t
|
_
_
j
a
a
b
a
u
`
k
v
i
}
9
b
i
a
m
a
`
j
h
a
`
j
}
b
i
r
i
a
p
a
g
k
_
s
a
_
~
~
g
b
k
j
y
_
a
e
y
c
j
i
f
k
k
`
e
i
v
v
a
c
`
_
s
e
_
b
y
v
k
c
_
s
b
g
v
`
`
|
a
_
_
x
k
f
k
k
j
u
c
k
e
a
q
b
o
}
k
n
v
e
s
e
e
`
e
c
i
j
j
b
z
_
e
u
e
a
p
j
k
a
v
b
c
x
y
_
_
e
u
y
k
j
^
t
`
s
i
b
f
i
i
_
t
z
j
j
b
z
c
e
a
_
a
j
v
x
e
j
k
j
x
x
l
_
u
v
~
i
g
k
y
r
o
b
j
h
k
f
`
i
g
_
e
~
e
i
p
s
i
r
u
a
s
_
i
j
b
c
x
`
k
s
t
k
q
_
v
o
z
k
g
s
c
i
b
¥
k
k
`
s
o
`
r
k
b
¦
b
p
_
c
f
e
f
`
_
p
¤
k
n
e
i
g
y
`
d
~
_
e
u
c
u
_
£
f
x
u
v
¦
b
z
s
_
x
p
s
c
_
g
_
a
_
b
s
q
v
}
f
_
k
c
e
a
b
l
r
k
p
o
q
f
u
n
i
i
j
d
n
o
_
¡
m
i
`
References
_
g
v
`
n
_
e
e
`
_
`
|
_
_
g
i
m
a
y
g
`
a
e
_
t
k
`
k
e
q
¡
^
`
x
e
x
m
x
q
e
i
c
t
a
i
l
i
z
b
`
v
b
x
h
e
_
a
n
c
c
f
u
_
j
e
b
_
c
k
a
x
~
e
a
a
_
c
_
_
j
s
g
a
i
j
_
k
q
b
k
v
a
y
y
c
t
_
y
e
t
k
u
i
s
`
_
i
j
k
`
c
_
a
g
~
e
v
i
u
`
t
a
k
g
k
k
b
v
z
k
e
_
b
z
j
e
i
r
v
_
c
a
u
c
q
v
6. Summary
p
_
_
`
a
}
}
c
h
e
h
e
n
o
p
q
r
b
a
`
i
a
_
k
`
a
s
a
j
g
|
b
_
t
i
y
s
c
b
g
a
a
`
f
i
`
c
o
c
q
e
e
«
^
`
u
b
k
l
`
k
k
}
c
^
e
_
b
g
~
a
j
_
v
k
c
e
e
n
o
p
p
o
r
f
i
`
x
a
j
j
u
v
i
a
b
_
b
g
z
k
_
j
v
z
e
c
_
b
y
k
v
c
p
c
q
o
o
l
_
|
b
_
l
_
t
k
s
e
_
x
x
u
y
y
i
b
a
n
o
p
p
r
i
y
`
a
i
a
j
a
|
w
f
_
a
`
_
v
a
`
|
_
b
u
_
j
c
k
c
d
k
_
`
s
a
_
g
u
y
_
v
i
a
b
_
s
v
k
j
j
_
b
i
d
_
s
v
j
a
`
i
s
c
e
c
i
y
k
`
a
b
i
c
e
c
a
a
b
_
y
a
c
e
c
a
j
g
a
b
i
c
d
e
c
k
b
v
a
`
k
c
^
e
c
j
k
a
c
e
_
v
v
_
b
i
c
e
c
i
_
`
`
a
y
y
z
i
c
e
c
d
i
b
v
a
c
f
_
b
g
i
s
y
_
c
d
e
n
q
r
a
j
k
y
u
j
_
`
k
x
i
g
k
t
i
a
j
a
}
_
i
k
g
i
l
k
`
c
e
b
a
u
v
`
k
_
l
a
}
k
i
a
b
b
_
i
`
k
s
g
i
s
k
_
s
k
_
s
s
a
y
i
_
v
k
g
i
v
z
_
y
a
a
j
i
b
v
a
k
`
i
b
k
b
a
_
k
s
v
`
i
d
a
b
k
b
v
k
c
v
_
j
|
e
c
u
`
e
e
j
i
b
e
i
y
`
a
i
a
j
e
b
}
k
y
v
e
i
s
e
c
q
c
o
z
_
l
`
_
a
`
}
v
|
c
u
f
e
t
_
i
b
y
g
`
a
u
b
n
x
d
v
_
c
r
e
n
f
o
k
p
`
q
a
p
s
r
a
j
z
i
k
b
t
m
i
_
y
b
_
x
j
u
z
`
_
`
k
_
y
i
v
a
k
`
b
c
i
a
u
_
b
v
g
i
i
a
b
_
_
c
b
f
g
k
`
a
a
s
u
a
`
j
y
k
f
i
f
`
x
x
u
a
_
`
j
v
e
i
a
b
t
k
s
k
k
b
e
b
c
q
t
a
l
`
i
v
a
u
c
d
`
_
a
`
v
i
t
u
j
v
c
m
a
_
_
a
u
`
_
s
c
e
e
c
k
`
u
s
a
b
c
e
e
_
b
g
m
v
`
_
l
i
s
c
d
e
n
o
p
p
q
r
k
k
j
a
x
t
v
_
b
z
_
`
t
_
t
a
k
l
y
`
z
i
v
b
a
`
b
a
u
|
k
j
c
i
b
z
_
a
u
a
_
k
s
`
b
e
d
a
c
d
_
_
k
u
e
c
v
a
`
s
c
}
a
c
`
k
`
k
e
n
u
|
e
o
t
_
b
y
k
_
j
u
_
v
i
a
b
a
}
_
d
k
`
s
a
b
_
j
u
j
v
i
x
a
j
j
u
v
_
b
v
_
t
x
j
k
`
}
a
_
b
s
`
c
a
s
e
x
c
_
s
q
x
d
f
p
k
y
a
v
k
s
p
v
s
`
b
_
a
b
j
g
j
k
u
v
c
i
s
s
v
a
`
a
u
`
_
}
c
a
y
r
c
k
e
^
f
s
k
`
c
i
b
a
b
e
t
k
b
v
s
a
}
d
_
`
v
i
y
u
j
_
v
k
_
b
g
_
s
k
a
u
s
d
a
j
j
u
v
_
b
v
s
c
f
k
`
a
s
a
j
y
i
a
s
z
_
c
a
e
a
c
`
e
b
b
i
o
e
t
x
k
`
u
s
a
b
c
e
_
b
g
m
a
u
v
`
_
l
i
s
c
d
e
n
o
p
p
o
r
k
k
j
a
x
t
k
b
v
_
b
`
_
l
_
s
z
_
_
b
|
k
y
j
`
e
k
a
a
y
l
o
a
v
c
_
`
k
c
y
i
_
s
p
v
v
s
a
i
z
b
i
c
k
b
e
k
e
_
v
n
s
g
u
c
q
`
k
s
c
r
a
f
}
s
f
i
s
`
a
y
a
q
i
`
_
b
v
i
a
k
b
d
s
_
^
`
v
i
k
y
j
v
k
k
k
a
b
b
q
y
k
b
v
p
`
¢
_
v
i
e
a
b
e
c
a
i
s
`
l
v
_
j
f
i
b
_
v
|
j
e
c
e
a
k
c
h
j
e
i
b
c
e
e
`
^
i
e
n
o
s
p
c
p
e
r
e
d
~
_
`
k
v
i
`
y
u
j
s
v
u
b
_
c
v
k
e
t
e
_
v
v
^
k
`
`
a
k
t
x
k
a
s
`
u
c
`
d
k
i
e
f
b
e
y
¡
_
`
k
s
i
_
s
s
c
_
s
e
s
a
h
y
i
i
_
j
v
j
k
i
_
t
g
s
i
c
v
z
y
u
j
_
`
k
}
}
k
y
v
s
i
b
z
k
_
j
v
z
|
|
a
t
k
b
c
f
t
e
e
k
s
x
i
`
e
`
i
v
e
_
`
k
k
g
e
c
q
c
e
_
y
v
b
k
g
`
i
_
_
v
i
a
j
a
b
a
a
c
}
d
e
n
o
i
p
b
p
v
z
r
k
f
y
i
s
v
s
|
k
a
s
}
s
t
m
k
_
b
b
x
v
u
a
`
}
c
_
b
t
g
i
_
i
c
k
f
b
v
v
_
t
a
i
s
`
e
d
b
_
i
`
a
b
b
g
e
d
c
_
c
b
p
g
q
e
e
_
`
s
y
b
_
p
s
q
s
o
_
a
u
j
j
x
g
k
s
r
~
e
m
s
f
~
i
t
i
u
`
k
_
k
|
|
b
k
`
_
g
k
c
j
e
a
c
}
z
_
s
v
b
b
a
a
a
e
c
^
z
_
v
v
u
v
`
a
c
_
_
`
x
i
t
e
_
_
v
b
j
v
i
y
a
i
z
b
s
a
c
}
}
}
_
o
k
d
c
y
v
q
k
o
s
`
_
s
a
b
p
b
_
q
g
j
o
`
_
_
c
j
^
k
e
y
m
a
e
b
}
m
i
z
u
_
b
e
`
b
_
c
v
i
f
a
e
b
m
y
a
_
s
v
y
z
_
_
g
j
l
k
_
i
`
t
c
d
x
e
_
m
y
e
v
a
_
`
b
}
g
a
`
_
_
k
`
|
a
|
s
_
a
c
j
¦
s
i
e
e
k
n
g
i
o
s
p
v
q
`
i
p
r
u
v
i
k
a
b
k
j
t
k
a
x
_
t
s
k
u
`
b
k
v
t
a
k
}
b
v
a
s
e
b
i
`
a
c
j
s
k
i
k
`
v
g
c
c
p
o
c
e
`
i
x
_
s
y
_
g
k
t
x
_
y
v
a
`
c
e
f
i
`
h
_
e
c
c
d
c
e
q
n
q
o
p
r
p
`
c
b
k
b
v
`
_
i
b
`
a
s
b
e
t
i
s
s
b
i
v
a
e
c
b
x
x
i
i
c
b
¡
e
y
c
a
u
_
u
z
`
c
c
f
e
e
c
b
s
p
s
c
}
}
k
i
a
s
^
y
s
`
e
c
i
k
g
i
c
f
e
m
e
c
`
_
v
z
_
i
j
c
c
e
f
¡
e
e
m
i
z
s
i
z
b
a
b
`
_
k
t
c
c
¡
e
e
_
`
j
i
c
_
s
e
v
^
_
e
_
_
c
b
f
e
g
m
e
z
_
`
k
|
t
c
_
c
e
_
e
n
b
o
_
p
`
p
c
e
r
m
_
k
b
_
s
_
u
`
_
k
g
t
k
k
c
b
v
e
s
d
e
a
}
s
v
v
t
a
s
x
z
k
`
i
y
x
_
`
_
t
k
v
k
`
s
g
u
`
i
b
b
g
i
_
b
x
_
y
k
k
s
k
_
`
y
z
`
_
b
i
_
v
i
a
b
k
a
s
x
z
k
`
k
k
^
i
z
a
s
x
z
k
`
k
d
`
a
`
_
t
_
b
g
_
t
x
_
i
b
_
v
_
v
|
x
i
y
_
j
j
a
y
_
v
i
a
b
i
b
v
z
k
_
b
_
^
_
s
i
b
w
o
s
u
`
k
f
s
s
k
s
s
t
k
b
v
w
d
`
a
`
k
s
s
_
b
k
t
i
y
_
j
x
`
a
x
k
`
v
i
k
s
c
e
k
a
x
z
|
s
e
k
s
e
c
q
q
q
c
o
o
q
p
e
e
g
`
i
x
_
v
z
i
c
e
¡
e
_
`
k
c
e
e
m
z
i
b
b
_
t
c
¡
e
`
i
_
s
v
_
_
c
f
e
m
e
k
|
c
e
f
_
`
_
j
c
f
e
m
i
s
z
a
`
k
c
e
_
j
c
e
a
q
a
_
e
k
y
v
i
a
b
_
j
t
a
`
v
_
j
i
v
|
s
v
u
g
i
k
s
_
b
g
_
i
e
a
`
g
k
v
v
_
e
c
i
f
e
s
e
c
j
j
o
`
e
m
_
u
j
c
_
x
p
i
e
_
b
_
`
c
_
b
_
g
k
c
e
d
e
z
_
u
z
_
b
c
e
e
e
z
_
`
t
_
c
e
k
g
g
|
c
e
e
a
x
_
j
c
m
e
e
¡
_
`
_
s
i
t
z
u
j
u
c
m
e
k
g
g
|
c
e
e
e
u
x
v
_
c
e
_
b
g
_
j
c
e
n
o
p
p
r
k
_
s
u
`
k
t
k
b
v
s
a
}
e
k
c
^
`
k
`
v
u
s
s
i
s
g
k
t
_
u
a
b
s
v
`
_
v
k
g
i
b
v
t
x
_
c
v
z
p
i
c
q
e
¡
p
e
e
c
_
j
_
i
c
a
s
x
z
k
`
i
y
x
_
`
_
t
k
v
k
`
s
g
u
`
i
b
b
g
i
_
b
x
_
y
k
k
s
k
_
`
y
z
`
_
b
i
_
v
i
a
b
k
a
s
x
z
k
`
k
_
i
a
s
x
z
k
`
k
d
`
a
`
_
t
t
k
_
b
g
_
t
x
_
i
b
_
v
_
v
|
x
i
y
_
j
j
a
y
_
v
i
a
b
i
b
v
z
k
_
b
_
_
s
i
b
w
e
c
d
j
c
e
c
_
`
_
c
k
c
e
e
_
b
g
i
s
z
`
e
m
e
d
z
|
s
i
y
_
j
_
b
g
a
x
v
i
y
_
j
x
`
a
x
k
`
v
i
k
s
c
e
k
a
x
z
|
s
e
k
s
e
c
q
q
q
c
o
o
p
e
c
k
`
t
|
j
k
b
c
e
¡
k
t
t
_
`
c
f
e
¡
k
t
k
`
|
c
^
e
_
b
g
~
a
|
j
_
k
`
v
s
c
e
e
n
o
p
p
r
f
t
i
k
e
_
View publication stats
s
_
i
g
d
`
b
j
e
c
_
a
b
c
s
n
t
u
_
o
v
q
z
`
e
^
s
g
e
`
_
k
a
e
x
_
b
u
k
`
d
i
_
p
c
_
g
c
z
f
k
z
v
u
g
_
f
d
e
e
_
j
j
c
_
g
f
j
v
p
k
e
e
h
_
y
a
b
g
k
e
b
_
_
l
|
y
a
e
s
y
c
~
k
_
e
v
b
^
c
e
y
u
v
k
t
|
a
y
_
`
g
i
_
j
i
b
}
_
`
y
v
i
a
b
c
e
z
`
a
t
e
~
_
k
t
a
s
v
e
c
c
q
q
e
b
v
_
i
`
x
a
j
j
u
v
i
a
b
_
b
g
