Full text of DOI:10.1080/10473289.2000.10464136

This article was downloaded by: [University of Massachusetts, Amherst]
On: 20 March 2013, At: 03:58
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of the Air & Waste Management Association
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uawm20
Characterization of Activated Carbon Fiber Filters
for Pressure Drop, Submicrometer Particulate
Collection, and Mercury Capture
Toshiaki Hayashi a , Tai Gyu Lee a , Melynda Hazelwood a , Elizabeth Hedrick a & Pratim
Biswas a
a Aerosol & Air Quality Research Laboratory, Environmental Engineering & Science
Division, University of Cincinnati, Cincinnati, OH, USA
Version of record first published: 27 Dec 2011.
To cite this article: Toshiaki Hayashi , Tai Gyu Lee , Melynda Hazelwood , Elizabeth Hedrick & Pratim Biswas (2000):
Characterization of Activated Carbon Fiber Filters for Pressure Drop, Submicrometer Particulate Collection, and Mercury
Capture, Journal of the Air & Waste Management Association, 50:6, 922-929
To link to this article: http://dx.doi.org/10.1080/10473289.2000.10464136
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising
directly or indirectly in connection with or arising out of the use of this material.

Hayashi et al.
TECHNICAL PAPER
ISSN 1047-3289 J. Air & Waste Manage. Assoc. 50:922-929
Copyright 2000 Air & Waste Management Association
Characterization of Activated Carbon Fiber Filters for Pressure
Drop, Submicrometer Particulate Collection, and Mercury Capture
Toshiaki Hayashi, Tai Gyu Lee, Melynda Hazelwood, Elizabeth Hedrick, and Pratim Biswas
Aerosol & Air Quality Research Laboratory, Environmental Engineering & Science Division, University of
Cincinnati, Cincinnati, OH
ABSTRACT
gaseous com poun ds,3 th eir poten tial for particulate m at-
ter an d Hg con trol h as n ot been well establish ed.
Th e use of activated carbon fiber (ACF) filters for th e cap-
ture of particulate m atter an d elem en tal Hg is dem on -
strated. The pressure drop and particle collection efficiency
ch aracteristics of th e ACF filters were establish ed at two
differen t face velocities an d for two differen t aerosols:
sph erical NaCl an d com bustion -gen erated silica particles.
The clean ACF filter specific resistance was 153 kg m -2 sec-1.
Th e experim en tal specific resistan ce for cake filtration was
1.6 × 106 sec-1 an d 2.4 × 105 sec-1 for 0.5- an d 1.5-µm m ass
m edian diam eter particles, respectively. Th e resistan ce
factor R was approxim ately 2, sim ilar to th at for th e h igh -
efficien cy particulate air filters. Th ere was a discrepan cy
in th e m easured particle collection efficien cies an d th ose
predicted by th eory. Th e use of th e ACF filter for elem en -
tal Hg capture was illustrated, an d th e breakth rough ch ar-
acteristic was establish ed. Th e capacity of th e ACF filter
for Hg capture was sim ilar to oth er powdered activated
carbon s.
Several experim ental and theoretical studies have been
carried out exam ining the pressure drop and particle col-
lection characteristics of fibrous filters. The pressure drop
and collection efficiency predictions for clean fibrous fil-
ters agree well with experim ental data.4 The velocity field
for flow past a system of parallel cylinders was solved by
Kuwabara.5 Fuchs and coworkers6-8 experim entally studied
m odel filters to com pare results with the theoretical solu-
tion by Kuwabara.5 Lee and Liu9,10 theoretically and experi-
m entally studied the collection efficiency of fibrous filters
and developed a filtration theory for particle capture by
diffusion and interception. Particle collection by different
m echanism s on single fibers and subsequent dust loading
has been extensively studied. The particle collection char-
acteristics in the interception region were studied by Wang
et al.11,12 and the interception, diffusion, and inertia regions
were studied by Payatakes et al.13-18 Kanaoka et al.19-21 stud-
ied the m echanism s of interception, diffusion, and inertia
on resultant particle capture efficiencies.
INTRODUCTION
Activated carbon fibers (ACFs) are processed by th e car-
bon ization of syn th etic fibers such as rayon or acrylic
m aterials1 an d h ave a h igh adsorption capacity for gases.
Filters h ave previously been m ade by assem blin g a m at of
ACFs. Such filters h ave foun d application in th e rem oval
of gaseous pollutan ts an d are curren tly used for th e cap-
ture of toxic gases em an atin g from in dustrial processes
an d gasolin e vapors from autom obile tan ks.2 Alth ough
th ese filters h ave been exten sively used for capture of
Th e pressure drop ch aracteristics caused by particle
loadin g are m ore difficult to predict on a firm th eoretical
basis. Most research ers h ave reported th at th e form of
deposit on a fiber depen ds on filtration con dition s an d is
difficult to predict. As a result, th e studies of pressure drop
ch aracteristics h ave been prim arily em pirical.22,23
Activated carbon powders h ave been effectively used
for th e capture of air toxics such as m ercury in com bus-
tion an d oth er exh austs.24,25 Two m eth odologies h ave
been proposed: in jection in to th e flue gas followed by
capture on a down stream bag filter, or th e use of a packed
bed geom etry. An altern ative approach would be to use
ACFs in th e form of filter m ats, as used in th is work.
Sorben t in jection m eth odologies in various form s h ave
also been proposed,26-30 an d th e ACF filters can be used
to capture th e sorben t particles. As th is work proposes
to exam in e particulate m atter collection by ACF filters,
pressure drop is an im portan t ch aracteristic th at n eeds
to be studied.
IMPLICATIONS
The availability of activated carbon as fibers and fibrous
filters increases the potential for their use in industrial
systems. This paper establishes the pressure drop and
particle capture characteristics of ACF filters. Furthermore,
their use in fibrous filter form for the capture of vapors
such as Hg offers an effective alternative in comparison
with powder injection or packed bed systems.
922 Journal of the Air & Waste Management Association
Volume 50 June 2000

Hayashi et al.
In th is paper, th e particulate m atter filtration ch arac-
teristics of an ACF filter are reported. Th e clean filter pres-
sure drop is establish ed as a fun ction of th e flow rate. Th e
pressure drop ch aracteristics of th e ACF filter are th en es-
tablish ed as a fun ction of dust loadin g. Th e particle col-
lection efficien cies of th ese filters for sph erical an d
agglom erate particles are establish ed. Th e capture of el-
em en tal m ercury (Hg) by ACF filters is exam in ed, an d th e
breakth rough curves are reported.
pressure drop as a function of mass loading (mass collected/
area of filter) was then determined. Two different atomizer
solution concentrations were used, and the mass median
diam eters (MMDs) were m easured by an optical particle
counter to be 0.5 and 1.8 µm approximately, consistent with
MMDs reported in the literature.33 A fixed facial velocity of
5 cm/sec was used. The experiments were conducted for dif-
ferent time durations, after which the filter was weighed to
determine the mass loading. The collection efficiency of the
filter was then established as a function of particle size. Two
aerosols were used: an atomized NaCl solution and a high
tem perature silica aerosol gen erated by oxidation of
hexamethyldisiloxane.27,28 When the experiments were con-
ducted with the polydisperse aerosol, the size distribution
upstream and downstream of the filter was measured using
a differential mobility analyzer-condensation particle counter
(DMA-CPC) combination. For the silica test aerosol case, the
high-temperature generated particles were classified using a
DMA, and the particle concentrations upstream and down-
stream were measured using a CPC.
EXP ERIMENTS
Three different system s were used in this study. First (Fig-
ure 1a), the pressure drop and the particulate capture char-
acteristics of the ACF filter were exam ined. Second, the Hg
capture characteristics of the ACF filter were exam ined us-
ing a system sim ilar to that used by Wu et al.30 (Figure 1b).
Th e ACFs were prepared by a carbon ization an d acti-
vation process, usin g cellulose fibers as th e precursor
m aterial. Th e tem perature ran ged from 100 to 1300 K,
an d details are provided in an oth er report.31 Th e result-
an t pores on th e surface h ad a diam eter of approxim ately
2 nm . Th e oth er ch aracteristics of th e prepared ACF fil-
ters are listed in Table 1. Th e filters were th en cut in to 72-
m m -diam eter disks an d placed in a filter h older for testin g
(Figure 1a). Th e pressure drop an d collection efficien cy
were m easured as a fun ction of th e flow rate. Two differ-
en t aerosols were used: sph erical NaCl particles gen erated
by atom ization (TSI, In c., Model 3076), followed by dif-
fusion dryin g an d an electrical m obility classified (TSI,
In c., Model 3071) com bustion -gen erated silica aerosol.32,27
Th e objective was to exam in e th e beh avior with sph eri-
cal particles (NaCl) an d agglom erated particles (SiO2).
A series of experim ents were perform ed and are listed
in Table 2. The clean filter pressure drop characteristics were
established by m easuring the pressure drop across the fil-
ter for facial velocities varying from 0.4 to 30 cm /sec. The
The system used to exam ine the capture of Hg by the
ACF filter is shown in Figure 1b. Air at a flow rate of 0.1 L/
m in was passed through an Hg bubbler placed in a con-
stant tem perature bath (Figure 1b). This was m ixed with a
clean air stream to obtain a total rate of 1 L/m in. A reactor
(60 cm long, 5 cm in diam eter) was used to ensure that the
Hg vapor was well m ixed in th e air stream an d passed
through a 76-m m -diam eter ACF filter. A series of im pingers
were then connected downstream of the filter to trap the
Hg n ot collected on th e ACF filter. Th e first sam plin g
im pinger contained a 0.1 M HNO3 solution and was used
to trap th e oxidized form s of Hg (HgO in th e experi-
m ents).34,35 A set of three other im pingers containing 0.4%
KMnO4 and 1.0% H2SO4 was used to trap elem ental Hg in
the gas phase.36 The concentration of KMnO4 was selected
to m inim ize any interferences in the Hg m easurem ent.
For all experim en ts, th e last im pin ger captured less
th an 1.0% of th e total m easured Hg, providin g assuran ce
th at all th e Hg was bein g trapped in th e im pin gers. Th e
im pin ger solution s were an alyzed to determ in e th e
Table 1. Characteristics of the ACF filter.31
2
Specific surface area (m /g)
1,450
1.80
0.70
0.80
17.4
2.56
0.103
3
True density (g/cm )
3
Pore volume (cm /g)
Table 2. List of experiments performed.
3
Apparent density (g/cm )
Fiber diameter, d (µm)
Set
Tested
Air Velocity
(cm/sec)
f
Filter thickness, L (mm)
Aerosols
Packing fraction, α (-)
I. Clean ACF filter pressure drop
-
0.4 to 30
5.0
NaCla
NaClb
Composition (Range)
II. Pressure drop vs. mass loading
III. Collection efficiencies of clean ACF filter
C
O
H
P
S
92–94.5%
2.9–3.5%
0.6–0.8%
1.32%
Ca
Zn
Si
0.16%
0.12%
Mg
K
190 ppm
As
10 ppm
10 ppm
0.7, 3.0, 5.0
5.0
SiO c
70 ppm
70 ppm
30 ppm
10 ppm
Cu
2
940 ppm
770 ppm
240 ppm
Fe
Al
a
Atomization of 10 mg/ml NaCl solution and 100 mg/ml NaCl solution; bAtomization of 1
Na
Cl
mg/ml NaCl solution; cThermal oxidation of hexamethyldisiloxane, 98+%, (CH )
0.2%
Ni
3 3
SiOSi(CH ) .
3 3
Volume 50 June 2000
Journal of the Air & Waste Management Association 923

Hayashi et al.
Figure 1a. Schematic diagram of the experimental systems for testing the ACF filter pressure drop and particle capture efficiency measurements.
Figure 1b. Schematic diagram of the experimental systems for testing the ACF filter Hg capture efficiency measurements.
elem en tal Hg con cen tration an d oxidized Hg species
con centration in the gas phase. Dilutions were usually re-
quired because of high concentrations of Hg (>100 ppb).
In m ost cases, dilutions were large enough to dilute any
excess oxidant, such as KMnO4. For im pingers with low
Hg concentration, m L volum es of a 50% hydroxylam ine
solution were added until the purple color of the KMnO4
disappeared. The am ount of Hg captured on the filter was
determ ined by extraction of the Hg species by acid diges-
tion. Filters were folded and placed in a disposable 50-m L
polypropylene centrifuge tube. A 20% (v/v) HNO3 was added
to the tube. The tube was capped and shaken vigorously for
about 1 m in, and then was placed in a 99 °C hot water
bath for 1 hr. The filter that was then dissolved into a slurry
was centrifuged until a clear supernatant was obtained. Th e
filter digestion was n ot h igh ly oxidative; h owever, th e
adsorbed Hg was tran sferred to th e solution . Th e result-
ant extract was analyzed by a cold vapor atom ic absorp-
tion (CVAA) system to determ ine the total Hg on the filter.
A m ercuryModule (Therm o Separation Products) was used.
The analytical step involves reduction of oxidized Hg to
Hg0, which is then determ ined by CVAA.
924 Journal of the Air & Waste Management Association
Volume 50 June 2000

Hayashi et al.
Th e system in Figure 1c was used to establish th e
breakth rough ch aracteristics of th e ACF filter. As th e ACF
filter h ad a very h igh capacity, an on lin e Hg an alyzer
(UV-1201 Mercury An alysis Un it, Sh im adzu) was used
to m easure th e con cen tration s before an d after th e fil-
ter. Th e in strum en t was calibrated usin g th e im pin ger
system described, followed by m easurem en t with a CVAA
an alyzer.
particles also penetrate deeper into the filter for the sm aller
MMD experim en ts. As a result of th is deeper pen etration ,
on ce th e tran sition to cake filtration h as occurred, th e
pressure drop for th e sm aller MMD aerosol experim en ts
is greater at th e sam e m ass loadin g.
Th e pressure drop in th e cake filtration region is ex-
pressed as33
(1)
RESULTS AND DISCUSSION
Clea n Filt er P r essu r e Dr op a s
wh ere P is th e total pressure drop across a filter, Po is th e
pressure drop across th e clean filter m edium , an d K2C is
th e specific cake resistan ce for a polydisperse aerosol
a Fu n ct ion of Velocit y
Th e pressure drop for a clean filter as a fun ction of th e
facial velocity is plotted in Figure 2. Th e prediction of th e
th eoretical m odel by Kuwabara5 is sh own , an d good agree-
m en t is obtain ed. Th e clean filter specific resistan ce, K1, is
153 kg m -2 sec-1. Because of th e open structure, loosely
packed fibers, an d larger fiber diam eter, K1 is lower th an
th at of h igh efficien cy particulate air (HEPA) filters (7970
kg m -2 sec-1, as reported in Novick an d Higgin s,37 an d 700–
9600 kg m -2 sec-1, as reported in Letourn eau et al.38).
(2)
Th e slope of th e graph in Figure 3 is used to com pute th e
experim en tal values of th e specific cake resistan ce, K2,
wh ich appear in Table 3. Th e ratio of K2 an d th e th eoreti-
cally determ in ed value (eq 2) is kn own as th e resistan ce
factor, R.39 Th ese values are listed in Table 3 an d, usin g
the theoretical expression proposed by Rudnick,39 the cake
porosity is calculated. Th e values of R are sim ilar to th ose
reported by Gupta et al.33 for dry aerosols form in g a cake
on a HEPA filter.
P r essu r e Dr op a s a Fu n ct ion of Du st Loa d in g
Th e pressure drop for th e ACF filter as a fun ction of th e
dust loadin g (M/A) is plotted in Figure 3 for two differen t
MMD aerosols. At th e early stages, th e particles are cap-
tured on th e surface of th e fibers an d fill th e voids. After a
certain tim e, a cake is form ed, an d th e pressure drop in -
creases at a faster rate with in creasin g dust loadin g. As
th e volum e occupied by th e larger particles is greater for
th e sam e m ass loadin g, th e tran sition to cake filtration
occu rs earlier th an th e sm aller MMD aerosol. Th e
Collect ion Efficien cy for Clea n Filt er s
Th e collection efficien cies for th e NaCl aerosol at differ-
en t filtration velocities an d for th e h igh -tem perature-gen -
erated silica aerosol at a fixed filtration velocity of 5 cm /
sec are sh own in Figure 4. Th e NaCl efficien cy data curves
are U-sh aped—Brown ian diffusion is th e dom in an t
Figure 1c. Schematic diagram of the experimental systems for testing the ACF filter breakthrough measurements for Hg capture.
Volume 50 June 2000
Journal of the Air & Waste Management Association 925

Hayashi et al.
Figure 4. Measured collection efficiency for NaCl particles at different
filtration velocities as a function of size. Also shown is the collection
efficiency for silica aerosol as a function of size for a filtration velocity of
5 cm/sec (lines are fitted to measured data points).
Figure 2. Pressure drop as a function of filtration velocity for a clean
ACF filter.
Figure 5. Comparison of the experimental single fiber collection
efficiencies for the NaCl particles (symbols) to the theory of Fuchs et
al.8 (solid lines).
Figure 3. Pressure drop as a function of dust loading (M/A) for two
MMD NaCl aerosols (lines are fitted to measured data points).
wh ere values for α, L, an d df are listed in Table 1. A com -
parison with a th eoretical prediction of Stech kin a et al.8
is sh own in Figure 5, an d th e agreem en t is n ot good. On e
reason for th e deviation could be th at th e th eory is de-
rived for circular cross-section fibers, wh ich is n ot th e case
for ACFs. Th e m easured m in im a in th e collection effi-
cien cy occur at sm aller particle diam eters, possibly in di-
catin g in terception to be relatively m ore dom in an t for
th e irregular cross-section ACFs. An oth er reason for th e
sh ift could be th e variable fiber diam eters.
Table 3. Characteristics of the ACF filter cake.
MMD (µm) α (-)
K (sec-1)
K
(sec-1)
2C
R (-) Calculated
Porosity (-)
g
2
6
5
0.5
1.8
2.0
2.0
1.62 × 10
8.06 × 10
2.01
1.86
0.962
0.970
5
5
2.40 × 10
1.29 × 10
Th e collection efficien cy of th e h igh -tem perature-
generated silica aerosol is sim ilar to that of the spherical
NaCl particles for sizes below 0.1 µm . The silica particles
are clearly agglom erated at larger sizes28 and result in higher
capture efficiencies (com pared with the spherical NaCl par-
ticles) for sizes greater than 0.1 µm . The discrepancy could
m ech an ism for th e sm all particles an d in terception for
th e larger particles. Usin g th e m easured filter collection
efficien cy, th e sin gle fiber collection efficien cy was calcu-
lated usin g th e log-pen etration equation 40
(3)
926 Journal of the Air & Waste Management Association
Volume 50 June 2000

Hayashi et al.
also be caused by an anom aly in the estim ated size by elec-
trical m obility classification of agglom erates. The equiva-
lent electrical m obility radius is a higher estim ate of the
diam eter for agglom erated structures. Further studies (Lee
et al.41) are currently under way to exam ine these factors.
Table 4. Hg capture efficiency (h) by ACF filter over a period of 1 hr.
Hg Feed
Conc.
(µg/m3)
Hg Mass
Hg Mass of
Total Hg Mass η (%)
Trapped in
Hg Collected
(Impingers +
Impingers (µg) on ACF Filter (µg) ACF Filter) (µg)
2a
6.76 × 10
0.394
0.218
39.53
39.924
83.925
99.0
99.7
Effect iven ess of Elem en t a l H g
3b
1.40 × 10
83.707
Ca p t u r e b y ACF Filt er
Th e first test con ducted was to exam in e th e effective-
n ess of th e ACF filter in trappin g Hg at h igh feed con -
cen tration s. Table 4 lists th e results of th e Hg capture
efficien cies for two differen t Hg feed rates— >99% in both
tests for a 1-h r experim en t. Th us, th e ACF filter is poten -
tially as effective as th e in situ sorben t m eth odology de-
veloped by Wu et al.30 Th e use of activated carbon in th e
fibrous filter form is advan tageous to in jectin g activated
carbon powders or use in a packed bed. Th e ACF can be
readily used in bag filter system s an d also h as low pres-
sure drop ch aracteristics. However, n ote th at th e cost of
th e ACF filter process for Hg capture is expected to be
sign ifican tly h igh er th an th e in situ sorben t m eth odol-
ogy developed by Wu et al.30
aHg inlet temperature is 30 °C; bHg inlet temperature is 40 °C.
To establish th e capacity of th e ACF filters, a break-
th rough experim en t was con ducted at 25 °C (Figure 6).
Th e m ass of th e ACF filter was 0.336 g, an d an in let feed
Hg con cen tration of ~40 µg/m 3 was used. Th e effluen t Hg
con cen tration in creased to approxim ately 62% of th e in -
let value after 10 h r of operation , after wh ich it appeared
to stabilize at th is level. Sim ilar observation s h ave been
m ade by Vidic an d McLaugh lin 25 an d Huan g et al.42 Vidic
an d McLaugh lin 25 attributed th is result to Hg escapin g as
HgS; h owever, th ey could n ot con clusively detect an y
sulfides in th eir effluen t stream . Furth erm ore, at som e
poin t all sulfur is expected to be con sum ed, an d HgS can -
n ot be con tin ually form ed (th eir experim en ts were con -
ducted for up to 600 h r). In th e presen t work, th e ACF
filter sam ples con tain ed m uch lower con cen tration s of
sulfur (0.2%) an d, h en ce, HgS probably was n ot form ed.
Alth ough n o detailed studies were con ducted, th e ACFs
probably still h ave som e capacity for Hg capture, th ough
th is occurs at a very slow rate. A careful exam in ation does
in dicate a con tin ued in crease of th e effluen t Hg con cen -
tration at a very slow rate. Otan i et al.24 observed th at th e
effluen t con cen tration reach es in let values after a lon g
duration (~30 days). An oth er possible reason could be th e
presen ce of m etals an d oth er con stituen ts th at catalyze
Hg oxidation . Clearly, m ore detailed studies would n eed
to be perform ed to explain th is observation .
Figure 6. Breakthrough curve for Hg capture of the ACF filter.
excess am oun t of sulfur in a pure n itrogen en viron m en t
at 600 °C) used by Vidic and McLaughlin.25 Note that som e
of th e Vidic an d McLaugh lin25 experim en ts were don e in
n itrogen , wh ereas th e experim en ts in th is work used air
as th e carrier gas. First, th e capacity was defin ed as th e
total m ass of Hg captured un til th e effluen t con cen tra-
tion reach ed 90% of th e fin al value (n ote th at th is is n ot
th e in let con cen tration , as th e fin al stable value was 62%
of th e in let value in th e experim en ts). Clearly, th e ACF
filter is com parable with oth er activated carbon powders
used. However, in creasin g th e sulfur con ten t (F-400S)
clearly in creased th e adsorption capacity in com parison
with th e ACF filter.
P ot en t ia l Ap p lica t ion s
There are several configurations in which ACF filters could
be used in in dustrial system s such as waste in cin erators
or coal-fired boilers. Th e ACFs could be woven an d used
in a bag filter con figuration wh erein particulate m atter
an d elem en tal Hg could be effectively captured with a
low-pressure drop. If sorben t in jection processes were
used,27 th e ACF filters could be used to trap th e sorben ts
an d h eavy m etals, an d con curren tly capture an y rem n an t
Hg in th e gas stream . ACF use in a bag filter form is pref-
erable to in jectin g activated carbon powders (from a cost
Com paring the capacity of different activated carbons
tested un der differen t con dition s is difficult. Table 5,
h owever, is a com parison of th e ACF filter results obtain ed
in th is study to various carbon s (Filtrasorb-400 [F-400],
HGR supplied by Calgon Corp., F-400 con tacted with
Volume 50 June 2000
Journal of the Air & Waste Management Association 927

Hayashi et al.
Table 5. ComparisonofACFfilterforHgcapturewithotheractivatedcarbonadsorber
(25 °C).
Po
R
V
α
pressure drop across clean filter (Pa)
resistan ce factor
air velocity (m sec-1)
a
Types of Activated
Carbon Used
F-400a
HGR
F-400Sa ACF Filter
(This Work)
packin g fraction
η
sin gle fiber collection efficien cy
air viscosity (Pa·sec)
particle den sity (kg m -3)
µ
Sulfur content
0
9.24 %
0.1
7.61 %
0.1
0.2 %
0.336
40
ρp
σg
Mass of AC used [g]
0.1
115
~1
geom etric stan dard deviation
3
Hg feed concentration [µg/m ]
55
55
Time for 90% breakthrough [hr]
~2
~13
~10
REFERENCES
1. Matsuo, T.; Ish izaki, N.; Fukuda, T. Activated Carbon Fiber: KF; Sen-i
Gakkaishi 1977, 33, 204-208 (in Japan ese).
C /C
at equilibrium
~1.0
~0.72
~45
~0.50
~400
~0.62
~52.5
Hg Hg,inlet
Adsorptive capacity [µg Hg/g AC]b ~19
2. Koh ata, T.; Fukuda, T.; Ish izaki, N.; Om ori, S.; Kom agata, H. Develop-
m en t Con cern in g Application of Activated Carbon Fiber; Kagaku
Kogaku 1980, 44, 343-350 (in Japan ese).
3. Fukuda, T.; Om ori, S.; Asakura, S.; Matsum oto, K. Recovery of Freon
by Activated Carbon Fiber; Kagaku Souchi 1989, 31, 55-61 (in Japa-
n ese).
aData is calculated from results reported by Vidic and McLaughlin;25 bCalculated up to
90% of final effluent concentration.
4. Davies, C.N. Air Filtration; Academ ic Press: Lon don , 1973.
5. Kuwabara, S. Th e Forces Experien ced by Ran dom ly Distributed Paral-
lel Circular Cylin ders or Sph eres in a Viscous Flow at Sm all Reyn olds
Num bers; J. Phys. Soc. Japan 1959, 14, 527-532.
6. Kirsch , A.A.; Fuch s, N.A. Th e Fluid Flow in a System of Parallel Cylin -
ders Perpen dicular to th e Flow Direction at Sm all Reyn olds Num -
bers; J. Phys. Soc. Japan 1967, 22, 1251-1255.
7. Kirsch, A.A.; Fuchs, N.A. Studies on Fibrous Aerosol Filters–II. Pressure
Drop in System of Parallel Cylinders; Ann. Occup. Hyg. 1967, 10, 23-30.
8. Stech kin a, I.B.; Kirsch , A.A.; Fuch s, N.A. Studies on Fibrous Aerosol
Filters–IV. Calculation of Aerosol Deposition in Model Filters in th e
Region of Maxim um Pen etration ; Ann. Occup. Hyg. 1969, 12, 1-8.
9. Lee, K.W.; Liu, B.Y.H. Experim en tal Study of Aerosol Filtration by
Fibrous Filters; Aerosol. Sci. Tech. 1982, 1, 35-46.
10. Lee, K.W.; Liu, B.Y.H. Th eoretical study of Aerosol Filtration by Fi-
brous Filters; Aerosol Sci. Tech. 1982, 1, 147-161.
11. Wan g, C.S.; Beizaie, M.; Tien , C. Deposition of Solid Particles on a
Collector: Form ation of a New Th eory; AIChE Journal 1977, 23, 879-
889.
12. Barot, D.T.; Tien, C.; Wang, C.S. Accumulation of Solid Particles on Single
Fibers Exposed to Aerosol Flows; AIChE Journal 1980, 26, 289-292.
13. Payatakes, A.C.; Tien , C. Particle Deposition in Fibrous Media with
Den drite-Like Pattern s: A Prelim in ary Model; J. Aerosol Sci. 1976, 7,
85-100.
stan dpoin t an d recovery), an d also to usin g activated car-
bon s in packed bed system s (h as a m uch h igh er pressure
drop). Clearly, m eth ods of ACF regen eration h ave to be
used; h owever, th e overall operation is expected to be su-
perior to activated carbon powder in jection .
CONCLUSIONS
Th e results of th is study in dicate th at ACF filters can be
readily used for particulate m atter an d Hg con trol. Th e
pressure drop ch aracteristics agree well with th eoretical
prediction s for fibrous filters. Th e resistan ce factor, R, for
th e ACF filter was estim ated to be 2.01, close to th ose
obtain ed for cake filtration in HEPA filters. However, th e
theoretical predictions for the particle collection efficiency
did n ot m atch th e experim en tal results, possibly because
of n on un iform ities in th e differen t fibers. Th e m easured
particle collection efficien cies were h igh er th an th e pre-
dicted values.
14. Payatakes, A.C. Den drites an d Air Filter Efficien cy; Filtration Separa-
tion 1976, 13, 602-607.
15. Payatakes, A.C. Model of Tran sien t Aerosol Particle Deposition in
Fibrous Media with Den dritic Pattern s; AIChE Journal 1977, 23, 192-
202.
16. Bh utra, S.; Payatakes, A.C. Experim en tal In vestigation of Den dritic
Deposition of Aerosol Particles; J. Aerosol Sci. 1977, 10, 445-464.
17. Payatakes, A.C.; Gradon , L. Den dritic Deposition of Aerosols by Con -
vective Brown ian Diffusion for Sm all, In term ediate an d High Par-
ticle Kn udsen Num bers; AIChE Journal 1980, 26, 443-454.
18. Payatakes, A.C.; Gradon , L. Den dritic Deposition of Aerosol Particles
in Fibrous Media by Inertial Im paction and Interception; Chem. Engng.
Sci. 1980, 35, 1083-1096.
19. Kan aoka, C.; Em i, H.; Myojo, T. Sim ulation of th e Growin g Process of
a Particle Den drite an d Evaluation of a Sin gle Fiber Collection Effi-
cien cy with Dust Load; J. Aerosol Sci. 1980, 11, 377-389.
20. Kanaoka, C.; Em i, H.; Tranthapanichakoon, W. Convective Diffusional
Deposition an d Efficien cy of Aerosol on a Dust-Loaded Fiber; AIChE
Journal 1983, 29, 895-902.
21. Kan aoka, C.; Em i, H.; Hiragi, S.; Myojo, T. Morph ology of Particulate
Agglom erates on a Cylin drical Fiber an d Collection Efficien cy of a
Dust-Loaded Filter. Presen ted at Aerosols, Form ation , an d Reactivity,
2n d In tern ation al Con feren ce, Berlin ; Pergam on , 1986, pp 674-677.
22. Leibold, H.; Wilh eim , J.G. In vestigation s in to th e Pen etration an d
Pressure Drop of HEPA Filter Media durin g Loadin g with Subm icron
Particle Aerosols at High Con cen tration s; J. Aerosol Sci. 1991, 22, S773-
S776.
23. Kan aoka, C.; Hiragi, S. Pressure Drop of Air Filter with Dust Load; J.
Aerosol Sci. 1990, 21, 127-137.
24. Otan i, Y.; Em i, H.; Kan aoka, C.; Uch ijim a, I.; Nish in o, H. Rem oval of
Mercury Vapor from Air with Sulfur-Im pregnated Adsorbents; Environ.
Sci. Technol. 1988, 22, 708-711.
25. Vidic, R.D.; McLaugh lin , J.B. Uptake of Elem en tal Mercury Vapors by
Activated Carbon s; J. Air & Waste Manage. Assoc. 1996, 46, 241-250.
26. Uberoi, M.; Sh adm an , F. High -Tem perature Rem oval of Cadm ium
Com pounds Using Solid Sorbents; Environ. Sci. Technol. 1991, 25, 1285-
1289.
Th e poten tial of th e ACF filters to capture Hg was
also illustrated. Th e Hg capture capacity was con sisten t
with th at reported for activated carbon powders. ACF fil-
ters offer a practical way to use activated carbon in in dus-
trial system s for Hg an d particulate m atter capture with a
low-pressure drop.
NOMENCLATURE
A
C
df
filter area (m 2)
Cun n in gh am slip correction factor
fiber diam eter (m )
dm m m ass m edian diam eter (m )
dp
E
particle diam eter (m )
overall collection efficien cy of filter
clean filter resistan ce (kg m -2 sec-1)
m easured specific cake resistan ce (sec-1)
polydisperse specific cake resistan ce (sec-1)
filter th ickn ess (m m )
K1
K2
K2C
L
M
P
total m ass of particles collected on filter (kg)
total pressure drop across filter (Pa)
928 Journal of the Air & Waste Management Association
Volume 50 June 2000

Hayashi et al.
27. Biswas, P.; Wu, C.Y. Con trol of Toxic Metal Em ission from Com bus-
tors Usin g Sorben ts: A Review; J. Air & Waste Manage. Assoc. 1998, 48,
113.
28. Owen s, T.M.; Biswas, P. Reaction s between Vapor Ph ase Lead Com -
poun ds an d in situ Gen erated Silica Particles at Various Lead-Silicon
Feed Ratios: Application s to Toxic Metal Capture in Com bustors; J.
Air & Waste Manage. Assoc. 1996, 46, 530-538.
29. Biswas, P.; Zach ariah , M.R. In situ Im m obilization of Lead Species in
Com bustion En viron m en ts by In jection of Gas Ph ase Silica Sorben t
Precursors; Environ. Sci. Technol. 1997, 31, 2455-2463.
30. Wu, C.Y.; Lee, T.G.; Tyree, G.; Arar, E.; Biswas, P. Capture of Mercury
in Com bustion System s by in - situ Gen erated Titan ia Particles with
UV Irradiation ; Environ. Eng. Sci. 1998, 15, 137-147.
40. Lee, K.W.; Ram am urth i, M. Aerosol Measurement: Principles, Techniques,
and Applications; Van Nostran d Rein h old: New York, 1993; p 188.
41. Lee, T.G.; Yan g, G.; Biswas, P. Com parison of Fractal Dim en sion s an d
Radius of Gyration of Agglom erates Determ in ed by Ligh t Scatterin g
to Electrical Mobility Measured Diam eters; Aerosol Sci. Technol. 2000,
subm itted for publication .
42. Huan g, H.S.; Wu, J.M.; Liven good, C.D. Developm en t of Dry Con -
trol Tech n ology for Em ission s of Mercury in Flue Gas; Hazard. Waste
& Hazard. Materials 1996, 13, 107.
31. Toyobo Co., Ltd. Preparation an d th e Properties of Activated Carbon
Fiber; Internal Technical Report; Toyobo Co., Ltd.: Osaka, Japan, 1979.
32. Biswas, P.; Li, X.; Pratsin is, S.E. Optical Waveguide Preform Fabrica-
tion : Silica Form ation an d Growth in a High -Tem perature Aerosol
Reactor; J. Appl. Phys. 1989, 65, 2445-2450.
33. Gupta, A.; Novick, V.J.; Biswas, P.; Mon son , P.R. Effect of Hum idity
an d Particle Hygroscopicity on th e Mass Loadin g Capacity of High
Efficien cy Particulate Air (HEPA) Filters; Aerosol Sci. Tech. 1993, 19,
94-107.
34. Moren cy, J.R. Con trol of Mercury in Fossil Fuel-Fired Power Gen era-
tion . Presen ted at th e PETC’s 10th An n ual Coal Precapture, Utiliza-
tion , an d En viron m en tal Con trol Con tractors Con feren ce, Pittsburgh
En ergy Tech n ology Cen ter, Pittsburgh , PA, 1994.
35. Ch u, P.; Porcella, D.B. Mercury Stack Em ission s from U.S. Electric
Utility Power Plan ts; Water, Air & Soil Pollut. 1995, 80, 135-142.
36. CFR, Title 40, Parts 61 an d 266, 1992.
37. Novick, V.J.; Higgin s, P.J. Ph ase 1 Ch aracterization of th e HEPA Filter
Media Used in th e Airborn e Activity Con fin em en t System at th e Sa-
van n ah River Site; Report WSRC-RP-89-793; Westin gh ouse: Savan -
n ah , SC, 1989.
About the Authors
Pratim Biswas is a professor and Tai Gyu Lee is a research
assistant professor in the Environmental Engineering and
Science Division at the University of Cincinnati. Elizabeth
Hedrick is a researcher at the National Exposure Research
Laboratory, U.S. Environmental Protection Agency, in Cin-
cinnati. At the time of this work, Melynda Hazelwood was a
U.S. Department of Energy intern from Cumberland College
in Kentucky, and Toshiaki Hayashi was a visiting scholar from
Toyobo Co., Ltd., Japan. Partial support is gratefully acknowl-
edged from U.S. Department of Energy grant No. DE-FG22-
95PC95222 and OCRC grant No. OCRC-98-1, B3.7.
38. Letourn eau, P.; Mulcey, P.; Ven del, J. Effect of Dust Loadin g on th e
Pressure Drop and Efficiency of HEPA Filters; Filtration Separation 1987,
24, 265-267.
39. Rudn ick, S.N. Sc.D. Th esis, Harvard Sch ool of Public Health , Cam -
bridge, MA, 1978.
Volume 50 June 2000
Journal of the Air & Waste Management Association 929