EPA-650/2-74-083



SEPTEMBER 1974
Environmental Protection Technology  Series


                    :::::;;::;:::;i:;i;:;;:;:;i#

-------
                                  EPA-650/2-74-083
 DYNACTOR  SCRUBBER
        EVALUATION
                by

    D. W. Cooper and D. P. Anderson

           GCA Corporation
       GCA/Technology Division
      Bedford, Massachusetts 01730
     Contract No. 68-02-1316, Task 6
         ROAP No. 21ADL-004
       Program Element No. 1AB012
    EPA Project Officer:  D.L.Harmon

       Control Systems Laboratory
  National Environmental Research Center
Research Triangle Park, North Carolina 27711
            Prepared for

 OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
       WASHINGTON, D.C.  20460

           September 1974

-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication.  Approval does not signify that the
contents necessarily reflect the views and policies of tho Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
                                  ii

-------
                             ABSTRACT

A novel aspirative spray scrubber, the Dynactor  (RP  Industries, Hudson,
Massachusetts), was tested for power consumption and collection ef-
ficiency at three flow rates, two temperatures, two dust loading levels,
for two dusts.  Total filter samplers and cascade impact or s were used
upstream and downstream from the collector.  Power was determined from
voltage, current, and phase angle measurements, A factorial design
series of tests at two levels of flow, concentration, temperature, and
dust type gave these average mass efficiencies:  99.0 percent for 4.0-
5.6 um aerodynamic diameter, 98.4 percent for 2.5-4.0  jim, 93.0 percent
for 1.3-2.5 ^m, 75.4 percent for 0.8-1.3 /im, 27.4 percent for 0.54-0.80^
and 47.4 percent for  <0.54 jxm.  Higher efficiency was fostered by:
lower flow rate, lower inlet temperature, higher mass loading.  Power
consumption was about one-third of that expected from a venturi scrubber
operated at a pressure drop (1.0 x 10  N/m  « 40 inches H-0) giving
equivalent collection efficiency.  Collection efficiency for both the
Dynactor and the venturi scrubber decreases dramatically for fine
particles smaller than 1
This report was submitted in fulfillment of Task Order No. 6 under
Contract No. 78-02-1316 by GCA/Technology Division under the sponsorship
of the Environmental Protection Agency.  Work was completed as of
July 26, 1974.
                                iii

-------
                              CONTENTS
ABSTRACT
LIST OF FIGURES
LIST OF TABLES
ACKNOWLEDGMENTS
                                                      Page No.
                                                         ill
                                                          v
                                                          vi
                                                        vlii
Section
    I
   II
  III
   IV
    V
   VI
  VII
 VIII
CONCLUSIONS
RECOMMENDATIONS
INTRODUCTION
TEST EQUIPMENT
RESULTS
DISCUSSION
APPENDICES
REFERENCES
  1
  2
  3
  'j
 22
 60
A-l
 72
                                iv

-------
                               FIGURES
No.                                                              Page No.
 1     Test System for Dynactor Two-stage Scrubber Evaluation,
       Including Filter Samples (F), Thermometers (T), and
       Pressure Gauge  (P)                                           6

 2     Details of Aerosol Concentration and Size Distribution
       Measurement Section                                          7

 3     Cumulative Mass Size Distribution of Iron Oxide Pigment
       Aerosol as Determined by Two Impactors (#4601 and #4602)
       With Original Substrates and With Reclaimed Substrates      13

 4     Flow Velocity Profile in Inlet Ducting, Dynactor Test
       Setup                                                       19

 5     Flow Velocity Profile in Outlet Ducting,  Dynactor Test
       Setup                                                       20

 6     Volume Flow Rate Versus Center-line Reading With Pitot
       Tube, Dynactor Test Setup                                   21

 7     Pressure Gain Produced by Dynactor Versus Flow Rate
       Through It                                                  24

 8     Dynactor Spray Flow and Nozzle Power Dissipation Versus
       Spray Pressure                                              27

 9     Inlet,Size Distribution, Iron Oxide at 1000 cfm
       (28 mJ/min)                                                  30

10     Inlet Size Distribution, Fly Ash at 1000  cfm (28 m3/min)    31

11     Dynactor Scrubber Collection Efficiency Versus Particle
       Aerodynamic  Diameter, Effects of Loading  and  Dust Type      47

12     Dynactor Scrubber Collection Efficiency Versus Particle
       Aerodynamic  Diameter, Effects of Flow Rate and Inlet
       Temperature                                                  48

13     Collection Efficiency,  1.3-2.5 urn,  Versus Flow for
       Different Dusts, Temperatures,  and  Concentrations           51

14     Collection Efficiency Curves for Venturi  Scrubbers and
       Dynactor Efficiency Data                                    62

Al    Single-stage Dynactor diffusion system cross sectional view  A-2

A2    Two-stage Dynactor scrubber                                  A-4

-------
                                TABLES

 No.                                                             Page No.

  1     Data From Two Andersen Mark III Cascade  Impactors
        Sampling Same Iron Oxide Aerosol (Mapico Black)  at
        28 Xpm cm New Media                                        11

  2     Data From Two Andersen Mark III Cascade  Impactors
        Sampling Same Iron Oxide Aerosol (Mapico Black)  at
        28 ipm on Reclaimed Media (Final Flow in #4601 was
        about 5 percent  less  than that   of #4602)                   12

  3     Andersen Model III Impaction Substrate Losses               16

  4     Dynactor Air-moving Capabilities:  Flow With Inlet and
        Outlet Open  to Atmosphere, Maximum Pressure Gain (No
        Air Flow)                                                  25

  5a     Test Matrix,  Fly Ash Aerosol                                33

  5b     Test Matrix,  Iron  Oxide  Aerosol                             34

  6     Summary of Results  of  16 Collection Efficiency Tests
        for Dynactor  (Factorial  Test Design)                        36

  7     Results  of Statistical Analysis  on Efficiency               38

  8     Significance  of Effects  of Flow, Dust, Temperature,  and
        Concentration on Scrubber Collection Efficiency             46

  9      Detailed Analysis of Interactions for Dynactor Efficiency
        on  1.3  -  2.5  nm Aerosol  Fraction  (Stage #6)                 50

10      Estimates of  Experimental Error for Single Test from the
        Residuals of  the Sum of Squares and from the  Replication
        Tests, Nos. 16, 17, 18                                      53

11      Results of Dynactor Collection Efficiency Tests Under
       Similar.Conditions except for Flow Rate (Iron Oxide,
       - 1 g/m )                                                   55

12     Results of Dynactor Collection Efficiency Tests Under
       Similar Conditions but With and Without Steam Addition
        (Iron Oxide,  500 cfm - 0.24 m /sec,  -v 1 g/m )                57

13     Results of Dynactor Collection Efficiency Tests Under
       Similar Conditions Except for the Addition of  10 ppm
       Surfactant to the Spray Water                               59
                                   v i

-------
                          TABLES (Continued)

No.                                                             Page No.

14a    Estimated Capital Cost of Dynactor System,  Based On
       Manufacturer's data (1973 dollars)                          65

14b    Estimated Capital Cost of High Energy Venturi Scrubbers,
       Yielding Similar Collection Efficiency to Dynactor (1973
       dollars)                                                    66
                                 vii

-------
                         ACKNOWLEDGMENTS
Information and support supplied by Stanley Rich, Vice President and
Technical Director of R P Industries, and by Dale Harmon,  of Control
Systems, Environmental Protection Agency, are acknowledged and  were
appreciated.
                              viii

-------
                                 SECTION I
                                CONCLUSIONS
This  evaluation was  one  of  a  series  of  such  evaluations  being  conducted by
the Environmental  Protection  Agency  to  identify novel devices which are
capable  of  high efficiency  collection of  fine particulates.  The Model
DY-12-F2 Dynactor  Scrubber  of R P  Industries  (Hudson, Massachusetts) which
was tested  had substantially  less  than  99% collection efficiency on fine
particulates, those  smaller than 2 /jm in  diameter, and thus did not satisfy
this  objective.
The following average mass  efficiencies were observed at the nominal rated
            3
flow  (0.47  m /s, 1000 cfm)  and  at  half  its rated flow:
          Size fraction                Efficiency        _
          (aerosol aero-           0.47 m /s      0.24 m /s
          dynamic diameter)         (1000  cfm)     (500 cfm)
              4.0-5.6 urn               98.8          99.2
              2.5-4.0 Mm               98.0          98.8
              1.3-2.5 Mm               91.2          94.8
              0.8-1.3 Mm               67.4          83.4
             0.54-0.8 Mm               28.1          26.7
              < 0.54  Mm               45.5          49.3
These efficiencies are similar  to  those expected from a venturi scrubber
with  a pressure drop of 1.0 x 10   N/m   (40 inch H-0).  The Dynactor oper-
ates  on  about one-third the power  of such a venturi.  A comparison of
                                                               3
costs for a venturi  scrubber and a Dynactor scrubber for a 19 m /s
(40,000  cfm) application indicated that the major difference between the
two would be about $40,000 to $50,000 per year savings in electrical
power costs (at $0.025/kWhr) for those using a Dynactor scrubber.

The following factors improved  spray scrubber collection efficiency:
lower inlet temperature, lower air flow, higher particle mass concentra-
tion,  higher nozzle pressure,  surfactant addition.

-------
                               SECTION II
                             RECOMMENDATIONS

Although the Dynactor does not give high efficiency collection for
particles smaller than one micron diameter, in those applications for
which a venturi scrubber might be suitable, the use of a Dynactor
scrubber should be considered as one alternative.  If collection
efficiency requirements and other considerations would require a
                                                              /     O
venturi scrubber with a pressure drop on the order of 1.0 x 10  N/tn
(40 inch H20),  a Dynactor scrubber could be substituted with a
significant savings in power consumption and a comparable cost for
equipment and installation.

-------
                            SECTION  III
                            INTRODUCTION

This work was done  to evaluate  the Dynactor scrubber with respect to  its
mass collection efficiency  as a function of particle size, the effect
of  several  parameters on this efficiency, the air-moving capability and
power consumption of the device, and  its cost.

The Dynactor uses a proprietary nozzle design to produce a water spray
which serves as an  air eductor  and as a scrubber, thus cleaning and pro-
pelling the gas simultaneously.  A description of the Dynactor, written
by  its manufacturer (RP Industries, Hudson, Massachusetts), is in Appen-
dix A.  We  tested one of the smaller units of its type, a two-stage de-
                                               3
vice with a nominal rating  of 1000 cfm (0.472 m /s), Model DY 12 F2.
The Dynactor was installed  in a test setup at GCA/Technology Division
and the following measurements  were made:
      • Air flow and pressure gain versus spray nozzle pressure
      • Electrical  power consumption versus spray nozzle pressure
      • Mass collection efficiency as a function of particle aero-
        dynamic diameter at two  levels of flow,  temperature,  and
        concentration,  for two  different dusts,  in a balanced
        test matrix
      • Mass collection efficiency as a function of particle  size
        for several additional  sets of conditions
      • Total mass collection efficiency at the  conditions  noted
        above
The dust was generated by a dust feeder - air ejector combination.   Its
concentration was determined by gravimetric analysis of filter samples

-------
obtained by  isokinetic sampling upstream and downstream from the Dynac-
tor.  The concentrations  in a set of aerodynamic size  intervals were
obtained from gravimetric  analysis of samples obtained by identical im-
pactors, one placed upstream and the other downstream  from the Dynactor,
from which data the mass collection efficiency as a function of particle
size was obtained.  These  data were analyzed using an F-test analysis
of variance to determine which factors  had  significant influence  on
collection efficiency in the various aerodynamic size  fractions and to
estimate experimental uncertainty.  Flow was measured using pitot tube
traverses and pressure gain was measured using Magnehelic pressure
gauges.  Electrical power  consumption was measured using an induction
coil ammeter and an oscilloscope, from which current, voltage,  and phase
angle could be obtained.

The work also included some cost estimates,  from which certain compari-
sons can be made with other control devices.  Comparisons with  other
experimental results for spray scrubbers have also been made,  especially
with regard to which factors can enhance collection efficiency.

-------
                            SECTION  IV
                          TEST EQUIPMENT

Most of  the  experimental work was done with the equipment shown in
Figures  1 and 2.  This equipment allowed us to measure collection effi-
ciency as a  function of particle size as well as the conditions of flow,
concentration, temperature, pressure drop, etc. which prevailed during
the efficiency tests.

Figure 1 gives the overall picture of the test setup.  Dust from a
screw feed was picked up by an air ejector/aspirator and blown into
ducting  leading to the Dynactor.  The flow in the ducting was produced
by the Dynactor and the relatively weak fan of the heater/blower, some-
times in conjunction with the fan shown at the very end of the flow
train.  Upstream, the turbulent mixture was sampled five or more duct
diameters from the dust feed by an Andersen Model III cascade impactor,
usually run with isokinetic flow, and by a filter assembly (glass fiber
absolute filter, 47-mm diameter) which always was operated isokinetic-
ally.  An identical sampling combination was used downstream, as de-
scribed more fully in Figure 2.  The temperature and pressure of the
mixture entering the Dynactor was measured as was,  sometimes, the tem-
perature of the mixture leaving the Dynactor.   The  pressure drop or
gain across the device was measured.   Each of the major components  of
the test equipment will be described  in greater detail  next.

-------
HEATER/
BLOWER
                    DUST
                  FEEDER

STAGE
* 2

STAGE
t I

     Figure 1.  Test system for Dynactor two-stage scrubber evaluation,
                including filter samplers (F),  thermometers (T),  and
                pressure gauge (P)

-------
 ISO-KINETIC PROBES
                                                                  FILTER
                    HEATED
                    DRYING
                    SECTION
                                      THERMBCTER
                                                                       PCMP
             CONDENSATION
             INDICATOR
Figure 2.  Details of aerosol concentration and  size  distribution
           measurement sections

-------
 HEATER/BLOWER

 The heater burned propane supplied continuously from a pressurized tank.
 Propane was selected  because its combustion products are almost exclu-
 sively carbon dioxide and water, with a negligible production of carbon
                                                        3
 aerosol.   The heater  was  rated  at a maximum of 350 x 10  Btu/hr (1.03 x
 10  joule/s).  Its  blower could provide 1400 cfm (0.66 m /s)  maximum
 flow.
 DUST  FEEDER AND  ASPIRATOR

 The dust  feeder  (Acrison,  Inc. Model  120)  operated with  a  vibrating hop-
 per that  channeled  the  dust  into  a  cavity  from which  a screw  feeder di-
 rected  an adjustable  constant volume  flow  to  the  aspirator.   The  aspir-
                                                           5    2
 ator  was  powered with pressurized air at 80 psig  (5.5 x  10 N/m    above
 atmosphere)  and  blew  the  dust into  the main ductwork  while deagglomer-
 ating the aerosol material.

 IMPACTORS

 Upstream  and downstream from the  Dynactor, we  used the Andersen Model
 III in-stack impactor to  size-fractionate  the  aerosol.   Except for early
                                                  -4   3
 tests which were run  at 28 .0pm (1 cfm or 4.7 x 10   m /s), the impac-
 tors  were operated  at 14  £pm to lessen the likelihood of particle re-
 bound, as  advised by its manufacturer .  The impactors were used with the
 glass fiber media impaction  substrates designed for them.  Later  in this
 report  is  a description of some tests  done to  ascertain  weight changes
 in the  impactor  substrates due to causes other than the  accumulation of
 particulate material.  The temperature of the  impactor at  the Dynactor
 outlet  and its drying section (Figure  2) was kept about  20 C above the
 temperature of the Dynactor  exhaust stream to  produce drying of the
droplets  present in the exhaust.  The drying section  volume was 2.4 x
  -3  3
 10   m  (75 cm long by 6.4-cm diameter), which yielded a residence time
of 10 seconds.
                                 8

-------
Preliminary tests showed that the concentrations downstream from the
Dynactor scrubber were much lower than those upstream.  To obtain
enough material on the downstream impactor for weighing and yet not top
much material on the upstream impactor (to prevent rebound and reen-
trainment), it was necessary to run these impactors for different total
durations, the downstream impactor sampling for about 10 times as long
as the upstream impactor.  The total downstream sampling time was sub-
divided into intervals, and the upstream samples were run briefly at the
time mid-points of these intervals, essentially the analogue of a midr
point quadrature.  The total filter samples had the same durations as
their impactor counterparts.  To prevent material from being captured
by the probes upstream when a sample was not being taken, these probes
were blocked with removable baffles.

Consideration was given to correcting the data obtained with the up-
stream impactor for the volume of air that was in the drying chamber at
the beginning of each sample, approximately 2.4 liters of air, relas-
tively free of aerosol due to sedimentation, etc., between samples.
Generally each upstream sample was for a minute or two, thus 14 to 28
                     o
liters (0.5 to 1.0 ft ) , so that the relatively clean air wpuid be £rom
about 8 to 16 percent of the total sample.  When we checked the flow
rate of the upstream impactor at the end of the test series, we found
that it had drifted from 14 4pm (0.5 cfm) to 17 ,0pm (0.6 .eftn), whereas
the downstream impactor had not drifted from 14 £pm..  The ,ayer,a£e con~-
tribution of this drift would be to make the concentrations upstream
seem 10 percent higher than they were, but the contribution from
relatively clean air would have made the concentrations seem 8 -to
percent too low, so these effects nearly cancelled each oifther.
net effect on flow and the 5-percent change in aerodynamic cutoff
eter in the upstream impactor were treated as negligible.  The down-
stream impactor sampled for 10 minutes or so generally and the 2 per-
cent (2.4/140) effect on total volume sampled was also ignored.

-------
 METTLER BALANCE (HIS) AND WEIGHING ERRORS

 The concentrations were determined from weight changes in the filters
 and impaction substrates.  To ascertain the reproducibility of our weigh-
 ing measurements,  we made 18 weighings each of two weights, 10 g and 100
 rag, over 8 days.  We obtained 16 readings of 9.9994 and two of 9.9993 g,
 and 14 readings of 0.1000 and four of 0.1001 g, from which we concluded
 our reproducibility was better than 0.1 rag.

 To check whether or not dessication made a substantial difference in
 the impactor substrate material, we made 24 weighings before and after
 dessicating the substrates for a day's duration.   The mean change in
 weight was 0.15 mg loss (for substrates averaging  0.20 g)  and 11 of the
 24 changes were 0.1 mg.  From this we decided to dessicate the substrate
 material before making the tare weighing as well as  before making the
 weighing with the  captured particulate material.

 Precautions were taken to enhance weighing accuracy  for all efficiency
 tests.   The substrates were  dessicated at  least 12 hours.   A static
 charge  eliminator  was  used in the Mettier  analytical  balance.  The  sub-
 strates  were weighed  singly  and  the weights  of  groups  of four were  com-
 pared with the  sum of  the four  individual  weights.  To lessen the  like-
 lihood  of  the wrong substrate being ascribed  to a given impaction stage,
 the substrates were numbered  so  that  their  last digit was  the same as
 the stage with which they were  to be used.

COMPARISON  OF ANDERSEN MARK III  IN-STACK CASCADE IMPACTORS

In a comparison  test done  before  the Dynactor efficiency tests, the two
identical  impactors were used to  sample the same aerosol,  iron oxide
powder generated from a Wright dust feeder into a wind tunnel.  Each
impactor drew 28 .0pm (1 cfm) from a Y-connection that was connected to
a single sampling probe.  The results of this test are given  in Tables
1 and 2 and in Figure 3.  The "reclaimed" media were rinsed in methanol
                                 10

-------
Table 1.  DATA FROM TWO ANDERSEN MARK III CASCADE IMPACTORS SAMPLING SAME IRON
          OXIDE AEROSOL (MAPICO BIACK) AT 28 Ipm ON NEW MEDIA
Impactor
Stage
1
2
3
4
5
6
7
8
Filter
Total
Mass
Mass collected
Impactor
#4601
0.4
0.3
0.6
0.8
2.0
9.6
3.3
0.1
0.6

17.7
Impactor
#4602
0.2
0.2
0.4
0.7
1.9
9.7
3.8
0.2
0.5

17.6
Percentage of
total mass
Impactor
#4601
2
2
3
5
11
54
19
1
3

100
Impactor
#4602
1
1
2
4
11
55
22
1
3

100
Cumulative mass
percentage
Irapactor
#4601
2
4
7
12
23
77
96
97
100

Impactor
#4602
1
2
4
8
19
74
96
97
100

Effective
Diameter
(lim)
D50
9.6
6.0
4.0
2.75
1.75
0.9
0.54
0.36


           Correlation  coefficient:
0.998

-------
Table 2.   DATA FROM TWO ANDERSEN MARK III CASCADE IMPACTORS SAMPLING SAME IRON
          OXIDE AEROSOL (MAPICO BLACK) AT 28 ipm ON RECLAIMED MEDIA (FINAL FLOW
          IN #4601 WAS ABOUT 5 PERCENT LESS THAN THAT OF #4602)
Impactor
Stage
1
2
3
4
5
6
7
8
Filter
Total
Mass
Mass collected
Impactor
#4602
0.1
0.1
0.4
0.7
2.2
11.3
3.5
0.1
0.6
19.0 R
Impactor
#4601
0.1
0.1
0.3
0.6
1.8
10.6
3.6
0.4
0.6
18.1 R
Percentage of
total mass
Impactor
#4602
1
1
2
3
12
59
18
1
3
100
Impactor
#4601
1
1
2
3
10
59
19
2
3
100
Cumulative mass
percentage
Impactor
#4602
1
2
4
7
19
78
96
97
100
Impactor
#4601
1
2
4
7
17
76
95
97
100
Effective
Diameter
(um)
D50
9.6
6.0
4.0
2.75
1.75
0.9
0.54
0.36
< 0.36
                  Correlation coefficient:
0.999

-------
u

o
M
Q


§
P"
I-M
w
Figure 3.
    10
     6
       w    4
    1.5
            .6



            .5






           .35
                                                     X  #4601, RECLAIMED


                                                     ©  #4602, RECLAIMED


                                                     +  #4601


                                                     A  #4602
                                                                      Xfe
                          10
                                                    50
70
                                                                90
95
                                                                                   98   99
                            MASS PERCENTAGE LARfER THAN BCD
    Cumulative mass size distribution of iron oxide pigment aerosol as determined by

    two impactors (#4601 and #4602) with original substrates and with reclaimed sub-

    strates.   Aerosol generated by Wright dust feeder

-------
 and reused.   The correlation coefficient for the  mass  captured  on the
 corresponding stages of the impactors  was 0.998,  indicating very close
 matching.   The data also indicate the  impaction substrates  can  be re-
 claimed by rinsing with methanol, which is significant,  as  they cost
 nearly $1.00 apiece.

 The smallest weight change that  could  be read from the balance  we used
 was 0.1 mg,  so that the weights  are  precise to within  about + 0.05 mg,
 which is also an estimate of the accuracy of the  difference between
 two weights which are nearly the same, the case that we  had to  deal
 with in the  main.   In the comparisons  of the two  impactors,  the ratio
 of  two weights for the same stages for the different impactors  diverged
 from unity significantly as the  weights approach  0.1 mg,  as expected
 from the weighing  precision.

 These comparison tests show that the impactors  are  nearly identical in
 performance,  which is  what  is  required in the tests  of the  Dynactor
 collection efficiency.

 LOSSES  FROM  IMPACTION  SUBSTRATE  MATERIAL

 It  has  been  reported by Bird,  et al. (1973)  that  the substrate  material
 used in the  impactor,  cut from glass fiber filters,  had a tendency
 to  lose weight  in  handling  and use.  At  the  beginning  of  our  tests we
weighed  four  such  substrates before and  after  inserting them  into  and
 removing them from the  impactor,  and we  found an  average weight  loss
 of  0.1 mg  per substrate.  Subsequent tests  involving actually sampling
 aerosols produced  some  very lightly loaded substrates  that  showed
weight  losses > 0.1 mg.   Blowing  off the  substrates  before using them
visibly removed  some loose  material.   Rinsing the substrates  in methanol
before using  them  removed an average of  1 mg of loose  solid material per
substrate.  There was noticed some losses  in weight  even with the  pro-
cedure of rinsing with methanol, dessicating for  12 hours or more,
                                 14

-------
weighing, using, dessicating again, weighing again.  To estimate this
loss we wanted  to determine an average weight loss due to handling and
sampling, so we sampled filtered air for 2 hours at 0.5 cfm.  The
average loss was 0.23 mg per substrate.  All these loss tests are
tabulated in Table 3.

The first column of Table 3 briefly describes how the impactor sub-
strate material was treated before it was tested; either it was used
as it came from the manufacturer (untreated) or it was washed in
methanol and dried again.  The second column tells how the substrate
was used in the tests: either just loaded into the impactor and then
unloaded immediately or loaded, used to sample a test aerosol as
part of the efficiency tests, and unloaded or loaded, used to sample
filtered air, and unloaded.  The weight loss is the difference
between the substrate weight before the test and the weight after the
test; the losses listed for the substrates used to sample the aerosols
are the losses  for the substrates which showed any loss, 25 of the 176
substrates from a series of 11 efficiency tests, which substrates were
invariably very lightly loaded (visual inspection).  The fourth
column gives the number of substrates having the indicated loss.  These
tests led to the following treatment of the data obtained with the
cascade impactor:
     1.  Weight changes of 0.3 mg or less were taken to be 0.3 mg
         for calculation of efficiencies, which efficiencies are
         given as inequalities.
     2.  Weight changes of 0.4 mg or more were used without
         correction.
We decided not to add a "loss correction" to the data, although
arguments can be made for doing so or not doing so.  The usual total
weight change was approximately 20 mg divided among the eight stages
and final filter and a correction, if used, would have been
approximately 0.2 mg per stage.  Disadvantages to making such a
                                 15

-------
Table 3.  ANDERSEN MODEL III IMP ACTION SUBSTRATE LOSSES
Treatment

Untreated
Most washed in
methanol






Washed in methanol





Washed in methanol




Test conditions

Load and unload (4)
Sample aerosols (176)







Sample filtered air (8)
for 2 hours, at am-
bient temperature



Sample filtered air (8)
for 15 minutes, am-
bient temperature


Weight loss
(nig)
-0.1
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
average loss:
-0.26
-0.1
-0.2
-0.3
-0.4
average loss:
-0.23
-0.1
-0.2
-0.3
average loss:
-0.12
Number

4
8
9
2
2
0
4


2
3
2
1


5
1
1


                      16

-------
 correction are  that the uncertainty in the  correction would  add  to
 uncertainty already present  in the data and that  the  correction  com-
 plicates  going  from the raw  data to the derived quantities  (concen-
 trations,  efficiencies) and  vice-versa.  This  problem of impactor.
 substrate  losses  is under further investigation by the Environmental
 Protection Agency (Control Systems).

 DUCTING

                                                         3
 For  the tests conducted at flow rates  of 500 cfm  (0.24 m IB)  and 1000
            3
 cfm  (0.47  m /s) the sheet metal ducting upstream  from the Dynactor
 was  8  inches in diameter by  about 12 feet long (0.20  m diameter  by
 4 m  long)  connected to  an expander to  take  it  from 8-inch diameter to
 16-inch diameter  (0.20  m to  0.40 m), followed  by  a 90° elbow,  6  feet
 (2 m)  more 16-inch  ducting and  another 90-degree  elbow connected to the
 Dynactor inlet.   For the 1500 cfm (0.71 m3/s)  tests,  the first section
 was  about  8 feet  (3 m)  of 12-inch diameter  (0.30  m) ducting.  The
 dust feeder was separated from  the sampling probes  by about 6 feet
 (2 m).  The Dynactor outlet  had a reducer to go from  the 16-inch
 diameter (0.40 m) ducting to 8-inch diameter ducting  (0.20 m), and
 the sampling probes  were about  3  feet  (1  m)  from  the  reduction.

 FLOW PROFILES IN DUCTING

 The tube inlets for  sampling the  dust  concentrations upstream and down-
 stream from the Dynactor  scrubber were  set up to be no  less than four
diameters downstream from flow  disturbances  and 1.5 diameters upstream.
The flow velocity profiles in the  immediate vicinity of the sampling
positions were measured to see  that they  approximated the turbulent
flow profiles expected.  By the Reynolds  analogy,  such a matching would
be a good indication of fairly homogeneous aerosol mixing.  Good  mixing
is expected because the Reynolds numbers  In the ducting are at least
1 x 10  and the sampling points are about 5 diameters or greater
                                17

-------
 downstream from the  point  of  aerosol  generation.   The  traverses  were
 done  at  the  positions  labeled "impactor"  in Figure 1.   Figure  4  shows
 the velocity profile obtained at  the  8-inch (0.20  m) diameter  ducting
 upstream (for somewhat  different  flow rates).

 Preliminary  tests  at the inlet indicated  a  typical, nearly  flat,
 turbulent  flow profile  without the heater blower on, but  the flow
 exhibited  swirl  once the heater blower was  used.   Installation of
 baffling,  two 4-inch diameter (0.10 m) ducting  pipes about  2 feet
 long  (0.2  m)  immediately downstream from  the blower, virtually
 eliminated the swirl effect;  Figure 4 was obtained with the blower
 on.   A more  detailed pitot traverse was made at the downstream
 sampling position.   The results of eight-point  and twelve-point
 traverses made downstream  are given in Figure 5, along with center-
 line  measurements.   The flow  closely  approximates  the turbulent  flow
 expected, with no  pronounced  anomalies.   The total volume flow from
 the equal-area,  eight-point traverse  was  compared with that from the
 equal-area,  twelve-point traverse: for the  two such comparisons we
 made, they agreed within 5 percent and within 2 percent.  Figure 6
 shows the centerline measurements made simultaneously with the eight-
 point traverse measurements used to characterize Dynactor air-moving
 capability; reproducibility was judged adequate for the use of the
 centerline pitot reading to set the Dynactor flow at the 500,
 1000, and 1500 cfm (14, 28, 42 m3/min) lev
Dynactor fractional collection efficiency.
1000,  and 1500 cfm (14,  28,  42 m3/min)  levels when testing  the
This completes this section on the equipment used in the Dynactor
evaluation and some of the tests performed on this equipment.
                                18

-------
JC
•k
1 ^m • "
«k
»-
U 9
s
LJ
S
m
.05 .10 .15
i i I
• • • • f • * *•
1
1
1
1
1
k
1
1
1 ! i
10

M
X.
E
5

              2468
      POSITION ALONG DUCT DIAMETER,
                     inches
Figure 4.  Flow velocity profile in inlet ducting,
         Dynactor test setup
                    19

-------
                X 12-POINT TRAVERSE
                0 8-POINT
I
•
•

-------
 1000
                                                 28
 800
o600

I

^"-400
                 *:
                                                 20
             .t
                                                  "
                                                10
 200
          1    I    1
                       1
1    1    1    1
          .1  .2   .3   .4   .5  .6   7  .8  3  ID
          Ap,CENTER-LINE  PITOT TUBE,
Figure 6.  Volume flow rate versus center-line reading
         with pitot tube, Dynactor test setup
                      21

-------
                              SECTION V
                               RESULTS

 INTRODUCTION

 This  section contains  the results  of the work done  to  characterize  the
 Dynactor's  air-moving  capabilities,  power consumption,  and  collection
 efficiency  on a mass basis as a function of  particle size,  determined
 with  cascade impactors.   In Section  VI these results will be  dis-
 cussed  and  the Dynactor  compared with venturi scrubbers.

 AIR MOVING  CAPABILITIES

 The Dynactor scrubber  acts somewhat  like a fan, providing a pressure
 boost rather than a pressure  drop  to  air moving through it, up to some
 maximum flow which depends upon  operating conditions.  We tested the
 pressure gain  from Dynactor inlet  to  Dynactor outlet with a pressure
 gauge connected  to the inlet  and outlet  ports, as shown by  P  in
 Figure  1.  The  total air  flow at 70°F was obtained from eight-point
 equal-area traverses in the 8-inch diameter duct immediately down-
 stream  from  the  Dynactor.   The manufacturer  informed us that the inlet
 of  the device should not be at a vacuum  in excess of 1/2 inch H00
         22
 (1.2 x  10   N/m  ) below ambient, and we  noted that for pressure dif-
                                                               o    2
 ferences across  the device  on the order  of 1  inch H-O (2.5 x 10  N/m )
 and greater  there was a significant accumulation of water spray in the
upper plenum chambers of the device,   indicating some disruption of the
normal flow patterns, so only a few tests were done under such condi-
tions.  The data for flow  (at 1 atmosphere pressure and 70°F)  versus

                                 22

-------
 pressure  increase  are  given in Figure  7  for  three  different  spray  noz-
 zle  pressures,  100,  150,  200 psig  (6.9,  10.4,  13.8 x  105  N/m2).   The
 usual  operation of the device, according to  the manufacturer,  is with
                                                               2      2
 pressure  differences  less than or  about  1/2  inch H20  (1.2 x  10  N An ).
 Table  4 has  the data  for  the Dynactor  flows  with inlet  and outlet
 open to the  atmosphere and  the pressure  increases  to  be expected (no
 flow).  The  flows  were somewhat less than expected by the manufacturer;
 the  maximum  pressures  matched expectations.

 POWER  CONSUMPTION: PUMPS

 The  Dynactor scrubber  spray  nozzles are  powered by two high-pressure
 pumps  that run  on  three-phase 207 volt electrical  power.  The  formula
 for  the power consumption of an electrical device  is  given by:

                          P = ^i  I V cos  

 in which
   n = number of phases (here,  3)
   P = power, watts
   I - current magnitude, amps
   V = voltage magnitude, volts
   * = phase angle between voltage and current
 The  power was determined  by measuring the voltage magnitude, the current
magnitude and the phase angle  at three spray nozzle pressures: 100, 150,
 and  200 psig (6.9, 10.4,  13.8 x  105  N/m2).   The current and voltage
measurements were made at a point in the circuit before it branched to
 feed the two pumps, giving total current in each phase wire and the
common voltage.   The voltage was 206 to 207 volts regardless of nozzle
pressure settings, both nozzles being set to the identical pressures for
each measurement.  The current readings in each of the phase wires  were:
                                23

-------
900 i



(
600

2
U_
O
3:
o
u_
300

N / m
.5xl02 I.OxlO2 I.SxIO2 2.0xl02 2.5xl02
r 	 i i i i '
52
A 200 psiy, -- 13.8 x 10 N/m
A
X A
X
x x *
150 psig x
V 7T
* Xx*
_o x x x
0
0 A A
O
100 psig X
O
0 ° 0
X
-
—
o
1 1-1 I n '
.2 .4 .6 .8 1.0 |.
           PRESSURE  INCREASE,
Figure 7.  Pressure gain produced by Dynactor versus
          flow rate through  it
                       24

-------
Ul
                  Table  4.   DYNACTOR AIR-MOVING CAPABILITIES:  FLOW WITH INLET AND OUTLET OPEN TO
                            ATMOSPHERE, MAXIMUM PRESSURE GAIN  (NO  AIR FLOW).


Unobstructed flow
cfm
m /min
m Is
Maximum pressure gain
inch H,0
2 i
nt/m
Spray nozzle pressure
100 psig
(6.9 x 105 N/m2)

640
18.1
0.302

1.0
2.5 x 102
150 psig
(10.4 x 105 N/m2)

780
22.1
0.368

1.5
3.8 x 102
200 psig
(13.8 x 105 N/m2)

880
24.9
0.415

2.0
5.0 x 102

-------
           Pressure                    Current, I
            (psig)                       (amps)
             100            13.7, 13.5, 14.0 (average: 13.7)
             150            13.7, 13.6, 13.9 (average: 13.7)
             200            13.5, 13.6, 13.9 (average: 13.7)
 Phase angle measurements were performed by connecting vertical oscillo-
 scope trace to measure voltage betveen phases, and horizontal to
 measure a small voltage drop between two terminals of the same phase
 lead in the switch box, the impedence being purely resistive.  Phase
 angles of both pump motors were the same and did not vary with pressure.
 The resulting Lissajous figure was not a perfect ellipse, indicating
 some waveform distortion due to the motor characteristics.  The mea-
 sured angles between two adjacent phase lines were:
     a = 49.6° and /? = 8.2° and since
     a = 30° + <0 and ft = 30 - 4>; <0 = 20°,
 and thus the total power is
     P = s/3 x 207 x 13.7 x COS 20° = 4616 watts - 4.6 kW

 POWER CONSUMPTION: DYNACTOR SPRAY NOZZLES

 The pumps that drive  the Dynactor spray nozzles are both  rated at  7  hp
 (5.2 kW).   By measuring the electrical power consumed,  we determined that
 the power consumption was about 6.2 hp (4.6 kW) total with the noz-
                            2
 zles at 200 psig (13.8  N/m ) and with them at half that  pressure  as
 well,  indicating power consumption was insensitive to nozzle  pres-
 sures  below this value.   Because the power  used by the  system is
 related to the kind of motor and pump used, the most basic estimate
 of  power consumption  is to calculate the power, P, expended by the
 nozzles,  which is the product of the pressure  drop across the nozzles,
Ap,  and the volume rate of flow of  water, 0 .   These values are shown
 in  Figure  8,  the flow and pressure  data were supplied by  the  manu-
 facturer.   (The-values are plotted  on a log-log plot because  the

                                 26

-------
     10
e
'e
TO
60
•  FLOW

X  POWER

                                     /

             1  gal/min
                      -53
            6.31 x  10   mJ/min
1 P s i g

1 hp - .746 kw


 till	
                          6.89 x 103 nt/m2
                                  I
I
50  60 70 80 90 100           200

          NOZZLE PRESSURE,  psig
                                     1.0
                                                  K
                                                  •5
                                                  .3
                                                      a
                                                      si
                                         o


                                         (X.

                                         t/5
                                         M
                                         Q

                                         
-------
applicable  fluid  flow  theory,  potential  flow,  predicts  the  spray
velocity, thus  spray volume  flow  rate, will  be proportional to  the
square  root of  the  pressure  drop  across  the  nozzle;  this  is supported
by  the  data.)

The nozzle  power  dissipation is given by
    p = ^AP                                                           (1)
The data supplied by the manufacturer were:
    Q   =6.6 gal/min              at        £p •  100 psig
        = 4.2 x  10~4m3/sec)                  (Ap -  6.89 x 105 N/m2)
    Q  = 8.1 gal/min              at        AP -  150 psig
       = 5.1 x  10~4m3/sec)                  
-------
 TEMPERATURE LIMITATION

 In preliminary testing, the Dynactor spray was started and the heater
 used to raise the temperature of the inlet gas to 300°F (200°C + 5°C)
 while generating and sampling an iron oxide aerosol at 1 gr/ft .
 Although we had hoped to be able to make tests at this temperature, the
 obvious damage done to certain plastic materials in the Dynactor made
 the device inoperable.  It was agreed by all parties that future tests
 at elevated temperatures should be  done at 200°F.   We were told
 that other models of the device do  not use these plastic parts and can
 operate at higher temperatures.

 MASS COLLECTION EFFICIENCY VERSUS PARTICLE SIZE

 The remainder  of the section on results will contain material  from
 tests done to  determine  collection  efficiency,  by  mass,  as a function
 of particle size,  or more  precisely as  a function  of particle  aero-
 dynamic  diameter.   The data from these  tests are presented in  Appen-
 dix B.   Sizing was  done  by using Andersen Model  III  In-Stack Impactors
 at 14 Ipm  flow rate (0.5 cfm),  and  the  material  on a given stage  of the
 impactor was classified  as  being of an  aerodynamic diameter greater
 than or  equal  to the aerodynamic cutoff diameter for that  impaction
 stage and  less than the  aerodynamic cutoff diameter  of the impaction
 stage immediately upstream.  The cutoff diameter is  the diameter  for
which the  collection efficiency of  the  given impaction stage would be
0.50.  The  cutoff diameters were taken  from  information supplied by
the  impactor manufacturer.  Figure  9 shows four measurements of the
inlet cumulative particle size  distribution  by mass  of the  iron oxide
pigment  (Mapico Black) as generated with  the Acrison dust  feeder.
Figure 10 gives the  same information for  the  fly ash aerosol.   The cut-
off diameters  for the impactors were: 13.7, 8.6, 5.6, 4.0, 2.5, 1.3,
0.80, and 0.54 ^m.  Conditions  such as humidity and feed rate would
be expected to alter the size distributions somewhat for different
tests.  The efficiency tests measured the outlet concentration in a
certain size interval versus the inlet concentration in that size

                                 29

-------
                TEST NO.
                   17
                   19
                   12
                   14
                           X - HEAVY GRAIN LOADING, AMBIENT TEMPERATURE
                           0 - HEAVY GRAIN LOADING, ELEVATED TEMPERATURE
                           A - LIGHT GRAIN LOADING, AMBIENT TEMPERATURE
                           V - LIGHT GRAIN LOADING, ELEVATED TEMPERATURE
   99

   98


   95

   90


   80

H  70

i  6°
w  5O
n
w


PH
t/D
30

20


10
                      X
                      o
                                             o
                                             X
                                                      e
                                                      x
                                                             o
                                                             x
                                                           r
                                                           9
                                                                     o
                                                                     X
                                                                 n — i
                                                                 8 I   10
                                                                 8.6
            T
             .5
T
 2
T
 3
              .54
                            1.3
    2.5
                                                      4.0
                                                             5.6
                        AERODYNAMIC DIAMETER, D   ,
        Figure 9.   Inlet size distribution,  iron oxide  at  1000  cfra  (28 m3/Tnin)
                                      30

-------
              TEST  NO.
                 21
                 22
                 9
                 10
                         X -  HEAVY GRAIN LOADING, AMBIENT TEMPERATURE
                         0 -  HEAVY GRAIN LOADING, ELEVATED TEMPERATURE
                         A -  LIGHT GRAIN LOADING, AMBIENT TEMPERATURE
                         V -  LIGHT GRAIN LOADING, ELEVATED TEMPERATURE
   99

   96


   93


   90
1
w
w
o
80

70

60

SO

40
                                                      O
                                                      I
                                                             e
                                                            I
CO
1/3  20
                                                                     &
                                                                     f
    10


    9


    2

    I
             ,1
             .54
                      .8
                            T
                            1.3
'2   \
    2.9
                                                I
                                                3
4.0
5.9
el  10
8.6
                         AERODYNAMIC DIAMETER, Dn , am
1*7
       Figure 10.  Inlet size distribution, fly ash at  LOOO cfm  .28 m  /win)
                                      31

-------
 interval,  thus eliminating  most of the  effect of variations  in  inlet
 size distribution on  collection efficiency.

 Fractional  Efficiency of  Collection;  Factorial Tests

 A factorial test design was used to determine the Dynactor efficiency
 and the  effects, if any,  of flow rate,  dust  concentration, temperature,
 and dust type.  Two levels  of  each  of these  parameters were  used, which
       4
 meant  2  =16  tests.   Tables 5a and 5b  give  the  levels of the parameters
 for both dusts  and the number  designating  the test at each particular
 set of parameters.  The data from these tests are in Appendix B.  In
 Tables 5a and  5b the  balanced  factorial design test matrix is enclosed
 in a heavier line box.  The design  allowed testing whether or not each
 of the parameters was  significant and whether or not interactions of
 two parameters  were significant,  using  standard analysis of variance,
 and allowed estimates  of  the experimental error of measurement.  The
 other  tests done with  iron  oxide  were made to investigate the efficiency
 of a single stage, the effect  of  a  lower spray nozzle pressure, the
 effect of the addition at the  inlet of water  vapor, and the effect of
 addition of surfactant to the  spray.  Three measurements were made at
 one set  of conditions  to provide  a  second estimate of experimental
 variation.   In  all, 25 tests have been reported and the results of
 another  seven not reported  because of various  failures during the
 experiments, primarily among the  first experiments done.

 Statistical Analysis of Data

 The data analysis was designed  through a cooperative effort with GCA,
 University of Dayton Research Institute, and  the Control Systems group
 at EPA.  The computer program used to analyze the data was the BMD02V
program  from the Biomedical series, 1966 revision (Dixon, 1973).
                                 32

-------
              Table 5a.  TEST MATRIX, FLY ASH AEROSOL


500 cfm
(14 m3/min)
~ 1 gr/ft3
<~
Heated3
Ambient
1 g/m3)
#7
#8
~ 0.1 gr/£t3
(~ 0,1 g/m3)
#27
#26
1000 cfm
(28 tn3/min)
~ 1 gr/ft3
(~ 1 g/m3)
#22
#21
~ 0.
(~ 0




1 gr/ft3
.1 g/m3)
#10
#9
alnlet air T = 95°C = 200°F




blnlet air T = 21°C - 70°F
                                 33

-------
                                   Table 5b.  TEST MATRIX, IRON OXIDE AEROSOL


Heated3
Ambient
b
First stage
Steam added
Surfactant
b
Low spray pressure
500 cfra
(14 m3/min)
~ 1 gr/ft3
(~ 1 g/tn3)
1 "
1 #4
#28, 31
~ 0.1 gr/ft3
(-0.1 g/m3)
#24
#23

1000 cfm
<28 m3/min)
~ 1 gr/ft3
(~ 1 g/m3)
#19
#16,17,18
#25
~ 0.1 gr/ft3
(~ 0.1 g/m3)
#14
#12
#32
#13
1500 cfm
(42 m3/min)
~ 1 gr/ft3
(~ 1 g/m3)
#30
#29
u>
           alnlet air T = 95°C
200°F
           blnlet air T - 21°C
70°F

-------
 The following were treated as independent variables for this analysis:
 total concentration (total filter), total impactor sample, impactor
 samples on each of stages 4,  5,  6,  7,  8, and impactor final filter.
 The efficiencies of each of these eight aerosol size fractions were
 put through an analysis of variance,  using the standard F-test.

 The computer program calculated  the following for each of the size
 fractions  used:
     • "grand mean" of all 16  tests
     • "marginal  means" of sets of eight tests having one
       or the other of the treatment parameters:  dust types,
       flow rates,  inlet concentrations, temperatures
     • "sum of squares" due to each  of  the four treatment
       parameters as well  as those due  to two-factor and
       three-factor interactions  between the  parameters.
 A decision had  to  be made whether to test for the four three-factor
 interactions and have only one estimate of the experimental error  (the
 residual sum of  squares)  or to limit the tests to one-  and two-factor
 analysis and have  five estimates  of the experimental error (the
 residual sum of  squares and the  four three-factor sums  of  squares).
 The  latter  choice  was made, increasing  the sensitivity  of  the  tests
 for  the one-  and two-factor contributions.   This  meant  that the appro-
 priate F-test to use was  a  comparison with confidence  estimate tables
 for  an F-test distribution having one degree  of  freedom in the numerator
 and  five in the  denominator,  F(l,5).

 The  results  of the  factorial  test are summarized  in  Table  6.  The mean
 efficiencies  for the tests  and the  corresponding  uncertainties are
 listed with  the  aerosol fraction  to which  they correspond.  The aerosol
 fraction is  either  the  size interval between impactor cutoffs or the
 total filter  or  the  sum of  the material collected on the impactor, the
 latter having somewhat  smaller mean sizes  than the total filter due to
 losses of the very  largest particles in the drying sections.  The effi-
ciencies for  the two different dusts for the total filter  and the total
impactor have been presented  separately because these would be expected

                                35

-------
     Table 6.  SUMMARY OF RESULTS OF 16 COLLECTION
               EFFICIENCY TESTS FOR DYNACTOR
               (FACTORIAL TEST DESIGN).
Aerosol fraction
Total filter
Iron oxide
Flyash
Total impactor
Iron oxide
Flyash
4.0 - 5.6 |_un
2.5 - 4.0 [in.
1.3 - 2.5 |im
0.8 - 1.3 |im
0.54 - 0.8 p.m
< 0.54 urn
Mean
efficiency
96.04 %
93.11
98.97
93.71
89.81
97.60
99.02
98.37
93.00
75.4
27.4
47.4
Number
of
tests, n
16
8
8
16
8
8
16
16
16
16
16
16
Estimated
uncertainty3
(±
-------
  to have very  different median  aerodynamic diameters,  as  seen  from  the
  size distributions shown in Figures 9 and 10.  The uncertainty figures
  are derived from the uncertainty estimates for a single .test, which
 will be explained below.

 Tables 7a through 7h present a much more detailed picture of the results
 of the efficiency tests.  Here are listed:
      • The aerosol fraction
      • The mean of all the efficiency tests in the factorial design
      • The means of the eight tests each at two different levels
        of flow, temperature,  and concentration, and two different
        dusts
      • The results of the standard  F-test analysis  of variance,
        for an  F ratio having  a  numerator of  1 degrees of freedom
        and a denominator of 5 degree  of  freedom
      • The mean square  error  associated  with  one measurement at
        this  size fraction.
 The analysis of variance  allows one to determine what likelihood  there
 is that the  differences  noted between measurements  come from differences
 in the  parameters under  study rather  than from extraneous variations.
 The "significance level" is the probability that one  would be  correct in
 ascribing  a  difference in the results to  a difference  in  the level  of
 the parameter  under test, here  flow, dust, temperature, concentration,
making  the usual statistical assumptions  about normal  populations.
Table 8 lists  the significance  levels for the effects  of  these parameters
on efficiencies of collection for the various aerosol  fractions.

The information in the Tables 7 a through 7h,  Table 8, and in Figures 11
and 12 allow us to draw the following  conclusions:
     1.  The lower flow rate yielded higher efficiencies for
         all size fractions and the  differences were usually
         statistically significant.
     2.   Fly ash was collected with greater efficiency than
         iron oxide for  all  size fractions, and these dif-
         ferences were usually statistically  significant.
                                 37

-------
   Table 7a.   RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:  Total Filter




GRAND MEAN OF EFFICIENCY TESTS:  96.04
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 m3/min (500 cfm)
28.3 mS/min (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
-0.2 g/m3 (0.1 gr/ft3)
-2.0 g/m3 (1.0 gr/ft3)
MEAN
96.69
95.40
93.11
98.97
96.50
95.59
94.87
97.21
RESULTS OF F-TEST ANALYSIS OF VARIANCE
EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(D(4)
F (1,5)
3.95
81.99
1.99
13.04
11.93
1.98
SIGNIFICANCE LEVEL
(IF > 0.90)
> .99
> .95
> .95
MEAN SQUARE ERROR: 1.677
                                      38

-------
   Table 7b.   RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:   Total Impactor




GRAND MEAN OF EFFICIENCY TESTS:  93.71
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 m3/min (500 cfm)
28.3 mVmin (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
~ 0.2 g/m3 (0.1 gr/ft3)
-2.0 g/m3 (1.0 gr/ft3)
MEAN
94.05
93.36
89.81
97.60
94.40
93.01
91.62
95.79
RESULTS OF F-TEST ANALYSIS OF VARIANCE

EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(DO)

F (1,5)
1.19
152.6
4.83
43.50
24.33
6.33
SIGNIFICANCE LEVEL
(IF > 0.90)
> .99
> .90
> .99
> .99
> .90
MEAN SQUARE ERROR: 1.593
                                     39

-------
   Table 7c.  RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:   Impactor stage #4 ,  4.0 - 5.6




GRAND MEAN OF EFFICIENCY TESTS:  99.02
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 m3/min (500 cfm)
28.3 mVmin (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C C~ 200°F)
-0.2 g/n3 (0.1 gr/ft3)
~2.0 g/m3 (1.0 gr/ft3)
MEAN
99.21
98 . 82
98.35
99.69
99.29
98.75
98.60
99.43
RESULTS OF F-TEST ANALYSIS OF VARIANCE

EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(3) (4)

F (1,5)
1.31
15.67
2.54
6.14
4.44
2.30
SIGNIFICANCE LEVEL
(IF > 0. 90)
> .95
> .90
> .90
MEAN SQUARE ERROR: .457
                                     40

-------
   Table 7d.   RESULTS OF STATISTICAL ANALYSIS OS EFFICIENCY
PARTICLE SIZE FRACTION:  Impactor Stage #5 , 2.5 - 4.0 \j.m




GRAND MEAN OF EFFICIENCY TESTS:  98.37
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 n»3/inin (500 cfm)
28.3 m3/min (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
-0.2 g/m3 (O.L gr/ft3)
-2.0 g/m3 (1.0 gr/ft3)
MEAN
98.75
97.99
97.47
99.26
98.57
98.16
98.00
98.74
RESULTS OF F-TEST ANALYSIS OF VARIANCE

EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(3) (4)
(2) (3)

F (1,5)
10.50
57.83
3.08
9.86
20.45
3.43
2.71
SIGNIFICANCE LEVEL
(IF > 0.90)
> .95
> .99
-
> .95
> .99
-
"
MEAN SQUARE ERROR: 0.221
                                     41

-------
  Table  7e.    RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:   Impactor stage #6, 1.3 - 2.5




GRAND MEAN OF EFFICIENCY TESTS:  93.00
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 n*3/min (500 cfm)
28.3 mS/min (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
-0.2 g/n3 (0.1 gr/ft3)
~2.0 g/m3 (1.0 gr/ft3)
MEAN
j
94.82
91.17 ;
92.37
93.62
93.54
92.46
92.59
93.41 1
RESULTS OF F-TEST ANALYSIS OF VARIANCE
EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(D(4)
(D(3)
	 (I) (2) 	
F (1.5)
75.80
8.89
6.57
3.86
40.71
25.69
11.55
8.1Q
SIGNIFICANCE LEVEL
(IF > 0.90)
> .99
> .95
> .95
> .99
> .99
> .95
^ .QS 	 	
MEAN SQUARE ERROR: 0.703
                                     42

-------
   Table 7f.   RESULTS OF STATISTICAL ANALYSIS ON  EFFICIENCY
PARTICLE SIZE FRACTION:  Itnpactor stage #1, 0.8 -  1.3




GRAND MEAN OF EFFICIENCY TESTS:  75.44
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 m3/min (500 cfm)
28.3 tP/tnin (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
-0.2 g/m3 (0.1 gr/ft3)
~2.0 g/m3 (1.0 gr/ft3)
MEAN
83.45
67.44
81.87
69.01
78.95
71.94
72.79
78.10
RESULTS OF F-TEST ANALYSIS OF VARIANCE

EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(1)(2)
(D(3)

F (1,5)
3.79
2.45
0.73
0.42
2.45
1.63
SIGNIFICANCE LEVEL
(IF > 0.90)
-
-
-
-
-

MEAN SQUARE ERROR: 270.6

                                     43

-------
   Table 7g.   RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:    Impactor stage #8 ,  0.54 - 0.8




GRAND MEAN OF EFFICIENCY TESTS:  27.39
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 m3/min (500 cfm)
28.3 m3/min (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
~ 0.2 mg/in3 (0.1 gr/ft3)
~ 2.0 Big/m3 (1.0 gr/ft3)
MEAN
26.66
28.12
16.92
37.86
25.04
29.75
26.51
28.27
RESULTS OF F-TEST ANALYSIS OF VARIANCE

EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(DO)

F d,5)
0.01
2.50
.13
.02
2.22
SIGNIFICANCE LEVEL
(IF > 0.90)
_
-
-
-

MEAN SQUARE ERROR: 700.9
                                     44

-------
   Table 7h.   RESULTS OF STATISTICAL ANALYSIS ON EFFICIENCY
PARTICLE SIZE FRACTION:   Final  filter  after  impactor, < 0.54




GRAND MEAN OF EFFICIENCY  TESTS:   47.38
MARGINAL MEANS
PARAMETER
FLOW
DUST
TEMPERATURE
CONCENTRATION
LEVEL
14.2 mS/min (500 cfm)
28.3 n^/min (1000 cfm)
IRON OXIDE
FLY ASH
~ 20°C (~ 70°F)
~ 95°C (~ 200°F)
-0.2 g/m3 (0.1 gr/ft3)
-2.0 g/m3 (1.0 gr/ft3)
MEAN
49.26
45.50
39.94
54.82
51.36
43.40
23.51
71.25
RESULTS OF F-TEST ANALYSIS OF VARIANCE
EFFECT
(1) Flow
(2) Dust
(3) Temperature
(4) Concentration
(2) (4)
(D(3)
F (1,5)
0.26
4.05
1.16
41.62
7.04
4.60
1.83
SIGNIFICANCE LEVEL
(IF > 0.90)
.90
> .99
> .95
> .90

MEAN SQUARE ERROR: 219.0
                                      45

-------
Table 8 .   SIGNIFICANCE OF EFFECTS OF FLOW,  DUST,  TEMPERATURE,
           AND CONCENTRATION ON SCRUBBER COLLECTION EFFICIENCY
Aerosol fraction
Total filter
Total impactor
4.0-5. 6 |am
2.5-4.0 (jm
1.3-2.5 urn
0.8-1.3 |am
0.54-0.8 mn
< 0.54 jim
Significance level
Flow
~ 0.90
--
--
> 0.95
> 0.99
~ 0.90
--
--
Dust
> 0.99
> 0.99
> 0.95
> 0.99
> 0.95
--
--
~ 0.90
Temp.
--
> 0.99
--
—
> 0.95
--
--
—
Cone.
> 0.95
> 0.99
> 0.90
> 0.95
~ 0.90
--
--
> 0.99
                              46

-------
PERCE
EFFICIENCY
0
O


99
98

95
90
80
50
10
O
- f GRAND MEAN
. + ~ 1 gr/ft3
(~ 1 g/m )
" 0 ~ .1 gr/ft3
(~ 0.1 g/ni

»
-
O
-
o
?
:° , ,
-»•
j
( i
O
O





1 1
.
4
1 1.
J 0
-
O
: 8
-
*
I GRAND MEAN
0
+ FLY ASH
O IRON OXIDE
•0 +
1 1 1 1
                                                                     0.5
                                                                     1.0
                                                                     2.0
                                                                    5.0
                                                                    10
I
w
Ed
                                                                    20
                                                                   50
                                                                   100
      OI254OI      2345

              DIAMETER, urn                  DIAMETER, ^m

     Figure 11.   Dynactor scrubber collection efficiency versus particle

                 aerodynamic diameter, effects of loading and dust type
                                     47

-------



99



98
g
w
*• »»
s
H
1 "°
H
a
•j
O
O
90





50

10
O
- J GRANT
_ , 500 c
. (14.i
.
Q 1000
w (28.:



-

-
-
_
-



^

•

J

G
V
?
-of*

1
0 1
) MEAN
:fm« : ;
3
i m /dxi)
( i
cfm
5 m Anin) + ( >
J

O


+
ff
O















1 1 1
-
| +
m
" 1 1.
O
i
o

-

•
i i
,
.





B ^
•f
I
o
J GRAND MEAN
+ 20°C (70°F]
_^ AC f^ f O/1O • 1? '
^ ^1 !F J ^* \ £*\AJ C 4
-^ ^^ -
1
f

1 1 1 1
0.5


1 O
1 .W


2.0


5.0


10





2O





50


OO
23401 2345
           DIAJETER, |jm                  DIAMETER, ^In
Figure 12.  Dynactor scrubber collection efficiency versus particle
            aerodynamic diameter, effects of flow rate and inlet
            temperature
                                                                     w
                                                                     P-.
                                                                     H
                                                                     U
                                48

-------
     3.   The lower temperature produced greater efficiencies than
         did the higher temperature, in seven of eight aerosol
         fractions,  but this was statistically significant in only
         two fractions.
     4.   Higher concentrations were collected with greater
         efficiency  than lower concentrations and this was sta-
         tistically  significant in most fractions.
The F-test analysis of variance also gave information on the significance
of interactions between the variables flow (1), dust (2), temperature
(3) and concentration (4).  Thus the interaction labeled (2) (4) in
the Tables 7a,  b, c, d, e, h is the interaction of dust and concentra-
tion, which can be interpreted in either of two ways:  the degree to
which different dusts gave a different dependence of efficiency upon
concentration or_ the degree to which different concentrations gave a
different dependence of efficiency upon dust type.  In Table 9 are
arrayed the averages of four tests each at the particular combinations
of parameters whose interaction was found to be significant for the
1.3 to 2.5 |_im aerosol fraction, Table 7e.  Thus, the first group con-
tains two levels of concentration and two types of dust, from which
we see that the difference in efficiency between high and low concen-
trations was more pronounced for iron oxide than it was for flyash.
The next set in Table 9 shows that the difference in efficiency between
the two types of dust was greater at the lower flow rate than at the
higher.  Similar interpretations would be appropriate for the other two
sets of data in Table 9: the effect of temperature was greater at the
higher flow rate and the effect of concentration was greater at the
lower flow rate.  The F-test values in Table 7e indicate all these
statements are statistically significant.  The flow interactions are
displayed in Figure 13.  The explanations of all these interactions
for all the stages would be difficult to make, but an example of one
such explanation would be that the difference in efficiency for the two
dusts is caused by collection mechanisms that increase with increased
residence time, thus being more significant at the lower flow rate than
at the higher.
                                 49

-------
Table 9.  DETAILED ANALYSIS OF INTERACTIONS FOR DYNACTOR
          EFFICIENCY ON 1.3 - 2.5 jim AEROSOL FRACTION
          (STAGE #6)
Efficiencies (N=4)
Concentration (4)
~ 0.1 g/tn
3
~ 1.0 g/m

Dust (2)
Iron oxide
Fly ash
Temperature (3)
20°C
95°C
Concentration (4)
3
~ 0.1 g/m
3
~ 1.0 g/m
Parameter: - Dust (2)
Iron oxide
90.6
94.2
Fly ash
94.6
92.7
Parameter: Flow (1)
500 cfm

93.6
96.0
94.6
95.0
93.4
96.3
1000 cfm

91.2
91.2
92.4
89.9
91.8
90.6
                             50

-------
   IOO
2
W
H
U
M
W


§
P"^
8
    98
    96
    94
    90
     89
                         Iron oxide

                          Fly ash
                               _L
                                                A  20°C
                                       A  95°C
                                                        _L
                                             QO.l g/nf


                                             • l. g/m3
                                                                              _L
m
            /min
        500
        14.2
1000
28.3
500
14.2
                                                     1000
                                                     28.3
500
14.2
1000
28.3
          Figure  13.  Collection efficiency, 1.3-2.5 (jim,  versus  flow  for
                     different dusts, temperatures, and  concentrations

-------
Estimates of Experimental Error

Two methods were used to get estimates of the experimental error.  The
first involved dividing the residual sum of squares by the number of
degrees of freedom (5) and taking the square root, for the data in the
          4
balanced 2  test matrix used in the analysis of variance.  This is a
standard method.  The residuals were those from the analysis of the
individual factors (flow, dust, temperature, and concentration) and
two-factor interactions.  Such residuals would include three-factor
interactions, if any, and mistakenly treat them as experimental error.
(Nearly always, these three-factor contributions to the sum of squares
were of the same magnitude as the residuals left even after the three-
factor interactions were taken into account.)  The second estimate of
the experimental error came from the three replications for iron oxide
                 3                33
at 1000 cfm (28 m /min), ~ 1 gr/ft  (~ 1 g/m ), at ambient temperature.
Both are given in Table 10.  In six of eight cases, the estimates from
the replication tests' were lower.  The standard deviation of an average
of N such tests would be the standard deviation for one test divided
    1/2
by N   .  For any one of the measurements to be significantly different
from a given value, it should be at least two standard deviations dif-
ferent from that value, which gives a guideline for comparing the
results of the tests done outside the factorial design test matrix.

Collection Efficiency Tests at 1500 cfm (42 m /min)

An 8-point pitot tube traverse was used downstream in 8 inch (0.20 m)
                                                          •j
diameter duct and yielded an outlet flow of 1510 cfm (A3 m /min).
Twenty point pitot tube traverse used upstream in 12 inch (0.30 m)
                                                     3
diameter duct yielded an inlet flow of 1250 cfm (35 m /min).  Magne-
helic gauges used for the above traverses were checked with an upright
manometer.  They were in good agreement which determined that the
gauges were not the cause of the above flow discrepancy.
                                 52

-------
Table 10.  ESTIMATES OF EXPERIMENTAL ERROR FOR SINGLE TEST
           FROM THE RESIDUALS OF THE SUM OF SQUARES AND
           FROM THE REPLICATION TESTS, Nos. 16, 17, 18
Aerosol fraction
Total filter
Total impactor
4.0-5.6 urn
2.5-4.0 urn
1.3-2.5 urn
0.8-1.3 (itn
0.54-0.8 urn
< 0.54 um
Standard deviation estimates
(M.S.E)1/2a
a)
1.29
1.26
0.68
0.47
0.84
8.4 (17.7)C
26.5
14.8
0.59 x range
a)
0.41
1.00
0.18
0.59
0.12
3.7
34.0
11.2
            mean square of residual sum of squares
       after two-factor analysis of factorial design
       tests.

       Best estimate of standard deviation from range
       of three measurements (Wilson, 1952).

      °Value including a discarded datum, test No.
       10 (12.5)
                             53

-------
 An attempt was made to seal leaks in the Dynactor and peripheral ducting,
 but it must be assumed that with a static pressure loss greater than
 2 inch 1^0 (5 x 10  nt/m ) at the outlet some leakage occured.  Flow
 through the Dynactor was calculated as the average between inlet and
 outlet volumes, but grain loading was calculated on the basis of the
 flows measured upstream and downstream.   Efficiency was determined by the
 ratio of the rate of particulate mass flow out of the Dynactor to the
 rate of particulate mass flow into the Dynactor.

 The results for the tests at 1500 cfm (42 m3/min) and at 1000 cfm
 (28 m /min) for iron oxide at 1  mg/m  are given in Table 11 to facilitate
 comparison.  At ambient temperature,  the lower flow rate produced a
 higher efficiency in six of the  eight aerosol fractions measured and
 indistinguishable efficiencies in two (4.0 -  5.6 ^m,  0.54 urn).   In
 three cases (total impactor,  2.5 to  4.0  urn,  1.3 to 2.5 um) the
 higher flow rate  value  was more  than  two of its estimated standard
 deviations  away from the  low  flow rate mean,  which would support the
 conclusion  that at ambient temperature the  effect of  increasing  the
 flow lowered  the  efficiency.  At the  elevated temperature, 95°C  (200°F),
 raising  the flow  seems  not to have decreased  the efficiency.  Three
 efficiencies were  lower and three were higher.   The data from the
 tests at 1000 cfm (28 m3/min) and  1500 cfm  (42  m/min)  displayed  lower
 efficiencies for increased  flow  at 20°C  but not at 95°C;  at 1000  cfm
 (28 m /min), six of six fractions showed higher  efficiencies  at 21°C
 than  at  95°C, but at 1500  cfm (42 m /min) six of seven fractions  showed
 higher efficiencies at 95°C than at 21°C, suggesting an  interaction
 between  the two variables.

 Collection Efficiency Tests With Inlet Air Humidified

Efficiency tests were performed using heated  air  to evaporate a fine
water spray upstream from the Dynactor inlet  to use the subsequent con-
densation of water vapor to enhance particle  collection  (diffusiophoresis

                                54

-------
Table 11.  RESULTS OF DYNACTOR COLLECTION EFFICIENCY TESTS UNDER
           SIMILAR CONDITIONS EXCEPT FOR FLOW RATE  (IRON OXIDE,
           ~ 1 g/m3)
Aerosol fraction
Total filter
Total Impactor
4.0-5.6 nm
2.5-4.0 urn
1.3-2.5 urn
0.8-1.3 urn
0.54-0.8 urn
< 0.54 urn
Test number
21°C (70°F)
1500 cfm
(42 m3/min)
94.8
88.9
> 97.5
97.0
87.1
75.0
< 19.4
--
29
1000 cfm
(28 m3/min)
95.4
93.9
> 99.2
98.5
93.1
83.7
46.4
< 82.5
16,17,18 avg.
95°C
1500 cfm
(42 m3/min)
94.7
93.8
> 98.4
98.6
89.4
83.1
75.8
--
30
200°F)
1000 cfm
(28 m3/min)
94.9
91.7
99.1
97.6
89.9
75.9
--
66.1
19
                               55

-------
The spray was Introduced downstream from the inlet sampling probes.  Two
                                                          5     2
4 1511 Sprayco pinjet nozzles operated at 80 psi  (5.5 x 10   N/m ) were
used, spraying 0.56 gal/min (3.5 x 10~5 m3/s) into air at 350°F (177°C).
The humidity ratio was determined two ways: by weight change of dessicant
material after the gas passed through a cyclone (10 mm) and by wet bulb/
dry bulb thermometer measurements (and comparison with a psychometric
chart).  During test #28, the wet and dry bulb thermometers were
inserted into the duct immediately upstream from  the Dynactor inlet.
During test #31 the gas sample was made to pass not only through a
cyclone but also through a heating section into which the wet bulb and
dry bulb thermometers were placed, and from which it passed into the
dessicant.  The ratio, by weight, of water vapor  to dry air for test
#28 was 0.032 as determined by dessicant weight gain and 0.057 as
determined by the thermometric measurements.  Some of this discrepancy was
thought to be due to condensation in the sampling line to the dessicant,
one of the reasons the heated section was introduced for test #31.  For
test #31, the humidity ratio was 0.070 as determined by the dessicant
and 0.071 and 0.086 as determined by thermometric measurements near the
beginning and end of the test.  No intentional changes were made in the
system in producing the higher humidity ratio the second time, so al-
though we have reported the tests as having two different average
humidity ratios (0.044, 0.076), they may have had nearly the same.

The data from these two tests are given in Table  12, along with data
from two tests most nearly matching the conditions of the steam addition
tests.  In each aerosol size fraction the ranges  of the steam added
tests and the tests without added steam overlap,  with the exception of
the interval 1.3 to 2.5 urn, for which the efficiencies without steam
addition are higher than those with added steam.  The addition of water
vapor to achieve an inlet humidity ratio of about 0.05 g water per g dry
air did not appreciably enhance collection efficiency, an unexpected
result.
                                 56

-------
Table 12 .  RESULTS OF DYNACTOR COLLECTION EFFICIENCY TESTS UNDER
           SIMILAR CONDITIONS BUT WITH AND WITHOUT STEAM ADDITION
           (IRON OXIDE, 500  cfm = 0.24 m3/sec, ~ 1 g/m3)
Aerosol fraction
Total filter
Total impactor
4.0-5.6 Urn
2.5-4.0 nm
1.3-2.5 urn
0.8-1.3 |im
0.54-0.8 ^m
< 0.54 (im
Tests
No steam
#4
96.4
93.9
99.1
98.3
96.0
85.2
--
73.3
#53
91.1
94.3
> 99.2
99.1
97.5
89.3
--
72.0
Avs.
93.8
94.1
> 99.1
98.7
96.8
87.2
--
72.6
Steam added
#28
--
95.4
> 99.3
99.3
95.0
85.7
48.4
< 64.4
#31
96.2
93.0
> 98.9
98.9
94.0
80.3
70.0
< 80.0
Avg.
96.2
94.2
> 99.1
99.1
94.5
83.0
59.2
< 72.2
  aThis  test done
at 95°C (= 200°F)
                                57

-------
 Collection Efficiency Test With Surfactant Added to the Spray

 The  addition of a  surfactant chemical to the spray used in the Dynactor
 might be  expected  to affect efficiency by changing the droplet size
 distribution in the  spray  somewhat and changing the wetting efficacy
 of the droplets, both changes due to the lowering of the surface tension
 of the water in the  spray.   A surfactant was added to the water
 reservoir in the Dynactor*s stages to form a 10 parts per million
 solution.  An efficiency test (test #15)  was run using this solution,
 but  otherwise at the conditions at which the replicative tests
 (#17, 18, 16) were run; that is,  1000 cfm (28 m3/min), ~ 1 gr/ft3
        3
 (~ 1 g/m  ),  iron oxide aerosol, ambient temperature.   Table 13 contains
 the  efficiencies from the  two types of test.   In six of six aerosol
 fractions,  the surfactant  additive gave a higher efficiency than had
 been obtained without it,  and in the other two fractions there were
 no distinguishable differences.   In the range 1.3 to 4.0 |am,  the
 efficiencies were  more than one standard  deviation apart but  less  than
 two.   For most applications,  the  differences  would not be substantial.

 Collection Efficiency Tests At  Reduced  Nozzle Pressure

 One  efficiency test was made  using the  Dynactor  spray  nozzles  at 100 psig
          5      3
 (6.9  x  10  N/m ), half their normal pressure.   This test,  #13  listed
 in Appendix B,  is directly comparable with  test  #12, also  listed in
 Appendix  B, and in every size fraction  it gave a  lower  efficiency.  The
                                                 3                   3
 total efficiency on iron oxide  at  1000  cfm  (28 m /min)  at ~ 0.1  g/m
 (~ 0.1 gr/ft  )  at ambient temperature was 81.8 percent  for  this
 reduced pressure, compared with 90.0 percent  at  the usual nozzle
 pressure  for  these same conditions.

 Some  of these results are discussed more fully in  the next  section,
which also contains a comparison between the Dynactor scrubber and  con-
ventional  venturi scrubber  technology.
                                 58

-------
Table 13 . RESULTS OF DYNACTOR COLLECTION EFFICIENCY TESTS UNDER
          SIMILAR CONDITIONS EXCEPT FOR THE ADDITION OF 10 ppm
          SURFACTANT TO THE SPRAY WATER


Aerosol fraction
Total filter
Total impactor
4.0-5.6 fim
2.5-4.0 (am
1.3-2.5 jitn
0.8-1.3 \an
0.54-0.8 urn
< 0.54 \m
Test number
Collection efficiencies
Surfactant
added
95.5
94.4
> 99.2
99.2
94.2
84.9
62,2
< 82.2
25
No
surfactant
95.4
93.9
> 99.2
98.5
93.1
83.7
46.4
< 82.2
16,17,18 avg.
                                59

-------
                            SECTION VI
                            DISCUSSION

In  this section, we compare the Dynactor Scrubber with  a widely-used
control device which also uses the  collection of particles by spray
droplets,  the venturi scrubber.  Costs are compared  for the two dif-
ferent scrubbers when used so as to obtain very similar efficiencies.
The effects of the factors studied  with the factorial design tests is
discussed  as well.

DYNACTOR SCRUBBER VERSUS VENTURI SCRUBBER

Both the typical venturi scrubber and the "ejector venturi" (Harris,
1964) use  a fine water spray to clean a gas stream.  The latter uses
this spray to supply some of the motive power to the air stream.  Fair-
ly similar flow versus pressure gain curves are obtained with the
Dynactor scrubber operating at 200  psig and 18.4 gpm water spray
           5 •    2              — 3 3
(13.8 x 10  N/m  and 1.17 x 10  m/s) and with the ejector venturi
                                         5     2              -3 3
operating  at 40 psig and 24 gpm (2.8 x 10   N/m  and 1.53 x 10  m /s),
according  to the published data (Harris, 1964), which would mean that
the latter acts as a fan about 4 times more efficiently than does the
Dynactor.   The collection efficiency information on the ejector venturi
is sketchy, but at a comparable power consumption by the nozzles in the
                                    o
two systems (2 hp/1000 cfm, 3.2 kw/m /s) the observed mass efficiency
for the ejector venturi (Takishima  et al., 1961) was 97.8 percent and
for the Dynactor scrubber 99.0 percent, for fly ash aerosols having
mass median diameters near 6 ^m.   Thus the two-stage Dynactor yielded
half the penetration compared to an ejector venturi operating at the
same spray input power per volume flow of air.   Much more experience
                                60

-------
                                 io3
  has been gained with the usual venturi scrubber, so that the remainder
  of  this  comparison segment will concern the Dynactor and the conven-
  tional venturi scrubber.  The prime function of the Dynactor scrubber
  is  to remove particulate material rather than to aid flow,  so the
  comparison should be made with venturi scrubbers having the same
  collection efficiency.   Figure 14 is derived from the Scrubber Handbook
  (Calvert  et al. ,  1972).   It gives the theoretical collection efficiency
  of venturi  scrubbers as  a function of the air flow pressure drop  pro-
  duced by  the venturi;  these curves match experimental results obtained
 with Venturis rather well.   From the grand mean mass collection effi-
  ciencies measured  for  the Dynactor we have plotted the interpolated
 collection  efficiency  at 1,  2,  and 4 urn aerodynamic diameter  to obtain
 the pressure .drop  associated with a venturi scrubber having the same
 collection  efficiency.   The  data indicate  that  the  corresponding
 venturi scrubber pressure drop  would be about 40-inch  H^O or  10 x
    2                                                    2
 N/m .   Using  a 40-inch  H20  venturi as  an  equivalent for comparison,
 we investigated the  question of relative costs.  Both  devices use
 water sprays and thus must consume  clean water., dispose of  contaminated
 water.  Both would be  expected  to have similar  maintenance  problems.

 Power Consumption

 The  Dynactor power consumption  is the power used by  the nozzles divided
 by the  efficiency of the pumps  and  associated plumbing.  The Dynactor
 nozzles use a total theoretical power (flow times pressure drop) of
 1.6  kw  for the 1000 cfm treated by  the DY-12-F2.  (The slight air  flow
 pressure  drop is negligible.)  The venturi scrubber power consumption
 is the product of the air flow  times the pressure drop, which would be
 4.7  kw for 40-inch H20 and 1000 cfm, plus any power other than the air
 flow pressure drop used in supplying the water spray.  Thus, comparable
 Dynactor  scrubbers and venturi scrubbers have intrinsic power  require-
•ments which  are nearly three times lower for the Dynactor.
61

-------
                                       4    2
                                     10  N/m
§
I-H


I
H
W
2
u
OU
    1.0

    0.9

    0.8

    0.7

    0.6


    as


    0.4



    0.3
    0.2
                 0.5
                      1.0
                   1.5
2.0     2.5
3.0
3.5
0.10

0.09

0.08

0.07

0.06


0.05



0.04




0.03
    0.02
    0.01

    0.0091-

    0.008

    0.007

    0.0061-
    0.005
 T    I     I
                                                          1    T
                                           O  DYNA.CTOR

                                              DY-12-F2

                                          	  VENTURI

                                              SCRUBBER
                                                                    o.oo
                                                        O.5O
                                                                        0.90
                                                                              CJ
                                                              UJ
                                                                        O.95
                                                                    O.99
                20
                                                        120
    Figure 14.
         40      60       80      100


          VENTURJ  PRESSURE DROP. A P

Collection  efficiency curves for venturi  scrubbers

and Dynactor efficiency data
        140
                                    62

-------
Water Consumption
The two-stage Dynactor  scrubber  we  tested  worked with nozzles having a
                                              -3  3                3
flow of 18.4 gal/min  for  1000 cfm or  2.5 x 10  m /s water per 1 m /s
of air.
Cost estimates  in the  Scrubber Handbook (Calvert  et  al.,  1972) used
                3             -33            3
8.4 gal/1000  ft  or  1.1  x 10   in  water per m  air as  typical for
venturi  scrubbers, which is somewhat less  than  half  that  used by the
Dynactor.

The amount  of water  actually consumed by both systems  would  depend on
the dust loadings and  on their relative abilities to recirculate the
contaminated  water.
Costs

Scrubber  applications vary, as do the trade-offs made  by scrubber
manufacturers in meeting their user's requirements.  Any cost  compar-
 ison has  to be used with an awareness of the uncertainties  in  such
 projections, and this one is no exception.  The one  which follows has
 an added  difficulty: it is based on two different sources of data,
 the open  literature on venturi scrubbers and the data  supplied by RP
 Industries, manufacturers of the Dynactor scrubber.
                                                                 2
We chose  to compare the two types of scrubber at 40,000 cfm (19 m  /s),
which  is  more nearly typical of industrial applications than the 1000
 cfm at which the Dynactor DY-12-F2 was tested.  The  systems were
 assumed to be operated 8,000 hours a year, nearly full-time.

 Following is an explanation of the information contained in the tables,
                                 63

-------
Estimated Capital Investment - Capital  investment  is  the sum  of  pur-

chased equipment cost and installation.    Venturi installation

was assumed to be 200 percent of  the  purchased  cost,  from material  in

the Scrubber Handbook (Calvert et al.,  1972).


Annual^ Operating Costs - The annual operating cost includes the  follow-

ing fixed, variable, and semi-variable  costs.


Fixed  costs -
      1.  Amortization of capital  investment  -  The  capital  invest-
         ments have been amortized  over  a period of  20 years.
         This reflects  the  expected lifetime of the  equipment
         based upon literature  review.   (Blecker and Nichols,
         1973)

      2.  Interest  on loan - An  interest  rate of 8  percent  of the
         total capital  investment was used.  It was  further
         assumed that the interest  is to be  paid after one year,
         but  is capitalized uniformly over the estimated 20-
         year lifetime  of the equipment.   This method is used  in
         similar engineering estimates.   (NAFCA, 1969)

      3.  Insurance - The cost of  insurance was estimated to be
         1.0  percent of total" capital investment.  This figure
         is suggested by Peters and Timmerhaus (1968) as a
         reasonable estimate.
Variable and  semi-variable  costs  -
      1.  Labor  and maintenance  - Annual  labor and maintenance
         was estimated to be  1.4 percent of  the  installed equip-
         ment cost.  This represents  the lower end of the range
         for this type of equipment.   (Blecker and Nichols,
         1973;  Calvert, Goldschmidt,  Leith, Mehta, 1972)

      2.  Electric power - Electric  power requirements for the
         high energy venturi  scrubber were obtained from the
         Scrubber Handbook.   (Calvert et al., 1972)  Power
         needed for the Dynactor System  was obtained from
         manufacturers literature.  Power costs were assumed to be
         be $0.025 per kilowatt-hour. (Boston Edison rate
         schedule, 1974)
                                64

-------
         Water - Water requirements for the Dynactor system
         were obtained from manufacturer's literature.  Water
         needs for the venturi scrubber were obtained from the
         Scrubber Handbook.  Water costs were assumed to  be
         $0.50 per 100 cubic feet. (Boston Public Works Dept.,
         July 1974)
  Table 14a.  Estimated capital cost of Dynactor system,  based on
              manufacturer's data  (1973 dollars)
FLOW CAPACITY:
         40,000 cfm = 1.13 x  103 m3/min = 18.9 m3/s
TOTAL CAPITAL INVESTMENT                                     $120,000
    Purchased Equipment                         $60,000
    Installation                                $60,000
            Estimated Operating Cost of Dynactor System
TOTAL CAPITAL INVESTMENT  (C.I.)                              $120,000
FIXED COST  (ANNUAL)
    Amortization  at 5% C.I.                                  $  6>000
    Interest on Loan  (8%  of C.I.)                   •         $    48°
    Insurance  (1.0% of C.I.)                                 $  1,200
VARIABLE AND SEMI-VARIABLE COST  (ANNUAL)
    Labor and Maintenance (1.0% of  C.I.)                     $  1,200
    Electric Power ($0.025/kw.-hr.)                         $ 11,936
    Water ($0.50/100  cubic feet)                             $  4»825
  Paid  in one  year;  amortized over 20 years.
                                 65

-------
Table 14b.  Estimated capital cost of high energy Venturi scrubbers,
            yielding similar collection efficiency to Dynactor
            (1973 dollars)
FLOW CAPACITY
     40,000 cfm = 1.13 x 103 m3/min = 1.89 m3/s
TOTAL CAPITAL INVESTMENT                                     $147,000
    Purchased Equipment                         $49, 000
    Installation                                $98,000
      Estimated Operating Cost of High Energy Venturi Scrubber
TOTAL CAPITAL INVESTMENT (C.I.)                              $147,000
FIXED COST  (ANNUAL)
    Amortization  at  5% C.I.                                  $  7,350
                                 #%
    Interest on Loan (8% of C.I.)                            $    588
    Insurance  (1.07.  of C.I.)                                 $  1,470
VARIABLE AND SEMI-VARIABLE COST  (ANNUAL)
    Labor  and Maintenance  (1.0%  of C.I.)                     $  1,470
    Electric Power  ($0.025/kw.-hr.)                          $63,300
    Water  ($0.50/100 cubic feet)                             $  6,420

a Paid  in  one year;  amortized over a  20-year period.
From the material on costs in the foregoing two tables, we conclude that
the primary difference in costs between the Dynactor system and an equiv-
alent venturi scrubber is the difference in electrical power costs, esti-
mated here to be a difference on the order of $50,000/yr for treating
                3
40,000 cfm (19 m /s) in favor of the Dynactor, based on $0.025/kw.-hr.
For the 1000 cfm model we tested, the difference in intrinsic power re-
quirements was a factor of three to one in favor of the Dynactor in com-
parison with an equivalent venturi.  Using this ratio, the difference
between Dynactor power cost per year and the power cost per year of an
equivalent venturi would be expected to be $42,000 for treating 40,000
cfm.  These yearly savings would be about 407, of the capital investment.

                                66

-------
EFFECTS OF FLOW, DUST,  TEMPERATURE,  CONCENTRATION,  ETC.

As noted in the results,  there  were  statistically  significant effects
on collection efficiency  due  to flow,  dust,  temperature, and concentra-
tion of particulates.   Here we  will  discuss  these  effects and link
these results with  those  of others.
Flow
Increasing the  flow  rate  will  increase  the  velocity  gradients, which
would be expected  to increase  deposition due  to  impaction and intercep-
tion and to increase the  turbulent eddy diffusivity, which is a linear
function of the Reynolds  number  (Calvert et al.,  1972), and  thus
increase the rate  of mass transfer to the droplets.  Increasing the
flow rate will  also  decrease  the residence  time,  which would give less
time for the collection mechanisms to act,  significant for the smallest
particles, where  diffusion would predominate  the collection  mechanisms,
and for the largest  particles, f°r which settling would become  import-
ant.  We found  somewhat higher collection efficiencies at the lower
flow rate, as did  Lancaster and  Strauss (1971) in their experiments
with spray scrubbers, using ZnO  particles with a number median dia-
meter of 1.0 jam.

Dust

Particles of fly  ash were consistently collected with greater efficiency
than particles  of  iron oxide  having the same  aerodynamic  diameter  in
our tests.  Different aerodynamic behavior  by particles having the same
aerodynamic diameter is unexpected, but perhaps  the  particles differed
in  the likelihood  with which  a particle/droplet  collision produced
capture or in the  degree  to which they served as nuclei during condensa-
tion or the degree to which high humidities facilitated their agglomera-
tion.  The iron oxide powder  seemed less hydrophilic than did the fly
ash, thus the iron oxide  would have been less wettable and more difficult

                                  67

-------
for the water droplets to entrap.  Lohs (1969) found that making
hydrophobic polystyrene particles into hydrophilic ones, by coating
their surfaces with a wetting agent, increased the capture of these
particles by a spray scrubber.  From their experiments with venturi
scrubbers, Calvert, Lundgren, and Mehta (1972) concluded that particle
wettability enhanced collection efficiency.

Temperature

Temperature can influence collection efficiency in a variety of ways.
Higher temperatures means higher viscosity for gases; for example, as
air goes from 20°C to 100°C, its viscosity increases a factor of
1.20  (Bird et al., 1960), which increases its resistance to particle
motion, hindering the various collection mechanisms.  For the sub-
micron particles, this can be offset by the increase in the Cunningham
slip  correction factor as temperature increases and by the increase
in the particle diffusivity due to Brownian motion.  Our experiments
showed a statistically significant decrease in collection efficiency
for 1.3-2.5 urn aerodynamic diameter particles as temperature increased
from  20 C to 95 C, as well as a general trend toward decreased effi-
ciencies for all the aerosol size fractions.  Lancaster and Strauss
(1971) measured a decrease in efficiency in going from 20°C to 30°C with
a spray scrubber operating on water-saturated air containing particulate
material.

Concentration

As particle concentration increases, particle agglomeration increases
due to coagulation.  Increased agglomeration means an aerosol having
larger mean size, which generally enhances collection efficiency in
spray scrubbers.  Increased concentrations yielded higher collection
efficiencies in all the aerosol size fractions in our tests.  Lancaster
and Strauss (1971), among others, reported increased efficiency with
                                68

-------
increased mass loading.   In our  tests with the Dynactor, the improve-
ment in collection was most dramatic for the smallest particles,
indicative of coagulation.

Surfactant Addition

The addition of wetting  agent, surfactant, to the spray water lowers
the surface tension  of the water, which would mean it improves the
ability of the droplets  to wet and  engulf particles and it tends to
decrease the droplet size of  the spray, the latter being determined
by the equilibrium between the force of surface  tension and the forces
tending to break  up  the  droplets.   Improved wetting should improve
collection efficiency in situations where poor wetting is an inhibitor,
which we believe  was the situation  with iron oxide.  Smaller droplet
sizes generally improve  collection  efficiency as well, for a given
droplet mass concentration.   Our results showed  a trend toward a slight
improvement in efficiencies using a surfactant additive.

Water Vapor Addition, Diffusiophoresis, Thermophoresis

Lapple and Kamack (1955) were among the first of many to note that the
addition of steam upstream from  a scrubber could produce substantial
improvements in collection efficiencies.  Lohs  (1969) attributed the
enhancement of efficiency, which he too measured, to the following
causes:  condensation on the  particles which made them into relatively
massive droplets, improved adhesion between particles and particles
and between particles and spray  droplets, diffusiophoresis.  Sparks
and Pilat  (1970)  calculated  the  contribution  of  diffusiophoresis to
collection by  spray  droplets  and concluded  the  effect could be dramatic,
increasing for  smaller  particles and  lower  gas/droplet relative
velocities.  On  the  other hand,  Slinn and Hales  (1971) analyzed the
roles played by  thermophoresis and  diffusiophoresis  in  the scavenging
of atmospheric  aerosols by  cloud droplets  and concluded  that thermo-
phoresis generally predominates.
                                  69

-------
If the condensation process begins with a drop which is at the ambient
temperature, then the condensation heats the drop and this heating
produces an opposing thermophoresis; if evaporation begins with the
drop at ambient temperature, these two mechanisms are reversed and
still oppose each other.  Relatively hot droplets which evaporate will
repel particles due to both diffusiophoresis and thermophoresis until
they cool to below ambient temperatures; relatively cool droplets
which condense moisture will attract particles by both mechanisms
until they heat to above ambient temperatures.  For droplets starting
at the ambient temperature (Too), condensation causes the droplet to
heat up due to the latent heat of condensation, X, and thermal forces
inhibit collection.  An approximate ratio of the two flux forces was
given by Slinn and Hales as

                              M  X
        thermophoresis   _     a	
        diffusiophoresis     5 R I*,
where
    M  « "molecular weight" of air,
     cl
    X  = latent heat of evaporation/condensation
    R  = gas constant
    TO,, = temperature of the fluid medium
from which they computed that thermophoretic transport will exceed
diffusiophoretic transport by a factor of about 6 and oppose it.
Lancaster and Strauss (1971) tried to separate the effects of these
flux force mechanisms from the effects of particle size increase and
adhesion improvement due to water vapor.  They used  cold and hot  sprays
to scrub water-saturated air and found no improvement with the cold
spray, which would have produced greater condensation upon the spray
droplets and would have produced an enhancing thermophoresis.  Lan-
caster and Strauss did find that steam addition helped increase col-
lection efficiency, decreasing the particle penetration by a factor
of (1- 5Q), Q being the mass ratio of steam to air; this increase
they attributed to particle build-up due to condensation.   A recent

                                 70

-------
survey of "flux force/condensation scrubbing," the use of steam with
spray scrubbers, concluded that particle growth probably predominates
over diffusiophoresis as the enhancement mechanism (Calvert et al.,
1973).   The relative roles of these mechanisms seem as yet in question,
but they are important considerations in the design of scrubbers: if
the improvement in collection efficiency is due to particle build-up,
then steam should be added as early in the flow as practicable; if
diffusio-phoresis is enhancing collection and doing so more vigorously
than particle build-up, then the  steam should be added just shortly
before the scrubbing takes place.

Although we did not  find  a major  improvement  in efficiency when we
used ~ 0.05 water  vapor  to air  ratio,  this ratio was  about one-third
that ratio recommended  in the  study  of "flux  force scrubbing"  by
Calvert et al. ,  (1973).
                                 71

-------
                            SECTION VIII


                             REFERENCES
 1.  Bird, A. N. Jr.,  J.  D. McCain,  and  D.  B.  Harris.   Particulate
     Sizing Techniques for Control Device Evaluation.   A.P.C.A. Meeting.
     Chicago.  Paper 73-282,  1973.

 2.  Bird, R. B., W. E. Stewart,  and E.  N.  Lightfoot.   Transport
     Phenomena.  Wiley, New York.   1960.

 3.  Blecker, H.G.,  and T. M.  Nichols.   Capital and Operating Costs  of
     Pollution Control Equipment Modules.   Prepared for U. S. Environ-
     mental Protection Agency,  Office of Research and Monitoring.  July
     1973.

 4.  Calvert, S., J. Goldshmid, D. Leith, and  N.C. Jhaveri.  Feasibility
     of Flux Force/Condensation Scrubbing for  Fine Particulate Collection.
     Control Systems,  N.E.R.C., Research Triangle Park, North Carolina.
     EPA-650/2-73-036. 1973.

 5.  Calvert, S., J. Goldschmid, D.  Leith,  D.  Mehta.  Scrubber Handbook.
     Control Systems Division,  Office of Air Programs,  E.P.A.  1972.

 6.  Calvert, S., D. Lundgren,  and D. S. Mehta.  Venturi Scrubber Per-
     formance.  J. Air Pollut.  Control Assn.   22:529-532.  1972.

 7.  Dixon, W.J.  HMD: Biomedical  Computer  Programs.  U. Calif. Press,
     Berkeley.  1973.

 8.  Harris, L.S. Energy  and Efficiency  Characteristics of the Ejector
     Venturi Scrubber. Presented  at the 57th  Annual APCA Meeting,
     Houston, Texas.  Paper 64-35.   June 1964.

 9.  Harris, L.S. Fume Scrubbing With the  Ejector Venturi System.  Chem.
     Engr. Prog. 62:No. 4, 55-59.   1966.

10.  Lancaster,  B.W.,  and W. Strauss. A. Study of Steam Injection Into
     Wet Scrubbers.  Indus. Engr.  Chem.  Fund.  10:362-369.  1971.

11.  Lapple, C.G. and  H.  J. Kamack.   Performance of Wet Dust Scrubbers.
     Chem. Eng.  Prog.  51:110.  1955

                                  72

-------
12.  Lohs, W.  Manufacture of Aerosols and Separation of Ultrafine Dusts
     in Spray Washers.  Staub.  29:No. 2, 43.  1969.

13.  National Air Pollution Control Administration.  Control Techniques
     for Particulate Air Pollutants.  U. S. Government, Washington, D.C.
     1969.

14.  Peters, S. and K. D. Timmerhaus.  Plant  Design and Economics for
     Chemical Engineers.  McGraw-Hill, New York.  Section Edition.  1968.

15.  Slinn, W. G. N. and J. M. Hales.  A Re-evaluation of the Role of
     Thermophoresis as a Mechanism of In - and Below - Cloud Scavenging.
     J. Atmos. Sci.  28:1465-1471.  1971

16.  Sparks, L. E. and M. J. Pilat.  Effect of Diffusiophoresis on
     Particle Collection by Wet Scrubbers.  Atmos. Environ.  4:651-660.
     1970  .

17.  Takashima, Y., W. T. Kyritsis, R. Dennis, and L. Silverman. Waste
     Processing Off-gas  Scrubber Studies.  Proceedings of the Seventh
     AEG Air Cleaning Converence.  Brookhaven National Laboratory.
     'p. 557-579.  October 1961.

18.  Wilson, E.B. Jr.  An Introduction to  Scientific Research.  McGraw-
     Hill,  Inc., New York.   1952.
                                   73

-------
                             APPENDIX A
               MANUFACTURER'S DESCRIPTION OF DYNACTOR

The  continuous gas/liquid contactor, the Dynactor,   is  a  proprietary
development  of R P Industries,  Inc.   Figure A-l is  a cross  section of
a  single  stage Dynactor diffusion contactor.   Liquid entering  the sys-
tem  under a  pressure to 140 to  200 pounds per square inch (typical)
is atomized  into thin films and droplets of average thickness  or dia-
meter  less than 1/64 inch.   This liquid discharge diffuses  or  expands
into the  reaction chamber causing air or gas  to be  aspirated by being
trapped within the moving shower of films and particles.  The  result-
ing  mixed fluid then continues  to travel down the reaction  column
with intimate  contact maintained between gas  and liquid.  This causes
physical  and chemical equilibria to occur by  the time the mixed fluid
exits  from the reaction column  into the separation  reservoir.  The
Dynactor  can be viewed as a macroscopic diffusion pump which makes
use  of diffusion principles in  order to aspirate large volumes of air
per  volume of  motive liquid.  By utilizing diffusion rather than
Bernoulli principles, the Dynactor aspirates  up to  4,800  standard
volumes of gas per volume of motive  liquid.   In comparison, venturi
eductors  will  aspirate not  more than 100 volumes of gas per volume
of motive liquid.

Because there  are no venturi or other  constrictions in the Dynactor,
energy requirements are  considerably lower than for conventional jet
or venturi eductor systems.  If gas  carries small solid particles
along with it,  such as activated carbon or powdered neutralizing and
precipitating  agents,  such  particles are wetted and captured by the

                                  A-l

-------
   f

»




f
«
	 H
           LIQUID  INPUT,  140 TO 200psi
                 AIR  INPUT,  LOW VELOCITY, AMBIENT  PRESSURE
            HIGH VELOCITY,  SUB-AMBIENT  PRESSURE
                   SHOWER OF THIN FILMS AND  PARTICLES
                   REACTION  COLUMN
                              GAS  OUTPUT
       RESERVOIR/SEPARATOR
                                                LIQUID  LEVEL DETER-
                                                MINING  TRAP
                                                LIQUID  OUTPUT
figure A-l.
Single-stage Dynactor diffusion system
cross sectional view
                         A-2

-------
liquid throughout the entire  length of the reaction chamber.  By con-
trast, venturi wet scrubbers  make effective contact between gas and
liquid only in the constricted  throat region.  Contact time, therefore,
in the Dynactor is about 20 times longer than in venturi devices.

Just as in oil and mercury diffusion vacuum pumps, it is also possible
to construct Dynactors having multistage gas inputs.  Figure A-2 is
a drawing of the two-stage Dynactor diffuser system employed in these
studies.  The internal configuration was constructed to maximize gas/
liquid turbulence and contact throughout the length of the six-foot
long, 12-inch diameter reaction column.
                                 A-3

-------
DIRTY
GAS
IN

?










*

PLENUM









^)


o
o 5
fll O
2 o




















W S3
0 O
<; M
iB
w S
f4 O
M CJ











RESERVOIR
SEPARATOR
PUMP 1











*

PLENUM









*>


M g
Si
^i t?
2 cj










GAS OUT
I







— *.. — . ;
w
2
'S
u
a
Q







RESERVOIR
SEPARATOR
PUMP
t
                  WATER  OUT
                                                                MAKE UP WATER  IN
                            Figure A-2.  Two-stage Dynactor

-------
                             APPENDIX B
                      DYNACTOR EFFICIENCY DATA

This appendix  presents  all  the data on the  Dynactor efficiency tests
used as a basis  for  the Dynactor evaluation.   It  includes the tests for
the balanced 2   factorial design study,  as  well as tests involving
flow at 1500 cfm (42.5  m3/min),  moisture  addition with a spray nozzle
upstream from  the  Dynactor,  use  of  surfactant  in  the Dynactor spray,
and testing of a  single stage  of the  two-stage device.

The efficiency data  sheets,  one  for each  individual set of test condi-
tions,  contain the following information:
    • Test identification number
    • Concentration  at  Dynactor  inlet
    • Concentration  at  Dynactor  outlet
    • Total  mass efficiency, based on the preceding two numbers
    • Aerosol material  used in test
    • Air  flow through  Dynactor
    • Volume rate of flow of Dynactor spray
    • Temperature at Dynactor inlet
    • Remarks,  where appropriate
    • Flow rate and sample duration for the various sampling  devices
    • Fractional  efficiency  data, including:
       • Limits, aerodynamic diameter,  of size interval
       • Concentration  of aerosol in that size interval at Dynactor inlet
       • Concentration  of aerosol in that size interval at Dynactor outlet
                                 B-l

-------
        •  Mass efficiency on particles of sizes in that size
           interval,  based on the preceding two numbers
        •  Total concentrations measured by'the Andersen
           impactors  at inlet and outlet and mass efficiency
           based on them
As discussed in the report, the substrate material onto which im-
paction occurred sometimes showed a weight loss or a weight gain that
was less than 0.3 mg.  These small or negative weight changes were
given the same value, <_0.3 mg, which was used to derive those ef-
ficiencies given as inequalities.  For the large end of the particle
size spectrum, 4 pirn and above, the inequalities were calculated with
the assumption that collection efficiency did not decrease as
particle size increased.
                              B-2

-------
               DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST if  4
AEROSOL   Iron oxide
                    INLET CONCENTRATION   .86  gr/ft3 =1979 mg/m '
                    OUTLET CONCENTRATION  .031 gr/ft3 =71.4 mg/i:J
                    TOTAL MASS EFFICIENCY   .964
FLOW:
TEMPERATURE:
500  ft3/min = 14.2 m3/min
         70 °F
21 °C
Total Spray:  18.4 gal/min
             = 1.16 x 10   m /sec
REMARKS:   Size distribution obtained from impactors at 1.0 cfm flow was
 interpolated to match sizing intervals for impactors at 0.5 cfm flow.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft-Vmin)
1.0
1.0
1.0
1.0
RATE
(m-Vmin)
= .028
= .028
= .028
= .028
DURATION
(rain)
1.
1.
15.
15.
TOTAL
(ft3)
1.0
1.0
15.0
15.0
VOLUME
(V)
= .028
= .028
= .43
= .43
        FRACTIONAL EFFICIENCY DATA,  ANDERSEN IMPACTOR,  MODEL  III
STAGE
AERODYNAMIC:
DIAMETER
1 £13. 6
2
3
4
5
6
7
8
Final
TOTAL,
8.6 - 13.6
5.6 - 8.6
INLET
CONCENTRATION
< 10.6
OUTLET
CONCENTRATION
(mg/m3)
< -7
< 10.6 ~ .7
41.0 .85
4.0 - 5.6 96.8 .85
2.5 - 4.0 i 247.
1.3 - 2.5 , 495.
0.80 - 1.3 155.
4.1
19.8
23.1
0.54 - 0.80 i < 10.6 11.8
filter £ 0.54
Andersen Irapactor
17.7 4.7
1074.
66.0
FRACTIONAL
MASS
EFFICIENCY
(percent)
>. 99.1
£90-1
> 99.1
99.1
98-3
96.0
85.2
£ o
73.3
93.9
                                 B-3

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST *  5
AEROSOL   Iron oxide
INLET CONCENTRATION ~ .8  gr/ftJ =~2000mg/ir
                               3
                            OUTLET CONCENTRATION .040 gr/ft  = 91.4mg/- "'
                            TOTAL MASS EFFICIENCY  ~ -95
FLOW:   500  ft3/min = 14.2 m3/tnin  Total  Spray:  18.4
TEMPERATURE:     205 °F =   97 °C
                            x 10
                                                               m
                                                                 /sec
REMARKS:  Size distribution obtained by impactors at 1.0 cfm flow was
interpolated to match sizing intervals for impactors at 0.5 cfra flow.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMP ACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft3/rain)
1.0
1.0
1.0
1.0
RATE
(m-Vmin)
= .028
= .028
DURATION
(min)
1.
1.
- .028 j 15.
= .028
15.
1
TOTAL
(ftj)
1.0
1.0
15.
15.
VOLUME
On-5)
- .028
= .028
= .43
= .43
        FRACTIONAL EFFICIENCY DATA,  ANDERSEN  IMPACTOR,  MODEL LIT


STAGE
1
2
3
4

AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6

INLET
CONCENTRATION
(mg/m^)
< 3.5
<3.5
26.9
84.8

OUTLET
CONCENTRATION
(mg/m3)
< -7
FRACTIONAL
MASS
EFFICIENCY
(perci n t)
> 99.1
< .7 I > 99.1
< -7
> 99.1
< .7 > 99.1
5 2.5 - 4.0
6 1.3 - 2.5
7 0.80 - 1.3
8 0.54 - 0,80
Final filter $0.54
TOTAL, Andersen Impactor
205.
1 417.
I
134.
10.6
17.7
898.
1.9 j
10.4 j
14.4
17.9
4.9 ;
49.7 -
                                                               99.1
                                                               97.5
                                                               89.3
                                                             <  0.0
                                                               72.0
                                                               94-3
                                B-4

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST ;t  1
AEROSOL   Fly  ash
            INLET CONCENTRATION ~ 0.7 gr/ft"
            OUTLET CONCENTRATION .0059gr/ft  - 13.6mji/"
            TOTAL MASS EFFICIENCY  ~ .99
             3
FLOW:   500   ftJ/min = 14.2 m3/min  Total  Spray:  18.4 gal/mig  3
                     0         0                 = 1 . 16 x  10   m /sec
TEMPERATURE:
    0
200  F
                               0
                            95  C
REMARKS:  Inlet concentration sampler malfunctioned,  so inlet  concentra
tion  (0.7 gr/ft-3)  estimated from other tests.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL  FILTER
OUTLET IMPACTOR
           FLOW RATE
      (f t-Vmin)  (m3/min)
         1.0    = .028
          .5    = .014
         1.0    = .028
          .5    = .014
                                            DURATION
                                              (tnin)
                                               3.
                                               3.
                                              69.
                                              90.
TOTAL VOLUME

 3.0  =  .085
 1.5  =  .043
69.   = 1.95
45.   - 1.27
        FRACTIONAL EFFICIENCY DATA. ANDERSEN IMPACTOR, >DDEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
VI. \METER
(urn)
£13.6
8.6 - 13.6
5-6 - 8.6
INLET
CONCENTRATION
(tng/m3)
37.7
118.
186.
4.0 - 5.6 | 141.
•>.5 - 4.0 H3.
1 3 - 2.5 ; 82.
0.80 - 1-3 I 21.2
0.54 - 0.80 < 7.1
filter £0. 54 < 7 • l
Andersen Impactor
704.
OUTLET
CONCENTRATION
(mg/m-^)
< .24
< .24
< .24
< .24
.5
3.2
4.0
2.9
4.0
14.9
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99. 9
> 99-9
> 99.9
> 99.8
99.5
96.1
81.1
< 58.9
< 43.3
97.9
                                 B-5

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  8
AEROSOL   Fly ash
               INLET CONCENTRATION  .90   gr/ft  = 2076
               OUTLET CONCENTRATION -0067gr/ft3 =15.4«ig/i
               TOTAL MASS EFFICIENCY  .993
FLOW:   500
TEMPERATURE:
  3             3
ft /rain = 14.2  TO /rain Total Spray: 18.4 gal/min
    70 °F =
21 °C
= 1.16 x 10 3 m3/sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/min) (m^/min)
1.0 = .028
.5 - .014
1.0 = .028
.5 = .014
DURATION
(roin)
2.
3.
OO •
90.
TOTAL VOLUME
(ft3) (in3)
2.0 = .057
1.5 = .043
88. = 2.49
45. =1.37
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR, MODEL  III
AERODYNAMIC
DIAMETER
STAGE (urn)
1 £13.6
2 8.6 - 13.6
3 5.6 - 8.6
4 4.0 - 5-6
5 2.5 - 4.0
6 1.3 - 2.5
7 0.80 - 1.3
8 0.54 - 0.80
Final filter £0.54
TOTAL, Andersen Impactorj
INLET
CONCENTRATION
(rag/m3)
61.2
160.
278.
151.
132.
80.
19.
< 7.0
< 7.0
881.
OUTLET
CONCENTRATION
(mg/m3)
< .2
< .2
< .2
< .2
.5
3.5
4.1
3.0
2.0
13.4
FRACTIONAL
MASS
EFFICIENCY
(percent )
> 99. 9
> 99. 9
> 99.9
> 99.8
99.6
95.6
78.3
< 57.8
< 72.2
98.5
                               B-6

-------
               DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #9
AEROSOL  Fly Ash

FLOW: 1000    f fVmin =28.3
INLET CONCENTRATION  .25  gr/ft  =568
OUTLET CONCENTRATION .0015gr/ftJ =3.3 mg/n3
TOTAL MASS EFFICIENCY .994
m /min Total  Spray:  18.4 gal/min
  _                 	 ti/:__-!rt~J*J
TEMPERATURE: 68 °F = 20 °
= 1. 16 x 10 m /sec
L»
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft-Vmin)
2.0
0.5
2.0
0.5
RATE
(m3/min)
= .057
= .014
= .057
= .014
DURATION
(min)
10.
26.
296.
300.
TOTAL VOLUME
(ft3) (m3)
20. = -57
13. = .37
592. = 16.8
150. = 4.3
        FRACTIONAL EFFICIENCY  DATA,  ANDERSEN IMPACTOR, MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(nm)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Irapactor
INLET
CONCENTRATION
(mg/m3)
7.3
16.3
29.6
22.3
20.9
13.0
4.9
< .8
< .8
114.
OUTLET
CONCENTRATION
(mg/m3)
<.07
<.07
<.07
<.07
.14
.82
1.2
.72
.40
3.3
FRACTIONAL
MASS
EFFICIENCY
(percent )
> 99. .8
> 99.8
> 99.8
> 99.7
99.3
93.7
74.5
< 50.9
< 10.4
97.0
                                B-7

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST # 10
AEROSOL  Fly ash
INLET CONCENTRATION 0-17  gr/ff = 396 mg/ir.J
OUTLET CONCENTRATION .0015gr/ft3 = 3.5 tng/«/
TOTAL MASS EFFICIENCY .991
 3
FLOW:   1000  ft /min = 28.3 in /min  Total Spray: 18.4 gal/mii^  3
                    o         o                  = 1.16 x 10   m /sec
TEMPERATURE:    200  F =   95  C
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft-* /min) (m3/tnin)
2.0 = .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(tnin)
10
20
240
240
TOTAL VOLUME
(ftj) On-*)
20 = .57
10 = .28
480 = 13.6
120 - 3.4
        FRACTIONAL EFFICIENCY DATA. ANDERSEN IMPACTQR.  MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6 '
8.6 - 13.6
5.6 '- 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(ntg/m3-)
3.2
12.7
25.1
15.5
14.1
9.9
1.4
< 1.1
< 1-1
82.3
OUTLET
CONCENTRATION
(mg/m3)
< .09
< .09
< .09
< .09
.15
.8
1.2
.7
.6
3.45
FRACTIONAL
MASS
EFFICIENCY
(percent )
> 99.6
> 99.6
> 99.6
> 99.4
99.0
92.0
12.5
< 36.1
< 44.4
95.8
                                B-8

-------
               DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  12
AEROSOL   Iron oxide
INLET CONCENTRATION  .244  gr/ft3 = 560 my/rr '
OUTLET CONCENTRATION .0244 gr/ft"' -  56 mg/V >J
TOTAL MASS EFFICIENCY  .900
FLOW:   1000  ftJ/min = 28.3m3/min  Total Spray: 18.4 gal/min '
TEMPERATURE:      70 °F     21 °C                = 1>16 X 10   m /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ftj/min) (m3/min)
2.0 = .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(min)
5
10
50
120
i 	
TOTAL
(ftj)
10
5
100
60
VOLUME
Cm-1)
.28
.14
= 2.8
= 1.7
        FRACTIONAL EFFICIENCY DATA,  ANDE'RSEN  IMPACTOR, MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
Cum)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
INLET
CONCENTRATION
Cmg/m^)
< 2-1
16.3
34.6
67.1
124.
1.3-2.5 188.
0 . 80 - 1.3 i 146 .
0. 54 - 0.80
25.4
filter $0. 54 < 2.1
Andersen Impactor
605.
OUTLET
CONCENTRATION
(mg/m3)
< .2
< .2
.4
1.0
3.8
14.6
21.4
18.2
6.4
65.9
FRACTIONAL
MASS
EFFICIENCY
(pnL)
> 99.0
> 99.0
99.0
98.5
96.9
92.3
85.3
28.5
< 0.0
89.1
                                B-9

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  13
AEROSOL  Iron oxide
INLET CONCENTRATION  .33  gr/ft  = 747mg/
OUTLET CONCENTRATION .059 gr/ft  = 135 ma/
                            TOTAL MASS EFFICIENCY .818
               3             3
FLOW:   1000  ft /rain = 28.3m /min Total  Spray:  9.2 gal/min3   3
TEMPERATURE:     80 °F =   27 °C                 =  -58 x  10"  m /sec
REMARKS:  Nozzles used at 100 p.s.i.g. =6.9 x 105 N/M2
SAMPLING DEVICE
INLET TOTAL FILTER
INLET IMP ACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/min) (m^/min)
2.0 = .057
.5 = .014
2.0 = .057
.5 = -014
DURATION TOTAL
(rr.in) (ftj)
4.0 8.0
4.0 2.0
40. 80.
40. 20.
VOLUME
.23
.057
= 2.3
.57
        FRACTIONAL EFFICIENCY DATA.  ANDERSEN IMPACTOR.  MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
Gun)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter $0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
< 5.3
7.1
33.6
54.8
124.
210.
147.
15.9
< 5.3
599.
j
OUTLET
CONCENTRATION
(mg/m3)
1.6
1.6
2.3
5.3
15.4
45.4
59.
15.9
5.5
152.
FRACTIONAL
MAS?
EFFICIENCY
(pcrct n i )
> 77.5
77.5
93.2
90.3
87.6
78.4
60.0
0.0
< o-o
74.6
                                B-10

-------
               DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  14


AEROSOL  Iron oxide
            INLET CONCENTRATION  .118 gr/ft3 = 271 mg/ir 5

            OUTLET CONCENTRATION .016 gr/ft3 = 36.6mg/tr3
                             TOTAL MASS EFFICIENCY  .866

FLOW:   1000  ft3/n>in = 28.3 m3/min  Total  Spray.  18.4 gal/inl?
TEMPERATURE:
203  F =
95 °C
                                 = 1.16  x 10
3  3
  tn
/sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft-Vmin) (nrVinin)
2.0 = .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(min)
4.
4.
40.
40.
TOTAL
(ftj)
8.
2.
80.
20.
VOLUME
OJ)
= .226
= .057
= 2.26
= .566
        FRACTIONAL  EFFICIENCY DATA.  ANDERSEN IMPACTOR.  MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter^0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
< 5.2
< 5.2
7.1
10.6
30.0
97.2
67.1
< 5.2
< 5.2
215.5
OUTLET
CONCENTRATION
(mg/m3)
< .5
< .5
< .5
< .5
1.4
10.4
15.9
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 95.3
. > 95.3
> 95.3
> 95.3
95.3
89.3
76.3
8.1 < 0
< .5
36.0
-
83.3
                                 B-ll

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  16
AEROSOL   Iron oxide
FLOW:   1000  ft /rain
TEMPERATURE:
          INLET CONCENTRATION 1.01  gr/ft  = 2311 mg/tr.
          OUTLET  CONCENTRATION  .047 gr/ft3  = 107
          TOTAL MASS EFFICIENCY .954
      28.3m3/min  Total Spray: 18.4 gal/min3
63 °F
                           17 °C
                               - 1.16 x 10
                                                              m
                                                                 /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMP ACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(f rVmin) (np/min)
2.0 - .057
0.5 = .014
2.0 = .057
0.5 • .014
DURATION
(min)
1.0
2.0
4.5
15.0
TOTAL VOLUME
(ft-0 (ra^)
2.0 = .057
1.0 = .028
9.0 = .255
7.5 = .212
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR.  MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(jim)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5,6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter $0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
49.5
70.7
159.
180.
254.
375.
258.
60.1
17.7
1424.
OUTLET
CONCENTRATION
(mg/m3)
< 1-4
< 1.4
< I-4
< 1-4.
4.7
25.4
33.0
17.4
< 1.4
81.5
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.2
> 99.2
> 99.2
> 99.2
98.1
93.2
87.2
71.0
> 92.0
94.3
                               B-12

-------
              PYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
                             INLET CONCENTRATION   .84 gr/ft  =1922
                             OUTLET CONCENTRATION  .042 gr/ft3 =  96 mg/tr3
TEST #  17
AEROSOL   iron oxide
                             TOTAL MASS  EFFICIENCY  .950
FLOW:   1000  ft3/min = 28.3m3/min Total Spray: 18.4 gal/min
TEMPERATURE:      70 °F =   21 °C
                                                 = 1.16 x 10
                                                              m
/sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ffVmin) (nvVmin)
2.0 - .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(min)
1.
2.
15.
15.
TOTAL VOLUME
(ft-5) (mj)
2.0 = .057
1.0 = .028
30. = .85
7.5 = .21
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0-54 - 0.80
filter £0.54
Andersen Iirvpactor
INLET
CONCENTRATION
(mg/tn3)
24.7
53.0
148.
187.
307.
382.
205.
21.2
< 10.7
1336.
OUTLET
CONCENTRATION
(mg/m3)
< 1.4
< 1.4
< 1.4
1.9
5.2
25.9
39.1
18.4
2.8
95.2
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.0
> 99.0
> 99.0
99.0
98.3
93.2
80.9
13.3
< 73.3
92.9
                                B-13

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  18
AEROSOL  Iron oxide
FLOW:    1000ft /rain
TEMPERATURE:
          INLET CONCENTRATION  .872  gr/ft3  = 1996
          OUTLET CONCENTRATION .038  gr/ft3  = 86.6mg/TT<3
          TOTAL MASS EFFICIENCY  .957
     28.3 tn /min   Total Spray: 18.4 gal/min
70 °F
21 °C
= 1.16 x 10"3 m3/sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft3/min)
2.0
0.5
2.0
0.5
RATE
= .057
= .014
= .057
= .014
DURATION
(min)
1.0
2.0
15.0
15.0
TOTAL

2.0
1.0
30.
7.5
VOLUME
(mj)
= .057
= .028
= .85
= .21
        FRACTIONAL EFFICIENCY DATA,  ANDERSEN IMFACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
Gin)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter $0.54
Andersen Impactor
INLET
CONCENTRATION
(rng/m^)
21.2
53.0
198.
205.
322.
343.
198.
28.3
< 10.7
1378.
OUTLET
CONCENTRATION
(mg/m3)
< 1.4
< 1.4
< 1.4
< 1.4
2.8
24.0
33.5
12.7
1.9
74.9
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.3
> 99.3
> 99.3
> 99.3
99.1
93.0
83.1
55.0
< 82.2
94.6
                                B-14

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  19
AEROSOL   Iron oxide
            INLET CONCENTRATION  .679gr/ft3  = 1553 uig/i/
                             OUTLET CONCENTRATION .035 gr/f t   = 79.1 mg /,r. '
                             TOTAL MASS  EFFICIENCY -949
FLOW:   1000  ft3/min = 28.3 ra3/min Total Spray:  18.4 gal/rain
TEMPERATURE:
203 °F
                            95 °C
                                                =1.16x10   m /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ftj/min) (m3/min)
2.0 = .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(min)
1.0
2.0
14.0
14.0
TOTAL
(ftj)
2.0
1.0
28-
7.
VOLUME
(n.^)
= .057
= .028
= .792
= .198
        FRACTIONAL EFFICIENCY DATA.  ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(pirn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(tng/m3)
24-7
56.5
191.
166.
272.
286.
170.
17.7
< 10.7
1194.
OUTLET
CONCENTRATION
(mg/m3)
< 1.5
< 1.5
< 1.5
< 1.5
1 -r
6.6
28.8
40.9
15.6
3.5
98.9
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 9.9.2
> 99.2
>99.2
> 99.1
' 97.6
89.9
75.9
11.4
< 66.7
91.7
                                 B-15

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST # 21
AEROSOL  Fly ash
INLET CONCENTRATION  -955  gr/ft3 -2186
OUTLET CONCENTRATION -Oil  gr/ft3 =25.3
TOTAL MASS EFFICIENCY  .988
FLOW:  1000  ft3/min - 28.3 m /min  Total  Spray:  18.4  gal/min3   3
	                 o         o                  = 1.16  x 10   m /sec
TEMPERATURE:     70  F =   21  C
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/tnin) On^/tnin)
2.0 = .057
0.5 = .014
2.0 = .057
0.5 = .014
DURATION
(min)
3.0
3.0
57.5
60.0
TOTAL VOLUME
(ft-*) (mj)
6.0 = .171
1.5 = .042
115. = 3.25
30. = .84
        FRACTIONAL EFFICIENCY DATA.  ANDERSEN IMPACTOR,  MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(lira)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter $0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/rn^)
40.0
75.4
153-
115.
80.1
73.0
28.3
< 7.1
< 7.1
575.
OUTLET
CONCENTRATION
(mg/m3)
< .4
< .4
< .4
< .4
1.2
6.8
11.0
5.9
1.1
26.4
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.8
> 99.8
> 99.8
> 99.7
98.5
90.6
61.3
< 16.7
< 85.0
99.4 .
                                B-16

-------
               DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  22
AEROSOL   Fly ash
INLET CONCENTRATION  1.30  gr/ft3 = 2984mg/Tr/'
OUTLET CONCENTRATION  .012 gr/ft3 = 27.4 mp,/r 3
                             TOTAL MASS EFFICIENCY  .991
FLOW:   1000  ft3/min = 28.3n,3/min  Total Spray: 18.4 gal/min
TEMPERATURE:    203  °F =   95 °C                 = 1>16 x 10"  m /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft3/min)
2.0
0.5
2.0
0.5
RATE
(m-Vmin)
= .057
= .014
= .057
= .014
DURATION
(min)
3.0
3.0
58.0
60.0
TOTAL
(ftj)
6.0
1.5
116.
30.
VOLUME
0J)
= .171
= .042
= 3.28
= .84

        FRACTIONAL EFFICIENCY DATA,  ANDERSEN IMPACTOR, MODEL  III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(nm)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0. 54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
84.8
155.
229.
127.
115.
80.1
30.6
< 7.1
< 7.1
822.
OUTLET
CONCENTRATION
(mg/m3)
< -4
< -4
< -4
< .4
1.4
9.2
9.2
4.6
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.8
> 99.8
> 99.8
> 99.7
98.8
88.5
70.0
< 35.0
1.8 < 75.0
26.9
96.7
                                B-17

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST # 23
AEROSOL  Iron oxide
INLET CONCENTRATION  .178 gr/ft3 =408 mg/tr.3
OUTLET CONCENTRATION .010 gr/ft3 = 23.2mg/mJ
                            TOTAL MASS EFFICIENCY .943
FLOW:    500 ft /min = 14.2  m /min Total  Spray:  18.4  gal/min
TEMPERATURE:     66 °F =    19°C                = 1.16 x 10"  m /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ftj/min) (m-Vmln)
1.0 • .028
0.5 = .014
1.0 = .028
0.5 = .014
DURATION
(min)
15
15
118
120
TOTAL
(ft3)
15.0
7.5
118
60
VOLUME
(™J)
= .424
= .212
= 3.34
= 1.70
        FRACTIONAL EFFICIENCY DATA.  ANDERSEN IMPACTOR.  MODEL III
STAGE
1
2
3
4
5 ,
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2;. 5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(rag/in-5)
< 1.4
 98.6
> 98.6
> 98. 6
> 98.6
97.8
91.0
76.6
< 0
< 25.0
85.6
                               B-18

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  24
AEROSOL  Iron oxide
           INUET CONCENTRATION  .187  gr/ft3 = 428
           OUTLET CONCENTRATION .014  gr/ft3 =32.6mp/r3
           TOTAL MASS EFFICIENCY  .924
FLOW:  500    ft  /mln  =14.2 ra /min  Total Spray: 18.4 gal/min
TEMPERATURE:
203 °F
95 °C
= 1.16  x 10  3 m3/sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft-Vmin)
1.0
'0.5
1.0
0.5
RATE
(nvVmin)
= .028
= .014
= .028
= .014.
DURATION
(min)
15
15
84
120
TOTAL
(ft13)
15.0
7.5
84
60
VOLUME
(m3)
= .424
= .212
= 2.38
= 1.70
        FRACTIONAL EFFICIENCY DATA,  ANDERSEN IMPACTOR, MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £.0. 54
Andersen Impactor
INLET
CONCENTRATION
(mg/n»3)
2.8
3.3
10.8
16.5
39.6
71.1
56.5
12.7
< 1-4
213.
OUTLET
CONCENTRATION
(mg/m3)
< .18
.24
.29
.35
1.5
7.2
9.8
6.5
2.4
28.3
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 93.8
92.9
97.3
97.9
96.3
89.9
82.7
49.1
< 0
86.7
                                 B-19

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  25


AEROSOL   Iron oxide
           INLET CONCENTRATION  .615 gr/f t   = 1406ms/jr.'


           OUTLET CONCENTRATION .027 gr/ftJ  = 62.9mg/rr.3


           TOTAL MASS EFFICIENCY  .955
               3             3
FLOW:   1000 ft /min = 28.3 m /min  Total Spray:  18.A gal/min,
TEMPERATURE:
70 °F
21 °C
= 1. 16 x 10   m /
                                                                  sec
REMARKS:  Approximately 10 ppm surfactant added to spray water.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft-Vmin) (m-Vmin)
2.0 = .057
0.5 - .014
2.0 = .057
0.5 - .014
DURATION
(min)
1.
2.
15.
15.
TOTAL VOLUME
(ft^)  99.2
> 99.2
> 99.2
> 99.2
99.2
94.2
84.9
62.2
< 82.2
94.4
                                B-20

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #   26
AEROSOL   Fly ash
INLET CONCENTRATION .121  gr/ft3  = 276 mg/m3
OUTLET CONCENTRATION. 002  gr/ft3  = 3.9 mg/m*
                            TOTAL MASS EFFICIENCY   98.6
FLOW:    500  ft  /min  =  14.2 m./min  Total Spray: 18.4 gal/m^n 3
TEMPERATURE:      85 °F  =    29 °C                 1.16 x 10   m /sec
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/min) (m^/min)
1.0 = .028
0.5 = .014
1.0 = .028
0.5 = .014
DURATION
(min)
25.
25.
268.
270.
TOTAL VOLUME
(ft3) (raj)
25.0 = .708
12.5 = .354
268. =7.58
135. =3.82
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTQR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(nm)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
12.7
26.0
41.8
24.3
24.0
17.5
10.5
< .8
< .8
158.
OUTLET
CONCENTRATION
(mg/m3)
< .1
< .1
< .1
< .1
< -1
.7
1.4
.9
.3
3.4
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.8
> 99.8
> 99.8
> 99.7
> 99.7
96.0
86.7
< 0
< 62.5
97.8
                                B-21

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #27
AEROSOL  Fly Ash
                    INLET CONCENTRATION  .136 gr/ft3  = 311 mg/ir.
FLOW:
TEMPERATURE:
                    OUTLET CONCENTRATION  .002 gr/ft'  =  4.2 mg
                    TOTAL MASS EFFICIENCY  .986
500   ft3/min =14.2 m3/min  Total Spray: 18.4 gal/rain^  3
                      o                  = 1.16 x 10   m /sec
         200 °F =
 n3
95 "C
REMARKS:
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft3/tnin)
1.0
0.5
1.0
0.5
RATE
= .028
= .014
= .028
= .014
DURATION
(min)
25
25
243
293
TOTAL
25
12.5
243
146.5
VOLUME
= .708
= .354
= 6.88
= 4.15
        FRACTIONAL EFFICIENCY DATA,  ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
INLET
CONCENTRATION
(mg/m3)
10.2
22.9
36.5
24.0
21.2
16.7
11.6
0.54 - 0.80 i 1.7
filter $0.54 < .8
Andersen Inpactor
145
OUTLET
CONCENTRATION
(mg/m3)
< .07
< .07
< .07
< .07
< .07
.58
1.4
.89
.46
3.4
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.8
> 99.8
> 99.8
> 99.7
> 99.7
96.5
87.7
47.5
< 45 . 8
97.7
                                B-22

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST # 28
AEROSOL  IRON OXIDE
         (steam)
INLET CONCENTRATION
gr/ftJ -
mg/in
OUTLET CONCENTRATION .021 gr/ft3 =47.8 mg/irJ
TOTAL MASS EFFICIENCY
FLOW: 500     ft3/rain = 14.2  tn3/min Total Spray:  18.4 gal/min
TEMPERATURE:
                    = 1.16 x 10"3  m3/sec
REMARKS:   Spray  introduced  in inlet  duct.  Humidity ratio of
           approximately .044.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/min) (in^/min)
Broken = line, si
0.5 = .014
1.0 = .028
0.5 = .014
DURATION
(min)
imple void
2
15
15
TOTAL
(ft-*)

1
15
7.5
VOLUME
(in3-)
=
= .028
= .42
= .21
        FRACTIONAL EFFICIENCY DATA.  ANDERSEN  IMPACTOR, MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(tarn)
»13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
67.1
110
216
216
251 j
318
198
28.3
< 10.6
1403
OUTLET
CONCENTRATION
(mg/m3)
< 1.41
[. . — — 	
< 1.41
< 1.41
< 1.41
1.88
16.0
28.3
14.6
3.77
64.5
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 99.3
> 99.3
> 99.3
> 99.3
99.3
95.0
85.7
48.4
< 64.4
95.4
                                 B-23

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #29
AEROSOL  Iron Oxide
          INLET CONCENTRATION .581  gr/ft3 =1329
          OUTLET CONCENTRATION .025 gr/ft3 = 57.
          TOTAL MASS EFFICIENCY  .948
FLOW: 1500    ft3/min = 42.5 m3/min Total Spray: 18.4 gal/min
TEMPERATURE:
70°F
21°C
1.16 x 10"3 m3/sec
REMARKS:
     Data corrected for leakage between inlet and outlet.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ft3/min) (m^/min)
1.0 = .028
0.5 = .014
2.0 = .056
0.5 = .014
DURATION
(min)
2
2
TOTAL
(ftj)
2.0
1.0
15 30
15
7.5
VOLUME
OJ)
= .056
= .028
= .85
= .21
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3-
4
5
6
7
8
Final .
TOTAL,
AERODYNAMIC
DIAMETER
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
INLET
CONCENTRATION
(mg/m3)
10.6
28.3
81.3
67.1
134
177
134
< 10.6
Eilter^0.54 < 10.6
Andersen Impactor! 633
OUTLET
CONCENTRATION
(mg/m3)
< 1.41
< 1.41
< 1.41
< 1.41
3.30
18.8
27.8
7.07
< 1.41
58.4
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 97.9
> 97.9
> 97.9
> 97.5
97.0
87.1
75.0
< 19.4

88.9
                                B-24

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #  30
AEROSOL  Iron Oxide
           INLET CONCENTRATION .571   gr/ft3  = 1306mg/ir?
           OUTLET CONCENTRATION .025  gr/ft3  = 57.6mg/m5
           TOTAL MASS EFFICIENCY  .947
FLOW:   1500  ft3/min  = 42.5 ra3/min Total Spray: 18.4 gal/min
TEMPERATURE:
190°F =
88°C
1.16 x 10"3 m3/sec
REMARKS:
     Data corrected for leakage between inlet and outlet.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ft3/min)
1.0
0.5
2.0
0.5
RATE
(nwmin)
= .028
= .014
= .056
= .014
DURATION
(min)
2
2
15
15
TOTAL
(ft3)
2.0
1.0
30
7.5
VOLUME
OJ)
= .056
= .028
= .85
= .21
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(urn)
£13.6
8.6 - 13.6
5.6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
35.3
63.6
124
106
163
198
148
14.1
< 10.6
852
OUTLET
CONCENTRATION
(mg/m3)
< 1.41
< 1.41
< 1.41
< 1.41
1.9
17.4
20.7
2.8
< 1.41
43.3
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 98.7
> 98.7
> 98.7
> 98.6
98.6
89.4
83.1
75.8
-.
93.8
                                B-25

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST #31
AEROSOL  Iron Oxide
         (steam)
FLOW:
TEMPERATURE:
                  INLET CONCENTRATION .504  gr/ft3 = 1153mg/tr '
                  OUTLET CONCENTRATION .019 gr/ft3 = 44.4 mgAr.''
                  TOTAL MASS EFFICIENCY  .962
     3             3
500ft /min =  14.2ra /min  Total Spray: 18.4 gal/min
                                                 -3  3
                                       =1.16 x 10   m /sec
REMARKS:   Spray in inlet duct humidity ratio of approximately .072.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW
(ftj/min)
1.0
0.5
1.0
0.5
RATE
(m^/min)
- .028
= .014
= .028
= .014
DURATION
(min)
5
5
25
25
TOTAL
TftT
5.0
2.5
25
12.5
VOLUME
= .14
= .07
= .70
= .35
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR. MODEL III
STAGE
1
2
3
4
5
6
7
8
AERODYNAMIC
DIAMETER
(urn)
»13.6
8.6 - 13.6
5-6 - 8.6
4.0 - 5.6
2.5 - 4.0
1.3 - 2.5
0.80 - 1.3
0.54 - 0.80
Final filter £0.54
TOTAL,
Andersen Impactor
INLET
CONCENTRATION
(mg/m3)
11.8
17.0
41.0
48.1
102
184
90.5
17.0
< 4.2
510
OUTLET
CONCENTRATION
(mg/m3)
< .8
< .8
< .8
< .8
1.1
11.0
17.8
5.1
< .8
35.9
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 98.9
> 98.9
> 98.9
> 98.9
98.9
94.0
80.3
70.0
-
93.0
                                B-26

-------
              DYNACTOR SCRUBBER EFFICIENCY EVALUATION DATA
TEST # 32
AEROSOL  Iron Oxide
    (1st Stage Test)
FLOW:
TEMPERATURE:
                    INLET CONCENTRATION .286  gr/ft3 = 654 mg/m*
                    OUTLET CONCENTRATION
                    TOTAL MASS EFFICIENCY  -
                        -  gr/ft   =   - mg/n>
       3             3
1000 ft /min = 28.3 m /rain  Total Spray:  18.4  gal/min
          80°F =
27°C
           -3   3
= 1.16 x 10   m /sec
REMARKS:    Downstream  sampling done between the first and second
            Dynactor  stages.
SAMPLING DEVICE

INLET TOTAL FILTER
INLET IMPACTOR
OUTLET TOTAL FILTER
OUTLET IMPACTOR
FLOW RATE
(ftj/min) (m3/min)
1.0= .028
0.5 = .014
— = —
0.5= .014
DURATION
(min)
3
3
-
10
TOTAL VOLUME
(ftj) (m3)
3.0 = .084
1.5 = .042
_
5.0 = .14
        FRACTIONAL EFFICIENCY DATA, ANDERSEN IMPACTOR, MODEL III
STAGE
1
2
3
4
5
6
7
8
Final
TOTAL,
AERODYNAMIC
DIAMETER
(lim)
»13.6
8.6 - 13.6
5.6 - 8.6
INLET
CONCENTRATION
(mg/m3)
14.1
28.3
70.7
4.0 - 5.6 80.1
2.5 - 4.0
127
1.3-2.5 184
0.80 - 1.3
0.54 - 0.80
filter £0.54
Andersen Impactor
130
16.5
< 7.1
650
OUTLET
CONCENTRATION
(mg/m3)
< 2.1
< 2.1
2.8
6.4
19.1
53.7
38.9
6.4
< 2.1
129
FRACTIONAL
MASS
EFFICIENCY
(percent)
> 96.0
> 96.0
96.0
92.1
85.0
70.8
70.0
61.4
-
80.1
                                B-27

-------
                                TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
. REPORT NO.
 EPA-650/2-74-083
                           2.
            3. RECIPIENT'S ACCESSION>NO.
. TITLE AND SUBTITLE

Dynactor Scrubber Evaluation
            5. REPORT DATE
            September 1974
                                                      6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
Douglas W. Cooper and Daniel P. Anderson
            8. PERFORMING ORGANIZATION REPORT NO.

            GCA-TR-74-21-G
. PERFORMING ORG \NIZATION NAME AND ADDRESS
GCA Corporation
GCA/Technology Division
Bedford, Massachusetts  01730
            10. PROGRAM ELEMENT NO.
            1AB012; ROAP 21ADL-004
            11. CONTRACT/GRANT NO.
            68-02-1316 Task 6
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
            13. TYPE OF REPORT AND PERIOD COVERED
            Final; Through 7/26/74
            14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT The reporf giV6s results of testing the Dynactor spray scrubber for power
 consumption and collection efficiency at three flow rates, two temperatures, and two
 dust loading levels, using two dusts. Total filter samplers and cascade impactors
 were used upstream and downstream from the collector.  Power was determined from
 voltage, current, and phase-angle measurements. A factorial design series of tests
 at two levels of flow, concentration, temperature, and dust type gave  these average
 mass efficiencies: 99.0% for  4.0-5.6 jum  aerodynamic diameter,  98.4% for 2. 5-4. 0
 tim,  93.0% for 1.3-2.5 jum, 75.4% for 0.8-1.3 /jm, 27.4% for 0.54-0.80 jum, and
 47.4% for < 0.54 /urn.  Higher efficiency was fostered by: lower flow rate,  lower
 inlet'temperature, and higher mass loading.  Power consumption was about one-third
 of that expected from a venturi scrubber with equivalent collection efficiency, but
 collection efficiency decreased dramatically for fine particles, those smaller than
 1
7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                          b.lOENTIFIERS/OPEN ENDED TERMS
                                                                  c.  COSATI Field/Group
Air Pollution
Scrubbers
Performance Tests
Dust Collectors
Efficiency
Air Pollution Control
Stationary Sources
Dynactor Scrubber
Fine Particulate
Collection Efficiency
Power Consumption
13 B
07A
14 B
13A
18. DISTRIBUTION STATEMENT

 Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF P,

    113
20. SECURITY CLASS (This page)
 Unclassified
                         22. PRICE
EPA Form 2220-1 (»-73)
                                        B-28

-------