Removal of suspended and dissolved solids from process ...
Transcript of Removal of suspended and dissolved solids from process ...
Lehigh UniversityLehigh Preserve
Theses and Dissertations
1973
Removal of suspended and dissolved solids fromprocess water-computer simulationThomas J. KopecLehigh University
Follow this and additional works at: https://preserve.lehigh.edu/etd
Part of the Chemical Engineering Commons
This Thesis is brought to you for free and open access by Lehigh Preserve. It has been accepted for inclusion in Theses and Dissertations by anauthorized administrator of Lehigh Preserve. For more information, please contact [email protected].
Recommended CitationKopec, Thomas J., "Removal of suspended and dissolved solids from process water-computer simulation" (1973). Theses andDissertations. 5104.https://preserve.lehigh.edu/etd/5104
.: .. _<,',_,_ ..... \ •.
REMOVAL OF SUSPENDED AND DISSOLVED SOLIDS FOOM PROCESS WATER-COMPUTER SIMULATION
by
Thomas J. Kopec
A Research Report
Presented to the Graduate Faculty
of Lehigh University
in Candidacy for the Degree of
Master of Science
in
Chemical Engineering
Lehigh University
Bethlehem, Pa.
May, 1973
CERTIFICATE OF APPROVAL
This research report is accepted and approved in partial fulfillment of the requirements for the degree of Master of Science in Chemical Engineering.
l!~~ (dat
ii
Dr. Curtis Clump Professor in Charg
Dr. Leonard Wenzel Chairman of the Department of Chemical Engineering
. \
ACKNOWLEIXiEMENTS
I would like to thank Dr. Curtis W. Clump for his supervision
and guidance during the course of this project. My thanks also go
to Dr. Alan S. Foust for his help in setting up the Bureau of Mines
program.
The author is also grateful to Dan Kwasnoski and George Haines
of the Bethlehem Steel Corp. for their time and effort in obtaining
data for the gas scrubbing operation at the Bethlehem Steel plant.
Thanks also go to Frank W. Koko and Richard Greene for their know
ledge and assistance in developing Program SOIJDS.
I would also like to express my grateful appreciation to the
Proctor and Gamble Co. for providing my financial support for my
graduate study.
iii
1 '·
i, ,,
TABLE OF CONTENTS
LIST OF FIGURES • • • • • • • • • • • • • • • • • • • •
LIST OF TABLES • • • • • • • • • • • • • • • • • • • •
ABSTRACT • • • • • • • • • • • • • • • • • • • • • • •
INTRODUCTION . . . . . . . . . . . . . . . . .. . . . . MODELING CONCEPTS I I I I I I I I I • I I I I I I I I I
PROCE3S DE3CRIPTI0N • • • • • • • • • • • • • • • • • •
DESIGN MODEL - PROGRAM SO LIDS • • • • • • • • • • • • •
ECO:OOMICS OF DESIGNED SYSTEM • • • • • • • • • • • • •
CASE STUDY . . . . . . . . . . . . . . . . . . . .. . .
V
V
1
3
6
7
15
18
21
RESULTS AND DISCUSSION • • • • • • • • • • • • • • • • 27
CONCLUSIONS AND RECOMMENDATIONS • • o • • • • • • • • • 29
APPENDIX • • • • • • • • • • • • • • • • • • • • • • • 31
RE:FER.ENCE3 • • • • • • • • • • • • , , , • • , • , , , 3 2
VITA • • • • • • • • • • • • • • • • • • • • • • • • • 34
iv
,,
'
f >"
'"
1
2
3
No, -l
2
3
4
LIST OF FIGURES
Removal of Suspended and Dissolved Solids from Process Water • • • • •• o ••••••
Capital Cost Versus Flow Rate Into Gas Scrubber •••••••••••••• • • • •
Operating Cost Versus Flow Rate Into Gas Scrubber •••••••••••
LIST OF TABLES
• • • • •
Equipment List • • • • • .. • • • • • • • • • •
Data Obtained From Bethlehem Steel . . . .. . . Build Up Rates For Dissolvable Solids
Economics For The Process ••••••
V
• •
• •
• • •
• • •
8
25
26
9
11
13
24
'I I l
l 1
I 1 1
l
l 1 1 1
ABSTRACT
The removal of dissolved solids from process water is a prob
lem which is facing the Bethlehem Steel Corp. today. The State of
Pennsylvania has set standards for the discharge of industrial
effluents. Bethlehem Steel cannot handle the removal of dissolved
solids from the bleed stream from their gas scrubbing system to meet
these standards with existing equipment.
A system was designed to remove the dissolved ions found pre
sent in this bleed stream. A fixed bed Ion Exchange system was the
technology used. The calculations were done by developing a computer
program to do the energy and material balances, size and cost the
equipment and calculate the utilities needed. The economic analysis
was done by a computer program developed by the Bureau of Mines. The
·program calculated the capital and operating costs for the process.
A study was made with the two computer programs. The water
flow into the gas scrubber was varied from 12,000 gallons per minute
to 8,000 gallons per minute. The programs were also run with the
dissolvable solids in the gas stream entering the scrubber set at
2.0 lbs. per minute and 10.0 lbs. per minute. The results show a
slight decrease in capital and operating costs as the water flow to
the scrubber is decreased. This is true for both dissolvable solids
rates. The capit~l and operating costs did drop appreciably for the
drop in disaolvable solids rates. The capital cost went from $3.3
1
million to $2.0 million. The operating cost was reduced from $1.6
million to $0.7 million. It was estimated that the capital costs
were within 30 per cent of the actual cost.
2
' .
J
INTRODUCTION
At present time, the Bethlehem Steel Corp. has five blast fur
naces in operation at Bethlehem, Pa. They produce a source of air
pollution in the form of solid particulate carried out of the top of
eac.h furnace by hot flue gas. The solid particulate can vary in com
position depending on the type of ore that is charged to the furnaces,
but will normally containSi02, Al2o3, Na2o, CaO, MgO, CaCl, Fa) and
Fe203.
Bethlehem Steel Corp. has in operation at this time a system to
remove this solid particulate from the gas. The dirty gas is first
passed from the furnace through a collection system to remove large
particles. The size of the particles in the gas leaving this system
will be 800 microns or less. The dirty gas is then passed through a
venturi scrubber to remove almost all of the solid particulate. The
gas is sent to another scrubbing system to remove the remaining amount
of particulate and then is burned as a low Btu fuel.
· Water is used as the fluid in the venturi scrubber to wash out
the particulate. Some of the particulate such as FeO and Fe203 have
essentially very little solubility in water. These solids will be
suspended in the water. Some of the solids such as Na2o and CaCl
will dissolve in the water. The water is treated and then recycled
back to the scrubber for reuse.
3
' ' . i I .
I I I
1
I I I I I J
'
l l J
l l
I
The treatment of the water consists of two operations: (1)
removal of suspended solids; (2) removal of dissolved solids. The
removal of the suspended solids is accomplished by the use of a
clarifier and a filter. However, a problem exists with the dissolved
solids. They begin to build up in concentration ap the water is re
cycled. The present solution is to bleed some of the recycled water,
which has a dissolved solids concentration of 2500 ppm, to the river
and add fresh river water. This lowers the dissolved solids concen
tration in the recycle stream.
Pennsylvania has laws governing the discharge of industrial ef
fluents into waterways, regarding the dissolved and suspended solids
concentration of the effluent. A maximum limit of 500 ppm was set
for dissolved solids and 200 ppm for suspended solids. The limit set
for the suspended solids can be met with existing equipment. A new
system is needed for the removal of the dissolved solids.
The most feasible approach, at the present, to the problem of
the removal of dissolved solids has been by the use of ion exchange
technology. Other methods may be applicable to this industrial
problem, but must be proven to be economically competitive as the
ion exchange process. One factor to the key to making ion exchange
economically attractive is the regeneration of the resin. An inex
pensiv~ regenerant and its recovery are the most important factors,
assuming that a suitable resin can be found to handle the stream
needed to be cleaned up.
4
A computer model, called SOLIDS, was developed to size and
cost the equipment need for the ion exchange process. The process
consists of the gas scrubbing and removal of the suspended and dis
solved solids (see Figure 1). The model also calculates the required
utilities and raw materials and their costs. The capital and oper
ating costs for the process were calculated by another computer,
which was developed by the Bureau of Mines• College Park process
evaluation group.
Bethlehem Steel Corp. supplied data from the actual gas scrub
bing operation. This data can be found in Table 2. Data that was
needed, but was not obtained was the chemical analysis of particulate
in the gas stream entering the gas scrubber. This makes the rate of
dissolvable solids in the gas stream an unknown. This rate is needed
to design the ion exchange system.
5
,i \ !
MODELING CONCEPTS
Today computer models are becoming more feasible to design
chemical processes. With simulation models, many process studies
can be made in a short period of time. These models can become
large in physical size, so the programmer should be aware of how
to develop the model. The model should have a definite structure
and adequate documentation so a user could follow the logic involved.
All major variables should be read into the model and not set in the
program. The model should be capable of printing all process condi
tions calculated.
The saving of computer time and core are also very important.
Overlays can be used to save computer core by writing over calcula
tions done previously. Simulation models should have generalized
calculations. A system could be developed which could design every
pump in the design. A generalized mass and energy balance package
is needed, but is difficult to develop.
6
i! ,I
. ~ i l
., ~
I l
PROCESS DESCRIPTION
The process (see Figure 1) designed removes suspended and d s
solved solids from the water from the gas scrubber. The treatment
of this water takes place in two sections: (1) suspended solids
removal} (2) dissolved solids removal.
The suspended solids section consists of a clarifier and a
filter. The underflow of the clarifier, containing the settled
solids, is sent to a rotary drum filter for final separation of the
solids. These solids are stored and reused later as part of the
feed to the blast furnaces. The overflow of the clarifier is split
into two streams. Most of it is recycled back to the scrubber feed
tank and the rest is sent to the dissolved solids section for clean
up.
The dissolved solids section makes use of a fixed bed ion ex
change system. The build up concentrations from the Bethlehem Steel
process can be found in Table 2. Four ions from this list were used
as the dissolved solids to be removed from the gas scrubber water.
They ares (1) sodium; (2) calcium; (3) chloride; (4) sulfate. These
ions are made to enter the process in a fixed ratio (see Table 3).
The removal of the sodium and calcium ion is done by the use of
a cation exchanger resin. The resin used for the design was Dowex
HCR-W. The hydrogen fonn of the resin was used. To regenerate the
spent resin, a six per cent by weight aqueous solution of sulfuric
7
l
.t I
'
' 'i
.t
. DIRTY GAS
GS- I
T- 4
P,1
TC ST :.r;t
TO RIVER
ST E i\ f/
BLEED
/
\\
FIGURE NO·
REMOVAL OF SUSPEND AND DISSO'L\/ED SOLIDS FROM FROCESS WATER
CLEA~ GAS Cl -1
RE.CYCLE STREt-1/
C :-l • I
E \' - I
R - I
·' '
,. c:_.··
T - 5
T - 6
T - I
~------------------------·----.------
E 2 B-1
FEED
p - 8 P • 13 P - I 2
p - 10
T - 3
P- 9
FEE J
I - E,
I L
I ii
Ii
Gas Scrubbing Section
Name
Gas Scrubber Scrubber Head Pump Scrubber Pump Scrubber Feed Tank
Suspended Solids Section
Clarifier Clarifier Pump Rotary Drum Filter Recycle Pump Feed Pump to Hold Tank
Dissolved Solids Section
Hold Tank Cation Exchange Beds Feed Pump - Cat. Beds Backwash Pump - Cat. Beds Cation Regenerant Tank Cation Reg. Tank Mixer Anion Exchange Beds Feed Pump - An. Beds Backwash Pump - An. Beds Anion Regenerant Tank Anion Reg. Tank Mixer Product Pump Spent Regenerant Pump - Cl Spent Regenerant Pump - C2
Table 1
Ei~UIPMENT LIST
9
Equipment No.
GS-1 P-1 P-2 T-4
CL-1 P-3 F-1 P-4 P-5
T-1 B-1 P-6 P-7 T-2
MX-1 B-2 P-8 P-9 T-3
MX-2 P-10 P-12 P-13
Spent Regenerant Section
Name
Cation Spent Reg. Tank Anion Spent Reg. Tank Spent Regenerant Pump - C2 Spent Regenerant Pump - A2 Reactor Evaporator Bottoms Pump Condenser Condenser Pump Ejector Cooling Tower System
TABLE 1 ( cont.)
10
Equipment No.
T-5 T-6 P-14 P-15 R-1
EV-1 P-17
CN-1 P-18
EJ-1 CT-1
TABLE 2
PROCESS DATA OBTAINED FIWM BErHLEHF.M STEEL CORP.
Gas Flow Rate - 140,000 SCFM
Gas Composition -MOLE%
26 16 3
52 3
water Flow to Gas Scrubber - 12,000 GPM
Build up of Dissolved Solids in Bleed Steam -
ION PPM
Fe 2 Zn 40 Al 6 Ca 110 Mg 75 so4 900 Cl 800 Si02 30 Na 850
Amount of Filter Cake - 75 tons/day (dry)
Analysis of Filter Cake -
Fe C Zn Si0 2 Ca Mg Al20J
11
wt.%
65.6 7.8 3.3 6.4 2.9 2.1 2.3
--- . ----~---·-------·-··--. --------··--·----
TABLE 2 ( cont.)
Particle Size Distribution of Particule Entering Gas Scrubber
Wt. % of Total Stream
6.0 10.0 30.0 50.0 70.0 90.0 98.0
12
Size (Microns)
700 660 440 280 100
75 50
TABLE 3
BUILD UP RAT:ES* FOR DISSOLVED IONS IN RECYCLED GAS SCRUBBER WATER
ION Wt. fraction
Na 0.3195 Ca 0.0414 Cl 0.3008 so4 0.3383
1.0000
* For every pound of dissolvable solids in the gas stream entering the scrubber, the distribution will be constant and will be in accord with the values in the table.
13
acid was used. The chloride and sulfate ions were removed by the
use of Dowex SBR anion exchange resin. The resin has only a chloride
form, so the resin must be pretreated with the anion resin regenerant,
which is a four wt. per cent aqueous solution of sodium hydroxide.
The cleaned-up proces3 water from the ion exchance system is
essentially free from dissolved ions, except for leakage f'rom the
resin. This stream is then added to the scrubber feed tank, where
it is mixed with the recycle stream. The exit stream from the tank
should be the same concentration as the previous pass. If not the
amount of recycle should be adjusted accordingly.
The dissolved solid section also produces two waste streams:
( 1) spent sodium hydroxide solution; ( 2) spent sulfuric acid solution.
These two stream.s are added together in a reactor and the following
neutralization takes place:
::::
There is an excess of sodium hydroxide, so all the sulfuric acid is
used. The exit stream from the reactor goes to an evaporator to
concentrate the stream. The bottoma of the evaporator are sent to
a pond for final drying. The evaporator operates under vacuum so
that the overhead product, pure water, is at a temperature where it
can be used as make up water or dumped into the river. A two-stage
jet ejector with intercondenser was used to produce the vacuum for
the evaporator.
14
DESIGN MODEL - PROGRAM SOLIDS
Program SOIJDS was developed to simulate the process in Figure
1. The program does the calculations for the mass and energy balances
needed for each piece of equipment. These balances are then used to
detennine the sizes of all the pieces of equipment. The purchased
costs are then calculated by the use of analytical expression for the
available cost data. These costs are updated to January 1, 1973 by
the use of the Marshall and Stevens Index. The index at this time
was 336.7. Installed costs are also calculated for each piece of
equipment by using a materials and labor factor to correct the updated
purchased cost. The utilities needed, such as process water, steam
and electricity are calculated per section. A summary at the end of
each section gives the totals for each utility used.
All these calculations were set-up so that they are flexible.
This flexibility shows itself in allowing the user to change many
process variables in the system. The program shows the results in
changes of equipment sizes, costs and utility use. The user has con
trol over this flexibility through the use of 41 data cards. F.ach
card contains the process variables for each piece of equipment. The
ion exchange system has six data cards to allow for changes in almost
all the variables in the system. The documentation for these data
cards can be round in the body of the program.
The program itself has documentation in each subroutine and in
the main line program explaining the logic and calculations involved.
The program also has a variable list in the main body of the program
to allow the user to follow the calculations. The documentation and
variable list are an important part of any program. Without them the
program becomes useless.
The overall design of the program has some strong points and
also some weaknesses. The program is able to handle the design of
pumps ranging from 12,000 GPM to 1 GPM. The cost data may need some
refinement, but the design mechanism can handle this range of flow
with no problem. This kind of flexibility is indicative of all the
pieces of equipment that are designed. The program also has physical
data programmed as a function of temperature to allow for temperature
changes in the system. To check on this flexibility the program pro
duces a large amount of output. The user can find a printout of the
size, cost and process conditions for each piece of equipment. The
printout is well spaced so the user can find what he is looking for.
Program SOIJDS does have some weak points. One of the major
points is its treatment of the spent regenerant streams. The current
design uses neutralization and concentration to treat these streams.
This is not very efficient use of the regenerant streams. A more
efficient system to recover some of the spent regenerants is needed.
There also is a question about some of the cost data obtained
and the range it is valid to. The program Jan only give an order
of magni~ude cost for the process because of this uncertainty. The
16
program is also limited in its design range in amount of physical
data prograrruned and its range. This information on the design
range can be found in the subroutine which has the data stored.
17
ECONOMICS OF DESIGNED SYSTm
The calculation of the equipment costs by Program SOLIDS is
only part of the economic analysis of the system. The total capital
and operating costs are also needed to determine if the process is
economically promising. To do these calculations another computer
program was used. The program that was used, Program CAOPCO, was
written by the Bureau of Mines• College Park process evaluation
group. It was written by the group to be used on an IBM 7094 com
puter. Dr. Alan S. Foust, McCann Professor, Department of Chemical
Engineering, Lehigh University, has taken this program and adapted
it for use on the CDC 6400 computer at Lehigh University.
The description and operation of the program can be found in
the Bureau of Mines Information Circular 8426. The circular also
gives a description of how the calculations are done and some of the
philosophy behind them. The data cards needed are listed in the cir
cular along with the needed formats and an explanation to their ase.
Some documentation of the program and data cards can also be found
in the body of the program.
Program CAOPCO, as adapted by Dr. Foust, could not handle the
economics for the solids removal system. The program had been set
up to handle solids or liquids with operating costs in the order of
magnitude of$ 0,01 per unit of reference material and higher. The
costs tor the treatment of water are of th~ order of$ 0,10 per 1000
18
gallons and higher. On the basis of one gallon, the cost would be
of the order of$ 0.0001. The program will only handle operating
costs per one unit of reference material, so it had to be changed
to work with costs of this order of magnitude. Changes to the data
cards as set up by the Bureau of Mines' group can be found in the
Appendix.
The output taken from Program SOLIDS to be put into Program
CAOPCO is as follows, (1) purchased cost per piece of equipment;
(2) number of pieces of equipment; (3) size of piece of equipment;
(4) utilities used per section; (.5) total raw materials used per
day; (6) quantity of reference material per day. Program CAOPCO
needs much more data than this for it to run. The rest of the input
is in the fonn of factors to take into account, maintenance, labor,
supervision, operating supplies, etc.
One problem that occurred with Program CAOPCO was with the
handling of the steam. The program wanted to put in a steam plant
if any steam was needed, whether the user wanted it or not. To
solve this problem, no steam was specified as needed by the process.
The amount of steam needed, lbs. per day, was read into the program
in place of raw water needed. The cost for steam,$ 0.0005 per lb.
was read into the program in place of the cost for raw water. This
can be seen in the output from the program.
The output from Program CAOPCO was in tabular form, making it
' very easy to read. The equipment was listed by section with their
individual costs and sizes and a total sectional cost. After the
19
equipment lists the following tables can be found: (1) Estimated
Capital Cost; (2) EGtimated Working Capital; (3) Estimated Annual
Operating Costs; (4) Raw Material and Utility Requirements; (.5)
Direct Labor, Men Per Shift; (6) Daily Sectional Requirements.
One table was not used in the analysis, which was Estimated Selling
Price.
Program CAOPCO, when modified, was more than adequate to
handle the economics of the designed system. The costs from the
design model were not in all cases accurate, but rather approximate
cost. The economic analysis gave capital and operating costs which
also were approximate. These costs can be made more accurate by
making the factors which enter into the program more in line with
the actual process. Better equipment costs could also be used to
increase the accuracy. This though is not needed for this design
project because the main goal was the order of magnitude of the
costs.
20
! ;
CASE STUDY
In order to test Program SOLIDS for its flexibility and to see
what the economics are for the designed system, a study was run using
both computer programs. As mentioned in the Introduction, the amount
of dissolvable solids in the gas stream entering the gas scrubber was
unknown. The amount of suspendable solids entering was known at about
104 lbs. per min. By studying other blast furnaces using many differ
ent types of ores as feed, it was concluded that the amount of dissolv
able solids entering the gas scrubber was in the range of 10 lbs. per
min. The accuracy of this number is in question, but it was used.
This study and one made of the feed conditions at the Bethlehem Steel
furnaces indicate that this number may be too high. The value of 10
lbs. per minute was used as an upper limit and 2 lbs. per minute was
set as the lower limit.
Another variable thought to have an affect on the economics of
the process is the flow of water entering the gas scrubber. Table 2
shows that the flow for the actual process is 12,000 gallons per
minute. The question was then raised, what happens to the capital
and operating costs of the process if this flow is lowered? To ans
wer it, the study was also set to include flows of 12,000, 10,000
and 8,000 gallons per minute. Lower flows were not used because many
of the process variables were set for these high flows.
21
The procedure used for the study was to run Program SOLIDS
using the upper limit for the dissolvable solids and then vary the
flow. The same procedure was also followed with the lower limit.
In order to run this study another process variable had to be con
sidered. This variable was amount of water that was recycled with
out clean up.
Program SOIJDS was set up to read in the concentration of the
dissolved solids in the water entering the gas scrubber. This con
centration increases because the dissolvable solids in the gas stream
as washed by the water. The recycle factor tells how much of this
water is cleaned up and how much is recycled without clean up of
dissolved solids. These two streams are then mixed together in the
scrubber feed tank. If the concentration of dissolved solids in the
tank exit stream is the same as the value read in, then the correct
amount of water was recycled. If not, the amount of recycle has to
be adjusted.
The dissolved solids concentration for the water entering the
scrubber was read into Program SOLIDS as 2000 ppm. If the calculated
value was less than 2000 ppm, then too much water was cleaned up and
more water should be recycled. If the calculated value was higher
than 2000 ppm, then not enough water was cleaned up and less water
should be recycled.
The data from each correct run from Program SOIJDS was then
put into Program CAOPCO. The economics obtained are listed in Table
4. Figures 2 and 3 show the capital and operating cost for the
22
designed system as a function of the flow of water to the scrubber.
The amount of dissolvable solids in the gas stream entering the
scrubber was used as a parameter.
23
TABLE 4
~NOMICS FOR THE PROCESS
Dissolvable Solids into Scrubber
2.0 lbs/min. lO.O lbs/min.
Flow l2,000 l0,000 8,000 l2,000 l0,000 8,000
(GPM)
Capital Cost 2,022,320 2,054,590 l,855,330 3,399,740 3,307,650 3,226,980
I'\) {$) ::-
Opera ting Cost 723,856 717,787 665,172 1,682,594 l,657,457 1,645,425
($/yr.)
Operating Cost 0.13 O.l6 O.l9 0.32 0.38 0.47
($/lOOO gal)
Recycle 0.990 0.987 0.985 0.951 0.942 0.928
:. ' l '. ' '. .T ··1··.,1· .. , I
•• I ' ' " : ! .
' '
T I ·,··· ·: .. :T '. ·r· :·''!'···;····1··:· ·,
! :. I I : i ! : . I '
. I'. '· · I FIGURE~ NO •.. 2 . . I i ii
, I ' i ; ( \ :•
..... ~APITAL:: COf3T : vEks~s FLOW. ~~E. INTO GAS SCRUBBER ! I I \ . I I ' I ; .
! . ! I· '..:. I . :: ! ..
4.20 ! I
3.80
... i ..... : ..... I
3. 4 O·
I i I I 1
i I I I
• • I
I ':.
· \ I
',,,.
. . . .. ..... I. I :.. . , I I
i i
I
I
I . . . I .
! I
I i
I . I
I'
~o.o LBS/MIN DISSOLVABLE SOLIDS IN GAS STREAM
. ENTERING 'THE .SCRUBBER'
' . --·--. ·-·- ,-•-·-· ·-·
I . : I
. i ' . i· .
'
' '
•.•.•.•. . ....... : r T r I r : . .-• • • · •.. . ; i i . . I • I : F~GURE; NO, ... 3.
:~:···· +•·t·•··CRTt•nT•rRSUf>I {LOW ~T~ INTO GAS SCRUBBER ' .. ' I· ' I 'I ' I
-·-.. ···-·. : ': I : i I l I .•• ·- ••• 1 ···-~. ·---·· ••• ·-· ..... -· ' •••••• , •• ,_ .. , ·-· -- . -~---
1
.. , . · · · 1 · I : : : ' ' : \ ' I : " . i ' i ' . . II .
:·::::·"\":::: .. ·,1 .. : .... 1 .:: ::, ·:· i i -:; : : ';_ _ _:__.L~-~.: ... I ~--~-i . .: .. JQ_.:O_ LBS/MIN DISSOLVABLE. SOLIDS. IN. GAS STREAM.
: -:::. :. ::::· : : ] ! :: , .. ::·ENTERING ;THE .SCRUBBER • I • ' I
I : .. I l . ·1 ' I ) 1
/ I ' \
I -----~--75·-r -- ·- --·-i··--:----1---·--: . -1---·---- :·---:---:--··· ··--·- -J --·--- - -
• ' ·1 •••••• • •• I •• • t , , l ' , ,
I : .. I ' : : I ' ' 0
I , I!) l3
'r--• ; .. , . :: . ·······--·--·····------..l----····1· .. i
1 •
I . 1 . • . . .
i ' I " '
i--·- .. --1·. 5 o·-r · ----: · · -- ·-··: ... ----- : ______ :_ ---------·-- ·- --·--1 I I ! I ' 1 · • •: .... • . . • • • ' . . . I.. ' • . : • .. • ' .
I . . . r-~--·- ·-·-·--·--· _____ ,_' --··-----!.. :>; : : : I •• • • I '
. '-. ! ..... I ..
: ~ : ••••• ' ·- I • ·--·-·· ---·---·-·· __ ,_. ·--- -- ·-' ~--'(f)-·-1-.-2s- : - , - -------·------···------------·-----
' I , i .. I ' ...... .
I
... !. .. . : !·· . ,. ' I ,
1 __ . : . . L. ___ \ ___ · _.__ ___ ; --- - -- -----··---,.
'j
RllSULTS AND DISCUSSION
The results of the economic study show trends that were ex
pected. Figure 2 shows the capital cost decreasing slightly as the
flow to the scrubber decreases. This is true of the 10.0 lbs. per
minute dissolvable solids rate. For the 2.0 lbs. per minute rate,
the curve shows a maxima at 10,000 gallons per minute. The capital
cost for the flow of 8,000 gallons was lower than that of 12,000
gallons per minute. The change in the capital cost for the two dif
ferent dissolvable solid rates was appreciable. The capital cost
for 10.0 lbs. per minute rate is in the range of $3.3 million and
for the 2.0 lbs. per minute rate the cost was around $2.0 million.
These results show that the changes in flow have only a small
effect on the capital. This is true for both dissolvable solids
rates. As for the maxima which occurs in the lower dissolvable
solids rate, it is probably due to the design of the system. The
system was set up originally for higher fiows and dissolvable solids
rates. If all the variables were optimized for lower flows and sol
ids rates then this maxima would probably disappear and the curve
would show the decreasing trend.
As for the difference in capital cost for the two solids rates,
the results show that the economics are very sensitive. The capital
cost dropped by $1.3 million for the drop in solids rate. This kind
of sensitivity says that more studies should be made to find the
27
,/
lk
economics of dissolvable solids rates inside and outside the range
in question.
The operating costs show the same decreasing trend. Again
there was an appreciable difference in operating cost between the
two solids rates. The operating cost for 10.0 lbs. per minute rate
was in the range of $1.6 million with about half of this cost used
on raw materials (regenerants). The cost for 2.0 lbs. per minute
rate was around $0.7 million and about one quarter of the cost was
used on raw materials.
The capital costs predicted by the Bureau of Mines are expected
to be within 30 per cent of the actual cost. An analysis of the pro
gram shows it to be in the order of a study estimate. The study
estimate uses average factors to obtain costs and also a knowledge
of the major items of equipment. The value of! 30 per cent is also
listed in the description in the Bureau of Mines program as an esti
mate of the accuracy of the capital costs obtained.
One of the most important questions to be asked of the entire
project is, will the design work? The design was based on known
technology and no new developments. The operating conditions and
the resin used in the ion exchange section were all suggested by the
Dow Chemical Co. literature. The other sections used standard pieces
of equipment and they were set for attainable conditions. Based on
this knowledge, it was estimated that the design will work. The
process conditions set may not be the optimum but they should make
the process run.
28
(X)NCLUSIONS AND RECOMMENDATIONS
It is quite evident that more work is needed in the area of
ion exchange for the removal of dissolved solids. The importance
of obtaining the best design is magnified by the fact that this type
of process produces no product which can be sold. The entire capital
and operating costs must then be absorbed by the existing facilities.
Better ways are needed to predict equilibrium relationships
between solids and liquids. Organic compounds have received a great
deal of attention for removal from process water, but what about
inorganic ions? Ion exchange technology has been used, but better
systems are needed for industrial waste streams with high flows and
dissolved solids concentrations.
Research is also needed in finding other methods for the re
moval of dissolved solids. One method which shows potential is
removal by the use of membranes. An economic analysis must be done
on such a system to determine if it can be economically competitive
with ion exchange.
A smaller area of research needed is in the solubility of many
inorganic ions in solution. Physical data is needed to show the be
havior of these ions in mixtures and also to predict the saturation
point for process water. It would seem that this type of research
may have to be done by the particular company with the dissolved
solids problem at the present time.
29
Design technology also plays an important role in solving the
problem at hand. It is obvious in ion exchange that a regenerant
recovery system is needed. This would show a decrease in the oper
ating cost for the process. A possible approach to this problem
would be to recycle the regenerant and let the concentration of the
spent regenerant in the total stream build up to some point. Then
fresh regenerant could be added and some spent regenerant bled off.
What is also needed is another way to treat this spent regenerant
stream other than by neutralization. A comparison would then tell
which is the more economical way of treatment.
30
APPENDIX
Changes Made in Data Cards From Bureau of Mines Program
Data Card No. 5
Columns 13 - 18
Fonnat changed from F6.2 to F6.5
Data Card No. 12
Columns 1 - 7 changed to 1 - 12
Format changed from F?.O to Fl2.0
(rest of data card push back five spaces)
Data Card No. 13
Columns 50 - 59
Format changed from FlO.O to Fl0.7
31
r
REFERENCES
l. Applebaum, Samuel B. Demineralization by Ion Exchange. New Yorki Academic Press, 1968.
2. Aries, Robert S. and Newton, Robert D. Chemical Engineering Cost Estimation. New York, Chemonomics, Inc., 1951.
). Bethlehem Steel Corp. Bethlehem, Pa. Production records. July, 1972.
4. Chilton, Cecil H., ed. Cost Engineering in the Process Industry. New Yorkz McGraw-Hill Book Co., 1960.
5. Dahlstrom, D. A. and Cornell, C. F. "Thickening and Clarification," Chemical Engineering (4), Vol. 78 (1971), pp. 63-69.
6. Dow Chemical Co., Midland, Michigan. Ion Exchange literature.
January, 1973.
7. Dryden, Charles E. and Furlow, Richard H. Chemical Engineering Costs. Department of Chemical Engineering, Ohio State Uni
versity, 1966.
8. Foust, Alan S., et al. Principles of Unit Operations. New York: John Wiley and Sons, Inc., 1960.
9. Happel, John. Chemical Process Economics. New Yorki John Wiley and Sons, Inc., 1958.
10, Harbison-Walker, Refractories Co. A Study of the Blast Furnace. Pittsburgh, 1911.
11. Johnson, Paul w. and Peters, Frank A. A Computer Program For Calculatin Ca ital and O eratin Costs, Information Circular
2 • U.S. Dept. of the Interior, Bureau of Mines, Washington,
D.c., 1969.
12. Kunin, Robert. Ion Exchan~e Resins, 2nd ed. New Yorks Wiley and Sons, Inc., 19 8.
John
13, Ludwig, Ernest E. Applied Process Design For Chemical and Petrochemical Plants. 3 Vols, Houstom Gulf Publishing
Co~, 1964.
32
14. McCabe, Warren L, and Smith, Julian C. Unit Operations of Chemical Engineering, New York: McGraw-Hill Book Co., 1967.
15, McGannon, Harold E. The Making, ShapinG and Treating of Steel. Pittsburgh, United States Steel, 196.
16. Perry, John H,, ed. Chemical Engineers' Handbook, 4th ed. New Yorks McGraw-Hill Book Co., 1963.
17, Peters, Max S. and Timmerhaus, Klaus D, Plant Design and Economics for Chemical Engineers. New Yorks McGraw-Hill Book Co., 1968.
18, Popper, Herbert. Modern Cost-Engineering Techniques. New York: McGraw-Hill Book Co., 1970,
19. Porter, H, F. "Filter Selection, 11 Chemical Engineering (4), Vol. 78 (1971), pp. 39-48.
20, Schweyer, Herbert E. Process Engineering Economics. New York: McGraw-Hill Book Co., 1955.
21. Stern, Arthur C, Air Pollution, Sources of Air Pollution and Their Control, Jrd vol. New York: Academic Press, 1968.
22, Strassburger, Julius H., ed. Blast Furnace - Theory and Practic~ 2 vols. New Yorks Gordon and Breach Science Publishers, 1969,
23, Strauss, W. Industrial Gas Cleaning, New Yorks Pergamon Press, 1966.
24. Strauss, w. Air Pollution Control, Part 1. New Yorks WileyInterscience, 1971,
25, Sweetser, Ralph H, Blast Furnace Practice. New York: McGrawHill Book Co., 1938.
26. Vilbrandt, Frank C. and Dryden, Charles. Chemical En ineerin Plant Design. New Yorks McGraw-Hill Book Co., 19 9,
27. Weast, Robert C., ed. Handbook of Chemistry and Physics, h 7th ed. Cleveland: The Chemical Rubber Co,, 1966.
28. Weber, A. P. "Selecting Turbine Agitator," Chemical Engineering (25), Vol. 71 (1964), PP• 169-174.
· 29. White, Walter F. Chemical Character of Surface Water in Pennsylvania, 1946~1949, Pennsylvania Department of Commerce, State Planning Board, 1951.
33
VITA
The author was born on May 4, 1949 in Passaic, New Jersey to
Bruno and Frances Kopec. In June of 1967, he graduated Garfield
High School with honors and in the Fall of 1967 he entered Newark
College of Engineering. The author received a B.S. in Chemical
Engineering in June of 1971 from this institution.
Upon completion of the requirements for a Masters Degree in
Chemical Engineering, the author went to work for E. I. DuPont de
Nemours and Co. at Deepwater, New Jersey as a development engineer
in the Elastomers Division.
34
I
I . -- -- .. . ___________________ \ __