ba-1960-0027.ch009

Water Purification by Zone-Freezing WILLIAM H. MINK, GEORGE F. SACHSEL, and R. B. FILBERT, Jr.

Downloaded by UNIV OF NEW SOUTH WALES on September 5, 2015 | http://pubs.acs.org Publication Date: January 1, 1960 | doi: 10.1021/ba-1960-0027.ch009

Battelle Memorial Institute,505King Ave., Columbus 1, Ohio

The economics of purification of saline water by zone-freezing were investigated using analog simulation as a tool to optimize the design. Estimated costs for the process were found to be too high to make it competitive with other processes now under development.

F or some time the U n i t e d States Department of the Interior has been carrying out a program aimed toward the selection of a n economical method of obtaining potable water from sea water. One method investigated at the Battelle Memorial Institute (1) is a n adaptation of the zone-purification process which h a d previously been used satisfactorily i n the purification of metals (8). I n the process, as applied to purification of sea water, a narrow zone of water is frozen i n a tube containing sea water. A s this zone is made to traverse the length of the tube, the formation of ice crystals tends to concentrate the salt i n the solution ahead of the crystals. T h i s results i n the concentration of the salt at one end of the tube and the depletion of salt at the other end.

In SALINE WATER CONVERSION; Advances in Chemistry; American Chemical Society: Washington, DC, 1960.

76

ADVANCES IN CHEMISTRY SERIES


The Vertical-Tube Unit Before costs could be estimated with any degree of accuracy for a possible largescale operation, it was necessary to select some type of apparatus i n which the process could operate i n a continuous manner. T h e apparatus which Battelle believed would offer the greatest advantage from a n economic and technical standpoint was the vertical-tube unit which is similar to the zone-void refiner described b y Pfann (4) a n d is shown i n Figure 1. I n this unit a number of vertical tubes pass through closely spaced, insulated, horizontal plates. T h e spaces between the plates are connected b y automatic valves to either the heating or the cooling part of a refrigeration system. It is thus possible to have any space between the plates be a melting or a freezing zone. Sea water is introduced near the top of the tubes. B y selective operation of the valves, freezing and melting zones move upward through the tubes. Sea water passes down the tube and is subject to a zone purification each time it passes a freezing zone. It leaves the bottom of the tube as purified water, while the salt is concentrated in the form of brine at the top of the tube, where it is removed periodically b y flushing. %ΈΜΠΏ

V///////X BRINE OUT

SEA WATER IN

NWVM

AIR A

Figure 2.

Β

C

0

BRINE ANO/OR SEA WATER FREEZING ZONE FRESH WATER

VALVE OPEN Ε

F

β

Schematic representation of vertical-tube process

T h e sequential operation of a two-stage, vertical-tube, zone-freezing unit is shown in Figure 2. I n this figure, for purposes of clarity, the individual zones are shown large in comparison to the tube. I n an actual unit the zone spacing would be a fraction of the tube diameter. I n Figure 2,4, a single tube is shown as it might appear during operation. I n the melting zone ice melts and the resulting water falls through an air space to the water below. A r o u n d the water is a dead zone and no heat is being trans ferred i n either direction. Below the water is an ice plug formed b y a freezing zone. T h e lower end of the bottommost ice plug melts, and water falls through an air space into the reservoir for pure water. However, since there is no passage for air into the tube, the level of pure water will rise in the tube as the ice melts. A t some time later the conditions shown in Figure 2,B, will exist i n the tube. Operation of the automatic valves has shifted each zone upward one space. T h e top plug of ice melts as before and becomes smaller. Water falls from this zone through the air space to the water below. T h e second plug has i n effect been moved upward, since the dead zone which had previously been above it has now become a freezing zone and the freezing zone which had previously been below it has now become a melting zone. A t the bottom of the tube a third plug begins to form.


MINK, SACHSEL, AND FILBERT—ZONE-FREEZING

77

In Figure 2,C, which shows the tube at some time after that illustrated i n Figure 2,B, the top plug has melted completely. T h e water i n the top of the tube is concentrated to a salinity considerably above that of sea water, so that at this point the brine is flushed out and replaced with fresh sea water. T h e remainder of the tube operates as described previously.


A s the zones continue to move upward as a result of the operation of the automatic valves, the plugs shown i n Figure 2,D, also move upward. T h e water above the top plug becomes more concentrated because its volume is being reduced and because of the segregation action of freezing. T h i s segregation action is the tendency to exclude salt from the ice formed, thus concentrating the salt i n the liquid phase. U p to this point there has been no net downward flow of water. Some time later the conditions shown i n Figure 2,E, exist. T h e ice plugs continue to move upward as before. However, air is permitted to enter the /bottom of the tube, so that as the bottom plug melts, the outside level of the reservoir is maintained within the tube, and purified water overflows. W h e n the air space has reached the desired volume as shown i n Figure 2J?, the air inlet is closed and the plugs continue to move upward, concentrating the saline water. In Figure 2,G, a cycle has been completed and the system is at the point shown i n Figure 2,A. T h e air space below each plug is necessary to obtain a net downward flow of water. T h e air space rises i n the tube and is displaced b y water as shown i n Figure 2 , C ; i n Figure 2,E, air displaces water.

Simulation of Operation of the Vertical-Tube Unit T o obtain both equipment and operating costs it was necessary to determine the optimum number of stages. T h e number of stages required to produce a given purity of water is i n turn a function of the configuration a n d size of the vertical-tube unit, the segregation coefficient, and the extent to which the brine is concentrated (or converted) before discharge. Although mathematical relationships have been developed for the zone-void refiner (4), it appeared that the quantity of data required for optimization of the vertical tube unit could be obtained more efficiently b y simulation on an analog computer. T h i s was accomplished b y simulation of each freezing zone individually and then b y means of a programming relay connecting and disconnecting the proper zone at the proper time. A simplified block diagram of the computer setup is shown i n Figure 3. T h e information obtained from the simulation was the salinity of the product water as a function of time. A n actual tracing of the output of the computer is shown in Figure 4. B y integrating the output it was possible to obtain an average value of product salinity for a given set of conditions. B y obtaining product salinities for different sets of operating conditions it was possible to develop an expression correlating the concentration of salt i n the output with different operating conditions. T h i s r e lationship is shown i n Figure 5, where the curves characterize a procedure to yield a product water of 500 p.p.m. salt from sea water with 35,000 p.p.m. of salt.

Experimental Studies Before the curves of Figure 5 can be used, two things must be determined: the variation of the segregation coefficient, k, with the temperature of the coolant outside the tubes; and the time required to freeze a zone of ice i n the tube (not needed for simulation, because it does not affect the product salinity) which affects the production rate from a given unit. T h e effect of temperature difference on the segregation coefficient was determined experimentally. Although the segregation coefficient data v a r y with concentration, the effect is small i n this type of system and was considered to have no significance i n the economic study. Tubes ranging i n size from 7/16 to 1 inch containing salt water were immersed in a constant temperature bath and the segregation coefficient was determined.



78


H

ZONE SIMULATOR

ZONE SIMULATOR

ZONE SIMULATOR

TIME OELAY

ZONE SIMULATOR

_

INITIAL CONDITION

REPETITIVE OSCILLATOR

Figure 3. Simplified block diagram for simulation of a four-stage vertical tube zone-freezing unit

Figure 4. The instantaneous salinity out from a two-stage zone-freezing unit

§

0.6

0.2

0.3

0.4

0.5

0.6

SEGREGATION COEFFICIENT, k

Figure 5. Conversion and segregation coefficient required to produce water with 500 p.p.m. salt for various number of stages In SALINE WATER CONVERSION; Advances in Chemistry; American Chemical Society: Washington, DC, 1960.

79

MINK, S AC H SEL, AND FILBERT—ZONE-FREEZING


It was found that, although tube size has no effect, the temperature difference has a large effect on the segregation coefficient, as shown i n Figure 6.

ο ι—; Ο

1 0.2

ι

I

I

1

0.4

0.6

0.8

1.0

SEGREGATION COEFFICIENT, k

Figure 6. Effect of temperature difference on segregation coefficient Using reasonable values of film coefficients and thermal conductivities, i t can be shown that the time to freeze to the center of a tube is given approximately b y : φ

=

3.2Ρ*·» At

where Τ = time required to freeze, hours D = inside diameter of tube, inches At = temperature difference between coolant and water, ° F . A s a check on the above equation, the time required to freeze i n a 1-inch tube was determined experimentally. I n one case, using a At of 17.8° F . , the observed time was 13.4 minutes and the calculated time was 18.7 minutes. I n another case, using a At of 9.7° F . , the observed time was 27.5 minutes and the calculated time was 34.8 minutes. I t can be seen that the theoretical calculation gives conservative values. T h i s m a y be due to the difficulty of observing when crystal formation begins, a n d to the possibility of supercooling, both factors tending to reduce the observed freezing time.

Estimate of Costs F o r optimizing the operating cost of the process, the variable operating cost (which is only that cost affected b y the design a n d operating parameters of the vertical-tube unit) was determined as a function of number of stages and conversion using informa tion from a heat exchanger manufacturer. T h e variable, operating cost is plotted i n Figure 7 for units containing two to six stages. I t can be seen from the curve that the lowest variable operating cost occurs at four stages, with a conversion of 0.13. U n d e r these conditions the At should be 3 ° F . Cost estimates were made for plants producing 100,000 gallons per day and 10,000,000 gallons per day of product water containing 500 p . p j n . of dissolved solids from sea water containing 35,000 p.p.m. of dissolved solids, using the procedure outlined b y the Office of Saline Water (#). F o r the larger plant, a total of 77,000,000 gallons per day of sea water is required, of which 67,000,000 gallons per day are discharged as brine


80



β.ο

5.01

I

0

0.1

I

I

I

I

I

0.2

0.3

0.4

0.5

0.6

CONVERSION, θ

Figure 7. Variable operating costs for the vertical-tube zone-freezing process containing about 40,000 p.p.m. of dissolved solids. I t was estimated that 387,000 kw.-hr. of electric power would be required per day. Capital costs are shown i n Table I and operating costs i n Table II. Investment plays a large part i n the operating costs;

Table I.

Capital Costs for the Vertical-Tube Zone-Freezing Process Capital Costs, Dollars 10 g a l / d a y 10 gal./day plant plant 5

Item Vertical-tube unit, installed Standard engineering equipment, installed Other plant costs Working capital Total investment Unit investment, dollars/gal./day

Table II.

892,000 636,000 1,054,000 57,000 2,639,000 26.40

7

10,020,000 52,940,000 51,769,000 2,420,000 117,149,000 11.70

Operating Costs for the Vertical-Tube Zone-Freezing Process Operating Costs, Dollars/Day 10* gal./day 10 gal./day plant plant 7

Item

Power Labor Amortization Other Total Cost per 1000 gallons of product water

27 103 578 251 959 9.59

2,710 3,227 25,700 10,504 42,141 4.21

it accounts for about 60% of these costs. Increasing the plant size beyond the 10,000,000-gallon-per-day capacity would result i n only a slight decrease i n the estimated operating costs, since multiple units would be required and other costs would change little. T h e possibility of better design does not hold m u c h promise for reduced costs either. T h e principal difficulty is that low freezing rates are required to obtain low


MINK, SACHSEL, AND FILBERT—ZONE-FREEZING

81

s e g r e g a t i o n coefficients. I f t h e f r e e z i n g r a t e i s decreased e n o u g h t o p e r m i t o p e r a t i o n o f a single-stage u n i t , t h e p r o d u c t i o n r a t e f r o m t h e u n i t a p p r o a c h e s z e r o . A t t h e p r e s e n t t i m e , t h e a u t h o r s h a v e n o p u b l i s h e d cost d a t a o n o t h e r c o n v e r s i o n processes o b t a i n e d o n t h e s a m e basis as t h o s e p r e s e n t e d here w i t h w h i c h t o c o m p a r e t h e r e s u l t s . T h e e s t i m a t e d costs a r e , h o w e v e r , a b o v e t h e d e s i r e d l e v e l set b y t h e Office of S a l i n e W a t e r .

Literature Cited ( 1 ) H i m e s , R. C., Miller, S. E., Mink, W. H., G o e r i n g , H. L., Ind. Eng. Chem. 51, 1345


(1959).

(2) Office of S a l i n e W a t e r , U. S. D e p a r t m e n t o f the I n t e r i o r , "A S t a n d a r d i z e d P r o c e d u r e f o r E s t i m a t i n g C o s t s of S a l i n e W a t e r C o n v e r s i o n , " March 1956. ( 3 ) P f a n n , W. G., J. Metals 7, 297 ( F e b r u a r y 1955). (4) P f a n n , W. G., "Zone Melting," p . 119, Wiley, New York, 1958. RECEIVED for r e v i e w July 7, 1960. A c c e p t e d July 15, 1960. W o r k supported by t h e Office o f Saline W a t e r .


ba-1960-0027.ch009

Recommend Documents