Transcript
(Data Recorder) (Leader)
(Inspector) Experiment #6, Reverse Osmosis
Group #5
Date Submitted: Wednesday, October 11, 2006
Ryerson University
Department of Chemical Engineering
CHE 415 Unit Operations II
Lab Report
Experiment #6 Reverse Osmosis
Experiment Performed on Thursday October 5, 2006
Report Submitted to
Dr.
By Group # 5 1. (Leader)
2. (Data Recorder)
3. (Inspector)
Date Submitted: Wednesday October 11, 2006
Marking Scheme
Formatting Answer to all 6 questions in each Report Section / 10
General Appearance; Grammar and Spelling / 5
Complete and Informative Tables and Graphs / 15
Contents Accuracy and Precision of Results / 20
Comparison with Literature Data / 10
Influence of Procedural Design on Results / 10
Logic of Arguments / 20
Sample Calculations / 10
_____
Total: / 100
Table of Contents
1.0 Introduction Page 2
2.0 Theoretical Background Page 3
3.0 Procedures Page 6
Figure 3.1: Reverse Osmosis System Page 7
4.0 Results and Discussion Page 8
5.0 Error Analysis Page 11
6.0 Conclusion Page 12
Appendix A : Metals Concentrations in Water Page 13
Appendix B: Results in Graphical Display Page 14
Appendix C: Sample Calculations Page 16
Appendix D: Raw Data Page 17
Reference Page 21
1.0 Introduction
Reverse osmosis is a process capable of producing high quality purified water through a semi-permeable membrane when a sufficient amount of pressure is applied. The semi-permeable membrane only allows the solvent, in this case water, to pass while retaining the solutes, in this case, minerals on the feed side. A high pressure is applied to the feed side so that the water may flow against its natural path, its osmotic flow, and try to balance the concentration on both sides of the membrane. As a result, water is forced to flow from a lower concentration of to a higher concentration and hence the name, reverse osmosis. Since the membrane is under such pressure it is very easy for the pores to be blocked by the solute trying to pass over to the product side and one way to reduce the blocking of pores by the solute is to utilize a cross flow in the feed stream. The Filtration Concepts Inc. Commercial Tap-Water Reverse Osmosis System Model # TW 1-1/2-120-60 used in this experiment makes use of the cross flow to extend the life of the semi-permeable membrane.
The primary objective of this experiment was to evaluate the performance of the reverse osmosis water purification system. The key comparison was the amount of clean water produced as a function of pressure, therefore a plot the product flow rate versus the pressure difference will be assessed as best as possible to determine the performance of the system. Furthermore, some other factors will be considered such as the effects of a recycle stream on the performance of the system, the total dissolved solids (TDS) in the product stream as well as the odor of the product water.
2.0 Theoretical Background
The osmotic pressure is the pressure exerted by the flow of water through a semi-permeable membrane separating two solutions with different concentrations of solute (cite) and the relationship for dilute solutions was shown by Van’t Hoff to be (Geankoplis, 2003):
(1)
Where, n = total moles of solute [kmole]
Vm = volume of pure solvent water [m3]
R = gas constant 8.313 kPa m3 / (kmole K)
T = temperature [K]
? = osmotic pressure [psi]
Only the osmotic pressure for a reverse osmosis system is affected by the solute concentrations in the feed side and since the feed solution was Toronto tap water the solute concentrations were found in Toronto’s 2004 water annual report (Appendix A). Assuming the water from downtown Toronto is supplied by the Ashbridges Bay water filtration plant, the metal concentrations was taken from an analysis of their effluent.
For diffusion type membranes like the one used in this experiment, the diffusion of solvent through the membrane can be calculated using the following equation (Geankoplis, 2003):
(2)
Where, Nw = solvent mass flux [kg/(s-m2)]
Pw = solvent membrane permeability [kg solvent/(s-m-atm)]
Lm = membrane thickness [m]
= P1 – P2 (hydrostatic pressure difference with 1 on feed side and 2 on product side) [atm]
= 1 - 2 (Osmotic pressure difference with 1 being feed side and 2 being product side) [atm]
Aw = the solvent permeability constant [kg solvent/ (s-m2-atm)]
Since Aw is constant, it can be seen in equation (2) that a plot of the solvent mass flux, Nw, versus the difference between hydrostatic pressure and osmotic pressure,) should give a linear graph with slope equal to the solvent permeability constant.
The solvent mass flux can easily be converted into the solvent mass flow rate by multiplying equation (2) by the area of the membrane giving the following equation:
(3)
Where, nw = solvent mass flow rate [kg/s]
Am = area of membrane [m2]
Thus, now if the solvent mass flow rate is plotted against the pressure difference between hydrostatic and osmotic pressure, the slope will be the solvent permeability constant multiplied by the area of the membrane. However, the permeability constants for membranes must be determined experimentally for the particular type of membrane used and therefore the value obtained in this experiment can be assumed to be the theoretical solvent permeability constant for this particular membrane.
In addition, when a recycle stream is added to the system it should theoretically increase the concentration of solutes in the feed stream and thus alter the osmotic pressure only. Therefore, osmotic pressure should increase proportionally with the solute concentration and as a result decrease the pressure difference between the hydrostatic pressure and itself. From equation (3) it can be seen that since is a constant for the specific membrane used, a decrease in the pressure difference should result in a decrease in the solvent mass flow rate.
Lastly, the performance of the reverse osmosis system can also be assessed based on the physical quality of the product such as the amount of the total dissolved solids (TDS) remaining or the odor of the product. Odor is a good general indicator of water quality as no one would drink water with an upsetting odor. TDS is the combined content of all inorganic and organic substances contained in the liquid that can pass through a two micrometer sieve, and a safe maximum level of 500 mg/L (500ppm) in drinking water is allowed by the United Stated Environmental Protection Agency (USEPA).
3.0 Procedures
Read and understood the reverse osmosis safe operations manual.
Before turning on the system the inlet valve was closed.
Then verified that a five micron filter cartridge was installed properly.
Now the inlet valve was opened before the system was turned on.
Ensured feed water was turn on and all pretreatment devices were in service mode.
Verified tubings were connected to appropriate tanks and drains.
Verified level controls were connected properly.
Opened high pressure valve all the way.
Opened recycle valve all the way.
Turned control switch to “On” position.
Monitored the high pressure gauge to ensure the pressure did not exceed 150 psi.
The recycle valve was set at desired percentage of recycle.
The high pressure valve was turned slowly to obtain various pressures and flow rates.
Before 150 psi was reached the high pressure valve was released.
The recycle percentage was changed and repeated from step 13.
After all trials were completed the system was shut off, feed water turned off, and product tank was drained.
Figure 3.1: Filtration Concepts Inc. Commercial Tap-Water Reverse Osmosis System Model # TW 1-1/2-120-60
4.0 Results and Discussion
The objective of this experiment was to predict the performance of the reverse osmosis system based on several factors. The first factor was how the total differential pressure, which is the difference between the hydrostatic pressure and the osmotic pressure, affected the flow rate of the product water. The systems was run with a 44%, 66%, and 88% recycle stream which was determined by first finding out the number of turns possible and then turning it equivalent to the percentage of recycle desired. It was discovered that there was a possible 9 turns for the recycle valve and thus 4, 6, and 8 turns were used to obtain a 44%, 66%, and 88% recycle stream respectively. All units were converted into S.I. units; temperature was taken as 298K for calculation purposes and the volumetric flow rate of the solvent was converted into a mass flow rate and plotted against the differential pressure. All of the graphs can be seen in Appendix B.
The first trial was done with a 44% recycle of reject into the feed, and when the mass flow rate of the solvent was plotted against the differential pressure it produced a noticeable linear relationship. When a trendline was added and its intercept set at 0, since when the hydrostatic pressure is equal to the osmotic pressure there should not be any flow across the membrane, it provided us we the equation y = 0.0125x. This equation was compared to equation (3) and it was seen that = 0.0125 kg/(s-atm). However, it was noticed that there was deviation in the data points with respect to the trendline.
The second and third trials were run under a 66% and 88% recycle stream and produced values of 0.0132 kg/(s-atm) and 0.0174 kg/(s-atm) respectively for the membrane permeability constant, , specific to this system. Because the membrane permeability constant,, is in fact a constant, the percentage recycled should not affect o change this value in any way. However, there is no way of acquiring a theoretical value and it must be obtained experimentally. Furthermore, as the percent recycle increased, there was a better fit for the data points leading us to believe that the results obtained from the third trial with 88% recycle was the most reliable result. Therefore, the experimental membrane permeability constant was taken to be 0.0174 kg/(s-atm) and when compared to the first and second trials produced errors of 28.2% and 24.1% respectively. Further interpretation of the graphs showed that the performance of the system was consistent through the whole range of pressures for the reason that the mass flow rate of the solvent increased proportionally with the pressure. If the performance of the system was not consistent through the range of pressures it will show on the graphs as a deviation from a linear relationship. Although the system was performing consistently it did not mean the system was producing water at its intended potential, but without the availability of theoretical values this cannot be assessed.
In addition to the amount of product that was being produced, the performance could also be based on the quality of the water that was being purified. The maximum amount of total dissolved solids (TDS) allowed in drinking water was set to be 500 mg/L or 500 ppm set by the USEPA. The system used in the laboratory had a TDS indicator for the product and was recorded to be steady at 6 ppm which is way below the maximum allowable limit. This shows that the performance of the reverse osmosis system was operating at a very high level and was very effective in removing most of the TDS from the product. Moreover, the product water did not have an appalling odor as odor is a good general indicator of water quality. Hence, the system was also very effective in removing the odor causing contaminants from the product.
Finally, it can be said that the reverse osmosis system used in this experiment was performing at a very high level with respect to the physical quality of the water and very consistently as well. On the other hand, the productivity of the system could not be assessed due to the lack of information available, but none the less this reverse osmosis system was considered to be performing superbly.
5.0 Error Analysis
There were a number of sources of error in regards to this experiment that was worth mentioning. The first would be the calculations for osmotic pressure; since it is directly related to the concentration of solutes in the feed stream there is no direct method available in the laboratory to make accurate measurements of such concentration, and furthermore, the values obtained from the Toronto Water Agency were only available for up to the year 2004.
Another error noticed during the duration of the experiment was the fluctuation in the pressure of the feed stream which greatly affected the high and low pressure gauges. It was noticed that when the feed pressure dipped so did the other two pressure readings, and thus producing inaccurate data points. The results of this experiment may improve if the feed pressure can be kept constant for the entire duration of the experiment.
Moreover, the pressure gauges and flow meters were not a good manner for obtaining accurate data as the graduations were large in comparison to the changes that was occurring. Recommendations would be to install digital pressure gauges as well as digital flow meters for easy access to accurate numbers.
6.0 Conclusion
The most reliable experimental value for the membrane permeability constant for this particular membrane was found to be 0.0174 kg/(s-atm) at a recycle of 88%, which meant that at a recycle of 44% there was a 28.2% error while at 66% recycle there was a 24.1% error involved. In spite of this, the performance of the system was considered to be consistent through the range of pressures that was utilized based on the linear relationship that can be seen in the graphs. Also, the performance of the reverse osmosis system was assessed based on some physical qualities of the product like TDS and odor. It was determined the amount of TDS in the product was steady at 6 ppm while producing water that was odor free. This meant that the system was operating at a high level of performance with respect to TDS as well as odor. Thus, the reverse osmosis system used in this experiment was determined to be generally performing efficiently.
Appendix A
Metal Concentrations in Toronto Water from Ashbridges Bay (mg/L or ppm)
Metal mg/L (ppm) MW (kg/kmol) Concentration (kmol/m3)
Arsenic 0.0015 74.92 2.00214E-08
Cadmium 0.001 112.41 8.89601E-09
Chromium 0.0069 52 1.32692E-07
Copper 0.018 63.55 2.83242E-07
Iron 0.61 55.85 1.09221E-05
Mercury 0.00013 200.59 6.48088E-10
Nickel 0.005 58.69 8.51934E-08
Lead 0.013 207.2 6.27413E-08
Zinc 0.046 65.39 7.03471E-07
Total 0.70153
1.2219E-05
Appendix B: Results in Graphical Display
Appendix C: Sample Calculations
Converting Product Flow Rate into Mass Flow Rate for first data point:
Converting GPM to LPM using first product flow rate data:
Converting psig to atm using first high pressure reading data:
Delta P from first data reading:
Calculating Osmotic Pressure on the feed side, ?:
Calculating Osmotic Pressure on the product side:
(Since it is safe to assume that the solute concentration in the product side is negligible, n=0, the osmotic pressure will also be zero)
Delta P –Delta ? for the first data point:
Membrane Permeability Constant, , was produced using Microsoft Excel’s built in linear regression package for a graph.??-??
Appendix D: Raw Data
4/9 turn
product flow rate Reject flow rate high pressure reading low pressure reading TDS(ppm)
(GPM) (GPM) (PSIG) (PSIG)
1.1 4.1 80 14 6
1.2 3.75 100 16 6
1.3 3.05 117.5 20 6
1.4 2.9 120 20.5 6
1.4 2.5 126 22.5 6
1.4 2.4 130 23 6
1.5 2 135 24 6
1.5 1.75 140 25 6
1.52 1.55 145 25.5 6
6/9 turn
1.1 3.75 91 20 6
1.2 3.4 100 22 6
1.25 3 110 24 6
1.35 2.6 115 25.8 6
1.45 2.2 120 22.5 6
1.5 1.6 130 25 6
1.5 1.25 135 26.5 6
1.5 0.85 140 27 6
1.55 0.7 145 29.2 6
8/9 turn
0.85 3.45 65 21 6
1 2.8 75 25 6
1.05 2.5 80 27.2 6
1.1 2.2 85 29 6
1.15 1.9 90 30.5 6
1.2 1.5 95 32.5 6
1.2 1.25 100 34.5 6
1.25 1.95 105 36 6
1.3 1.5 110 38 6
1.4 0 118 40.5 6
Table 1: Experimental Data
Product Flow Rate Reject Flow Rate High Pressure Reading (P1) Low Pressure Reading (P2)
4/9 Turns
(LPM) (LPM) (atm) (atm)
4.1639 15.5201 6.4422 1.9524
4.5425 14.1953 7.8027 2.0884
4.9210 11.5455 8.9932 2.3605
5.2996 10.9777 9.1633 2.3946
5.2996 9.4635 9.5714 2.5306
5.2996 9.0850 9.8435 2.5646
5.6781 7.5708 10.1837 2.6327
5.6781 6.6245 10.5238 2.7007
5.7538 5.8674 10.8639 2.7347
6/9 Turns
4.1639 14.1953 7.1905 2.3605
4.5425 12.8704 7.8027 2.4966
4.7318 11.3562 8.4830 2.6327
5.1103 9.8420 8.8231 2.7551
5.4888 8.3279 9.1633 2.5306
5.6781 6.0566 9.8435 2.7007
5.6781 4.7318 10.1837 2.8027
5.6781 3.2176 10.5238 2.8367
5.8674 2.6498 10.8639 2.9864
8/9 Turns
3.2176 13.0596 5.4218 2.4286
3.7854 10.5991 6.1020 2.7007
3.9747 9.4635 6.4422 2.8503
4.1639 8.3279 6.7823 2.9728
4.3532 7.1923 7.1224 3.0748
4.5425 5.6781 7.4626 3.2109
4.5425 4.7318 7.8027 3.3469
4.7318 7.3815 8.1429 3.4490
4.9210 5.6781 8.4830 3.5850
5.2996 0.0000 9.0272 3.7551
Table 2: Experimental Data in S.I. Units
Delta P = P1-P2 Solvent Mass Flow Rate Delta P - ?
4/9 Turns
(atm) (kg/s) (atm)
4.4898 0.069399 4.4895
5.7143 0.075708 5.7140
6.6327 0.082017 6.6324
6.7687 0.088326 6.7684
7.0408 0.088326 7.0405
7.2789 0.088326 7.2786
7.5510 0.094635 7.5507
7.8231 0.094635 7.8228
8.1293 0.0958968 8.1290
6/9 Turns
4.8299 0.069399 4.8296
5.3061 0.075708 5.3058
5.8503 0.0788625 5.8500
6.0680 0.0851715 6.0677
6.6327 0.0914805 6.6324
7.1429 0.094635 7.1426
7.3810 0.094635 7.3807
7.6871 0.094635 7.6868
7.8776 0.0977895 7.8773
8/9 Turns
2.9932 0.0536265 2.9929
3.4014 0.06309 3.4011
3.5918 0.0662445 3.5915
3.8095 0.069399 3.8092
4.0476 0.0725535 4.0473
4.2517 0.075708 4.2514
4.4558 0.075708 4.4555
4.6939 0.0788625 4.6936
4.8980 0.082017 4.8977
5.2721 0.088326 5.2718
Table 3: Data Points for Graph in S.I. Units
Osmotic Pressure Feed Side ? Osmotic Pressure Product Side ?
(atm) (atm)
2.99E-04 0
Table 4: Osmotic Pressures in S.I. Units
Reference
Geankoplis, C. (2003) Transport Processes and Separation Processes Principles. United States. Prentice Hall.
Toronto Water Agency. 2004 Annual Report. City of Toronto, Canada, 2004. Retrieved Monday October 9, 2006 from:
United States Environmental Protection Agency (USEPA). Ground Water & Drinking Water. United States, Updated February 28, 2006. Retrieved Monday October 9, 2006 from:
1
