Ozone Water Purification Engineering Senior Project Final Written Report 5-11-07

May 2007 Engineering Senior Project

Jennifer Clay Stephanie Koplar

Composed by student project members: Mechanical Engineering Concentration Mechanical Engineering Concentration

Dr. Timothy Whitmoyer Dr. Donald Pratt Ray Diener Ariela Vader

Natural Sciences Central America Team

Messiah College Messiah College

Advised By: Engineering Department Messiah College Engineering Department Messiah College President of Elizabethtown Crystal Pure Water Biology Department Messiah College In Cooperation with Water for the World Water for the World

The Collaboratory The Collaboratory

Table of Contents 1. Introduction…………………………………………………………………………...………3 1.1 Abstract 1.2 Water Problems in Honduras 1.3 Meeting the Problem 1.4 Literature Review 1.5 Solution 2

Original Design…………………………………………………………………………..…9 2.1 Specifications 2.2 Design Components 2.3 System Testing 2.4 Documentation

3

Implementation……………………………………………………………………………12 3.1 Construction 3.2 Testing and Redesign 3.3 Operation

4

Schedule…………………………………………………………………………………...20 4.1 Gantt Chart

5

Budget……………………………………………………………………………………...21

6

Conclusions………………………………………………………………………………..22 6.1 Objective Comparison 6.2 General Comparison

7

Future Work………………………………………………………………………………..23 7.1 General Advice 7.2 Further Research 7.3 Further Testing and Documentation 7.4 Trip Tasks 7.5 Future Design

References……………………………………………………………………………………..25 Appendices

2

1. Introduction 1.1 Abstract This water purification system uses ozone as the method of disinfection, and was designed as a demonstration model. The system is designed to overcome water quality problems (mainly bacteria) that a team sponsored by The Collaboratory for Strategic Partnership and Applied Research encountered in rural areas of Honduras (WFTW Honduras 2006 trip report). In conjunction with producing clean drinking water, this system also addresses investment costs, maintenance costs, and ease of use, among other factors. The project team members are Jennifer Clay and Stephanie Koplar, mechanical engineering seniors at Messiah College. The project faculty advisors were Dr. Timothy Whitmoyer and Dr. Donald Pratt. Ariela Vader, Messiah College Faculty and Ray Diener, president of Elizabethtown Crystal Pure Water also advised this project. This project is done in cooperation with the Natural Science and Central America teams of Water for the World within The Collaboratory for Strategic Partnership and Applied Research. 1.2 Water Problems in Honduras Many communities in Honduras do not have access to clean drinking water. They experience numerous problems and sicknesses associated with unsanitary water supplies including illnesses such as Hepatitis. Many water-related problems such as diarrhea are so common that Honduran people treat them as part of daily life, and would not consider these ailments sicknesses. Testing of their water showed that their water contains bacteria, which could be the cause of most, if not all of their water-related illnesses. Another issue that the Honduran people face is dirt in their water. They recognize that water containing dirt is unhealthy and are forced to spend large amounts of money buying bottled water instead of using the municipal water source. A third item regarding the Honduran water that needs to be addressed is its taste. Many of the community members choose to get their drinking water from a local tad pole infested “spring” because they like the taste of that water better than the municipal water. 1.3 Meeting the Need Purpose In response to the problem of contaminated water in Honduras, we decided to design a small prototype water treatment system. The system would be capable of handling the water quality problems encountered by The Collaboratory team in rural regions of Honduras, and would be used as a demonstration tool to help educate the Honduran people about water purification and expose them to water treatment using ozone. This project was to be an initial step in designing an ozone water purification system that could eventually be implemented in a rural community in Honduras.

3

Benefit of Clean Water Drinking unclean water causes countless sicknesses and deaths throughout the world. This is a very significant problem for communities that do not have access to clean water sources. Drinking clean water allows people to avoid harmful water bourn ailments. Consuming water free of bacteria and other harmful contaminants reduces sickness and allows people to live healthy lives. Some of the benefits of health include an increase in productivity, as well as an increase in the quality of life. Choosing between alternatives Because the need for clean water is universal, there are several water purification methods available on the market today. The most prominent ones include chlorine, ultraviolet light, reverse osmosis, and ozone. We have chosen to use ozone for several reasons. Ozone based systems are less technically complicated and less expensive than reverse osmosis systems, which makes it more applicable for the people in Honduras. Unlike chlorine which has a considerable contact time to oxidize and destroy microorganisms, ozone oxidation begins immediately after contact with the microorganism’s cell membrane wall (Cruver 33). Biozone’s website stated that disinfecting with 1 ppm chlorine at a water temperature of 59°F and a pH value of 7 will require a retention time of 75 minutes. The disinfection efficiency achieved will be 99.9 percent. Ozonating the same water sample with ozone and achieving a disinfection efficiency of 99.9 percent using the same temperature and pH and a concentration of 1 mg/l ozone will require a retention time of only 57 seconds (www.biozone.com). Another benefit is ozone’s short half life which allows the unstable gas to convert back to oxygen in about 30 minutes which makes it safer than many disinfectants which stay in the water. Ozone also has intrinsic properties that can reduce color and odor of the water. Although initially the ozone may produce a smell, because of its short half life, the ozone will quickly dissipate unlike chorine which has a much longer half life. When ozone reverts back to O2, it improves the freshness of the water. Ozone also flocculates, or amasses various kinds of sediment and oxidizes metals, allowing them to be easily removed by filtration. Ozone has the ability to penetrate and break down the cell walls of bacterium, and deactivate harmful contaminates. Because of its ability to purify the water from a wide range of bacterium, ozone eliminates the need for chemicals that may be harmful to one’s body. Ozone can also provide residual disinfection aiding in the sanitation of water containers. 1.4 Literature Review Currently, there are several ozone based purification systems. Four systems are listed below; the first two are dealing with the same volume that we may experience, and the last two are large scale items, but similar in theory. The purpose for our model is for demonstration purposes only, and therefore must be compact and portable. We also need our system to be relatively inexpensive. 1. The closest existing state of the art system that we found is made by Ozone Solutions and can be found on their website: . The H2O Mini

4

Injection System weighs about 18 pounds and has a flow rate of approximately 1.5gpm. The water first passes through a mesh filter and then using a venturi, ozone is injected into the water. The cost of this system is $2,525 including an ozone generator and an air dryer. A picture of the model and description of its components can be seen below:

Photo 1 - H2O Mini Injection System by Ozone Solutions

2. Another existing portable model using ozone is made by Vortex Water Technologies. A description and picture of their system is as follows: The self-cleaning Vortex® Water Machine produces clean, clear water from almost any source without using chemicals. The purification system goes beyond simple filtration, using a patented five-way process that catalyzes ozone by UV light, destroying contaminants while infusing the water with fresh oxygen. The self-contained system may be installed above or below counters, or wall mounted with an optional kit to install a dedicated faucet. Self-regulated flow rate is up to 0.5 gpm, resulting in low maintenance cost of under $0.07 per gallon. The unit is 5.9 x 17.5 (150 x 445 mm) and weighs five pounds (2.2 kg) when empty. The slim design makes the machine a stylish addition to any home.

Photo 2 - Vortex® Water Machine by Vortex Water Technologies

5

3. Another existing state-of-the-art system is made by WaterChef, Inc. The cost of this system is $49,900. Although this model deals with higher volume applications producing approximately 15,000 gallons per day, the process is very similar to our system. The schematic for this system can be seen in our appendix. 4. Am also makes a larger scale application which can be found on the internet site :< http://www.watertanks.com/article/1760/>. This system uses a different injection method. Instead of using a venturi to pull air across a UV light by creating a vacuum, this model uses an air pump to push air past the UV light.

Photo 3 - Water system by Pure Water

See Appendices for a comparison of our objectives with these state-of-the-art systems. 1. 5 Solution Many of the water purification systems we discovered through our research were larger systems requiring large tanks. We have decided to design a small point of use system using ozone to disinfect the water. Although the system is similar to the first one in our lit review it is significantly cheaper and uses the capabilities of both ozone and ultraviolet light to disinfect the water. We choose to design a small system without cumbersome tanks so that it can be transported and used to demonstrate the theory of using ozone to disinfect water to the people of Honduras. Objectives The following are the objectives we wrote for our system, to make sure that it was operating properly, and would provide a solution to the problem we were addressing.

•

To build a prototype water treatment system to reduce the level of Fecal Coliform in the water to show positive for coliform in less than 5.0% of the samples This is the EPA water standard with regard to bacterial contamination. We wanted to meet the EPA standard to ensure the water was properly disinfected.

•

To build a prototype water treatment system to reduce the level of turbidity to less than .5 NTU (nephelolometric turbidity units) 6

Readings of turbidity must be less than .5 NTU to meet the EPA standard for turbidity. We wanted to meet this standard to ensure the water is safe to drink.

•

To build a prototype water treatment system that will produce 1 gallon of pure water in a 30 minute period. Originally we had an objective of 2-5 gallons per minute flow rate. After assessing the purpose and requirements of our water purification system we changed our flow rate objective to be in terms of quantity of water, realizing that the quantity of water we could produce was more important for a demonstration system like the one we intended to build.

•

To create operational and maintenance manuals for the system that have a Flesch-Kincaid grade level of less than 9.0 according to Microsoft Word We wanted to produce documentation to accompany our system that would be reasonable for the education level we might experience. Ninth grade is generally when the Honduras people begin to study a specific field. We wanted to produce documentation that a person with a general education level in Honduras could understand.

•

To build a prototype water purification system weighing less than 50lbs. The system needed to be broken down and taken on the plane with us to Honduras. We set the weight limit at 50 pounds to ensure it would meet the airline weight restrictions.

•

To build a prototype water purification system that costs less than $300. We wanted to keep the cost of the system as low as possible to ensure it would be practical for the Honduras people, recognizing Honduras is a developing country and the people do not have the same financial resources that we are accustomed to.

Project Description A basic overview of our final system design and its components is given below. The water purification system uses ozone and UV light as the methods of disinfection. The purification system consists of the following stages. 1. 2. 3. 4. 5. 6. 7.

20 x 20 mesh screen 20 micron particle filter 5 micron particle filter UV Light Carbon filter Ozone injection 1 hour wait period

7

Figure 1 - Schematic of final system design

Water Pump A water pump is needed to create the pressure to pump the water through the system. Particle Filtration A 20x20 mesh screen works in conjunction with 20 and 5 micro-pore sized filters to remove particles from the water stream. UV Light The water passes directly across an UV light which changes the DNA of the bacteria and keeps them from reproducing. This ensures bacterial decontamination. Carbon Filtration A standard block carbon filter is used to improve the taste and odor of the water. Ozone Generation Ozone is generated by the passage of air across an ultra violet (UV) light which converts the oxygen (O2) in the air into ozone (O3). The ozone will provide residual disinfection and improve the taste of the water by adding oxygen to it. Ozone Injection The method of ozone injection in to the water stream is venturi injection. By a change in pipe diameter, there is a pressure drop in the venturi which creates a vacuum causing air to pass across the UV light and be injected into the water stream.

8

Ozone Contact Time Requirements A certain amount of time is required for the dissolved ozone to interact with the contaminants in the water and thus disinfect the water. This time is known as the contact time. According to a Ken Mow at Ozotech Inc., an approximation of the ozone concentration contact time required is given by the equation: C*t=1.6 where C is the concentration of ozone in milligrams per liter of water, and t is the contact time in minutes. Since ozone is injected as the final stage of the system, the consumer will have to wait a given amount of time before using the water for it to be completely disinfected. According to our experimentation, the contact time required is at least 20 minutes (see “Testing and Redesign” in the Implementation section). However, the user should wait an entire hour for the ozone to dissipate and revert back into oxygen. Waiting the additional time will ensure that no ozone will be tasted in the water. The dissipation of ozone follows an exponential curve with a half life of 20 minutes. Power Supply Power is needed for the UV light within the ozone generator as well as an electric water pump. The system components will run off of standard AC power (standard US and Honduran wall outlets).

2. Original Design 2.1 Specifications The following are the design specifications we came up with to make sure the system does what we need it to. • System will fit within a volume of 10 cubic feet • System will weigh less than 75 pounds • System will use Commercial off-the-shelf (COTS) parts found in the US. • System will deactivate 1 fecal coliform per 1ml of water • System will reduce Turbidity to less than 5 NTU (nephelolometric turbidity units) • System will have a flow rate of 1-5 gallons per minute • System will cost less than $400 • UV light should be replaced once a year • The pump should be replaced every 5 years • The tank should be replaced every 10 years • The venturi should be replaced every 5 years • The filters should be replaced when the back pressure doubles indicating clogged filters • System should be checked once a month to ensure that the water is being purified • There will be clear instructions to check the indicator light on the generator to prevent water from running through the system without ozone being injected.

9

2.2 Basic Design Our original design is as follows: 1. Screen 2. Ozone injected into the water 3. Water enters contact tank 4. 20 Micron particle filtration 5. 5 Micron particle filtration 6. Carbon filtration

Figure 2 - Schematic of original system design

2.3 Design Components The following are descriptions of the original components we purchased for our system. Generator For many reasons explained in our Engineering Design Report (EDR) we decided to use an ultraviolet light to generate the ozone in our application. The original generator we purchased was the Purezone UV5X. It is specified to produce 125mg/hr, and has dimensions of 12”x4”x2”. The UV light bulb has an expected lifetime of 10,000 hours. The cost of this unit is $125. The air inlet consists of 3 small holes. The outlet is attached to an ozone resistant tube with an inner diameter or ¼ inch. Injector We chose a venturi injector from Mazzei Injector Corporation. The model we chose was the 384x, made of ozone resistant Kynar with ½ water inlet and outlet, ¼ inch ozone inlet. See EDR for justification and appendices for performance tables.

10

Filtration The components we used for filtration are listed below. See the EDR for further explanation.

• • • •

Y strainer, 20x20 stainless steel mesh 10” 20 micron polypropylene filter 10” 5 micron polypropylene filter 10” Carbon Block filter

Pump We bought Zoeller Impeller pump from a local plumbing store called R.F. Fager Company. The specifications are as follows: • Model 311 • 115 volts single phase / 60 Hz • Transfers up to 337 gallons per hour • Lifts water from 15 feet below pump level • Garden hose connections • Includes an extra impeller and gasket • 6’ hose with strainer and 6’ power cord • 90 day limited warrantee 2.3 System Testing In our original plan we intended to perform several tests to gain a better understanding of the system, change design flaws, and prove our systems ability to disinfect water. Short descriptions of the tests we originally planned on performing are given below; see the Engineering Design Report for more information on these tests. Fecal Coliform Testing Fecal Coliform plates were used to quantify the disinfection capabilities of the system. The water was tested before and after purification to show the systems affect on waterborne bacteria. Turbidity Testing Turbidity tests were performed using a spectrophotometer on water samples before and after they are treated by our system. These tests quantify the cloudiness of the water and thus, the capability of the particle filters. Contact Time Testing Testing was performed to determine how long it takes for the ozone to disinfect the water, eliminating the fecal coliform. These tests were performed using Fecal Coliform plates after the water has been in contact with the ozone for certain amounts of time.

11

3. Implementation Our final design changed quite a bit from the original design above. This section describes some of the process that we went through in order to produce the final design. The final design setup can be seen in the appendices. 3.1Construction Constructing the prototype is simply a matter of connecting the pre-manufactured components in the proper order. Refer to the appendices for pictures of the components in this description. A picture of our final setup can also be found in the appendices. The following shows the steps of how to connect our final design. Connect the following: 1. Y screen to inlet of pump 2. 20 micron filter with tubing to outlet of pump* 3. 20 micron filter with tubing to 5 micron filter* 4. 5 micron filter to UV light 5. UV light to Carbon filter* 6. Carbon filter to venturi inlet 7. L to venturi outlet** 8. Male to male fitting to L** 9. Second L to male to male fitting** 10. Second male to male fitting to second L** 11. Union with screen to second male to male fitting** 12. 4 foot long tube to union for water outlet 13. Ozone generator to gas inlet of venturi *note follow filter housing directions labeled on lid **note venturi, L’s, male fittings, and union comprise the S-configuration shown in apendices -Fasten metal tubing clamp tightly around tube every time tubing is connected to a barbed fitting 3.2 Testing and Redesign After testing our system, we realized that many problems had to be overcome before we would have a working model. This section describes some of the problems we encountered, and what steps we took to overcome these problems in order to meet our objectives. Our first fecal coliform tests seem to imply that our system without the contact tank was purifying the water. Along with a desire to keep the system simple and to have an “on demand” system, these results made us believe that a contact tank was unnecessary for our system. However, these results were never repeatable which may have been a result of bad plates, or just a random occurrence.

12

Testing 3/9/2007 5:15pm Results 3/10/2007 6:55pm Orientation 384x, Purezone, 20u, 5u, carbon FC/mL % sample Pre Post % Reduction 100 120 0 100 100 235 2 99.15 Table 1- Fecal Coliform results using the original orientation. These results were unrepeatable.

After doing more fecal coliform tests with the original configuration (minus the contact tank), we recognized that our system was unable to consistently disinfect the water. Our objective was to reduce the fecal coliform 100%. We were getting maximum FC reduction results between 81-85%. See the charts below for more detail: Testing 3/11/2007 8:15pm Results 3/12/2007 10:15pm Orientation: 384x, Purezone, 20u, 5u, carbon FC/mL % sample Pre Post % Reduction 100 16 6 81.25 Table 2 - First unacceptable Fecal Coliform results using original orientation Testing 3/26/2007 7-9pm Results 3/27/2007 9:15pm Orientation: 384x, Purezone, 20u, 5u, carbon FC/mL % sample Pre Post %Reduction 100 20 3 85.00 Table 3 - Second unacceptable Fecal Coliform results using original orientation

After considering the low reduction rates, we observed the ozone smell of our system and postulated that there was not enough ozone being injected into the water to handle the bacteria in the water (which had a relatively low initial count of fecal coliform). We reasoned that the three filters in series caused too much back pressure at the venturi outlet which decreased the suction rate and ultimately decreased the amount of ozone injected. We performed several tests rearranging the configuration of the system, trying to determine by smell when we were creating more ozone. We experimented with orientations that took out one or more filters or reoriented the tubing around the venturi to see if any of that made a difference. The smell of ozone significantly increased when we only had two filters in the system instead of three. (See logbooks for more process details). These tests were done simply to understand the effect of back pressure on the ozone injection rate and were not necessarily intended for implementation. In conjunction with this knowledge of back pressure effects, we decided to move the 20 micron filter before the venturi, leaving the other two filters after the venturi in an attempt to decrease 13

the back pressure. We did notice a slight increase in the intensity of the ozone smell. Because we presumed that a lack of ozone in our water was the source of our disinfection problem, we decided to buy tests that would quantify the amount of ozone in our water stream. We purchased two types of tests, one uses a visual comparison method (Ozone Solutions Vacuviales – see appendices) and the other set of tests uses reagents which are compared in a colorimeter and then output a given result (Hach reagents – see appendices). We were able to borrow a colorimeter from the Chemistry Department to read the given ozone concentration. Initially, we only had the Vacuviales available for use and our original results using these tests showed no dissolved ozone in the water stream. We then tested for ozone concentrations with just the pump and ozone injection and although we were able to smell the ozone, we were still unable to get readings for dissolved ozone. We contacted Ozone Solutions, our vacuvial manufacturer, and asked one of their representatives why we were unable to get dissolved ozone results even though we smelled it. We were told that UV based ozone generators are not nearly as efficient as corona ozone generators, and this may be the reason we were not getting enough ozone. We were also told that a concentration of only .02 ppm would yield smell and that smelling ozone does not necessarily mean that the ozone has been dissolved into the water. There is a difference between ozone in gaseous form and ozone that is dissolved in the water stream. Although we could detect the smell of ozone, this measurement was misleading because an increase in the ozone smell does not directly correlate to an increase in the amount of ozone dissolved. This led us to rethink our components and system configuration. Based on our design calculations, the amount of ozone generation our generator was capable of (125mg of O3/hr) should have yielded about .55 ppm, assuming 100% of the ozone is dissolved into the 1 gpm water stream. For this reason we contacted the ozone generator manufacturer, Purezone, to find out at what suction rate this specification was for so that we could rearrange our system and compare suction rates, but they were unable to tell us. In order to characterize an approximate suction rate, we began to take pressure measurements, comparing them with the performance tables for the venturi to analyze our corresponding suction rates. After doing some pressure testing, we found that we were generating approximately a 61% pressure drop across the venturi with an inlet pressure of around 10 psi. After speaking with Mazzei, our venturi manufacturer, we were told that at these low operating pressures that we would need a pressure difference greater than 75% to generate a decent suction rate. According to the pressure tables provided by Mazzei, (see table for 384-x model of venturi in appendices) our pressure difference would only generate approximately .4 SCFH using a high pressure of 10 psi and a low pressure of 5 psi. These are approximate values of the average pressures seen below. After talking to Mazzei we thought this may be a little low, but we were unsure of how this compared to the correct suction rate for a maximum ozone generation since none was provided by Purezone.

14

Pressure Testing 4/5/2007 Orientation: 20u/384x/Purezone/5u/carbon filter

% Pressure Pressure (Psi) HI LOW Drop 10.25 3.90 61.95 9.60 3.80 60.42 9.25 3.70 60.00 AVG 9.70 3.80 60.82 Table 4 – Percent pressure drop across 384x venture proved to be insufficient.

Our sponsor provided us with another venturi and ozone generator to experiment with because we believed our generator and/or venturi was the source of low ozone concentrations and thus our inability to disinfect the water. Although unable to test positive for ozone using our original components, we did test positive using the new generator and the new venturi (see Table 5). The new ozone generator used a UV light as well and was manufactured by Sterilight. Ozone Tests 4/5/2007 Dissolved Configuration Repitions O3 (ppm) Purezone generator, 384x, 5u filter, check valve 4 0 Purezone generator, 384x, no filters, no check valve 3 0 Sterilight generator, 484, no filters, no check valve .3-.4 2 Table 5 - Dissolved ozone results proving that new components yielded higher ozone concentrations

The new generator has a specification of producing 100mg O3/hr at a 5 SCFH suction rate. Our new venturi was manufactured by the same company as our previous one, Mazzei, but was the 484 model. We later found out through a Mazzei representative that the 484 venturi model operated better under lower input pressures than the 384x and could be the reason for increased dissolved ozone levels. Once we tested positive for dissolved ozone without the filters, we wanted to test the system with all the components using the new ozone generator and the new venturi. First, we first did pressure testing to see how the pressure drops compared. The pressure drop seemed very low with all three filters so we decided to take out the carbon filter before testing for dissolved ozone. However, even after removing the carbon filter, no ozone was dissolved in our water. After removing all the filters we were able to produce .1-.2 ppm of ozone. Note that this result is slightly lower than the above result which could be explained by the fact that these results were taken after the water was recycled. We were also able to generate .1-.2 ppm of ozone when we included all the filters, but had the venturi last in the system in order to eliminate any back pressure. According to the 484 pressure tables with air

15

suction, an inlet pressure of 4.3psi and an outlet pressure of .05 psi should yield an approximate suction rate of about 5.5 SCFH. See the appendices for more detail. Dissolved HI Pressure (Psi) LOW Pressure (Psi) % Pressure ORIENTATION Initial Final Intitial Final Ozone Drop 20u, 484 venturi, Sterilight ozone, 5u, carbon filter 8.20 n/a 6.20 n/a 24.39 n/a 20u, 484 venturi, Sterilight ozone, 5u, carbon filter 8.60 8.00 6.20 5.90 27.91 n/a 20u, 484 venturi, Sterilight ozone, 5u 8.30 7.50 2.20 2.20 73.49 0.00 20u, 484 venturi, Sterilight ozone, 5u 8.50 n/a 2.35 n/a 72.35 n/a 98.82 .1-.2 8.50 7.50 0.10 0.20 484 venturi, Sterilight ozone 98.84 .1-.2 4.30 n/a 0.05 n/a 20u, 5u, carbon filter, 484 venturi, Sterilight ozone Table 6 - Optimal orientation comparison using dissolved ozone and % pressure drop results

After verifying that ozone was dissolved into the water when the venturi was at the end of the system, we decided to do some fecal coliform testing with different contact times. (See appendices for process details). Because the highest amount of FC was eliminated at the 20min contact time, we concluded that the contact time for ozone should be at least 20min. However, further testing should be done because of the discrepancy within the results. These results showed that our system was still not fully eliminating the fecal coliform in the water. For this reason, we decided to add the UV light into our system to ensure complete disinfection. Note that this still does not take into account the system orientation with regard to mass transfer. Testing Results Orientation

4/11/2007 3:30pm 4/12/2007 3:45pm 20u, 5u, carbon, 484, Sterilight ozone generator FC/mL

% sample Pre Post % Reduction Contact Time 100 18 100 7 61.11 15min (w/carbon) 100 2 88.89 20min (w/carbon) 100 5 72.22 60 min (w/no carbon) Table 7 - Fecal Coliform results comparing various contact times

Several pressure tests and corresponding ozone concentration tests were taken to determine the optimal configuration. The pressure tests were done using a computer automated connection and the ozone tests were done with the colorimeter reagents. It was brought to our attention that the process by which the ozone was dissolved into the water could be improved. For further information on the theory behind increasing the mass transfer see the “Dissolved Ozone” section under Implementation Conclusions for further detail. Based on these mass transfer theories, we tried two different orientations that we thought would help maximize mass transfer. One was the S-configuration (see appendices) and the other was the U-configuration. Using tap water, we tested the dissolved ozone with the Vacuvials with each configuration and found that the S configuration yielded a higher amount

16

of ozone which confirmed previous results using the colorimeter reagents. See results below. 4/23/2007 Dissolved ozone (ppm) U-configuration .01-.05 S-configuration .08-.09 Table 8 - Dissolved ozone (ppm) results for "U" and "S" configurations

In our final design, we moved the venturi to the end of the system to eliminate back pressure thus maximizing our suction rate. We also added the S configuration after the ozone injection to assist in the mass transfer. These both proved very beneficial in reducing the bacteria. See the final testing results below. Testing 4/23/2007 1am Results 4/24/2007 12am Orientation: 20u, 5u, UV, carbon, 484, Sterilight ozone (including S orientation) FC/mL % sample Pre Post % Reduction 100 1640 15 99.09 Table 9 – First fecal coliform results using UV light Testing 4/25/2007 2:30pm Results 4/26/2007 1:40pm Orientation: 20u, 5u, UV, carbon, 484, Sterilight ozone (including S orientation) FC/mL % sample Pre Post % Reduction 100 1430 6 99.58 Table 10 - Second fecal coliform results using UV

Along with the preceding tests, we also tested the turbidity reduction and the flow rate. To do turbidity testing, we borrowed a spectrophotometer from our Biology Department. See results in the table below:

Turbidity Testing 3/26/2007 Turbitity (NTU) Pre Sample 40.3 Post Sample 1.1 Table 11 - Turbidity results

Although, we were able to significantly reduce turbidity, we did not meet our objective of turbidity being less than .5NTU. For this reason, we decided to buy a finer particle filter (.5 micron) to replace one of the old ones. The only problem was 17

that when we implemented this filter in the system, our pump was not able to generate enough pressure to push the water through it at a reasonable flow rate. Due to funding and time constraints we decided to leave the system as it was reasoning that the turbidity reduction was acceptable for our application, but should be reduced in the future to meet EPA standards. Flow rate was measured by two main methods. The first method involved recording how much time was needed to fill a certain container with water and then dividing by the amount of water in the container. The second method involved weighing the final sample and then using the specific gravity of water and the time to fill the container to yield a flow rate. We used the second method for our final results because it proved to be more accurate. Flow Rate Testing 4/23/2007 Orientation: 20/5/UV/carbon/Sterilight/484/S Weight Flow Rate (lb) (gpm) 63.67 13.7 1.55 Table 12 - Flow rate results

Time (s)

Although this was slightly lower than our original objective of 2gpm, we believe that this flow rate is sufficient for our application. Implementation Conclusions There are two main factors that affect the ability ozone has to disinfect contaminated water sources; the amount of ozone generated and the amount of ozone that gets dissolved into the water stream. Generating Ozone One factor with regard to the amount of ozone generated is the amount of ozone the generator is capable of producing. The other factor is the suction rate that the venturi is producing to pull the air across the generator and into the water stream. If the generator is not being operated at a high enough suction rate, it may not be producing the maximum of ozone possible. Other methods of producing ozone, such as corona discharge, may be capable of producing higher amounts of ozone under similar suction conditions. Our EDR explains why this method of ozone generation was rejected. The two generators that we used and tested were similar UV based ozone generators. We chose to use the Sterilight generator because it had specifications for both the Ozone generation rate as well as the suction rate required to produce that generation rater. We do believe that the Purezone generator is capable of working with a water purification application such as this, but further testing would need to be performed to make sure it was generating sufficient amounts of ozone. The amount of suction that the injector can produce is dependant upon the geometry of the venturi and the pressures at its inlet and outlet. The venturi manufacturer provided us with specification tables which compare the inlet and

18

outlet pressures and the flow rates to estimate the amount of suction the injector can produce. It is our assumption that these tables were generated from experimental data collected by the manufacturer. See the appendices for performance tables for the 384x and 484 Mazzei venturis. From our testing and discussions with the venturi manufacturer, Mazzei, we came to some general conclusions about the various venturis. The 384x works better under high pressures and requires a large percent pressure drop from inlet to outlet to create the most suction. The 484 works better with low pressure differences and at lower operating pressures. This information explains why the 484 worked well at the end of our system after the particle filtration. According to this information a caution should be made that removing the filters from the system, operating it only with the pump, venturi and generator, may not be ideal conditions for ozone injection and should not be used estimate the amount of ozone the generator and injector are capable of producing. Furthermore, any change in the pump or venturi location within the system will change the pressure characteristics would require analysis to determine if the venturi is providing the same suction rate, or if a different model injector would be preferable. Dissolving Ozone In order for ozone to handle fecal coliform contamination, it must be dissolved into the water stream so that the ozone molecules come in direct contact with the bacteria. This happens through a process known as mass transfer or diffusion. Mass transfer is dependant directly upon the area of interaction between the ozone molecules and the water molecules; smaller ozone bubbles means a higher surface area to volume ratio and more interaction between the ozone and water. Increasing the turbulence of the water stream would also increase the amount of ozone that dissolves into the water stream by providing more direct interaction between the water and ozone molecules. From our experimentation adding the S configuration on the end of our system helped improve the amount of ozone that was dissolved. The S configuration consists of two L joints, a screen, and 4 feet of tubing. The L joints increase the turbulence of the water stream, while the screen helps break up the ozone bubbles. Some type of mixing device could also be purchased to help the ozone dissolve into the water better, such an idea was rejected because of the high cost of these devices. Our Results According to the 5 psi pressure drop we measured across the venturi, the tables show us that we should be producing a suction rate of about 5.5 standard cubic feet per hour (SCFH). With this suction rate our generator should be capable of its maximum ozone production rate of 100mg Ozone/hour. At a flow rate of 1.55gpm, our system would theoretically yield .284 ppm or mg/liter of ozone if all of the ozone were to dissolve into the water stream. Because of our system’s inability to provide 100% mass transfer of the ozone into the water, we were only able to measure dissolved ozone concentration levels of up to .09 ppm.

19

3.3 Operation The following table shows how bacteria decontamination in our system depends on the back pressure (relates to suction) and mass transfer. The first test shows how the original configuration had too much back pressure so not enough ozone was generated to purify the water. In the second test the ozone was injected at the end, so there was low back pressure but not enough mass transfer took place to purify the water. The third test, with our final system design including the S-configuration, had both low back pressure and high mass transfer and was able to purify the water by 99.6%. The fourth test shows the capability of ozone to purify the water by itself, without the aid of the UV light. Note, in the third and fourth tests the water had an extremely high initial FC count, approximately 10 times that of the Honduran source. Therefore, we are confident that our system will be able to completely reduce fecal coliform in their water sources. Back Pressure

Mass Transfer

High

High

Low

Low

Low

High

Low

High

Configuration

FC Count Pre Post

ozone, 20u, 5u, 15 carbon 20u, 5u, 20 carbon, ozone 20u, 5u, UV, 1430 carbon, ozone 20u, 5u, 1430 carbon, ozone

4

73.3%

5

75.0%

6

99.6%

16

98.9%

Table 13 - Fecal coliform reduction summary

4. Schedule 4.1 Gantt Chart See appendices

20

Reduction

5. Budget Project Expenditures

Components Purezone UV 5x Ozone Generator Mazzei 384x Venturi Sterilight Generator & 484 Venturi Sterilight UV light Zoeller 115V Mini Vac Pump Carbon Filter Carbon Filter Housing 20u Particle Filter (2) 20u Particle Filter Housing 5u Particle Filter 5u Filter Housing Screen Fittings, tubing, and Teflon tape Agar FC testing Vacuvial Test kit Colorimeter Reagents Casing Contact Tank check valve Total Table 14 Project Expenditures

Budgeted Cost $120.00 $100.00 $75.00 $20.00 $20.00 $10.00 $10.00 $5.00 $100.00 $60.00 $40.00 $560.00

Actual Cost w/ Tax & Shipping $129.97 $45.05 $0.00 $0.00 $70.35 $8.66 $0.00 $4.54 $32.08 $0.00 $32.08 $14.39 $73.94 $57.00 $89.00 $31.25 $10.52 $598.83

Final Design Estimated Costs Estimated Components Cost Sterilight Generator & 484 $225 Venturi $120 Sterilight UV Light $70 Zoeller 115V Mini Vac Pump $9 Carbon Filter $22 Carbon Filter Housing $5 20u Particle Filter (2) $22 20u Particle Filter Housing $5 5u Particle Filter $22 5u Filter Housing $14 Screen $21 Fittings, tubing, and Teflon tape $535 Total Table 15 Estimated cost of final design

21

Estimated Cost w/o gifts $129.97 $45.05 $225.00 $120.00 $70.35 $8.66 $30.00 $4.54 $32.08 $10.00 $32.08 $14.39 $73.94 $57.00 $89.00 $31.25 $973.31

Funding Source Eng. Dept. Eng. Dept. Ray Diener WFTW Eng. Dept. Eng. Dept. WFTW Eng. Dept. Eng. Dept. WFTW Eng. Dept. Eng. Dept. Eng. Dept. Eng. Dept. Stephanie Stephanie Eng. Dept.

6. Conclusions 6.1 Objective Comparison The following table compares our results to our objectives. Result Objective 1.5 gpm 2 gpm 99.60% 100% 1.1 NTU .5 NTU Turbidity 28 lbs 50 lbs Weight $535 $300 Cost Table 16 Comparison of results and objectives Flow Rate Bacteria

Flow Rate We were not able to reach the original desired flow rate of 2 gallons per minute, but we were able to meet the specification of 1-4 gpm. We did meet the objective of being able to purify one gallon of water in a half an hour period, which meets the need of the purpose of the system allowing water to be purified for demonstrations. Bacteria We were able to reduce the bacteria in the water by 99.8%. This does not quit meet the goal of completely eliminating the fecal coliform, however we are confident that it will adequately purify the water under normal conditions, instead of the extremely high levels of contamination that we put through the system to do this testing. Turbidity We were able to reduce the turbidity from 40.3 NTU to 1.1 NTU. This does not quite meet the .5 NTU objective. Because of problems related to bacteria decontamination and the time we needed to focus on them, we were not able to do extensive turbidity testing or purchase a finer filter that would have reduced the turbidity further. Weight We were able to meet the weight limit of 50 pounds. The combined weight of the system’s components is 28 pounds; this does not include the weight of the carrying case. Cost The cost of the system is significantly more than we had anticipated and we did not reach our objective of creating a system for under $300. Two things that led to the increased cost were the higher cost of the ozone generator and venturi, and the added expenditure of the UV light. Although the system is more expensive than we had expected we still believe it is a viable option for the people of Honduras. The benefits of clean water significantly out weigh the capital cost of such a system.

22

Operation and Maintenance Manuals We were unable to produce the operation and maintenance manuals that we had intended. Because of problems with the amount of ozone that was getting dissolved into the water stream and the ability of the system to handle fecal coliform contamination we did not have enough time to write the manuals. 6.2 General Comparison Although we did not meet all of our objectives we believe that our project was successful. We were able to gain a great deal knowledge with regard to using ozone to purify water. We identified the two major factors affecting the ozone concentration and therefore the systems ability to purify water. These factors were the pressure across the venturi (which affects suction rate) and the mass transfer of the ozone dissolving into the water stream. We were able to producing a working prototype that can be used in Honduras to demonstrate this water purification method.

7. Future Work 7.1 General Advice

• • • •

Don’t confuse smell of ozone with amount of dissolved ozone, Ozone smell can be misleading. Perform dissolved ozone tests frequently. Perform pressure testing to quantify the effect configuration changes are having on the venturi. Intuition regarding pressure and flow can be misleading. Pay close attention to information gained regarding fluid dynamics.

Make sure you have a sufficiently contaminated source to do testing on. Inconclusive results because of an uncontaminated source wastes time. 7.2 Further Research

• • •

•

Research affect pH levels have on using ozone for water purification Research temperature has on using ozone for water purification Investigate using both UV and ozone to purify the water. Will the ozone attack bacteria that have already been deactivated by the UV light? How will this affect the ozone’s ability to offer residual disinfection (i.e. Will the ozone be used up on the already deactivated bacteria?) Gain further understanding of bromide and factors related to its presence when using ozone for water purification

• Research amount of time needed for ozone to aid filtration. 7.3 Further Testing and Documentation • Test system with Purezone generator to see the amount of ozone it is capable of producing at a given suction rate • Perform more accurate turbidity tests • Perform further testing to characterize system and make sure results are repeatable • Write operation and maintenance manuals 23

7.4 Trip Tasks

• •

Determine if bromide is present in the water sources at potential implementation sites Demonstrate ozone system to local people. Gage their reaction to this type of technology

•

Determine specifications a system would need to meet 7.5 Future Design

• • • • •

Redesign to meet EPA standards for turbidity Redesign to remove UV light and purify water with ozone alone Redesign to improve mass transfer and raise dissolved ozone concentration levels Redesign to inject ozone before filtration, considering pressure constraints of venturi and mass transfer considerations Design for permanent installation

24

References http://aem.asm.org/cgi/reprint/26/3/39 http://www.coolantconsultants.com/ozone_technologies.htm Cruver, Dr. James E. Disinfection. Water Technology. P.32-38. 1989 Couch, Benjamin. Disinfection Byproducts. Pacific Ozone Technology Diener, Ray; water purification specialist. Elizabethtown Crystal Pure Water http://www.epa.gov/safewater/mcl.html#mcls Hach Company P.O. Box 389 Loveland, Colorado 80539-0389 Phone: 800-227-4224 Fax: 970-669-2932 http://www.hach.com/ Company that makes testing supplies including the colorimeter and reagents http://www.lenntech.com/ozone/ozone-generation.htm Mazzei Injector Corporation 500 Rooster Dr. Bakersfield, CA 93307 Phone (661) 363-6500 Fax (661) 363-7500 www.mazzei.net -venturi manufacturer and supplier (384x and 484) -willing to help with other questions as well (ask for Justin or Mike – they seem to be the most knowledgeable) www.mcmaster.com Melligan, Myrle. Triple O Systems; www.tripleo.com Meyer, John. Messiah College Engineering Department Lab Technician. Microbac laboratory Camp Hill, PA 763-0582 Lab that does bromide testing on an iron chromatograph (IC) for $30 per sample http://www.northcoastmarines.com/ozone.htm Ozone Solutions, Inc. 789 7th St NW Sioux Center, IA 51250 USA Toll Free: (888) 892-0303 25

Ph: (712) 722-0337 Fax: (712) 722-1787 http://www.ozoneapplications.com/info/cd_vs_uv.htm www.ozonesolutions.com http://www.ozonemeters.com/K-7402_product.html for Vacuvials -Vacuvial Supplier (ozone testing) -Great resource for ozone information and materials -Willing to answer questions, very helpful (ask for Scott) Ozotech Inc.; www.ozotech.com Stephen Miller Ken Mow – vice president, very knowledgeable 1-800-796-9671 Agua Next http://www.o3ozone.com/ozone_generators_air_purifiers/ozone_generators_air_water_ purifiers/uv_ozone_generator/uv_pro_550_ozone_generator.htm www.pacificozone.com www.pacinst.org/reports/water_related_deaths/water_related_deaths_report.pdf# http://patft.uspto.gov/netacgi/nphParser?Sect1=PTO2&Sect2=HITOFF&p=1&u=/netahtm..., # 6,348,155 Stewart, Keith 931-243-6633 Small ozone water purification systems for around the world Specially designs corona discharge ozone generators and air dryers Sells portable system for medical teams, www.integrityglobal.com Upper Allen Sewage Treatment Plant 717-697-9548 Treatment plant down by the yellow breaches Source of contaminated water www.waterfiltersonline.com www.watercenterfilters.com www.wcponline.com www.wqa.org Water Quality Association Ozone Task Force. Ozone. United States, 2004.

26

Appendices

Manual References 384x Product Manual: http://www.mazzei.net/tables/384x.pdf 484 Product Manual: http://www.mazzei.net/tables/484.pdf Sterilight UV: http://www.r-can.com/product.php?cat=7&sub=4 Sterilight UV Specificiations: http://www.r-can.com/product.php?cat=7&sub=4 Sterilight Ozone: http://www.r-can.com/product.php?prod=255&sub=4 Vacuvial Catalog Page: http://www.ozonesupplies.com/K-7402.html Vacuvial Information: http://www.ozonesupplies.com/files/Vacuvials.pdf Vacuvial MSDS: http://www.ozonesupplies.com/files/K-7402MSDS.pdf

State of the Art Comparison

Flow Rate Bacteria Turbidity Weight Cost

Objective

Our System

Ozone Solutions

VortexPlus

WaterChef

American Tank Co.

2-5gpm 100% .5 NTU 50 lbs $300

1.5 gpm 99.60% 1.1 NTU 28 lbs $535

1-1.5gpm 99.9 n/a 35-40 gpm $2,525.00

.5 gpm n/a n/a 5.8 lbs. n/a

10-20,000gpm n/a n/a n/a $49,900

n/a n/a n/a n/a n/a

Table 1 - State of the art comparison

Purchasing Equpment

Components Purezone UV 5x Ozone Generator Mazzei 384x Venturi Sterilight Generator & 484 Venturi Sterilight UV light Zoeller 115V Mini Vac Pump Carbon Filter Carbon Filter Housing 20u Particle Filter (2) 20u Particle Filter Housing 5u Particle Filter 5u Filter Housing Screen Fittings, tubing, and Teflon tape Agar FC testing Vacuvial Test kit Colorimeter Reagents check valve Table 2 - Equpment suppliers

Purchased From

Funding Source

North Coast Pets Mazzei Donated by Ray Donated by WFTW R.F. Fager R.F. Fager WFTW R.F. Fager Filtech R.F. Fager Filtech R.F. Fager R.F. Fager VWR Ozone Solutions Hach R.F. Fager

Eng. Dept. Eng. Dept. Ray Diener WFTW Eng. Dept. Eng. Dept. WFTW Eng. Dept. Eng. Dept. WFTW WFTW Eng. Dept. Eng. Dept. Eng. Dept. Stephanie Stephanie Eng. Dept.

Complete Costs

Components Purezone UV 5x Ozone Generator Mazzei 384x Venturi Sterilight Generator and 484 Venturi Sterilight UV light Zoeller 115V Mini Vac Pump Carbon Filter Carbon Filter Housing 20u Particle Filter (2) 20u Particle Filter Housing 5u Particle Filter 5u Filter Housing Screen Fittings, tubing, and Teflon tape check valve Agar for FC testing Vacuvial Test kit Colorimeter Reagents Casing Contact Tank Total Table 3 - Complete project expenditures

Budgeted Cost ($) $120.00 $100.00 $75.00 $20.00 $20.00 $10.00 $10.00 $5.00 N/A $100.00 $60.00 $40.00 $560.00

Actual Cost ($) Before Tax unknown unknown GIFT GIFT $66.37 $6.38 GIFT $4.29 $22.15 GIFT $22.15 $13.58 $69.75 $9.92 $57.00 $74.00 $21.50 $367.09

Shipping Cost ($) $9.97 unknown N/A N/A N/A $1.79 N/A N/A $7.83 N/A $7.83 N/A N/A N/A unknown $15.00 $9.75 $52.17

Actual Cost w/ Tax & Shipping $129.97 $45.05 $0.00 $0.00 $70.35 $8.66 $0.00 $4.54 $32.08 $0.00 $32.08 $14.39 $73.94 $10.52 $57.00 $89.00 $31.25 $598.83

Estimated Cost w/o gifts $129.97 $45.05 $225.00 $120.00 $70.35 $8.66 $30.00 $4.54 $32.08 $10.00 $32.08 $14.39 $73.94 $10.52 $57.00 $89.00 $31.25 $983.83

Dissolving Ozone and Ozone Testing Different tests can be done for testing ozone in the air verses dissolved ozone in the water. We utilized two different methods for testing dissolved ozone. One was Vacuvials from a company called Ozone Solutions. These were small ampoules filled with vacuum pressure that filled with water when the tips were broken off. There are chemicals in the vials that make them change color in the presence of ozone. The color can then be visually compared to known standards and the level of ozone in the water estimated in terms of how many parts of ozone there are per million parts of water (ppm). Dissolved Ozone Reagent Kit 0.05-2.0 PPM K-7402 30 tests Color comparator $74.00

Photo 1 Ozone testing equipment, Vacuvial system

To perform more accurate dissolved ozone tests we used the Chemistry Department's portable colorimeter; it is a DR/850 Portable Colorimeter from Hach. We purchased the following reagents from Hach to perform the dissolved ozone tests with this colorimeter. High Range Ozone Reagents • Product # 2518025 • Price $21.50 • Ozone Reagent, HR, AccuVac® Ampoules • AccuVac® Ampoules contain the precise amount of reagent for a single test and can be used as a measurement curette Photo 2 Colorimeter, • Method: Indigo used for ozone testing • Range: up to 1.50 mg/L • Package contains 25 vials

Ozone Solutions, Inc. | 789 7th St. NW | Sioux Center, IA 51250 USA | Ph: (712) 722-0337 | Fax: (712) 722-1787

Dissolved Ozone Vacuvials Ozone detection for less than $75 K-7402 Dissolved ozone vacuvials are the quickest & least expensive way to measure your dissolved ozone levels. 30 colorimetric tubes for instant ozone detection.

Features low cost (less than $75) simple use easily distinguishable colors 30 tests in one kit accurate kit includes 30 vacuvials, color comparator charts, sample cup, indicator solutions & plastic carry Product ID: K-7402 In Stock: Yes Price: $74.00

Specifications Range (PPM): Minimum Detectible Limit: Method: Test Included with Kit:

0-0.6 & 0.6-2.0 0.025 PPM DDPD METHOD (Proprietary chemistry) 30

Visual Detection Method

Easily determine the dissolved ozone level visually with color comparator tubes. These tubes are included with every kit!

Ozone Solutions, Inc. | www.OzoneSupplies.com | [email protected]

Dissolved Ozone Vacu-vials Instantly and inexpensively determine your dissolved ozone levels without spending thousands of dollars on complex in line monitoring equipment. ®

How to use Vacu-vials for fast, simple, photometric analysis.

Step 1. SNAP. Fill the snap cup with your water sample, then snap the tip of a Vacu-vial ampoule as shown in the diagram above. Vacuum will cause it to fill instantly with sample fluid and mix with the pre-measured reagent inside.

Step 2. MIX. Mix the contents of the filled ampoule by tilting it up and down several times, each time allowing the small bubble inside to travel from end to end.

Step 3. READ. Compare the filled ampoule with the included color comparators and read the result.

K-7402 is the dissolved ozone visual color comparator kit. It includes everyting your need to test the dissolved ozone level. Range PPM

MDL PPM

Cat. No.

Kit * Price

Refill Cat.No.

Refill Price

0-2.0

0.025

K-7402

$74.00

R-7402

$30.80

* K-7402 includes solution, sample container, color comparator & 30 ampoules.

– Refill is 30 ampoules

…or you can use a SAM to read the ampoules.

Single Analyte Meters (SAMs) provide unprecedented economy, simplicity, and accuracy in dedicated photometers. Light Source: Light-emitting diode. Optical Path: 13 mm light path. Photodetector: Silicon photodiode. Power Source: One 9-volt alkaline battery (40 hours of use). Precision: 0.1 ppm

Signal Analyte Meter (SAM) $615.00

Page 1 of 1

CHEMetrics, Inc. 4295 Catlett Rd., Calverton, VA 20138 (800) 356-3072 (540) 788-9026 Fax (540) 788-4856 E-mail [email protected]

24 Hour Emergency Numbers: Creation Date: Revision Date:

(703) 590-9204 (540) 439-3860 04/24/90 (2243-8) 08/06/04

MATERIAL SAFETY DATA SHEET I. CHEMICAL IDENTIFICATION

VIII. EXPOSURE CONTROLS/PERSONAL PROTECTION

TRADE NAMES: OZONE CHEMets® and Vacu-vials®

OSHA PEL: 0.75ppm Formaldehyde TWA, 200ppm Methanol TWA ACGIH TLV: 0.3 ppm C Formaldehyde, 200ppm Methanol TWA PROTECTIVE EQUIPMENT: Safety glasses.

CATALOG NOS.: R-7402 and R-7403 DESCRIPTION: Reagent ampoules for the determination of ozone in water. Each CHEMet™ ampoule contains approximately 0.50 mL of liquid reagent sealed under vacuum. Each Vacu-vial™ ampoule contains approximately 2 mL of liquid reagent sealed under vacuum. NFPA RATINGS: HEALTH: 1 FLAMMABILITY: 0 REACTIVITY: 0

II. COMPOSITION/INFORMATION ON INGREDIENTS COMPONENT: Formaldehyde Solution, 37% CAS NO.: 50-00-0 PERCENT: < 0.2 COMPONENT: Potassium Phosphate Monobasic CAS NO.: 7778-77-0 PERCENT: 2.0 COMPONENT: Methanol CAS NO.: 67-56-1 PERCENT:

4.0

COMPONENT: Deionized Water CAS NO.: 7732-18-5 PERCENT: >93.0 COMPONENT: Other components CAS NO.: N/A PERCENT: < 1.0 Any component of this mixture not specifically listed (e.g. "other components") is not considered to present a carcinogen hazard.

III. HAZARDS IDENTIFICATION

IX. PHYSICAL AND CHEMICAL PROPERTIES STATE: Liquid APPEARANCE: Colorless ODOR: None SOLUBILITY IN WATER: Complete pH: 4 BOILING POINT: 95°C MELTING POINT: 0°C VAPOR PRESSURE: N/A SPECIFIC GRAVITY: 1 VAPOR DENSITY: N/A

X. STABILITY AND REACTIVITY HAZARDOUS DECOMPOSITION PRODUCTS: When heated to decomposition, formaldehyde fumes and oxides of carbon are emitted. Stable under normal conditions.

XI. TOXICOLOGICAL INFORMATION CARCINOGENIC STATUS: Formaldehyde: ACGIH: A2 Suspected carcinogen, IARC: Group 2A carcinogen. No other data available at this time.

XII. ECOLOGICAL INFORMATION Methanol, in high concentrations, is dangerous to aquatic life, and is expected to biodegrade rapidly. No other data available at this time.

XIII. DISPOSAL CONSIDERATIONS Dispose of in a manner consistent with Federal, State, and Local Regulations.

ACUTE TOXICITY: Irritation, cough, blindness, weakness, nausea, vomiting, convulsions. CHRONIC TOXICITY: Irritation, dermatitis, CNS effects causing headaches or impaired vision MEDICAL CONDITIONS AGGRAVATED BY EXPOSURE: skin or respiratory disorders

IV. FIRST AID MEASURES EYE AND SKIN CONTACT: Immediately flush eyes and skin with water for 15 minutes. INGESTION: Do not induce vomiting. If victim is conscious and alert, give 24 cupfuls of milk or water. Never give anything by mouth to an unconscious person. Seek medical attention. INHALATION: Remove individual to fresh air. If not breathing give artificial respiration (do not use mouth to mouth resuscitation), if breathing is difficult, administer oxygen and seek medical attention.

V. FIRE FIGHTING MEASURES FLASH POINT: N/A AUTOIGNITION POINT: N/A FLAMMABILITY LIMITS: UPPER: N/A LOWER: N/A EXTINGUISHING MEDIA: Dry chemical or carbon dioxide

VI. ACCIDENTAL RELEASE MEASURES Take up with absorbent material. Place in small containers for disposal.

VII. HANDLING AND STORAGE Always wear eye protection when working with these ampoules. WARNING: Do not break the tip of the ampoule unless it is completely immersed in your sample. Breaking the tip in the air may cause the glass ampoule to shatter. If this product is used as directed, the user will not come in contact with or be exposed to any of its chemical components. Wash thoroughly after handling. Avoid contact with eyes. Fragile. Liquid in glass. Handle with care. Product should be stored in the dark and at room temperature; however, temperatures up to 120°F or even below freezing will not normally affect reagent performance.

XIV. TRANSPORT INFORMATION Not regulated.

XV. REGULATORY INFORMATION EUROPEAN INFORMATION: EU Symbols: XN - HARMFUL Risk Phrases: Harmful by inhalation, in contact with skin and if swallowed. Harmful: danger of serious damage to health by prolonged exposure through inhalation, in contact with skin and if swallowed. Safety Phrases: In case of accident or if you feel unwell, seek medical advice immediately (show the label where possible). CANADA INFORMATION: WHMIS Classification: D2A All chemical components of this product are listed on Canada’s DSL list. U.S. INFORMATION: This product contains methanol and formaldehyde which are subject to SARA Section 313 reporting requirements. All chemical components of this product are listed on the TSCA Inventory.

XVI. OTHER INFORMATION THE ABOVE INFORMATION IS BELIEVED TO BE ACCURATE AND REPRESENTS THE BEST INFORMATION CURRENTLY AVAILABLE TO US. ALL PRODUCTS ARE OFFERED IN ACCORDANCE WITH THE MANUFACTURER'S CURRENT PRODUCTION SPECIFICATIONS AND ARE INTENDED SOLELY FOR USE IN ANALYTICAL TESTING. THE MANUFACTURER SHALL IN NO EVENT BE LIABLE FOR ANY INJURY, LOSS OR DAMAGE RESULTING FROM THE HANDLING, USE OR MISUSE OF THESE PRODUCTS. CHEMets® and Vacu-vials® are registered trademarks of CHEMetrics, Inc.

Turbidity Testing Turbidity testing was done on a spectrophotometer from the Biology Department. This measures the amount of light that can pass through a water sample. The instrument must first be calibrated using standard solutions of known turbidity levels. The machine uses a linear curve to measure the turbidity levels depending on the amount of light that passes through the sample. After calibration the user must clean the sample Photo 1 Turbidity testing equipment jar, place and cover it in the device, press ‘READ’, and note the given output of turbidity. The spectrophotometer gives readings in nephelolometric turbidity units (NTU).

Photo 2 Spectrophotometer

Pressure Testing Procedure 1: Differential Pressure Gage We used an automated gage to measure differential pressures of less than 10psi. This gage was connected to a computer and set up by John Meyer. 1. Move the cart and computer a significant distance away (about 8 ft.) from water purification system to prevent any water from touching the computer. 2. Using ¼ inch tubing, connect the high pressure end of the gage (marked ‘HIGH’) to the location which will experience the maximum pressure and then connect the low pressure end of the gage to the location that will experience the minimum amount of pressure. 3. When both ends are connected and stable, run the system. Loosen the screw for the HIGH end of the gage to allow water to drip out. Continue to let the water drip until the entire tube is full of water from the measurement location to the gage. Once the tube is full, tighten the screw. Do the same procedure for the LOW end of the gage. For the low end of the gage, there are two smaller diameter screws to loosen. Once both tubes are full of water and the screws are tightened, data recording can begin. 4. To login, follow the instructions on the side of the computer provided by John Meyer. 5. Once you are logged in, click on the “Virtual Logger” icon. 6. Then go to “Edit” and then “Settings.” Once in the Settings window, set the channel to 2. To correlate the voltage output of the transducer to pressure set the minimum and maximum voltage to -1V and 5V to 5psi and 10psi, respectfully. These correlations can be derived from the formula: y=2.5*x -2.5 where y is the pressure and x is the voltage recognizing that the transduces has a 1-5 voltage range output which corresponds to a 0-10 psi pressure difference. In order to make sure that the data will be recorded in a new file, you must click enable new logger. 7. Once all the settings are established, click ‘START’ to run the test and ‘STOP’ to stop the test. Each time you do a new test, make sure to rename the file, otherwise the new test will copy over the old test. Procedure 2: Atmospheric gage pressure Originally we used large pressure gages to measure the gage pressure at the inlet and outlet of the venturi. These gages were not linked. There wass no automation, only a visual comparison. These gages are extremely large and heavy and should be lifted with care. These gages are for pressures less than 30psi. 1. Connect the ¼ inch tubing from the end of the gage (located in the back of the output dial) to whatever location in the water system you are interested in. 2. Run the system. Wait until the system is fairly study to record data. 3. When doing a differential type test, hook up both meters, and record the results.

Photo 3 – Picture of larger pressure gages

Fecal Coliform Testing Materials needed: -Bunsen burner -Strike -Tweezers - Sealed filter paper -Ethanol -Labeled FC plates made with Agar

-Autoclaved filter and clamp - Clean plastic flask with an outlet to connect to vacuum - Sterilized graduated cylinder -Distilled water

Procedure: 1. Connect the Bunsen burner to a gas outlet. 2. Connect the plastic flask to the vacuum. 3. Set up the autoclaved filter on top of the plastic flask. You do not need to clamp it down, yet. 4. Turn on the gas and use the striker to start the Bunsen burner. 5. Dip tweezers in ethanol and then place over the flame. Remove from the flame and let the flame die out. 6. Open a sealed filter paper being very careful not to touch it with your hands. Use the tweezers to separate the filter paper from the thin blue attachment. 7. Place the filter paper inside the filter and clamp shut. 8. Take the distilled water container and place the end over the flames to ensure decontamination. To moisten the filter paper, pour a small amount of distilled water in the filter, turn on the vacuum, and allow the water to run through the filter paper. 9. Typically, one would first take a control sample (distilled water) first. Fill the graduated cylinder to the 100mL mark with distilled water and pour into filter. Once all the water has gone through, close the vacuum, and unclamp the filter. 10. Dip tweezers in ethanol and then place over flame. Remove from the flame and let the flame die out. 11. Carefully use the tweezers to remove the filter paper from the filter. You may want to break the seal, by pressing on the rubber part that connects the filter to the plastic flask. Breaking the seal allows an easier removal of the filter paper so tearing is minimal. 12. Place the filter paper in the FC plate labeled for the control sample. 13. Repeat steps 5-12 for the various samples in your experiment except modify step 9 to fit your application based on the concentration of the sample you want to run through the filter paper. After the sample is completely poured into the filter, pour a very small amount of distilled water into the filter after the sample to ensure that any fecal coliform on the edges will go through the filter paper. 14. Insert the FC plates inverted (bottom up) into the 44.5 degree Celsius incubator for 24 hours. 15. After 24 hours examine the FC plates using a low power (10-15x magnification) dissection microscope if needed. Count the colonies and record the results (Fecal coliform colonies are blue, regardless of shade) *Note that the order one conducts their experiment is important. To prevent washing the filter superfluously, conduct your experiment so that samples are taken from cleanest to dirtiest so that no unwanted bacteria would get into the water stream.

Contact Time Testing Procedure 1. Run the water through the system (20u, 5u, UV, carbon, venturi) into a container. 2. Start timer when pump is turned off and the full amount of water is in the container, 3. Take sample for maximum contact time from the container. Note that this sample will never be run through the carbon filter, therefore the ozone is allowed to decompose on its own. 4. Reorient system to run purified water through the pump and carbon filter 5. At desired time increments run purified water samples through carbon filter. 6. Collect at least 200 ml of water sample in small container, seal and label with time. 7. Perform FC plate tests on various samples to determine the effect of time ozone is in contact with the water sample.

Flow Rate Testing Procedure 1 (by weight) 1. Weigh the container that will hold the water. Record the measurement, and zero out the scale. 2. After running the system at a steady flow rate, begin timing how long it takes to fill up the bucket. 3. When time is up, measure and record the weight of the bucket. 4. Use the measured weight and time recorded in conjunction with the specific gravity of water and conversion factors to yield a flow rate with common units. Procedure 2 (by volume)* 1. After system reaches steady state, start the timer as you begin to fill up a labeled beaker with water. 2. Use the measured volume and the recorded time in conjunction with conversion factors to yield a flow rate with common units. *Note, this procedure may be less accurate that taking flow rate measurements by weight due the uncertainty in the time and the small quantity of water that is measured.

Schematics

Figure 1 - Schematic of original system design

Figure 2 - Schematic of final system design

System Photo

Photo 4 - Ozone water purification system

System Components

Photo 5 - Pump and screen

Photo 6 - 20 micron filter and filter housing

Photo 7 - 5 micron filter and filter housing

Photo 8 - Carbon filter and filter housing

Photo 9 - Sterilight UV light

Photo 10 - Sterilight ozone generator

Photo 11 - Mazzei Venturi

Figure 3 - Schematic of venturi

Photo 12 - Venturi and S configuration. Flow proceeds from right to left. The components beginning at the right are the venturi, L, male to male fitting (copper), L, screen, tubing

How to Read an Injector Chart Injector charts can be confusing because of the multiple columns and unfamiliar terms. This is a quick lesson on how to interpret injector charts to determine which model will work for your ozone application.

Image of an injector: Water flows from left to right; ozone is introduced into the middle

Below is a chart for a very popular ozone injector. The injector is capable of injecting both liquids and gases. For ozone, we can completely ignore the 3rd and 4th columns because they apply to liquid suction only. The first column is the injector inlet pressure, which is the pressure provided from a pump. The 2nd column is the injector outlet pressure, which is the pressure exerted on the injector outlet from delivering the water where it needs to go. The next column called MOTIVE FLOW states the flowrate of water going through the injector. The last column called AIR SUCTION lists the amount of air, or ozone, that can be sucked into the water stream. As can be seen from the chart, as injector outlet pressure (2) increases, injector suction decreases (6). This is true even though the motive flow (5) stays relatively constant.

Example: A pump delivering 18 GPM @ 15 PSI can inject a maximum of 20 SCFH (10 lpm) of air if 7 PSI of back pressure exists. If more suction is needed, two options exist: Increase the size of the pump, or decrease injector outlet pressure by increasing the diameter of the pipe, reducing the number of elbows or lowering the height the delivered water. psig = pounds per square inch gauge gpm = gallons per minute scfh = standard cubic feet per hour

Fact: The terms “venturi” and “injector” are used synonymously in the ozone industry”

http://www.ozoneapplications.com/info/reading_injector_charts.htm accessed Mar 21st 2007

Material Compatibility with Ozone Ozone Compatible Materials Chart * Material ABS plastic Acetal (Delrin®)

Rating (Source: Cole Parmer) [Ozone Concentrations not specified]

B - Good C - Fair

Aluminum

B - Good

Brass

B - Good

Bronze

B - Good

Buna-N (Nitrile) Butyl

D - Severe Effect A - Excellent

Cast iron

C - Fair

Chemraz

A - Excellent

Copper

B - Good

CPVC

A - Excellent

Durachlor-51

A - Excellent

Durlon 9000

A - Excellent

EPDM EPR Epoxy

A - Excellent up to 100-deg F A - Excellent N/A

Ethylene-Propylene

A - Excellent

Flexelene

A-Excellent

Fluorosilicone

A - Excellent

Galvanized Steel

In Water (C - Fair), In Air (A - Excellent)

Glass

A - Excellent

Hastelloy-C®

A - Excellent

HDPE

A- Excellent

Hypalon®

A - Excellent

Hytrel®

C - Fair

Inconel

A - Excellent

Kalrez

A - Excellent up to 100-deg F

Kel-F® (PCTFE)

A - Excellent

LDPE

B - Good

Magnesium

D - Poor

Monel

C - Fair

Natural rubber

D - Severe Effect

Neoprene

C - Fair

NORYL®

N/A

Nylon

D - Severe Effect

PEEK

A - Excellent

Polyacrylate

B - Good

Polyamide (PA)

C-D (Not recommended)

Polycarbonate

A - Excellent

Polypropylene

C - Fair

Polysulfide Polyurethane, Millable PPS (Ryton®)

B - Good A - Excellent N/A

PTFE (Teflon®)

A - Excellent

PVC - Water

A - Excellent

PVC - Air

B - Good

PVDF (Kynar®)

A - Excellent

Santoprene

A - Excellent

Silicone

A - Excellent

Stainless steel - 304

B - Good/Excellent

Stainless steel - 316

A - Excellent

Steel (Mild, HSLA)

D - Poor

Teflon

A - Excellent

Titanium

A - Excellent

Tygon®

B - Good

Vamac

A - Excellent

Viton®

A - Excellent

Zinc

D - Poor

Note: These materials were tested at ozone levels exceeding 1,000 PPM. Ratings -- Chemical Effect

A.

Excellent. -- No effect

B. Good -- Minor Effect, slight corrosion or discoloration. C. Fair -- Moderate Effect, not recommended for continuous use. Softening, loss of strength, swelling may occur.

D. Sever Effect -- Not recommended for ANY use.

N/A = Information Not Available.

* Remember that different materials react differently to wet or dry ozone. DRY ozone has been dried to a -60 deg F or lower, WET ozone contains small amounts of moisture. Contact Ozone Solutions to determine if your material is compatible.

http://www.ozoneapplications.com/info/ozone_compatible_materials.htm Accessed Mar. 21st 2007

Ozone Conversions: Physical Properties, Standard conditions P = 1013.25 MB, T = 273.3 K Density of ozone , 2.14 kg/m3 Density of oxygen, 1.43 kg/m3 Density of air, 1.29 kg/m3 Density of water, 1000 kg/m3 USEFUL CONVERSION FACTORS (for water) 1000 liters = 1 m3 = 264 US gallons 1 gal = 3.785 liters = 3785 ml OZONE CONCENTRATION IN WATER 1 mg/l = 1 PPM O3 = 1 g O3/m3 water {By weight} OZONE CONCENTRATION IN AIR BY VOLUME 1 g O3 / m3 = 467 PPM O3 1 PPM O3 = 2.14 mg O3/m3 OZONE CONCENTRATION IN AIR BY WEIGHT 100 g O3 / m3 = 7.8% O3 1% O3 = 12.8 g O3/m3 OZONE CONCENTRATION IN OXYGEN BY WEIGHT 100 g O3/m3 = 6.99% O3 1% O3 = 14.3 g O3/m3

Convert gaseous O3 concentration from g/m3 to ppm by volume --- [PPM O3 = C · 467] (Example: 2.14 g/m3 at standard conditions = 1,000 ppm) Also: If we know concentration in g/m3 and flowrate in LPM, we can calculate output in g/hr Concentration(g/m3) X Flowrate(lpm) X 0.001(m3/liter) = O3 output (g/minute) (Example: 28.7 g/m3 at 2.9 lpm flowrate) 28.7 g/m3 X 2.9 lpm X (1 m3/1,000 liters) = 0.083 g/minute 0.083 g/minute x 60 minutes = 4.9 g/hr

http://www.ozoneapplications.com/info/ozone_conversions.htm Accessed Mar 21st 2007

Effect of Ozone on Bacteria

1 - Computer generated image of a bacteria cell 2 - Close-up of ozone molecule coming into contact with bacterial wall 3 - Ozone penetrating and creating hole in bacterial wall 4 - Close-up effect of ozone on cell wall 5 - Bacterial cell after a few ozone molecules come into contact 6 - Destruction of cell after ozone (cell lysing)

As a comparison based on 99.99% of bacterial concentration being killed and time taken: Ozone is 25 times of that of HOCl (Hypochlorous Acid) 2,500 times of that of OCl (Hypochlorite) 5,000 times of that of NH2Cl (Chloramine). Further more, ozone is at least 10 times stronger than chlorine as a disinfectant. Chlorine reacts with meat forming highly toxic and carcinogen compounds called THMs or tri-halomethanes - rendering meats lesser quality products. THMs was also implicated as carcinogens in developing kidney, bladder, and colon cancers. Chlorine also results in the production of chloroform, carbon tetrachloride, chloromethane besides THMs. On the other hand, ozone does not even leave any trace of residual product upon its oxidative reaction.

http://www.ozoneapplications.com/info/bacteria_destruction.htm accessed Mar 21, 2007

Ozone Properties: Property

Ozone

vs. Oxygen

Molecular Formula:

O3

O2

Molecular Weight:

48

32

Color:

light blue

colorless

Smell:

- clothes after being outside on clothesline - odorless - photocopy machines - smell after lightning storms

Solubility in Water (@ O0.64 deg C):

0.049

Density (g/l):

2.144

1.429

Electrochemical Potential, V:

2.07

1.23

Typical O3 half-life vs. Temperature Gaseous

Dissolved in Water (pH 7)

Temp (C)

half-life *

Temp (C)

half-life

-50

3-months

15

30-minutes

-35

18-days

20

20-minutes

-25

8-days

25

15-minutes

20

3-days

30

12-minutes

120

1.5-hours

35

8-minutes

250

1.5- seconds

* These values are based on thermal decomposition only.

No wall effects, humidity, organic loading or other catalytic effects are considered.

Ozone Solubility The solubility of ozone depends on the water temperature and the ozone concentration in the gas phase: Units in mg/l or ppm.

O3 GAS

5o C

10o C

15o C

20o C

1.5%

11.09

9.75

8.40

6.43

2%

14.79

13.00

11.19

8.57

3%

22.18

19.50

16.79

12.86

http://www.ozoneapplications.com/info/ozone_properties.htm

Ozone vs. Chlorine: ACTION IN WATER Oxidation Potential (Volts)-

CHLORINE

OZONE

1.36

2.07

Moderate Moderate

Excellent Excellent

No

Yes

Color Removal

Good

Excellent

Carcinogen Formation

Likely

Unlikely

Moderate

High

None

Moderate

Variable

Lowers

2-3 hours

20 min.

Operation Hazards: Skin Toxicity Inhalation Toxicity

High High

Moderate High

Complexity

Low

High

Capital Cost

Low

High

Monthly Use Cost

Moderate-High

Low

Air Pre-treatment

None

Filer and dehumidify air

Disinfection: Bacteria Viruses Environmentally Friendly

Organics Oxidation Micro flocculation pH Effect Water Half-Life

http://www.ozoneapplications.com/info/ozone_vs_chlorine.htm Accessed Mar 21st 2007

ID

Task Name

1

Research/scheduling about spectrometer

Mon 12/11/06

Start

Finish Fri 2/16/07

2

Investigate Pressure measurement opportunities

Mon 12/11/06

Fri 3/9/07

Write up Testing Procedures

Mon 12/11/06

Fri 5/4/07

Sep 3, '06 Sep 10, '06 Sep 17, '06 Sep 24, '06 Oct 1, '06 Oct 8, '06 Oct 15, '06 Oct 22, '06 Oct 29, '06 Nov 5, '06 Nov 12, '06 Nov 19, '06 Nov 26, '06 Dec 3, '06 D S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S S

3

4

5

Fecal Coliform

Mon 12/11/06

Fri 2/2/07

6

Contact Time

Mon 12/11/06

Fri 2/2/07

7

Turbitity

Mon 12/11/06

Fri 2/16/07

8

Screen

Mon 12/11/06

Fri 2/16/07

9

Bromide/Bromate

Mon 12/11/06

Fri 2/23/07

10

Flow Rates/Pressure

Mon 12/11/06

Wed 2/28/07

11

12

Pre-Carbon Outlet

Mon 2/12/07

Fri 3/16/07

13

Order Contact-Tank

Mon 2/12/07

Fri 3/16/07

Actual Testing

Mon 2/12/07

Fri 5/4/07

14

15

16

Fecal Coliform

Mon 2/12/07

Fri 3/2/07

17

Contact Time

Mon 2/12/07

Fri 3/2/07

18

Turbitity

Mon 2/26/07

Fri 3/16/07

19

Screen

Mon 2/26/07

Fri 3/16/07

20

Bromide/Bromate

Mon 3/12/07

Fri 3/30/07

21

Flow Rates/Pressure

Mon 3/26/07

Fri 4/13/07

23

Write Operational Manual

Mon 2/12/07

Fri 3/30/07

24

Write Maintance Manual

Mon 2/12/07

Fri 3/30/07

25

Final Presentation

Mon 4/16/07

Fri 4/27/07

26

Final Design Report

Fri 3/30/07

Fri 5/4/07

22

Project: Spring Gannt Chart Date: Fri 5/11/07

Task

Progress

Summary

External Tasks

Split

Milestone

Project Summary

External Milestone Page 1

Deadline

c 10, '06 Dec 17, '06 Dec 24, '06 Dec 31, '06 Jan 7, '07 Jan 14, '07 Jan 21, '07 Jan 28, '07 Feb 4, '07 Feb 11, '07 Feb 18, '07 Feb 25, '07 Mar 4, '07 Mar 11, '07 Mar 18, '07 Mar 25, '07 Apr 1, '07 Apr 8, '07 Apr 15, '07 Apr 22, '07 Apr 29, '07 MTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F

Project: Spring Gannt Chart Date: Fri 5/11/07

Task

Progress

Summary

External Tasks

Split

Milestone

Project Summary

External Milestone Page 2

Deadline