Summer 2009 CDC Research

Overview

Abstract

In some AguaClara plants, a surface foam develops at the end of rapid mix. The first experiment goal was to understand the chemical conditions required for this surface foam to develop at the end of rapid mix and the first baffle. The first test trials were conducted with a constant supply of clay and varying amounts of alum but these did not exhibit any form of surface foam formation. Subsequent trials included organic matter: humic acid, but these only produced large non persistent bubbles. It was not until a stronger surfactant, liquid soap, was added to the baffle spacing that a surface foam with strong persistent bubbles developed. From these experiments it was concluded that air entrainment along with a surfactant in the raw water are the main factors behind surface foam formation. The raw water contains decaying matter which acts as the surfactant while the waterfall at the LFOM creates the perfect condition for air entrainment and thus surface foam formation. With this in mind, the research is now focused on retrofitting AguaClara's designs so that no air entrainment occurs in the entrance tank and rapid mix chamber by eliminating waterfalls and or implementing hydraulic jumps.

Introduction and Objectives

For the summer of 2009, our team has two main goals:

• We will attempt to recreate the foam in a laboratory setting that has been forming in many of the AguaClara plants in Hoduras. Once that is completed, we will design a way to retrofit the current plants to fix the problem
• We will learn about the current design for a Nonlinear Chemical Dose Controller and then update MathCAD code for the controller and hopefully be able to build a fully functional prototype by the end of the summer.

These two goals are very important to the overall goals of AguaClara for a number of reasons. The foam that forms in the current AguaClara plants both increases the amount of work that plant operators have to spend to keep the water clean, and reduces the overall effectiveness of the plants. Although the foam cannot flow very far in the plant itself, it can be blown around by wind, and also small bubbles in the water could flow as far as the sedimentation tank, where they would come out, possibly causing the floating floc problem. We suspect that the surfactants and natural organic matter lower the surface tension energy requirements, creating these small bubbles that stay in solution long enough to reach the sedimentation tank.

As AguaClara continues to grow and serve larger and larger communities, we will be building plants with much bigger capacities, where the amount of foam would be impossible to clean. Currently, the plant operators simply scoop the foam off the surface of the water with buckets - not the most sanitary or effective solution. Therefore, it is imperative that we find a solution not only for the current plants in operation, but also for the future plants we design. This summer we learned how the foam is created, and now we are exploring what design changes we can make that will reduce or even eliminate foam formation.

Also, as we evolve to build larger plants, the Linear Chemical Dose Controller (LCDC) won't be able to provide a sufficient dose of chemicals to treat the larger flow rates. The CDC team from past semesters has found that the LCDC can only provide chemical flows up to 400mL/min, which is too low for larger plants. Thus, a Nonlinear Chemical Dose Controller (NCDC), which won't be linearly dependent on water height like the LCDC, is needed for bigger plants. This summer we will need to improve upon the initial design of a NCDC in order to make it more dependable and robust.

It is our hope that at the end of the summer, the plant operators in Honduras will have a quick-fix way to eliminate the foam forming in their plants, and that future AguaClara plants will have larger capacities, allowing them to provide clean, foam free drinking water for more and more people.

Summer 2009 CDC Research Team's goals and meeting minutes.

Experimental Methods

In an attempt to recreate surface foam in a lab setting, similar to that found in AguaClara Plants in Honduras, we first had to recreate a similar design in the lab. We did so by first having a tank which mixed tap water and clay in order to have a constant turbidity of approximately 100 NTU. This tank represented the incoming plant water. Water from this tank was then mixed with Natural Organic Matter (NOM) and sent into Tank 2 which simulated the rapid mix tank and first baffle of the flocculation tank. Once in this tank, the turbidity of the water was once again measured, and alum was dripped onto the surface of the water. The water in the rapid mix side of tank 2 was mixed with a stirrer before traveling into the first baffle portion of the tank. It was here that the water was able to settle a bit and a webcam took pictures of the surface water every two minutes to determine if surface foam was created. When an aerator was added, it was placed in the first baffle portion of tank 2. (You are missing an important component here and this is what conditions are required in rapid mix and how you created those conditions. There are two goals in rapid mix: one is large scale turbulent mixing and this requires a minor loss coefficient of one or greater and the second is smaller scale turbulent mixing requiring an energy dissipation rate between 0.5-1.0 W/kg. Hopefully, we can mix down to a scale where molecular diffusion would be effective. I would only include the hydraulic way we created rapid mix. Include a design equation for how this can be calculated for future reference)

Our experiment included the following parameters: Aluminum Sulfate (Alum) dosage, concentration of Natural Organic Matter (NOM), location of alum addition, aeration of water and addition of a surfactant. In the first week of our experiment, we tested for alum dosage in order to determine whether it caused foam formation. Alum dosages of 35 mg/L, 45 mg/L, 55 mg/L and 65 mg/L were added to the rapid mix chamber.

The following week the group used Humic Acid in order to determine the effects of NOM on surface foam formation. Varying the dosage at 1mg/L, 2mg/L, 5 mg/L and 10 mg/L, Humic acid was mixed in the rapid mix chamber with 45 mg/L of alum.

After deciding to add an aerator, we kept the parameters the same as in the NOM experiments, however we added an aerator to the first baffle portion of tank 2.

Results and Discussion

Our first experiment sought to determine whether or not alum was the sole factor in the formation of surface foam. After dosing the water with varying alum concentrations, it was determined through photographs and observation that no foam was formed. Sample photos are included in Figure 1.

Although the foam was not formed, this data was crucial to our research. Not only did it minimize the list of potential foam formation factors, it also help build our understanding of foam formation. As a result of foam not forming due to alum addition at the surface of the water, we also were able to rule out alum addition under the water surface as intuitively it was a solution to the form formation.

Our second experiment sought to determine whether or not the addition of Natural Organic Matter (NOM) contributed to the formation of surface foams. At all concentrations of Humic Acid, we again found there was no surface foam formed. Although the NOM did lower the surface tension better enabling bubbles to form, we believe they did not form due to a lack of air bubbles which naturally are found in AguaClara plants. As a result, we modified our experiment to include an aerator to provide them.

In our next experiment, we varied the concentration of NOM but included an aerator to provide bubbling directly into what would be the first baffle spacing. At concentrations below 2 mg/mL we found that no foam was formed. However at higher concentrations of Humic Acid, we found that large bubbles would rise and then quickly pop in the center of the tank. In the meantime, smaller bubbles would form around the edges of the tank and were slightly more persistent as seen in Figure 2.

Bubbles form when water molecules form bonds around air pockets. A surfactant is generally an organic molecule that has both hydrophobic and hydrophilic (what?). Due to this polarity, surfactants form micelles in water which helps to stabilize air bubbles and prevent them from aggregating. Humic acid is a weak surfactant, so though it did reduce the aggregation of air bubbles it still occurred. This resulted in large, non persistent air bubbles that formed a foam at the surface of the water.

Although, a surface foam was formed at concentrations of Humic Acid greater than 2 g/mL, it was not the foam that is found in AguaClara plants. The foam we created in lab contained large non-persisent bubbles, however the foam we sought to create contained small persistent bubbles. At this point we began searching for a stronger surfactant to further prevent the aggregation of bubbles, thus replicating the foam found in Honduran AguaClara plants.

In order to support our hypothesis, we added soap to our last experiment with Humic Acid simply to see if would create the surface foam we wanted. Indeed the soap caused a persistent foam, similar to that found in AguaClara plants and can be seen in Figure 3 below.