THE EFFECTS OF INORGANIC COMPONENTS OF WATER ON TASTE
Last year, we were told, Miami's drinking water was contaminated. Although the water problem has apparently been resolved, we were nonetheless interested in the subject of water testing, especially after we heard complaints about the taste of Western drinking water. Certain locations, according to some students, apparently had superior taste (or, inferior in the case of Peabody). Our curiosity resulted in the following lab.
The purpose of this research project is to compare physical and chemical characteristics of water to drinking preferences of students. We hypothesize that students on western campus can detect, by taste, the differences in chemical characteristics between the water of various Western locations. We expect that the results from the taste survey will coincide with the laboratory analyses, proving our hypothesis.
In this lab, we hope to establish a connection between the survey results and laboratory tests. Since the prior problems of the drinking supply have been corrected, the drinking water can be assumed to be EPA standards. Yet opinions of the Western students indicate that the drinking water at various Western locations tastes different, suggesting some variation exists. Are the samples different to the human sense of taste? If so, what characteristics cause this variation? We hope that this lab will answer these questions.
>>>>RELEVANCE OF RESEARCH QUESTION<<<<
Prior Western Research on Water Quality:
The onlY similar Natural Systems Project to ours was one that dealt with cadmium in water from 1997. While it was focused more on a view of water purity with regards to cadmium only. They observed water sources for drinking, and natural bodies of water, but they did use very similar methods of testing. Our project looks at purely drinking water, and its purity with respect to a number of factors.
Our research will hopefully elaborate on the methods used in the 1997 project. In our design, the larger subject of water taste is taken into consideration. We realize, however, that human taste is a very complicated scientific topic that cannot be fully accounted for in our project. Perhaps the class will learn something about their sense of taste, if our project succeeds.
>>>>MATERIALS AND METHODS<<<<
2. Water from McKee, Mary Lyon, Peabody, Alexander, and distilled water
1. The samples were gathered from their respective locations: the drinking fountains at Alexander Hall, Boyd Hall, Mary Lyon Hall, McKee Hall, and Peabody Hall. From each location, two samples were taken--one for the purpose of drinking, and one for the purpose of testing. Each sample was given a number (1-5) to represent its origin for later reference (this should disassociate the sample with it's origin, aiding in the generation of unbiased results). Also, the samples were gathered at the same general time and used promptly (on the same day preferably) to ensure that the taste of the samples is optimized.
2. Each member of the class was then given one small cup of the each water sample.
3. The class individually tasted each of their samples and rated them according to taste in the following manner: one sample should be determined as the best, and one sample should be determined as the worst.
4. The results of the taste tests were then recorded in a table on a provided sheet of paper. The best sample received a 1, the worst received a -1, and the rest received zeros. All of this data was gathered for later interpretation.
5. The class then divided themselves into groups of three.
6. Each group received, once again, one of each of the five samples, with numbers 1-5. The groups then performed one chemical test on their five samples (the tests were assigned, and were one of the following: pH, total hardness, iron, copper, manganese, and sulfide.
7. The samples were performed according to the procedures in the appendix of this packet.
8. The data from all the test were recorded on sheets that were provided, and turned in for analysis.
It is unbiased because they don't know what water they're drinking from when they rate the best and worst water taste. The groups also won't know what water sample their testing so they couldn't have a biased opinion of the water.
Test for pH
1. Range Finding Indicators Solution
2. Test Tube
1. First fill one test tube (0230) to the 5.0mL line with the sampled water.
2. Then, proceed to add ten drops of the *Range Finding pH Indicator Solution (2220)
3. This will then produce a color. This color will then indicate the approximate pH value of the sampled water.
The chart is provided below
pH 3.0 Red
pH 4.0 Red-orange
pH 5.0 Orange
pH 6.0 Yellow
pH 7.0 Yellow-green
pH 8.0 Green
pH 9.0 Blue-Green
pH 10.0 Blue
pH 11.0 Purple
Test for Iron
1. *Iron Reagent #1
2. *Iron Reagent #2
3. Spoon, 0.05g
4. Octect Comparator, Iron
5. Test Tube, 5.0mL
*Warning: Reagents marked with a * are considered hazardous substances.
1. First, fill a test tube (0230) to the 5mL line with the sampled water.
2. Add 5 drops of *Iron Reagent #1 (4450). Then cap and mix.
3. Use the 0.05g spoon (0696) to add one level measure of *Iron Reagent #2 Powder (4451). Then mix until the powder completely dissolves. Wait 3 minutes.
4. If Iron is present in the sampled water, a pink color will develop. Then insert the test tube into the Iron Comparator (7474). Match the sample color to the color standard.
5. Depending upon the amount of iron in the water a pink to deep red color will be present. A trace of iron is apparent if the water that turns a shade of pink. However, a high concentration will produce a dark red color
Test for Copper
1. *Copper 1 Reagent
2. Standard Color Reagent
3. Test tubes, A & B
*Warning: Reagents marked with a * are considered hazardous substances.
1. First fill test tube "A" with the sampled water.
2. Then proceed to drop dive drops of *Copper Reagent (6446) and mix. If the water turns yellow then copper is present in our water sample.
3. Next, proceed to fill the second test tube (0804) to the lower line "B" with distilled water.
4. Then add the Standard Color Reagent (6613) to test tube "B", one drop at a time, counting the drops and mixing after each addition. Then proceed to hold the two tubes about one-half inch above a plain white surface and look down through the tubes to compare colors. Continue adding color reagent to the second tube until the colors match the reaction in the first tube.
5. The test is then calculated as:
Copper (ppm) = 0.025 x No. Drops Color Reagent
1. Hardness Reagent #6 Tablet
2. *Hardness Reagent #5
3. *Sodium Hydroxide Reagent with Metal Inhibitor
4. Calcium Hardness Indicator Tablets
5. Hardness Dr Titration Tube
6. Direct Reading Titrator, 0-200
*Warning: Reagents marked with a * are considered hazardous substances.
1. First fill the Hardness Dr Titration Tube (0769) to the mark with our sampled water.
2. Then proceed to add 5 drops of *Hardness Reagent #5 (4483) and mix. Add one hardness reagent #6 tablet (4484) and stir till the tablet is disintegrated. A red color then should develop.
3. Next, proceed to fill the direct reading titration (0382) with the hardness titration reagent #7 (4487).
4. Then add the hardness titration reagent (4483) one-drop at a time, stirring to mix after each drop, until the red color changes to clear blue.
5. The result is read then directly from the direct reading titrator. Record the Hardness in ppm Calcium Carbonate.
Test for Sulfide
1. *Sulfide Reagent A
2. Sulfide Reagent B
3. Sulfide Reagent C
4. Test Tube, 5.0mL
6. Octet Comparator, Sulfide, 0.2 and 2.0 ppm
1. First fill the test tube (0230) to the 5mL line with the sampled water.
2. Next proceed to add 15 drops of *Sulfide Reagent A (4458). Then cap and mix. Do not forget that the test sample now has a high acid content.
3. Then add 3 drops of Sulfide Reagent B (0354). Next, cap and mix. Then wait one minute.
4. Use the 1.0mL pipette (0354) to add 1.0mL of Sulfide Reagent C (4460). Then cap and mix.
5. Next, if sulfide is present, a blue color will develop. Then insert the test tube into the Sulfide Comparator (7477). Finally, match the sample's watercolor to the standard color.
1. 2, 25mL mixing bottle
2. Ascorbic Acid Reagent Powder Pillow
3. Rochelle Salt Solution
4. Alkaline-Cyanide Reagent
5. P.A.N Indicator Solution
6. Color Comparator
7. Sample Tube
8. 1mL plastic dropper
1. Fill a clean 25mL mixing bottle to the 25mL mark with demineralized water. This is the reagent blank.
2. Fill the second 25mL mixing bottle to the 25mL mark with the sample. This is the prepared sample.
3. Add the contents of one Ascorbic Acid Reagent Powder Pillow to each bottle. Swirl to dissolve.
4. Add 1.0mL of Alkaline-Cyanide Reagent to each bottle. Swirl to mix. (A turbidity may form, but will dissipate after step 5.
5. Using the 1mL calibrated plastic dropper, dispense 1.0mL of P.A.N. Indicator Solution, 0.1%, to each bottle. Swirl to mix. An orange color will develop if manganese is present.
6. Allow the color to develop for a minimum period of 2 minutes
7. Pour at least 5mL of the prepared solutions into two clean viewing tubes.
8. Insert the tube of prepared sample into the right top opening of the color comparator.
9. Match the color of the comparator to the sample and read the mg/L manganese through the scale window.
In the taste survey, our results for McKee was 0, Mary Lyon was 1, Alexander was 0, the distilled water was -6, and Peabody was 5. These results indicate that the class'opinion ranked Peabody's water as the best tasting, and the distilled water as the worst tasting.
If the water contains more acid than a pH of 5.0 and more alkaline than a pH 8.5-9.0, this test then should be then looked at more carefully for possible industrial contamination.
The average pH that was found in McKee was 7.06, Mary Lyon's pH was 7.2, Alexander was 7.0, distilled was 7.34, and Peabody's water was 7.08. The pH scale is from 1-14. The neutral pH is 7.0 and the lower the pH the more acidic, the higher the pH the more basic. The pH for Alexander was the best while the distilled water was a little more basic than the rest of the locations. We believe that there was an unidentifiable error in the testing of the distilled water because we expected the pH to be 7.0.
Iron is found to be present in water supplies that have been contaminated by acid mine wastes.
The maximum contaminant level for iron is .3 ppm.
For McKee, it had +/- .2 ppm of iron, while Mary Lyon had +/- .5 ppm. Alexander had +/- .3 ppm, distilled water had +/- .4 ppm, and Peabody had +/- .3 ppm.
A normal level of copper, found in any drinking supply, would be 0.03 ppm. The pure taste of water can taste bitter at low levels such as 1.0 ppm. Basically, the more copper in a water supply the more bitter it will be. The maximum contaminant level of water is 1.0 ppm.
The copper content in McKee was 0 ppm, Mary Lyon had .05 ppm, Alexander had .075 ppm, the distilled water had 0 ppm, and Peabody had .35 ppm. The highest level of copper was found highest in Peabody's water. We believe that this high amount of copper was caused by the copper water fountain.
Total hardness is taking into account because it represents the total amount of calcium and magnesium ions. However, these ions are expressed as calcium carbonate. It is important to note that even in natural water there is a high amount of calcium and magnesium ions. The normal level of hardness is 25 grains/gallon.
McKee and Mary Lyon had 26 grains/gallon of hardness, Alexander and Peabody had 25 grains/gallon, and the distilled water had only 1 grain/gallon. They had similar amounts of total hardness except for distilled water as we expected.
Sulfide found in water supplies of excessive amounts is a toxic substance that acts as a respiratory depressant in both humans and fish. Also a small about of sulfide such as a few hundredths of a milligram per liter can cause a noticeable odor in our water.
Alexander, Mary Lyon, and Peabody all had <.2 ppm of sulfide, while McKee had <.15 ppm, and distilled water had 0 ppm.
Manganese can produce strong acidity causing damage to your internal system if ingested in large amounts. The maximum level of manganese is .05 mg/L.
In McKee, Mary Lyon, and Alexander contained .15 mg/L of manganese in their waters. Distilled had 0 mg/L while Peabody had .1 mg/L.
We believe that the test kit may have been old, therefore, making our results inaccurate. The reason we think that is because our control, the distilled water, had the highest level of manganese that could be in drinking water and it shouldn't have any manganese at all.
[Test] (1) McKee (2) Mary Lyon (3) Alexander (4) Distilled Water (5) Peabody
Sulfide (ppm) <0.15 <0.2 <0.2 0 <0.2
pH 7.5 7.4 7.4 7.7 7.2
6.9 7.1 6.6 7.3 7.1
6.9 7 6.6 7.2 7.1
6.9 7.1 7.1 7.4 7.4
7.1 7.4 7.3 7.1 6.6
Copper (ppm) 0 0.05 0.075 0 0.35
Hardness(grains/gal) 26 26 25 1 25
Iron (ppm) +/- .2 +/- .5 +/- .3 +/- .4 +/- .3
Manganese (mg/L) 0.15 0.15 0.15 0 0.1
Survey 0 1 0 -6 5
When our procedures were completed, we were confronted with the problem of extracting a meaningful conclusion from our data. At the beginning of the project, we expected that this conclusion would arrive in the form of some correlation between our survey results and any number of chemical data. In simple terms, we wanted to see our chemical tests jive with the survey.
The relationship we wanted to see between our data sets was a relationship that we assumed would be fairly obvious. After completing our procedures, however, we were surprised and, perhaps, a little disappointed-our raw data showed no obvious correlation to our survey. We were not without hope, though.
Long before we reached that point, we had established that we would use statistics to confirm our conclusions. Our choice in doing was planned to be the t-test, but, as we had learned, this test works only in the comparison of fairly similar data sets-but which data sets were similar?
There was no clear answer to this question. Faced with this fact, we made our statistical confirmation into a statistical analysis. To proceed towards our conclusions, we had to establish one thing-which chemical tests were the closest related to our survey.
For the purpose of finding this, paired correlation testing was chosen to analyze our raw data. Such an approach would reveal to us perhaps the simplest sort of statistical similarities-linear correlation. Linear correlation, specifically of the positive variety, would show us which sample sets were significantly related to our survey. Such relation would allow us to generate a null hypothesis and move on the next step in our statistical process-paired t-tests to confirm the correlation that we may find.
We ran paired correlation tests on each of the chemical tests, matching them to the survey data set. Tests resulting in values close to positive one would have close to perfect positive linear correlation. After viewing the results, we were pleased to see significant correlation in three of the six chemical tests. Tests in which high positive correlation was observed were as follows: Sulfide (.897), Hardness (.852), and Copper (.746).
Moving on, it was these three tests which were most likely to be significantly similar to the survey, and, on this understanding, we based our development of the null hypotheses for the next part of our process. The other tests, pH (value of -.567), Iron (-.131), and Manganese (.657) were determined to be insignificant in relation to the survey due to the values that they yielded.
Next, we ran paired t-tests on the Sulfide, Hardness, and Copper data sets, with the second members as, appropriately, the survey set. Setting .05 as the cutoff for our values, we observed the P-values for our samples to be as follows: (Sulfide) .9788, (Hardness) .0043, and (Copper) .9551.
Further, we ran paired t-tests on each of the three tests with the other two tests as the second members. The results were as follows: (Sulfide and Copper) .3990, (Sulfide and Hardness) .0137, and (Copper and Hardness) .0137.
According to the P-values displayed by Hardness, we rejected the null hypothesis that the Hardness test was significantly similar to the Survey. Also, according the to T-tests of Hardness with Sulfide and Copper, the hardness test was not statistically similar to either one of those tests.
On the other hand, Sulfide and Copper, when tested against both with the survey and with each other, produced P-values that persuaded us to accept the null hypothesis that those tests were significantly similar to the survey results. From our statistical analysis, then, we have good reason to state that the Sulfide and Copper tests were very similar to the survey results. Therefor, we are confident that the levels of Sulfide and Copper in the samples were detected and expressed in similar form in the survey.
After carefully analyzing the results from our experiments and surveys, we concluded that chemicals do have an impact on the way water tastes. We primarily draw this conclusion from the fact that the water from Peabody was rated the best tasting. We believe that this had to do with the fact that this water contained the most copper. Of all the tests we performed, copper affected the taste of the water the most. It usually adds a bitter taste to the water, but, in this case, it gave the water the most character and resulted in the best tasting by a large margin.
In another impacting test, sulfide has a major impact on the way water tastes. In the case of Peabody, the amount of sulfide was consistent with the EPA standards and was lower than the maximum contaminant level. That is another possible reason why the water in Peabody was rated best. The other waters were below the contaminant level, but impacted the taste in a negative way.
The purpose of our project was to see if the chemicals within water affected the taste. From our results, we could definitely tell that the amount of inorganic chemicals impacted the taste. It was interesting to note that distilled water was rated the worst water of the five we tested. This might have to do with the fact that people are accustomed to chemicals in their water. The distilled water tested with no copper in it. This is a definite reason why people liked the water in Peabody and did not like the distilled water. This points to the tendency of people to grow accustomed to water tastes and chemicals.
We have learned very much from this lab. The amount of contaminants in water does affect its taste. Whether that taste is good or bad, obviously depends on the people. However, the people on Western thought that the water in Peabody was the best. This leads to the simple conclusion that people grow accustomed to their surroundings. Chemicals in water do affect the test. We had a lot of fun finding that out.
This project was called a student generated discovery lab. In our project, this statement could not have been more true. We discovered something that we had not previously known. We also discovered ourselves and each other. We learned from our findings, but also from each other. That makes this lab a true success.
For Further Info on this Topic, Check out this WWW Site: refer to "Relevance of Research Question" in packet .
Return to the Topic Menu
IMPORTANT: For each Response, make sure the title of the response is different than previous titles shown above!