Chem. Senses 26: 85-89,
2001
© Oxford University Press 2001
Determination of the Taste Threshold of Copper in Water
Institute of Nutrition and Food Technology, University of Chile, Macul 5540, PO Box 138-11, Santiago, Chile
Correspondence to be sent to: Isabel Zacarías, PO Box 138-11, Santiago, Chile. e-mail: izacaria{at}uec.inta.uchile.cl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Copper effects on human health represent a relevant issue in modern nutrition. One of the difficulties in assessing the early, acute effects of copper ingested via drinking water is that the taste of copper may influence the response and the capacity to taste copper in different waters is unknown. The purpose of the study was to determine the taste threshold of copper in different types of water, using soluble and insoluble salts (copper sulfate and copper chloride). Copper-containing solutions (range 1.0–8.0 mg/l Cu) were prepared in tap water, distilled deionized water and uncarbonated mineral water. Sixty-one healthy volunteers (17–50 years of age), with no previous training for sensory evaluation, participated in the study. A modified triangle test was used to define the taste threshold value. The threshold was defined as the lowest copper concentration detected by 50% of the subjects assessed. To evaluate the olfactory input in the threshold value obtained, 15 of 61 subjects underwent a second set of triangle tests with the nose open and clamped, using distilled water with copper sulfate at a concentration corresponding to the individual’s threshold. The taste threshold in tap water was 2.6 mg/l Cu for both copper sulfate and copper chloride. The corresponding values for distilled deionized water were 2.4 and 2.5 mg/l Cu for copper sulfate and copper chloride, respectively. In uncarbonated mineral water the threshold values were slightly higher, 3.5 and 3.8 mg/l Cu for copper sulfate and for copper chloride, respectively, which are significantly higher than those observed in tap and distilled waters (P < 0.01, Kruskal–Wallis test). The taste threshold did not change significantly when the nose was clamped. In conclusion, the median values for copper taste threshold were low, ranging between 2.4 and 3.8 mg/l Cu, depending on the type of water.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Copper is an essential nutrient for humans, necessary for enzymes and proteins indispensable for life. However, it is also toxic when ingested in large amounts. Most of the cases that suffer toxic effects are due to accidental or deliberate ingestion of many milligrams or grams of copper salts. Studies by Cohen et al. and Béguin-Bruhin et al. using distilled water and mineral spring water suggested that copper concentrations >3 mg/l Cu would produce changes in taste and color, a concentration at which toxic effects are not observed (Cohen et al., 1960
; Béguin-Bruhin et al., 1983). For this reason it was generally assumed that changes in the color and taste of drinking waters are relevant and efficient in protecting humans from toxic effects. In recent years though, interest in the early adverse effects of copper has led to the question of to what extent color or taste in drinking water may be sufficient to protect from early toxic effects. In fact, there are several anecdotal reports of individuals that suffered gastrointestinal symptoms (mainly nausea, vomiting, diarrhea and abdominal cramps) after ingestion of copper-contaminated waters.
Tap water usually contains a few milligrams of copper per liter. Copper concentration in drinking water varies according to the natural mineral content, pH, hardness of the water, anion concentrations, oxygen concentration, temperature and characteristics of the plumbing system. The principal source of copper in water is corrosion of copper pipes (Olivares and Uauy, 1996a,b; WHO, 1996, 1998; Environmental Protection Agency, 1997). Although most drinking waters contain <3 mg/l Cu, values as high as 22– 28 mg/l Cu have been reported, especially if acidic water was stagnant in copper tubes (WHO, 1993; Bent and Bohn, 1995).
Studies carried out by us in order to characterize the early adverse effects of copper contained in water showed that the first symptom to be observed is mild and transient nausea, which appears at 4 mg/l Cu (delivered as copper sulfate) in the most sensitive individuals (M. Olivares et al., submitted for publication); about two thirds of the individuals tested repeated the positive response when rechallenged with the same dose. These results posed the question of to what extent a mild non-specific symptom such as nausea may be influenced by the capacity to taste the ‘bad’, metallic taste reported for copper by the same individuals. To shed light on this aspect we carried out this study with the objective of determining the taste threshold of copper in different waters using some of the copper salts commonly found in drinking waters.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Subjects and design
Apparently healthy volunteers 18–60 years of age were invited to participate in this study. They received oral and written explanations of the protocol and answered questionnaires aimed at applying exclusion criteria. The volunteers were to be free of chronic health problems, especially oral and nasal conditions that may have interfered with tasting or olfactory capacities, and to have no previous experience with sensory evaluations. Those that fulfilled these requirements and volunteered signed a written informed consent before beginning the study. The Institutional Review Board approved the protocol at INTA, University of Chile.
Taste evaluations were performed early in the morning, in a blind fashion, at least 2 h after a light breakfast. Because it was assumed that perceiving copper taste at low concentrations could be difficult, the subjects were instructed to keep the sample bolus in their mouth for at least 1 min, spit it out, rinse their mouth with a sucrose solution (0.005 M) (which at this concentration gave a pleasant neutral sensation and eliminated bitter and other off tastes) and wait for at least 60 s before tasting the next sample (Amerine et al., 1965; Witting de Pena, 1980; Meilgaard et al., 1991). Individuals gave their judgments after tasting all the samples assigned for that session. At the end of the evaluation salty crackers were served with coffee or tea in order to eliminate any persistent or unpleasant taste.
To avoid any persistent taste influencing the following tasting events, in each session the volunteers received one single copper concentration, ranging from 1.0 to 8.0 mg/l Cu, as CuSO4·5H2O or CuCl2. Panelists were tested three times per week, beginning at 1 mg/l Cu and incrementing at 1 mg/session until a positive report defined the threshold value for that person. A positive report was defined as correct identification of the copper-containing sample (see below).
Screening test
This was performed with the purpose of defining individual taste threshold zones (Amerine et al., 1965; Witting de Pena, 1980; Meilgaard et al., 1991). Briefly, it consisted of presenting the subject with eight cups, each one containing 20 ml of solution, encoded with random three digit numbers. The copper concentrations were presented in ascending order from 0 to 8 mg/l Cu at room temperature (18–23°C) in a blind fashion. When the subject perceived a distinct taste sensation, he or she recorded and described the particular taste. This value and those near it represented the threshold zone for that individual. This test was applied prior to the first set of trials only, using copper sulfate as the copper salt. Once the threshold zone was defined, in the next session the subject underwent a modified triangle test.
Modified triangle test
The triangle test is a three-product test in which the subject’s task is to identify the different one (Meilgaard et al., 1991). To decrease the possibility that the result was due to random guessing in this study we modified the method, presenting five samples to each subject: four identical ones and one containing the test solution. The panelist was instructed to identify the odd sample and indicate the perceived taste. Five cups of 20 ml each were presented at room temperature (18–23°C); the cup itself did not impart a taste or odor to the sample. The samples were coded with a three digit number and the set of five cups was presented to the subject in a random order (ABBBB, BABBB, BBABB, BBBAB and BBBBA, where A was the cup containing the test solution and B the placebos).
Triangle test with nose ‘clamped’
With the purpose of evaluating the effect of olfactory input on taste threshold and the capacity to confirm the taste threshold when the nose was ‘clamped’, a subgroup of 15 individuals underwent a second series of triangle tests, one with the nose open and the other with the nose clamped, during which they received copper sulfate solution prepared in distilled water at the concentration corresponding to his/her previously defined taste threshold.
Sample size calculation, expression and analysis of results
To obtain statistical significance when differences between groups were at 20%, using an 
of 5% and ß of 80%, the sample size was calculated as 60. Results were expressed as the median threshold value, defined as the lowest copper concentration detected by 50% of the subjects evaluated. Statistical analysis included non-parametric tests and the Kruskal–Wallis ANOVA median test (Colton, 1974), analyzing by copper salt and by water type.
Test waters
Tap water (TW)
This was obtained from taps located in our laboratory, which fulfil Chilean quality norms (Empresa Metropolitana de Obras Sanitarias, 1996). Its chemical composition was 169 mg/l Ca, 46 mg/l Na, 2 mg/l K, 12 mg/l Mg, 68 mg/l Cl, 280 mg/l SO4. The pH was 7.4. These values are also in agreement with recent data published by Lagos and coworkers (Lagos et al., 1999).
Uncarbonated mineral water (UCMW)
This was purchased in the local market (Cachantun; Chile). Its composition before the addition of copper was 216 mg/l Na, 4 mg/l K, 76 mg/l Ca, 10 mg/l Mg, 28 mg/l Cl, 85 mg/l SO4, and 25 mg/l SiO3. The pH was 6.3.
Distilled deionized water
This was purchased as a single batch from Corning (Mega-Pure System MP-6ª). The pH was 5.8.
Copper salts
Because the most common soluble copper compounds in water are copper chloride, copper nitrate and copper sulfate (Environmental Protection Agency, 1998), for this study we chose Cu SO4·5H2O and CuCl2 (pro-analysis grade), both purchased from Merck (Darmstadt, Germany).
Solutions
These were prepared daily, at room temperature, using a 100 ml stock solution prepared with distilled deionized water containing 0.5 mg/ml Cu. This solution was added to each test water in amounts such as to obtain the test concentrations 1–8 mg/l Cu. The final copper concentration in the test solutions offered to the volunteers was verified by atomic absorption spectrometry (Perkin Elmer 2280). Results were within 5% of the expected concentration.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Screening test
The results of this test helped define the perception zone of each subject (Figure 1A). Of the 60 panelists, 43 and 77% perceived the copper taste at concentrations of 2 and 3 mg/l Cu, respectively.
|
Modified triangle test
Cumulative frequencies of threshold values for copper sulfate and copper chloride in tap water are shown in Figure 1B. Of the individuals, 31 and 38% were able to identify the copper taste at a concentration of 2 mg/l Cu as copper sulfate and copper chloride, respectively. Corresponding values for 3 mg/l Cu were 62 and 58%, respectively. Fifty percent of the subjects reported taste perception at a concentration of 2.6 mg/l Cu for both copper salts.
In DDW, tasting of 2 mg/l Cu (as copper sulfate or chloride, respectively) was reported by 39 and 44% of the subjects (Figure 1C). The corresponding figures for 3 mg/l Cu were 65 and 56%, respectively. Fifty percent of the subjects perceived taste at concentrations of 2.4 and 2.5 mg/l Cu, respectively.
In UCMW, tasting of 2 mg/l Cu (as copper sulfate or chloride, respectively) was reported by 23 and 31% of the subjects (Figure 1D). The corresponding figures for 3 mg/l Cu were 45 and 42%, respectively. Fifty percent of the subjects perceived taste at concentrations of 3.5 and 3.8 mg/l Cu, respectively.
ANOVA analysis of the combined data for both salts and the three types of waters showed that the results obtained with TW and DDW were similar. In contrast, UCMW gave significantly higher threshold values (P < 0.01). For all three waters and the two copper salts the vast majority of the subjects reported a bad, unpleasant or disgusting metallic, acid, bitter or salty taste.
Effect of ‘clamping’ the nose
When the trials were repeated offering the individuals the copper concentration at which they first reported tasting, 7/15 individuals confirmed their threshold when their nose was open while 8/15 individuals confirmed the threshold when their nose was clamped. When the second set of trials performed in a subgroup of 15 individuals are compared, in 10 of the 15 individuals tested the response was the same, either confirming or not confirming the threshold concentration, with the nose open or clamped. 
2 analysis showed no significant differences in taste thresholds reported with the nose open or clamped (Table 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The results of this study help establish the taste threshold for copper in three types of water and with two copper salts, some of them frequently present in daily life. The values 2.5–2.6 mg/l Cu for DDW and TW and 3.7 mg/l Cu for UCMW show that water quality does influence the capacity to taste copper sulfate and copper chloride. We would like to speculate that the higher threshold observed for UCMW may be attributed to copper chelation by polyvalent organic anions (WHO, 1998
). The thresholds found are in a range of concentrations slightly below those at which the earliest adverse effects due to acute copper exposure appear, therefore raising the possibility that the unpleasant taste of copper in the water may be a factor influencing the appearance of mild symptoms. However, because in these studies the individuals did not swallow the solution, it was not possible to evaluate the actual relationship between taste reporting and the appearance of symptoms. None of the subjects experienced symptoms.
The fact that drinking waters usually have <3 mg of copper coincides with the fact that in most places people do not taste copper in water. Taking into account that nausea and other gastrointestinal symptoms may appear at as low as 4 mg/l Cu (as copper sulfate) (M. Olivares et al., submitted for publication), the perception of taste in tap water may be an important indicator to sensitive individuals preventing them from ingesting a specific water, but it must be taken into account that 
40% of people will not taste copper content at concentrations <4 mg/l Cu and therefore will not have the chance to become aware of any risk. This means that color and taste of a specific drinking water are not necessarily good indicators to alert people to avoid drinking a specific water.
It is interesting that very little differences were observed between the salts used, sulfate or chloride. Total sulfates in drinking water are usually high (250 mg/l in Santiago) (Empresa Metropolitana de Obras Sanitarias, 1996) and they are not tasted, therefore, it is unlikely that sulfate instead of copper could be responsible for the taste reported. Béguin-Bruhin et al. postulated that tasting copper in water would depend on its chemical species and its solubility. In his study the taste of copper was described as metallic, bitter, chemical and astringent. In our study the subjects also described a variety of tastes, adding acid, bitter and salty to those previously described (Béguin-Bruhin et al., 1983).
In studies of some heavy metal ions, such as Ag+, Hg2+ and Pb2+, and compounds like ferrous sulfate, tasting and olfactory capacities appear clearly related (Moncrieff, 1967; Hettinger et al., 1990; Meilgaard et al., 1991). In our study repeat testing at the copper concentration that defined individuals tasting thresholds showed that only half of the subjects confirmed their thresholds. Having the nose open or clamped did not significantly change the threshold concentrations, either as individuals or as a group.
Since the methodologies used in this study were based on those described by Cohen et al. and Béguin-Bruhin et al. (Cohen et al., 1960; Béguin-Bruhin et al., 1983), it is interesting to compare our results with those obtained by them. The threshold for copper in distilled water obtained by us (2.4–2.5 mg/l Cu for copper sulfate and copper chloride) is lower than those reported by Cohen et al. and Béguin-Bruhin et al. (6.6 and 3 mg/l Cu, respectively). For spring water their values using copper chloride were 1 and 13 mg/l Cu, while our mean values were 3.5 and 3.8 mg/l Cu, depending on the salt used. Unfortunately, tap water was not evaluated in these earlier studies. It is difficult to speculate about possible explanations for the differences obtained. Several factors may be involved, such as differences in the chemistry of the waters used, cultural characteristics of the study subjects or adaptation to basal concentrations of copper in drinking water in the locations of the studies, which in the case of Santiago are clearly low (<0.01 mg/l).
In conclusion, the tasting thresholds of copper sulfate and copper chloride are between 2.5 and 3.7 mg/l Cu in TW, DDW and UCMW. It is highest when copper is delivered in UCMW, probably because part of the copper dose is chelated by organic anions present in higher amounts in this type of water. The copper salt used, sulfate or chloride, had little influence on the capacity to taste copper. These findings will be relevant information to help better understand the effects of acute copper exposure.
|
Acknowledgments |
|---|
Part of this paper was presented at the 16th International Congress of Nutrition July 27–August 1, 1997, Montreal Canada. This work was supported by the Centro de Investigación Minera (CIMM), INTA, University of Chile Research Program, Santiago, Chile.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Amerine, M.A., Pangborn, R.M. and Roessler, E.B. (1965) Principles of Sensory Evaluation of Food. Academic Press, New York, pp. 53–66.
Béguin-Bruhin, Y., Escher, F., Roth, H.R. and Solms, J. (1983) Threshold concentration of copper in drinking water. Lebensm. Wiss. Technol., 16, 22–26.
Bent, S. and Bohm, K. (1995) Copper induced liver cirrhosis in a 13 month old boy. Gesundheitswesen, 57, 667–669.[Medline]
Cohen, J.M., Kamphake, L.J., Harris, E.K. and Woodward, R.L. (1960) Taste threshold concentrations of metals in drinking water. J. Am. Water Works Ass., 52, 660–670.
Colton, T. (1974) Statistics in Medicine. Little, Brown & Co., Boston, MA.
Empresa Metropolitana de Obras Sanitarias (EMOS) (1996) Análisis Físico Químico de Agua. Redes Comuna Macul, Santiago, Chile.
Environmental Protection Agency (EPA) (1997) Copper (first draft). Health and Ecological Criteria Division, Office of Water, Washington, DC.
Environmental Protection Agency (EPA) (1998) Drinking Water and Health. Technical Factsheet on Copper. Office of Water, Washington, DC (http://www.epa.gov/OGWDW/dwwh/tioc/copper.html).
Hettinger, T., Myers, W. and Frank, M. (1990) Role of olfaction in perception of non-traditional taste stimuli. Chem Senses., 15, 755–760.
Lagos, G., Maggi, L.C. and Reveco, F. (1999) Model for estimation of human exposure to copper in drinking water. Sci. Total Environ., 239, 49–70.[Medline]
Meilgaard, M., Civille, G. and Carr, B. (1991) Sensory Evaluation Techniques, 2nd edn. CRC Press, Boca Raton, FL.
Moncrieff, R.W. (1967) Chemical Senses. Leonard Hill, London, UK.
Olivares, M. and Uauy, R. (1996a) Limits of metabolic tolerance to copper and biological basis for present recommendations and regulations. Am. J. Clin. Nutr., 63, 846S–852S.
Olivares, M. and Uauy, R. (1996b) Copper as an essential nutrient. Copper nutrition in humans: essentiality and toxicity. Am. J. Clin. Nutr., 63, 791S–796S.[Abstract]
WHO (1993) Guidelines for Drinking Water Quality. World Health Organization, Geneva, Switzerland, vol. 1, pp. 219–223.
WHO (1996) Guidelines for Drinking Water Quality, 2nd edn. World Health Organization, Geneva, Switzerland, vol. 2, pp. 219–226.
WHO (1998) Guidelines for Drinking Water Quality, Addendum to 2nd edn. World Health Organization, Geneva, Switzerland, vol. 2, pp. 219–226.
Witting de Pena, E. (1980) Evaluación Sensorial: Una Metodología Actual para Tecnología de Alimentos. Talleres Gráficos USACH, Santiago, Chile.
Accepted August 13, 2000
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. D. Cuppett, S. E. Duncan, and A. M. Dietrich Evaluation of Copper Speciation and Water Quality Factors That Affect Aqueous Copper Tasting Response Chem Senses, September 1, 2006; 31(7): 689 - 697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. T. Lawless, S. Schlake, J. Smythe, J. Lim, H. Yang, K. Chapman, and B. Bolton Metallic Taste and Retronasal Smell Chem Senses, January 1, 2004; 29(1): 25 - 33. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||