Arab J Geosci DOI 10.1007/s12517-014-1428-5

ORIGINAL PAPER

Lab-evaluation of applicability selected hydrogeological tracers in physical/chemical conditions Fateme Jafari & Werner P. Balderer & Hamid R. Jahani & Saman Javadi

Received: 14 December 2013 / Accepted: 10 April 2014 # Saudi Society for Geosciences 2014

Abstract In this research, the effects of a number of physical/ chemical factors such as pH, sunlight, temperature, and salinity, as well as adsorption in porous media on the applicability of selected groundwater tracers are evaluated. Tracers from different categories as fluorescent dye tracers (uranine, eosin, and rhodamine B), chemical salts (NaCl and KCl), and non-fluorescent dye tracer (KMnO4) have been tested. This research was conducted in the laboratory. The results show that uranine losses its florescence in acid environments, while in alkaline conditions its florescence increases. The results also show that due to photochemical decay, eosin is the most unstable tracer if subjected to sunlight. KMnO4 turns to brown under sunlight and high temperature conditions, and may lose its usefulness as a tracer. Results also confirm that the fluorescence intensity of rhodamine B decreases with increasing temperature and/or salinity. Uranine and eosin have high resistance against high temperature and salinity conditions. An important factor in ground water tracing is adsorption of tracer in porous media environment. Our research show that rhodamine B would easily adsorb to fine grain porous media, while uranine and eosin are of high resistance against adsorption; KMnO4 is adsorbed easily too. Keywords Solids mass transfer coefficient . Core-annulus structure . Turbulent diffusion . Gas-solid fluidization F. Jafari (*) Water Research Institute, Ministry of Energy, Tehran, Iran e-mail: [email protected] W. P. Balderer Department of Earth Sciences, Swiss Geotechnical Commission, ETH Zurich, Zurich, Switzerland H. R. Jahani Geo9 Groundwater Explorers, Perth, Australia S. Javadi Tehran University, Tehran, Iran

Introduction The use of tracer dyes is a technically valid and costeffective method for characterizing contaminant fluxes and hydraulic properties in complex hydrogeologic systems (Arsnow et al. 2010). The application of tracers provides a method for the investigation of hydrological system (Leibundgut et al. 2009). Davis et al. (1985) defined tracer as matter or energy carried by groundwater which will give information concerning the direction of movement and/or velocity of the water and the potential contaminants which might be transported by the water. Groundwater tracing is amongst the most reliable methods applied in hydrogeology where hydrodynamic characteristics of the aquifer are to be assessed. Of the most important applications of this technique, one could point to determination of aquifer dispersion coefficient (Seiler et al. 1989), groundwater flow direction and effective average linear velocity (Einsiedl 2005; Morales et al. 2007), subsurface hydraulic connection among water resources (Buzády et al. 2006), and pollution movement studies (Imes and Fredrick 2002; Ammann et al. 2003). At the same time, selection of the most suitable tracer/s is one of the most critical tasks in a groundwater tracing study which could result in the failure of the experiment if inappropriate. Different environmental factors affect the tracer both quality and quantity wise. Lindqvist (1960) studied photochemical decay rates for some fluorescent dyes. According to Benischke (2005), the photochemical decay rate for uranine is high, though this rate is less for rhodamine B. This would cause uranine to be a less favorable tracer. Smart and Laidlaw (1977) studied temperature on some tracers such as uranine and rhodamine B, in Bristol and Sheffield, England. Their results indicated high resistance of uranine against temperature, while rhodamine B loses its fluorescence by increasing the temperature.

Arab J Geosci Table 1 Excitation and emission wavelengths for the selected fluorescent dyes Tracer

Formula

CAS number

Color index number (CI)

Excitation wavelength (nm)

Emission wavelength (nm)

Uranine Eosin Rhodamine B

C20H10O5Na2 C6H6Br2N2Na2O9 C28H31ClN2O3

518-47-8 95,917-83-2 86,893-15-4

45,350 45,380 45,170

490 515 555

512 537 580

Mon et al. (2006) worked on adsorption of four dye tracers on the sandy sediments in the laboratory using column method. Smart and Laidlaw (1977) examined the effect of different suspended sediment concentrations on the selected fluorescent tracers. They also applied different sediment types and studied the effect of sediment type on tracer adsorption. In this research, we selected uranine, eosin, and rhodamine B from fluorescent dye tracers which are among the most applied tracers in groundwater tracing studies. The other studied tracers include potassium permanganate from nonfluorescent dye tracers, which suits short distance tracing and has fast and simple detecting system, and sodium and potassium chloride from chemical salts of the most used tracers, again with the advantage of fast and simple field measurement. Some tracers might have been examined in earlier researches, but have been studied again to compare them with other tracers. The behavior of each tracer is evaluated under different physical/chemical factors including pH, temperature, salinity, and sunlight, as well as adsorption in porous media.

Measuring systems Shimadzu RF 1501 spectrofluorophotometer was used to detect fluorescent dyes uranine, rhodamine B, and eosin after calibration and adjusting for the appropriate excitation and emission wavelengths (Table 1). In order to measure sodium and potassium chloride, electrical conductivity of the samples was measured using Multimeter HACH Sension 156 instrument after calibration against salt concentration. For potassium permanganate, spectrophotometer HACH DR-2010 was used to measure salt concentration following calibration of concentration against color intensity. All measurements were performed according to ASTM standards.

Materials and methods Tracers Table 1 shows the formula, color index (CI), CAS number and excitation, and emission wavelengths for the three studied fluorescent tracers. Uranine is one of the most used dye tracers for hydrogeologic purposes (Benischke 2005). The tracer is an orange-red powder well soluble in water. Eosin is used less frequently than uranine and is supplied as a red crystal powder. Rhodamine B is another groundwater tracer used mainly in karstic terrains but less frequently than uranine and eosin. It is supplied as a powder with dark greenish color (Benischke 2005). Sodium chloride is the most used salt tracer for hydrogeologic purposes (Käss 1998). It is widely supplied as cooking salt, and its solubility in water is 360 g L−1 at 25 °C. Potassium chloride has some advantages when compared with sodium chloride, although it is used less frequently. The last tracer is potassium permanganate which is rarely used.

Fig. 1 Physical model for determining tracer adsorption rates

Arab J Geosci Fig. 2 a Change in the initial fluorescence of uranine against pH, b change in the initial fluorescence of eosin against pH, c change in the initial fluorescence of rhodamine B against pH, d change in NaCl concentration readings against pH, e change in KCl concentration readings against pH, and f change in KMnO4 concentration readings against pH

Evaluation of physical/chemical effects methods pH effect Naturally, groundwater pH is around 7, but depending on the geological and environmental conditions it may deviate slightly towards acidic or alkaline states. The pH may affect fluorescence of some tracers to some extents (Smart and Laidlaw 1977). It can be used in dye detection procedure while measuring more than one tracer in a sample. In order to evaluate

the pH effect, different concentrations of tracers such as 1, 5, and 10 ppb for fluorescent tracers; 1,500 and 1,600 μs cm−1 for sodium and potassium chloride; and 1, 5, and 10 mg L−1 for potassium permanganate were prepared and measured in both acidic and alkaline states. Temperature effect The impact of temperature on tracers is considerable especially where groundwater flow in a hydrothermal

Arab J Geosci Table 2 The effects of pH on selected fluorescent and non-fluorescent tracers Tracer

Acidic pH

Alkaline pH

Uranine Eosin Rhodamine B Sodium chloride Potassium chloride Potassium permanganate

Noticeable quenching in FI in pH below 6 quenching in FI in pH below 4 quenching in FI in pH below 5 Stable in natural waters pH ranges Stable in natural waters pH ranges Stable in natural waters pH ranges

Noticeable increase in FI in pH above 8 No considerable change No considerable change Stable in natural waters pH ranges Stable in natural waters pH ranges Stable in natural waters pH ranges

Fig. 3 a The effect of temperature on the fluorescence of uranine, b the effect of temperature on the fluorescence of eosin, c the effect of temperature on the fluorescence of rhodamine B, d the effect of temperature on NaCl concentration readings, e the effect of temperature on KCl concentration readings, and f the effect of temperature on KMnO4 concentration readings

Arab J Geosci

system is of interest. It is studied by increasing and decreasing sample temperature (between 0 and +70 °C) and observing the tracer behavior. For sodium and potassium chloride, this was carried out following the correction of electrical conductivity for temperature. So, the electrical conductivity of the sample would be only a function of its salt content and not its temperature.

Fig. 4 a The effect of addition of two different salts on the fluorescence of uranine, b the effect of addition of two different salts on the fluorescence of eosin, c the effect of addition of two different salts on the fluorescence of rhodamine B, and d the effect of addition of two different salts on KMnO4 concentration reading

Salinity effect Salinity effect is of utmost importance particularly in brine or coastal environments, and where saline lenses are present in the aquifer. To understand the consequences of this parameter, the salinity of the samples was increased using two common types of salts, sodium chloride and potassium chloride. This

Arab J Geosci

increase led to increment of electrical conductivity to tens of thousands of micro per centimeter. The tracer concentrations were then measured.

duplicate sample with the same tracer concentration was kept in darkness and analyzed later to check for other effects. Adsorption in porous media

Sunlight effect The stability of tracer against light is taken into consideration where water is expected to be exposed to sunlight, e.g., while emerging in surface water or during sample storage and transport. To analyze this effect, tracers were exposed to sunlight all in same amount and similar condition in front of the sun. A Fig. 5 a The effect of sunlight on uranine fluorescence, b the effect of sunlight on eosin fluorescence, c the effect of sunlight on rhodamine B fluorescence, d the effect of sunlight on NaCl concentration reading, e the effect of sunlight on KCl concentration reading, and f the effect of sunlight on KMnO4 concentration reading

Accurately determining water fluxes and mass transfer in natural soils remains an important hydrological issue (Gerke et al. 2013). All tracers are to some extent adsorbed in the geologic environment while flowing in the aquifer. This is one of the most important factors responsible when the amount of recovered tracer is considerably less than the injected amount.

Arab J Geosci Table 3 Adsorption rates for different initial concentrations of selected tracers in three sediment types Tracer

Initial concentration Adsorption rate % Class A Class B Class C

Uranine

1 ppb

10 ppb 20 ppb Eosin 1 ppb 10 ppb 20 ppb Rhodamine B 1 ppb 5 ppb 10 ppb Sodium chloride 1 mg L−1 5 mg L−1 10 mg L−1 Potassium chloride 1 mg L−1 5 mg L−1 10 mg L−1 Potassium 10 mg L−1 permanganate 50 mg L−1 100 mg L−1

17.2

20.6

21.6

13.2 10.0 1.6 1.4 1.1 83.7 80.9 79.4 32.0 30.0 27.3 30.8 29.0 25.8 74.8 73.0 80.6

15.1 14.0 1.8 1.6 1.2 58.9 82.9 80.4 34.3 31.2 28.8 32.8 29.4 26.0 83.4 87.0 86.5

18.9 18.2 2.0 1.8 1.4 88.4 84.9 81.4 34.1 33.5 29.1 34.1 31.9 27.3 87.1 85.7 90.4

To analyze this effect, three different sediment classes were selected as porous media, named as A, B, and C, with clay contents of 0.5, 2.5, and 5 %, respectively. As shown in Fig. 1, tracers of different concentrations were injected through the sediment column, and the outflow water was sampled within the specified period (Fig. 1). The breakthrough curves were drawn for each experiment, and the amount of tracer recovered was computed.

Results and discussion pH effects The results of pH effect on tracers are shown in Fig. 2a–f. Four different initial concentrations of each tracer have been used, samples 1 to 4. Their initial concentration and FI (Fluorescence Intensity) are presented in graphs. According to Fig. 2a, increasing pH to 8 or higher will cause increase in the uranine fluorescence intensity (FI) noticeably. Base on Fig. 2a, FI from 186 reaches to 1,000 when pH increases to more than 8. Decreasing pH to below 5 will diminish the FI totally in one sample and drop FI severely in two other samples. The maximum intensity for uranine is observed in pH 10 and higher. The uranine FI also disappears at pH 4.5 and lower. This is in accordance with Smart and Laidlaw’

(1977) findings. Compared to uranine, eosin is less sensitive to pH condition (Fig. 2b). The changes in eosin intensity occur in acidic pH; it is almost stable in alkaline environment. The fluorescence intensity of eosin starts to decrease in pH 5. Rhodamine B behavior is more or less similar to uranine, although it is again less sensitive than uranine (Fig. 2c). All discussed pH effects are reversible (Fig. 2). In pH range 3 to 12, no considerable change was seen in electrical conductivities of sodium and potassium chloride samples (Fig. 2d, e). Increase in EC is seen in pH values less than 3 and more than 12, although no change is also seen in color intensity for potassium permanganate (Fig. 2f). In addition, the results of the pH effect are presented in Table 2. Temperature effects The results of temperature effect on tracers are shown in Fig. 3a–f. Uranine did not show considerable change in fluorescence intensity except some increase in high temperature (Fig. 3a). Eosin, too, did not appear to be affected by temperature (Fig. 3b). Unlike these two, the fluorescence intensity of rhodamine B diminished to when temperature approached 70 °C (Fig. 3c). The process was reversed when temperature raised, i.e., the fluorescence intensity decreased. Sodium and potassium chloride showed stability against temperature and no effect was seen (Fig. 3d, e). Long-term temperature increase on potassium permanganate caused change in its color due to production of brown MnO2 sediment in the sample. Increasing the temperature for less than 1 hour did not cause any effect on potassium permanganate samples (Fig. 3f). Salinity effect According to Fig. 4a, d, the increase in salinity has no impact on the uranine fluorescence until the sample EC approaches 100 μs cm−1 (Fig. 4a). From this point onward, the sample fluorescence increases. This test was carried with both NaCl and KCl salts, and results were similar. Regarding the eosin, increasing the solution salinity with NaCl and KCl did not lead to the same effect (Fig. 4b). KCl salinity did not change the solution fluorescence anymore while increasing NaCl salinity made an inverse effect on the solution fluorescence, yet this quenching is not considerable that it reaches at maximum almost 20 % in the conductivity value range of 100,000 μs cm−1. Rhodamine B was the most sensitive fluorescent tracer against salinity increase. It lost around 30 % of its fluorescence in conductivity of 100,000 μs cm−1 (Fig. 4c). Yet, it was less sensitive to NaCl salinity comparing with KCl salinity. Potassium permanganate color intensity was not affected while

Arab J Geosci

Fig. 6 Minimum adsorption rates for some tracers

increasing solution salinity to up to 10,000 μs cm−1 with both salts (Fig. 4d). Sunlight effect The results of sunlight effect on tracers shown in Fig. 5a–f. Figure 5a–c shows the effect of sunlight on the fluorescent tracers. It clearly confirms the impact of photochemical decay on the fluorescence intensity of the fluorescent tracers. In uranine samples, quenched to almost less than half and diminished after 15 days (Fig. 5a). The photochemical decay rate for

eosin was significantly higher than uranine so that after 4 h the fluorescence property of all samples disappeared totally. Eosin FI dropped from 600 in 4 h to about 10 conversely, uranine stay fairly stable (from 950 reaches to 920) during first 4 days (Fig. 5a, b). In rhodamine B samples, drop in the fluorescence happened but with a lesser extent as compared to both eosin and uranine. After 15 days exposing to sunlight, rhodamine B samples maintained almost 20 % of their initial fluorescence intensity (Fig. 5c). According to the results, among the three fluorescent tracers, rhodamine B is of the highest resistance against the sunlight, while eosin is the least resistant. The electrical conductivity of the salt tracers showed no changes when exposed to sunlight, but potassium permanganate samples increased their color intensity after being exposed for 1 day, because it caused change in its color due to production of brown MnO2 sediment in the sample (Fig. 5d–f). Adsorption in porous media Table 3 shows the adsorption behavior of selected tracers in three different sediment classes. As expected, the amount of

Table 4 The effects of selected physical/chemical factors on the studied tracers Tracer

Uranine

Eosin

Rhodamine B

Sodium chloride

Potassium chloride

Potassium permanganate

Physical/chemical factor Low pH

High pH

Sunlight

Temperature

Salinity

Adsorption in porous media

− Severe fluorescence quenching* + Moderate quenching of fluorescence in pH<4* + Moderate quenching of fluorescence in pH<5* ++ Stable

+ Considerable increase in fluorescence* ++ Stable in alkaline environment ++ Stable in alkaline environment ++ Stable

− Adaptable in short periods** −− Severe fluorescence quenching** + Adaptable in short periods** ++ Stable

++ Stable in high temperatures ++ Stable in high temperatures −− quenching of fluorescence* ++ Stable

+ stable

++

++

++

++

++ Stable in high salinities + Stable in high salinities − quenching of fluorescence** − High amount of tracer required −

− Moderate adsorption −

Stable

Stable

Stable

Stable

++

++

−

−

High amount of tracer required +

Moderate adsorption −−

Stable

Stable

Adaptable in short periods**

Adaptable in short periods**

Stable

High adsorption

++ Adaptable; + Semi-adaptable; − Semi-inadaptable; −− Inadaptable Abbreviations: ppb part per billion; FI Fluorescence Intensity; EC Electrical Conductivity; μs micro Siemens *Reversible **Not reversible

++ stable −− High adsorption

Arab J Geosci

adsorbed tracer in class A is less than in class B, and the amount in class B is less than in class C. For example, for 1ppb uranine sample, percentages of adsorption in sediment classes A, B, and C are 17.2, 20.6, and 21.6 %, respectively. According to these results, rhodamine B with adsorption of 88.4 % of 1-ppb sample is of the highest adsorption rates among all other tracers. In the next step to rhodamine B, potassium permanganate has an adsorption rate of 87.1 % for 10 mg L−1 sample. Figure 6 shows the minimum adsorption rates for the selected tracers.

Conclusions Selecting the proper tracer is an extremely important exercise in a groundwater tracing program. To do this, it is necessary to have sufficient knowledge of the aquifer conditions where the tracer is to flow through and the environmental factors that may affect the tracer. According to the results of this research, one limiting factor for using potassium permanganate and rhodamine B is the presence of fine grain materials in the aquifer matrix. These two tracers are also less applicable in high temperature environments (i.e., geothermal systems). Using potassium chloride and sodium chloride as a groundwater tracers in alluvial aquifers where clay minerals exist may lead to considerable tracer loss. Table 4 summarizes the effects of selected physical/chemical factors on the studied tracers.

Acknowledgment The authors wish to thank Water Resources Institute of Iran, Office of Applied Research for financially supporting this research coded WRE-82097.

References Ammann A, Hoehn E, Koch S (2003) Ground water pollution by roof runoff infiltration evidenced. Water Res 37:1143–1153 Arsnow GF, Vancil MA, Schreiber RP, Ramacciotti CN (2010) Dye tracer study—tried and true method yields surprising results. Proc Annu Int Conf Soil Sediments Water Energy 15:335–350 Benischke R (2005) Fluorescent tracers in hydrology. Austria: Institute for Geothermic and Hydrogeology, Graz Buzády A, Erostyák J, Paál G (2006) Determination of uranine tracer dye from underground water of Mecsek Hill, Hungary. J Biochem Bioph Methods 69:207–214 Davis SN, Campbell D, Bentley HW, Flynn TJ (1985) Ground water tracers. National Water Well Association, Worthington Ohio Einsiedl F (2005) Flow system dynamics and water storage of a fissuredporous karst. J Hydrol 312:312–321 Gerke KM, Sidle RC, Mallants D (2013) Criteria for selecting fluorescent dye tracers for soil hydrological applications using uranine as an example. J Hydrol Hydromech 61(4):313–325 Imes JL, Fredrick BS (2002) Using dye-tracing and chemical analyses to determine effects of a wastewater discharge to Jam Up Creek on water quality of Big Spring, Southeastern Missouri. U.S. Geol. Surv. Fact Sheet 103-02: 6 Käss W (1998) Tracing Technique in Geohydrology. Germany: Umkirch, LKr. Breisgau-Hochschwarzwald Leibundgut Ch, Maloszewski P and Külls Ch (2009) Tracers in Hydrology. John Wiley & Sons: 432 Lindqvist L (1960) A flash photolysis study of fluorescein. Ark Kemi 16(8):79–137 Mon J, Flury M, Harsh JB (2006) Sorption of four trarylmethan dyes in a sandy soil determined by batch and column experiments. J Geoderma 133:217–224 Morales T, Valderrama IF, Uriarte JA, Antigüedad I, Olazarn M (2007) Predicting travel times and transport characterization in karst conduits by analyzing tracer breakthrough curves. J Hydrol 334(1–2): 183–198 Seiler K, Maloszewski P, Behrens H (1989) Hydrodynamic dispersion in karstified limestones and dolomites in the upper Jurassic of the Franconia Alb. J Hydrol 108:235–247 Smart PL, Laidlaw IMS (1977) An evaluation of some fluorescent dyes for water tracing. J Water Resour Res 13(1):15–33