Cost-Effective, Wireless, Portable Device for Estimation of Hexavalent

Subscriber access provided by University of Sunderland

Article

Cost effective and wireless portable device for estimation of hexavalent Chromium, Fluoride and Iron in drinking water Debmalya Santra, Subhradeep Mandal, Angshuman Santra, and Uttam Kumar Ghorai Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b03337 • Publication Date (Web): 03 Oct 2018 Downloaded from http://pubs.acs.org on October 4, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 18

Page 1 of 17

Title: Cost effective and wireless portable device for estimation of hexavalent Chromium, Fluoride and Iron in drinking water Authors: Debmalya Santra†§, Subhradeep Mandal†§, Angshuman Santra†, Uttam Kumar Ghorai†* †Department

of Industrial Chemistry and Applied Chemistry, Swami Vivekananda Research

Center, Ramakrishna Mission Vidyamandira, Belur Math, Howrah-711202, India

Abstract: The quality of drinking water often remains unknown to the people because of the inadequacy of cost-effective testing systems that can be used in field. Major portable instruments for water quality analysis include Ion Selective Electrodes (ISE) or Colorimeters. These are low cost devices but in case of multiple analyte detection like hexavalent Cr, Fluoride (F-) and Iron (Fe) with single instrumentation, no portable systems are available till date as per the authors’ knowledge. In this paper, we demonstrate the working of a low cost (approximate price INR 1500 or US $20) portable colorimetric system that can be operated with android smartphones wirelessly to estimate the contamination levels of Cr(VI), F-, or Fe in drinking water. This system also generates absorption spectra by recording absorbance of the analyte using Light Dependent Resistor (LDR) sensor. An android application software named “Spectruino” is developed to calculate the concentration of the analytes. We strongly believe that this cost-effective portable system will be very useful to ensure the drinking water quality throughout the continent to improve human health.

1. Introduction: About 60 percent of India’s total produced crop is being irrigated with rapidly decreasing ground water 1. Water supply of the Indian continent is served by inter-state rivers 2. Now, most of the 3-4

drinking water from these sources are contaminated by F-, Cr(VI), Fe

etc. Fluoride

contamination is considered to be the top priority to ensure the quality of drinking water after microbial pollution

5, 6

. Presently, 32 countries in the world have reported endemic levels of

fluorosis7. Hexavalent Cr is a predominant species than trivalent Cr, commonly found in water as contaminants causing carcinogenic and mutagenic effects to human health common contaminant of ground waters

8, 9

. In India, Fe is a

10, 11

the risk of cancer, arthritis, liver problems

. Gradual buildup of Fe in human body can increase

12

etc. Permissible limits for F-, Cr(VI) and Fe are about

1ppm, 0.5 ppm, 1 ppm 13-15 respectively. The techniques which are available for detection of those heavy metal and non-metal contaminants include Spectrometry, Surface Plasmon resonance sensors

(SPR),

Electrochemical

Sensors,

ISE

16,17

etc.

AAS

(Atomic

Absorption

Spectrophotometers) or SPR sensors are very accurate but the costs of those types of instruments are very high. Colorimetric and ISE methods are more portable and field effective to analyze the


Page 3 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 2 of 17 contamination levels in water. But the conventional colorimeters and ISEs 18 are very much costly. On the other hand, smartphone operated sensors are getting popularity as they are small, easy to use and portable. In the present socio-economic condition, 339.95 million people in India uses smartphones and the number is likely to increase to about 442.5 million by the year 2022 18. In this condition, smartphone operated portable colorimeters will be highly effective to determine the contamination levels in drinking water. Some smartphone operated colorimeters are being reported since 2014. C. Chen et al.20 have proposed a smartphone-based spectrometer based on 3D print technology assembled on the rear facing camera of smartphone, which is mechanically aligned with all the optical components like fiber bundle, diffracting grating, lenses and slit. In this method, absorbance of a rhodamine b (Rh B) was measured by illuminating the sample (Rh B) with a green light emitting diode (LED) with central wavelength of 532 nm. An iPhone based digital image colorimeter for detecting tetracycline (TC) in milk was reported by P. Masawat et al.21 in 2015. This paper states about the fabrication of a colorimeter that utilizes image matching algorithm to determine TC in milk. TC solution extracted from milk samples using solid phase extraction was captured and the concentration was predicted by comparing color values with those collected in a database. Another smartphone based simultaneous pH and nitrite colorimetric determination was reported for paper microfluidic devices by N. L Ruiz 22. In this method, the authors used image processing method which was coded into an Android application. Several sensing areas which contained the corresponding reagents have been used to produce selective color changes when a sample solution is placed in the sampling area. Some pH indicators like chlorophenol red and phenol red were used in determining pH and nitrite concentration of a solution. Detection of uric acid and glucose were done by X. Wang et al.23on multilayer-modified paper with smartphone as display device. In this method, the authors used a pre-processed paper and they captured image of the paper with sample. The image was then converted to grayscale and the gray values were calculated for quantitative analysis. Md. A Hossain et al.24 has proposed a smartphone based portable flurometer for pH measurements of environmental water. Here, the authors used some chemo sensors to measure pH of a solution. LED (which is used as flash for camera) on mobile was used as light source and camera of a smartphone was used as sensor. Digital image of the fluorescence originated from chemo sensor was captured and the application/ software installed in the smartphone analyses the image by measuring the color of the image. So, it can be seen from the previous works of various researchers that the smartphone-based colorimeters are developed till date, are using white light or multiple light emitting diodes (LED) for illumination source. The existing methods cannot produce absorption spectra of the solution. In these previous works it is also seen that in most cases, mobile cameras are used as sensors. On the other hand, LDR has a very interesting feature. These types of sensors are generally built with



Page 4 of 18

Page 3 of 17 PbS, PbSe, InSb

25

etc. When light falls on the sensor, the electrical resistance drops down and

when it is placed in dark, the electrical resistance goes up

26

. We know that, when light passes

through a solution, a portion of the light gets absorbed by the solution Error! Reference source n ot found.. Our work is to construct a low cost, portable, light-weight device to determine the concentration of Cr(VI), F-, and Fe in drinking water using these phenomena. In 2017, we have also developed a portable colorimeter28 which can be used in different fields of application such as dye degradation monitoring, water quality monitoring, and industrial application including remote detection of analyte in field environment. 2. Experimental Section: 2.1 Principle of recording absorption spectra from the device: Recording the absorption spectra is based on getting voltage readings from the LDR sensor which is placed on one end of the sample cuvette holding chamber. The used light source is a tri colored LED (Light Emitting Diode), capable of emitting red, green & blue lights when current is supplied through corresponding anodes. The voltage on the anodes is controlled through an Arduino microcontroller

29

(Arduino Uno). The Red, Green and Blue anodes are connected to the

microcontroller through digital output pins 6, 5 and 3 respectively

30

. The current through the

anodes are varied by setting various duty cycle of the PWM (Pulse Width Modulation). Figure 1 illustrates the change in the average current (Iavg) (Measured Using Metravi XB33C multimeter) through various anodes at different duty cycle values set from the Arduino board. Initially, the Arduino is programmed to set a digital value as PWM duty cycle

31

(Dc) which is

received through serial port ranging from 0 to 255. From Figure 1, it can be seen that the change of Iavg with Dc is linear.

Figure 1: Relationship of current through LED with PWM duty Cycle for 3 different LEDs (Red, Green, and Blue).

Figure 2: Relationship of current through LED with PWM duty Cycle for 3 different LEDs (Red, Green, and Blue).


Page 5 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 4 of 17

A CIE 1931 chromaticity diagram

32

of resolution 1366x768 is taken in Matlab® software and

different values of R, G & B pixels are noted at different wavelength mentioned in the CIE diagram. Then R, G & B values got from Matlab are plotted against the wavelengths at 20 nm intervals followed by fitting each graph with mathematical functions. In the plot (Figure 2), Red, Green and Blue (R, G, B) values are assumed to be taken as PWM values of the microcontroller. For reducing error, each graph is fitted with several fitting functions given below (Table 1, 2 and 3). Table 1 Wavelength Range

Color function (For Red Color)

400 – 460 nm

R=22787.6 − (166.74× 𝝀)+(0.40821× 𝝀𝟐 ) − (0.0003× 𝝀𝟑 )

461 – 540 nm

R=0

541 – 570 nm

R=(12× 𝝀– 6603)

571 – 680 nm

R=255

Table 1: Color fitting functions for red color. Table 2 Wavelength Range

Color function (For Green Color)

400 – 460 nm

G= 0

461 – 500 nm

G=(7.13× 𝝀 – 3295.6)

501 – 570 nm

G= 255

571 – 610 nm

G=3735 − (6.14× 𝝀 )

611 – 680 nm

G=0

Table 2: Color fitting functions for green color. Table 3 Wavelength Range

Color function (For Blue Color)

400 – 489 nm

B= 255

490 – 550 nm

B=1918.8836 − (3.48152× 𝝀 )

551 – 610 nm

B= 0

611 – 680 nm

B= −27436.88+(124.74× 𝝀) − (0.183× 𝝀𝟐 ) − (0.00009× 𝝀𝟑 )

Table 3: Color fitting functions for blue color.

In the above mentioned tables, R, G & B represent the value of pixel of Red, Green & Blue color respectively and 𝜆 represent the wavelength of light. In order to produce a required color, the equivalent wavelength is to be put in the abovementioned equations and the corresponding values



Page 6 of 18

Page 5 of 17 of the R, G &B is obtained. As the Dc value also ranges from 0-255, the R, G & B values obtained from those equations are set as the Dc value in the microcontroller. A desktop User Interface (UI) and an android application is made in which software routines send the values of R, G & B to the microcontroller via USB serial port or Bluetooth™ using HC-05 Bluetooth module 33. For the sensing circuit, a voltage divider network is made using a 5 mm LDR and 10 KΩ resistor. The output of the voltage divider network is connected to ADS1115 16 bit analog-to-digital (A/D) converter 34. The serial output from A/D converter is connected to Arduino Uno. The main function of ADS1115 is to measure the analog output (V out) from the voltage divider network and send to Arduino in digital numbers (D) ranging from 0-32768. The analog voltage from the sensing circuit and the digital output is related as: D=

Vout × 5

215

………. (1)

To get better precision, several values are averaged. So, the averaged D values are given by: Davg=

∑61 𝐷

…………. (2)

5

As per abovementioned equation, 5 subsequent outputs from ADS1115 are averaged. As per the programming used in Arduino, each acquisition from the ADS1115 sensor takes about 100 milli seconds and when 5 values are averaged, Arduino sends out the Davg value to the connected desktop pc or android mobile via serial port or Bluetooth.

The working of the desktop UI that runs the device is briefly described in supporting information (S1). The absorbance of a sample at the scanning wavelength is calculated by using this formula: 0 𝐷𝑎𝑣𝑔

A = ln

𝐷𝑎𝑣𝑔

…… (3)

0 Where, 𝐷𝑎𝑣𝑔 is the Davg value during baseline / blank scan and Davg represents the value during

sample scan. The above mentioned equation is analogous to Beer’s law: A=

𝐼0 𝐼

…… (4)

Where I0 is the incident light intensity and I is the intensity of the transmitted light. In our work, 0 we have replaced I0 and I with 𝐷𝑎𝑣𝑔 and 𝐷𝑎𝑣𝑔 respectively.After calculating the absorbance of the

sample, the software routine also stores the data in a computer readable “.txt” format. This data file can be imported to any graphing software to view the absorption spectra.


Page 7 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 6 of 17 2.2 Principle of Photometric measurements: The detailed description of the steps involved in carrying out the photometric measurements are given in supporting information section (S2). For easy and fast operation, the android app (Figure 3) is developed using MIT Appinventor 2 online android compiling platform 35.

Figure 3: Schematic connection and wiring diagram of the instrument with PC and android device

2.3 Preparation of Calibration Curve 2.3.1 For hexavalent Chromium estimation: For estimation of Cr(VI), well known 1, 5 diphenyl carbazide method is adopted

36

. At first,

standard 100 mg/L Cr solution is prepared by dissolving 0.02538 grams of AR grade K2Cr2O7 (Merck) into 100 ml deionized water. Then it is diluted to 1 mg/L. From 1 mg/L stock solution, 5 concentrations ranging from 0 to 1 ppm of Cr(VI) is prepared in 100 ml volumetric flasks.1,5 diphenyl carbazide reagent (DPC) is prepared by dissolving 5 g AR grade 1,5 DiphenylCarbazide powder (Merck) in 100 ml of AR grade acetone (Merck). Then 2 ml of DPC solution is added to each volumetric flask containing chromium standard solution. Then the rest of the volume of the flask is made up with DI water. The solution is kept in rest for 2 minutes for color development and then the absorbance is measured using our fabricated device and Shimadzu UV-3600.



Page 8 of 18

Page 7 of 17 2.3.2 For Fluoride Estimation: For estimation of fluoride in drinking water, widely adopted SPADNS-Zirconyl lake method is adopted

37

.Reagent 1is prepared by dissolving 958 mg AR grade SPADNS (4, 5-dihydroxy-3-

(parasulfophenylazo)-2, 7-naphthalene disulfonic acid, trisodium salt) (Merck) in 500 ml water. Reagent 2 is prepared by dissolving 133 mg zirconyl chloride octahydrate, ZrOCl2.8H2O, in about 25 ml distilled water. Then 350 ml conc. HCl is added to reagent 2 and diluted to 500 ml with DI water. Equal volume mixture of reagent 1 and 2 is used as color developing reagent. Reference solution is prepared by adding 10 ml of reagent 1 solution to 100 ml distilled water. 7 ml conc. HCl is diluted to 10 ml and added to the diluted reagent 1 solution. A series of standard solutions are prepared containing fluoride concentration ranging from 0 to 1 ppm. To prepare the standard solution, 10 ppm stock solution (prepared by dissolving 0.0221 g anhydrous NaF in 1000 ml DI water) was serial diluted with DI water. Color development is done by using 50 ml sample. 5 ml each of SPADNS solution and zirconyl-acid reagent is added and mixed well. The solution is kept in rest for 2 minutes then the absorbance is measured using our fabricated device and Shimadzu UV-3600. 2.3.3 For Iron Estimation: For estimation of iron in drinking water, phenanthroline method is adopted 38. In this method iron (II) is complexed with 1, 10 phenanthroline and the color of the complex is orange. The standard iron solution contains 0.25 g/l of iron (II) which is prepared from Mohr’s salt. 25 ml of the standard iron solution is pipetted into a 500 ml volumetric flask and diluted up with distilled water. 10 ml of an unknown sample solution is pipetted into a 250 ml volumetric flask and diluted with distilled water to the mark. Next two 25 ml aliquots of this solution is pippeted into two 50 ml volumetric flasks. Then using a 10 ml graduated cylinder, 4ml of 10% hydroxylamine hydrochloride solution is added and then 4 ml of 0.3% o-phenanthroline solution is added to each of the volumetric flask. The mixture is kept to stand for 10 minutes. Next, each flask is made up to the mark by mixing D.I water. Finally a calibration curve is plotted between absorbance and concentration of iron sample. 2.3.4 Smartphone application software and desktop user interface: Screenshots of smartphone application software named “Spectruino” is given in figure 4a, 4b, 4c and 4d. Those screenshots depict the output given to the user when the concentration of the analytes are above WHO (World Health Organization) limit. Figure 4e and 4f represents the concept and realistic view of the fabricated instrument respectively.


Page 9 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 8 of 17

Figure 4: (a), (b), (c) are the screenshots during determination of iron, fluoride and chromium. (d) shows the application icon in an android device (Motorola MOTO E). Real time view of the device is illustrated in figure (e) and (f). The first figure is about the concept view of the device. The later image shows the internal wiring and the connected modules to carry out the whole process as described in this work.

3. Results and Discussion: The sensitivity of the instrument depends on the sensitivity of the LDR sensor. The resolution of the analog to digital converter (ADS 1115) is given below 39: 𝑅𝑒𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 𝑖𝑛 𝑚𝑉 =

5 215

× 1000

………… (5)

Figure 5: Output from the instrument (Davg values) has been plotted with number of data points in presence and absence of the sample to record how much amplitude of signal is produced when sample is placed



Page 10 of 18

Page 9 of 17 The lowest possible measurable voltage output (resolution) from the sensor is 0.1525 mV with same step size. That means, when a minimum signal with amplitude of 0.1525 mV is produced by the VDN, the value of Davg becomes 1 and with every 0.1525 mV increase of the voltage signal, Davg value is incremented to the next positive integer till 32768. To find the signal to noise ratio of our instrument, a series of data is recorded after placing samples and blanks in the sample holder. Figure 5 illustrates the signal response of the instrument in presence and absence on the sample. The ratio of the signal to noise is found to be 9.47:1 and the corresponding SNR (in dB) is calculated using the formula written below 40: 𝑆𝑖𝑔𝑛𝑎𝑙

𝑟𝑚𝑠 𝐷𝑎𝑣𝑔

𝑆𝑁𝑅 𝑖𝑛 𝑑𝐵 = 20 × 𝑙𝑜𝑔10 (

𝑁𝑜𝑖𝑠𝑒 𝑟𝑚𝑠 𝐷𝑎𝑣𝑔

) …………… (6)

The SNR is found to be -19.1147dB. All the calibration curves for Cr(VI), F-, and Fe estimation are plotted using both our instrument and a standard SHIMADZU UV-3600 41 spectrophotometer. Figure 6a illustrates the calibration curves for hexavalent chromium determination, prepared using both our instrument and UV-3600. The calibration curves are straight lines in both cases. The

Figure 6: Calibration curves are given in (a) for hexavalent Chromium determination using our instrument and UV-3600. In (b), standard deviation is compared in case of both two instruments.

values of correlation coefficients are 0.99786 and 0.99847 respectively for our instrument and UV-3600. The equation obtained from the linear fit for our instrument is given below: 𝐶𝑜𝑛𝑐𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐶𝑟(𝑉𝐼)(𝑝𝑝𝑚) = 1.01405 × 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒( 𝐴𝑟𝑏 𝑈𝑛𝑖𝑡𝑠) ……. (7) This equation is fed in both the Desktop UI and android application named “Spectruino” to determine unknown Cr(VI) concentration. Comparison of the data obtained using our instrument is given on figure 6b. Some standard concentrations of Cr(VI) were prepared and then the concentrations were determined experimentally using our instrument and UV-3600. As it can be


Page 11 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 10 of 17 seen from the figure 6b, the difference between the results obtained by our instrument does not differ much with respect to the results obtained from UV-3600. To determine fluoride concentration in drinking water first, the calibration curves are prepared like the previous method. Figure 7(a) illustrates the calibration curves for our instrument and UV-

Figure 7: Calibration curves are given in (a) for Fluoride estimation using our instrument and UV-3600. In (b), standard deviation is compared in case of both two instruments.

3600. In case of SPADNS-Zirconyl acid reagent method, the color of the solution fades off due to formation of ZrF6− 42. So, the absorbance of the solution is likely to be highest with 0 fluoride concentration. With increasing concentration of fluoride, the absorbance of the solution decreases and the calibration curves produce negative slopes when fitted with linear equation. The correlation coefficients are 0.99907 and 0.99915 for our instrument and UV-3600 respectively. The fitted equation for fluoride determination is given below: 𝐶𝑜𝑛𝑐𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐹 - (𝑖𝑛 𝑝𝑝𝑚) = (−1.6654 × 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒) + 0.151434 …….. (8) The equation is fed in the software (“Spectruino”) to determine unknown F- concentrations in water samples. For validation of calibration curves, solutions containing standard concentrations of F- were tested using our instrument. The experimental concentrations are very much close to real concentrations but the difference between them slightly increases with increasing concentration (figure 7b). To determine the concentration of iron in drinking water, the calibration curves are prepared using phenanthroline method. The curve is given in figure 8a. The absorbance of the solution increases with increasing iron concentration in water. The correlation coefficients are 0.993 and 0.999 for our instrument and UV 3600 respectively. The fitted equation which is fed in “Spectruino” software for iron estimation is:



Page 12 of 18

Page 11 of 17 𝐶𝑜𝑛𝑐𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐹𝑒 (𝑝𝑝𝑚) = 0.00111 + (0.44017 × 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒) − (0.23145 × 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 2 ) + (0.06615 × 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 3 )…………….. (9)

A range of standard Fe solutions are prepared and the experimental concentrations are tested against real concentration (figure 8b).

Figure 8: Calibration curves are given in (a) for Iron estimation using our instrument and UV-3600. In (b), standard deviation is compared in case of both two instruments.

Field analysis of the samples are done using our instrument in 11 different districts of West Bengal (Kolkata, Birbhum, Murshidabad, Howrah etc.) for F-, Cr(VI) and Fe. The results are given in figure 9.

Figure 9: Test result of 11 samples for hexavalent Chromium (a), Fluoride (b), and Iron (c). The comparison of LoQ is given in (d)


Page 13 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 12 of 17 In figure 9a, 9b and 9c, results of Cr(VI), F- and Fe estimation are plotted respectively. The results of Cr(VI) indicate elevated chromium contamination in Durgapur, Bankura and North 24 Paragnas. The WHO limit for Hexavalent Cr in water is about 0.5 mg/L whereas, in those districts, the contamination levels are close to that. The fluoride contamination is higher in the districts of Birbhum, Burdwan and Durgapur. But, reason for such result is likely to be the deposit of naturally fluoride containing rocks

43

in those areas. In general, the fluoride contamination levels

vary from 0 mg/L to 48 mg/L. Our result copes with those previous findings. The iron contamination problem is very prominent in these regions. The average iron concentration is found to be 10 mg/L whereas the WHO limit for safe drinking is about 1-1.5 mg/L. Tests are done to determine the Limit of Detection (LoD) or detection limit (DL) based on calibration curve for Cr(VI), F- and Fe estimation. The LoD values are calculated using the formula 44 given below: 𝐿𝑜𝐷 (𝑖𝑛 𝑝𝑝𝑚) =

3×𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑆𝑙𝑜𝑝𝑒 𝑜𝑓 𝑐𝑎𝑙𝑖𝑏𝑟𝑎𝑡𝑖𝑜𝑛 𝑐𝑢𝑟𝑣𝑒

…………. (10)

The LoD values are calculated for both our instrument and UV-3600 and given in figure 9d. The LoD values for both two instruments show very little difference of 0.002 ppm, 0.01 ppm and 0.002 ppm for Cr(VI), F- and Fe estimation respectively. To investigate the accuracy of the absorption spectra, we have prepared different concentrations of methyl orange dye (MO) and put them in same pH environment (pH 4.0). Then we have recorded absorption spectra of those MOs with our instrument. After obtaining those absorption data using “Spectruino” we have smoothed them using 10 point Fourier smoothing. The resulting spectra are given in figure 10a. As can be seen from the resulting spectra given in figure 10, the absorption maxima does not shift significantly with increasing concentration of MO. From literature values, it is observable that MO at pH 4, absorbs at about 464 nm

45

. Our result complies with those

literature values with an error of ±5 nm.

Figure 10: Absorption spectra of MO at different concentration recorded with our instrument is given in (a). Light sensitivity test results are given in (b).



Page 14 of 18

Page 13 of 17 The results of water analysis in several districts fall in the range of previously produced results by different researchers. All these outcomes finally point out that the method of analyzing contaminants is fairly accurate and feasible compared to the systems which are already present. The ambient light conditions are different in the field and to further investigate the ambient light sensitivity, we have prepared a standard 0.5 ppm fluoride solution and tested it in 4 different conditions. The light flux is measured with METRAVI 1332A light meter. The results of the test are given in figure 10b. As we can see from figure 10b, the results do not vary much with changing light conditions. So, it can be used during day or night without producing significant errors in detection. The measurement of uncertainty of the results are done and the experimental details are given in supporting information (S8). The total cost of the instrument is under INR 1500 (approx.) or US $20 (approx.).

4. Conclusions: In short, we can conclude that, we have successfully fabricated a cost-effective wireless, portable device that is calibrated for estimation of fluoride, hexavalent chromium and iron. And, the result obtained from our device is comparable with the results obtained from Shimadzu UV 3600. Thus, satisfying the need of the standard spectrophotometric systems in field, the ability of obtaining absorption spectra from the sample will also help to determine the absorption maxima of any unknown colored sample. This ability of the instrument will help one to carry out tests in remote areas to estimate the contamination levels of drinking water. Nevertheless, usage of smartphone as operating device makes the device easy to use and cheap. We hope that it will help a lot of people to know about the quality of drinking water and will create a good effect in human health.

Associated content Supporting Information Available The supporting information is available free of charge on the ACS Publications website. Different techniques developed by various groups, working principle of the desktop UI, process and steps of photometric measurements, the comparison of fits for Fe determination, total hardness analysis of drinking water samples, determination of total iron, measurement of uncertainty.

Author information *Corresponding author: [email protected]; [email protected] FAX: 2654-1123; Phone: 91 33 26549181


Page 15 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 14 of 17 Present address: §

Debmalya Santra, Dept. of Metallurgical Engineering & Materials Science, IIT Bombay.

§

Subhradeep Mandal, Rubber Technology Centre, IIT Kharagpur.

Notes The authors declare no competing financial interest.

Acknowledgement: The authors wish to thank Swami Shastrajnananda (Principal) and Swami Ekachittananda (Vice Principal) of Ramakrishna Mission Vidyamandira for their endless and enthusiastic support towards this project. The authors gratefully acknowledge Department of Science & Technology (DST), Govt. of India and Department of Higher Education, Govt. of West Bengal for providing financial assistance to carry out this work. The authors want to thank Prof. Kalyan Kumar Chattopadhyay, Dr. Nilesh Mazumder and Dr. Sumit Mandal for fruitful discussion. The authors also extend their gratitude towards Dr. Nirmalya Shankar Das and Mr. Soumyadip Mondal for helping us with the graphical contents presented in this paper.

References: 1.

India Groundwater: a Valuable but Diminishing Resource, http://www.worldbank.org/en/news/feature/2012/03/06/india-groundwater-critical-diminishing, (Accessed April 5, 2018).

2. Water scarcity and India, http://www.greencleanguide.com/2011/07/09/water-scarcity-and-india , (Acessed April 8, 2018). 3. GuhaMazumder, D. N.; Haque, R. ;Ghosh, N.; De B., K.; Santra, A.;Chakraborty, D.; Smith, A. H.; Santra, Amal; Chakraborty, D.; Int. J. Epidemiol. 1998, 27, 871-877. 4. Suthar, S.; G., Vinod K.; Jangir, S.;Kaur, S.; Goswami, N.; Environ. Monit. Assess. 2008, 145, 1-6 5. Mukhopadhyay, K.; Ghosh, A.; Das, S. K.; Show, B.; Sasilkumar P.; Ghosh, U. C.; RSC Adv. 2017, 7, 26037-26051. 6. Ayoob, S.; Gupta, A. K.; Critical Reviews in Environmental Science and Technology 2006, 36, 433-487. 7. Eastmond, D.A.; Macgregor, J.T.;Slesinski, R.S.; Crit. Rev. Toxicol. 2008, 38, 173-190.



Page 16 of 18

Page 15 of 17 8. Mohan, D.; Chander, S.; J. Hazard. Mater. 2006, 137, 762-811. 9. The Dark Side of Iron - Why Too Much is Harmful,https://www.healthline.com/nutrition/whytoo-much-iron-is-harmful#section4 (Accessed April 9, 2018). 10. WHO Guidelines for Drinking Water Quality, http://www.who.int/water_sanitation_health/dwq/2edaddvol2a.pdf (Accessed April 10, 2018). 11. Martin, J. H.; Knauer, G. A.; Nature 1982, 299, 611-612. 12. Puri, Avinash ; Kumar, Manoj; Indian Journal of Occupational and Environmental Medicine 2012, 16, 40-44. 13. Dakiky, M.; Khamis, M.; Manassra, A.; Adv. Environ. Res. 2002, 6, 533-540. 14. Gorchev, H. G.; Ozolins, G. M.; WHO chronicle. 2011, 38, 104-108. 15. Gordon, B.; Callan, P.; Vickers, C.; WHO chronicle. 2008, 38, 381-382. 16. March, Gregory; Nguyen, Tuan; Piro, Benoit.;Biosensors 2015, 5, 241-275. 17. Ozbek, N.; Akman, S.; Talanta. 2012, 94, 246-250. 18. Novell, M.; Parrilla, M.; Crespo, G. A.; Rius, F. X.; Andrade, F. J.; Anal. Chem. 2012, 84, 46954702. 19. Number of smartphone users in India from 2015 to 2022, https://www.statista.com/statistics/467163/forecast-of-smartphone-users-in-india/ (Acessed April 15, 2018). 20. Ding, H.; Chen, C.; Qi, S.; Han, C.; Yue, C.; Sens. Actuators, A. 2018, 274, 94-100. 21. Masawat, P.; Harfield, A.; Namwong, A.; Food Chem. 2015, 184, 23-29. 22. Lopez-Ruiz, N.; Curto, V. F.; Erenas, M. M.; Benito-Lopez, F.; Diamond, D.; Palma, A. J.; Capitan-Vallvey, L. F.; Anal. Chem. 2014, 86, 9554-9562. 23. Chen, X.; Chen, J.; Wang, F.; Xiang, X.; Luo, M.; Ji, X.; He, Z.; Biosens. Bioelectron. 2012, 35, 363-368. 24. Hossain, A.; Canning, J.; Ast, S.; J. Rutledge, P.; Li Yen, T.; Jamalipour, A.; IEEE Sens. J. 2015, 15, 5095-5102. 25. Bae, W. K.; Joo, J.; Padilha, L. A.; Won, J.; Lee, D. C.; Lin, Q.; Koh, W. K.; Luo, H.; Klimov, V. I.; Pietryga, J. M.; J. Am. Chem. Soc. 2012, 134, 20160-20168. 26. Light Sensors,https://www.electronics-tutorials.ws/io/io_4.html (Accessed April 2, 2018). 27. Swinehart, D. F.; J. Chem. Educ. 1962, 24, 8, 1349-1354. 28. Santra, D.; Santra, A.; Ghorai, U. K.; A portable method to determine concentration and absorption pattern of colored sample, Indian patent application filing number 201731037755. 29. Arduino Datasheet (Farnell),https://www.farnell.com/datasheets/1682209.pdf (Accessed April 3, 2018). 30. Arduino UNO Rev3,https://store.arduino.cc/usa/arduino-uno-rev3 (Accessed April 5, 2018). 31. Duty Cycle,https://en.wikipedia.org/wiki/Duty_cycle (Accessed April 5, 2018). 32. 1931 Color Space,https://en.wikipedia.org/wiki/CIE_1931_color_space (Accessed April 9, 2018).


Page 17 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 16 of 17 33. HC-05- Bluetooth Module,https://components101.com/wireless/hc-05-bluetooth-module (Accessed April 10, 2018). 34. ADS1115 Datasheet,http://www.ti.com/lit/ds/symlink/ads1115.pdf (Accessed April 12, 2018). 35. MIT Appinventor 2,http://ai2.appinventor.mit.edu/?locale=en#6006809374621696 (Accessed April 13, 2018). 36. Walsh, A. R.; O'Halloran, J.; Water Res. 1996, 30, 2393-2400. 37. Ottaviani, M.; Magnatti, P.; Dojmi di Delupis, G.; Mikrochim Acta. 1984, 84, 313-316. 38. Saywell, L. G.; Cunningham, B. B.; Anal. Chem. 1936, 9, 67-69. 39. Arduino ADS1115 Module Getting Started Tutorial,http://henrysbench.capnfatz.com/henrysbench/arduino-voltage-measurements/arduino-ads1115-module-getting-started-tutorial/ (Accessed April 12, 2018). 40. Signal-to-noise ratio,https://en.wikipedia.org/wiki/Signal-to-noise_ratio (Accessed April 15, 2018). 41. SHIMADZU UV-3600 Plus,https://www.shimadzu.com/an/molecular_spectro/uv/uv3600plus.html (Accessed April 19, 2018). 42. Kolthoff, I.M.; Stansby, M. E.; Anal. Chem. 1934, 6, 118-121. 43. Ghosh, A.; Mukherjee, K.; Ghosh, S. K.; Saha, B.; Res. Chem. Intermed. 2013, 39, 2881-2915. 44. Uhrovčík, J.; Talanta 2014, 119, 178-180. 45. Farahani, N.; Kelly, P.; West, G.; Hill, C.; Vishnyakov, V.; Coatings 2013, 3, 177-193.



Page 18 of 18

Page 17 of 17

For TOC only:


Cost-Effective, Wireless, Portable Device for Estimation of Hexavalent

Recommend Documents