Διερεύνηση εναλλακτικών μεθόδων επεξεργασίας πόσιμου νερού για την αποτροπή διάβρωσης σιδηροσωλήνων ύδρευσης

Καλτσά, Αλεξάνδρα; Kaltsa, Alexandra

dc.contributor.author	Καλτσά, Αλεξάνδρα	el
dc.contributor.author	Kaltsa, Alexandra	en
dc.date.accessioned	2020-12-30T07:47:31Z
dc.date.available	2020-12-30T07:47:31Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/52697
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.20395
dc.description	Εθνικό Μετσόβιο Πολυτεχνείο--Μεταπτυχιακή Εργασία. Διεπιστημονικό-Διατμηματικό Πρόγραμμα Μεταπτυχιακών Σπουδών (Δ.Π.Μ.Σ.) “Επιστήμη και Τεχνολογία Υδατικών Πόρων”
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	Πόσιμο νερό	el
dc.subject	Δίκτυα ύδρευσης	el
dc.subject	Διάβρωση	el
dc.subject	Μεταλλικά κουπόνια	el
dc.subject	Σιδηροσωλήνας	el
dc.subject	Drinking water	en
dc.subject	Corrosion	en
dc.subject	Distribution systems	en
dc.subject	Coupons	en
dc.subject	Iron pipe	en
dc.title	Διερεύνηση εναλλακτικών μεθόδων επεξεργασίας πόσιμου νερού για την αποτροπή διάβρωσης σιδηροσωλήνων ύδρευσης	el
dc.title	Investigation of alternative methods of drinking water treatment to prevent corrosion of iron pipe distribution systems	en
heal.type	masterThesis
heal.classification	Διαχείριση υδατικών πόρων	el
heal.classification	Water resources management	en
heal.language	el
heal.access	campus
heal.recordProvider	ntua	el
heal.publicationDate	2020-10-30
heal.abstract	Η εξασφάλιση καθαρού νερού αποτελεί μία διαχρονική ανάγκη παγκοσμίως. Για την καλή προστασία των υδατικών πόρων και την αποδοτική τους χρήση απαιτείται η ορθή διαχείρισή τους. Η διαχείριση των υδατικών πόρων συνδέεται άμεσα και με την διαχείριση των δικτύων ύδρευσης η οποία αποτελεί ένα από τα σημαντικότερα πεδία μελέτης τόσο στις αναπτυγμένες όσο και στις αναπτυσσόμενες χώρες. Ο βασικότερος παράγοντας στην ορθολογική διαχείριση των δικτύων ύδρευσης είναι το ίδιο το νερό, που διατρέχει το δίκτυο, το οποίο πρέπει να έχει συγκεκριμένα ποιοτικά και ποσοτικά χαρακτηριστικά. Ένα μεγάλο μέρος του ρυθμού της διάβρωσης του δικτύου οφείλεται στην αντίδραση αυτού του διερχόμενου νερού με τα υλικά των σωληνώσεων. Οι ουσίες που διαλύονται στο νερό έχουν σημαντική επίδραση τόσο στην διάβρωση όσο και στις μεθόδους για τον έλεγχό της. Τα μέταλλα και τα κράματά τους, έχουν την ικανότητα να διαβρώνονται καθώς ενώνονται με διάφορα άλλα στοιχεία με τα οποία σχηματίζουν άλλες πιο σταθερές ενώσεις. Τα πρώτα, όντας αναγόμενα στην μεταλλική τους μορφή είναι και ενεργειακά αναβαθμισμένα σε σύγκριση με την πρώτη ύλη τους, επομένως έχουν την τάση να επιστρέψουν στην αρχική και σταθερή οξειδωμένη μορφή τους, η οποία είναι σε χαμηλότερη ενεργειακή στάθμη, σύμφωνα με τον δεύτερο θερμοδυναμικό νόμο. Σε αυτή την διαδικασία συμβάλει η ένωσή τους με το οξυγόνο που τα βοηθάει να οξειδωθούν και να μετατραπούν σε οξείδια ή σε άλλες ενώσεις από τις οποίες ενδεχομένως προήλθαν, με ταυτόχρονη απελευθέρωση θερμότητας. Αυτή η τάση των μετάλλων να επανέλθουν στην αρχική τους κατάσταση, είναι και το κύριο αίτιο της διάβρωσης. Η παρούσα διπλωματική εργασία έχει ως στόχο τη διερεύνηση εναλλακτικών μεθόδων επεξεργασίας του πόσιμου νερού για την αποφυγή της διάβρωσης ενός δικτύου ύδρευσης. Η εργασία που πραγματοποιήθηκε αφορά ένα σύστημα ύδρευσης το οποίο τα τελευταία χρόνια τροφοδοτείται από υπόγειο νερό με συγκεκριμένη σύσταση και θεωρείται μη διαβρωτικό. Στην μελέτη αυτή εξετάζεται η αντικατάσταση του υπόγειου νερού με νερό από επιφανειακές πηγές. Ένας από τους λόγους που μπορεί να είναι αναγκαία αυτή η στροφή από υπόγειες πηγές σε επιφανειακές είναι τόσο το οικονομικό (π.χ. κόστος άντλησης) όσο και το περιβαλλοντικό κόστος. Συχνά τα επιφανειακά νερά, λόγω της φυσικοχημικής τους σύνθεσης, παρουσιάζουν διαβρωτικά χαρακτηριστικά που οδηγούν σε ανεπιθύμητες επιπτώσεις, όπως είναι το χρώμα, η γεύση και η οσμή του νερού ή/και αυξημένες συγκεντρώσεις διαλυτών μετάλλων. Αρκετές από τις παραμέτρους που επηρεάζουν την διάβρωση είναι στενά συνδεδεμένες μεταξύ τους καθώς μια ενδεχόμενη μεταβολή σε μία από αυτές μπορεί να φέρει μεταβολή σε μια άλλη. Σημαντικό παράδειγμα τέτοιας σχέσης είναι μεταξύ του pH, της αλκαλικότητας καθώς και της αναλογίας των ιόντων Cl- και SO4-2. Οι Larson & Skold (1958) ανακάλυψαν ότι η αναλογία του αθροίσματος των Cl- και SO4-2 προς την αλκαλικότητα, η οποία αργότερα ονομάστηκε δείκτης Larson, ήταν σημαντική καθώς μπορούσε να υποδείξει την χαμηλή ή υψηλή διαβρωτικότητα του νερού. Νερά με δείκτη Larson–Skold μεγαλύτερο από 1.2 δείχνουν μια τάση προς υψηλούς ρυθμούς διάβρωσης ενώ εκείνα με τιμές μικρότερες από 0.8 δεν παρουσιάζουν διαβρωτικό χαρακτήρα. Στην παρούσα εργασία διερευνάται η συμβολή αυτών των διαφορετικών φυσικοχημικών χαρακτηριστικών στο δυναμικό διάβρωσης και στη συνέχεια μελετώνται οι πιθανές μέθοδοι επεξεργασίας των νερών ώστε να μην παρατηρούνται φαινόμενα διάβρωσης. Πιο συγκεκριμένα, διερευνήθηκε η επίδραση της αλκαλικότητας μέσω της προσθήκης διαφορετικών χημικών, η επίδραση του pH καθώς και η επίδραση των διαφορετικών συγκεντρώσεων χλωριόντων (Cl-) και θειικών ιόντων (SO4-2). Επιπλέον, εξετάστηκε και η περίπτωση ανάμιξης διαφορετικών αναλογιών υπόγειου και επιφανειακού νερού ως μια θεραπευτική λύση κατά της διάβρωσης. Για την διεκπεραίωση των πειραμάτων χρησιμοποιήθηκαν μικρού μεγέθους αντιπροσωπευτικά μεταλλικά διαβρωμένα δείγματα από τους υπό εξέταση σιδηροσωλήνες ύδρευσης (κουπόνια) ώστε να εφαρμοστεί η μέθοδος απώλειας βάρους κουπονιών σε στάσιμο νερό (coupon tests in stagnant water). Αρχικά, οι αναλύσεις και οι δοκιμές έγιναν με πραγματικά υπόγεια και επιφανειακά νερά όμως στη συνέχεια για τη διεκπεραίωση των υπόλοιπων πειραμάτων και λόγω της χρήσης μεγάλης ποσότητας νερού ήταν απαραίτητο να παρασκευάζονται συνθετικά νερά, υπόγεια και επιφανειακά, με τη χρήση χημικών ουσιών σε κατάλληλες ποσότητες ώστε να προσομοιώνουν τα φυσικοχημικά χαρακτηριστικά των πραγματικών νερών. Στην παρούσα διπλωματική εργασία η πειραματική περίοδος διήρκησε από τον Οκτώβριο του 2019 μέχρι τον Μάρτιο του 2020. Αρχικός στόχος ήταν να μελετηθεί η συμπεριφορά των συνθετικών νερών συγκριτικά με τα πραγματικά. Από την σύγκριση των ποσοστιαίων απωλειών βάρους στα κουπόνια, παρατηρήθηκε πως τα πραγματικά και τα συνθετικά νερά έδωσαν παρόμοια αποτελέσματα και έτσι στα υπόλοιπα πειράματα που πραγματοποιήθηκαν έγινε χρήση μόνο συνθετικών νερών. Επίσης, εξετάστηκε η επίδραση του pH αλλά και της μεταβολής της αλκαλικότητας, στην διάβρωση, με προσθήκη των χημικών NaHCO3 και NaOH των οποίων οι δόσεις υπολογίστηκαν με την βοήθεια του δείκτη Larson. Η προσθήκη χημικών έγινε με σκοπό να αυξηθεί η αλκαλικότητα των δειγμάτων. Τα δείγματα που είχαν υποστεί χημική επεξεργασία συγκριτικά με εκείνα χωρίς επιπρόσθετη επεξεργασία, είχαν χαμηλά ποσοστά απώλειας βάρους και χαμηλούς δείκτες διαβρωτικότητας κι αυτό διότι η αύξηση της αλκαλικότητας λειτούργησε ως αναστολέας της διάβρωσης με αποτέλεσμα να προκύψουν χαμηλοί δείκτες Larson (0.8, 0.95 κλπ.) καθώς και χαμηλά ποσοστά απώλειας βάρους (0,48%, 0,61%, 0,52% κλπ.). Επιπλέον, μελετήθηκε η επίδραση διαφορετικών συνδυασμών της αναλογίας των χλωριόντων και των θειικών ιόντων σε σταθερή αλκαλικότητα. Από την μελέτη των ποσοστιαίων απωλειών βάρους προέκυψε ότι όσο οι συγκεντρώσεις των Cl- αυξάνονταν τόσο αυξανόταν και η απώλεια βάρους. Ενδεικτικά, για σταθερή αλκαλικότητα ίση με 174mg/l και συγκεντρώσεις χλωριόντων ίσες με 120mg/l οι απώλειες βάρους έφτασαν σε ποσοστά 1,82%, 1,77% και 1,80% ενώ αντίστοιχα οι δείκτες Larson αυξήθηκαν κι αυτοί. Επιπλέον, παρατηρήθηκε πως τα θειικά ιόντα δεν επηρέασαν σημαντικά την απώλεια βάρους, καθώς ακόμα και σε αυξημένες συγκεντρώσεις (120, 180mg/l), με σταθερή αλκαλικότητα στα 174 mg/l και συγκεντρώσεις χλωριόντων στα 40 και 80 mg/l, οι ποσοστιαίες απώλειες βάρους προέκυψαν χαμηλές ( 0,76%, 0,99% κλπ.). Τέλος, μελετήθηκε η επίδραση διαφορετικών αναλογιών υπόγειου και επιφανειακού νερού, στην διάβρωση των κουπονιών. Από τα αποτελέσματα των ποσοστιαίων απωλειών βάρους των διαφορετικών προσμίξεων παρατηρήθηκε ότι στις προσμίξεις 100:0 και 75:25 (υπόγειο : επιφανειακό) επικρατούσαν υψηλά επίπεδα αλκαλικότητας και χαμηλά επίπεδα ιόντων. Αυτό είχε ως αποτέλεσμα να προκύψουν χαμηλά ποσοστά απώλειας βάρους (0,41%. 047%) και χαμηλοί δείκτες Larson (0.19, 0.48). Αντιθέτως, οι προσμίξεις 0/100, 25/75 και 50/50 (υπόγειο: επιφανειακό) χαρακτηρίστηκαν από υψηλά επίπεδα αλκαλικότητας αλλά και υψηλές συγκεντρώσεις χλωριόντων (80mg/l). Αυτό είχε ως αποτέλεσμα, τα ποσοστά απώλειας βάρους να αυξηθούν (0,99%, 0,97%) και να προκύψουν δείκτες Larson υψηλοί (1.37, 1,0, 0,72). Κλείνοντας, από τα αποτελέσματα της παρούσας μελέτης βγήκε το συμπέρασμα ότι οι παράγοντες που παίζουν σπουδαίο ρόλο στο δυναμικό της διάβρωσης είναι κατά κύριο λόγο η αλκαλικότητα και οι συγκεντρώσεις των χλωριόντων. Οι αυξημένες συγκεντρώσεις των χλωριόντων λειτουργούν ως επιταχυντές της διάβρωσης ενώ αντίθετα, υψηλά επίπεδα αλκαλικότητας λειτουργούν ως αναστολείς. Μία λύση στην θεραπεία των φαινομένων διάβρωσης, αποτελεί η χημική επεξεργασία με προσθήκη χημικών διαλυμάτων, ώστε να αυξηθούν οι συγκεντρώσεις της αλκαλικότητας ή η ανάμιξη με νερά τα οποία έχουν υψηλά επίπεδα αλκαλικότητας και χαρακτηρίζονται μη διαβρωτικά. Τα θειικά ιόντα δεν επηρεάζουν σημαντικά το ρυθμό της διάβρωσης καθώς είναι λιγότερο δραστικά και διαβρωτικά από τα χλωριόντα. Τέλος, το pH θα πρέπει να διατηρείται σταθερό σε ένα εύρος τιμών από 7.2-7.7, διότι τιμές χαμηλότερες σε συνδυασμό με τους υπόλοιπους παράγοντες συμβάλουν στο φαινόμενο της διάβρωσης.	el
heal.abstract	Introduction The most important management factor of a water distribution system is the water itself, which must have specific qualitative and quantitative characteristics. Corrosion of water distribution system pipelines is considered a problem of major importance and must be treated by the managers of these systems. Corrosion is defined as the destruction, wear and/or malfunction of a material caused by chemical, electrochemical or mechanical interactions with its surroundings. Water passing through the system has the ability to affect the corrosion of pipelines, through its physicochemical properties. Many factors affect corrosion rate and must be taken under consideration such as water flow, pH, alkalinity and chlorides and sulfates ions ratio. The scope of this thesis is the investigation of alternative methods regarding the treatment of drinking water in order to avoid corrosion of water distribution system. More specifically, it refers to the contribution of different physicochemical characteristics to corrosion dynamic and the possible methods of water treatment in such a way to avoid any corrosion occurrence even after alteration of water supply (from groundwater to surface water). The main reason causing the change from groundwater to surface water is the cost, which is not only financial (abstraction expenses) but also environmental. Surface water due to its physicochemical composition feature corrosive characteristics that lead to undesirable results (color, taste, odor and/or increased concentrations of dissolved metals). Furthermore, this study examines the behavior of alkalinity when adding different chemical solutions and the effect of pH and of the concentrations of chloride (Cl-) and sulfate (SO4-2) ions to the corrosion rate. Finally, yet importantly, cases of different mixtures of groundwater and surface water were examined as a possible treatment to the negative consequences of corrosion. Experimental details A series of experiments were performed using the method of coupon tests in stagnant water. The coupons used during the experiments were parts of corroded galvanized iron pipe taken as a sample from an actual water distribution system. The water samples that were used on the experiments were groundwater and surface water (actual and synthetic). Synthetic water was prepared through adding the appropriate doses of chemical solutions to obtain the same chemical composition with the actual water. The experiments conducted were aiming in analyzing and comparing the behavior between synthetic and actual water, examining the influence of changing pH and alkalinity levels by adding chemical solutions with doses that were calculated according to the Larson Index (LI) and in determining the impact of different chloride and sulfate ions ratio combinations. Finally, another goal of the experiments performed was the examination of various mixtures of groundwater and surface water and their contribution in coupon corrosion. The cleaning process of coupons was performed according to the regulations of ASTM and G1 method of Standard Practice for Preparing, Cleaning and Evaluating Corrosion Tests Specimens (ASTM, 1999). Upon completion of the cleaning process the coupons free from corrosion byproducts, were weighed in order to determine the initial weight. At the end of each experimental cycle, the coupons were weighed in order to determine the final weight and the percentage of weight loss. Every 24 hours the water samples were examined according to: a) the percentage of transmittance (T%) on three wavelengths (700, 635, 590 nm) and b) the pH. Upon accomplishment of the above, the water samples were renewed for the next 24 hours. The duration of each experimental cycle was 40 days. During the experiments, actual groundwater and actual surface water were used with characteristics shown in table 1. Due to the use of great water amount, it was necessary to produce synthetic water samples with identical physicochemical characteristics to the actual ones. The actual and synthetic water samples were compared in order to ensure the same quality and the same behavior towards corrosion dynamics. The characteristics in Table 1 represent the annual average of the actual water quality. Analysis of these annual values indicated two seasonal discrete qualitative profiles of the actual surface water, during the dry (Season 1) and the wet (Season 2) period. Based on the above, a synthetic surface water of the dry period was prepared (Season 1) and was examined as it represents the period with the most critical water traits in comparison with the corrosion dynamics. The qualitative characteristics of the synthetic surface water during dry period (Season 1) are presented in Table 2. As seen in Table 1 the actual groundwater is non corrosive (LI=0.19) compared to the actual surface water (LI=1.80) while as seen in Table 2, synthetic surface water of Season 1, appear to have an intense corrosive character since Larson Index is equal to 1.37. Water with Larson Index lower than 0.8 considered to be non-corrosive while water with Larson Index higher than 1.2 is considered extremely corrosive (Agatemor & Okolo, 2008; Vasconcelos et al., 2015). Table 1: Αnnual average of the actual water quality (groundwater, surface water). *N.D. = Non Detected LOD = Limit Of Detection Table 2: Qualitative characteristics of synthetic surface water of dry season (Season 1). Description of Experiments a) Actual and Synthetic water comparison experiment (untreated) The purpose of this experiment was to study whether the synthetic water samples (groundwater and surface water) prepared in the laboratory had the same qualitative characteristics (Table 1) and the same behavior with the actual ones (groundwater and surface water). These specific water samples were compared without any additional treatment. b) Experiment to study the comparison between actual and synthetic corrosive surface water with treatment and Larson Index 0.8 Actual and synthetic surface water, with the annual qualitative characteristics, were used to perform this experiment. The purpose of this study was to compare actual and synthetic surface water with additional treatment and to examine the effectiveness of this treatment to the reduction of corrosion. Surface water samples with an LI of 1.8 (as seen in Table 1) are considered extremely corrosive. The main purpose was to treat surface water samples, in order for this index to reach 0.8 by changing only the alkalinity and to ensure that both actual and synthetic surface water have the same behavior. The treatment involved the addition of NaHCO3 with an initial concentration of 2.3 g/l. The chemical solution was added in doses suitable to increase the alkalinity of these water samples. c) Experiment to study the synthetic surface water during dry period (Season 1) with chemical treatment and Larson Index 0.8. Synthetic surface water during dry period (Season 1) was used to perform this experiment. Season 1 represented the period with the worst and most critical water characteristics in relation to its corrosion potential. The Larson Index of the synthetic surface water samples of dry period (Season 1) is 1.37, as seen in Table 2, and so it is considered extremely corrosive. The main purpose of this study was to treat this water samples in order for this index to reach 0.8 by changing only the alkalinity. The treatment involved the addition of two chemicals, NaHCO3 and NaOH with an initial concentration of 2.3 g/l and 2 g/l, respectively, in specific doses to increase the alkalinity of the samples. Furthermore, these chemically treated samples of synthetic surface water of dry period (Season 1) were compared with synthetic surface water samples (with the annual qualitative characteristics, Table 1) with the addition of NaHCO3 and with synthetic surface water samples (with the annual qualitative characteristics, Table 1) without any additional treatment. d) Experiment to study the synthetic surface water during dry period (Season 1) with chemical treatment and Larson Indexes 0.95 and 1.0. Synthetic surface water during dry period (Season 1) was used to perform this experiment. The aim was to study different alkalinity concentrations by adding appropriate doses of chemical solutions. The change in alkalinity would also lead to a change in Larson Index from 1.37 to 0.95 and 1.0. The treatment involved the addition of NaHCO3 and NaOH with an initial concentration of 2.3 g/l and 2 g/l, respectively, in different doses to change the alkalinity of these water samples and reduce the corrosiveness of the water. Furthermore, synthetic surface water during dry period (Season 1) samples with the different Larson Indexes (0.8, 0.95 and 1.0) were compared. A synopsis of the experiments performed and Larson Indexes are stated in table 3. Table 3: Record of experiments and Larson Indexes Study of critical chemical parameters In these experiments was studied the criticality of some chemical parameters at the rate of corrosion. For these experiments were used only synthetic water samples, since the experiments of comparison with the actual ones had preceded and their similarity was successfully established a) Experiment to study the effect of different levels of chloride and sulfate ions of synthetic surface water of dry season (season 1), on the rate of corrosion Synthetic surface water of dry period (Season 1) was used with and without chloride and sulfate ions. The aim of this experiment was to study the effect of changing the levels of chloride (Cl-) and sulfate (SO4-2) ions, on the corrosion rate. Synthetic surface water of dry period (Season 1), containing chloride and sulfate ions in a ratio 81:120 mg/l (Table 9), was the control sample, while other synthetic surface water of dry period (Season 1) were prepared with different ion ratios. Chloride levels were studied and categorized as low, medium and high and are shown in Table 4. b) Experiment to examine the effect of different pH values of synthetic surface water of dry season (Season 1), on the corrosion rate In this experiment was used synthetic surface water of dry season (Season 1) with chloride and sulfate ions in a ratio of 81:120 (control sample) and pH constant at 7.6. The aim of this experiment was to study the effect of pH in corrosion rate. Also, the comparison between the pH of control sample (pH = 7.6) and two different pH of Season 1 (pH = 7.0 and pH = 8.0) was studied. In this case, the pH of Season 1 equal to 7.0 was adjusted by adding CO2 while in the Season 1 water sample with pH equal to 8.0 was added a little dose of caustic soda (NaOH). The lowest value of the selected pH (7.0) was based on the annual pH of non-corrosive groundwater, while the highest value (8.0) was based on the Langelier index (LSI). Specifically, the average LSI of corrosive surface water is 0.03 with a calculated pH of 7.5. Based on the equation LSI = pH-pHs the maximum pH of the water that was accepted in order to reach an LSI index <0.5 was 8.0. The Langelier Index (LSI) can take positive, negative or even zero values. More specifically when: • LSI = 0, water is considered neutral or in chemical equilibrium. • LSI <0, water tends to become corrosive • LSI> 0, water is considered to be supersaturated with respect to calcium carbonate (CaCO3) with possible membrane formation while at the same time showing a tendency to form rust (USEPA, 1984; Laurie et al, 2001., Vasconcelos et al., 2015). c) Experiment to study the effect of mixing different ratios of synthetic groundwater and synthetic surface water of dry season (Season 1), to corrosion rate. The aim of this experiment was to study whether there is an improvement in the corrosion rate if different ratios of non-corrosive and corrosive water are mixed. For the need of this experiment, synthetic (non-corrosive) groundwater and synthetic (corrosive) surface water of dry period (Season 1) were used. The combinations between synthetic non-corrosive groundwater and corrosive Season 1 water are shown in Table 5. As mentioned, the synthetic groundwater is characterized as non-corrosive as it has Larson Index equal to 0.19 in contrast to Season 1 water, which has Larson Index equal to 1. Table 4: Presentation of different levels of chlorides and ratios of chloride and sulfate ions. Table 5: Presentation of groundwater and surface water ratios. Results & Discussion a) Actual and Synthetic water comparison experiment (untreated) The results shown in Figure 1 were obtained by comparing the percentage weight losses of actual and synthetic water in both groundwater and surface water, with the annual average of qualitative characteristics. Figure 1: Percentage of weight loss of actual and synthetic water over 40 days. Coupons immersed in groundwater (actual and synthetic) gave a weight loss of 0.3 ± 0.04% and 0.4 ± 0.11%, respectively, while those immersed in surface water (actual and synthetic) gave a percentage 0.89 ± 0.1% and 0.97 ± 0.03%, respectively. It is obvious that water with higher corrosive character creates greater alterations in the metal coupon and therefore more material is lost in the form of corrosion byproducts. From the comparison between the actual and synthetic groundwater as well as the actual and synthetic surface water, no difference of more than 0.37% and 0.08%, respectively, was noted. In addition, the transmittance (T%) of the water samples at wavelengths of 700, 635 and 590nm was determined in all experiments. The study of the color of red water did not yield safe conclusions. "Red water" or colored water is water with high turbidity, which may come from sources that contain iron and have not been properly treated or due to corrosion of the iron pipes through which drinking water passes (AWWA, 2002). One reason for the presence of red water is the stagnant conditions that prevailed in these experiments, as it has been shown that even in a non-corrosive water the iron concentration will increase more rapidly as the stagnation time in the pipes increases (Ministry of Health, Canada, 2009). The phenomenon of nucleation is another important reason for the presence of red water. Due to this phenomenon in the most corrosive water, the corrosion products that are created have the ability to form a nucleus on the metal surface and then the new corrosion by-products are deposited on it. Thus, water is observed to be purer because the corrosion by-products cannot escape into the water as they are attached to the formed nuclei. In non-corrosive water, on the other hand, these nuclei do not form quickly, which is why water continues to be redder for longer as oxides can escape (Ansari et al., 2019). Furthermore, the methodology of these experiments states that the conclusions come from the study of weight loss of coupons and not from the examination of the color of water samples (Sander A. et al., 1997., ASTM Committee, 1999). Therefore, due to the above, the study of the color of red water cannot give safe results and so these measurements are not reliable to draw conclusions. b) Experiment to study the comparison between actual and synthetic surface corrosive water with treatment and Larson Index 0.8 The purpose of this study was to compare actual and synthetic surface water, with the annual qualitative characteristics, with additional chemical treatment. The treatment involved the addition of NaHCO3 chemical solution with initial concentration 2.3 g /l, in an appropriate quantity to alter the alkalinity of the samples, and therefore the Larson Index. Surface water has an LI of 1.8 and is characterized by a very high corrosion ability. The main purpose was to process actual and synthetic surface water appropriately in order to reach an LI of 0.8 by changing only the alkalinity and to compare them in order to be sure that they behave the same. The results shown in Figure 2 were obtained by comparing the percentage weight losses of actual and synthetic surface water by chemical treatment. Figure 2: Presentation of the comparison of weight loss (%) of actual and synthetic surface water, with annual qualitative traits, with and without treatment, over 40 days. Figure 2 shows the percentage of weight loss of actual and synthetic surface water with and without chemical treatment. Coupons immersed in the actual corrosive surface water without treatment gave a weight loss of 0.89 ± 0.1% while those immersed in the actual corrosive surface water with the addition of NaHCO3 gave a percentage of 0.47 ± 0.02%. Respectively, the coupons immersed in the synthetic corrosive surface water without treatment had a weight loss of 0.97 ± 0.03% while the same water with the addition of the chemical solution NaHCO3 gave a percentage of 0.48 ± 0.04%. It is obvious that the surface water samples without chemical treatment (LI = 1.8) have a strongly corrosive character in contrast to those who were chemically treated. The result of the chemical treatment was the change in alkalinity and therefore the Larson index has been adjusted (LI = 0.8). c) Experiment to study the synthetic surface water during dry period (Season 1) with chemical treatment and Larson Index 0.8. Synthetic surface water during dry period (Season 1) was used to perform this experiment Season 1 represents the period of surface water with the worst water characteristics in relation to its corrosion dynamic (dry season). The Larson Index for Season 1 has been reported to be 1.37, which probably gives the water a highly corrosive character. The aim was to study the effectiveness of adding alkalinity as a method to reduce the corrosive potential of water. The most commonly used chemicals to add alkalinity to drinking water include NaHCO3 and NaOH. Water treatment, in the present experiment, involved the addition of NaHCO3 and NaOH with an initial concentration of 2.3 g/l and 2 g/l respectively, to change the alkalinity of the samples so that the Larson Index will be 0.8. Furthermore, these chemically treated samples of synthetic surface water of dry period (Season 1) were compared with synthetic surface water samples, with the annual average of qualitative characteristics, with the addition of NaHCO3 and with synthetic surface water samples, with the annual average of qualitative characteristics, without any additional treatment. The results shown in Figure 3 shows the data obtained by comparing the percentage of weight losses of the different water conditions. Figure 3: Comparison of weight loss (%) of Season 1 with chemical treatment (NaHCO3, NaOH) and Synthetic Surface water, with annual qualitative traits, with and without chemical treatment, over 40 days. As shown in Figure 3, when NaOH was added in the synthetic surface water of dry period (Season 1), a weight loss of 0.52 ± 0.03% was achieved, while the addition of NaHCO3 in Season 1, gave a weight loss of 0.55 ± 0.03%. As far as the synthetic surface water, with the annual average of qualitative characteristics and the addition of NaHCO3 is concerned, the weight loss of these coupons had no significant difference according to their comparison to synthetic surface water of dry season (Season 1) and the addition of the same chemical (NaHCO3), as small differences between them were noted. In particular, for the first one a weight loss of 0.48 ± 0.04% was calculated while for the second one the weight loss was 0.55 ± 0.03%. As far as the synthetic surface water with the annual average of qualitative characteristics and without any chemical treatment is concerned, it is obvious that the synthetic water without the addition of no chemical shows a weight loss of 0.97 ± 0.03%, compared to Season 1 with the addition of NaHCO3 which gave a weight loss of 0.55 ± 0.03% and Season 1 with the addition of NaOH which gave a weight loss of 0.52 ± 0.03%. In addition, the weight loss of the coupons of synthetic surface water without chemical treatment, is comparatively higher than those of the synthetic surface water with the annual average of qualitative characteristics and the addition of NaHCO3. d) Experiment to study the synthetic surface water during dry period (Season 1) with chemical treatment and Larson Indexes 0.95 and 1.0. The aim was to study the weight loss of coupons at different alkalinity concentrations by adding appropriate amounts of chemical solutions. The change in alkalinity would also lead to a change in the Larson Index from 1.37 to 0.95 and 1.0. The treatment included the addition of chemicals NaHCO3 and NaOH with an initial concentration of 2.3 g/l and 2 g/l respectively. The results shown in Figure 4 were obtained by comparing the percentage of the weight losses between the different water conditions. Figure 4: Comparison of weight losses (%) of synthetic surface water of dry season (Season 1) with LI = 0.8 and LI = 0.95 and 1.0, with addition of NaOH and NaHCO3 respectively, over 40 days. As shown in Figure 4, water with Larson index 1.0 and addition of NaOH, being more corrosive, caused the metal coupons to have a higher weight loss equal to 0.61 ± 0.03% compared to the one with Larson index 0.8, the weight loss of which observed equal to 0.52 ± 0.03%. In contrast, water with Larson index 0.8 and addition of the chemical NaHCO3 gave a weight loss equal to 0.55 ± 0.03% while the one with Larson index 0.95 gave 0.59 ± 0.04%. e) Experiment to study the effect of different levels of chloride and sulfate ions of synthetic surface water of dry season (season 1), on the rate of corrosion The results shown in Figure 5 were obtained by comparing the percentage of weight losses with different ratios of chlorides and sulfates. Figure 5: Comparison of different concentrations of chloride and sulfate ions with the percentage of weight loss of coupons. Figure 5 shows the percentage of weight loss of coupons when changing water quality by adding three different ratio levels of chloride to sulfate ions: low (40:60, 40:120, 40:180 mg/l), medium (80:60, 80:120 mg/l) and high (120:60, 120:120, 120:180 mg/l). The samples with the different ratio levels were compared, separately, to the control sample with an initial ratio of 81:120 mg/l. Coupons immersed in the control sample with a chloride to sulfate ion ratio of 81:120 mg/l presented a weight loss of 0.99 ± 0.08%. Coupons placed at the three low levels of chloride to sulfate ions (40:60, 40:120, 40:180 mg/l) yielded a weight loss of 0.73 ± 0.06%, 0.75 ± 0.04% and 0.77 ± 0.04% respectively. Those immersed in water samples with the two medium ion levels (80:60, 80:180 mg/l) yielded a weight loss of 0.98 ± 0.04% and 1.09 ± 0.03% respectively. Finally, the coupons placed at the three different high ion levels (120:60, 120:120, 120:180 mg/l) yielded a weight loss of 1.81 ± 0.07%, 1.74 ± 0.14% and 1.8 ± 0.12% respectively. As the levels of chlorides in drinking water increase, being more active than sulfates ions, the corrosion index of water (Larson index), which implies its greater corrosive ability, also increases (USEPA, 1984., Babaeia AA et al., 2018, Tavanpour et al., 2016). In contrast, sulfate ions do not appear to play a significant role in the rate of corrosion. More specifically, at low chloride concentrations (40mg/l) it is observed that although sulfate concentrations increase (60, 120, 180 mg/l) this does not affect weight loss (0.75 ± 0 04%). In addition, looking at the ratio 120:60 mg/l it seems that although the concentration of sulfate ions is much lower compared to chlorides, the resulting weight loss is very high (1.82 ± 0.07%). f) Experiment to examine the effect of different pH values of synthetic surface water of dry season (Season 1), on the corrosion rate The results shown in Figure 6 were obtained by comparing the percentage of weight losses of the different pH values. Figure 6: Comparison of the percentage of weight losses of the two different pH values at 7.0 and 8.0 with the control sample at pH = 7.6. Figure 6 states that the coupon samples at pH 7.0 showed a weight loss with an average of 1.39 ± 0.1% while the coupons at pH 8.0 and 7.6 recorded a weight loss of 0.95 ± 0.01% and 0.99 ± 0.08%, respectively. Some studies have found that in the pH range from 7 to 9, both the rate of corrosion and the degree of iron tuberculation in water systems increased as the pH increased, while in another study it was found that iron concentrations decreased steadily as the pH increased from 7.6 to 9.5 (Sarin et al., 2003). In the present case, it was observed that the corroded coupons, which were placed in water with values of 7.6 and 8.0, gave similar weight loss while on the contrary those placed in water with value of pH 7.0 gave much higher weight loss. g) Experiment to study the effect of mixing different ratios of synthetic groundwater and synthetic surface water of dry season (Season 1), to corrosion rate. Comparing the percentage of weight losses of different mixing ratios of synthetic groundwater and Season 1, the results shown in Figure 7 were obtained. Figure 7: Comparison of the percentage of weight losses between three different mixes of synthetic non-corrosive groundwater and corrosive surface water of dry period (Season 1) 50:50, 25:75 and 75:25 related to the control sample, over 40 days. Figure 7 states the percentage of weight losses of the three different ratios 50:50, 25:75, 75:25 (synthetic groundwater: Season 1). Coupons immersed in the water ratio 50:50 resulted to a weight loss of 0.98 ± 0.07% while in the ratio 25:75 there was a weight loss of 0.97 ± 0.1%. Accordingly, the coupons of the control sample were characterized by a weight loss of 0.99 ± 0.08%. On the contrary, the ratio of synthetic non-corrosive groundwater and corrosive surface water of Season 1 75:25, as expected, gave less weight loss to the metal coupons with a percentage of 0.47 ± 0.03%, since most of the water used was non-corrosive. The new water resulting from the 75:25 (groundwater: Season 1) mixture is characterized by higher alkalinity concentrations (245.25 mg/l) and lower chlorides concentrations (45.7 mg/l) which also leads to a lower Larson index (LI = 0.48), compared to the other two ratios and the control sample. Conclusions All the experiments that took place during this experimental period are shown in Figure 8, in order to have a complete insight of which is the most crucial factor affecting the corrosion rate. Figure 8: Presentation of weight loss (%), Larson indexes, chloride and sulfate ions and alkalinity concentrations for each experiment separately. From the experiments of different ratios of chloride and sulfate ions it appears that at constant alkalinity, as Cl- concentrations increase so weight loss do. More specifically, for constant alkalinity equal to 174 mg/l and chloride concentrations equal to 120 mg/l the weight loss reaches percentages of 1.82%, 1.77% and 1.80% while respectively the Larson indexes also increase, as water turns out more corrosive. In addition, at the same constant alkalinity of 174 mg/l and chloride concentrations of 40 and 80 mg/l, it is observed that sulfate ions do not significantly affect weight loss even at their increased concentrations (120, 180mg/l). The respective weight losses (%) have low percentage rates (0.76%, 0.99% etc.). From the results of the various mixtures of non-corrosive synthetic groundwater and corrosive synthetic surface water of dry period (Season 1) it is observed that in the ratios 100:0 and 75:25 high levels of alkalinity and low levels of ions prevail. This results in low weight loss rates (0.41%, 047%) and low Larson Indexes (0.19, 0.48). On the contrary, the ratios 0:100, 25:75 and 50:50 are characterized by high levels of alkalinity but also high concentrations of chlorides (60-80 mg/l). As a result, weight loss rates increase (0.99%, 0.97%) and Larson indexes are high (1.37, 1.0, 0.72). Finally, observing the results for the samples that had been chemically treated with NaOH and NaHCO3 , one can see that compared to the synthetic surface water with the annual average of qualitative characteristics and without additional treatment, all the rest resulted in low weight loss rates and low Larson Indexes. In particular, the addition of chemicals was performed in order to increase the alkalinity of the samples, which as stated in figure 8, is at very high levels. Although there is a sufficient concentration of chlorides in the samples, the increase of alkalinity acts as an inhibitor of corrosion and generate low Larson Indexes (0.8, 0.95, etc.) as well as low weight loss rates 0.48% , 0.61%, 0.52% etc. From all the above it seems that the factors that most affect corrosion are mainly alkalinity and chloride concentrations. Elevated concentrations of chlorides act as accelerators for corrosion while high levels of alkalinity act as inhibitors. In other words, water that is characterized by high concentrations of chlorides is corrosive. If however, a chemical treatment is performed by adding chemical solutions (such as NaOH and NaHCO3) to increase the alkalinity, then the corrosion effect is inhibited. Similarly, suitable mixtures of non-corrosive groundwater and corrosive water can also decrease the corrosiveness of water. Sulfate ions do not affect significantly the rate of corrosion, as they are less active and less corrosive than chlorides. Finally, the pH should be kept constant in the range of 7.2-7.7 because lower values in combination with other factors contribute to the corrosion effect.	en
heal.advisorName	Μαντζιάρας, Ιωάννης	el
heal.advisorName	Mantziaras, Ioannis	en
heal.committeeMemberName	Παπακωνσταντής, Ηλίας	el
heal.committeeMemberName	Μαμάης, Δανιήλ	el
heal.committeeMemberName	Μαντζιάρας, Ιωάννης	el
heal.committeeMemberName	Papakonstantis, Ilias	en
heal.committeeMemberName	Mamais, Danos	en
heal.committeeMemberName	Mantziaras, Ioannis	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Πολιτικών Μηχανικών	el
heal.academicPublisherID	ntua
heal.fullTextAvailability	false