CEE 370 Fall 2015

Laboratory #5 Drinking water treatment

Objective
To determine if the Robert’s Meadow Reservoir meets the current water quality standards
established by the US EPA for water supply.

Introduction

The most common and economical treatment for drinking water is “conventional
treatment”. Conventional treatment consists of the following processes: coagulation,
flocculation, sedimentation (clarification) and filtration (USEPA, 2015, Figure 1) and is
often followed by disinfection prior to a holding tank for achieving CT and then sent into
the distribution system (Figure 2). Conventional treatment may also incorporate the use of
filter media (such as granular activated carbon (GAC) or powdered activated carbon (PAC).
This level of treatment will result in water that either meets or fails in meeting federal
standards for pathogen control and for chlorinated disinfection byproducts (DBPs). In order
to determine this, water quality testing is done.

Figure 1. Schematic of conventional water treatment process (USEPA, 2015)

Figure 2. Schematic of conventional water treatment process with chlorine addition in the
treatment process

Current Drinking Water Standards in the US

Surface Water Treatment Rule (USEPA, 1989) requires 0.5 log removal of Giardia and
2.0 log removal of viruses by disinfection when followed by conventional treatment with
coagulation and filtration. To achieve this level of removal, water suppliers are required to
provide a certain “CT” (i.e., the product of disinfectant concentration C and contact time T;
Table 1). The values of CT depend on the disinfectant, the temperature and the pH.
To account for short- circuiting in chlorine contact tanks, the USEPA requires that the t10
1
be used to calculate the “effective” retention time rather than the calculated hydraulic
retention time (HRT) where:

𝐻𝑅𝑇 = 𝑉/𝑄 Eq. 1

such that V is the volume of the tank (m3) and Q is the volumetric flow rate (m3/s).

This t10 is essentially the time it takes for the fastest 10% of the influent water to exit
the tank. It is determined by a tracer study on the full-scale tank, and it is usually
represented as a percentage of the HRT. A well-baffled tank can reach t10 levels up to 50%
of the HRT or more.

The Stage 2 Disinfectants/Disinfection Byproduct Rule (D/DBPR) (USEPA, 2006)

requires that DBPs be kept below 80 µg/L for trihalomethanes (THMs) and 60 µg/L for
haloacetic acids (HAAs). This means that the amount of precursor organic matter in the
water after coagulation and filtration must be low enough so these DBPs do not exceed
these limits even when the water has reacted with the chlorine for many days (the typical
maximum value for distribution system residence times).

1
Table 1. CT Values for inactivation of Giardia Cysts by Free Chlorine at 20°C (USEPA, 1991)

3
Background/Problem Description

The Robert’s Meadow Reservoir (also referred as Hoxie Reservoir or the Leeds Reservoir)
in the Leeds section of Northampton served for 34 years as the principal water supply for
the town, and later as a supplemental supply. Following two large fires in Northampton in
1870, it was decided (February 1871) that the community should look to Roberts’ Meadow
Brook as a reliable water supply for its first municipal system. Construction of the reservoir
began in May 1871 and by September of that year it was filled with about 4 million gallons
of water. By 1873 there were 517 families that had been supplied with piped water, not to
mention a large number of businesses and 107 fire hydrants. Eventually two more dams
would be constructed in Roberts’ Meadow system.

Recognizing the need for more water than could be supplied by Roberts’ Meadow, the city
began planning for two new supplies in Whately and Williamsburg, the West Whately
Reservoir and the Mountain Street reservoir, respectively. Construction proceeded in the
first few years of the 20th century. On March 30, 1905 the city started using the new upland
sources exclusively. The Whately system would later be improved with the construction of
the much larger Francis P. Ryan Reservoir immediately adjacent to the West Whately
Reservoir.

Table 2 summarizes the drainage area, safe yield and reservoir storage of the four surface
water sources in Northampton.

Table 2. Northampton’s Surface Water Supplies (Northampton Watershed Resource

Protection Plan, PVPC, June 1994)

Source reservoir Drainage Safe Yield Reservoir

Area (acres) (MGD) Storage (MG)
Roberts Meadow 6,900 2.00 85
West Whately 1,182
3.79 750
Ryan 2,762
Mountain Street 541 1.17 375

Low water levels in the Mountain Street Reservoir forced the city to begin using Roberts
Meadow again in October 1932. The state department of health found high bacterial counts
from the Roberts Meadow supply, so the city installed a chlorination system for that supply
that went into service on Nov 9, 1932. At this time it was also recognized that the water had
high levels of organic matter (color) and algae (Gazette, May 4, 1932). The general water
degradation was probably exacerbated by runoff from upstream farms. From this point on,
the Roberts Meadow system was used only when the levels in Mt. Street Reservoir were
low and the overall supply needed a supplement. In late 1950, two wells were constructed
in the Florence section of Northampton to help provide some additional water. Roberts
Meadow Reservoir was finally taken out of service in 1978- 1979, after 108 years of use.
After this point it was still listed as a back-up source for emergencies, but it was apparently
never used again.

3
With the near failure of the Whittenton Pond Dam in Taunton in 2005, the Commonwealth
of Massachusetts began an aggressive campaign to inspect major public dams. At this time
they decided that the Upper Roberts Meadow Dam was at risk and needed to either be
repaired or removed. Following careful analysis by the city’s consultant, the decision was
made to remove it based largely on economic considerations. Most recently (Hampshire
Daily Gazette; October 29, 2010), the Northampton Board of Public Works voted to have it
removed. However, experience will tell you that in politics as with sports, “it is never over
until it is over”. There may still be appeals and possibly legal challenges to this decision.
Nevertheless, the major portion of the Roberts Meadow supply is impounded by the Middle
Roberts Meadow Dam which will remain regardless of the fate of the Upper Dam.

Purpose

The purpose of this laboratory study is to determine if the Roberts Meadow supply could
be treated using conventional (i.e., low cost) technology and meet existing standards
for disinfection and disinfection byproducts. Specifically, it would need to be of high
enough quality so that after coagulation with alum and filtration, the remaining dissolved
organic matter (DOM) is below a key threshold. That threshold is not defined by a simple
UV254 absorbance nor is it fully defined by a total organic carbon level (TOC). Instead it
is defined as a level that will not produce excessive THMs and HAAs in the distribution
system when chlorine is added at sufficient dose and contact time to meet “CT” regulations
for disinfection.

The US EPA requires that conventional treatment plants meet a certain percent removal of
TOC. For the Roberts Meadow Supply, it would probably be 35% when the raw water
TOC is below 4 mg/L and *45% when it is above this value).

In the laboratory, the conventional treatment process can be simulated with the classic Jar
Test Procedure outlined below, followed by filtration on a glass fiber filter and then by the
addition of chlorine. It must be done in this order to accurately reproduce what would occur
in a full-scale water treatment plant. The chlorinated sample is then held for a period of
time to simulate the residence time in the clear well (CT tank) and the distribution system.

Analysis

Measurement of Natural Organic Matter

We will use a Genesys 10s Ultraviolet-Visible (UV-Vis) Spectrophotometer for monitoring

the concentration of natural organic matter in the coagulated water. In our case, we are
using absorbance as a “surrogate” measure of the amount of dissolved natural organic
matter (NOM) remaining in a treated samples. Thus the sample must be filtered prior to
analysis. While we could pick almost any wavelength as our “surrogate” parameter, we
normally use 254 nm by virtue of an arbitrary convention.

This instrument must be turned on and warmed up well before use. The first step in its use
is to set the wavelength using the panel controls. It is then “calibrated”, or more accurately,
“zero-ed” by placing a cuvette with distilled water into the cell changer and pressing the
4
button corresponding to ‘measure blank’ on the screen. Make sure you’ve selected the
“Absorbance” readout. Then place your sample in a separate cuvette and press the button
on the cell changer indicating the position of the sample. You will see a read out in
absorbance units per centimeter. Also on the screen will be a reminder of the wavelength
that this instrument is tuned to. The concentration of any absorbing compound is linearly
related to its absorbance in accordance with Beer’s Law (Eq. 2):

𝐴𝑏𝑠 = 𝐴 × 𝐶 × 𝑥 (Eq. 2)

Where Abs is the measured absorbance, A is the absorptivity (M-1cm-1) in which case it is
called the molar absorptivity), C is the concentration and x is the light path length of the
sample cell (usually 1 cm, which is what we’ll be using).

Measurement of Chlorine Residual by DPD Titration

The amount of chlorine added to water is called the “chlorine dose”. The amount that
persists or can be measured after addition is called the “chlorine residual”. After addition of
aqueous chlorine to water for treatment purposes, it is important to measure how much is
left at any given time. This is because chlorine can become rapidly dissipated and its
effectiveness is directly linked to its concentration and exposure time. In this treatment
study you will be adding a range of chlorine doses (1-5 mg/L), but the residual chlorine will
be a bit lower, sometimes a lot lower. At the end of each of the tests, you will be measuring
the final chlorine residual using the DPD titration method as follows:

1. Add 5 mL of DPD buffer from auto pipet bottle to a separate Erlenmeyer flask
without the diluted sample. Add 5 mL of DPD indicator solution from the other auto pipet
bottle to the flask
2. Add 100 mL of a diluted sample to the Erlenmeyer flask. The solution would turn
pink indicating the presence of chlorine.
3. Record the initial level of the titrant (ferrous ammonium sulfate, FAS) in the
burette. Titrate with FAS until the pink color just disappears and record the final level in the
burette. (The sample you’re titrating must have a chlorine residual near or below 4 mg/L as
Cl2 (0.056 mM); add more if there is less than 4.5 mL remaining.)
4. The chlorine residual in mg/L as Cl2 in your sample is equal to the number of mLs
of titrant used: 1 mL of FAS = 0.1 mg/L of residual Chlorine as Cl2.

Relationship between Chlorine Residual and DBP Concentration

Actual measurement of disinfection byproduct (DBP) concentrations is beyond the scope of

the CEE 370 lab. However, it is possible to use chlorine residual measurements to estimate
DBP formation from coagulated and filtered waters.
First, we need to define an important water quality parameter know as chlorine demand
(Eq. 3):

Cl2 Demand (mg/L) = Chlorine Dose (mg/L) – Chlorine Residual (mg/L) (Eq. 3)

5
From this we can estimate DBP formation. The reason is that waters with almost entirely
“organic” chlorine demand, such as this one, will result in a certain percent conversion of
reacted chlorine to organic chlorine by-products. The correlation we will use comes from
research studies conducted at UMass. Equations 4 and 5 give estimates of the THM and
HAA formation under the same conditions that were used for chlorine demand test (i.e.,
same pH, time, and temperature).

𝑇𝐻𝑀 = 1.919(𝐶𝑙2 𝑑𝑒𝑚𝑎𝑛𝑑)0.47 (𝑝𝐻)1.245 (𝑡𝑖𝑚𝑒)0.053(𝑡𝑒𝑚𝑝)0.204 (Eq. 4)

𝐻𝐴𝐴 = 35.24(𝐶𝑙2 𝑑𝑒𝑚𝑎𝑛𝑑)0.178 (𝑝𝐻).0.314 (𝑡𝑖𝑚𝑒)0.141 (𝑡𝑒𝑚𝑝)0.125 (Eq. 5)

where both THM and HAA are in units of µg/L, chlorine demand is in mg/L, time is in
hours, and temperature is in °C.

Procedure

Table 3 summarizes the responsibilities for every set number, in addition to the amount of
bicarbonate and alum dose to be added.

Table 3. Summary of Test Conditions for Each Lab Period

Set # Bicarbonate addition (mL of Alum Dose (mg/L)

78 g/L NaHCO3)
1 0 0, 5, 10, 15, 30, 60
2 2 mL 0, 5, 10, 15, 30, 60
3 0 0, 10, 20, 45, 60, 100
4 2 mL 0, 10, 20, 45, 60, 100

Part A1. Jar Testing

1. Receive testing assignment (Set #) from TA
2. Measure out 1-L (or 700 mL if smaller beakers are used) volumes of the raw water
by pouring from a carboy to a plastic 1-L graduated cylinder and dispense to the
each of the six beakers
3. Add Bicarbonate solution (NaHCO3) as needed (for sets #2 and #4) to each beaker
4. Measure out an additional volume (e.g, 250 mL) of raw water into a separate 500-
mL beaker and add the same dose ratio of (NaHCO3) to this, accounting for the
smaller volume. Use this sample for measurement of raw water pH, temperature and
UV-Vis scan. While you’re taking the scan, record the raw water UV absorbance at
254 nm. You may wish to do this while the 6 beakers are in their slow mix (#8) or
settling phase (#10).
5. Place beakers under the 6-paddle stirrer (jar test machine) and stir at high speed
(~100 rpm).
6. One-by-one add the requisite volume of alum stock to achieve the desired doses for
your set# (rapid mix phase)
7. Once the last beaker has been dosed, wait 60 seconds and reduce the mixer speed to
20 rpm. Maintain this slow mix phase for 20 minutes (flocculation). During this
time you may wish to make the measurements on the small sample in the 500 mL
6
 beaker (e.g., raw water pH, temperature, UV254 and the UV-Vis scan).
8. Remove the six beakers from the 6-paddle stirrer. Gently measure coagulated pH
of each coagulated sample and the raw water from step 5.
9. Allow all 6 beakers to sit quiescently (settling phase) for 30 minutes.
10. One-by-one, carefully decant (pour, without disturbing settled solids) ~500-700 mL
into a graduated cylinder. Stop pouring once the settled solids (if any) begin to flow
out of the beaker

Part A2. Analysis of Settled and Filtered Water

11. Pour about a small volume from the graduate cylinder into a turbidimeter cell and
measure turbidity (settled turbidity).
12. Place a new Glass Fiber filter (GF/C) into the Millipore filtration apparatus. Re-
assemble the apparatus and turn on the vacuum pump. Then pour the full remaining
decanted sample volume from the large graduated cylinder into the reservoir.
13. Once filtration is complete, withdraw a few mL with a pasteur pipet and use this to
measure absorbance on the filtered sample (filtered UV254 absorbance).
14. Pour the filtered sample from the vacuum flask back into the graduated cylinder and
record the exact volume. Then pour this into a 500-mL amber bottle. If the volume
of water equal or exceeds 500 mL, pour about 490 mL, leaving a few mL of
headspace in the bottle. In either event, it is important that you know exactly what
the volume of sample is in the bottle. Label it, and set aside for chlorine demand
testing.
15. Repeat Steps 12-15 for each settled sample

Part B. Chlorine Demand Testing

1. Add the requisite volume of chlorine stock to achieve the desired chlorine dose
(See Table 4 below). Cap and mix by inverting several times.
2. Make sure the amber bottle is properly labelled (group name, jar#, conditions) and
place it on the bench for room-temperature incubation (reaction).
3. At the end of each of the 5 chlorine contact times, locate your bottles, pour out 100
mL and measure residual chlorine for each of the five. After you are done, cap the
bottles and return them to the bench to continue incubating. Except for the first
recording time (@1 hr), you will have to return during a non-lab time to make the
measurement. Please make arrangements for this with your lab partners. Only one
person from each group need be present to make these measurements. Please
contact the instructor or TA if the door is locked. Be aware that you can modify the
reaction time for your convenience, but it is important you record the exact time of
analysis. Also you should be aware that some bottles may run out of chlorine before
the last scheduled reaction time. If you notice one that has dropped to zero residual,
there is no need to continue measuring it.
4. Measure and record final pH in each sample after the last chlorine measurement.
5. When the last measurement is made, measure and record the final water
temperature for at least one sample.

7
 Table 4. Summary of test conditions for each lab period

Chlorine Dose Approximate Reaction Times

1 1.5 mg/L 1 hr, 14 hrs, 26 hrs , 2 days, 3 days
2 2.5 mg/L 1 hr, 14 hrs, 26 hrs , 2 days, 3 days
3 3.5 mg/L 1 hr, 38 hrs, 3 days, 5 days, 7 days
4 5 mg/L 1 hr, 38 hrs, 3 days, 5 days, 7 days

Steps for completing the lab write-up

Sharing of Data
a) Transfer your coagulation data to the excel spreadsheet template that you were sent by
email.
b) Send the filled-in spreadsheet by email to the instructor and to lab TA. Do this within 24
hours of your lab period.
c) Continue to fill in the chlorine residual portion of the spreadsheet as you collect more
data.
d) Send the completed spreadsheet (including chlorine residual data) to the instructor and
TA by Tuesday, April 17th at noon.

Presentation of your own Data and Discussion

e) Graph the raw water absorbance spectrum and a spectrum of coagulated/filtered water
(use the one with the highest dose). Graph both on the same set of axes with wavelength
(200-400 nm) on the x-axis. Are they similar? What does this tell you about using 254 nm
as an indicator of organic matter concentration?
f) Graph settled water turbidity and UV absorbance (254 nm) versus alum dose in mg/L.
Alum dose is the independent variable, and therefore it should be identified with the x-axis.

Interpretation Regarding Current EPA Regulations

g) The instructor and Lab TAs will provide you with the entire class’s data at some point
during the second week of the lab. You should incorporate this dataset into your group’s
lab writeup.
h) Use the power function equations (#2 and #3) to estimate THM and HAA levels for the
various treatment scenarios (alum dose, pH, chlorine dose) with special attention to the 3
day chlorine reaction time as this is the normal maximum water age in the Northampton
system.
i) Comment on the feasibility of meeting both THM and HAA standards while still
maintaining a measureable chlorine residual (i.e., >0.1 mg/L) for the 3-day reaction time.
j) Propose at least 2 scenarios (chlorine dose and soda ash addition) that will meet these
requirements. If none do, propose at least 2 scenarios that come closest. Comment on the
“optimal” alum dosage for this scenario.
k) Calculate a chlorine contact tank volume for the two scenarios needed to meet the
“CT” requirement assuming a flow of 2 MGD at Roberts Meadow. Use 0.5°C as a
conservative design condition (Table 1). Although the USEPA requires an additional 0.5
log Giardia removal by chlorination, assume the City wants to achieve a full 1 log removal
credit to provide an additional margin of safety. You should also assume a t10 of 50% of
8
the hydraulic retention time of the tank.

References:

USEPA, 1989. Drinking Water, National Primary Drinking Water Regulations, Filtration,
Disinfection, Turbidity, Giardia Lamblia, Viruses, Legionella, and Heterotrophic Bacteria,
Vol 54 (124).
USEPA, 2015. Conventional Treatment. Retrieved on October 29, 2015 from:
http://iaspub.epa.gov/tdb/pages/treatment/treatmentOverview.do?treatmentProcessId=1934
681921
USEPA, 2006. National Primary Drinking Water Regulations: Stage 2 Disinfectants and
Disinfection Byproducts Rule; National Primary and Secondary Drinking Water
Regulations: Approval of Analytical Methods for Chemical Contaminants, Vol 68(159).
USEPA, 1991. Guidance Manual for Compliance with the Filtration and Disinfection
Requirements for Public Water Systems Using Surface Water Sources. Retrieved from:
http://water.epa.gov/lawsregs/rulesregs/sdwa/swtr/upload/guidsws.pdf