Accepted Manuscript

Efficient Treatment of Oil Sands Produced Water: Process Integration Using

Farshad Mohammadtabar, Rajesh G. Pillai, Behnam Khorshidi, Armin

PII: S1383-5866(18)34055-3
DOI: https://doi.org/10.1016/j.seppur.2019.03.070
Reference: SEPPUR 15437

To appear in: Separation and Purification Technology

Received Date: 17 November 2018

Please cite this article as: F. Mohammadtabar, R.G. Pillai, B. Khorshidi, A. Hayatbakhsh, M. Sadrzadeh, Efficient
Treatment of Oil Sands Produced Water: Process Integration Using Ion Exchange Regeneration Wastewater as a
Chemical Coagulant, Separation and Purification Technology (2019), doi: https://doi.org/10.1016/j.seppur.
2019.03.070

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Efficient Treatment of Oil Sands Produced Water:

Process Integration Using Ion Exchange Regeneration

Wastewater as a Chemical Coagulant

Farshad Mohammadtabar1, Rajesh G. Pillai1†, Behnam Khorshidi1, Armin Hayatbakhsh2,

Mohtada Sadrzadeh1

1
Department of Mechanical Engineering, 10-367 Donadeo Innovation Center for Engineering, Advanced

Water Research Lab (AWRL), University of Alberta, Edmonton, Alberta T6G 1H9, Canada

2
EXEN Pro Ltd, 144 4th Ave SW, Calgary, Alberta T2P 3N4, Canada

KEYWORDS: Oil sands; Produced water treatment; Boiler blowdown;

Coagulation/flocculation; SAGD; Process integration

1
ABSTRACT

The potential of ion exchange regeneration wastewater (IERW) containing magnesium ions to

act as a coagulant for steam-assisted gravity drainage (SAGD) boiler blowdown (BBD) water

was investigated with the aim of reducing the water consumption in a SAGD plant. Conventional

techniques for the treatment of BBD water require significant operational costs as they rely on

the extensive use of chemical coagulants or electric power. Since the proposed process design in

this study uses a wastewater stream for the treatment of the BBD water, operational costs and

energy consumption are predicted to be minimized. The effect of temperature, the mixing ratio of

BBD and the IERW water, and the stirring intensity on the removal of organic matter and

dissolved minerals were investigated. Usage of IERW as the coagulant enhanced the removal of

impurities; removal efficiency of 81.1% and 98.6% were obtained for organic matter and silica

concentration, respectively. The treatment of BBD water by IERW, however, has led to an

increase in the concentration of calcium in the effluent, which was effectively mitigated by soda

ash softening. Furthermore, resource recovery from the coagulated sludge was explored by SEM,

EDX, FTIR, XPS, and XRD, and revealed that the precipitated compounds were mostly

composed of calcium, magnesium, and silica. Additionally, the successful usage of IERW for the

treatment of another BBD wastewater with a higher concentration of organic matter was

demonstrated. In conclusion, the IERW water proved to be highly effective in removing

contaminants from SAGD produced water.

INTRODUCTION

Steam-assisted gravity drainage (SAGD) technology is considered a practical process to recover

bitumen from oil sands reservoirs [1–4]. The SAGD process uses two parallel horizontal wells,

which are drilled above each other deep underground into the oil sand reservoir. To increase the

2
temperature and thus reduce the viscosity of the bitumen, steam is injected through the upper

well and the low viscous drained bitumen collected into the lower well as a mixture to be

pumped to the surface for bitumen extraction. This process requires a high volume of fresh

water, and currently about 80% of the boiler feed water (BFW) is coming from the recycled oil

sands affected water.

Recently, significant attention has been given to water treatment methods of SAGD plants since

a poor quality feed for the steam generator will lower the efficiency of the boiler, BFW purity

should be at an acceptable level. In SAGD industry, once through steam generators (OTSG) are

widely used to generate steam. To provide feed water for OTSG, water treatment processes

should reduce silica and hardness concentration to <50 mg/L and <1 mg/L, respectively. Also,

the BFW should have total dissolved solid (TDS) and oil content, lower than 7000 mg/L and 0.5

mg/L [1,5,6]. Ion exchange regeneration treatment is a commonly used method to remove

hardness from the BFW by replacing the calcium and magnesium ions with sodium ions [7]. In

this process, a concentrated sodium chloride solution regenerates the ion exchanger. The purpose

of the regeneration is to replace the calcium and magnesium ions, which were removed from the

wastewater and retained in the ion exchanger, with sodium ions and return the resin to its

original state. Therefore this process results in wastewater with a high concentration of sodium,

calcium, and magnesium (Scheme S1, supporting information), which is called the ion exchange

regeneration waste (IERW) [8,9]. After removing hardness, the treated water is guided to the

boiler, and the IERW is moved to the disposal system. The OTSG steam generation process

typically produces 80% steam quality resulting in a concentrated solution, which is known as

boiler blowdown (BBD). The BBD's impurity concentration is much higher than the BFW. In a

typical SAGD plant, an amount of the BBD is recycled back to the water treatment section, and

3
the rest is partially directed to the system [2]. In conventional approaches, BBD can be treated

for reuse with the help of membrane filtrations, evaporation, chemical and biological treatment

techniques. Previous research has shown that the concentration of contaminants in the BBD is

higher than in other types of wastewater in a SAGD plant. Membrane filtration is an effective

method to treat the BBD due to its high TDS, organic matter and silica removal efficiency

[1,10,11] and Fenton’s oxidation and acidification are effective for the removal of silica and

organic matter from this wastewater [3]. However, these methods increase the capital cost and

energy consumption of the SAGD plant and can result in significant waste [1]. What’s more, the

use of water treatment applications for the reuse of BBD demands specialized and expensive

equipment. Therefore, it is highly beneficial to use inexpensive but efficient technology to treat

the BBD.

In conventional water treatment plants, flocculation is an essential technique to separate

impurities from contaminated water solutions [10,12,13]. At present, various types of coagulants

(such as lime, soda ash, caustic) are used in wastewater plants to improve the efficiency of the

chemical process [14]. Previous studies reported the successful use of magnesium at high pH for

achieving a high removal efficiency of unwanted contaminants from wastewater [12,14–20].

Magnesium appeared to be an effective coagulant in the reduction of silica from the heavy oil

wastewater [21]. Silica removal using magnesium substances also proved to be useful as a pre-

treatment for filtration processes [18]. The influence of magnesium hydroxide in the separation

of organic matter from wastewater was also investigated. It was demonstrated that organic matter

removal of wastewater could be improved by increasing the concentration of magnesium

hydroxide [16]. Based on our earlier studies, The SAGD produced water contains various types

of organic substances such as aromatic acids, phenols and naphthenic acids [2,3].

4
At a high pH, magnesium precipitates as magnesium hydroxide according to the following

reaction [14]:

(1)

The resulting magnesium hydroxide that carries positive charges acts as an effective coagulant

by charge neutralization; and, according to DLVO theory, this substance can remove negatively

charged organic matters and silica particles [22]. The DLVO theory explains the stability of

colloidal particles as a balance between the electrostatic repulsion due to the surface charges and

van der Waal’s interaction of the colloids in the suspension [23]. In the water treatment,

adsorption occurs when the attraction force between the solid and the impurity ion is stronger

than the attraction force of the water and the unwanted solute [24,25]. What’s more, calcium

carbonate can be removed by adsorption to the magnesium hydroxide floc [14,15]. Thus, the

active surface of the magnesium hydroxide can attract organic matter, calcium carbonate, and

silicate through adsorption [21].

Although extensive research has been carried out using chemical coagulants to treat the BBD,

the majority of those coagulants are not environmentally friendly and require a high dosage of

coagulants. Furthermore, usage of chemical additives may overload the water treatment system

and increase the operating resources and costs. Therefore, if a waste stream in the SAGD plant

can be used as a coagulant, this will lead to a more cost and energy effective process for BBD

treatment. Therefore, the major objective of this study is to investigate the feasibility of using the

currently unusable IERW as the coagulant to treat the BBD under different treatment conditions

(Scheme S2). This wastewater contains a high concentration of magnesium and calcium that can

be potentially utilized as effective coagulants in reducing the silica and organic content of BBD

5
water. This is the first report, to our knowledge, discussing the possibility of using ion exchange

regeneration wastewater as a coagulant for the treatment of SAGD process-affected water and

exploring the resource recovery from coagulated sludge waste. A full factorial design was

applied to study the contribution of temperature, the ratio of IERW to BBD and mixing intensity

on the outcome of the treatment; the level of these control factors was selected based on the

feasibility of these conditions in a SAGD plant. Another important aspect of using IERW as the

coagulant is that by applying this method of treatment, the extra cost and energy for the disposal

of both IERW and BBD can be reduced. Finally, the chemical composition of the resulting

sludge from this water treatment process was analyzed to explore the feasibility of resource

recovery.

2. MATERIALS AND METHODS

In this study, two BBD water samples (BBD-1 and BBD-2) with varying dissolved organic

content were used. BBD-1 and IERW samples were provided by EXEN Pro Ltd. through

collection and delivery by Black Pearl Resources Inc., located in Calgary. The IERW was used

as a coagulant to treat the BBD. The characteristics of BBD water samples and IERW are

presented in Table 1. Additionally, another BBD wastewater sample was prepared by increasing

the pH and silica concentration of the boiler feed water received from a SAGD plant located in

the Athabasca oil sands region of Alberta, Canada [11]. The properties of the synthesized BBD

(BBD-2) are written in Table 1.

6
Table 1: The properties of BBD-1, IERW and BBD-2 water at room temperature.

BBD reported

Parameter Unit IERW BBD-1 BBD-2 in literature

[1,2,20]

TDS ppm 66625±375 6525±25 2212±35 4026-17200

Conductivity ms/cm 103.1 13.2 1.8 -

pH - 6.20±0.11 11.66±0.01 11.60±0.10 10.50-12.33

Turbidity NTU 0.25±0.05 0.86±0.06 - 0.20-53

UV absorbance at 254 -
- 0.07±0.01 0.72±0.04 0.55-0.87
nm

SUVA254 - 1.04±0.15 0.77±0.08 - 2.75-5.21

TOC ppm 6.71±0.07 92.12±0.11 274±2.40 695-2482

Silica ppm as SiO2 5.22±0.48 77.60±1.50 165±0.13 65-238

Magnesium ppm 2201±351 0.24±0.03 57.3±1.10 0.68-0.08

Calcium ppm 9455±1059 2.97±2.59 310±2.20 4.25-4.80

Sodium ppm 22165±3244 1806±130 1830±59 819-5199

Chloride ppm 80650±4203 40±6 - 494-6715

Sulfate ppm 320±57 1230±310 - -

2.1. Design of experiments

A full factorial design was employed in this process to analyze the treatment of BBD-1 by

varying three control factors. STATISTICA software was used for the experimental design and

data analysis. This design consists of three dependent variables and two levels. In addition, one

replication was done to minimize variability in the experimental results. The control factors and

7
the selected levels for this design are provided in Table 2. These control factors were chosen

based on the common industrial practices to treat the wastewaters using chemical coagulant. For

all of the experiments, the mixing (coagulation) and precipitation time were 30 minutes.

Table 2: Full factorial design factors and levels.

Levels
Factors Unit
-1 1

IERW to BBD ratio Volumetric ratio (IERW:BBD) 1:12 2:12

Temperature °C 40 80

The speed of the stirring rpm 0 60

According to the experimental design, the total number of runs was 16 experiments with one

replication. This approach is useful to understand the significant factors and the interaction

effects between the variables. For all of the experiment, the response variables considered were

turbidity, ultraviolet (UV) absorbance at 254 nm, silica concentration and total organic carbon

(TOC). The blocking was applied to hinder the effect of nuisance factors, which can affect the

outcome of the experiment, by using the same batch of BBD-1 and IERW for all experiments.

Some of the nuisance factors for this experiment can be the different concentrations of TDS,

silica and organic matter of the water samples for different batches. The different experimental

runs based on this design are presented in Table 3.

8
Table 3: Different experimental runs based on the design of the experiments.

Level of design factors

1 -1 -1 -1

2 -1 -1 1

3 -1 1 -1

4 -1 1 1

5 1 -1 -1

6 1 -1 1

7 1 1 -1

8 1 1 1

The analysis of variation (ANOVA) was applied to study the effect of the control factors on the

outcome of the experiment. In this analysis, a probability value (p-value) indicate the

significance of the design factors on the response variable (turbidity). This ANOVA table was

obtained with a 95% confidence level so in this experiment; a design factor with a p-value lower

than 0.05 will have a significant contribution to the outcome of the experiment [42]. In this table,

the sum of square (SS) is the variation in the observed result for each factor and the degrees of

freedom (DOF) represents the number of values that have the freedom to vary in a statistical

study. The mean square can be calculated by the sum of the square over the degree of freedom.

The F-value is the mean square of measurements divided by the mean square of the residual. It

should be noted that a larger F-value indicates that the factor has a relevant effect on the

experiment.

9
2.2. Batch test procedure

Jar tests were performed with a total volume of 1 L for the IERW coagulant and BBD. The

IERW to BBD ratios were adjusted to 1:12 or 2:12, and the solution temperature was set to the

40 °C or 80 °C as stated in Table 2. These dosages are selected based on the water flow process

of the actual plant for IERW and BBD streams of the SAGD operation. The operating

temperatures for the chemical treatment process were selected based on the operating condition

of a typical SAGD process. In this process, the temperature of the produced water is usually as

high as 80 °C; however, it is also feasible to lower the temperature down to 25 °C. Two mixing

methods were used for this treatment, including the 30 minutes of stirring at 60 rpm with a

magnetic stirrer and no stirring, which involved simply pouring both solutions into the 1 L

beaker. This procedure was followed by emptying the mixture solution into a 1 L graduated

cylinder and placing it in an oven to keep the solution at the desired temperature. The solution

was then allowed to settle for 30 minutes. Finally, samples were taken from the supernatant and

slurry phase for characterization.

2.3. Characterization of wastewater and sludge samples

An inductively coupled plasma optical emission spectroscopy (ICP-OES) was used to determine

the concentration of silica, calcium, sodium, and magnesium. Dissolved organic matter (DOC)

was analyzed with the non-purgeable organic carbon method using the Shimadzu TOC-VCSH

analyzer and was reported as total organic carbon (TOC). UV absorbance was measured using a

Hach DR 5000 spectrophotometer. Ion chromatography (IC, DIONEX ICS-3000 ion

chromatography system) was employed to identify the concentration of chloride and sulfate

anions. The Zetasizer Nano Z system was used for the zeta potential measurement. All samples

for the IC, UV absorbance, and ICP analysis were filtrated by a 0.45-μm PVDF membrane

10
(Millex, EMD Millipore). Therefore, the presented concentrations of anions and cations in this

work represent the dissolved phase of these substances. . As silica is one of the most stable form

of silicon and due to its abundance in the sand, Si mostly, exist as SiO2 in the wastewater.

Turbidity and pH were analyzed using a turbidity meter, (expressed in nephelometric turbidity

unit (NTU)), and pH meter, respectively. Field emission scanning electron microscopy (FESEM,

Zeiss Sigma 300 VP) was used to observe sludge samples after precipitation. The elemental

compositions of the precipitated solids were evaluated using energy-dispersive X-ray

spectroscopy (EDXS), X-ray photoelectron spectroscopy (XPS, Kratos AXIS ULTRA

spectrometer) and X-ray powder diffraction (XRD, Rigaku Ultima IV X-ray diffractometer)

methods. The chemical functional groups present in the produced sludge were analyzed by

diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy.

3. RESULT AND DISCUSSION

3.1. Removal efficiency analysis of treated water

Figure 1 presents the removal percentage of TOC and silica for different experimental trials after

the chemical treatment. Based on this figure, the removal percentage for silica varied from

97.77% to 99.44% and the removal percentage for TOC was from 78.64% to 83.22%. The

removal percentages were obtained by comparing the initial value of the parameters in the BBD-

1 to the treated supernatant samples. The mean removal percentage of all eight runs for TOC and

silica concentration was 81.06% ± 0.82% and 98.63% ± 0.61%, respectively. Therefore, using

the IERW for the BBD-1 treatment resulted in a high removal percentage of organic matter and

silica concentration for all of the experimental runs, which was in agreement with previous

removal percentages in the literature [16,26,27]. As illustrated in Figure 1, the removal

11
percentages showed a small variation between the different runs suggesting the insignificant

effect of control factors on the removal of silica particles and organic matters. Thus, it can be

concluded that the dilution effect of using this water-based coagulant was insignificant since

using two different volumetric ratios of IERW:BBD approximately achieved the same amount of

contaminant removal. This can be explained by the coagulation-flocculation of silica and organic

particles induced by magnesium hydroxide. In the case of BBD-1, the zeta potential was -14.65

mV (±0.95) which indicates that the organic matters and silica particles have a negative surface

charge. According to DLVO theory, the negatively charged silica and organic molecules are

rapidly making contact with each other due to Brownian motion and are continually being

repelled by electrostatic repulsion due to their similar surface charge. After adding the IERW to

the BBD-1, the ionic strength increased which resulted in compression of the double layer

thickness of suspended materials like silica and different organic matter ions. At higher ionic

strength, the attraction forces between these particles became larger and larger flocs were formed

[28,29].

12
Figure 1: Removal percentage of silica and TOC for all of the experimental trials.

Table S1 summarizes the properties of the supernatant for all of the experimental runs. It shows

that the treated water for all of the experimental trials had high calcium content. However, in the

supernatant samples, the concentration of calcium and sodium increased significantly after the

treatment. Since the calcium concentration varied based on the IERW dosage, the average

concentration of calcium was presented based on the two different dosages of the IERW, 598 and

1102 ppm for the lower and higher dosages of IERW, respectively. Furthermore, the mean

concentration of magnesium for all of the supernatants was about 0.85 ppm. It has been reported

that a SUVA value higher than 4 implies the dominant presence of aquatic humic compounds.

The presence of these humic substances is the indicator of higher hydrophobicity, which can

have an adverse effect such as requiring a higher dosage of coagulant in the chemical treatments

and causing severe organic fouling in the membrane filtrations [30]. Table S1 also shows the

SUVA value calculated from the UV absorbance at 254 nm (UV254). As can be seen, the

13
SUVA254 value was between 2.26 to 3.07, which suggests the presence of hydrophobic humic

and hydrophilic non-humic acid [31]. Thus, the SUVA values of supernatants for all of the

experimental runs were in the acceptable range.

Table S1 also shows the turbidity values for all of the experimental trials. Turbidity is a term

used to determine water quality; this parameter is determined by optical measurement of light,

which scatters due to suspended and dissolved particles [32–34]. The turbidity varied from 1.1 to

17.3 NTU for different runs. This result indicates a noticeable change in the results for different

conditions indicating a significant effect of the controllable parameters on the turbidity of the

treated water. The significance of the different factors is more evident when comparing run 1 and

8 since all of their factor levels are different. This is because after adding the coagulant, the silica

and organic matters were removed from dissolved phase, but the formed flocs were not

precipitated in the given time and remained suspended in the supernatant phase; the turbidity

measurements confirm this explanation.

3.2. Statistical analysis

The efficiency of this chemical process was analyzed to find the significant controllable factors.

The performance of this treatment was studied using turbidity since the purpose of this process is

to remove the suspended solids through sedimentation and filtration [34–38]. The treated water

after the flocculation had high turbidity in all of the experimental runs, however the BBD-1 had

low turbidity. It is worth noting that the reason for the low turbidity value of BBD-1 is probably

due to the relatively high concentration of organic matter and the small amount of suspended

solids. The general difference between the means of the turbidity values in different experimental

conditions can be determined using the analysis of variation (ANOVA). The results of the

ANOVA test considering the turbidity as the response variable is presented in Table 4. For this

14
experiment, the minimum F-value indicating the significant factors is 6.59; this value was found

from the F-distribution table. Therefore based on the p-value and F-value, this statistical analysis

indicates that all of the controllable factors are significant. Moreover, the p-values and F-values

state that the mixing intensity is the most significant factor as it has the lowest p-value and has

the highest effect on turbidity. These results also indicated that the temperature had a higher

effect on the turbidity compared to the dose of the IERW coagulant [39,40].

Table 4: ANOVA table of control factors for turbidity.

ANOVA-Turbidity
Factors (symbol)
SS DOF Mean square F-value P-value

Temperature (A) 27.3060 1 27.3060 37.8161 0.003547

Speed of stirring (B) 84.6001 1 84.6001 117.2040 0.000413

Dose of IERW (C) 7.8408 1 7.8408 10.8587 0.030067

Residual 2.8883 4 0.7221 - -

The variance of turbidity based on different factor levels can be investigated with mean plot. The

effect of temperature on the mean plot of turbidity is illustrated in Figure 2. It was found that by

increasing the temperature from 40 °C to 80 °C, the turbidity value of supernatant decreased

from 7.94 NTU to 4.24 NTU. This behavior can be attributed to the fact that by increasing the

temperature, the size of the flocs can become larger which leads to the precipitation of the

aggregated particles. In previous studies, a similar trend was observed where it was seen that at

higher temperatures, the removal efficiency of silica from heavy oil wastewater was improved

[21]. It was also demonstrated that the aggregation in the lower temperature could result in a

higher breakage and lower re-growth of the flocs [41,42].

15
The influence of coagulant dosage on turbidity removal was investigated by using two different

levels of dosage; the results are depicted as a mean plot in Figure 2. These results indicate that

there was a decrease in turbidity value when the dosage increased, resulting in a supernatant with

lower turbidity which indicates a more efficient treatment. This trend is expected since

increasing the dosage of coagulant will provide more magnesium ions to form insoluble

magnesium hydroxide resulting in an improved aggregation process, which can remove more

negatively charged ions by neutralization and adsorption improving the flocculation [14–19].

Figure 2 also shows the coagulation performance of IERW under different mixing conditions

using a mean plot. As can be seen in this figure, the mixing had a significant influence on the

supernatant properties. After applying the mixing, the turbidity decreased from 9.34 to 2.84

NTU. This result confirms that mixing intensity can be used to improve the removal percentage

of suspended solids by affecting the strength and size of the aggregated particles that are being

formed due to flocculation. It can be concluded that mixing can remove more contaminants

through sedimentation by allowing larger flocs to be formed [43]. Therefore, mixing the

solutions enhanced the adsorption and charge neutralization process. It should be noted that the

mixing intensity must be at a proper range. Furthermore, a very high or low stirring speed can

have an adverse effect on the flocs properties. For example, a high mixing intensity can result in

breakage of the floc into clusters [44].

16
Figure 2: The mean plot of temperature, IERW to BBD ratio and mixing intensity for the

turbidity.

Based on the variation of the turbidity and the design factors, the optimized level for each

parameter can be estimated. For temperature, removal efficiency was highest at 80 °C, and the

removal was higher when the IERW and BBD-1 were mixed for 30 minutes at 60 rpm.

Furthermore, increasing the dose improved the flocculation and precipitation so 2:12 volume

ratio is recommended coagulant dose to achieve better water quality.

From the regression coefficients, the regression model of the process was obtained to study the

functional relationship between factors and response variables, and to find the optimized

condition for this process. This model can be written as follows:

17
 (2)

Where A, B and C represent temperature, mixing intensity, and IERW to BBD ratio,

respectively. The relationship and significance of each variable can also be observed in this

model. This equation shows more precisely that mixing had the highest effect on the turbidity of

the supernatant. Based on the regression model of turbidity, the optimum experimental condition

can be achieved by using the higher factor level. Therefore, the supernatant sample of run 8,

which was completed at the highest factor levels, showed the most suitable water quality.

3.3. Sludge characterization

The treatment of BBD-1 with IERW lead to coagulation followed by sludge formation and water

recovery. The physicochemical characterization of the precipitated flocs was further performed

to explore resources recovery from the sludge waste. A mass balance was conducted to

investigate the removal percentage of different elements in the sedimentation process. The total

input and output mass of calcium, magnesium, silica and sodium from run 8 are given in Table

S2. The samples were prepared by evaporating the slurry phase of the run 8. The water recovery

of this run was 80 % and the wet cake sample showed a total dry solid concentration of 9 %. The

observed silica removal in Table S2 is in agreement with the silica removal observed in Figure 1.

This table also confirms that the formation of flocs is due to neutralization and adsorption of

negative charged particles such as silica and calcium carbonate to the surface charge of the

magnesium hydroxide.

The produced sludge after IERW treatment was analyzed using SEM, EDX, FTIR, XPS and

XRD techniques. SEM and EDX analyses show the morphology and composition of the sludge

residue (Figure 3 (a), (b) and (c)). The SEM micrographs show continuous amorphous phase

with the semi-crystalline domain, and an average size of the particles, which was about 5 µm.

18
The elemental composition of the sludge was obtained from the EDX analysis include Ca, Mg,

Cl, Si, Na, O and C. It was demonstrated that the concentration of Mg and Ca was high in the

precipitated solids. This result shows that particles were entrapped by magnesium due to

flocculation mechanism, which confirms the silica and organic matter removal by using the

IERW as the coagulant [45–48].

In order to identify the major chemical species and functional groups present in the sludge, FTIR

analysis of the solid precipitates were performed which is demonstrated in Figure 3 (d). The peak

at 1007 cm-1 can be attributed to the Si-O band and the peaks from 1156-1200 cm-1 can be

assigned to silica bonding. The 1631 cm-1, 650 cm-1 and 3398 cm-1 bands can be the

characteristics of bending vibration of O-H. The 874 cm-1 peak observed at the low-frequency

region can represent Mg-O vibration. According to literature, peak at 1447 cm-1 with high

intensity can be due to the presence of CO32- [18,49,50]. Therefore, FTIR data are in agreement

with EDX results suggesting the existence of Mg, Ca and Si in the precipitated sludge.

19
Figure 3: SEM, EDX and FTIR analysis of the precipitated sludge obtained after the treatment

of BBD water with IERW. Representative SEM images with a magnification of a) 100x, b) 330x,

and c) EDX results. (d) Representative FTIR spectra of precipitated sludge obtained after the

treatment of BBD water with IERW.

To further investigate the elemental composition of the precipitated sludge and their ionic state,

XPS analysis was performed. Figure 4 shows the XPS survey spectrum of the precipitated sludge

and the high-resolution spectra of Ca 2p and Mg 2p. The peaks in the survey spectrum were

20
assigned to Ca, Mg, Si, Na, Cl, O, S, and C in the solid sample. The peak at 167.3 eV in the

survey indicate oxygenated sulfur species, possibly –SO4. These results are in agreement with

EDX, FTIR and removal percentages mentioned in Table S2. Furthermore, different atomic

percent of these elements, given in Figure 4 (a), confirms the precipitation of corresponding

compounds after the chemical treatment. Figure 4 (b) shows the high-resolution spectra for Mg

2p with a peak binding energy around 50.0 eV, which is close to those reported for magnesium

hydroxides in the literature [51]. These results shows that magnesium was mostly precipitated as

magnesium hydroxide [52]. The Ca 2p high-resolution spectra clearly shows well-defined peaks

at 351.0 eV and 347.5 eV corresponding to 2p 1/2 and 2p 3/2 respectively with a peak area ratio

of 1:2. The Ca2+ chemical shifts between various compounds (e.g., sulfates and carbonates) fall

within a small range (< 1eV), so it is not possible to identify the nature of Ca 2+ compounds from

XPS data alone. Therefore, XRD analysis was performed on the sludge samples as the diffraction

data can be used as fingerprints for sample identification.

21
Figure 4: XPS analysis of the precipitated sludge. (a) XPS- survey spectrum of the precipitated

sludge showing component elemental peaks identified and their relative abundance. The high-

resolution XPS spectra for Mg 2p and Ca 2p are shown in (b) and (c), respectively.

The XRD pattern of the precipitated sludge collected after the treatment of BBD water with

IERW is given in Figure 5. The presence of CaSO4, Mg(OH)2, CaCO3 and NaCl were identified

after fingerprinting the standard spectra of these compounds with the XRD pattern of the sludge.

The XRD results further confirm the EDX, FTIR and XPS results and show successful

sedimentation of BBD-1 water in the presence of IERW coagulant. Moreover, based on the

presence of calcium sulfate, which is considered as an industrial product, it can be concluded that

this sludge has the potential to be used as a by-product through the extraction of calcium sulfate.

22
Figure 5: XRD Powder pattern of the dried sludge indicating the presence of different crystal

patterns in the slurry phase.

3.4. Softening of BBD after IERW coagulant treatment

Table S1 shows that the concentration of calcium after the coagulation process by the IERW and

BBD-1 mixing is high. If the supernatant would be recycled, this hardness can reduce the

efficacy of the SAGD plant by causing scaling on the pipes, exchangers and the boilers. Thus,

soda softening was applied to remove the abundant calcium ions from the produced supernatant.

23
For this purpose, the supernatant of the most efficient run (run 8) was extracted from the mixture

solution. The properties of this supernatant are available in Table S2. Based on the IC analysis,

the concentration of the chloride and sulfate in this solution was 10990 ppm and 1030 ppm, so

presumably, most of the remaining calcium in the solution is in the form of non-carbonate

hardness (permanent hardness) such as CaSO4 and CaCl2. This observation also demonstrates

that the coagulation process increased the chloride concentration of the BBD-1, significantly,

while the concentration of sulfate decreased most likely due to the formation of calcium sulfate.

Soda ash (sodium bicarbonate) was chosen to remove calcium ions from the solution because

this coagulant has high efficiency in the removal of permanent hardness from water. For this

purpose, 5 g of soda ash was added to 1 liter of supernatant with the same experimental method

as the run 8. This dosage was selected based on some primary experiments, which analyzed the

hardness removal at a different dosage of soda ash. The reactions involved in the removal of

calcium are as follow [53,54]:

(3)

(4)

After the sedimentation, the solution was analyzed for calcium removal. The properties of this

solution are provided in Table 5. From this table, it can be observed that the concentration of

calcium decreased from 1084 ppm to zero ppm, but the sodium concentration increased from

4397 ppm to 6973 ppm. This result shows that after the two treatments, almost all of the

hardness was removed. It is worth noting that in a SAGD plant, the volume of BBD is only a

portion of the BFW; the volumetric ratio of BBD used in this study was about 10% of the BFW.

It can be said that in the industrial practices, the treated BBD will be mixed with the BFW before

entering the boilers. Therefore, the treated BBD-1 in this study can be diluted almost ten times

24
and then be reused as the BFW. Thus, after dilution, the treated water will have an acceptable

concentration range of TDS and sodium too. This provide an opportunity to use a lower volume

of fresh water for the boilers.

Table 5: Comparison of the properties of the supernatant after soda softening.

Supernatant before soda Supernatant after soda

Calcium ppm 1084 0

Magnesium ppm 1.29 0.01

Sodium ppm 4311 6973

pH 10.75 10.96

TDS ppm 11350 16665

Conductivity ms/cm 28.3 30.3

3.6. Applicability of IERW as a coagulant in the treatment of wastewaters

In order to assess the treatment efficiency of IERW with other types of wastewater, a BBD-2

sample with a higher concentration of organic matter was treated with the IERW coagulant (see

Table 1). IERW water was added to the BBD-2 sample with the same experimental conditions as

run 8. In this experiment, three different dosages of IERW was used for the treatment of BBD-2,

and the precipitation of the mixture solution was observed for 1 hour. Figure 6 shows the effect

of using different dosages of IERW for the treatment of the Athabasca BBD-2. It can be seen that

high silica and organic matter removal were achieved and the color of BBD-2 became lighter

after the treatment with IERW due to the removal of organic matter. An increase in the IERW

dosage shows higher removal of TOC and silica, possibly due to the presence of more

25
magnesium ions. These results confirm that the precipitation of the magnesium hydroxide was

effective in removing the silica particles and organic matters in the BBD-2 by the coagulation-

flocculation process.

Figure 6: Demonstrate the treatment of BBD-2 water using IERW as coagulant. The silica and

TOC removal percentages using IERW for the treatment of BBD-2 are depicted in the graph.

The solution labelled (a) is BBD-2 and the (b), (c) and (d) represent the mixture solution after

sedimentation with 2:12, 4:12, 6:12 IERW/BBD volumetric ratio, respectively.

CONCLUSIONS

In the present work, a systematic study has been conducted to investigate the feasibility of using

IERW as a coagulant for the treatment of BBD. It was observed that the IERW is capable of

reducing some major impurities. These contaminants can be responsible for reducing the plant

water recycling and efficiency of the OSTG, from the BBD and this takes into account not only

the silica removal but also the organic matter presented in the BBD. The efficiency of silica and

TOC removal is 98.72% and 81.34%, respectively, which can be considered effective in a

26
chemical treatment process. The turbidity measurements indicate that when the temperature

decreased from 80 °C to 40 °C, the number of settled particles was reduced. Moreover, it was

observed that using mixing and increasing the temperature and dosage of the IERW improved

the flocculation process significantly and larger flocs were formed and precipitated. However,

the concentration of calcium increased after this treatment; this problem was solved by using

soda ash softening process to remove the calcium from the solution. The sludge characterization

supported the hypothesis of silica and organic matter removal by coagulation-flocculation of

magnesium hydroxide. The usage of IERW was found to be effective for the treatment of

Athabasca BBD. Therefore, even by considering the use of a softening treatment, this process

can be practiced to reduce the extensive use of freshwater as BFW and to eliminate the risk of

disposal well plugging.

ACKNOWLEDGMENT

The authors would like to thank the EXEN PRO Ltd for their financial support and Natural

Sciences and Engineering Research Council of Canada (NSERC).

ASSOCIATED CONTENT

Supporting Information. The supporting information includes details of ion exchange

AUTHOR INFORMATION

Corresponding Authors


(R.G.P.) E-mail: [email protected]. Fax: +1 (780) 641-1601


(M.S.) Email: [email protected]

27
Present Addresses

† National Research Council Canada - Nanotechnology Research Centre, 11421 Saskatchewan

Drive, Edmonton, AB, Canada T6G 2M9

REFERENCES

[1] G. Hurwitz, D.J. Pernitsky, S. Bhattacharjee, E.M.V. Hoek, Targeted removal of dissolved

organic matter in boiler-blowdown wastewater: Integrated membrane filtration for

produced water reuse, Ind. Eng. Chem. Res. 54 (2015) 9431–9439.

doi:10.1021/acs.iecr.5b02035.

[2] R.G. Pillai, N. Yang, S. Thi, J. Fatema, M. Sadrzadeh, D. Pernitsky, Characterization and

comparison of dissolved organic matter signatures in steam-assisted gravity drainage

process water samples from athabasca oil sands, Energy and Fuels. 31 (2017) 8363–8373.

doi:10.1021/acs.energyfuels.7b00483.

[3] S. Guha Thakurta, A. Maiti, D.J. Pernitsky, S. Bhattacharjee, Dissolved organic matter in

steam assisted gravity drainage boiler blow-down water, Energy and Fuels. 27 (2013)

3883–3890. doi:10.1021/ef4002154.

[4] A. Maiti, M. Sadrezadeh, S. Guha Thakurta, D.J. Pernitsky, S. Bhattacharjee,

Characterization of boiler blowdown water from steam-assisted gravity drainage and

silica-organic coprecipitation during acidification and ultrafiltration, Energy and Fuels. 26

(2012) 5604–5612. doi:10.1021/ef300865e.

[5] J.. Fanaritis, J.P., Kimmel, Review of once-through steam generators, J. Pet. Technol. 17

(1965).

[6] K.W. Gwak, W. Bae, A review of steam generation for in-situ oil sands projects,

28
 Geosystem Eng. 13 (2010) 111–118. doi:10.1080/12269328.2010.10541317.

[7] D. Flynn, Nalco water handbook, Third Edit, McGraw-Hill: New York, Chicago, San

Francisco, Lisbon, London, Madrid, Mexico City, Milan, New Delhi, San Juan, Seoul,

Singapore, Sydney, Toronto, 2009.

[8] J.P. Van Der Hoek, A. Klapwijk, Reduction of regeneration salt requirement and waste

disposal in an ion exchange process for nitrate removal from ground water, Water. 9

(1989) 203–210.

[9] J. Kim, M.M. Benjamin, Modeling a novel ion exchange process for arsenic and nitrate

removal, Water Res. 38 (2004) 2053–2062. doi:10.1016/j.watres.2004.01.012.

[10] A. Al-As’Ad, M.M. Husein, Treatment of steam-assisted gravity drainage water using low

coagulant dose and fenton oxidation, Environ. Technol. (United Kingdom). 35 (2014)

1630–1638. doi:10.1080/09593330.2013.877086.

[11] M. Sadrzadeh, J. Hajinasiri, S. Bhattacharjee, D. Pernitsky, Nanofiltration of oil sands

boiler feed water: Effect of pH on water flux and organic and dissolved solid rejection,

Sep. Purif. Technol. 141 (2015) 339–353. doi:10.1016/j.seppur.2014.12.011.

[12] C. Luo, Distribution of velocities and velocity gradients in mixing and flocculation

vessels: Comparison between LDV data and CFD predictions., PhD Thesis. (1997).

doi:10.3141/2100-07.

[13] J. Bratby, FLOCCULATION, (n.d.).

[14] G.. Semerjian, L., Ayoub, High pH magnesium coagulation flocculation in wastewater

treatment, Adv. Environ. Res. 7 (2003) 389–403.

[15] D. Vandamme, I. Foubert, I. Fraeye, B. Meesschaert, K. Muylaert, Flocculation of

chlorella vulgaris induced by high pH: Role of magnesium and calcium and practical

29
 implications, Bioresour. Technol. 105 (2012) 114–119.

doi:10.1016/j.biortech.2011.11.105.

[16] J. Leentvaar, M. Rebhun, Effect of magnesium and calcium precipitation on coagulation-

flocculation with lime, Water Res. 16 (1982) 655–662. doi:10.1016/0043-1354(82)90087-

2.

[17] G.M. Ayoub, S. Lee, B.E.N. Koopman, Seawater induced algal flocculation, 20 (1986)

1265–1271.

[18] I. Latour, R. Miranda, A. Blanco, Silica removal with sparingly soluble magnesium

compounds. Part II, Sep. Purif. Technol. 149 (2015) 331–338.

doi:10.1016/j.seppur.2015.05.037.

[19] M.N. Sepehr, M. Zarrabi, H. Kazemian, A. Amrane, K. Yaghmaian, H.R. Ghaffari,

Removal of hardness agents, calcium and magnesium, by natural and alkaline modified

pumice stones in single and binary systems, Appl. Surf. Sci. 274 (2013) 295–305.

doi:10.1016/j.apsusc.2013.03.042.

[20] L. Shamaei, B. Khorshidi, B. Perdicakis, M. Sadrzadeh, Treatment of oil sands produced

water using combined electrocoagulation and chemical coagulation techniques, Sci. Total

Environ. 645 (2018) 560–572. doi:10.1016/J.SCITOTENV.2018.06.387.

[21] Y. Zeng, C. Yang, W. Pu, X. Zhang, Removal of silica from heavy oil wastewater to be

reused in a boiler by combining magnesium and zinc compounds with coagulation,

Desalination. 216 (2007) 147–159. doi:10.1016/j.desal.2007.01.005.

[22] N.D. Tzoupanos, a I. Zouboulis, Coagulation-Flocculation processes in water /

wastewater treatment : The application of new generation of chemical reagents, 6th

IASME/WSEAS Int. Conf. HEAT Transf. Therm. Eng. Environ. (2008) 309–317.

30
[23] M. Tadokoro, S. Tsumeda, N. Tsuhara, H. Nakayama, Y. Miyazato, K. Tamamitsu, A.

Vinu, K. Ariga, Electric double-layer capacitance of carbon nanocages., J. Nanosci.

Nanotechnol. 9 (2009) 391–395. doi:10.1166/jnn.2009.J042.

[24] P.A.R. Tom D. Reynolds, Unit operations and processes in environmental engineering,

The second, PWS Publishing Company, 1996.

[25] L.M. Soldatkina, A.N. Purich, V.V. Menchuk, Adsorption of dyes on magnesium

hydroxide, Adsorpt. Sci. Technol. 19 (2001) 267–272. doi:10.1260/0263617011494150.

[26] B.Y. Gao, Q.Y. Yue, Y. Wang, W.Z. Zhou, Color removal from dye-containing

wastewater by magnesium chloride, J. Environ. Manage. 82 (2007) 167–172.

doi:10.1016/j.jenvman.2005.12.019.

[27] G.M. Ayoub, S.W. BinAhmed, M. Al-Hindi, F. Azizi, Coagulation of highly turbid

suspensions using magnesium hydroxide: Effects of slow mixing conditions, Environ. Sci.

Pollut. Res. 21 (2014) 10502–10513. doi:10.1007/s11356-014-2857-0.

[28] D.B. Kent, M. Kastner, Mg2+ removal in the system Mg2+-amorphous silica-H20 by

adsorption and Mg-hydroxysilicate precipitation, Geochim. Cosmochim. Acta. 49 (1985)

1123–1136.

[29] J. Bratby, COAGULATION and FLOCCULATION, Third Edit, Published by Uplands

Press Ltd, 1980.

[30] S. Inamdar, N. Finger, S. Singh, M. Mitchell, D. Levia, H. Bais, D. Scott, P. McHale,

Dissolved organic matter (DOM) concentration and quality in a forested mid-Atlantic

watershed, USA, Biogeochemistry. 108 (2012) 55–76. doi:10.1007/s10533-011-9572-4.

[31] J. Weishaar, G. Aiken, B. Bergamaschi, M. Fram, R. Fujii, K. Mopper, Evaluation of

specific ultra-violet absorbance as an indicator of the chemical content of dissolved

31
 organic carbon, Environ. Sci. Technol. 37 (2003) 4702–4708. doi:10.1021/es030360x.

[32] A.C. Ziegler, Issues related to use of turbidity measurements as a surrogate for suspended

sediment, Water. (2002) 2001–2003. http://ks.water.usgs.gov/Kansas/rtqw/.

[33] J.C. McCarthy, T.E. Pyle, G.M. Griffin, Light transmissivity, suspended sediments and

the legal definition of turbidity, Estuar. Coast. Mar. Sci. 2 (1974) 291–299.

doi:10.1016/0302-3524(74)90019-X.

[34] A.A. Bukhari, Investigation of the electro-coagulation treatment process for the removal

of total suspended solids and turbidity from municipal wastewater, Bioresour. Technol. 99

(2008) 914–921. doi:10.1016/j.biortech.2007.03.015.

[35] T.J. Mohammed, E. Shakir, Effect of settling time, velocity gradient, and camp number on

turbidity removal for oilfield produced water, Egypt. J. Pet. (2017).

doi:10.1016/j.ejpe.2016.12.006.

[36] M. Yan, D. Wang, J. Ni, J. Qu, W. Ni, J. Van Leeuwen, Natural organic matter (NOM)

removal in a typical North-China water plant by enhanced coagulation: targets and

techniques, Sep. Purif. Technol. 68 (2009) 320–327. doi:10.1016/j.seppur.2009.05.021.

[37] G.D. Lanciné, K. Bamory, L. Raymond, S. Jean-Luc, B. Christelle, B. Jean, Coagulation-

Flocculation treatment of a tropical surface water with alum for dissolved organic matter

(DOM) removal: influence of alum dose and pH adjustment, J. Int. Environ. Appl. Sci. 3

(2008) 247–257. http://www.jieas.com/fvolumes/vol081-4/3-4-6.pdf.

[38] A.L. Ahmad, S. Ismail, S. Bhatia, Water recycling from palm oil mill effluent (POME)

using membrane technology, Desalination. 157 (2003) 87–95. doi:10.1016/S0011-

9164(03)00387-4.

[39] E. Saljoughi, M. Sadrzadeh, T. Mohammadi, Effect of preparation variables on

32
 morphology and pure water permeation flux through asymmetric cellulose acetate

membranes, J. Memb. Sci. 326 (2009) 627–634. doi:10.1016/j.memsci.2008.10.044.

[40] Z. Almansoori, B. Khorshidi, B. Sadri, M. Sadrzadeh, Parametric study on the

stabilization of metal oxide nanoparticles in organic solvents: A case study with indium

tin oxide (ITO) and heptane, Ultrason. Sonochem. 40 (2018) 1003–1013.

doi:10.1016/j.ultsonch.2017.09.012.

[41] C.S.. Fitzpatrick, E. Fradin, J. Gregory, Temperature effects on flocculatio, using different

coagulants, Water Sci. Technol. 50 (2004) 171–175.

http://onlinelibrary.wiley.com/doi/10.1002/cbdv.200490137/abstract.

[42] J.K. Morris, W.R. Knocke, Temperature effects on the use of metal-ion coagulants for

water-treatment, J. Am. Water Work. Assoc. 76 (1984) 74–79. doi:10.0.

[43] S. BinAhmed, G. Ayoub, M. Al-Hindi, F. Azizi, The effect of fast mixing conditions on

the coagulation–flocculation process of highly turbid suspensions using liquid bittern

coagulant, Desalin. Water Treat. 53 (2015) 3388–3396.

doi:10.1080/19443994.2014.933043.

[44] P.T. Spicer, S.E. Pratsinis, J. Raper, R. Amal, G. Bushell, G. Meesters, Effect of shear

schedule on particle size, density, and structure during flocculation in stirred tanks,

Powder Technol. 97 (1998) 26–34. doi:10.1016/S0032-5910(97)03389-5.

[45] B. Ramavandi, Treatment of water turbidity and bacteria by using a coagulant extracted

from Plantago ovata, Water Resour. Ind. 6 (2014) 36–50. doi:10.1016/j.wri.2014.07.001.

[46] Y. Liao, X. Tang, Q. Yang, W. Chen, B. Liu, C. Zhao, J. Zhai, H. Zheng, Characterization

of an inorganic polymer coagulant and coagulation behavior for humic acid/algae-polluted

water treatment: polymeric zinc–ferric–silicate–sulfate coagulant, RSC Adv. 7 (2017)

33
 19856–19862. doi:10.1039/C7RA00232G.

[47] R. Li, C. He, Y. He, Preparation and characterization of poly-silicic-cation coagulant from

industrial wastes, Desalination. 319 (2013) 85–91. doi:10.1016/j.desal.2013.03.024.

[48] S. Aoudj, N. Drouiche, M. Hecini, T. Ouslimane, B. Palaouane, Coagulation as a post-

treatment method for the defluoridation of photovoltaic cell manufacturing wastewater,

Procedia Eng. 33 (2012) 111–120. doi:10.1016/j.proeng.2012.01.1183.

[49] J.T. Gourley, W.A. Runciman, Multiphonon infrared absorption spectra of MgO and CaO,

J. Phys. C Solid State Phys. 6 (1973) 583–592. doi:10.1088/0022-3719/6/3/021.

[50] R. Choudhary, S. Koppala, S. Swamiappan, Bioactivity studies of calcium magnesium

silicate prepared from eggshell waste by sol-gel combustion synthesis, J. Asian Ceram.

Soc. 3 (2015) 173–177. doi:10.1016/j.jascer.2015.01.002.

[51] V. Fournier, P. Marcus, I. Olefjord, Oxidation of magnesium, SIA. 34 (2002) 494–497.

doi:10.1002/sia.1346.

[52] I. Al-Mutaz, I. Al-Anezi, Silica removal during lime softening in water treatment plant,

Int. Conf. Water Resour. Arid Environ. (2004) 1–10.

http://repository.ksu.edu.sa/jspui/handle/123456789/3456.

[53] I. Al-Mutaz, I. Al-Anezi, Silica removal during lime softening in water treatment plant,

Int. Confference Water Resour. Arid Environ. (2004) 1–10.

http://repository.ksu.edu.sa/jspui/handle/123456789/3456.

[54] F. Mohammadesmaeili, M.K. Badr, M. Abbaszadegan, P. Fox, Byproduct recovery from

reclaimed water reverse osmosis concentrate using lime and soda-ash treatment, Water

Environ. Res. 82 (2010) 342–350. doi:10.2175/106143009X12487095236919.

34
Highlights

 The new process design uses existing wastewater in the SAGD process for the treatment
of another waste stream.
 Explored the feasibility of using IER wastewater as a coagulant for the treatment of BBD
water.
 The removal efficiency of 81.1% and 98.6% were obtained for organic matter and silica,
respectively.
 Characterization of sludge materials was performed to identify their chemical
constituents and resource recovery.
 The new process improves the overall water treatment efficiency, minimizing operation
cost and energy consumption.