Sustainable life-cycle assessment of mixing approaches in water storage tanks

Poor mixing in water storage tanks can cause stagnant zones that could pose negative public health effects. The present study uses Life Cycle Assessment to decide among the only three mixing options available, namely sprinkler, multiple inlets, and a mechanical mixer for the first time. These options were compared using different life-cycle assessment (LCA) tools using an 80-year lifetime as the functional unit while assuming that all three options result in acceptable water quality. Using SimaPro modeling software as well as the IPCC 2013 GWP 100a V1.0 and Cumulative Energy Demand methods, these three mixing approaches were compared with and without waste recycling. Results showed that application of a sprinkler is the least expensive option. Damage-cost analyses for categories of human health, ecosystem quality, and resources showed that a sprinkler caused the least damage and cost, while a mixer resulted in the most damage and cost.


INTRODUCTION
coliform bacteria occurred more frequently when water temperature was increased from 5 C to >20 C. Higher microbial growth in storage tanks can also be observed because of tank design as the tanks are usually over-sized to accommodate emergency supplies, resulting in stagnant, microbe-present water that may only be used in infrequent emergency situations. This stagnation may also allow for sediment accumulation that can work as a growth medium for bacteria (Boulos et  indicated that the novel sprinkler system resulted in parallel downward streamlines that removed most of the stagnation zones in the tank. Another example of a proposed mixing system is using a multiple-inlet system, consisting of an internal pipe with multiple outlets into the tank (Beatrix ). Such a system can provide enough mixing with a properly designed passive mixing system, which has often 10 percent volume turnover or less. In this system, mixing in the tank occurs during the fill cycles, occurring from momentum of flow being injected into the tank water. The velocity difference between the inlet jet and tank water creates turbulence, which develops circulation patterns throughout the tank. Another effort to increase the level of mixing is using a mechanical mixer, consisting of a mixing arm being rotated by an electric motor (Fisk ). Using a mixer can significantly improve the degree of mixing in storage tanks. The mixer is an active, submersible mixing system for acceptable drinking water quality in storage tanks and reservoirs. The mixer can rapidly eliminate stratification, uniformly distribute disinfectants and prevent water quality deterioration. Efficient and effective mixing of large volumes is made possible by the mixer, which establishes a stable flow structure throughout the storage volume.
Design of tanks for particulate contaminant behavior can also have an effect (Won et al. ). Tank mixing has been studied experimentally (Alizedah Fard & Barkdoll ) as well as numerically (Huang et al. ). found that LCA can be used for cost optimization of environmental impact for waste water treatment plants.
Since no studies of the environmental impacts of mixing approaches have been performed, this paper describes lifecycle environmental and cost assessments that were performed for the aforementioned mixing approaches to enhance the level of mixing in storage tanks. The goal of this study is, for the first time, to inform the decision-makers about different aspects of these three studied systems to aid in selection of a mixing system that includes not only initial costs, but also ingoing costs and environmental impact.
Thus, the analysis and the life-cycle assessments are broadly applicable to municipal authorities and water utilities.

PROCEDURE
The main goal of this analysis is to evaluate and compare the environmental impacts of three approaches to minimize stagnation in water storage tanks. A functional unit with an operational life span of 80 years was used in this LCA analysis. Eighty years is a typical design life for these types of systems (Maupin et al. ). This is justified, since the current analysis assumes all mixing options provide the conditions necessary to achieve the same level of mixing in the tank. A tank size of 5,725 m 3 (27 diameter × 10 m height) was used for all devices to ensure comparability. The effect of the mixer materials of plastic, steel, and rubber on the water quality are not considered here since these materials are commonly used with water and, therefore, are not considered dangerous to water quality. The cost and environmental impact of the manufacturing of the materials is, however, considered thereby necessitating an LCA approach. This study investigates the environmental impacts of the three technologies related to overall energy consumption, greenhouse gas emissions, human health, ecosystem quality, and resources. This study also compares their life-cycle assessment results to identify the most sustainable mixing approach for water tanks suffering from stagnation.
The system boundary is defined as extending from raw materials' acquisition to the manufacturing and usage of the final product ( Figure 1).  Figure 2 shows the methodology of this study. Phase one includes energy inputs and materials used. This phase was used as inventory for LCA and LCC. Phase two assesses the impacts of the process/system/activity on human health, ecosystem, and resources. In this stage, the Cumulative Energy Demand method was utilized to evaluate the environmental effects based on the energy consumption in different materials/stages. The forms of energy considered here are renewables (hydropower and wind), fossil fuels (natural gas and coal), and the US energy grid mix. In the third stage, a sensitivity analysis was performed to identify the critical role of the energy source.

Environmental life-cycle assessment
All major life-cycle stages are considered for each mixing option, including the production of construction materials, equipment, and operation and maintenance. The specific construction processes considered for the existing tank are preparing the site, and disinfecting the tank before use.
Steel and PVC pipes from the old mixing systems would be sent to the recycler. The operation and maintenance stage consists of draining and disinfecting the tank and the replacement of equipment. Energy consumption by the municipal pumping station was omitted from the analysis because there were no differences between the options.
Comparative life-cycle assessments rely on a measure of the function of the studied system and provides a reference on which the inputs and outputs can be related. This study uses a functional unit that assumed 80 years of operating time for all three options and equipment replacement was calculated based on this period. Each mixing system was considered for a tank with a volume of 5,725 m 3 (27 m diameter × 10 m height) and it is assumed that all systems can provide the highest-quality of finished drinking water. In other words, all material inputs and emissions are normalized based on the capacity of the highest-quality water each approach can provide. As there is no technical (water quality) comparison between these options in the literature, this study assumes that all these mixing options are able to provide the same degree of mixing in the studied tank. In addition, as access to clean energy is not possible everywhere, a US mixed energy grid was assumed for energy source and the results were presented based on this assumption in the previous sections.

Mixer system
Product manufacturing: This system consists of two major parts, a mixer which is installed in the tank and a control Waste materials and waste scenarios: For this system, the main source of waste was from stainless steel and copper, and rubber. Two waste scenarios were considered.
In the first one, all materials were considered recyclable and in the second one nothing was recycled. A summary of the materials used for mixer system is tabulated in Table 1.

Multiple inlet system
Product manufacturing: This system consists of two main parts: first, the main steel pipe, which is transporting inflow inside the tank; second, the rubber inlets, which provide turbulent flows at multiple locations.
Product usage: It is estimated that rubber inlets need to be replaced with new ones after 10 years of service. As this system is not a mechanical system, it does not consume energy.
Waste materials and waste scenarios: The main source of waste considered is from stainless steel and rubber inlets. Two waste scenarios were considered, both recyclable and non-recyclable. In the first one, all materials were considered recyclable and in the second one non-recyclable.
Summary of used materials for multiple-inlet system is described in Table 2.

Sprinkler system
Product manufacturing: In this system, the main inlet riser steel pipe transports water from outside of the tank to the top sprinkler distribution system; then, the PVC sprinkler systems work as a distribution system.
Product usage: Replacement of PVC pipes every 10 years was included, but the riser pipe was not replaced.
No electricity was included due to the passive nature of the sprinkler system.
Waste materials and waste scenarios: In this system, the main source of waste comes from PVC pipes and the steel riser pipe. Two waste scenarios were considered: all materials recycled and nothing recycled. The summary of the used materials for the sprinkler system is described in Table 3.

Impact assessments
SimaPro modeling software was used for the impact assessment. IPCC 2013 GWP 100a V1.0, Cumulative Energy Demand (CED), and Eco-indicator 99 were used to study all mixing approaches. The first two methods were chosen  health, ecosystem quality, and resources were compared.
All mixing approaches are studied in the case of the damage they pose to three categories. These categories are human health, ecosystem quality and resources.
Two waste scenario models were created for each product: in one scenario, no materials were recycled and in the other one scenario materials were recycled. It is important to model both scenarios to see the effects of the ideal and worst-case scenarios. It should be noted that recycling all materials would not be possible in some areas as recycling facilities may not be available.   overall energy consumption and greenhouse gas emissions measured by the methods are summarized in Table 4.

RESULTS AND DISCUSSION
As can be seen in

Cost analysis
Alongside the preceding technical analysis, it is important to estimate the life-cycle cost for each product. Table 6 provides the associated cost for each approach (Pax Water Technologies ; Red Valve ).
The mixer's manufacturing cost is the highest of all three systems. This high cost emanates from the fact that the mixer's motor, stainless steel equipment, and control unit are expensive. The operating cost of the mixer is also the highest one, as the motor needs to be changed every five years. In addition, the electricity cost of the mixer adds to the motor cost and makes the mixer the most expensive option considered.

Damage-cost analysis
It is important to compare all options from both cost and life-cycle assessment points of view. Competitive maps are useful tools for decision making. These maps have two  axes, one is for cost and the other one is for a qualitative factor. In this analysis method, the higher the value of cost or the qualitative factor, the less favorable is the process or product. In fact, each option has a coordinate on the map and the closer the coordinate to the origin the better the option is. It means this option has the minimum damage and cost in comparison to other options. Table 7 and Figures 8-10 show competitive maps for health damage-cost, ecosystem quality-cost, and resource-cost, respectively.
According to the results, the sprinkler system is the best option based on all three analyses and the mixer option was Since electricity generation does not pose significant damage to the three categories for the case of clean energies, material usage is the main factor to be considered for lifecycle assessment. On the other hand, when fossil fuels (coal or natural gas) were considered as the main source of energy, the sprinkler option had the minimum damage for most of the categories. Therefore, the source of electricity can significantly change the results.

Sensitivity analysis
To lessen the uncertainty in the electricity source, a sensitivity analysis of three different scenarios was performed.
The three sources were as follows: Scenario 1-Energy demand was only supplied from US mixed energy grid.
Scenario 2-Energy demand was only supplied from hydropower.
Scenario 3-Energy demand was supplied equally from a US mixed energy grid and hydropower.       The scale shows a relative environmental impact factor. An impact factor value of 100 is assigned to the highest impact option and the other values are relative to that.

RECOMMENDATIONS FOR FUTURE IMPROVEMENT
In the current study, the same degree of mixing was assumed for all options. However, further analysis is necessary to compare these options when they have different mixing efficiencies. Such an analysis can be done by EPANET software.
In addition, by modeling the water distribution network in EPANET software, other important health factors (like chlorine concentration) in the network can be calculated.
Therefore, we can perform a broad analysis not only with the available indicators in Simapro software, but also with other health indicators. As a result, a comprehensive analysis could be performed comparing these options to enable making a conclusion about the feasibility of each one.

DATA AVAILABILITY STATEMENT
All relevant data are available from an online repository or repositories in Alizedah Fard () found at https://digitalcommons.mtu.edu/etdr/706/.