One of the major drivers for this project is the changes made with the release of Directive 60 from the Alberta Energy Regulators (AER). The directive changes the allowable release of methane emissions, with a particular focus on instrumentation and chemical pumps. The most relevant section of the directive reads:

This provided the framework for tackling the problem, the implication of the above is that the greenfield (newly constructed) production wells will need to either have a vapor recovery system, switch to electrical powered systems or switch to air power pneumatic systems.

Through the generosity of our sponsor, we were provided detailed schematics (PFDs and P&IDs) of the inner components of the current well designs. Due to the proprietary nature of this technology, the team has made a PFD representation of the wellhead system under analysis. The diagram below shows the processing infrastructure to one of the production wells. The emissions produced from the asset are highlighted by red circles. 

A mixture of gas, water and condensate are produced to surface from a reservoir, deep beneath the earth. This mixture is routed to a heated sand separator unit which removes many of the solids, such as sand. After leaving the sand separator, the pressure of the mixture is reduced in a gas separator vessel. This allows the heavier parts, the water and condensate, to drop to the bottom while the gas exits the top and travels down the sales line. The condensate is measured and, as it is of high value, reintegrated into the sales line downstream. The water is sent to a produced water tank where it is trucked out at a later date.

Analyzing the data provided to us, we investigated each part of the production system for potential emission sources. There were four emission culprits identified from regular operations: pneumatic instrumentation, pneumatic chemical pump, catalytic heaters and the produced water tank. These sources are described in more detail below.

Through investigative analysis, we evaluated a myriad of alternatives in the interest of arriving at a solution. It was found that the most likely path to success was to convert the pneumatic instrumentation and pumps to electrical power and analyze vapor recovery technology for a high-pressure application.

A variety of pneumatic devices are used in the operation of the gas pad, including controllers, pressure and level switches, transducers, valves, etc.​ These pneumatic devices use pressurized natural gas for actuation and signals, and will release the natural gas to the atmosphere as part of normal operations.​ High-bleed devices will bleed and emit more than 6 standard cubic feet (scf) of methane per hour and can average about 4 tonnes of methane bled per year. ​

Catalytic heaters work by combusting natural gas at low temperature. There is no flame and the only products of the reaction are CO₂ and water vapor.​
Catalytic heaters do not have a vent, so direct capture of the emissions is not plausible. The original design uses three heaters, one in the sand separator building and two in the gas separator building.

Methanol is injected into the gas lines to prevent the formation of hydrates, which occur mainly due to pressure and temperature fluctuations. When formed, hydrates can halt production and may have to be “pigged” out of the line. There are three types of chemical pumps: pneumatic, low-bleed pneumatic, and electric.​ The design uses a pneumatic and solar powered pump for downhole and sales line hydrate mitigation.

Water drops out of the gas in the separator unit and is moved to a produced water tank, which is open to atmosphere outside the gas separator building. The pneumatic fuel gas from the instrumentation and pump are also routed to the tank. This is done as a safe venting practice so there is no risk of accumulating flammable gasses in the building. The water goes from high pressure to low pressure and “flashes” the solution gas to atmosphere.

After extensive research, the most applicable turbine on the market is a C30 microturbine, developed by Capstone Turbine Corporation. This microturbine replicates an open Brayton Cycle, which can be modelled as a closed cycle. To determine cycle efficiencies, we will be assuming the case of an ideal Brayton Cycle. 

Using the standard cold-air assumption, i.e. constant specific heats of air (k=1.40), the ideal efficiency can be modeled vs. the pressure ratio with the following equation: 

Using a design pressure ratio, of 3.39, from a given research paper the ideal efficiency is ~27%. 

This efficiency value corresponds with corresponds with the manufacturer’s specifications. 

As turbines operate most efficiently at their rated power at steady state, we will require a buffer system to store energy to power the pad while the turbine is off. 

Due to the impracticality of turbines available on the market, we are forced to source a microturbine that is well oversized for our pad design.  

The power requirement for a four-well pad is ~24.5 kWh per day. At the optimal running conditions, the C30 is rated for 30 kW and capable of producing 720 kWh per day. This is oversized for our application. 

Looking for a real-world solution, we opted for the C30 30kW microturbine produced by Capstone Turbine Corporation. This microturbine features air bearings and operates on an open Brayton cycle. 

This solution also has a small physical footprint on site, as the turbine unit occupies an area of 2.56m² and comes as standalone unit that can have programmed or remote-controlled operation. This technology is well utilized in industries such as telecom and petroleum, and so has the benefit of precedence in industry. 

Some comparisons to the pressure ratios found in conventional turbines reveal that these turbines are designed with a lower pressure ratio, 3.39, vs 5-20 for conventional turbo machinery. Due to this, the efficiency is 27%, which is fairly close to the listed efficiency at rated power. 

The major issue in the implementation of this design is the disparity between the turbine generation capacity and the pad power consumption. Turbines must be sized closely to their application as they operate best at steady state, near the rated power, for optimal efficiency and longevity. In the case of a C30, the turbine cannot be throttled without being penalized with low overall efficiencies (BELOW 20%) and shorter operational lifespan. 

To take full advantage of this technology, would be to utilize it to power a web of 4-well pads. This would allow this system to utilize its high efficiency and low emissions potential, however, to scale it to our application, we will need a battery bank to act as a buffer to store energy. 

Using the same battery system as the solar system, the turbine would require 32 batteries to power the pad for ~3 days. Based on an average battery system charging rate of ~125 A, we calculated that the batteries would require 27 hours to charge, utilizing ~3.04 kWh of the turbine output per hour. Conversely, the pad would utilize ~1.02 kWh of the turbine output per hour. Based on the load demands, the system would utilize 4.06 kWh of the turbine output per hour, only 13.5% of the turbine’s rating.

Qnergy offers three commercial models of its Stirling engine, the PowerGen 600, 1200 and 5650. Each unit has its advantages and applications in relation to the power supply for gas production. Below is a table that shows the different applications based on the size of the generator.

The graph below shows CO₂ emissions from the engine as it relates to changes in ambient temperature. Built from correlations based on information from the Qnergy website, it can be concluded that the PowerGen 600 and 1200 have small variation in their emissions. However, the PowerGen 5650 is much more sensitive.

The aggressive fuel efficiency curve for the 5650 shows why the engine is sensitive in terms of ambient temperature. The temperature differential which operates the engine is easier to achieve at lower temperatures (requires less fuel to produce the same energy) and all the efficiency curves will display the exponential characteristic. The 5650 is engineered in such a way that the profile is displayed at realistic temperatures.

Even with the higher efficiency, two PowerGen 600 units to power the four well pad would be the optimal choice. The 5650 is an oversized unit for the application and thus is not selected. Using the 600 units provides the least wasted work while meeting the power requirement of the wells.

Qnergy also offers a remote monitoring service that comes standard with the PowerGen units. This monitoring service provides real-time status updates of the units and can alert operators when faults occur. The picture below is an example of the remote monitoring data available.

Another advantage of the PowerGen unit is that they are reliable enough to not require a battery intermediary system to support the well, something that both the PV solar units and turbine units will require. The team sized a battery pack capable of supporting well operations for 8.5 hours for maintenance and emergencies associated with the Stirling engine.

An economic analysis was performed on each solution and then compared to each other to find the economically optimal solution.

An important aspect of solar power that should be considered is downtime due reliability issues. With some assumptions, this issue was quantified in our economic analysis. The assumptions made are listed below:

• Each year, the system will be down for 4 days in winter (storms) and 4 days in summer (forest fires)
• Average gas production per well is 80 e³m³/day
• Average gas price is ~0.08 $/m³

The results estimate an average loss in revenue of ~$50,000/year per well.

The graph below represents the total reduction of CO₂e per year, per well in tonnes of CO₂e compared to the existing solution, currently in use. The PV solar solution shows the highest reduction of 97%, the Sterling engine solution shows the second highest reduction of 93%, and the turbine solution shows the lowest reduction of 87%. While the total reductions between the three solutions vary, it is clear that all three represent a significant reduction in CO₂e over the existing solution.

The following graph is a representation of the year 1 capital costs of the three design solutions, broken down by cost categories. It is interesting to note that while capital costs of the PV solar solution are relatively lower compared to the other two solutions, the installation and battery costs are relatively higher.

The next set of graphs represent the costs of each solution over a span of 10 years and the Present Value elimination costs of 1 tonne of CO₂e per year for each solution. The capital costs in year 5 and year 10 represent the purchase of new batteries for replacement. The PV solar system has the lowest initial capital cost, however, factoring in the expected lost revenue due to system downtime, the option bares the highest cost of ownership at ~$3,300/CO₂e. The reliability of the Stirling engine option reduces the amount of revenue loss due to downtime. Over the 10 year forecast, this alternative has the lowest cost of ownership resulting in ~$800/CO₂e.

The table below summarizes the total present costs and elimination costs for each design. For comparison’s sake, the PV solar solution without downtime was added. We can see that without the losses from the downtime, the elimination costs become the lowest of the three solutions. However, this scenario with no downtime is unrealistic and is only used for a theoretical comparison. 

PowerGen 1200 Case

The PowerGen series is not included in the above comparison but has favorable costs. It could be used in certain applications, however, in the oil and gas industry modularity is prized. The single power supply for a pad is not as appealing as a per well power solution. The 600 is in between and offers the best utilization of power generated and modest modularity. The lower capital cost makes the 1200 appealing in specific cases and is why it is included here.

CCA Rule Change

Eligible energy systems and equipment as defined by the requirements of Classes 43.1 and 43.2 of Schedule II of the Income Tax Regulations Act can apply an accelerated Capital Cost Allowance (CCA) rule [11]. Currently any eligible systems and equipment purchased after November 20th, 2018 and implemented before 2023 can apply the ”First-Year Enhanced Allowance” rule that allows 100% of the CCA to be deducted in year 1. Each year after the allowable rate declines until 2028 [11].

Example

Below is an example of the half year rule being omitted and the First-Year Enhanced Allowance being applied in full. The CCA is fully realized in fiscal 2023 with no undepreciated capital cost (UCC) remainder for subsequent years.

Below is an example of the half year rule being omitted and the First-Year Enhanced Allowance being applied in the 2024. This results in the full amount of capital cost being available for CCA purposes and an accelerated CCA rate of 75% in the first year. In this case there is a remainder of UCC and it is depreciated at year over year enhanced allowance rate of 50% for the remainder of the UCC balance.

The final design recommendation is to change the chemical pump and instrumentation to electronic equivalents. The new system design will be energized by two Qnergy PowerGen 600 engines. This recommendation does not achieve the initial target of a net zero pad, however, the design does reduce emissions by an estimated 92% and is projected to eliminate downtime due to power related issues. The majority of real-world problems with power at these remote sites are related to batteries. By eliminating the large reliance on battery banks, production will experience less interruptions thus maximizing revenue. It takes a very small amount of downtime to make up for the larger capital cost of the PowerGen 600 units. The re-usability of the units and their prolonged length of the life expectancy provides assurance that surplus revenues will be achieved when compared to other alternatives.

A Venturi Jet Ejector (VJE) based VRU is a potential avenue for capturing the relatively small amounts of emissions given off by the produced water tank and catalytic heaters. As the produced gas moves through the unit, a suction force is created due to a negative pressure phonon around the diffuser. This allows the emissions captured in the produced water tank to be sucked into the produced gas line for processing further down the line. The picture below illustrates the operation of a VJE.

The major disadvantage with these systems is the large drop in pressure across the device. To counteract this drop in pressure, a compressor must be added to the line after the VRU.

An inline booster compressor would be appropriate for this application, however after sizing a unit, the motor required ~410 kW of power. Given the remoteness of the gas wells, the additional capital costs, and the high amount of power input required this VJE based VRU system is rendered unfeasible.

The Electroswing Adsorption System is a developing technology that uses electrochemical cells to capture carbon dioxide as it passes through the system. The exhaust from the catalytic heaters and power generator will be passed through the Electroswing Adsorption System. As the exhaust passes through the system, an electrical current will be induced. This electrical current allows the electrochemical cells to capture and hold onto the carbon dioxide within the exhaust fluid while allowing the remainder of the fluid to pass through. The electrical current can then be reversed, and the carbon dioxide is released and stored in a vessel on-site. With current testing showing ~70% effectiveness in capturing carbon dioxide, this technology could become a large contributor in the global reduction of carbon dioxide.