Strategic Selection of RNG Refinement Systems

By Anu D. Vij and Mike Pawlowski, Ship & Shore Environmental

Special to The Digest

There is increasing emphasis on Renewable Natural Gas (RNG) applications in today’s Green Economy. Technology is moving at a rapid pace for RNG. Refinement of streams from a variety of sources of methane into quality pipeline gas requires a certain amount of planning for the destruction of off gas or tail gas. The objective is to extract as much of the methane from the source as possible while removing the impurities from the incoming stream. With a variety of methods available for removal of contaminants, a certain amount of planning on destruction of the byproducts from the purification processes is required. Economics of the refinement methods used will dictate the process selection for each stream. Every method will produce a certain quality and quantity of tail gas which represents the components removed by refinement.

Importance of Proper Selection

Each refinement system has its advantages and disadvantages. Selection of the proper method of refinement is driven based on the quality of the incoming stream. The impact these processes have on the quality and quantity of tail gas vary greatly. Building a matrix of the impurities in the incoming steam and methods of removal is helpful. The end result of quality methane can have different paths depending on treatment vs separation. In example, if problematic constituents such as H₂S and Siloxanes end up in the tail gas they can cause issues of corrosion, fouling, as well as post-combustion treatment.  When possible, treatment to remove these from the process is preferred when compared to separation which would push them into the tail gas.

When H₂S combusts, SO₂ is formed and corrosion will occur in unprotected A36 steel. This may require higher grades of metallurgy for contact surfaces and with high amounts, post combustion scrubbing of SO₂ may be required. If high amounts of SO₂ are present in a direct fired oxidation system, a pre-scrubber quench would be required. All of these outcomes increase the cost of ownership of the abatement system.  These additional costs need to be weighed against systems which remove H₂S from being present in the tail gas.

Siloxanes, which are prevalent in landfill gas applications, present a different problem. When siloxanes combust, a very fine grade of SiO₂ (sand) is formed and can deposit on the heat transfer surfaces of regenerative or recuperative oxidizers. This particulate may also require post-oxidation collection if the content is enough to create high levels of particulate. With a direct fired oxidizer, cooling of the flue gas would be required prior to collection adding to the cost of ownership of the abatement system.

Different Refinement Systems

Refinement methods such as Membrane Separation have relatively narrow ranges of quality and quantity of the off gases.  Abatement of tail gasses from these streams are more predictable from both a content as well as a flow perspective.  PSA systems have their advantages, but the off gas quality and quantity tends to have more variation, making the abatement strategy different if not more difficult.  Process variation can be dealt with, but some solutions require supplemental fuel for proper destruction across the entire output spectrum. Typically, an output of tail gas above 6% methane will be most effectively handled by a direct fired thermal oxidizer.  For lower methane equivalent contents, recuperative or regenerative oxidation would be the lowest operating cost solution.

Direct Fired Thermal Oxidizer

A direct fired thermal oxidizer (Fig. 1) operates in the following manner. Once startup conditions are established in the oxidation chamber by the use of a start-up burner, the transition from start-up fuel to tail gas can proceed. The volume of tail gas is increased and the burner cuts back to maintain the proper chamber temperature for effective destruction of the tail gas (typically ~1,500^oF). Combustion conditions are managed by proportional control of the tail gas, and the use of a Variable Frequency Drive (VFD) or dampers to control the air fuel ratio and chamber temperature. This is the simplest form of oxidation, and the lowest initial cost. The limitations present themselves in the form of BTU content of the tail gas. It must be high enough to sustain combustion without the addition of supplemental fuel and secondarily, the volumes of system need to fall into a range that balances retention time and turbulence within the oxidation chamber. Operation outside of these ranges can result in high consumption of supplemental fuel.

Regenerative Thermal Oxidizer

A Regenerative Thermal Oxidizer (Figs. 2 and 3) is a highly energy efficient form of oxidation. Rates of 95% or higher thermal efficiencies can be achieved through the use of heat exchange beds of ceramic modules. A typical RTO has two chambers which cycle between absorbing heat from combustion and preheating the incoming gas.

Depending on the process inlet conditions and based on the concentrations of VOC’s in the system, dilution air and/or a Hot Gas Bypass may be required to properly balance the combustion process for an RTO.

An abundance of caution must be utilized to ensure that hazardous conditions do not exist outside of the 1,500^oF controlled combustion chamber. Managing the LEL within the non-explosive range is critical. RTO systems have fairly good turndown and with the combination of dilution air as well as a Hot Gas Bypass, they can handle most of the wide-ranging swings that can be presented within a renewable gas upgrade facility.

Importance of Vendor Selection

Selecting the proper vendor is critical. Search for one that has the experience and product lines that can encompass the right solution approach. Avoid firms with a one-technology solution. RNG is a great utilization of resources, but at the end of the day the economics will weigh into the equation. The initial cost of abatement equipment is just part of the equation. Operating costs and overall cost of ownership is the driver for most of the economic models currently is use.