News and interest in the opportunity to produce low-carbon hydrogen from natural gas, often referred to as blue hydrogen, has significantly increased in recent months. The blue hydrogen product consists of reforming methane (natural gas) into hydrogen in the presence of water and (sometimes) oxygen. The basic scheme is as follows (lighter blue values indicate possible inputs/outputs, depending on the scheme selected):
There are a number of considerations when imagining and defining a blue hydrogen project that can have a significant impact on the optimal configuration and economics.
Making blue hydrogen consumes energy. The overall energy requirement of making blue hydrogen is about 1.5 – 1.8 GJ of natural gas per GJ of hydrogen product, depending on how low of a carbon footprint you target, the scale of the facility and the technology line-up. This provides the primary “cash cost” portion of making hydrogen.
The valuation of CO₂ credits is critical. Depending on how your organization values CO₂ credits in the period beyond start-up can have a significant impact on technology selection. If the CO₂ credits are believed to have a high value, then the economic calculation will favour capturing very high percentages of CO₂. This must consider that the incremental cost of increasing CO₂ capture will become exponentially more expensive as emissions approach zero or will force technology selection changes. Once you lock in certain technology selections, the cost of increasing CO₂ capture can become prohibitively expensive. A key practice to follow is to perform sensitivities on the pricing of natural gas, hydrogen and CO₂ credits, along with capital costs and other inputs, to determine how sensitive the economics are to such variability. In some cases, aggressive pricing assumptions without considering the risk that said assumptions aren’t quite right can lead to malinvestment chasing perfection.
Consider all the CO₂ Emissions. There are four primary emissions sources that need to be considered:
- Combustion products from natural gas supporting the reforming process (either an SMR furnace, ATR preheat furnace or utility boilers).
- Process CO₂ not captured and routed to fuel, often from PSA tail gas, or left in the H₂ product (either as CO₂ or converted back to CH₄) that will be emitted when the hydrogen is used.
- Combustion products from natural gas or process gases that are routed to utility steam generation or an onsite power cogeneration facility.
- Scope 2 emissions associated with imported power. Depending on the carbon intensity of the local power grid, the carbon credit valuation and desire to approach ‘zero emissions’ may drive the configuration to on-site power generation.
Combustion emissions can be avoided by moving to hydrogen as a combustion fuel, but this leaves the challenge of what to do with the uncaptured process CO₂. Alternatively, combustion emissions could be captured using a flue-gas CO₂ capture process.
What markets/customers are targeted? The markets targeted for blue hydrogen have different requirements when it comes to delivery pressure, purity and quantity.
- If you are blending hydrogen into natural gas, the purity need not be very high. The product could contain some CH₄ or CO₂ (but not much CO) and meet the specifications for such blending. This may allow for the use of methanation in the final purification step, resulting in lower H₂ losses than a PSA unit. However, if the hydrogen is to be routed to power generation or other on-site combustion uses, a PSA option might result in lower overall H₂ losses because hydrogen would not be “consumed” to make methane from CO and CO₂. This may also change the logic in the upstream process scheme on CH₄ and CO slips through the synthesis gas production process.
- If you are producing hydrogen for refinery use, the purity requirement will depend on the requirements of the individual hydroprocessing steps. Many hydroprocessing units can accept purities below 98%, or even down to 95%, and can manage impurities (inerts, CH₄) via purge. But high pressure ULSD units and hydrocrackers often need >99% purity to avoid excess purges and to achieve the desired product specifications. This will require a PSA unit to purify the hydrogen, result in about 10% of the total hydrogen product being available only at lower pressure. If there is on-site power/steam generation, this could be routed there as a low GHG fuel, but that must come into the overall project definition.
- If you are going to produce hydrogen for fuel-cell use (vehicles) or for liquefaction for transport, the purity level must be much higher. This also requires a PSA, but the size of the unit and losses will be more significant than in the high purity refinery use case. Fuel cells require very high purity hydrogen, and liquefaction requires removing materials that would freeze in that process, including nitrogen, oxygen, CO₂ and other species. Liquefaction itself is a very energy intensive process, consuming a significant quantity of the contained heating value in the product.
Related to the markets requirements of quality and conditions, the nascent markets for hydrogen may result in different timing for these products and at different market growth rates. Blending into natural gas at low concentrations doesn’t face significant technical challenges and displacing grey hydrogen in refinery and ammonia production use also can happen at scale relatively quickly. Exports and fuel-cell applications may happen more slowly. The question of flexibility and future-proofing designs against potential market changes should also be considered carefully, as future-proofing can become economically unattractive if timelines extend.
When it comes to scale, logistics may override economics. For projects in Western Canada, one of the challenges is the logistics of moving large reactors/exchangers into the region due to both dimensional and weight restrictions. Some key components are manufactured by a limited number of fabricators (as specified by the technology licensors), and these will need to be imported via a port, such as Duluth, Minn., and then moved by rail into Saskatchewan, Alberta or British Columbia. This is not different than the experience of the upgrading and refining industries moving reactor vessels. While the technology licensors claim very large capacity units, such facilities as a single train may not be logistically possible, and this will also impact the economics and execution model. It may, however, provide an opportunity to stage the construction and start-up of the facilities, particularly if the available product demand is expected to be less than the ultimate capacity.
If you have any questions, don’t hesitate to reach out to a knowledgeable engineering contractor that understands both the technical issues and the local environment and economic climate. Stay tuned for part two of this series.