Infrastructure and distribution
Part 1 of this article, published in the Q3 issue, introduced the study concept and methodology for examining four transportation vectors to convert natural gas from Ras Laffan Industrial City in northeast Qatar into an H2 product at South Hook LNG terminal in Milford Haven, West Wales, UK. The value chain for each option was defined, and CAPEX and OPEX were calculated for each unit within the process. Part 2 considers H2 production with carbon capture and the LNG value chain.
H2 production for all four options is based on steam methane reforming (SMR) technology. SMR was selected based on the availability of in-house information and the desire to have comparative options. H2 production consists of syngas generation, CO2 removal and H2 extraction.
Syngas generation technologies include SMR, autothermal reforming (ATR) and partial oxidation (POX) with a gas-heated reformer (GHR) for additional efficiency. CO2 removal is based on an amine system, which is a mature technology. H2 extraction technologies include pressure swing adsorption (PSA) and membranes. PSAs are proven technology producing higher-purity products.
Blue H2 production focuses on minimizing the quantity of CO2 released to the atmosphere when a hydrocarbon feedstock is used to produce the H2. The authors’ company’s blue H2 SMR technology targets > 90% CO2 capture. A simplified process diagram is shown in Fig. 3.10 The process flow diagram in Fig. 4 shows the unit processes required in the H2 production process and includes the H2 production block in the comparison.
Fig. 3. Blue H2 process block.
Fig. 4. SMR + CCS process block diagram.
Treated natural gas initially mixes with an H2 recycle from the downstream process, including boiloff gas (BOG) from storage, and then passes through a feed preheater. The heated desulfurized gas mixes with steam and passes through the prereformer. In the prereformer, higher-chain hydrocarbons are converted into methane (CH4), CO2, carbon monoxide (CO) and H2. Additional steam is injected into the gas leaving the prereformer underflow control to maintain the correct steam/carbon ratio for the reforming process.
The process gas then enters the gas heated reformer (GHR). In the GHR, methane is partially converted to H2 and CO in an endothermic reaction. The reforming in the GHR means that the advanced Terrace Wall steam reformer (TWR) does not require supplemental fuel gas firing compared to a traditional SMR.
The reforming reaction is strongly endothermic and requires high process temperatures to favor greater equilibrium concentrations of CO and H2. The syngas then passes through the shift reactors, where the CO present in the syngas is “shifted” to CO2 through the water-gas shift reaction.
After the shift reactors, the syngas is further cooled before it is sent to the amine unit, where the CO2 is removed. The syngas product leaving the amine unit is then divided into two streams: one stream is sent to the TWR furnace as a fuel gas for heating the catalyst tubes, and the second stream is sent to the PSA unit. The H2 is separated from the syngas in the PSA absorbent, and the residual tail gas is also sent as a fuel providing supplemental firing of the reformer furnace, if required. From the PSA unit, the H2 product stream is sent to the battery limit (BL). Some of the H2 is separated from this stream to be recycled to the front of the process underflow control to be mixed with the incoming feed gas.
The CO2 leaving the amine unit is the final product for three of the options. This stream needs to be dehydrated and compressed for reinjection into offshore wells. These costs are not included in the study comparison.
For the LNG value chain, the CO2 leaving the amine unit is compressed and liquefied for transport to Qatar; once returned, it is sent for reinjection into offshore wells. These costs are considered in this study. Returning the CO2 to Qatar is the most conservative approach, although sequestration closer to the import facility would be a more economical solution.
The LNG vector consists of the four key process blocks shown in Fig. 5. LNG liquefaction takes place in Qatar before transportation to the UK. Regasification and H2 production take place in the UK, delivering H2 to the natural gas grid. The captured CO2 is then liquefied and transported back to Qatar.
Fig. 5. LNG vector process blocks.
Fig. 6 shows a more detailed block flow diagram of the LNG vector with the key material balance. Tables 2, 3 and 4 are related to the different process blocks in Fig. 6. Each table shows the capacity of the individual unit and whether one or more units are required based on the largest-capacity references found in literature. A 30-yr CAPEX plus annual OPEX for the major feeds and utilities are converted into a specific H2 value and location for these costs.
Fig. 6. LNG block flow diagram with material balance.
LNG liquefaction (Table 2) is a portion of the QG2 mega trains using the APCI AP-X process (Fig. 7), which is a combination of a typical C3MR LNG liquefaction unit with an additional N2 subcooling section to increase the standard capacity up to 7.8 metric MMtpy. Revenue from byproducts, such as NGL and condensate, has not been considered.
Fig. 7. APCI AP-X process.16
LNG storage is based on the installed LNG storage capacity in Qatar as part of the QG2 value chain, prorated for the study capacity. Jetty infrastructure costs are not included. The storage tanks are full-containment type cryogenic tanks. The assumed 0.1 vol%/d BOG is sent to liquefaction for recovery and included in the material balance.
LNG transportation is based on Qmax LNG carriers as part of the QG2 value chain prorated for the study capacity. This results in a utilization of 42% for the one ship considered; however, in practice, this ship is utilized as part of the full fleet, so there is some additional CAPEX attributed to this study, which could be adjusted marginally downwards. It is assumed that the 0.12 vol%/day boiloff rate (BOR)11 is used as the ship fuel and any excess is reliquefied onboard, which is considered in the material balance.
LNG storage is based on the installed LNG storage capacity at South Hook as part of the QG2 value chain, prorated for the study capacity. Jetty infrastructure costs are not included. The storage tanks are full-containment type cryogenic tanks. The assumed 0.1 vol%/d BOG is recovered as part of the feed to the SMR.
LNG regasification (Table 3) is based on an assumed minimum installed capacity of 15.6 metric MMtpy as part of the QG2 value chain and prorated for the study capacity. Submerged combustion vaporizers (SCV) are installed at South Hook and use natural gas as fuel.
H2 production is based on two SMR units with associated CO2 capture. The key utilities are power, water for steam production and cooling water. The SMR + CCS process described earlier is the same for the different vectors. Surplus steam is minimized, and maximum CO2 capture is targeted in the blue H2 SMR + CCS process. The H2 product from this unit is sent for compression for injection into the natural gas grid. The CO2 captured is sent to the next unit for compression and liquefaction.
CO2 liquefaction (Table 4) includes compression, drying and liquefaction using a typical mechanical refrigeration unit, and any BOG is recovered and reliquefied. The main utility is power for the compression and refrigeration compressors.
CO2 storage is based on having sufficient storage to service the number of ships and their size. Refrigerated storage spheres with capacities of 12,000 m3 are assumed, with a margin to allow for delay in ship arrival. The same storage capacity in Qatar and the UK is assumed, and new ship berths, loading/unloading arms and pipework have been included in both locations. This is to acknowledge that the smaller CO2 ships and their added movements will require different and additional infrastructure; therefore, the specific value compared with the LNG storage is significantly higher.
CO2 transportation is based on references for refrigerated (7 barg, –50°C) ship sizes of 50,000 m3, which require three ships and result in a high ship utilization. It is assumed that these ships can operate on LNG as a fuel, if not now at least in the future.
The total specific value for the entire value chain is $1,610/metric t H2, split between a CAPEX of $446/metric t H2 and an OPEX of $1,164/metric t H2.
The LH2 vector consists of four key process blocks, shown in Fig. 8. H2 production and LH2 liquefaction take place in Qatar before transport to the UK. The captured CO2 is a product for further processing before reinjection in Qatar. Regasification of the LH2 takes place in the UK to deliver H2 to the natural gas grid.
Fig. 8. LH2 vector process blocks.
Fig. 8 shows a more detailed block flow diagram of the LH2 vector with the key material balance. Table 7 and Table 8 are related to the different process blocks in Fig. 8. Each table shows the capacity of the individual unit and whether one or more units are required based on the largest-capacity references found in the literature.
H2 production is the same SMR + CCS process as that described for the LNG vector, other than a slightly larger capacity due to the usage of H2 as a fuel for shipping. Infrastructure and revenue from byproducts such as natural gas liquids (NGL) and condensate have not been considered.
H2 liquefaction (Table 7) by cryogenic refrigeration has similarities to both nitrogen (N2) and LNG liquefaction, but presently at very small scale. Refrigeration compressors, brazed plate-fin heat exchangers and turboexpanders are the main equipment used, plus catalysts to convert ortho-H2 to para-H2 to prevent excessive evaporation in the process. Fig. 9 shows a simplified flow diagram for an H2 liquefaction plant.12
Fig. 9. LH2 block flow diagram with material balance.
Fig. 10. Simplified LH2 scheme.
It is an energy-consuming and inefficient process due to the very low boiling point of H2 at –253°C. Existing technologies require 25%–35% of the energy content of the H2 for liquefaction.13 In addition, BOG management adds duty to prevent loss of product.
Limited data is available on H2 liquefaction, and the largest existing liquefiers have a design capacity of up to 50 tpd.14 H2 liquefiers as large as 500 tpd are being developed, but no good data is available in the public domain. This study requires 517 tpd of H2, so a single unit could be a good fit; however, the best data was based on a 200-tpd unit quoted in literature.15 This study is based on three pro-rated units. Energy requirements quoted are in the range of 8 kWh/kg–12 kWh/kg H2.15,16 Lower energy requirements for H2 liquefiers are being researched with a target of 6 kWh/kg H2, but for the purpose of this study, realistic unit sizes and energy requirements are used. It is expected that in the near future, larger and lower-energy units will become available, so this is included as a sensitivity only for LH2 liquefaction.
LH2 storage is based on having sufficient storage to service the number of ships and their sizes. The same storage capacity in Qatar and the UK is assumed based on vacuum-insulated storage spheres of 50,000-m3 capacity, with margin to allow for any delay in ship arrival. At present, the largest LH2 tank capacity is 4,000 m3, located at NASA. Although H2 has high energy density on a mass basis (approximately 120 MJ/kg) compared to other fuels, on a volume basis it requires at least three times more volume than other fuel gases (8 MJ/l vs. 20.8 MJ/l for LNG).17 Due to the low storage temperature of –253°C, BOG generation of 0.18 vol%/d is estimated; therefore, LH2 storage requires advanced materials and technology, resulting in costly tanks.
LH2 transportation has few references, with only one pilot ship with a capacity of 1,250 m3 in operation, developed by Kawasaki.8 However, LH2 carriers with a capacity of 160,000 m3 (4 × 40,000 m3 spheres) are cited as being developed.18 As with storage, the low volumetric energy density and high BOG (0.2 vol%/d)18 result in costly equipment. It is assumed that the BOG is consumed by the ship’s engines for propulsion, as onboard reliquefaction is unrealistic.
LH2 regasification (Table 8) at any substantial flowrate has even fewer references than the rest of the LH2 value chain processes. At low flowrates, ambient air vaporizers can be used, but it is assumed at the capacity required for this study that something similar to the SCVs used for LNG regasification will be required. For the purposes of this study, power usage is assumed for the OPEX; if natural gas is used, then this part of the LH2 value chain would generate CO2, so it is assumed that this would be avoided in practice. Future developments must focus on recovering the substantial energy available from the process.
The total specific value for the entire value chain is $2,647/metric t H2, split between a CAPEX of $944/metric t H2 and an OPEX of $1,702/metric t H2.
For the sensitivity case using optimistic CAPEX and OPEX targets expected in the future after development of the technologies, the total specific value for the entire value chain is $2,257/metric t H2, split between a CAPEX of $804/metric t H2 and an OPEX of $1,452/metric t H2.
The final part of this article will consider the ammonia and methylcyclohexane value chains, safety considerations and the study conclusions.
This article was presented at the GPA Europe Virtual Conference on May 25, 2021.
The information and data contained herein is provided by Wood, solely in respect of the paper itself and should not be considered to have consequence outside of this hypothetical study. Wood makes no representation or warranty, express or implied, and assumes no obligation or liability, whatsoever, to any third party with respect to the veracity, adequacy, completeness, accuracy or use of any information contained herein. The information and data contained herein is not, and should not be construed as, a recommendation by Wood that any recipient of this document invest in or provide finance to any similar project. Each recipient should make its own independent evaluation to determine whether to extend credit to projects with which they are involved.H2T
10 Wood, “Hydrogen supply programme—novel steam methane/gas heated reformer phase 1 final study report,” 2020.
11 Al-Breiki, M. and Y. Bicer, “Comparative cost assessment of sustainable energy carriers produced from natural gas accounting for boil-off gas and social cost of carbon,” Energy Reports, Vol. 6, Nov. 2020.
12 U.S. Department of Energy, Hydrogen and Fuel Cell Technologies Office, “H2A hydrogen delivery infrastructure analysis models and conventional pathway options analysis results—Interim report” March 7, 2014, online: https://www.energy.gov/eere/fuelcells/downloads/h2a-hydrogen-delivery-infrastructure-analysis-models-and-conventional
13 International Energy Agency, “The future of hydrogen: Seizing today’s opportunities,” OECD, Paris Cedex 16, 2019, online: https://doi.org/10.1787/1e0514c4-en
14 Linde, “Large-scale liquid hydrogen production and supply,” online: https://lngfutures.edu.au/wp-content/uploads/2019/10/Cardella-U.-Large-Scale-Liguid-H2-Production-and-Supply.pdf
15 U.S. Department of Energy, Argonne National Laboratory, “Hydrogen delivery scenario analysis model (HDSAM),” online: https://hdsam.es.anl.gov/index.php?content=hdsam
16 U.S. Department of Energy, “DOE hydrogen and fuel cells program: Hydrogen storage,” Vol. 25, 2009,” online: http://www.hydrogen.energy.gov/storage.html
17 The National Academies Press, “The hydrogen economy: Opportunities, costs, barriers, and R&D needs,” online: https://www.nap.edu/read/10922/chapter/6
18 Kamiya, S., M. Nishimura and E. Harada, “Study on introduction of CO2 free energy to Japan with liquid hydrogen,” Physics Procedia, Vol. 67, 2015.
Nicola Chodorowska is a Managing Consultant in the specialist engineering and consulting group at Wood in Reading, UK. She holds a BEng degree in chemical engineering and is a Fellow of the Institution of Chemical Engineers.
Maryam Farhadi is a Process Engineer in process engineering and capital projects at Wood in Reading, UK. She holds an MSc degree in process systems engineering from the University of Surrey.