Converting the gas grid to hydrogen, starting with Leeds.

Low carbon electricity is within our sights but low carbon heating is more difficult. Most UK homes use natural gas (methane) and switching to renewables (electricity or biomass) is expensive, even when it is practical. Northern Grid Networks (NGN) have another suggestion - converting the gas grid to supply hydrogen.  This is carbon free at the point of use but not usually where it is made - the NGN proposal is to convert from methane and sequester the carbon stripped off. This will give, at their estimate, 59% reduction in emissions. However once have we have a hydrogen distribution network other options, including carbon negative ones are possible. The big questions is, is it practical to convert the grid to hydrogen? NGN propose that we use Leeds as a pilot city and they have produced a detailed plan: H21 Leeds City Gate. They have convinced me that it is technically possible, and the biggest barrier is cost, but even this is not ridiculous compared to other means for carbon savings. The diagram below shows the concept.

Overall concept for H21 - including gas supply to the steam reformers, salt caverns for storage and hydrogen pipelines. In the city hydrogen would flow through the existing gas distribution network. Image 1.6 from [1].

Overall design

We will use the existing gas distribution network to deliver hydrogen instead of methane.
Leeds will be isolated from the rest of the gas grid and supplied with hydrogen instead of methane. Existing gas pipelines will be used - but only the distribution system up to medium pressure (7 bar). The high pressure gas transmission system (gas 'motorways') will not be affected. The part of the network they have selected supplies 264,000 customers, of which 99% are domestic but there are some big industrial customers too: 27% of heat demand comes from just 127 industrial sites.

Hydrogen from steam reforming will be stored in salt caverns - up to 40 days winter supply.
The hydrogen will be produced from methane by steam reforming in four giant process plants. Gas use is highly variable through the day and through the seasons so storage will be used to even out the load. There are already salt caverns in the area storing hydrogen for industrial purposes; more will be excavated in the nearby Permian salt fields on the coast. They will construct enough storage for 40 days winter supply.

The White Rose CCS scheme must be revived, or something similar, to sequester the carbon.
The steam reformers generate CO2 which will be captured and sequestered (CCS) nearby. Leeds is conveniently close to the planned White Rose CCS facility. Unfortunately funding for White Rose was withdrawn last year. However, if this project were to be revived then Leeds would make a reliable CO2 supply for it. The CO2 generated from steam reforming is very pure and with the storage buffering demand production will run more or less steadily through the year.

This approach is good but future developments would make it much better.
This approach loses energy in methane reforming and requires extra energy for the CCS, so it is fairly expensive. However the methane reforming technique is a tried and trusted technology, a good choice for this pilot. Once the distribution network is established, other possibilities under development include:

• Making hydrogen by electrolysis from water - with research into new catalysts bringing improvements all the time, such as work at Stamford (Single-catalyst water splitter from Stanford produces clean-burning hydrogen 24/7) or at Munich and Leiden (Efficiency of water electrolysis doubled).  This can make use of spare electricity, for example when we have too much PV in summer.
• Making hydrogen from methane with zero carbon emissions - stripping out the carbon as a solid instead 'Let's get cracking on low cost hydrogen'. This would mean no need for White Rose for sequestration.
• Making hydrogen from biomass rather than methane and sequestering the carbon giving negative carbon emissions - so called BioEnergy-CCS. See Timeline: How BECCS became climate change’s ‘saviour’ technology

Technical issues with hydrogen

Hydrogen has a lower energy density so it will need to flow three times faster.
Hydrogen is much lighter than methane and has a lower energy density. You need about 3 times as much by volume for the same energy. Fortunately these factors offset each other and so you only need a little more pressure to deliver the same amount of energy. However, the hydrogen will have to flow faster - and if it flows too fast it can erode surfaces and damage equipment. The team have modelled supply and demand on the Leeds grid to determine where it needs strengthening, either to manage pressure or to keep the flow rate down. Upgrades will be needed in a few places.

Hydrogen makes metal brittle but most of the network is polyethylene.
Hydrogen is also renowned for damaging metal pipework by making it more brittle. Fortunately most of the network is now polyethylene and keeping the presssure low also minimises this problem. Leakage is expected to be very low, at most 0.001% of the volume supplied - and hydrogen, unlike methane, is not a powerful greenhouse gas.

Safety issues: hydrogen is less likely to go bang than methane
Hydrogen burns over a wide range of concentrations from 4% to 75%. However, it won't go bang until it reaches at least 10% - methane will explode as low as 4% - and hydrogen is so light it diffuses away quickly. Like methane it is odourless so something will be added to make it smelly. Also, the flame is very pale and hard to see but tiny additives could be mixed in to give it colour.

No carbon monoxide but more steam.
Methane burns to a mix of CO2 and steam whereas hydrogen produces only steam. This is generally an advantage and certainly there is no risk of carbon monoxide poisoning. However anyone with a flueless gas fire must take care to have plenty of ventilation to avoid condensation problems. Also if you have a gas oven you might find the additional moisture affects the cooking a little.

Hydrogen produces more NOx unless catalytic burners are used.
Generally speaking, the hotter the burn the more NOx that is produced and hydrogen burns somewhat hotter than methane: at over 2000°C. Catalytic combustion can reduce the flame temperature so that the NOx levels are virtually zero, but this does add to the cost of the appliances.

Practicalities of the conversion

All appliances: boilers, cookers, gas fires and industrial plant will need to be converted or replaced.
We have done this before, only in reverse. Between 1966 and 1977 we switched over from town gas which is 50% hydrogen to natural gas; every gas appliance in the country had to be converted to allow for the different flow rates and flame properties of methane. At the peak of this process, we converted 2.3 million homes a year, on average 2 to 3 appliances in each home. There were hundreds of different kinds of appliances and conversion kits were developed for almost all, so only old equipment had to be completely replaced.  We have more recent experience too, as the Isle of Man only converted in 2010. NGN estimates the cost for Leeds will be around £3,000/home.

Manufacturers should develop hydrogen-ready appliances that can be converted quickly.
For this to work, conversion kits will have to be developed for almost all modern appliances. If this whole project stops at Leeds then developing these special purpose kits will have been an expensive dead end. However if we finally roll out to the whole country this would be a relatively small part of the cost. NGN recommends that the government defines standards for 'HySwitch' hydrogen-ready appliances that are designed for easy conversion. If manufacturers make these available and householders are persuaded to buy them conversion will be relatively manageable. For example NGN estimate that converting a 'HySwitch' boiler might take 2 hours compared to 6 hours for a standard combi or 11 hours for a standard system boiler. Similarly, converting a 'HySwitch' cooker might take 1 hour compared to 13 hours for a standard gas cooker. Their cost estimates are based on 50% of boilers, 40% of cooking appliances and 30% of gas fires being HySwitch by the time it comes to convert, in mid 2020s. If HySwitch does not work out then the conversion will be much more expensive.

Then there are the industrial customers. There are only about 3,100 in Leeds but their equipment is much more specialised and NGN estimates converting them will cost £80,000 each, on average.

The grid wil be converted in segments. Households could be cut off for up to 5 days.
The grid will be converted in segments of around 2500 homes. Each segment will be converted over a period of 5 days, by an army of more than 900 engineers. The maximum time a household will be cut off would be 5 days.

Carbon savings and costs

Carbon emissions will be reduced by 59%, or more with cleaner electricity.
Carbon capture and sequestration is not perfect - so this hydrogen will be lowish carbon, not zero carbon. NGN plan for capturing 90% of the CO2, and also the reforming process is only about 70% efficient, so more methane will be needed to produce the same amount of energy as hydrogen. Overall carbon emissions should be down to 86 g/kWh, compared to 210 g/kWh for methane (including extraction and production). This is a reduction of 59%.

Given the heat demand for Leeds, this project will save 731 thousand tonnes CO2/year.

Some of this carbon comes from electricity - the more renewable electricity we have the less emissions. This chart shows where the emissions are from. The grid and the storage compressors could have zero emissions. That would mean overall emissions down 70%.

Carbon emissions from H21 and methane, compared. SMR means steam methane reformer. Chart 7.1 from [1]

Making hydrogen from methane is easy and we already do it. However, we don't have to start from methane. If we started with biomass then this this sort of scheme can be net negative carbon, helping to strip CO2 out of the atmosphere. This sort of scheme is absolutely essential to keep climate change to 2°C. NGN do not discuss the possibility but it would be an obvious development.

Running costs alone come to 2.3p/kWh, adding in the one-off cost would be 3.5p/kWh 
The conversion to hydrogen means we will use more methane for the same energy. Also there are running costs associated with the steam reformers, storage and CCS. The annual running costs are estimated at £139 million /year, or about 2.3p/kWh. The cost of carbon savings come to £190/tonne (my calculations). This sounds a lot but it is better value than installing more efficient boilers or solid wall insulation (see Cost of Carbon Savings).

The one-off costs come to just over £2,000 million, of which half is due to converting appliances and the remainder is the new infrastructure for the steam reformers, storage, pipelines and necessary network upgrades. If you amortice this over 30 years, then this adds another 1.2p/kWh or £190 tonne. However, this is without factoring in the cost of financing.

CCS infrastructure costs are not included.
These estimates include running costs for carbon capture and sequestration but not the White Rose infrastructure costs. The assumption is that we will need White Rose or an equivalent sequestration service anyway, so the development cost for that will be funded by other means.

Energy bills for NGN customers increase 2.9% due to the pilot, but rolled out across the country the increase would be more.
Clearly, energy bills will increase. NGN estimate that they would need to increase prices for their customers by 2.9%. However this is on the basis that all the NGN customers share the cost of the pilot.  If Leeds customers had to pay for the entire scheme alone their bills could go up to 10p/kWh and if the same scheme were rolled out across the country that is what we would all be paying, at least in the worst case. In practice there will be lessons learned and economies of scale from a larger scale rollout so 10p/kWh.

Also, by the time this pilot is seriously under way some of those developments I mentioned above might be available to make it cheaper and/or lower carbon.

Even though the technology is mature, this project is complex and has risks
There are a lot of pieces to this project. NGN have described their plan broken down into 60 projects in 16 work packages. None of the technology is grounbreaking but the sheer complexity adds risk and the riskiest part of the scheme - the carbon sequestration infrastructure - is not even included. Also there are risky assumptions in the cost of converting appliances.

It requires co-operation from government, appliance manufacturers and householders
This scheme requires co-operation from a variety of actors:

• From government agencies, to create standards for hydrogen-ready appliances.
• From appliance manufacturers, to develop and supply conversion kits and new appliances
• From householders, who will be expected to buy hydrogen-ready appliances and may be cut off from gas for up to 5 days.
• From whoever gets White Rose going, or a similar CCS project.

A hydrogen grid is an enabling technology that could be transformational. 
If I lived in Leeds I am not sure I would relish being a guinea pig on this. On the other hand it would be great to be a participant in the revolution of low carbon heating. We need a magic bullet and a hydrogen network could be a huge big step on the way. The cost is manageable, if government and appliance manufacturers come on board to smooth the conversion proccess and once we have CCS, which we need anyway. Then once Leeds has demonstrated the principle, rolling out to other cities would be relatively straightforward. It is an enabling technology for further developments including BioEnergy-CCS giving net negative emissions. I have been sceptical of the hydrogen economy in the past but this plan is convincing - I think the risk could be worth taking.



[1] H21 Leeds City Gate (July 2016) Northern Gas Networks