Thanks. We need to correct a few misconceptions from this article. First, the IEA estimates are high, which is not a typical for the agency when it comes to RE technologies. Nel, a leading producer of electrolyzers in Europe, is willing to write large (GW scale) contracts at $300/kW, well below the $840/kW quoted from the IEA. They have written such a contract with Nikola, though I am skeptical about Nikola being able to live up to its side of the contract. Even so, the issue for Nel is scaling up to achieve lower production costs. Moreover, in Asia, electrolyzer costs have already fallen into the $300/kW range. Further global scale up of the industry should support declining prices. Getting to the multi GW scale is critical.
The second issue regards operating electrolyzers at high load factors. The idea goes like this, the high capex of electrolyzers mean they have to be operated near 100% of the time to achieve a low levelized cost of hydrogen. This is a common framing, but it is misleading and commits a sunk cost fallacy. Once capacity has been built, the levelized cost of production is irrelevant. All prior investment is a sunk cost. What matters for obtaining optimal return is that the capacity is operated whenever it marginally profitable to do so. So this depends on the marginal value of hydrogen and the real-time marginal cost of power. Whenever there is power below a certain parity price for the price of hydrogen, the electrolyzer will run. Running strictly off grid power means that load factors are limited to 30% to 50%. However, there is a way to boost the load factor by coupling electrolyzer capacity to RE generation that in excess of what can be delivered to the grit.
For example, suppose you build a 100MWac solar farm that has say 200MWdc of actual panels. This means the inverter and interconnection capacity is limited to 100MWac and the farm can never put more than 100MW onto the grid. The PV capacity, however, is twice this amount. Whenever the PV modules are generating above about 105Mdc, the excess power cannot be put on the grid. This is called clipping, and it a way to increase the reliability and capacity factor of solar power. So presumably there is a PPA offtake contract for this 100MWac. So what can an electrolyzer or battery do with this surplus power above 105MWdc? Suppose we co-locate 50MWdc of electrolyzer capacity. So when the sun is cranking out more than 105MWdc, the electrolyzer has access to power at zero marginal cost. We know this has zero marginal cost because it is excess to what can physically be put on the grid and there are no other buyers for the power. (Now if we add a battery to the mix, then the battery also competes with electrolyzer for this surplus power, but the battery is limited in how much energy it can store and what rate it can charge. Thus, even with a battery, they system still obtains conditions under which surplus power has zero or very low marginal cost. So to keep the analysis simple, we will ignore the marginal demand from a battery, though it is an important part of optimizing the economics of a hybrid system.) Thus, the electrolyzer operates when there is surplus power (zero marginal cost). This may be only a few hours per day, but this is only a lower bound on the load factor. Because again, the electrolyzer can also run profitably whenever the grid power price is below parity for hydrolysis. To make the most of this we will assume that the interconnection and inverter are bi-directional. Thus, whenever the sun is not shining the electrolyzer can pull cheap power off the grid. Even when the solar is delivering 100MWac per PPA a fraction of this can be bought back for the electrolyzer.
This may be surprizing to many, but it makes total economic sense. The PPA is generally sold to a utility which is trying to make money selling power. It has an interest in both cheap supply and price responsive demand. So consider a PPA structured so that the buyer pays $40/MWh for upto 100MWac of solar, however the sell retains the option to buy back up to 50MWac at wholesale price whenever the price is below $40/MWh. This is actually beneficial for the PPA buyer to offer the buyback option. Otherwise, the buyer is obligated to pay the seller $40/MWh on a full 100MW even when the wholesale market is a cheaper source of power for the buyer. That is, under a standard PPA, the buyer is actually losing money whenever the wholesale price is below $40/MWh. So the buyback option for the electrolyzer limits how much the PPA buyer can lose. The arrangement also helps the grid to avoid extremely low prices and the need for curtailments. This is beneficial for all competing generators, especially other solar generators that need to fetch reasonable prices while the sun shines.
So in conclusion for this little example of solar PV+electrolyzer, the electrolyzer operates whenever wholesale price is below parity or when there is surplus PV power that can't be sold on the grid. Thus, this placement of an electrolyzer has a higher profitable load factor than an electrolyzer that draws power only from the grid. I would also point out that the electrolyzer shares inverter and interconnection capacity with the solar PV, which improves the utilization of those assets beyond what solar alone or electrolyzer alone would obtain. Thus, the combination achieves higher capital efficiency and larger load factors.
I would also point out that the opportunity to add battery storage to solar is quite similar. The difference there is that the battery discharges back into the grid with the wholesale price is high. The higher expected daily prices set the parity price up to which the battery can profitably charge. Suppose you have 200MWdc PV with inverter and interconnection at 100MWac and a 100MWac / 400MWh battery. While solar PV output is above 105MWdc, you have a surplus with zero marginal cost with which to charge the battery. So usually there is enough surplus solar to top off the 400MWh storage capacity. Suppose this gets discharge at peak prices in th evening. There is also an opportunity to recharge the battery overnight from the grid for early morning generation. Suppose the you have two hours at a price of $55/MWh in the early morning. You can sell 200MWh. So overnight you will charge the batteries for the cheapest two hours, so long as the price is below parity, abut $50/MWh. Buying at say, $20/MWh, provides $30/MWh of marginal profit, or $6,000 for the 200MWh discharge. Let's suppose the evening discharge was 400MWh at $60/MWh. This was charged from surplus solar at zero marginal cost, so the marginal profit here is $60/MWh on 400MWh, or $24,000 for the evening. Putting these together the 400MWh battery is discharging 600MWh per 24-hour cycle and generating $30,000 in marginal profit. The point here is that the battery that is co-located with the PV supply is able to achieve higher capacity factor (discharge utilization) and has access to power at lower marginal cost than the grid. And it is able to share inverter and interconnection capacity with the solar. So both solar and battery achieve higher capital efficiency when combined.
This is very similar to how combining with an electrolyzer improves financial performance. But a key difference is that the battery is limited by how much it can store (400MWh) and by how much it can profitably deliver in a narrow window of time (200MWh in the early morning). An electrolyzer is not really limited in this way (assuming tank capacity and offtake agreements are more than adequate for high load factors). So the battery will be hunting for the cheapest power to buy for charging, while the electrolyzer will simply run whenever power is available below parity prices. They have different and complementary buying profiles. Thus, it makes sense to pair solar (and/or wind) with some combination of battery and electrolyzer capacity. The combination will make optimal use of surplus RE generation (above the interconnection capacity) and lead to higher capital efficiency. The superhybrid operations are even harder to describe than the simpler setups. There is even a case where it can be marginally profitable to run the electrolyzer off of the battery. For simple example, if peak power prices are not of sufficient duration to use the full charge of the battery the excess gets dumped to the electrolyzer before RE goes back into surplus generation again. I would also point out that this can be a little more energy efficient than using the inverter to pull AC from the grid to supply DC to the electrolyzer. Even though these operating modes can be infrequent it could add a few points to both the electrolyzer load factor and the battery capacity factor and load factor. So it is all about driving up the capital efficiency. The flip side for the grid is that more RE can be integrated into the grid this way to be available to supply the grid when wholesale prices climb higher. And this is absolutely essential to realizing as soon as possible a 100% renewable grid.
I think it important that we not look at batteries or electrolyzers in isolation to RE. Stand-alone systems are not capital efficient. This kind of siloed thinking can lead us to think that these technologies are too expensive and will take too long to reach scale. I do think that the superhybrid (solar+wind+battery+electrolyzer) is the Holy Grail of deep decarbonization. The challenge is not really cost at this point. The challenges is how to integrate these technologies to optimize financial performance. Companies are going to need to get a lot of experience doing this to figure out how the combination can create the most value for the grid (and investors). The electrolyzer bit is still at demonstration scale, but the how superhybrid needs to find its optimal scales efficiency. As innovative companies advance on this learning curve, I think well find an inflection point where superhybrids run ahead of conventional demand for generation capacity. That is, at some point, the economic value extracted from the hydrogen market cracks open an energy market much bigger than the power grid has served. Through superhybrids, the grid becomes a net producer of gases rather than a net consumer. This will take demand for wind and solar power to much higher levels than mere service to the grid. So there is a tipping point where these new economics start to take over. But anyone whose imagination is limited to batteries or electrolyzers as stand-alone assets tied to the grid only will not see the synergies that make this inflection point much more immediate.
