|
23 | 23 |
|
24 | 24 |
|
25 | 25 | <p><a href="https://aefarrell.github.io/posts/hydrogen_blending/">Previously</a> <a href="https://aefarrell.github.io/posts/fugitive-hydrogen/">I evaluated</a> hydrogen as a fuel gas from the perspective of an end user – someone who purchases utility natural gas, at pressure, for use in combustion devices like boilers and heaters. From that perspective, hydrogen is not an unreasonable conversion, with material compatibility being the primary concern. In this post I’m going to look at it from the perspective of the gas utility.</p> |
26 | | -<p>From my previous analysis, I showed that the same piping operating at the same pressures delivers approximately the same energy, in terms of higher heating value, in systems in full hydrogen service as those in natural gas service. So, for an end user of natural gas (such as me, it’s how I heat my home) making some modifications to the fired equipment and getting a stream of hydrogen versus natural gas is a plausible pathway to low-carbon heating. That doesn’t entirely hold up for utility providing the gas, however, as there is an additional cost associated with compressing hydrogen over natural gas, which might make such systems impractically expensive to operate. At least that is the question I’m looking to answer here: is distributing hydrogen fuel gas to residential or industrial customers through a distribution network like natural gas feasible or not?</p> |
| 26 | +<p>From my previous analysis, I showed that the same piping operating at the same pressures delivers approximately the same energy, in terms of higher heating value, in systems in full hydrogen service as those in natural gas service. So, for an end user of natural gas (such as me, it’s how I heat my home) making some modifications to the fired equipment and getting a stream of hydrogen versus natural gas is a plausible pathway to low-carbon heating. That doesn’t entirely hold up for the utility company providing the gas, however, as there is an additional cost associated with compressing hydrogen over natural gas, which might make such systems impractically expensive to operate. At least that is the question I’m looking to answer here: is distributing hydrogen fuel gas to residential or industrial customers through a distribution network like natural gas feasible or not?</p> |
27 | 27 | <p>The full economic analysis of hydrogen as a fuel gas versus some other low carbon source of energy would so strongly depend on local factors – the local cost of electricity versus hydrogen, whether that region is subject to a carbon tax and how that tax works, etc. – that I don’t think much can be generalized. The economics of hydrogen, where I live, where natural gas is abundant and widely used, export infrastructure is limited, and the carbon tax largely excludes all but the largest industrial emitters, is pretty different from a place where all natural gas is imported at large expense, or with a very different approach to carbon pricing.</p> |
| 28 | +<p>Instead of pursuing a complete technoeconomic analysis of hydrogen, a pretty ambitious topic for a short blog post, I am going to focus on the single factor that is often bandied about online to justify the conclusion that hydrogen distribution is <em>infeasible</em>: the work needed to compress the gas in the pipeline system. I want to show where that number comes from and think about what it might tell us about the system overall.</p> |
28 | 29 | <section id="the-situation" class="level2 page-columns page-full"> |
29 | 30 | <h2 class="anchored" data-anchor-id="the-situation">The Situation</h2> |
30 | 31 | <p>We already know that natural gas distribution systems are feasible, there is one delivering natural gas to my house right now and it is also delivering natural gas to the chemical plant I work at, the gas fired power plant that is powering my laptop right now, etc. To some extent we also already know that hydrogen distribution systems are feasible as they already exist, the longest hydrogen transmission pipeline in Europe is >1000km long and there are >700km of hydrogen pipelines in the United States<span class="citation" data-cites="zendehboudi-2025">.<sup>1</sup></span> However those are primarily for supplying hydrogen as a <em>feedstock</em> to chemical and petrochemical facilities, not quite the same use case as hydrogen as a fuel gas.</p> |
31 | | -<div class="no-row-height column-margin column-container"><div id="fn1"><p><sup>1</sup> Sendehboudi and Gharbani, <em>Hydrogen Production, Transportation, Storage, and Utilization</em>.</p></div></div><p>A reasonable approach to answering this question is to compare a hypothetical hydrogen transmission system to a natural gas system. This is basically what <a href="https://aefarrell.github.io/posts/hydrogen_blending/">I’ve already done for pipe-flow</a> when looking at hydrogen blending: once the hydrogen is in the pipe and at pressure, everything works from that point down. What remains to be seen is whether it is feasible to get it into the pipe and at pressure. Specifically how much more work does it take to compress hydrogen to line pressure than natural gas?</p> |
| 32 | +<div class="no-row-height column-margin column-container"><div id="fn1"><p><sup>1</sup> Sendehboudi and Gharbani, <em>Hydrogen Production, Transportation, Storage, and Utilization</em>.</p></div></div><p>A reasonable approach to considering the feasibility of a hypothetical hydrogen transmission system is to compare it to an equivalent natural gas system. This is basically what <a href="https://aefarrell.github.io/posts/hydrogen_blending/">I’ve already done for pipe-flow</a> when looking at hydrogen blending: once the hydrogen is in the pipe and at pressure, everything works from that point down. What remains to be seen is whether it is feasible to get it into the pipe and at pressure. Specifically how much more work does it take to compress hydrogen to line pressure than natural gas?</p> |
32 | 33 | <p>The standard equation for determining the work, <img src="https://latex.codecogs.com/png.latex?%5Cdot%7Bw%7D_%7Bg%7D">, to compress a mass flowrate <img src="https://latex.codecogs.com/png.latex?%5Cdot%7Bm%7D"> of gas from a pressure of <img src="https://latex.codecogs.com/png.latex?p_1"> to <img src="https://latex.codecogs.com/png.latex?p_2"> is<span class="citation" data-cites="gpsa-2012"><sup>2</sup></span><span class="citation" data-cites="meherwan-2008"><sup>3</sup></span><sup>4</sup></p> |
33 | 34 | <div class="no-row-height column-margin column-container"><div id="fn2"><p><sup>2</sup> GPSA, <em>Engineering Data Book</em>.</p></div><div id="fn3"><p><sup>3</sup> Boyce et al., <span>“Transport and Storage of Fluids.”</span> 10–42.</p></div><div id="fn4"><p><sup>4</sup> Strictly speaking this is an approximation as it neglects the change in kinetic energy of the fluid, but for small compression ratios, less than ~5, it is appropriate</p></div></div><p><img src="https://latex.codecogs.com/png.latex?%0A%5Cdot%7Bw%7D_%7Bg%7D%20=%20%5Cdot%7Bm%7D%20%5Cint_%7Bp_1%7D%5E%7Bp_2%7D%20v%20dp%0A"></p> |
34 | 35 | <p>This is related to the isentropic work through the isentropic efficiency, <img src="https://latex.codecogs.com/png.latex?%5Cvarepsilon_%7Bi%7D"></p> |
|
44 | 45 | <p><img src="https://latex.codecogs.com/png.latex?%0Ar%20=%20%7B%20%5Cdot%7Bm%7D_%7BH2%7D%20%5Cover%20%5Cdot%7Bm%7D_%7BNG%7D%20%7D%20%7B%20%7B%5CDelta%20h%7D_%7BH2%7D%20%5Cover%20%7B%5CDelta%20h%7D_%7BNG%7D%20%7D%0A"></p> |
45 | 46 | <p>It is important to note that state 2 is not the same for hydrogen and natural gas. Since the integration is along an isentropic path, state 2 is at a pressure of <img src="https://latex.codecogs.com/png.latex?p_2"> and a temperature <img src="https://latex.codecogs.com/png.latex?T_2"> defined by <img src="https://latex.codecogs.com/png.latex?s_1%20=%20s_2"> and the entropy of hydrogen and natural gas are, in principle, different.</p> |
46 | 47 | <p>Compressors typically don’t raise pressures all the way from, say, atmospheric pressure to the 200-1500psi working pressures of natural gas transmission lines in a single stage. For one, as gases are compressed they heat up and that large temperature rise can damage a compressor. Usually compression is accomplished with a series of stages with interstage cooling. This work ratio is really only valid for a single stage.</p> |
47 | | -<p>Suppose we are evaluating a system that uses a multi-stage compressor to take gas at ambient conditions, in this case suppose 1bar and 15C, to a relatively high transmission line pressure of 100bar using 4 stages, Figure 1. The overall compression ratio is 100, with 3 stages this gives a per stage ratio of</p> |
| 48 | +<p>Suppose we are evaluating a system that uses a multi-stage compressor to take gas at ambient conditions, in this case suppose 1bar and 15C, to a relatively high transmission line pressure of 100bar using 3 stages, Figure 1. The overall compression ratio is 100, with 3 stages this gives a per stage ratio of</p> |
48 | 49 | <div id="53ef2d74-ac32-490f-a9d5-20f149965bb0" class="cell" data-quarto-private-1="{"key":"jupyter","value":{"source_hidden":true}}" data-execution_count="2"> |
49 | 50 | <div class="cell-output cell-output-display cell-output-markdown" data-execution_count="2"> |
50 | 51 | <p><img src="https://latex.codecogs.com/png.latex?%5Cbegin%7Balign%7D%0A%5Ceta_t%20&=%20100%5C;%5Ctext%7B%20%20%7D(%5Ctext%7Btotal%20compression%20ratio%7D)%0A%5C%5C%5B10pt%5D%0An%20&=%203%5C;%5Ctext%7B%20%20%7D(%5Ctext%7Bnumber%20of%20stages%7D)%0A%5C%5C%5B10pt%5D%0A%5Ceta%20&=%20%5Ceta_t%5E%7B%5Cfrac%7B1%7D%7Bn%7D%7D%0A%5C%5C%5B10pt%5D%0A&=%20100%5E%7B%5Cfrac%7B1%7D%7B3%7D%7D%0A%5C%5C%5B10pt%5D%0A&=%204.64%0A%5Cend%7Balign%7D"></p> |
@@ -340,8 +341,10 @@ Figure 2: Relative work required to compress hydrogen versus methane for th |
340 | 341 | <section id="final-thoughts" class="level2 page-columns page-full"> |
341 | 342 | <h2 class="anchored" data-anchor-id="final-thoughts">Final Thoughts</h2> |
342 | 343 | <p>I wrote this post to address some misconceptions that I’ve encountered regarding hydrogen<sup>7</sup> and in particular the rhetorical device of finding one single fact about hydrogen and taking that to mean some project or another has been “debunked”. Real engineering projects are just too complex for that to be a useful exercise. Reality always depends on a great many factors.</p> |
343 | | -<div class="no-row-height column-margin column-container"><div id="fn7"><p><sup>7</sup> is this all just an extended response to a thread on mastodon? I mean… sort of</p></div></div><p>Is the fact that a hydrogen fuel distribution system would require >3× the energy to operate mean that such a system is impractical? That really depends. It could be that a large, continent spanning, transmission system for hydrogen such as natural gas distribution employs in North America is rendered totally infeasible by the increased power demands. But then again, why should hydrogen be so geographically constrained? Natural gas is constrained by geology but presumably one could make green hydrogen wherever there is water and renewable power. Perhaps blue hydrogen is best built on top of the existing natural gas infrastructure – send natural gas across the continent and convert it to hydrogen closer to the end use. I am doubtful that one could come up with a sweeping conclusion from all of this that would say anything beyond one’s ignorance of the specific conditions of niche industries and use cases for hydrogen versus the panoply of alternative low carbon energy sources.</p> |
344 | | -<p>I think the dreams of existing gas fired power plants simply retrofitting to hydrogen and continuing on as before are looking increasingly like a relic from a bygone era. The price of renewables and storage continues to plumet and the economics of these schemes seem increasingly out of touch with that reality. But for other industries, with other heating demands, perhaps there is a compelling case to be made.</p> |
| 344 | +<div class="no-row-height column-margin column-container"><div id="fn7"><p><sup>7</sup> is this all just an extended response to a thread on mastodon? I mean… sort of</p></div></div><p>When considering my gas bill, at home, the dominant factor for the cost of natural gas is the cost to procure it in the first place (i.e. source it from wells, process it in gas plants). Transmission costs are relatively low (granted I live in a province where it just comes out of the ground, so the transmission <em>distances</em> are pretty low too). I imagine this would be the same for a hydrogen fuel gas system as well: the dominant cost will be the making of hydrogen.</p> |
| 345 | +<p>Is the fact that a hydrogen fuel distribution system would require >3× the energy to operate mean that such a system is impractical? That really depends. It could be that a large, continent spanning, transmission system for hydrogen such as natural gas distribution employs in North America is rendered totally infeasible by the increased transmission costs. But then again, why should hydrogen be so geographically constrained? Natural gas is constrained by geology but presumably one could make green hydrogen wherever there is water and renewable power. Perhaps blue hydrogen is best built on top of the existing natural gas infrastructure: send natural gas across the continent and convert it to hydrogen closer to the end use.</p> |
| 346 | +<p>I think it is clear, considering the problem more broadly, that hydrogen is unlikely to be a simple drop in replacement with the current infrastructure: infrastructure optimized for natural gas is not infrastructure optimized for hydrogen. But there are real opportunities for repurposing some parts of the natural gas infrastructure for hydrogen that shouldn’t be ignored.</p> |
| 347 | +<p>I think the dreams of existing gas fired power plants simply retrofitting to hydrogen and continuing on as before are looking increasingly like a relic from a bygone era. But this has less to do with the practicalities of delivering hydrogen and much more to do with the price of renewables and storage, which continue to plumet. But for other industries, with other heating demands, perhaps there is a compelling case to be made.</p> |
345 | 348 | <p>I say all of this as someone who is broadly skeptical of the hype around hydrogen. I think it is being pursued mostly as a saviour of fossil fuels and not as a technology that actually best solves the problems which face us as we transition to a low carbon future. But there are also a lot of really smart engineers working on projects centered around low-carbon hydrogen, and I imagine they know what they are doing.</p> |
346 | 349 | </section> |
347 | 350 | <section id="references" class="level2"> |
|
0 commit comments