9 ms·

I'm excited about anything in electrolysis/fuel cell development, but the figure of merit is efficiency. It's neat to use sulphuric acid to reduce the freezing

I'm excited about anything in electrolysis/fuel cell development, but the figure of merit is efficiency. It's neat to use sulphuric acid to reduce the freezing point of your electrolyte solution, but if it's not efficient you might as well just heat your electrolyzer. Nothing is a power system without efficiency figures. Otherwise it's just a chemistry experiment.

nine_k 5y ago
Since it uses two catalysts and controls for heat production (and loss), it must be pretty efficient, more so than straightforward electrolysis. It can be slower to produce the gasses, though. High efficiency, low throughput.
wrycoder 5y ago
Exactly. Hydrogen electrolysis is most efficient at elevated temperatures, and that was a problem in some exceptionally cold areas. This is a development that makes electrolysis practical in those regions.
- thereisnospork 5y ago
  >This is a development that makes electrolysis practical in those regions. It is neither a development[0] nor does it particularly make electrolysis practical in those regions[1] (or more so than it was yesterday). Unless there is something about keeping solar cells warm that is beneficial, I'm really not seeing it.[2] >Hydrogen electrolysis is most efficient at elevated temperatures negligibly until you get into the realm of solid oxide cells (100's of C) -20 to +40 is negligible, and moreover a moot point because of [1] [0]>> The method works by using electrolytes with low freezing points, such as dilute sulfuric acid, to allow the use of water at lower temperatures. A common electrolyte for making H2, at room temperature and otherwise. Not novel (or interesting). [1]>> resulting in an interior working temperature of around 10°C. Approximately 30-40 percent inefficiency in an electrolyzer manifests as heat -- all they did here is wrap their electrolyzer in a blanket. Any electrolyzer of appreciable (read: useful) power has trouble keeping cold, not warm. The square-cube law applies here, electrolyzer power (and heat generation) scales w.r.t. volume, heat loss with area. For intuitive purposes, a device capable of fueling a car (driving 24/7) would be outputting about as much heat as that car's engine. [2]There might be, (chemistry is my meat and potatoes, more so than physics), but I've always seen the issue being more with keeping solar cells cool.
roughly 5y ago
Depends. If the purpose of this is to take sunlight when it’s abundant and turn it into hydrogen as a storage mechanism, then the efficiency of the exchange may not matter as much.
- thereisnospork 5y ago
  Efficiency always matters; opportunity costs. If sunlight to hydrogen is your goal a novel solution has to at bare minimum be more efficient than solar PV -> electricity -> electrolyzer[0] to be useful. [0]Known, boring technology with 99.9% of the engineering warts and technical debt already solved.
  - mlyle 5y ago
    It has to be more -cost- efficient, not more thermodynamically efficient.
    - thereisnospork 5y ago
      One tends to follow the other. Arguing that, in the judgement of process for the conversion of one form of energy (sunlight) to another (hydrogen), thermodynamic efficiency, or the measure of how well a process converts one form of energy to another isn't relevant is somewhere between obnoxiously pedantic and wishful thinking. Funnily enough I read the article after your comment, and it is even less useful, interesting, than my initial assessment (not even direct solar to H2). It has approximately the novelty and utility of porting DOOM to <insert processor here> and running it in a refrigerator.
      - angus-prune 5y ago
        In the real world energy isn’t always fungible. The location of the energy can be an important factor in something’s overall efficiency. If you need the end product at a particular location, the transport costs can overwhelm any differences in efficiency. The same also applies if you have excess energy in a particular location that you want to use. In either case, the transport costs can make using the less efficient process, the more energy efficient choice overall.
matzab 5y ago
This is explicitly aimed at producing energy in cold environments (high altitude, polar regions), so it's less of an issue (and not exactly the point). Anyway, they seem to aim at 12% STH efficiency and provide some examples (assuming 65% hydrogen fuel cell efficiency): Then, powering a Raspberry Pi computing device (PRPi = 4W) for an autonomous measurement device year-round at the > Neumayer III station (70.68°S, 8.27°W, Tavg,y = −15.5 °C) would prospectively require a module area of 0.41 m2, whereas at Paranal observatory (Chile, 24.63°S, 70.40°W, Tavg,y = 0.4 °C) 0.18 m2 would suffice.
darksaints 5y ago
The figure of merit is cost, and it's only correlated with efficiency. If efficiency was king, we'd all be driving 60% efficient cars...the technology exists but the cost is not acceptable.
usrusr 5y ago
Agreed, compared to a novel chemical approach isolation and counterflow heat exchange seem almost embarrassingly lowtech to the mostly hypothetical "how to to electrolysis when it's cold outside" problem. Perhaps there might be some actual merit when the end product of your process is compressed/liquefied H2? Still reads more like the usual make up some hypothetical use case for the highly impractical niche you chose to push the boundaries of scientific knowledge. But who am I to judge, I did semantic web in multi-agent systems back when ai was symbolic.

nine_k 5y ago

Since it uses two catalysts and controls for heat production (and loss), it must be pretty efficient, more so than straightforward electrolysis. It can be slower to produce the gasses, though. High efficiency, low throughput.

wrycoder 5y ago

Exactly. Hydrogen electrolysis is most efficient at elevated temperatures, and that was a problem in some exceptionally cold areas. This is a development that makes electrolysis practical in those regions.

thereisnospork 5y ago
>This is a development that makes electrolysis practical in those regions. It is neither a development[0] nor does it particularly make electrolysis practical in those regions[1] (or more so than it was yesterday). Unless there is something about keeping solar cells warm that is beneficial, I'm really not seeing it.[2] >Hydrogen electrolysis is most efficient at elevated temperatures negligibly until you get into the realm of solid oxide cells (100's of C) -20 to +40 is negligible, and moreover a moot point because of [1] [0]>> The method works by using electrolytes with low freezing points, such as dilute sulfuric acid, to allow the use of water at lower temperatures. A common electrolyte for making H2, at room temperature and otherwise. Not novel (or interesting). [1]>> resulting in an interior working temperature of around 10°C. Approximately 30-40 percent inefficiency in an electrolyzer manifests as heat -- all they did here is wrap their electrolyzer in a blanket. Any electrolyzer of appreciable (read: useful) power has trouble keeping cold, not warm. The square-cube law applies here, electrolyzer power (and heat generation) scales w.r.t. volume, heat loss with area. For intuitive purposes, a device capable of fueling a car (driving 24/7) would be outputting about as much heat as that car's engine. [2]There might be, (chemistry is my meat and potatoes, more so than physics), but I've always seen the issue being more with keeping solar cells cool.

roughly 5y ago

Depends. If the purpose of this is to take sunlight when it’s abundant and turn it into hydrogen as a storage mechanism, then the efficiency of the exchange may not matter as much.

thereisnospork 5y ago
Efficiency always matters; opportunity costs. If sunlight to hydrogen is your goal a novel solution has to at bare minimum be more efficient than solar PV -> electricity -> electrolyzer[0] to be useful. [0]Known, boring technology with 99.9% of the engineering warts and technical debt already solved.
- mlyle 5y ago
  It has to be more -cost- efficient, not more thermodynamically efficient.
  - thereisnospork 5y ago
    One tends to follow the other. Arguing that, in the judgement of process for the conversion of one form of energy (sunlight) to another (hydrogen), thermodynamic efficiency, or the measure of how well a process converts one form of energy to another isn't relevant is somewhere between obnoxiously pedantic and wishful thinking. Funnily enough I read the article after your comment, and it is even less useful, interesting, than my initial assessment (not even direct solar to H2). It has approximately the novelty and utility of porting DOOM to <insert processor here> and running it in a refrigerator.
    - angus-prune 5y ago
      In the real world energy isn’t always fungible. The location of the energy can be an important factor in something’s overall efficiency. If you need the end product at a particular location, the transport costs can overwhelm any differences in efficiency. The same also applies if you have excess energy in a particular location that you want to use. In either case, the transport costs can make using the less efficient process, the more energy efficient choice overall.

matzab 5y ago

This is explicitly aimed at producing energy in cold environments (high altitude, polar regions), so it's less of an issue (and not exactly the point). Anyway, they seem to aim at 12% STH efficiency and provide some examples (assuming 65% hydrogen fuel cell efficiency): Then, powering a Raspberry Pi computing device (PRPi = 4W) for an autonomous measurement device year-round at the > Neumayer III station (70.68°S, 8.27°W, Tavg,y = −15.5 °C) would prospectively require a module area of 0.41 m2, whereas at Paranal observatory (Chile, 24.63°S, 70.40°W, Tavg,y = 0.4 °C) 0.18 m2 would suffice.

darksaints 5y ago

The figure of merit is cost, and it's only correlated with efficiency. If efficiency was king, we'd all be driving 60% efficient cars...the technology exists but the cost is not acceptable.

usrusr 5y ago

Agreed, compared to a novel chemical approach isolation and counterflow heat exchange seem almost embarrassingly lowtech to the mostly hypothetical "how to to electrolysis when it's cold outside" problem. Perhaps there might be some actual merit when the end product of your process is compressed/liquefied H2? Still reads more like the usual make up some hypothetical use case for the highly impractical niche you chose to push the boundaries of scientific knowledge. But who am I to judge, I did semantic web in multi-agent systems back when ai was symbolic.