Water Condensor by MIT using Metal Organic Framework

[trophosphere]

Massachusetts Institute of Technology (MIT) has created a water harvester that uses sunlight to condense water out of air at low humidity. The special thing about it is the class of material used to bind the water vapor called metal-organic framework (MOF) produced at UC Berkeley. Prototype unit was reported to pull 2.8L of water in 12 hours in an environment with 20-30% humidity.

More Information:
News Article 1
News Article 2
Article (Paywalled)

[CatalinaWOW]

Good stuff, but for many desert dwellers 20-30% humidity is a muggy horridly wet day.

[BradC]

Good stuff, but for many desert dwellers 20-30% humidity is a muggy horridly wet day.

Day, sure. But it actually absorbs at night when it's generally considerably more humid.

[MrOmnos]

Good stuff, but for many desert dwellers 20-30% humidity is a muggy horridly wet day.

Day, sure. But it actually absorbs at night when it's generally considerably more humid.

But no sunlight at night which means good sized batteries for storing energy. Non the less great idea.

[BradC]

Good stuff, but for many desert dwellers 20-30% humidity is a muggy horridly wet day.

Day, sure. But it actually absorbs at night when it's generally considerably more humid.

But no sunlight at night which means good sized batteries for storing energy. Non the less great idea.

Didn't read any of the articles huh?

[MrOmnos]

Nope. I just assumed things. Sorry my bad.

[EEVblog]

"There is a lot of potential for scaling up the amount of water that is being harvested. It is just a matter of further engineering now." 

[station240]

Figured I would attempt the maths for what volume of air is needed to get 2.8L of water.

Assume 5.6g of water per cubic meter of air (20% of 28 grams of water per cubic meter of air at 30 °C)
At nightime when the air is colder, it can hold less volume of water, but the amount of air and water doesn't change.

Therefore to get 2800mL of water (1gram = 1mL), you need to process at least 500 cubic meters of air. Sounds plausible, for a small scale emergency water supply. For a large scale water capture system, this is never going to work easily, too easy to deplete a local area of water vapour.

[mtdoc]

"There is a lot of potential for scaling up the amount of water that is being harvested. It is just a matter of further engineering now." 

I'm disappointed Dave. Are you not able to distinguish between actual science and engineering breakthroughs and commercial scams?  Who knows if this will eventually lead to any commercial product.  That's not what this report in Science was about.

This was brought up in another thread.  I'll repeat what I posted there:

That is a summary of a legitimate scientific research article published in Science - one of the 2 most prestigious scientific journals in the world (the other one being Nature).

It's about as far from BS as it gets.

-- - the "debunking" meme is way past it's prime. It may work well as click-bait, but an unfortunate side effect is it leaves many unable to distinguish between real science/engineering and solar-roadways type BS.

[CatalinaWOW]

This has real potential, but there is a lot that could kill it.  Would be a perfect fit for coastal equatorial deserts where temperatures are warm and relatively high humidity occurs at night.  Other deserts remain dry in the winter when temperatures are much lower and water content of the air drops dramatically.  Whether the capital costs will be low enough to make sense for a device that only works well part of the year will be an interesting question.

[JimRemington]

That is a summary of a legitimate scientific research article published in Science - one of the 2 most prestigious scientific journals in the world (the other one being Nature).

Both of those journals have far too frequently published articles that later turned out to be complete BS, based on fabricated or incorrectly interpreted data.

While this article may be legitimate, it is not a good idea to revere any scientific journal, including those two.

[mtdoc]

That is a summary of a legitimate scientific research article published in Science - one of the 2 most prestigious scientific journals in the world (the other one being Nature).

Both of those journals have far too frequently published articles that later turned out to be complete BS, based on fabricated or incorrectly interpreted data.

While this article may be legitimate, it is not a good idea to revere any scientific journal, including those two.

All human endevors are imperfect. ALL scientific journals occasionally see failures in the peer review and editorial process. But the scientific community generally considers Science and Nature to be the best of breed. The instances of this are rare and later rectified.   On the other hand, if a dishonest scientist fabricates data then that can only be caught after publication and attempts to repeat done by other scientists.  But again, the scientific process included a mechanism for flushing this out.  But of course any scientist who does such a thing has then ended their career and ruined their reputation.

My point was never to imply that the journal Science is infallable.  My point was that dismissing this report as BS because you think it seem implausible or because of  back of envelope calculations, etc shows a lack of ability to discriminate between real science and solar-roadways type scams.

As  I posted in the other thread on this (which should really be merged with this one - mods?), anyone who wants to question the authors (from the MIT ME dept and Cal Chemistry dept) conclusions in this report  then the only legitimate way to do that is to read the original source article in Science.  Then, if they feel they're qualified to question Kim et al.'s methods or dispute their data they should post it here.

If they have a valid criticism of their methods or data, they should write a letter to the editor of Science and it will be published.  In doing so they would instantly achieve some notoriety in the scientific community.

If that sounds sarcastic - it really is not meant to be.  Findings published in Science are widely read in the scientific community and hold great weight. Careers are literally made by having a research article published in this journal. If there is a fundamental flaw in the methods or data of an article published there, then publishing that could in itself could make a career.

[TheAmmoniacal]

Nature, Science and Cell have the highest impact factors, and are the most recognized scientific journals. But this does not mean that they are good scientific journals, far from it. They are highly popularized and generalized journals, they push hype and force authors to edit it down to absolute essentials. They are very much the tabloids of scientific journals, far from what journals should and ought to be. One major issue is that journals in Nature cannot be replicated, and are very difficult to read /even for scientists in the field/, because they are so condensed and made to have a broad appeal.

You publish in these journals if you want to further your academic career, not if you want to further your scientific field.

[mtdoc]

Nature, Science and Cell have the highest impact factors, and are the most recognized scientific journals. But this does not mean that they are good scientific journals, far from it.

Well I guess that depends on how you define good.  They are different than the best field specific journals.  But they don't publish rubbish as some of the lesser field specific journals do.

They are highly popularized and generalized journals, they push hype and force authors to edit it down to absolute essentials.

Yep, I agree - Though I think the hype factor is a more recent development. I think they force authors to condense their work to the essentials because of their generalized nature.  But I don't view that as necessary a bad thing. 

They are very much the tabloids of scientific journals, far from what journals should and ought to be  One major issue is that journals in Nature cannot be replicated, and are very difficult to read /even for scientists in the field/, because they are so condensed and made to have a broad appeal.

Well calling them "tabloids" is hyperbole IMO. I think they are exactly what they are meant to be - that is general scientific journals publishing the highlights of science that will have broader appeal. I'm not sure I understand what you mean by cannot be replicated - do you mean photocopied?   Also I don't find them difficult to read at all. Maybe a language thing?

You publish in these journals if you want to further your academic career, not if you want to further your scientific field.

I don't disagree.  But as I'm sure you are aware, results published in these journals are often also published in more extensive and complete form in a top field specific journal.

[JimRemington]

They are highly popularized and generalized journals, they push hype and force authors to edit it down to absolute essentials.

Agreed, even though I have published in those journals and frequently review research papers for them. The reproducibility problem is especially severe.

Further objections that are very frequently raised about Science, Nature and Cell (and their numerous spawn) include the fact that unqualified reviewing editors have first rejection rights, so that many excellent papers that are not considered "sexy" enough are thrown out in the first round, without formal review. Then, papers are often sent to unqualified referees, and legitimate issues raised by qualified referees are often ignored, especially if an article is considered sufficiently flashy. Finally, it is sometimes years before a completely false publication is rooted out, causing great setbacks in specialized fields.

[TheAmmoniacal]

Nature, Science and Cell have the highest impact factors, and are the most recognized scientific journals. But this does not mean that they are good scientific journals, far from it.

Well I guess that depends on how you define good.  They are different than the best field specific journals.  But they don't publish rubbish as some of the lesser field specific journals do.

They are highly popularized and generalized journals, they push hype and force authors to edit it down to absolute essentials.

Yep, I agree - Though I think the hype factor is a more recent development. I think they force authors to condense their work to the essentials because of their generalized nature.  But I don't view that as necessary a bad thing. 

They are very much the tabloids of scientific journals, far from what journals should and ought to be  One major issue is that journals in Nature cannot be replicated, and are very difficult to read /even for scientists in the field/, because they are so condensed and made to have a broad appeal.

Well calling them "tabloids" is hyperbole IMO. I think they are exactly what they are meant to be - that is general scientific journals publishing the highlights of science that will have broader appeal. I'm not sure I understand what you mean by cannot be replicated - do you mean photocopied?   Also I don't find them difficult to read at all. Maybe a language thing?

You publish in these journals if you want to further your academic career, not if you want to further your scientific field.

I don't disagree.  But as I'm sure you are aware, results published in these journals are often also published in more extensive and complete form in a top field specific journal.

They are difficult to read because they use non-standard terminology a lot, leave out important details and often make leaps in logic (as a result of condensing the script, which often leads to jumping between "what we did" without connecting it with reasoning, thought-process, or justification).

Leaving out these important details makes the experiments themselves impossible to replicate by other research groups.
And good luck trying to get good answers from them by email.

[mtdoc]

Thanks for your inside scoop Jim. I've never published in those journals but as a graduate student in the 1980s a postdoc in my lab did have a "cover article" published in Science and I recall some of the politics.

After your post I think I realize now what Ammominiacal (and you) meant about reproducability. You are referring to inability to reproduce the research due to lack of detailed materials and methods in the journal article. Yes, I can see how that would be an issue.

As mentioned - at least in my experience, often an article in Science or Nature has referenced or later followed by a more traditional, complete article in another field specific journal. But apparently that must not be true in all fields - or maybe less common now then it was when I was last active in research.

[TheAmmoniacal]

To be fair, Nature and Science most often provide a lot of the details in supplementary materials. It's just often not enough, not sure if it's deliberate or of it just gets lost in the editing process/condensing.

[edavid]

Another factor is that MIT has the most aggressive press office in existence.  They will do anything to publicize MIT.  I am always extra skeptical of anything they have touched.

[mtdoc]

Another factor is that MIT has the most aggressive press office in existence.  They will do anything to publicize MIT.  I am always extra skeptical of anything they have touched.

Ha! Yes that might explain why this was picked up by so many science "journalism" type venues. 

To be fair though, that is a separate issue from the research itself and journal publication (though perhaps some pressure is applied to submit to Science?).   IME, most scientists are not looking for general publicity and often try to avoid it. There are notable exceptions though!

[Habropoda]

I wondered why information was so difficult to tease out of the paper.  Thanks for explaining.

Because of the obtuseness of the paper, the bunk has started with the media grabbing onto the 2.8L figure and showing a picture of the little prototype, implying or sometimes outright stating that it produced that amount in a 24 hour real world test with one solar thermal cycle.  It’s only a matter of time before we see animations of women and children walking miles to get water and of water bottles magically filling up from tiny devices.  Talk about clickbait, it’s just too temping for the media.  Let the debunking continue!

It looks like the 2.8L figure was produced by a simulation drawn from small scale lab tests and not the prototype.  The simulation was for a 24 hour period using multiple thermal cycles with one kilogram of MOF, rather that just one cycle in a normal solar day.

Is anybody seeing in the report just how much water they got out of the prototype in their real world test?  How many square meters of material would it take to get that 2.8L in 24 hours outdoors? 

It may be that this will be useful technology but it would be good to know the real limitations and not the BS the media is spitting out.

edit: corrected simulation parameters.

[rfeecs]

Some more progress on this:

"In field tests, device harvests water from desert air":
http://news.mit.edu/2018/field-tests-device-harvests-water-desert-air-0322

The full paper:
"Adsorption-based atmospheric water harvesting device for arid climates":
https://www.nature.com/articles/s41467-018-03162-7

[mtdoc]

Some more progress on this:

"In field tests, device harvests water from desert air":
http://news.mit.edu/2018/field-tests-device-harvests-water-desert-air-0322

The full paper:
"Adsorption-based atmospheric water harvesting device for arid climates":
https://www.nature.com/articles/s41467-018-03162-7

But, but,  tHunderFool said it was a hoax..

As per Evelyn Wang, the PI:

Last year’s paper drew a great deal of attention, Wang says. “It got a lot of hype, and some criticism,” she says. Now, “all of the questions that were raised from last time were explicitly demonstrated in this paper. We’ve validated those points.”

[rfeecs]

Trying to figure out how much water they actually harvested.  From the paper:

Due to the limited quantity of MOF-801 used in the device (~3 g), accurate measurement of the quantity of harvested water was not possible, albeit we expected ~0.75 g of water production. Therefore, we used validated computational predictions⁴ based on the measured conditions during the water harvesting cycles (ambient and condenser temperatures, RH, and solar flux) to evaluate the deliverable water capacity.

So too little to accurately measure?

[TheAmmoniacal]

It's the same joke as last time, nothing new.