LED Light for Plants

[blasterman]

What I've researched in the meantime only confirms what I've stated before. In terms of terrestrial plants, particularly fruiting and vegetative plants, orange/red light plays a very significant role. The problem is that LED's aren't a very efficient source of orange/red light compared to HPS, which is the dominant commercial light source for this kind of light, and this disparity isn't going to change on the near horizon. Because Cree is getting better at making hyper efficient low CRI cool-white emitters is meaningless to a plant or symbiotic algae because white light doesn't matter. Also, because this forum is weighted heavily towards LED technology doesn't change this fact. Another problem is you have several posters here that are selling low powered LED fixtures and not being very genuine in their responses.

LEDs can generate specific amounts of 660-670nm light that other sources can't do well, but evidence that this is critical for commercial plant growth is sporadic and tends to be mentioned when it's in the commercial advantage of the proponent.

In regrads to corals, it's a different story. I've tested different colored LED lights on Acropora, Montipora, and several LPS species, and the only color that matters is blue. The closer to 440nm, the better. This is an obvious evolutionary response to the rapid absorption of red light in water. The good thing about LEDs in this respect is they are very efficient sources of far blue, so this is why LEDs are enjoying such success in reefing along with the fact they scale so well.

The real frustration is that rather than talking about this, simply order some generic 3watt LEDs and test it for yourself. Using C-channel, super glue, generic blue/red LEDs and cheap drivers you can easily build solid state light panels equal to any commercial source for a fraction the price.

 

[slebans]

What I've researched in the meantime only confirms what I've stated before. In terms of terrestrial plants, particularly fruiting and vegetative plants, orange/red light plays a very significant role.

Blue, green and many other colors(wavelengths) play equally significant roles in the process of photosynthesis.

The problem is that LED's aren't a very efficient source of orange/red light compared to HPS, which is the dominant commercial light source for this kind of light, and this disparity isn't going to change on the near horizon.

12 months ago I would have agreed with your statement. Leds producing red wavelengths could not come close to an existing HPS bulb's 40% efficiency in converting electricity to light. Once you place the bulb within a refector though - you would lose an additional 15-20% of the output. By late last year, providing you did not drive the leds too hard - you could find a red led that would match an HPS for output electrical to light efficiency. Recently, Osram released a Red 660nm led with a claimed efficiency of 40%.(Sorry for the linewrapping).
http://catalog.osram-os.com/catalog...Oid=0000000300018fab056f0023&act=showBookmark

I agree that HPS will remain the dominant source of lighting for greenhouse growers. Two things have to happen to change this outcome.
1) Led's must become efficient enough that you can pack the required watts into the same space as existing HPS fixtures and be able to economically/reliably remove the waste heat.
2) The operational cost savings of running LED versus HPS must be able to finance the capital cost for the new led light fixtures.

Because Cree is getting better at making hyper efficient low CRI cool-white emitters is meaningless to a plant or symbiotic algae because white light doesn't matter.

Not sure why you state that white light does not matter.What if the white light is 2700K at a CRI of 90? Regardless, I believe that even CREE cool-white emitters produce light that is useful to plants.

Also, because this forum is weighted heavily towards LED technology doesn't change this fact. Another problem is you have several posters here that are selling low powered LED fixtures and not being very genuine in their responses.

You sure hit the nail on the head that with that statement.

LEDs can generate specific amounts of 660-670nm light that other sources can't do well, but evidence that this is critical for commercial plant growth is sporadic and tends to be mentioned when it's in the commercial advantage of the proponent.

As you have mentioned in previous posts - plants grow just fine under HPS. I think the issue here is the relative importance led grow light suppliers attach to wavelengths between 660-680nm. Technically, they are correct in their assumption that 680nm is important for the process of photosynthesis. What many of them do not seem to understand is that 680nm is the most efficient wavelength absorbed by Chlorophyl A. This is because the energy of a photon at 680nm contains the exact energy required to raise the dimer chlorophyl A molecule at the center of an antennae complex to the necessary excited state that will allow the process to liberate an electron. Basically, allowing for variances in specific wavelength absorption rates, all absorbed photons from 380nm to 680nm will have exactly the same effect in the process to free up electrons. The difference is that the photons of shorter wavelengths than 680nm contain excess energy that the plant must then dissipate as waste heat. And since a photon is a photon is a photon - for a given amount of output watts - you will generate a lot more photons at 680nm than 380nm.

This document really helped me to understand this issue.
16.3 Photosynthetic Stages and Light-Absorbing Pigments
(hint - if you print the page to PDF all of the side bar diagrams are reproduced within the resultant document at full size/resolution)
http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=mcb&part=A4434

Finally, I would state that the majority of my research on these topics has been centered around the production of lettuce/spinach.

Just my $.02

Stephen Lebans

 

[thepaan]

So Slebans is right on here and I'd like to add to it. Also, I concur on the ncbi link - that was one of my resources as well.

I think it is important for anyone interested in participating in the discussion about plant lighting to understand how a plant uses said light.

Photosystems

If you read the link in Slebans post you will find it explained that there are two photosystems: PSI and PSII. Everything begins in PSII where light excites the core of PSII, creating a proton gradient within the system with enough force to cleave water into it's component gases. The electrons from this action are passed through mobile proteins to PSI where carbon fixation occurs and energy is created and stored in the form of ATP (temporary) and sugars (long term). Each photosystem uses an array of pigments to collect enough light and focus it towards the core (primary acceptor in the image below) in much the same way a sattelite dish focuses a signal. Thus, to achieve action, it is more important to have enough light (a strong signal) than the exact wavelength.

Absorption vs. Action

There are a lot of graphs showing what wavelengths are for plants and I think they are tossed around too casually and most of the time out of context so let me explain. In a Google search you will come out with two very different looking graphs - one is action spectra the other is absorption.

Some people have dark skin so they appear dark and their skin absorbs more visible light while some people appear light so their skin absorbs less visible lght. However, everyone absorbs some light and thus our skin has color. If it did not absorb visible light then our skin would appear perfectly white like this plastic beverage cup I have on my desk. Each spectrum of light our skin absorbs is collectively called the absorption spectra. To get an idea of the difference between this and action I would like you to think of how we get a sunburn. Not all light causes sunburns. We know this because artificial lights cannot give sunburns and so it was determined that UV wavelengths must cause sunburns. From all this we can gather that the action spectra for sunburns is very different from the absorption spectra of our skin. Similarly, the action spectra for photosynthesis is very different from the absorption spectra of chlorphyll.

Another difference you will find is between in vitro (within the glass) and in vivo (within the living) results. In vitro results have isolated the specific parts the scientist wants to study so they do not include the whole system. There are many reasons why the in vitro results are invalid for determing lighting requirements but let us leave it at that.

Absorption spectra in vitro:

Action spectra in vivo from http://members.misty.com/don/photosyn.html:

Action/Moles vs. Action/Watts

So now, we understand why we look at action instead of absorption and we can see that the action spectra has a much broader response curve than expected (e.g. you can grow a plant in only green light). But lets look closer at what exactly the above image shows. The measurements show quanta of light (Remember light is a particle as well as a wave?) used to convert an number of CO2 consumed at each wavelength. Because the number of CO2 consumed directly translates to growth, we can assume that the best growth is had at the best ratio which is the highest point on the curve. The remainder of the curve is relative to the highest point such that where it drops to .5 the plant is only consuming 50% as much CO2 - growing half as much - as at 1.0. But when was the last time you saw a data sheet for an LED list the output intensity in moles? We want watts, or at least we can convert. This link:

http://envsupport.licor.com/docs/TechNote126.pdf

has a decent comparison of action/moles vs action/watt. When we convert to watts the blue light gets a penalty because lower wavelengths have more energy.


Why LEDs?

So now we have an idea of what wavelengths drive photosynthesis best. With LEDs we can get a very narrow band in that range (around 660 nm) and not waste any electricity on other wavelengths. Unfortunately there is more to growing plants than producing growth. Examples are shaping that growth and triggering certain plant responses like flowering and fruiting. Fortunately, for the majority of plants, these responses can all be taken care of with a minimal (8-20% of red) amount of blue light (around 450 nm) and altering the length of light exposure.


Why not HPS?

HPS is not bad but it isn't optimal either. HPS is widely accepted for the volume of light (remember, I said earlier the amount is more important than the exact spectrum). It is suboptimal because it causes unecessary damage to the photosystems by overloading PSII. If we use Slebans's link to ncbi above, we see that PSII has more chlorophyll b than a where PSI has more a than b. Because PSII starts the chain but processes electrons at a faster rate due to better stimulation, it damages itself. See:

http://www.plantphysiol.org/cgi/reprint/153/3/988.pdf

for reference.

I hope I answered all of your questions. Feel free to point out any errors I may have made as I'm not exactly an expert - this is more of a hobby of mine.

 

[Oznog]

thepaan- there's an important note added:
One more note 5/3/2010:The action curve is shown to be of a measure of photosynthetic action per photon, rather than per unit energy. To convert this to photosynthetic action per unit energy, this curve needs to be weighted by wavelength. That would make the blue peak having a value around 55-60% of that of the red peak.

That's a major biasing factor. The red is much lower energy per photon, and what we really want is photosynthesis per watt of light energy. As such, the Action Spectrum chart is MUCH more favorable to red than it indicates.

Now the Action Chart... I'm still a little confused on the method (and thus the meaning). There are two options:
One is that I apply 1 mol of photons externally and measure the response. Some will be reflected and not absorbed at all, but that will not be counted.
Two, we could apply 1 mol of photons, note that say... only 70% of a wavelength is absorbed, and scale the response up to ABSORBED photons, multiplying the response measurement by 1/0.7. This is not altogether irrational, because unused reflected light may illuminate other leaves. Esp in those grow operations where they install reflectors around the plants (but I expect this is not an extremely efficient process.

I have difficulty here because the ABSORPTION chart says 525-625nm just isn't absorbed by A,B, or carotenoids at all. If the wavelength isn't absorbed, how can it be used as per the Action Spectra chart?

One theory is that the chart's only saying A,B,or carotenoids are capable of absorption, but other substances may block or reflect light before it reaches them. However, AFAIK, the green color of plants comes DIRECTLY from the chlorophyll at the surface, which reflects green & yellow, and thus does not utilize them.

That would indicate the Action Chart uses #2 and only factors in absorbed light, and is somewhat misleading because the 1 mol of applied photons in the green range will not have 50% of the "action" of light in the deep red range, because (as per the Absorption Chart) the rate of absorption is only ~5% that of the deep red!

I have difficulty believing that the Action Chart uses the #2 idea... because it's more complicated to factor in absorption. But if it doesn't then there's no way to reconcile it with the Absorption Chart, which shows absorption of greens and yellows are negligible and thus could not possibly have Action at the listed rate. And that one link http://www.general-cathexis.com/images/ActionSpectra001.png describes it as "per energy incident", which does suggest it is NOT #2, but the "note" added 5/3/2010 says it's kinda wrong about "energy" being incident, and was using mols, so I don't know how literally to read this.

 

[thepaan]

That's a major biasing factor. The red is much lower energy per photon, and what we really want is photosynthesis per watt of light energy. As such, the Action Spectrum chart is MUCH more favorable to red than it indicates.

Exactly! most charts you see will have either blue higher or the peaks nearly even but the red is actually higher and when you convert to W it becomes even more favorable for red. But lets not forget the efficiency of the LEDs themselves - where blue are used as a base for white so they have had a lot more time and energy put into them to make them more efficient.

Now the Action Chart... I'm still a little confused on the method (and thus the meaning)

O.K. So, the way this works is you have some light of a specific wavelength shone upon a leaf which is enclosed in a bubble of sorts. Inside this bubble they measure how many molecules of carbon dioxide are used up. They also measure the light that hits the surface of the leaf, called incident light. Light incident on the surface (measured in moles) divided by the number of carbon dioxide gives you a ratio. This ratio is normalized to 1 at the highest point on the curve then the remaing wavelengths are relatively adjusted. The energy terminology comes from counting the photons - each photon of a specific wavelength has a specific amount of energy. In other words, quanta of light (as long as wavelength is specified) can be considered energy of light.

I have difficulty here because the ABSORPTION chart says 525-625nm just isn't absorbed by A,B, or carotenoids at all. If the wavelength isn't absorbed, how can it be used as per the Action Spectra chart?....

First, it is very small but, both chlorophyll a and b have a non-zero absorption all the way across the visible spectrum. That they are green is only due to that being the dominant wavelength reflected - not the only one and not 100%. This absorption chart only shows absorption of those pigments as they exist in a tube. There are other protein-pigements besides chlorophyll. For absorption of an actual (bean) leaf see page 4 of this document: http://www.plantphysiol.org/cgi/reprint/46/1/1.pdf .

Here is where you have to seperate the idea that absorption of known components will be similar to action. The way a living plant uses light is very different from how the known components, after being broken apart, absorb it. We need to let go of the absorption chart as being relevant because it isn't. Let us say we want to grow a plant as quickly as possible but are going to use only a single wavelength of light. Because carbon fixation directly correlates to growth, it only matters which wavelength produces the best result compared to other wavelengths. We are not trying to stimulate any one component of the system but the system as a whole. To this end, knowing the component parts is irrelevant and knowing each component's absorption spectra is also irrelevant.

I hope this helps.

 

[Oznog]

Exactly! most charts you see will have either blue higher or the peaks nearly even but the red is actually higher and when you convert to W it becomes even more favorable for red. But lets not forget the efficiency of the LEDs themselves - where blue are used as a base for white so they have had a lot more time and energy put into them to make them more efficient.

Actually the "lumen" figure is so biased against deep red and blues, it's inappropriate to use (which is why dental blues are rated in mW).

"Lumen" is a measure based on units of "brightness" as seen by the human eye. Now exactly how they decided what 1 lumen of "brightness" of blue vs 1 lumen of green vs 1 lumen of red is, that's pretty odd, but they did it. Infrared and deep UV are 0 lumens regardless of output power. Deep red reports as a very poor lumen/photon ratio because of a poor eye response/photon ratio, but IIRC the actual efficiency- mW of light energy out vs mW of electrical energy in- calculated to about 20%, same as other LEDs. Just poor visibility, which is irrelevant because we care about plant response, not eye response.

I tried to find the Osram Golden Dragon "deep reds" mentioned above. I found them at Digikey, but the price was ludicrously high for the output. It wasn't competitive with LEDEngin. Wondering when the market will pick up on deep reds.
Bottom line, the "deep red"s are typically NOT bad for power output, even if the lumen output is very small. LEDEngin rates 525mw out at 1A @ 2.8V, efficiency of 18.75%. Well, that's if you maintain 25C, which you probably won't, but LEDs of any color suffer similar derating effects.

I tried to look up the Osram Golden Dragon "deep red" mentioned above. All I saw was Digikey, but the price was very high for the device output. It wasn't price-competitive with LEDEngin. I wonder when the market will see the value of making deep reds?

 

[Oznog]

O.K. So, the way this works is you have some light of a specific wavelength shone upon a leaf which is enclosed in a bubble of sorts. Inside this bubble they measure how many molecules of carbon dioxide are used up. They also measure the light that hits the surface of the leaf, called incident light. Light incident on the surface (measured in moles) divided by the number of carbon dioxide gives you a ratio. This ratio is normalized to 1 at the highest point on the curve then the remaing wavelengths are relatively adjusted. The energy terminology comes from counting the photons - each photon of a specific wavelength has a specific amount of energy. In other words, quanta of light (as long as wavelength is specified) can be considered energy of light.

I do see that. In this case, there might be a slight bias against wavelengths which are reflected instead of ones which are absorbed but not used. In a reflective growing container, some portion of reflected light may hit another leaf and a certain % may be utilized. However, light blocked by non-photopigment substances and not reflected is a total loss.

First, it is very small but, both chlorophyll a and b have a non-zero absorption all the way across the visible spectrum. That they are green is only due to that being the dominant wavelength reflected - not the only one and not 100%. This absorption chart only shows absorption of those pigments as they exist in a tube. There are other protein-pigements besides chlorophyll. For absorption of an actual (bean) leaf see page 4 of this document: http://www.plantphysiol.org/cgi/reprint/46/1/1.pdf .

Here is where you have to seperate the idea that absorption of known components will be similar to action. The way a living plant uses light is very different from how the known components, after being broken apart, absorb it. We need to let go of the absorption chart as being relevant because it isn't. Let us say we want to grow a plant as quickly as possible but are going to use only a single wavelength of light. Because carbon fixation directly correlates to growth, it only matters which wavelength produces the best result compared to other wavelengths. We are not trying to stimulate any one component of the system but the system as a whole. To this end, knowing the component parts is irrelevant and knowing each component's absorption spectra is also irrelevant.

I hope this helps.

Sorry I'm not trying to be difficult, but I still have trouble. OK, the Absorption chart only shows what light will be absorbed, not utilized (which is basically similar to its visible appearance of reflected colors), and painting the plant black would give it 100% "absorption"; but pointless.

But yet the Absorption chart still seems to present a limiting factor: if there is no absorption of a wavelength, it is physically impossible to utilize it. Light cannot be reflected AND used. And that's true regardless of whether it was measured on the leaf or the leave were thrown in a blender and Chloro A separated out and measured for absorption without any other structures of the leaf presenting absorption.

So the Absorption chart doesn't state how much photosynthesis is performed per mol or mW, only that it's absorbed, so it doesn't tell us the growth potential. But the problem remains, unabsorbed light couldn't be utilized for sure. This conflicts with the photoactive response charts of others in the green/yellow regions, regardless of whether it used mols or mW, because there's NO absorption shown in this range.

I'm thinking the way to logically resolve this is to simply discredit the data shown in this particular chart! It shows neither dimension nor scale... I suspect perhaps the Y-axis is not absorption=0 at y=0. Because your "bean leaf" link has an absorption chart showing something similar, but the range is 60%-95%. I think they depicted 60% at y=0 here, an undisclosed offset which is very misleading, and the whole thing is just drawn for aesthetics rather than depicting actual data like the bean leaf charts.

I guess this chart being bogus explains it. The "bean leaf" PDF shows it much better, and it all makes sense from that.

 

[thepaan]

Actually the "lumen" figure....

No where do I speak of lumens. Lumens are irrelevant here.

I guess this chart being bogus explains it.

You can think of it like that for our purposes but it isn't actually bogus.

A plant cell contains within it many different pigments. Many are known but many of these pigments are also unknown. The above chart represents the absorption of only three pigments after they have been removed from a living plant. Isolating each part allows scientists to study it without interference from the rest of the plant but without that interaction it is also impossible to determine how it would behave. So what we have are two different representations: the bean leaf being the whole plant and the above chart being a few components of the plant. I think the key point I'm trying to make here is that these few components are not all the components and their values are not necessarily additive. They could be multiplicative or one could offset the absorption spectra of another when everything is connected.

Take, for example, chlorophylls a and b. The difference between them is two hydrogen atoms and one oxygen atom. With such a minor variation the absorption of each changes noticeably - by adding just 3 atoms. Then you connect all these parts you know about back together and it still doesn't look like the original because there are parts you don't know about. It all becomes rather daunting when you look at from that perspective.

So we go back to the living plant. How well does a living plant use each wavelength? This is what we want to know because this is what we are dealing with.

 

[Oznog]

No where do I speak of lumens. Lumens are irrelevant here.

That was in regards to the regret that these deep-red LEDs aren't efficient. Actually, they seem to have the similar efficiency to other LEDs, and a high power output. Just a very low LUMEN value, which is irrelevant.

I noted that when comparing the "red" and "red-orange" Luxeon Rebels for making signal lights. The red had a substantially lower lumen value, and I thought they were inferior efficiency than red-orange and wished they'd "fix" this and give me a stronger true red. But in looking at the human eye response, I realized I was wrong- the human eye response drops off sharply outside the orange region, and the true red was a very similar radiated power. The purpose in that project WAS visual, so the shortcoming was in the human eye, not the LED technology, and thus red-orange was preferred over a true red.

You can think of it like that for our purposes but it isn't actually bogus.

A plant cell contains within it many different pigments. Many are known but many of these pigments are also unknown. The above chart represents the absorption of only three pigments after they have been removed from a living plant.

Did you note my point that if there's zero absorption at a particular wavelength, then Action would HAVE to be zero? That's critical. I did note your point that this is for separated pigments and the molar Action to create chemical energy from light does not follow Absorption. I see that. But, Action cannot take place without Absorption.

Well, I'm looking at the link you provided:
http://www.plantphysiol.org/cgi/reprint/46/1/1.pdf Fig 7 on pg 4 of 5, the Absorption line. And the Action line, really.

That's not the blendered, separated, ex vivo pigment, but I think it hints at the problem with that colorful Absorption chart. I think the colorful chart MUST have offset the results down by a LOT, probably ~60%. Which makes it look like the absorption of Chloro-A is like 100x greater at 675nm than at 510 nm. With next to zero absorption within Chloro-A,B, or Carotenoids as it appears in that chart, then Action would be impossible!

But it's not the case, as the Bean Leaf PDF chart shows the Action's like 35% greater on a mol basis at 675 vs 510nm, which would be 80% better on an energy basis. There's action, less, but not 100x less- so there must be a lot of absorption. So I think the big color Absorption chart just has so much wrong with it, it's "bogus". Chloro A/B/Carotenoids don't have zero absorption (thus no action is even possible) through the green/yellow range.

I'm still troubled and unsure of this conclusion, because EVERY chart that I get for Absorption shows NO absorption in the 500-600nm range. Nothing. Total zilch for A and almost zilch for B. So it's not this one chart. I think it'd be unusual for everyone to have put out an offset chart.

The only other explanation I could see would be that Chlorophyll has such poor efficiency that only a tiny % of Absorption is used. In which case the Absorption chart is correct and peaks are all from nonproductive absorption (like being painted with green). So the Absorption could have those very low areas and yet only have modest drops in the actual Action. But it'd need to be much less efficiency than I'd understood.

 

[Oznog]

OK, so I'm seeing some other stuff that says it's carotenoids and xanthophyll here which somehow absorb wavelengths that Chloro A&B don't, and somehow pass energy along to Chloro A&B indirectly, using a wavelength that Chloro A&B can't even absorb directly.

That kinda makes more sense now, but I wish I had a better explanation of the process. In particular the efficiency (as per the Action chart) seems unexpectedly high for an indirect path into Chlorophyll's energy.

 

[thepaan]

There really is almost no absorption in the pigments chlorophyll a and b in the green and yellow range. There are several reasons why this could be different than a living plant. As I explained earlier, it could be when connected together each component's absorption blends together. It could be there are components we don't know about. It could be that lower wavelengths (higher energy) lose some of that energy when passing through the leaf and get absorbed as a higher wavelength (lower energy). The wavelengths are obviously being absorbed by the plant but, as far as is known, not by the chlorophyll (or not much). I'm not sure what else to write here because that is about as good as I can explain it.

 

[Inkidu]

I never noticed these Osram deep reds before.

http://www.mouser.com/ProductDetail...-W5AM-1T3T-1-L-Z/?qs=dzkBnbp7AzQjkL7H45NWtQ==

They seem cheaper and more efficient than LedEngin deep reds.

Has anyone used them or have an opinion?

Thanks for any help.