It all depends on what you mean by affect. Two em waves in the same space will have a different overall amplitude at any frequency.
If you mean as in the overall color of light, that will change based on how much/what frequency waves are combined. Think about adding a bit of black sand to a jar of white sand – from a distance it will appear grey but the actual colors of individual grains of sand (frequency of “individual” em waves) won’t change.
For wifi, data transmission is via phase modulation of the em wave, so the signal is resilient against adding different frequencies/amplitudes but may suffer if the same frequency is transmitted at a different phase.
It depends on whether the light is within a medium or just in vacuum. Afaik light in vacuum behaves entirely linear (so waves of different frequencies don’t interact). But there are materials where light does indeed interact with light of different frequencies. One effect like this is so-called four-wave mixing. https://en.wikipedia.org/wiki/Four-wave_mixing?wprov=sfla1
In general you can take a look at non linear optics
I will! Thank you! Also, it’s super fun that there’s exceptions based on the medium; I had no idea. I was picturing air or vacuum when I conceived of the original question, so now I have other things to look into!
I second the other poster’s suggestion to look into nonlinear optics. A really common application is frequency doubling, also known as second harmonic generation, which doubles the energy of the photons. So an 800 nm laser (red) can be converted to 400 nm (green) with this method.
Another optics-based phenomenon that I think maybe strays too far from the intent of your initial question, but is too cool not to share, is laser Wakefield acceleration. A very high power laser pulse will push electrons out of its path in plasma or materials via the ponderomotive force. This charge separation creates electric field gradients on the order of billions of volts per centimeter, which can accelerate electrons or other charged particles to relativistic energies. So you can start with a green laser pulse and wind up producing gamma-ray beams, either by slamming the electrons into a stopping material or by Compton scattering other low energy photons off the relativistic electrons.
I really appreciate the extra info! It’s fascinating.
I’m recently an ex-fundie, so learning about all the cool stuff happening in science is like finding out your childhood house has a million secret rooms you never knew about.
I think when people say it’s only predicting the next word, it’s a bit of an oversimplification to explain that the AI is not actually intelligent. It’s more or less stringing words together in a way that seems plausible.
They’re very good at predicting the next word, so their choice of “a” or “an” is likely to make sense in context. But you can absolutely ask a GPT to continue a sentence that appears to use the wrong word.
For instance, I just tried giving a GPT this to start with:
My favorite fruit grows on trees, is red, and can be made into pies. It is a
And the GPT finished it with:
delicious and versatile fruit called apples!
So as you can see, language is malleable enough to make sense of most inputs. Though occasionally, a GPT will get caught up in a nonsensical phrase due to this behavior.
GPT creates plausible looking sentences, it has no concept of truth or anything like that. Since if you have an “an” it’s overwhelmingly likely that the next word will begin with a vowel it will choose one which plausibly fits with the corpus of text that came before. Likewise for an “a”.
There is no compromise in ability. It doesn’t have anything to “say” or whatever. What it produces is more like nonsense poetry than speech.
To be overly simple about it, the LLM uses statistics and a bit of controlled RNG to pick its words. Words in the LLM have links to each other with statistical probabilities attached. If you take the sentence “I fed a peanut to an elephant” and “I fed a peanut to a elephant” and then asked 100 people which is more correct, there will be a percentage which favors one over the other. Now with LLMs its not choosing using weighted coin flips, but rather picking the most likely next word (most of the time). So if the 100 people choose “an elephant” over “a elephant” 65% of the time in its training data, then the LLM will be inclined to use “an elephant.” However, Its important to know that the words around “an elephant” will also bias its choice to use the word ‘an’ for the word ‘elephant’.
Really, its largely based on the training data and the contexts to which ‘a’ and ‘an’ are used. Or in other words, the LLM knows because people figured it out for the LLM. People did all the thinking, LLM’s just use statistics on our bottled phrases to know when to use which. Of course, because it got its data from people - it will sometimes get it wrong which is based on how often people got it wrong.
Well, my example of the word ‘elephant’ has the same property as ‘herb’ where the use of ‘a’ or ‘an’ can depend on who you ask. I chose my example trying to anticipate this exact question, and I believe I gave you an answer.
Let me put it this way: it depends… It depends on the data the LLM (Chat GPT for example) has been given to train its output. If we have an LLM dataset which uses only text by people in the United Kingdom, then the data will favor “a herb” as the ‘h’ is pronounced, where data from the United States will favor the other way as the ‘h’ is usually silent when spoken out loud.
As a fairly general rule, people use the article “an” before a vowel sound (like a silent “h”) and “a” before a consonant sound (like a pronounced, or aspirated, “h”). Usually the data gathered is from multiple English speaking countries, so both “an herb” and “a herb” will exist in the training data, and from there the LLM will favor picking the one that is shown more often (as the data will biased.)
Just for fun, I asked the LLM running on my local machine. Prompt: "Fill in the blank: “It is _ herb” Response: “It is an herb.”
If it generates “I ate” and the next word can be “a” or “an”, then it will just generate one or the other based on how often they appear after “I ate”. It hasn’t decided by this point what it has eaten. After it has generated the next token, for example “I ate an”, then its next token is now limited to food items that fit the grammatical structure of this sentence so far. Now it can decide: did I eat an apple? An orange? An awesome steak? etc
There are many ways to do spectroscopy because of the wide range of wavelenghts of light. I won’t go into detail, but essentially what spectroscopy does is either:
Put energy into a sample and see what is absorbed (absorbtion spectroscopy)
Put energy in a sample and see what comes out (emission spectroscopy)
The reason those two methods produce characteristic results for each element is the following: An atom is made up of a nucleus of a certain charge and electrons canceling that charge around it. Those electrons are confined to so-called orbitals due to quantum weirdness (the “quantisation” of the orbitals is literally the origin of the word quantum). Those orbitals have different energies (you can imagine that an electron being very close to the nucleus is more strongly attracted than an electron which is farther away).
Because the electrons need to always be on those orbitals with fixed energies, only certain energies of photons can interact with them (if a different energy photon wanted to interact with an electron it would need to push the electron “between” two orbitals which is forbidden by quantum mechanics)
So now only certain energies of photons (which relate directly to wavelength) are absorbed, the rest passes uninterrupted leading to bands in the spectrum where lots of photons are absorbed.
Now depending on how many electrons your atom has and how far away they are from the nucleus those absorbtion bands will vary, giving you a good idea which atom you are looking at.
Emission spectroscopy works the other way around, instead of you seeing what is absorbed, you randomly put energy (often using heat) into the atom. When the atom wants to go back to its most stable state it has to emit a photon, this photon needs to correspond to a gab between two orbitals (because else the electron either starts or ends outside of an orbital (which is forbidden))
For molecules the elecyrons of the individual atoms are mixed together into their own molecular orbitals that follow the same logic the commenter above had written with respect to energy levels and photons.
I’m specifying this because the OP was asking about individual elements within a molecule, and that’s not how that works. The electrons are shared so you don’t get the emissions from the elements composing the atoms in the molecules on their own.
Waiting for an actual spectrographer to weigh in, but I think there are databases of molecular emission spectra that can be used to match a sample for complex molecules.
Each element has a known set of emission lines. Mixing elements together in molecules can shift these lines some and add them together. en.wikipedia.org/wiki/Emission_spectrum has some examples across the visible spectrum.
Afaik emission spectra are measured for astronomy and passive remote sensing (since generally you’re just capturing what’s already being emitted). Most spectrometers or spectroradiometers can be used to measure emission or reflection, so then it’s just a question of if you want to measure a sample’s reflectivity by shining a known source of light at it or it’s emission by exciting it with heat or electricity or lasers.
I think Raman spectroscopy is used to excite a crystal lattice with a laser to identify its structure based on the wavelengths emitted outside the laser band, so it has a specific application on crystallography, just like X-ray diffraction.
Also don’t forget mass spectrometers literally rip apart molecules and sort the atoms by mass, so the relative composition of elements can be measured directly.
Not all methods can pick up on all materials, and some methods are easier to use or faster or more accurate or you can so multiple things at once or probes something else than just what the material is.
For instance, EDX is a method where you hit the sample with an electron beam from a scanning electron microscope (SEM) and measure emitting x-rays. You can probe lots of elements with this, but not the very lightest of elements (skipping couple at the beginning of the periodic chart, don’t remember which). It is very accurate in terms of spectra and because it is an electron beam it has very high physical resolution. However, you need to use pretty high energies which can destroy or modify the sample when doing the scans. Also, you need an electron microscope with the capability, it is common to skip it and use the money for improving the electron microscope in other means. And then you need to put your sample in a high vacuum, which might be a problem. It is not exactly fast and SEMs are used a lot so the tool might be booked a lot, just a practical issue but a very real one. Also SEM costs hundreds of thousands of euros or millions. Nor cheap.
XRD is a method where you blast your sample with x-rays ans look at how they diffract from it. You can use it to probe which materials your sample is made of but you also get information about its crystal structure and things like distance between two atomic layers (very accurately). Issues are for instance that you might need to grind your sample into dust basically to do the measurement (it only probes the very surface of the sample). And it is not a high resolution method in the physical dimensions, you can tell what the entire sample is made of but not really what is this specific spot made of.
Then methods like Raman and infrared spectroscopy use lasers to excite molecules on the sample and then look at what the sample spews out. They both can be used to know what materials the sample is made of (at the laser spot), but not everything is “Raman active” or “infrared active”. Like I mentioned, they probe molecules and not necessarily individual atoms. Essentially they look at how molecules vibrate and rotate and how the electron cloud around the atoms stretch and move when being hit by lasers. EDX might tell you a material is made of carbon 12, but how is it arranged? Amorphous carbon (no crystal structure)? Buckyballs (small clusters of couple tens of carbon atoms in a ball)? Carbon nanotubes (sheet of carbon rolled into a tube)? Graphene (2D sheet of carbon aroms)? Raman and/or IR spectroscopy can tell you that. Now to be fair, EDX can also differentiate between those (or the electron microscope can as a whole) but it will have though time telling how well the atoms are arranged (missing atoms, doping, extra atoms etc).
Of course you can just take white light (usually maybe 300-1000nm or so), shine the sample with it and look at the spectra. Either transmissed light (light that goes through) or reflected. Then you can run into issues like, well most stuff doesn’t let white light through that easily for transmission, and not all samples reflect that well. Here you have looooots of different wavelengths so just making that wide frequency band well is difficult, hence it is usually limited to around the visible spectrum, and this also is a problem in spectral resolution (tends to be lower). And all frequencies interact with the sample a bit differently, so here afaik you don’t really get any more info that literally what the sample reflects / passes through. So no crystal structure or anything fancy.
if elements drive the colors, how do you parse out individual elements from a compound?
You might not be able to. What you might see is how many percent of X and Y and Z you have and from this you could determine what type of a compound you have. You would probably start to look at phase diagrams of those elements and from there you might be able to determine what compound you have.
Then you could also run into issues with spectral resolution and non-idealistic measurement conditions. For instance you might see a peak of spectra at some wavelength, but the peak is not a single line, it probably looks like Gaussian curve or Lorentzian. Now you can have multiple peaks very close by but because of the resolution of the system isn’t high enough to see them as individual peaks, you would see one big peak. To get around this you probably need to do some math and try to fit multiple peaks into your measurement data and see what peaks make up the big boi peak.
Is there a consistent pattern to how the emission lines relate to what the element looks like before going through the prism?
Pretty much yes. The theoretical peaks are what they are but your measurement data is noisy (like previous example). Your electrons are not at the same energy. You get secondary electrons that can mess up things. Your laser isn’t at one frequency and might change a bit from measurement to measurement. If you need accurate results you might need to make some calibration steps. Like, measure something you know is 99.999% copper, then adjust the setup so that it identifies it correctly. For lasers you should measure the frequency and adjust the results based on that. You might also need to measure like spectrum of a xenon lamp that has very well known, strong peaks, and adjust according to that.
Is there a resource that shows examples of emission lines along with the visible color?
There are lots of books on spectroscopy of different materials that work as reference. NIST also has a database www.nist.gov/pml/atomic-spectra-database but I have not used it myself. But usually the tool itself has a database that you just query from and it tells you.
Awesome, so if I’m reading that right UV can contribute to growth, and IR doesn’t contribute much at all. The blue and red end of the visible spectrum encourages growth, however it can pretty much occur at any visible wavelength, just not as efficient as the bands listed on the image.
Blue and red are the biggest drivers for photosynthesis, but other wavelengths have effects on the plants as well. For example, far red light (infrared) has a huge influence on flowering and stem elongation in many plants. They really rely on light for a lot of cues for regulation.
That chart is a good starting point, but it doesn’t tell the whole story. It only shows the absorption spectrum for Chlorophyll A, which is a key pigment for photosynthesis in all plant species, but there is also Chlorophyll B, and there are numerous other so called accessory pigments (beta carotinoids are the most common examples), that can work in conjunction with chlorophyll in a photosystem to collect light of different wavelengths. Algae in particular show a greater diversity of accessory pigments. They have evolved use light at different depths below the water surface, where it reaches them with a different spectrum, because water itself absorbs part of it. Have a look at red algae and brown algae, for example.
So if you want to grow algae in a controlled environment, you should do some research about the particular absorption preferences of the species you’d like to grow.
Plants use visible light for photosynthesis. Visible light ranges from low blue to far-red light and is described as the wavelengths between 380 nm and 750 nm. The region between 400 nm and 700 nm is what plants primarily use to drive photosynthesis and is typically referred to as Photosynthetically Active Radiation (PAR). Plant biologists quantify PAR using the number of photons in the 400-700 nm range received by a surface for a specified amount of time, or the Photosynthetic Photon Flux Density (PPFD) in the units μmol/s.
During photosynthesis, plants take in carbon dioxide (CO2) and water (H2O) from the air and soil. Within the plant cell, the water is oxidized, meaning it loses electrons, while the carbon dioxide is reduced, meaning it gains electrons. This transforms the water into oxygen and the carbon dioxide into glucose. The plant then releases the oxygen back into the air, and stores energy within the glucose molecules.
It’s called a photosystem and theirs 2. Just search up photosystem. The first one is actually the second one and the second one is actually the first one.
I’m late to this, but I’d like to bring up something I haven’t seen anyone else mention. But first, some more details regarding what has been discussed:
In most situations, it’s correct to say that EM waves basically don’t interact with one another. You can cross two laser beams, and they’ll just continue on their way without caring that the other one was present. A mathematically equivalent scenario is waves on a string: the propagation of a wave isn’t affected by the propagation of another, even when they overlap. Another way to put this is that they obey the principle of superposition: the total amplitude at any given point on the string is just the sum of the amplitudes of the individual waves at that point. You may want to argue that the waves do interact because there are interference effects, but interference is exactly what you get when they don’t interact, i.e. when the principle of superposition holds.
However, this is only true for so-called linear systems. I won’t go delve too deep into the math of what this means, but I think looking at the wave on a string example can give you some intuition. The behavior of waves on a string can be explained mathematically by treating the string as a large number of tiny points connected by springs. If the force on a given point by a neighboring spring is directly proportional (i.e. linear) to the spring displacement (Hooke’s law), then you find that the entire system obeys the wave equation, which is a linear equation. This is the idealized model of a string, and the principle of superposition holds for it perfectly. If, however, the forces acting on points within the string have a non-linear dependence on displacement, then the equation describing the overall motion of the string will be non-linear and the principle of superposition will no longer hold perfectly. In such a case, two propagating waves could interact with one another as the properties of the wave medium (the “stretchiness” of the string) would be influenced by the presence of a wave. In other words, the stretchiness of the string would change depending on how much it’s stretched (e.g. if a wave is propagating on it), and the stretchiness influences the propagation of waves.
Something analogous can happen with EM waves, and has been mentioned by others. In so-called non-linear media, the electromagnetic wave equation becomes non-linear and two beams of light (propagating EM waves) can influence one another through the medium. This makes sense when you consider that the optical properties of a material can be changed, even just temporarily*, when enough light is passed through it (for example, by influencing the state of the electrons in the material). It makes sense then that this modification to the optical properties of the material would influence the propagation of other waves through it. In the string example, this is analogous to the string itself being modified by the presence of a wave (even just temporarily) and thereby influencing the propagation of other waves. Such effects require sufficiently large wave amplitudes to be noticeable, i.e. the intensity of the light needs to be high enough to appreciably modify the medium.
What about the case of light propagating in a vacuum? If the vacuum itself is the medium, surely it can’t be altered and no non-linear effects could arise, right? In classical electromagnetism (Maxwell’s equations), this is true. But within quantum electrodynamics (QED), it is possible for the vacuum itself to become non-linear when the strength of the electromagnetic field is great enough. This is known as the Schwinger limit, and reaching it requires extremely high field strengths, orders of magnitude higher than what we can currently achieve with any laser.
*I want to emphasize that we’re not necessarily talking about permanent changes to the medium. In the case of waves on a string for example, the string doesn’t need to be stretched to the point of permanent deformation; non-permanent changes to its stretchiness are sufficient.
askscience
Oldest
This magazine is from a federated server and may be incomplete. Browse more on the original instance.