NDAs being what they are, someone in the industry might be able to point us to public sources that hint in the correct direction... it seems risky otherwise.
The press release always mentions the cool new technology du jour as a possible spinoff of the discovery. Today, it's transistors.
When the Superconducting Super Collider was proposed, the newspapers said that research done at the SSC would lead to breakthroughs in computer graphics and flat panel displays.
> Gawel and his collaborators have now created and imaged the long-sought ring molecule carbon-18. Using standard ‘wet’ chemistry, his collaborator Lorel Scriven, an Oxford chemist, first synthesized molecules that included four-carbon squares coming off the ring with oxygen atoms attached to squares. The team then sent their samples to IBM laboratories in Zurich, Switzerland, where collaborators put the oxygen–carbon molecules on a layer of sodium chloride, inside a high-vacuum chamber. They manipulated the rings one at a time with electric currents (using an atomic-force microscope that can also act as a scanning-transmission microscope), to remove the extraneous, oxygen-containing parts. After much trial-and-error, micrograph scans revealed the 18-carbon structure. “I never thought I would see this,” says Scriven.
The molecule (an all-carbon cycle of 18 atoms) was prepared in a very unusual way - by directly manipulating the atoms using an atomic force microscope.
In other words, each molecule is made individually. This is not the way that chemists typically work, and will not result in quantities of material that can be seen.
The abstract says nothing about chemical stability, but I suspect C-18 quite unstable and may never be prepared in gram quantities.
Higher cycles containing more carbons may be more stable, but this is likely to remain a curiosity for some time.
Still, this is a new kind of "allotrope" of carbon. The cyclic, relatively rigid nature of the structure and the potential for electrons to circulate under applied fields could lead to some unusual applications.
It's still a good first step towards that future being possible. Now "we" know that it's actually maybe possible as well as properties of the material which might lead to better techniques for producing it or similar things using less exotic processes.
The orbitals of the pi bonds are probably in awkward angles resulting in ring strain, and thus instability. It can be explained by the antibonding orbital overlap of the pi bonds.
In the ideal configuration, when you have single-triple-single bond, they "want" to be in the same line. A molecule where they are bles have more energy. (Imagine that it's like a spring, but don't take the analogies to literally.)
The molecules where the bonds have wrong angles usually have more energy and they end to decompose in other molecules where the bonds have the correct angles or they have other bonds. [Oversimplifying warning, Chemistry is more complicated.]
In this case, since all atoms form a ring, everything is symmetric within the plane so forces cancel out. So maybe it’s more stable than one would think, similar to cubane.
But upon further consideration , the potential energy surface may be more like a saddle point, with each atom having a stable local minimum in-plane but unstable local maxima perpendicular to the plane.
(I’m not a chemist).
The atoms must be in a minimum of energy. They always oscillate a little due to thermodynamic and quantum effects. If they are in a saddle point the molecule blends until the atoms reach a minimum, until the atoms rearrange themselves in a different molecule or until the molecule split. (Or until they react with another molecule ...)
I studied a little of Chemistry, but this is out of the scope that I know well. IIUC correctly from https://en.wikipedia.org/wiki/Cubane the stability of cubane is not due to symmetry. The stability comes from the fact that locally it looks like a alkane, i.e. a molecule with Carbon and Hydrogen that only have single bonds. They are quite stable, the most common example is gasoline that is mostly alkanes.
Double bonds are more reactive. For example vegetable oils (that have also Oxygen, not only Carbon and Hydrogen) may have or not have double bonds. The one with many double bonds become rancid easier. https://en.wikipedia.org/wiki/Rancidification
Probably yes, but there may be some additional rules. First you need an even number of Carbons, but there may be additional rules.
I have a gut feeling that in this case it's better if the number of carbons is like 4n+2 (i.e. even, but not a multiple of 4, just like 18 :) ). This rule is important when you have a similar structure with double bonds and Hydrogen. https://en.wikipedia.org/wiki/Aromaticity I'm not sure if it translates to triple bonds.
(Some handwaving: Here the pi bonds in the plane act like the H in the aromatic compounds, and the pi bonds perpendicular to the plane form an aromatic system.) (Note that handwaving is never a replacement of knowledge, so this many be very bad.)
I feel like a larger ring would immediately twist into some kinda figure-8 shape. I feel like this one would too, were it not laying on a surface. Actually I feel like the surface plays a major role in this molecule's ability to exist.
> In other words, each molecule is made individually. This is not the way that chemists typically work, and will not result in quantities of material that can be seen.
I was totally thrown off, because your second and third paragraphs aren't quoted. You write in the same style as the newspaper, and just as well. Good job!
(For anyone else wondering, parent quoted just the paragraph >"Gawel... says Scriven", then added their thoughts.)
Molecular assemblers are almost taboo to discuss in chemistry and have been for decades. Sort of like the AI winter. Things feel like they are starting to thaw.
By analogy to the "AI winter" I assume they belong to the age of broken dreams... people are afraid to talk about them because too many people have been burned by too many enthusiastic promises and the problem turns out to be much harder to solve than originally expected.
>In other words, each molecule is made individually. This is not the way that chemists typically work, and will not result in quantities of material that can be seen.
Is there not merit in doing it via this 'hard way' first before optimizing stuff and finding a production pathway? (IANAchemist)
It's not really a question of hard way vs easy way. It's more like the difference between having a balloon filled with helium and making a single helium atom in a collider. One you can buy at the grocery store, the other requires a team of trained scientists and a facility full of equipment.
A milllimole (10^20 moleculess) is considered small scale by many chemists. Working with a single molecule as described in the artice is just short of science fiction, and hardly any chemists have ever done it. It requires one of the most expensive instruments in the world.
IANAchemist either, but I've heard of this being done for prototyping in the semiconductor industry; manually creating single devices to test their electrical/quantum properties while the production teams are trying to figure out how to manufacture the things industrially.
I have been a chemist, and also played with explosives as a child. So it's my experience that C≡C bonds are very unstable. I mean, liquid HC≡CH (acetylene) must be diluted with acetone, and stored in tanks packed with diatomaceous earth, to keep it from falling apart explosively. A lot like nitroglycerine, really.
And AgC≡CAg (silver acetylide, with the Ag-C bonds being almost ionic) is a damn fine primary detonator. They used to use it in party poppers and cigarette/cigar "loads". But you could use it instead of lead azide or mercury fulminate, and it's much easier to make. But it's also much more expensive.
So anyway, I can't imagine that you'd want gram quantities of this stuff. Even if you could prepare it.
Edit: A gram of silver acetylide would punch a hole through 15-20 gauge mild steel. Uncontained. Just sitting there, in a little pile.
No one would really be confused by this. Most carbon isotopes are carbon-12, with some carbon-13. Something as large as carbon-18 would immediately suggest that something fishy is going on.
I don't know the details of the "wet" chemistry step described, but I imagine it would be possible to extract CO2 from the atmosphere and transform it over many step to eventually get to the required precursors.
The practical reality is that each transformation step has imperfect yeilds and I would guess that the equipment required to do the molecule by molecule transformation in the last step would require significant energy to run.
So to attempt to this today, you would end up producing more CO2 into the atmosphere than you removed.
We can draw carbon from the atmosphere, sure. It's an inefficient process costing a lot of energy, but given they're making the final rings molecule by molecule, it isn't as if we'd need a lot.
...for the purposes of scrubbing CO2 from the air? There's absolutely no reason we'd use this as the final product. Graphite or diamond would do the trick just as well, and either is cheap to make.
The question that springs to my mind from this is 'could rings be interlinked?'
I'm not very familiar with chemistry. Are there other instances where other molecules can interlink in a way that, while not having bonds between them cannot be separated without breaking the bonds of one or the other?
At the very least it's fun to imagine a form of ringmail made of these.
It's fun to imagine. I'll admit I did as well, but you wouldn't want to wear it. Broken rings no doubt turn into nanotubes, and free carbon nanotubes have roughly the same properties as asbestos.
Depends on the bond pattern. Alternating single and triple bonds (polycarbyne or polyyne) would be arrow-straight, rigid, semiconducting, untwisting, and explosive in bulk. Triple C-C bonds are an open invitation for cross-linking.
In order to be stable, they have to be capped with a less-reactive group than hydrogen. If you break a ring, and get a C18 with naked electrons instead of protons on the ends, that's going to react with something, quickly and probably energetically.
That bond-bending energy has to go somewhere, too. If you break the ring, you're probably going to break the molecule near the antipode, too, giving you two highly reactive C9s, or a C10 and a C8. Like if you cut a bowstring at full draw, you're probably going to break one of the limbs of the bow.
Very much so. There are different synthetic methods that could be used to create these so called “catenanes”, from a slapdash wait-and-see process where the starting material is thrown together and by hope and statistics a product comes out of the reaction with two interlocked rings. Otherwise one could build up a reaction with a starting material in a shape that’s conducive to a ring structure then attempt complexation/liganding etc etc finish and close the rings so they interlock.
For some interesting reading in self-assembling structures see: Hao Li et al. Quantitative self-assembly of a purely organic three-dimensional catenane in water, Nature Chemistry (2015). DOI: 10.1038/nchem.2392
No. Benzene is an aromatic ring. It has C-C=C sequences that easily flip to C=C-C, making electrons in the ring highly mobile. A C≡C-C sequence does not easily flip to C=C=C, or vice versa. Those electrons are a lot less mobile.
Also, alternating single and double bonds leave a bonding site on each carbon for a functional group. Alternating single and triple bonds use up all four (easy) carbon bonds, and you can't cram another bond in there without an insanely powerful acid.
This is really cool. Both the action of assembling it and how they imaged it.
One of the amazing things about carbon is that it can lots of interesting things (conductor, insulator, semi-conductor, Etc) with no added elements, just different bonding and structures. Further, it is an excellent thermal conductor. As a result it has the potential to be the material that replaces silicon in computer chips on a very large scale.
> ...showed that the 18-carbon rings had alternating triple and single bonds.
Stable alternating single/triple bonds? As in the bonds don't flicker around? How is that even possible?
1. Surely the bonds can't stay in the same place
2. Surely they would 'nebulise'/spread out/whatever the right term is, over the ring
3. if 2. doesn't apply, why on earth is it more energetically favourable to have a very high energy (ok, my assumption) triple bond next to a much lower energy (ditto) single bond, and for it not to immediately snap into a pair of double bonds?
Given my n00b level of teh chemistry, what is going on here?
These alkyne (C≡C) bonds are indeed strained and unstable, but there is an energy barrier to jumping to alkane (C-C) or alkene (C=C). Otherwise, polyynes like Ichthyothereol [1] would not be stable.
Resonance is not possible with this configuration because the orbitals are not symmetric. Benzene is often drawn with alternating alkane and -ene bonds, ⌬. It's not obvious, but benzene is radially symmetric in its electronic configuration. Each carbon has p-orbitals pointing vertically from the disc, forming a cloud, rather than discrete pi bonds between neighbors, and the electrons can "skate around the ring".
The first bond between carbons is always sigma. In alkenes, there is one sigma bond, in -ynes, there are two pi's at 90 degrees to each other. I'm a bit fuzzy on my o-chem, but iirc there is an energy gap between this configuration and cumulene C=C=C, which makes spontaneous transition unlikely.
In this c18 cycloalkyne, the p-orbitals are further strained into a bent cross shape if looking toroidally. This makes for an even bigger difference in energy levels, which means it can't flip to alkene or even swap positions.
OMG which part of "I Is A N00b" did I not make clear? :-)
The role of spdf and sigma/pi electrons, I'm aware of these things but don't have a mental model for them. And "p-orbitals pointing vertically from the disc"??.
That was brutal but you've given me a lot to aim for so I'll get to reading this afternoon - thanks!
In order for a pi-bond to easily flip parallel to a different sigma-bond, there has to be some factor making the other end of the new bond more positive, or the other end of the existing bond more negative. And pi-bonds basically occupy two side slots. You can think of them as left&right for one slot and top&bottom for the other.
Aromaticity is basically alternating single and double bonds. So when one pi-bond flips to the next bond down the chain (preferably into the same slot, but not necessarily--molecules can twist and flex), you get two single bonds in a row on one side, and two double bonds in a row on the other. Ordinarily, this would be a strong incentive to flip back to the way it was before, but if a double bond further down the chain can be flipped, it might propagate further down. In a ring, like benzene or a phenyl group, that wave of bond-flipping can just chase its own tail forever. In a polymer, it can go all the way down to the end of the (long) molecule before bouncing back.
When you have alternating single and triple bonds, that triple bond uses up both mutually-perpendicular slots for the side bonds. No other pi-bond can flip over next to the adjacent sigma-bond, because the bonds on the other side just bounce it right back, like having Manute Bol and Hakeem Olajuwon on either side of the paint when the second forward of your high school junior varsity basketball team tries to go around to one side or the other to score a new pi-bond. It's just not going to happen, unless you pump Jeremy Cohen full of so much cocaine, amphetamines, and sugar that he can charge into a gigantic center and somehow bounce the defender into the top row of the bleachers.
Oh my. I had no idea of "pi-bonds basically occupy two side slots" or "uses up both mutually-perpendicular slots for the side bonds". This geometry aspect is completely new.
I'll have a goggle but if you have any introductory resources for this, that would be very helpful. Books, sites, appropriate terms to search for I may not know, anything for a beginner just to get started, I'd appreciate it. Thanks.
Electrons around an atom occupy certain probability functions that correspond to the electron's energy level. The most basic and lowest energy of these at a given tier of orbital distance is nested spherical shells (s). Next is 3 different 2-lobed dumbbell shapes (p), oriented mutually perpendicularly (px, py, pz).
The spherical mode (s) and the dumbbell-shaped modes (px, py, pz) can also hybridize when each has a single electron in it. Those hybrid modes can point lobes out in different directions. A stable mode can hold up to 2 electrons, each spinning in opposite directions.
When an electron probability function of one atom overlaps with that of an adjacent atom, their functions can further hybridize into a bonding function. (They can also hybridize into anti-bonding functions, but ignore that for now.)
When the bonding function puts the electron pair mostly along the line connecting the centers of the bound atoms, that's a sigma bond. It's the lowest-energy kind of bond, so it always (aka exceedingly rare exceptions) happens before any other kind of bond.
In a carbon triple bond, the spherical orbital (s) will hybridize with just one of the dumbbell orbitals (p), forming one function (spx+) that mostly occupies one lobe of the former dumbbell, and one that mostly occupies the other (spx-). These two will then form direct sigma bonds with atoms on either side, separated by 180 degrees. That leaves the other two dumbbell modes (py, pz), mutually perpendicular to the two hybrids. Those can align with the parallel dumbbell modes of the adjacent atom and create bonding functions where the electron positions sort of average out to between the two bound atoms, but mostly occupy places that do not overlap with the sigma bond. Those are pi bonds. They happen between two aligned p orbitals. (There is another kind of indirect bond, a delta bond, that happens between two aligned d orbitals, which have more complex shapes. Those would be the fourth bond in a quadruple bond.)
When those carbons are triple-bonded, their p-orbital atoms are leaning toward the adjacent bound atom. It's much harder for the atom on the opposite side to get their attention to maybe participate in a pi bond on that side.
If you look at the bond through a microscope, you're just going to see a big blob of electron probability between the nuclei, with 6 electrons somewhere in it most of the time, where the triple bond is. It's not organized into discrete spaces for each separate bond. It only has separate slots when you look at the mathematical model.
I don't have any specific resources, other than any modern chemistry textbook you can find. Possible search terms "electron orbitals", "bonding orbitals"
I do recognise spdf and the orbitals, I didn't realise the clouds interacted (well duh!), and that's the clearest description of pi and sigma bonds I've come across. That was brilliant, thanks!
62 comments
[ 1.3 ms ] story [ 102 ms ] threadCan someone 'in the know' on semiconductor device fabrication expand on this more?
When the Superconducting Super Collider was proposed, the newspapers said that research done at the SSC would lead to breakthroughs in computer graphics and flat panel displays.
> Initial studies of the properties of the molecule, called a cyclocarbon, suggest that it acts as a semiconductor.
The molecule (an all-carbon cycle of 18 atoms) was prepared in a very unusual way - by directly manipulating the atoms using an atomic force microscope.
In other words, each molecule is made individually. This is not the way that chemists typically work, and will not result in quantities of material that can be seen.
The abstract says nothing about chemical stability, but I suspect C-18 quite unstable and may never be prepared in gram quantities.
Higher cycles containing more carbons may be more stable, but this is likely to remain a curiosity for some time.
Still, this is a new kind of "allotrope" of carbon. The cyclic, relatively rigid nature of the structure and the potential for electrons to circulate under applied fields could lead to some unusual applications.
Edit: Expanded a bit
The molecules where the bonds have wrong angles usually have more energy and they end to decompose in other molecules where the bonds have the correct angles or they have other bonds. [Oversimplifying warning, Chemistry is more complicated.]
Some nice graphics: https://www.sciencedirect.com/topics/chemistry/triple-bond Perhaps the text is too technical, but the graphics are nice.
But upon further consideration , the potential energy surface may be more like a saddle point, with each atom having a stable local minimum in-plane but unstable local maxima perpendicular to the plane. (I’m not a chemist).
I studied a little of Chemistry, but this is out of the scope that I know well. IIUC correctly from https://en.wikipedia.org/wiki/Cubane the stability of cubane is not due to symmetry. The stability comes from the fact that locally it looks like a alkane, i.e. a molecule with Carbon and Hydrogen that only have single bonds. They are quite stable, the most common example is gasoline that is mostly alkanes.
Double bonds are more reactive. For example vegetable oils (that have also Oxygen, not only Carbon and Hydrogen) may have or not have double bonds. The one with many double bonds become rancid easier. https://en.wikipedia.org/wiki/Rancidification
I have a gut feeling that in this case it's better if the number of carbons is like 4n+2 (i.e. even, but not a multiple of 4, just like 18 :) ). This rule is important when you have a similar structure with double bonds and Hydrogen. https://en.wikipedia.org/wiki/Aromaticity I'm not sure if it translates to triple bonds.
(Some handwaving: Here the pi bonds in the plane act like the H in the aromatic compounds, and the pi bonds perpendicular to the plane form an aromatic system.) (Note that handwaving is never a replacement of knowledge, so this many be very bad.)
Depends how many graduate students they can hire.
(For anyone else wondering, parent quoted just the paragraph >"Gawel... says Scriven", then added their thoughts.)
https://en.m.wikipedia.org/wiki/Gray_goo
Wrong url?
Is there not merit in doing it via this 'hard way' first before optimizing stuff and finding a production pathway? (IANAchemist)
A milllimole (10^20 moleculess) is considered small scale by many chemists. Working with a single molecule as described in the artice is just short of science fiction, and hardly any chemists have ever done it. It requires one of the most expensive instruments in the world.
I have been a chemist, and also played with explosives as a child. So it's my experience that C≡C bonds are very unstable. I mean, liquid HC≡CH (acetylene) must be diluted with acetone, and stored in tanks packed with diatomaceous earth, to keep it from falling apart explosively. A lot like nitroglycerine, really.
And AgC≡CAg (silver acetylide, with the Ag-C bonds being almost ionic) is a damn fine primary detonator. They used to use it in party poppers and cigarette/cigar "loads". But you could use it instead of lead azide or mercury fulminate, and it's much easier to make. But it's also much more expensive.
So anyway, I can't imagine that you'd want gram quantities of this stuff. Even if you could prepare it.
Edit: A gram of silver acetylide would punch a hole through 15-20 gauge mild steel. Uncontained. Just sitting there, in a little pile.
Nit: Article confusingly uses name carbon-18 to denote a molecule with 18 atoms of carbon. Carbon-18 suggests otherwise an isotope of carbon.
”Carbon (6C) has 15 known isotopes, from ⁸C to ²²C, of which ¹²C and ¹³C are stable”
⇒ ¹⁸C exists (albeit with a half-life of about 92ms)
The practical reality is that each transformation step has imperfect yeilds and I would guess that the equipment required to do the molecule by molecule transformation in the last step would require significant energy to run.
So to attempt to this today, you would end up producing more CO2 into the atmosphere than you removed.
...for the purposes of scrubbing CO2 from the air? There's absolutely no reason we'd use this as the final product. Graphite or diamond would do the trick just as well, and either is cheap to make.
I'm not very familiar with chemistry. Are there other instances where other molecules can interlink in a way that, while not having bonds between them cannot be separated without breaking the bonds of one or the other?
At the very least it's fun to imagine a form of ringmail made of these.
In order to be stable, they have to be capped with a less-reactive group than hydrogen. If you break a ring, and get a C18 with naked electrons instead of protons on the ends, that's going to react with something, quickly and probably energetically.
That bond-bending energy has to go somewhere, too. If you break the ring, you're probably going to break the molecule near the antipode, too, giving you two highly reactive C9s, or a C10 and a C8. Like if you cut a bowstring at full draw, you're probably going to break one of the limbs of the bow.
For some interesting reading in self-assembling structures see: Hao Li et al. Quantitative self-assembly of a purely organic three-dimensional catenane in water, Nature Chemistry (2015). DOI: 10.1038/nchem.2392
Maybe 18 carbons is the minimum (C≡C-C)ₙ, because smaller rings would be too strained. But larger rings would probably be more stable.
Also, alternating single and double bonds leave a bonding site on each carbon for a functional group. Alternating single and triple bonds use up all four (easy) carbon bonds, and you can't cram another bond in there without an insanely powerful acid.
One of the amazing things about carbon is that it can lots of interesting things (conductor, insulator, semi-conductor, Etc) with no added elements, just different bonding and structures. Further, it is an excellent thermal conductor. As a result it has the potential to be the material that replaces silicon in computer chips on a very large scale.
> ...showed that the 18-carbon rings had alternating triple and single bonds.
Stable alternating single/triple bonds? As in the bonds don't flicker around? How is that even possible?
1. Surely the bonds can't stay in the same place
2. Surely they would 'nebulise'/spread out/whatever the right term is, over the ring
3. if 2. doesn't apply, why on earth is it more energetically favourable to have a very high energy (ok, my assumption) triple bond next to a much lower energy (ditto) single bond, and for it not to immediately snap into a pair of double bonds?
Given my n00b level of teh chemistry, what is going on here?
Resonance is not possible with this configuration because the orbitals are not symmetric. Benzene is often drawn with alternating alkane and -ene bonds, ⌬. It's not obvious, but benzene is radially symmetric in its electronic configuration. Each carbon has p-orbitals pointing vertically from the disc, forming a cloud, rather than discrete pi bonds between neighbors, and the electrons can "skate around the ring".
The first bond between carbons is always sigma. In alkenes, there is one sigma bond, in -ynes, there are two pi's at 90 degrees to each other. I'm a bit fuzzy on my o-chem, but iirc there is an energy gap between this configuration and cumulene C=C=C, which makes spontaneous transition unlikely.
In this c18 cycloalkyne, the p-orbitals are further strained into a bent cross shape if looking toroidally. This makes for an even bigger difference in energy levels, which means it can't flip to alkene or even swap positions.
https://en.m.wikipedia.org/wiki/Ichthyothereol
The role of spdf and sigma/pi electrons, I'm aware of these things but don't have a mental model for them. And "p-orbitals pointing vertically from the disc"??.
That was brutal but you've given me a lot to aim for so I'll get to reading this afternoon - thanks!
Aromaticity is basically alternating single and double bonds. So when one pi-bond flips to the next bond down the chain (preferably into the same slot, but not necessarily--molecules can twist and flex), you get two single bonds in a row on one side, and two double bonds in a row on the other. Ordinarily, this would be a strong incentive to flip back to the way it was before, but if a double bond further down the chain can be flipped, it might propagate further down. In a ring, like benzene or a phenyl group, that wave of bond-flipping can just chase its own tail forever. In a polymer, it can go all the way down to the end of the (long) molecule before bouncing back.
When you have alternating single and triple bonds, that triple bond uses up both mutually-perpendicular slots for the side bonds. No other pi-bond can flip over next to the adjacent sigma-bond, because the bonds on the other side just bounce it right back, like having Manute Bol and Hakeem Olajuwon on either side of the paint when the second forward of your high school junior varsity basketball team tries to go around to one side or the other to score a new pi-bond. It's just not going to happen, unless you pump Jeremy Cohen full of so much cocaine, amphetamines, and sugar that he can charge into a gigantic center and somehow bounce the defender into the top row of the bleachers.
I'll have a goggle but if you have any introductory resources for this, that would be very helpful. Books, sites, appropriate terms to search for I may not know, anything for a beginner just to get started, I'd appreciate it. Thanks.
The spherical mode (s) and the dumbbell-shaped modes (px, py, pz) can also hybridize when each has a single electron in it. Those hybrid modes can point lobes out in different directions. A stable mode can hold up to 2 electrons, each spinning in opposite directions.
When an electron probability function of one atom overlaps with that of an adjacent atom, their functions can further hybridize into a bonding function. (They can also hybridize into anti-bonding functions, but ignore that for now.)
When the bonding function puts the electron pair mostly along the line connecting the centers of the bound atoms, that's a sigma bond. It's the lowest-energy kind of bond, so it always (aka exceedingly rare exceptions) happens before any other kind of bond.
In a carbon triple bond, the spherical orbital (s) will hybridize with just one of the dumbbell orbitals (p), forming one function (spx+) that mostly occupies one lobe of the former dumbbell, and one that mostly occupies the other (spx-). These two will then form direct sigma bonds with atoms on either side, separated by 180 degrees. That leaves the other two dumbbell modes (py, pz), mutually perpendicular to the two hybrids. Those can align with the parallel dumbbell modes of the adjacent atom and create bonding functions where the electron positions sort of average out to between the two bound atoms, but mostly occupy places that do not overlap with the sigma bond. Those are pi bonds. They happen between two aligned p orbitals. (There is another kind of indirect bond, a delta bond, that happens between two aligned d orbitals, which have more complex shapes. Those would be the fourth bond in a quadruple bond.)
When those carbons are triple-bonded, their p-orbital atoms are leaning toward the adjacent bound atom. It's much harder for the atom on the opposite side to get their attention to maybe participate in a pi bond on that side.
If you look at the bond through a microscope, you're just going to see a big blob of electron probability between the nuclei, with 6 electrons somewhere in it most of the time, where the triple bond is. It's not organized into discrete spaces for each separate bond. It only has separate slots when you look at the mathematical model.
I don't have any specific resources, other than any modern chemistry textbook you can find. Possible search terms "electron orbitals", "bonding orbitals"