LearnChem · Part V — Why that product and not another?
Put one oxygen atom into a chain of carbons and the chain starts doing things it never could.
Why does clove oil numb a toothache?
Almost all of that oil is one compound, and the oxygen on it is what this page is about.
Module 26 put you at a fork in the road. One branch ended at a gate that swings both ways. The other ended at a gate that only opens one way. Which gate you walked through decided what you were left holding. That was a choice between two roads out of one molecule. This page does something different. It changes the molecule itself. One oxygen atom goes into the chain. Then the chain does things a plain carbon chain never does.
Clove oil has been rubbed on aching teeth for centuries. A 2021 review records its use as a topical painkiller in dentistry. Most of the oil is one compound, called eugenol. Eugenol is built on a six-carbon ring. Two oxygens sit next to each other on that ring. One of them is an O-H group. That means an oxygen with a hydrogen bonded straight to it. The other oxygen links the ring to a methyl group. A methyl group is a carbon carrying three hydrogens. So there are hydrogens close to both oxygens. Hold that picture for the question below.
Eugenol's methyl group puts three hydrogens one atom from an oxygen. Can water take one of those?
Answer from your gut. The reveal follows.
The O-H on the ring makes this a phenol. A phenol is an O-H bonded straight to a benzene ring. The other oxygen has a carbon on each side. An oxygen with a carbon on both sides is an ether. The methyl hydrogens are bonded to carbon, not to oxygen. They cannot leave as hydrogen ions. One oxygen here can let a hydrogen go. The other cannot. Why clove oil numbs a tooth is a question about nerves, and this page does not answer it. What this page took from the oil is those two oxygens.
Here is the picture to carry through the page. When an O-H does let its hydrogen go, a negative charge is left behind. Sometimes that charge is penned on the one oxygen. Sometimes it is spread over a whole ring. The rest of this module is about how far.
Oxygen only gives up a hydrogen it has, and only as easily as the leftover charge can spread.
That sentence has two halves. This page takes them in order. The first half is about having a hydrogen at all. So the next section takes an oxygen that has none.
— The element is the same twice over, and on its own it settles nothing.
An ether is the control case for this whole page. It holds the oxygen and takes the hydrogen away. Ethanol and dimethyl ether are made of exactly the same atoms. Both are two carbons, six hydrogens and one oxygen. In ethanol the oxygen carries one of those hydrogens. In dimethyl ether the oxygen sits between the two carbons instead. Nothing else about the two molecules differs by much. They weigh the same.
Swap ethanol's O-H hydrogen for a second carbon. What does the boiling point do?
Predict, then read the arithmetic below.
Four common solvents form explosive peroxides while they sit in storage. They are diethyl ether, diisopropyl ether, THF and 1,4-dioxane. THF is short for tetrahydrofuran. A peroxide is an oxygen-oxygen bond, and it can detonate. The danger is worst when a sample is concentrated by evaporation. It is worst of all when one is distilled toward dryness. Diisopropyl ether is the quickest of the four to go bad. An opened bottle needs testing once it has sat unused for three months. It must not be kept past six months. The other three are slower. An opened bottle needs testing after twelve months of no use. It must not be kept past five years. These are facts about how such bottles are stored and handled. They are not steps for anyone reading this page to carry out.
University of Nevada, Reno, Environmental Health and Safety, 'Peroxide-Forming Chemicals Policy'. Online at unr.edu/ehs/policies-manuals/peroxide-forming-chemicals. Diisopropyl ether is on the policy's List A. Diethyl ether, THF and 1,4-dioxane are on its List B. The three-month and six-month windows are List A's. The twelve-month and five-year windows are List B's. The policy states that explosions commonly occur during distillation or other concentration. It says a distillation is stopped with liquid still left in the flask. Sigma-Aldrich, 'Peroxide Forming Solvents', a technical article, at sigmaaldrich.com. That is the source for the slow reaction with atmospheric oxygen in storage. It is also the source for light and heat speeding that up. Neither source says which part of the molecule the oxygen attacks.
Two comparisons sit in the panel below. The first is ethanol against dimethyl ether. Same atoms, same weight, and the only difference is where the hydrogen sits. That comparison isolates the O-H. The second is dimethyl ether against propane. Propane has no oxygen at all, and neither molecule has an O-H. That comparison isolates the oxygen on its own. Three water-solubility figures follow. Those are an ether, an alcohol and a plain hydrocarbon, the same three roles one carbon larger.
The hazard above is worth tying to the chemistry. An ether oxygen still holds electrons that no bond is using. Water can hook onto those electrons, which is why an ether dissolves at all. An ether also reacts slowly with the oxygen in the air. That happens inside an ordinary closed bottle, with nobody doing anything to it. Light and warmth both speed it up. What it leaves behind is a peroxide. Where on the molecule the air attacks is not something these two sources say. This page will not invent the part it cannot cite. Between them they do pin down which solvents, how long, and which step is dangerous. The dangerous step is concentration. Boil the ether off and whatever peroxide has formed is left behind in less and less liquid. That is why a flask is never taken to dryness. Section 6 comes back to that oxygen for a different reason.
The hazard box above is not decoration. It carries numbers, and numbers can be applied.
One bottle of diethyl ether was opened this morning, one fourteen months ago. Which needs testing before distillation?
So the ether is the control case. It behaves exactly as the keep-sentence says. There is no hydrogen on that oxygen. Nothing can be handed over. The big changes in the panel follow from that missing hydrogen. The small ones are the oxygen's own. Now put a hydrogen back on it. The second half of the sentence starts to matter.
— Take the hydrogen off the oxygen, and what changes tells you what it was doing.
Water and ethanol both have an O-H. Both can hand that hydrogen to something else. You already have a number for how readily a thing does that. It is pKa, and a lower pKa means a stronger acid. Ethanol looks like water with a small carbon tail bolted on. The question is whether that tail makes any difference.
Compare ethanol with water itself. Is ethanol the stronger acid, the weaker one, or about the same?
Predict, then drive the bench.
The table below is measured data, not a simulation. Every row is quoted for water at twenty-five degrees Celsius. Eight rows load by default, and they all come from one table. That table is a compilation. Five different compilers' work was gathered into it. It is not a single primary paper, and it does not say who measured what. A checkbox adds four more rows. Each of those four carries its own source on its own line.
| compound | pKa | source |
|---|
Water is one of the four extra rows, so tick the box to see it. Read the water row against the ethanol row. Water has the lower pKa, so water is the stronger acid. Ethanol is water with a carbon tail, and the tail makes it weaker. Here is why, and it is the cheapest proof on the page. Take the hydrogen off water and what is left is a very small ion. Water molecules pack in close around it on every side. That packing is the leftover charge spreading out into the water. Take the hydrogen off ethanol and the carbon tail is in the way. The charge has less room to spread into, so that O-H holds on harder. No ring is involved anywhere in this. Now read methanol against ethanol. On this table they are identical, to the digit. Do not read that as a tie that has been settled. A second source, fetched independently, gives ethanol a slightly higher value than methanol. The two sources disagree at exactly this step. So this page reports both and ranks neither. Tick the box for tert-butanol too. In water it is the weaker acid and methanol is the stronger. That row comes from the unverified table, so take the direction and not the size.
One more thing about that table is worth testing.
Measure methanol and tert-butanol as bare gas, with no water at all. What happens to their order?
Read the direction first, because the direction is the finding. With no solvent present, tert-butanol lets its hydrogen go more easily than methanol. In water it is the other way round, and nothing about the alkyl groups changed between the two measurements. What changed is the water. Methoxide is what is left of methanol once the hydrogen has gone. A small ion of that kind is wrapped tightly by water, and the wrapping lowers its energy a lot. That wrapping is the charge spreading too, out into the water instead of into the molecule. A big, bulky ion cannot be wrapped as closely, so it gains less. In the gas phase there is no wrapping to gain, and bulk helps instead. Bulk spreads the leftover charge over more volume. So the order in water is a solvation effect, and this page does not dress it up as anything else. Treat the size of the gas-phase gap as about thirty kJ per mole and no tighter. The two ends come from different measurement families, one an average of six determinations and the other a single one. Their difference carries more uncertainty than either end does.
Every number in this section is an equilibrium position. None of them is a speed. They say where the balance sits, not how fast it gets there. The next section asks not which is stronger, but how much has actually gone.
— An O-H's pKa is set by what hangs off it and by the water around it.
A pKa is a single number. It can look like a verdict. It is not one. Put a lot of molecules into water. Some of them will have let their hydrogen go. The rest still have it. Which fraction you get depends on the pKa and on the pH. So the honest question is not whether a thing is an acid. It is how much of it has actually let go.
Ethanol sits in ordinary water at pH 7. Has a noticeable fraction of its O-H hydrogens gone?
Predict, then set the controls below and look.
The bench below is simulated, not measured. It computes one thing, live, from numbers this page has already sourced. Every pKa it uses is read off the table in the last section. No new measurement goes into it. The slider sets the pH of the water. The menu picks the compound. The curve shows the fraction that has let its hydrogen go, at every pH.
One formula draws that whole curve, and it is worth doing by hand. Subtract the pH from the compound's own pKa to get a difference. Raise ten to the power of that difference, then add one to it. The fraction that has let go is one divided by that total. Now put the pH exactly equal to the pKa and work it through. The difference is zero, ten raised to the power zero equals one, and the total underneath is therefore two. One divided by two is a half, so at pH equal to pKa exactly half the molecules have let go. That holds for every compound on the menu, whatever its own pKa. The same formula pins two limits down, and both are proofs rather than readings. Ten raised to any power is positive, so one plus that number always exceeds one. One divided by something greater than one is always less than one. The fraction can therefore never reach one hundred per cent. It can never drop below zero either, since one divided by a positive number stays positive. Those two bounds are proved rather than observed, so they hold right across the whole slider.
The bench answers how much at any pH you dial in. You can also run it the other way. Pick the fraction first, and ask which pH would give it.
Use that formula on ethanol. At roughly what pH does one molecule in a hundred first let go?
Nothing in that formula mentions structure. It takes a pKa in and gives a fraction out. All the chemistry is hiding inside the pKa itself. So the real question is what sets a pKa in the first place. The keep-sentence says it is how far the leftover charge can spread. The next section puts that to the test.
— The fraction that has let go is set by pKa minus pH, not by a name.
Phenol is an O-H bolted straight onto a benzene ring. Ethanol is an O-H on a plain carbon chain. On the table in section three, phenol's pKa is far lower than ethanol's. It is far more acidic, by a very large margin. Phenol is still a much weaker acid than the acid in vinegar. That comparison belongs to the next module. This page will not finish it here. Acidity is a matter of degree. It is not a label a molecule either has or lacks. The section tests that with a nitro group, which is one nitrogen carrying two oxygens.
Put a nitro group opposite phenol's O-H, then one carbon nearer instead. Which drops the pKa more?
Predict, then read the arithmetic below.
Think of the leftover charge as a shout. Penned on one oxygen, it is a shout inside a phone booth. It is loud and sharp and all in one place, which is an expensive place for a charge to sit. So that O-H holds on tightly, and its pKa is high. Where the charge can reach further, the same shout crosses an open field. No single spot carries much of it. It is a cheaper place for the charge to end up. So that O-H lets go far more easily, and its pKa is low. Now the break, because every image on this course has one. A phone booth and a field are fixed rooms, sized by the building. How far a real charge spreads is not a room, but is set by what is attached to the oxygen. It is also set by the water around the whole molecule. And the image names only the effect, which is how far the charge gets. It does not name the mechanism, which is still argued over.
Here is evidence that does not depend on any mechanism story. Take phenol and hang a nitro group on the ring. Put it at the para position, directly opposite the O-H. Then move the same group to the meta position, one carbon back toward the O-H. Both compounds are already on the table, at 7.15 and 8.36. Subtract each of those from phenol's own 9.99 and you have two drops. The panel below does that subtraction live. It then divides the larger drop by the smaller one.
Chemists do not agree on why the charge ends up more spread out on a phenol. The usual answer is resonance, meaning the charge is shared into the ring once the hydrogen leaves. In 1986 Siggel and Thomas challenged that. They measured how tightly the innermost electrons were held, before and after the hydrogen left. They found little change in how spread out the electrons were. Much of the spreading, they argued, was already there in the neutral molecule. Burk and Schleyer wrote a reply, and Rablen ran the comparison again in 2000. No review since has closed the argument, so here is what this page asserts and what it does not. It asserts that the leftover charge on a phenol spreads further than an alcohol's does. It does not assert how, and the meta and para numbers above are evidence neither side disputes. That is why they carry this section, and the picture does not. A textbook that hands you the answer here is handing you one side of a live argument.
That was two numbers, one group and two positions. Now use them on something the panel did not compute.
3-nitrophenol and 4-nitrophenol sit in the same water at pH 9. Which one is further along?
It is the same group and the same ring, in two positions, with two different answers. That is the part no mechanism argument touches. Where the group sits changes how far the leftover charge gets. How far it gets changes the pKa. The keep-sentence has now been tested on an alcohol, on a ring and on an ether. One thing is still missing from it.
— Where the charge can reach is the question, and how it reaches is still argued over.
An oxygen atom has more on it than a hydrogen. It also carries two pairs of electrons that no bond is using. Those pairs are called lone pairs. Now take two ethanol molecules, add acid, and warm them. They join up and lose a molecule of water between them. What is left is an ether, with the oxygen bridging two carbons. Something on the attacking molecule had to do the joining. The question is what.
Two ethanol molecules join into an ether. Which part of the attacking molecule does the joining?
Predict, then read the explanation below.
Here is what happens when two ethanol molecules make an ether. Acid and heat are both needed, and the temperature is kept moderate. One molecule's O-H group turns into a group that can leave, taking the oxygen with it. It leaves as a molecule of water. Exactly how the acid manages that is the question at the end of this section. What it leaves behind is a carbon short of electrons. The second molecule attacks that carbon. It attacks with a lone pair from its own oxygen, not with its hydrogen. The lone pair makes a new bond from oxygen to carbon. The result is one oxygen with a carbon on each side, which is an ether.
The keep-sentence at the top of this page is about one hydrogen. It says when an oxygen gives that hydrogen up, and how readily. It says nothing at all about what you just read. The lone pairs are not covered by it, and they were never meant to be. That is a stated limit, not a gap somebody forgot. Letting a hydrogen go is one half of what this oxygen does. Attacking with a lone pair is the other half. The next module is built on that second half. It also takes the one thing this page has kept away from. An alcohol can be oxidised, and the next module is where the oxidising goes.
One loose end is worth pulling. The acid in that recipe has not been accounted for.
Add strong acid to an alcohol, with no second alcohol nearby. What happens at that oxygen?
So the page closes where it started. An oxygen with no hydrogen on it gives nothing up, whatever else is around. An oxygen with a hydrogen gives it up as readily as the leftover charge can spread. Penned on one oxygen with nowhere to go, the charge is expensive and the hydrogen stays. Give it room and it is cheaper, and the hydrogen goes. That room can be a ring, or it can be the water packed round a small ion. Both of those are spreading, which is why the sentence says spread and nothing narrower. That is the sentence, and it holds on every case this page has shown. What it does not reach is the rest of the oxygen. That is the next module's first job.
— The keep-sentence covers a hydrogen leaving, and it claims nothing past that.
C2H4(g) + H2O(g) ⇌ C2H5OH(g) — Industry makes ethanol this way, over phosphoric acid on silica. Count the gas moles across the equation, two on the left and one on the right. Module 18's rule says squeezing that mixture should push it to the right. Industry does squeeze it, at {PLO} to {PHI} atmospheres. Now find the heat change, from three formation enthalpies NIST reports for these gases. Ethanol is minus 234 kJ per mole and ethene is plus 52.4 kJ per mole. Water vapour is minus 241.826 kJ per mole. Subtract both reactant values from the product value, keeping every sign as it stands. The answer comes out exothermic, so the same rule argues for running cold. Industry runs hot instead, between {TLO} and {THI} kelvin, because the catalyst only works fast enough at that heat. The equilibrium yield pays for it, and only about {PCT} per cent of the ethene reacts per pass. The rest is recovered by separating the ethanol and recycling the ethene, reaching about {OVERALL} per cent overall. No equilibrium constant is quoted here, because none could be found for these conditions. One last contrast is worth making. Module 21 was electrons moving from iron to oxygen. The acidity half of this module has been a proton moving instead. Type the heat change below, in kJ per mole to one decimal place, with its sign.
1.
2. An alcohol nobody has named turns out to have a pKa of 15.8. Cyclohexanol, which is on the table with its own source, sits at 16.0. Water's row is behind the checkbox on that table, and phenol's row is not. Use only this module's own rule, with no new source and no new bench. Say whether the unnamed alcohol is a stronger or weaker acid than water. Then say whether it is stronger or weaker than phenol.
Put 15.8 on the scale beside the rows you already have. Water is at 14.0, and 15.8 is the higher number of the two. A higher pKa means a weaker acid, so this alcohol is the weaker acid of the two. Methanol and ethanol are at 15.5 and cyclohexanol is at 16.0. So 15.8 lands in among the ordinary alcohols, between those two figures. Phenol is at 9.99, which is nearly six pKa units lower again. So the unnamed alcohol is a far weaker acid than phenol. There is no ring on it for the leftover charge to reach into.
Oxygen only gives up a hydrogen it has, and only as easily as the leftover charge can spread.