LearnChem · Part V — Why that product and not another?
A carbon next to an oxygen gets attacked for two reasons, and neither one alone decides it.
Why does vinegar smell so sharp, and why does an open bottle of wine turn into it?
That sharpness is a carboxylic acid. A carboxylic acid is a carbon holding a double-bonded oxygen and an O-H group. This page asks why that particular carbon behaves as it does.
Module 27 left an oxygen deciding whether to let its hydrogen go. Its sentence was this. Oxygen only gives up a hydrogen it has, and only as easily as the leftover charge can spread. That is module 27's sentence, quoted, and this module does not re-derive it. A carboxylate is what remains when a carboxylic acid loses that hydrogen. The leftover charge then spreads over two oxygens at once. That is the furthest it spreads anywhere in this course. This module asks about the carbon beside the oxygen instead.
Leave a bottle of wine open and it turns to vinegar. Cork it properly and it does not. The difference is oxygen. Acetic acid bacteria need oxygen for nearly everything they do. Starved of it they slow down and die. Given oxygen, they put the wine's ethanol through two enzymes. Alcohol dehydrogenase turns ethanol into acetaldehyde. Aldehyde dehydrogenase turns acetaldehyde into acetic acid. Acetaldehyde is a real, named intermediate here. It is not a tidy invention. So one bottle holds three different carbons over its life. A nucleophile is an electron-rich attacker. It goes for a carbon that is short of electrons. The gate below asks which carbon it would find easier.
In that bottle, is acetaldehyde's carbon attacked more easily than acetic acid's, or less easily?
Answer from your gut. The reveal follows.
Vinegar is dilute ethanoic acid and it is a food. Glacial ethanoic acid is the concentrated liquid. It is not a food. It burns skin and eyes. Its vapour irritates the airways. The substance is the same in both. The concentration is the hazard. No source found here gives a concentration at which one becomes the other. So this page states none.
Source: NOAA CAMEO Chemicals, Acetic Acid (Glacial), https://cameochemicals.noaa.gov/chemical/2272
The aldehyde is the easier target. A nucleophile attacks acetaldehyde's carbon more readily than acetic acid's. That ordering is what this whole module is built on. The next five sections earn it. Nothing here has proved it yet. What the bottle does show is that the aldehyde is real. A review of acetic acid bacteria names both enzymes and both products. Alcohol dehydrogenase acts first and aldehyde dehydrogenase acts second. The panel below reads out how strong such a vinegar can get. It also shows one safety datasheet disagreeing with itself about acetaldehyde.
Picture the carbon in the middle of all this. On one side sits an oxygen that will take the charge. On the other side sits a crowd that is in the way. Attacked, on this page, always means attacked by a nucleophile. That is the electron-rich attacker from module 25.
A carbon is attacked when its neighbour will take the electrons, and nothing is in the way.
Before that ordering can be earned, a simpler question is owed. Why does a nucleophile go near a carbon next to an oxygen at all?
— Ethanol, acetaldehyde and acetic acid are three different carbons, not one story told three times.
Oxygen is greedier for electrons than carbon is. In a carbon-oxygen double bond, the shared electrons sit closer to the oxygen. The carbon is left short of electron density. Short of electrons is exactly what a nucleophile looks for. That is module 25's machinery, unchanged, pointed at one group. The group has a name of its own. A carbonyl is a carbon joined to an oxygen by a double bond. So a carbon is attacked here in module 25's own sense. A nucleophile brings a pair of electrons and offers that pair to the carbon. What nobody has checked yet is whether the bond's own strength helps or hinders.
A carbon-oxygen double bond is stronger than a carbon-oxygen single bond. Does that make its carbon hard to attack?
Predict, then read the arithmetic below.
Easy, and the bond's own strength is beside the point. The carbonyl reacts even though it is the strongest bond on this page. It does not react because it is weak. The readout below puts three bond strengths side by side.
Read the numbers above in order. The carbon-oxygen double bond costs more energy to break than the carbon-carbon double bond does. It also costs more than twice what the carbon-oxygen single bond costs. None of that stops the carbon being attacked. What decides the attack is where the shared electrons sit. Two further measurements above say the same thing in other ways. A dipole moment measures how unevenly charge sits across a whole molecule. Three of them are printed above, one for each of three carbonyl compounds. Compare those three against each other before the check below. The bond lengths above come at it from a third direction. Both double bonds are much shorter than methanol's single carbon-oxygen bond. One comparison shows why the neighbour has to be a willing one. A carbon-carbon double bond has a carbon that is easy to reach and nothing crowding it. A nucleophile still leaves an alkene alone. The neighbour there is another carbon, and a carbon will not sit holding a negative charge. Drop that first condition and the sentence would predict an attack that never happens.
The rest of this module asks what makes one carbonyl carbon poorer, or harder to reach, than another.
Look at the three dipole moments above. Which way does adding carbon groups push a carbonyl's polarity?
Every carbonyl carbon is short of electrons. That does not make them all equally easy to attack. The next section measures how far apart two of them can be.
— Everything in this section is a measurement and not a story: lengths, angles, dipoles, bond energies.
Water can add across a carbonyl double bond. The carbon then holds two O-H groups instead of one double-bonded oxygen. That product is called a hydrate. The reaction is an equilibrium. It runs both ways and settles somewhere. Where it settles has been measured. Bench 1 below carries five of those measurements. They come from one set of course notes. Two of the five rows are the ones to compare first. A substituent is any group hung on a carbon in place of a hydrogen. Pivaldehyde carries one substituent on its carbonyl carbon. That one substituent is bulky. Propanone, better known as acetone, carries two small ones. One compound has fewer things in the way, and the other has smaller things in the way.
Is pivaldehyde more hydrated at equilibrium than acetone, or less hydrated?
Predict, then drive the bench.
Methanal is the first row on bench 1. It is better known as formaldehyde. The US Environmental Protection Agency classifies it as carcinogenic to humans by the inhalation route. That is the EPA's own wording, and this page uses no stronger one. Read its row as data. Nothing on this page is a procedure to carry out.
Source: US EPA IRIS Summary for Formaldehyde (CASRN 50-00-0), https://iris.epa.gov/static/pdfs/0419_summary.pdf
Bench 1 below holds five rows, exactly as the source prints them. Two honesty flags come with them. The source states no temperature for any of the five rows. So the temperature column says that, rather than guessing at one. The source also never labels the convention behind its own constants. This bench recovers that convention instead of assuming it. Multiplying each constant by water's own molar concentration reproduces the source's own percentage column. The self-check below runs exactly that on acetone's row.
| compound | Keq (M-1) | % hydrated | temperature | source |
|---|
Now read pivaldehyde's row against acetone's row above. Pivaldehyde has one substituent and acetone has two. Pivaldehyde's one substituent is far bulkier than either of acetone's. If bulk alone decided this, pivaldehyde would sit below acetone. The table says otherwise. Its row sits between acetaldehyde's and acetone's. Counting what is in the way is one reading. Measuring how big it is is another. Call the first one the crowding term. The second reading is what each substituent does to the electrons. An alkyl group is a plain branch of carbon and hydrogen. An alkyl group pushes a little electron density toward the carbonyl. That feeds the oxygen. Call this second reading the electron term. Acetone has two groups doing the feeding and pivaldehyde has one. One thing on this bench looks like a contradiction. Section 2's dipole figures rose as carbon groups were added. This bench falls as carbon groups are added. Those two are not the same quantity. A dipole moment belongs to the whole molecule. How readily one carbon is attacked belongs to that carbon and its neighbour. The very groups that raise the molecule's dipole are the ones feeding the oxygen and crowding the carbon.
One row on that table was kept out of this comparison, and it is the odd one.
Hexafluoropropanone is the bulkiest carbonyl on bench 1, and the most hydrated. Which term is winning there?
Two terms, then, and they do not always pull the same way. A reader who has met only one of them will get part of this bench wrong. The next section brings in a compound whose reputation needs checking.
— The five rows of this bench are not clustered together; they sit decades apart.
Carbon dioxide has a carbon with an oxygen on each side. Both of them pull. By the reasoning of section 2, that carbon should be very short of electrons indeed. Carbon dioxide also has a reputation for doing nothing. It reacts with water to give carbonic acid, and it does so famously slowly. Living things do not put up with that. They carry an enzyme whose whole job is to speed this one reaction up. So the reputation is built on something real. The question is what exactly that something is, and the gate below asks it.
Carbon dioxide reacts with water very slowly. Given unlimited time, how much of it becomes carbonic acid?
Predict, then tick the box on bench 1, above.
Tick the box below to add carbon dioxide to bench 1's table. Its number comes from a different source and a different convention. It cannot simply be dropped in beside the other five. The readout beside the box does the conversion in the open.
Start with why that arithmetic was needed at all. The five rows on bench 1 leave water out of their constant, and their numbers are per molar. Carbon dioxide's own source folds water in, and its ratio carries no units. Setting those two figures side by side as printed would compare two different quantities. So the readout divides the apparent ratio by water's own molar concentration, and lands in bench 1's convention. Only after that division can the two be read against each other. Once they are, carbon dioxide's hydration constant and acetone's are of the same order of magnitude. This page does not say which of the two is larger. The gap left over is small. The two figures come from two different sources, and the five-row table states no temperature. That is not a margin this evidence can carry. So carbon dioxide sits alongside a ketone rather than far below one. What is genuinely slow about carbon dioxide is its rate, and the readout above prints that range. A rate is a different question from where the balance ends up. Module 16 and module 17 separated those two questions in this course. This is that same separation again. A living cell cannot wait at the uncatalysed rate, which is why it carries an enzyme.
The same separation applies to compounds this bench never measured.
A compound has carbon dioxide's equilibrium constant but reacts much faster. Does it reach a different percentage hydrated?
Both terms are now on the table. So is one distinction that belongs to neither of them. Section 5 puts five kinds of carbonyl in order, using the two terms only.
— Two numbers from two sources cannot be compared at all until they are in the same units.
An aldehyde and a ketone are not the only carbonyls. Put a second oxygen on the same carbon and you have a carboxylic acid. Put an oxygen carrying a carbon chain there and you have an ester. Put a nitrogen there and you have an amide. Each of those neighbours brings lone pairs of its own. A lone pair is a pair of electrons that is not in a bond. A lone pair next door can spread into the carbonyl, and that spreading is called resonance donation. This section asks what all that donation does to the carbon in the middle.
An ester carries a second oxygen beside its carbonyl carbon. Is it attacked more readily than a ketone's?
Predict, then read the arithmetic below.
Think of a letter slot with a box behind it. The carbon is the slot and the neighbouring atom is the box. A letter goes in only if the box has room and the slot is clear. Those two conditions have nothing to do with each other. An aldehyde is a clear slot on an empty box. A ketone has two groups leaning across the slot, and they feed the box as well. An ester or an amide has a second neighbour feeding that box whatever the slot looks like. The image breaks in three places, and all three matter. First, a posted letter stays posted, while hydration runs backwards as readily as forwards. Every number on this page is an equilibrium constant, not a one-way count. Second, a full box is full for everyone, but a strong enough nucleophile still reaches a crowded, well-fed carbon. Both terms are comparisons between cases, never labels. Third, the letter is delivered and gone, while the electron pair stays shared between the attacker and the carbon. The charge that ends up on the oxygen is the same pair that was in the double bond.
Here is the order those two terms give. An aldehyde is attacked most readily, then a ketone, then an ester, then an amide. In the questions that follow, above means attacked more readily and below means attacked less readily. A carboxylic acid belongs on that order too. Soderberg's chain names esters and protonated carboxylic acids together. The acid's own O-H oxygen feeds the carbonyl much as an ester's oxygen does. So the acid's carbon sits with the ester's, well below an aldehyde's. That is the ordering section 1 opened with, now earned. This order is not a quotation from any one source. It is a join of two sourced pieces, and the join is worth seeing. The aldehyde-over-ketone half comes from bench 1's own hydration rows, and the readout below prints that ratio live. The rest comes from Soderberg's own stated chain, which names resonance donation from the second neighbour as the cause. That chain runs one species further than this section does. Where it ends up is the next section's question, and it is left alone here on purpose. No source on this page states the whole run in one sentence. There is also a trap sitting in the middle of this order. More oxygens does not automatically mean a weaker carbonyl. Infrared light makes a bond stretch, and a stiffer bond absorbs at a higher frequency. Saturated esters absorb higher than saturated ketones do, despite the ester's extra oxygen.
There is one thing the sentence you keep does not reach. It says nothing about which group leaves once an attack has happened. That property has its own name, leaving-group ability. The sources give it as a third cause behind this family's behaviour. It decides what happens after the attack, never whether the attack happens. Soderberg's argument for it runs through basicity. An alkoxide is an oxygen carrying a negative charge, and a thiolate is a sulfur carrying one. A thiolate is the weaker base of the two, so it departs more easily. A thioester is an ester with a sulfur in place of that oxygen. So a thioester earns its place on this order after the attack, not before it. The check below asks you to place it on that later ground, and the settled sentence will not get you there.
One member of this family has not been placed yet, and it carries a sulfur.
Leaving-group ability counts once the attack has happened. Does a thioester then sit above an ordinary ester?
Everything so far has been about a neutral molecule. Section 6 takes a hydrogen off one of them. That changes the neighbour itself, not merely how well it is fed.
— Two independent terms, read off in turn, and neither one is ranked above the other.
A carboxylic acid can do something no other carbonyl on this page can do. It can let its O-H hydrogen go. Module 27 explained why that happens, and this module does not redo that work. What module 27 never asked is what the loss does to the carbon next door. Take the hydrogen away and the neighbouring oxygen is no longer a neutral oxygen. It becomes one of a pair sharing a whole negative charge. The carbon itself has not changed at all. Its neighbour has. Whether that makes the carbon an easier target or a harder one is the question below.
Take the O-H hydrogen off a carboxylic acid. Is that carbon then more attackable, or less attackable?
Predict, then read the bond lengths below.
The carbon becomes less attackable, not more. Once the proton is gone, the carboxylate's carbon is the least attackable carbon in this module. Its neighbour is already holding a full negative charge, and it will take no more. Bench 2 below is a map of eight carbons, and it is not a line-up. Pick a family from the control. Before the bench shows you anything, type your own count of how many things sit in the way. Then check that count. Two separate readouts appear afterwards, drawn on two different scales. One counts what is in the way. The other says how well the neighbour is already fed. They are read together, and they are never merged into a single score.
How many things sit in the way at this carbon? Type a count from 0 to 3, then check it against the source.
An alkyl halide is a carbon chain carrying a halogen such as chlorine. It has no carbonyl at all, and it still belongs on this map. The halogen is the neighbour that will take the electrons, and it leaves with the bonding pair. A primary carbon has almost nothing in the way, while a tertiary one is surrounded. Bench 2 counts the primary case here, so the count it wants is one. A tertiary carbon would count three, and that gap is module 25's own ordering. One further difference is worth stating plainly. In that substitution the neighbour departs with the pair. At a carbonyl the neighbour keeps the pair and stays attached. It is the same sentence with two different endings.
The two readouts are not the same kind of evidence, and the bench says which is which. For an aldehyde, a ketone and carbon dioxide, the feeding readout is a measured number taken straight off bench 1. For an ester, an amide, a carboxylate, a carboxylic acid and an alkyl halide, no number was sourced at all. Those families read as a sourced ranking instead, and the bench labels them that way. A ranking is weaker evidence than a measurement, and it is not dressed up as one here. The crowding readout is a plain count in every case.
| acid | pKa | temperature |
|---|
Why the acid gives up that hydrogen is module 27's sentence, not a new one. Oxygen only gives up a hydrogen it has, and only as easily as the leftover charge can spread. On a carboxylate the leftover charge spreads over two oxygens at once. That is the furthest it spreads anywhere in this course. The readout above is the physical proof of it. In the neutral acid the two carbon-oxygen bonds are plainly different lengths. In the carboxylate they come out equal. The table above it points the same way from a second direction. Adding chlorines to acetic acid drops its pKa each time, which is a substituent effect on the same O-H hydrogen. The last row of that table sits at a different temperature from the other six. Its own caption says so.
There is a second thing the sentence does not reach. In acid, the first thing that happens is not an attack on the carbon. A proton lands on the carbonyl oxygen instead. The sentence says what makes a carbon attacked. It says nothing about what an acid does before any of that. Read honestly, this strengthens the argument rather than denting it. An oxygen already holding a proton is far readier to take a pair of electrons. So the same two terms explain why acid speeds these reactions up. That protonation is the first move of the esterification from module 18, and the exercises below come back to it.
Two members of this family have never been set directly against each other.
Does a carboxylate sit above an amide, or below it?
Two terms have carried every case on this page. Will the neighbour take the electrons, and is anything in the way? An aldehyde, a ketone, an acid, a carboxylate, carbon dioxide and an alkyl halide all answered to those same two questions. One carbon from the opening is still unplaced. Ethanol's carbon has an oxygen next to it as well. That oxygen is joined by a single bond, so there is no second pair for it to take. What an alcohol does with its own hydrogen was module 27's question. In the bottle, enzymes change that carbon's neighbour, and then change it again. What differs between the three carbons is what each neighbour is, and what sits around it. None of them holds a position on a scale. The exercises below apply the two terms to carbons this page never measured.
— Take the hydrogen off, and the carbon's own neighbour is not the same neighbour any more.
CH3COOH (l) + C2H5OH (l) ⇌ CH3COOC2H5 (l) + H2O (l) — Module 25 set a nucleophile against a primary alkyl halide and against a tertiary one, and the primary carbon won. Explain that result using only this module's two terms. Then read the equilibrium printed above. Its equilibrium constant is {KC}. The source that gives it states no temperature, so none is quoted here. From an equimolar start, {PCT} per cent of the acid is converted by the time the reaction settles. The equilibrium is pushed toward the ester by taking water out. Distillation, a large excess of one reagent, and drying salts all do that. Sulfuric acid is the catalyst here and nothing more. It protonates the carbonyl oxygen first. No source found here supports calling it a drying agent.
1.
2. In your own words, what is different at a carbon-oxygen double bond? The worked answer is behind the button below.
In module 25's substitution the neighbour leaves with the pair of electrons. At a carbon-oxygen double bond the neighbour keeps that pair and stays attached to the carbon. The sentence you keep is the same in both cases. Only the ending differs. That is why one reaction replaces a group and the other adds one.
3. Di-tert-butyl ketone carries two tert-butyl groups on its carbonyl carbon. A tert-butyl group is a bulky branch of four carbons. They are ordinary alkyl groups, like acetone's, only far larger. Bench 1 above has pivaldehyde at 19 per cent hydrated and hexafluoropropanone at 99.9996 per cent. Using the two terms only, where does this ketone fall against acetone?
Less hydrated than acetone, and by some way. The crowding term gets much worse, because two tert-butyl groups are far larger than two methyls. The electron term barely moves, because a tert-butyl group feeds the oxygen much as a methyl does. One term worsens while the other stays put, so the result drops. This is also the case that explains hexafluoropropanone. Its result comes from its fluorines pulling charge away from the oxygen's side. It is not simply the bulkiest ketone on the bench.
4. Soderberg states that nitrogen is a powerful electron-donating group by resonance, more so than oxygen. An amide's second neighbour is a nitrogen. An ester's second neighbour is an oxygen. Work out what the stronger donation does to the carbon, then predict where an amide sits against an ester.
A carbon is attacked when its neighbour will take the electrons, and nothing is in the way.