Reversible aromatic substitution reactions

If we take naphthalene then we can react it with electrophiles. For example if we treat it with nitric acid under mild conditions.

Now the nitration reaction does not appear possible to reverse, but the two nitronapthelenes are different in terms of thermodynamics. In the case of the 1-nitronaphthalene there is a steric repulsion between the hydrogen on the 8 position and the nitro group. In the case of the 2-nitronaphthalene there is no strong steric repulsion between the hydrogens on the 1 and 3 positions and the nitrogroup.

There is a reason why the less thermodynamically favored product will form, this is a kinetic reason. If we react naphthalene with an electrophile at the 1 position we go via a cationic intermediate. Two of the five lower energy resonance forms keep one of the two rings aromatic. This will lower their energy.

By now you should be learning that the more low energy resonance forms you can draw the lower the energy the ion has, also if you can draw resonance forms which are extra stable then the energy of the combination of the resonance forms will be lower. Lower energy means more stable. If we repeat the intellectual process for the reaction at the two position on naphthalene then we get a less stable cationic intermediate.

So what happens when naphthalene reacts is that the kinetic product (product which is formed more quickly) is higher in energy than the product whose formation is slower. If the formation of the product where the naphthalene has reacted at the one position (alpha isomer) can be reversed then if we stew the reaction mixture for a long time then we will get the beta isomer as our product.

While nitration is very irreversible the reactions of sulfuric acid and things like tert-butyl cations are reversible. Since 1870 (Merz, Chem. Ber., 1870, 3, 196) it has been known that naphthalene-1-sulfonic acid when heated at 130 oC in sulfuric acid will transform into naphthalene-2-sulfonic acid.

Another example is the reaction of tert-butyl chloride with benzene using aluminium chloride as the cataylst. If we do the reaction then we will have a Friedal-Crafts alkylation reaction. The first stage will be the reaction to form tert-butyl benzene.

Then a second molecule of tert-butyl chloride reacts to form 1,4-di-tert-butyl benzene.

Then things get a bit complex, one of the tert-butyl groups will come off the benzene ring and then we can react the tert-butyl cation at the meta site to form the 1,3-di tert-butyl benzene.

This can then react with more of the tert-butyl chloride to form the final product, I suspect that only one product will form. The reason is that the steric effects of the tert-butyl groups will oppose the loss of the protons from the cationic intermidates.

The steric energy of 1,2,3-tri-tert-butylbenzene is 75.3531 kcal per mole, 1,2,4–tri-tert-butylbenzene is 38.2984 kcal per mole and 1,3,5–tri-tert-butylbenzene is 14.3038 kcal per mole. On the other hand the steric energy for 1,3-di-tert-butylbenzene is 9.5103 kcal per mole. The steric energy of the tert-butyl cation is only 3.6054 kcal per mole. I think that a steric argument explains why the 1,3,5-product is a major product from the reaction of benzene with tert-butyl chloride.

Arenes Chemistry Organic chemistry

Nitration of toluene

OK we are onto the final part of the lesson on aromatics, we need to consider the synthesis of TNT. It is interesting to note that non military and peaceful uses of explosives result in greater consumption / use. There are some explosives which are more suitable for military use rather than for mining. Years ago I visited a slate mine in England (Honister Pass) which was reopened by the late Mark Weir.

He was new to mining, and he attempted to mine a block of slate from the ine. He used plenty of black powder (gun powder) and he managed to break off a big lump of slate. The only problem was by using too much explosives he managed to reduce the slate block to small lumps which are close to worthless. At the mine he and his staff soon learnt that an Italian rock saw (nicknamed “the Italian job“) was a better method of freeing big blocks of slate.

I reason that TNT would be a dire explosive for slate mining, it is a rather fast explosive (detonation velocity 6900 ms-1) which has a high shattering ability (brisance). While TNT is not quite as shattering as RDX (8650 ms-1) would be it is far too shattering for slate mine use. For mining I think things like ANFO (4200 ms-1) or black powder would be more suitable when used in moderation. If I was a running a slate mine I would go for the black powder as the ANFO requires a rather strong boost charge to trigger it. If only a small amount of ANFO was to be used to loosen some rock then the boost charge will contribute a lot to the bang.

A word of advice do not play with explosives or experiment at home with them, the sentences for violations can include death or a serious injury resulting in permanent disability. These two harsh penalties can be dispensed by the energetic material, there is no appeal process. But after the energetic material has imposed its penalty then the criminal justice system may choose to impose a penalty.

This is no joke, some years ago some friends of mine at another university (University of Bleep) had a horrible event. A foolish student stole some chemicals from a teaching lab and illegally made a batch of an explosive. The student smuggled it out of the teaching lab and took it home. The student was later killed in an accidental explosion which destroyed the kitchen of the flat they were living in. Please do not bother asking me for more details as I will not answer that question.

The industrial synthesis of TNT is based on the reaction of a mixture of nitric and sulfuric acids with toluene. The reaction will occur in three stages, the first stage is the reaction of toluene with the nitration mixture to form the nitrotoluenes. From what we know already we should be able to understand how and why the main products of this reaction are 2-nitro-1-methyl benzene and 4-nitro-1-methyl benzene.

If we now consider the reaction of the 2-nitro-1-methyl benzene with the nitration mix. Before we start there are a couple of things to consider, there is something which is on my dirty washing list. Something which needs to slither back the swamp and crawl under a rock. Something which needs to never show its face again. I will let you look at it for a moment.

Bad bad bad, never let me see you drawing such a nitro group !

Now consider what is wrong, what is wrong is that we have a nitrogen to which we have five bonds. Now while things like sulfur and phosphorus (and more heavy p block atoms) can expand above and over the octet. Nitrogen can not do so, structure has a nitrogen which has five bonding pairs around it. This will not work well ! I know that the symmetry of this wrong nitro group might appeal to some people but it is still wrong.

Now we have seen it we should make it go away again, we should replace it with something far better. If we consider only the nitro group then this group of three resonance forms explains things better.

If we consider more of the resonance forms of nitrobenzene we can see that partial positive charges appear because of resonance on the carbons ortho and para to the nitro group.

Now lets look at the ortho-nitrotoluene. I calculated the partial charges on the carbons in the molecule, it is clear that those carbons which are ortho and para to the nitro group have partial positive charges while the carbons at the meta positions have slight partial negative charges.

This result suggests to me that the following resonance form makes a large contribution to the overall reality.

Now we should consider the reaction of the electrophile from the nitration mixture with 2-nitrotoluene. If we get it to react at the three position then we will go via the cation with the following resonance forms. The one on the left will have a very high energy, the nitro group will also be trying to remove electron density from the same carbons as were the positive charges in these three resonance forms are.

The formation of the other isomers of dinitrotoluene from 2-nitrotoluene will occur via the cations with the following resonance forms.

The last thing to do is to consider the order of reactivity. The reaction at the sites ortho and para to the nitro group will be slowed down by two things. These two things are the inductive pull of the nitro group on the electron density in the benzene ring and also the resonance withdrawal of electron density from both the benzene ring and the cationic intermediate by the nitro group.

The reaction at the positions meta to the nitro group will only be slowed by the inductive effect, so as the reaction. These positions are ortho and para to the methyl group. As a result the product of the second stage of the nitration of toluene will be a mixture of 2,4-dinitrotoluene and 2,6-dinitrotoluene.

Arenes Banned chemicals benzene Chemistry Insecticides Organic chemistry Uncategorized

Phenol and chlorobenzene with electrophiles

Due to the very electron rich nature of phenol it is possible to nitrate phenol using gentle conditions. One method is to use a solution of nitric acid (6 % w/w) with a solution of the phenol in ethylene dichloride with some tributyl ammonium bromide. Now I do not like the sound of this system, I would suggest instead using aliquat 336 in aromatic kerosene with dilute nitric acid. My dislike of the method presented in Organic Process Research & Development 2003, 7, 95−97 by Ashutosh V. Joshi, Mubeen Baidoosi, Sudip Mukhopadhyay, and Yoel Sasson is that the solvent they choose to use is carcinogenic. I would advocate using a solution of the phenol in aromatic kerosene (How about Solvesso 150ND) do the reaction with the nitric acid and then extract the nitrated phenols with sodium hydroxide before making the sodium hydroxide solution acidic to allow the nitrophenols to be recovered.

You should be able to understand the mechanism of the nitration of the phenol using what you know already about the reactions of the 2-tert-butylanisole and toluene. Now we have to move onto something else.

What happens if we have an arene bearing a group which withdraws electron density by the inductive pull effect but donates by the resonance effect ? This might sound like a rather strange group but we already have some which do this.

We have anisole which has a slight electron withdrawing effect by means of its inductive pull and a strong electron donation effect by means of resonance. But a more clear case is chlorobenzene. The chlorine atom has a strong inductive pull and it donates by resonance. Here is a diagram showing the resonance and inductive effects.

Now many years ago there was a wonder chemical, a non toxic insecticide which is about as toxic to humans (acute effect) as aspirin. It was regarded as a safe alternative to things like lead, arsenic, thallium and nicotine for use on food crops. It offered humans a means of expunging insects from farmland and the extermination of disease carrying insects such as body lice and malaria mosquitos. Sounds great but we later found that there were some horrible problems.

  1. Many of the insect pests became immune to its effects
  2. The agent killed off lots of useful insects such as honey bees
  3. The agent breaks down into a long lived substance which is toxic to some mammals (such as bats)
  4. The agent breaks down to something which is a xenoestrogen, the agent itself might also be active as a xenoestrogen.

In the USA there have been problems at Lake Apopka (Florida) as a result of xenoestrogens entering the lake, these have harmed birds and alligators (Cynthia V. Rider et. al. Environ Toxicol Chem. 2010 September ; 29(9): 2064–2071. doi:10.1002/etc.233).

The compound in question is DDT, DDT was made by the electrophilic aromatic substitution reaction of chlorobenzene with choral using an acid catalyst. In the first stage a molecule of chlorobenzene reacts with choral (trichloroacetaldehyde) to form the first of the new C-C bonds.

Then a second molecule of chlorobenzene will react with the 2,2,2-trichloro-1-(4-chlorophenyl)ethan-1-ol. This will protonate to form a resonance stabilized cation.

We finish off the DDT synthesis with a second reaction of a chlorobenzene molecule on the new cationic intermediate.

Now we have formed a molecule of DDT, here is the crystal structure of DDT in case you are interested to know what it looks like.

It is important to keep in mind that the synthesis of DDT is not able to make a perfectly pure product, commercial DDT used to contain some of the ortho,para and the ortho,ortho isomers. Here is the molecular structure of the ortho, para isomer of DDT.

Now we have been going through the synthesis of a molecule which has been banned. It might never have been banned if some sectors of society had used DDT in a more responsible manner. If we ignore the harmful effects on wildlife (the man reaoson for the ban) and concentrate on the problem of DDT resistance in insects then it is likely that if DDT had never been used for agriculture but had been only used for controlling insects which transmit disease to humans (eg Malaria control) then DDT might still be in use. I reason that the widespread outdoor use of DDT created a legion of DDT resistant insects while causing the various problems.

In case you think that the chemistry required for the formation of DDT from chloral and chlorobenzene has gone away, it has not. The synthesis of bisphenol A from acetone and phenol is using close to identical chemistry. Bisphenol A is made by the reaction of phenol and acetone using an acid catalyst

When you are ready for the final part of the lesson on aromatic substitutions please click here.

Chemistry drugs Insecticides Organic chemistry Uncategorized


There are a series of about four main natural compounds found in humans and other animals which are the estrogens. These are female sex hormones which are steroids. These are E1 (Estrone), E2 (Estradiol), E3 (Estriol) and E4 (Estetrol).

Xenoestrogen means an artificial estrogen, some artificial estrogens such as EE2 which is commonly used in oral contraceptives is not particularly bad. The xenoestrogens all are molecules which can fit the estrogen receptor, some are steroidal like the natural estrogens while others are not steroids. Here is a picture of EE2. The reason why EE2 is used instead of estrogen in oral contraceptives is that many of the natural sex hormones are not very active when swallowed. They tend to be converted into other things by the body before they can get into the main part of the blood system. Here is a picture of EE2.

I have seen many people making the argument that urine from women who are taking “the pill” will turn male fish into female fish. But when I look at the evidence it is interesting that the pregnant woman releases estrogens which have a far higher estrogenic effect on the environment than a woman who is “on the pill”. Here is a bar chart showing the amount of the three most important estrogens released per person per day.

You can see here now a woman who is menstruating does emit more estrogens in their urine than a man. A woman who has gone through the menopause emits less estrogen than before she stopped, but a little more than a man. When pregnant women emit a lot of estrogens in their urine. The three compounds I have data for are not equal in strength as estrogens. The relative strength of the estrogens will depend on the receptor which they are binding to. But in the paper I was reading you can give a score of 2.8 micrograms worth of E2 to a man, the menopausal woman has a score of 3.5 micrograms worth of E2, the menstruating woman has a score of 6.0 micrograms of E2 and the pregnant woman 589 micrograms worth of E2 per day.

The typical dose in the oral contraceptive is about 30 to 35 micrograms per day, as women will fully metabolize between 20 and 50 % of the EE2 which they swallow. Then the dose to the environment of EE2 from a woman on the pill is between 15 and 28 micrograms for the days that they are taking it. As EE2 is twice as potent an estrogen as E2 (the standard one), I can reason that the woman on oral contraceptives is somewhere in the range of 20 to 40 micrograms worth of E2 per day.

I have seen data that in 2001 that indicated that 43 % of Dutch women of reproductive age took oral contraceptives while in the USA it was 28 % of women of reproductive age. So if we use these values it can be estimated that on average women of reproductive age are releasing between 12 and 24 microgram worth of E2 per day.

Compared with the 6 micrograms worth of E2 per day this is not a vast increase. What is interesting is the question of how stable is EE2 compared with the other estrogens in a sewage plant. The paper I was reading indicates that many sewage plants are able to remove more than 80 % of both E2 and EE2. Thus the sewage plant will mitigate any effect of EE2 use on the fish in the river.

The truly scary worst of the worst xenoestrogen is DES, the jury is out regarding the question of “does it cause breast cancer when an adult woman is exposed to it ?”.


But it is very clear that in utereo exposure to DES does increase a woman’s chance of getting an unusual gynological cancer at a young age (teens or 20s). The cancer is such that DES is close to a perfect storm, the cancer is one which is not normally seen in women before the menopause. Also a normal gynological examination is likely to fail to spot signs of this cancer. A word of warning if you are easily shocked or horrified then do not go googling “DES daughter cancer”, the topic is a nasty horror show.

Other xenoestrogens include bisphenol A, in rats this is a very weak estrogen. But there are some related compounds which are much stronger as estrogens in rats.

Another one which is interesting is DDE which is a breakdown product of DDT. This is an interesting xenoestrogen as it does not have a hydroxyl group. Here is a picture of DDE.

Thankfully in Sweden the birds of prey whose numbers declined due to the use of some chlorine containing pesticides have started to increase again.

Another xenoestrogen of note relates to the surfactant triton X-100, this surfactant is not an estrogen but it breaks down in nature into a surfactant. The reason it is a problem is that the branched alkyl chain is only very slowly digested by bacteria. But when the corresponding compound with a C9 linear chain (Nonoxynol9) is released into rivers it is degraded much faster.

I have drawn all the xenoestrogens to try to suggest a similar structure to estrogens like E1. I may come back and write some more later on this topic.


Reaction of bromine / toluene and methyl benzoate and the nitration mixture

When we react bromine and toluene using the iron(III) chloride we have to have a step to form the true electrophile. We need a catalyst to make the reaction go. Here is a short film in which I had a solution of bromine in toluene. I added some steel staples and then I lifted them out with a magnet.

What you should notice is that when the steel was in the mixture the mixture generated bubbles of hydrogen bromide gas, when the steel was taken out the gas production stops. The surface of the steel reacts with a little of the bromine to form some ferric bromide. This ferric bromide is the catalyst for the reaction.

The first step is normally considered to be the reaction of the bromine molecule with the Lewis acid. Commonly this is thought (and taught) to be the reaction of bromine and the Lewis acid to form a pair of separated ions.

This will then be followed by the reaction with the aromatic ring. Now when the reaction occurs we have a choice of reacting in three different locations. We can react at the ortho, meta and para sites.

The rate of reaction at these three sites will not be equal. If we consider for a moment what the mechanism for each of the three possible reactions will be. If we use what we have found out from the nitration of benzene and the reaction of the anisole with the P4S10 then we should be able to make a set of three mechanisms like these.

We have to decide which ones of these will be faster, now to get to the cationic intermediate we have to pass through a transition state. Here is a graph of energy against the progress of the reaction for one molecule which I drew today.

Now as the transition state is similar to the cationic intermediate in energy we can reason that the lower the energy of the intermediate the lower the energy of the transition state will be. We have to overcome the energy barrier of TS1. Here is the diagram with the activation energy barrier marked on it.

Now if we have a second path to a different product, shown in red. Then I hope that you can understand that if the reaction is operating under kinetic control then we will not form the red product. The reason is that as the activation barrier of TS1(red) is larger the rate at which the intermediate for the red reaction is entered is smaller than the rate at which the black one is entered. Thus more of the black will be formed than the red one.

Now the key thing for us to understand is that the energy of the cationic intermediate is not dictated by only one resonance form. All the resonance forms have an effect upon the energy of the cation. If we imagine for a moment that all the resonance forms were stable cations then we can imagine how those which are formally drawn as secondary carbocations have higher energies than those which are formally drawn as tertiary carbocations.

But it is the combination of the resonance forms which determines the energy of the cation. If we ignore crazy high energy resonance forms like this one

and limit ourselves to the three lower energy forms which are commonly drawn then we can make some easy progress. If we assume for arguments sake that the three lower energy forms all contribute equally to the overall energy. Then if form A has an energy of 100, form B 70 and form C 100 then the average of these will be 90. So when one of the forms is more stable then we can understand how we lower the energy of the cation.

If we make more of the resonance forms lower in energy then the overall energy of the cation will be even lower. If we consider for a moment the reaction of bromine and 1,3,5-trimethyl benzene (mesitylene). Then we can see how all three of the resonance forms are more stable than the resonance forms which would exist if benzene was reacted.

So as a result if I was to react mesitylene with bromine then it should react faster than toluene under the same conditions. What I am going to do when I get the chance is to react some bromine with a mixture of mesitylene, chlorobenzene, toluene, tert-butyl benzene and some other aromatic compounds. If I get a GCMS trace of the mixture before and after adding the bromine we should be able to see the relative reactivity of the different compounds.

Now there is something which is an important difference between the reaction of the bromine and the P4S10. This is the size of the electrophilic reagent. In the case of bromine it is small while for P4S10 it is large. For large electrophiles we tend to see less reaction at the ortho sites as the electrophile is less able to reach the site. I think that the steric effect on the ortho / para ratio is a rather advanced idea which at Chalmers is not normally something we would expect a first year to recall / understand. But it is still an interesting thing.

The final stage of the bromination reaction is for the cation to lose the proton, the proton then reacts with the anionic iron bromo complex to reform the Lewis acid and form a molecule of HBr.

In the same way as a group attached to the ring can increase the stability of the cation, there are groups which lower the stability of the cation. If we consider a group which is able to withdraw electron density by both the inductive pull effect and the resonance effect for a moment. One such group would be a carboxylic acid or one of its derivatives. If we consider methyl benzoate for a moment. I have drawn some resonance forms which make it clear how the carbonyl group can pull electron density towards itself by means of the resonance effect.

There are at least two ways of thinking about it, one way is that if you are trying to increase the stability of the cationic intermediate by moving a positive charge onto a carbon which already is being used by another group in the same way. Then we will have contest for which group gets to put its positive charge onto the carbon. Here is an organic chemistry joke.

Q: What did the carbonyl group say to the Wheland intermediate (Arenium ion) ?

A: Clear off ! You are not putting your positive charge there ! I was here first !

Now we can consider it in a more serious way, if we take a more simple molecule (formaldehyde). we should understand that the oxygen is the most electronegative of the elements in the molecule. According to my quantum mechanical calculations we have charges on the different atoms.

Now if we repeat the calculation for acrylaldehyde (Acrolein, also known as propenal) we get a more electron rich oxygen. This indicates that the quantum mechanical calculation is indicating that the carbonyl group is withdrawing electron density from the alkene. This time I have put the partial charges on the non hydrogen atoms only.

Now we do this for fumaraldehyde, this is a trans alkene where there are two aldehyde groups which are playing “tug of war” with the electron density in the pi system.

What we should look at is the fact that when we have two aldehydes playing “tug o’ war” with the electron density that they get less electron density from the alkene. We can apply the same ideas to the cationic intermediate.

As the carbonyl group is trying to withdraw electron density from both the benzene ring and the cationic intermediate it will lower the amount of electron density in the benzene ring and it will also lower the stability of the cationic intermediate.

Now in organic chemistry a lot of things are dictated by kinetics, while a thermodynamic driving force might favor the formation of all three isomers of methyl nitrobenzoate the rates of formation of these things will be different.

There are some cases where the electrophilic substitution of aromatic things are reversible. But generally we run most of these reactions under conditions where the reverse reaction is very slow. As a result they do not reach equilibrium.

In our case we can regard the nitration to be irreversible, so which ever isomer is formed quickly will be the isomer which will be the major product. The ester group has two effects there is the inductive pull on the electron density in the ring. This occurs through the sigma bonds. There is also the resonance effect through which the ester group withdraws electron density.

While the reaction at all three sites is slowed down by the inductive removal of electron density by the ester group. The resonance effect only has an effect on the reaction at the ortho and para positions. So as a result the main product of the reaction is the meta isomer. The reaction at the meta site will be slower than the reaction of benzene, so if we were to make a 1:1 mixture of methyl benzoate and benzene and combine it with one equivalent of an electrophile then the majority of the electrophile would be consumed by the reaction with the benzene. You should recall how the young Mark Foreman reacted a mixture of xylene and ferrocene with an electrophile, when he did this the ferrocene reacted rather than the xylene.

When you are ready to consider the next aromatic reaction please click here, this will be phenol with nitric acid.