Sciencemadness Discussion Board

Toluene Nitration Using Nitrate Salts

AvBaeyer - 22-8-2015 at 18:21


Introduction

Nitration is one of the most useful and widely used reactions in organic synthesis. In the nineteenth century, nitration employing potassium or sodium nitrate in sulfuric acid was common but this was superseded by the nitric acid/sulfuric acid methods when nitric acid became readily available. However, the use of nitrate salts as nitrating agents under various conditions has continued to attract interest because of potentially milder reaction conditions and results that may differ from the nitric acid/sulfuric acid methods.

My interest in the use of nitrate salts came about because I needed a small amount of o-nitrotoluene and I did not have any nitric acid at hand. The lack of nitric acid was unfortunate since Magpie has posted a detailed procedure for the nitration of toluene and isomer separation in the Prepublication section. A quick web-search turned up several available references describing the use of nitrate salts which looked promising for my needs. Mellor et al. [1] reported the use of metal nitrates/sulfuric acid in dichloromethane for the nitration of toluene. Either ceric ammonium nitrate or potassium nitrate gave comparable results. The reported yield of nitrated products was quite good with an ortho/para nitrotoluene ratio about the same as the nitric acid/sulfuric acid method. Mawardi [2] published a study on the nitration of alkylbenzenes, including toluene, using cupric nitrate trihydrate in acetic anhydride. This work was of particular interest since for toluene the ortho isomer was the major product. Another paper by Bell [3] reported the cupric nitrate trihydrate in acetic anhydride method for 2-phenylcyclohexanone. Bell stated that the reaction as he ran it had an induction period leading to an exotherm. This was of particular concern since acetyl nitrate is formed in this reaction which in turn is known for its potential instability [4]. Having this information in hand, I chose first to employ the method described by Mellor but employing sodium nitrate in place of potassium nitrate. For reasons which will become clear from the results of these initial experiments, the cupric nitrate/ acetic anhydride method was also explored in some detail.

Regarding the exotherm reported by Bell [3], I believed that this was due in part to the reaction of acetic anhydride with the water of hydration of the cupric nitrate trihydate. This was born out by experiment.

Experimental

General.

All chemicals were obtained from reputable sources. Toluene and dichloromethane were distilled and center cuts were used. Reactions were followed by thin layer chromatography (tlc) on silica gel plates and visualized by uv light at 254 nm. All reactions were run in a 50 ml 3-neck flask equipped with a thermometer, a 25 ml dropping funnel, and a stopper in the third neck. The reactions were all stirred magnetically. After final work up, reaction products were subjected to vacuum (about 2 mm) for final solvent removal.

TLC method. Stirring was briefly stopped, any solids allowed to settle and an approximately 2 microliter sample of the supernatant was withdrawn. The plate was spotted and developed in hexanes using a double development. Under these conditions, o-nitrotoluene had Rf of about 0.60 and p-nitrotoluene had Rf of about 0.45. This is consistent with reported values in a similar system [5].

Reaction of toluene with sodium nitrate and sulfuric acid in dichloromethane. Powdered sodium nitrate (1.21 gm, 14.2 mmole) and toluene (1.84 gm, 20 mmole) were taken up in dichloromethane (35 ml). Sulfuric acid (96%, 0.8 ml) was added dropwise using a calibrated pipet. A red-brown coloration was observed around the sodium nitrate which quickly dispersed. TLC after about 10 min showed both nitrotoluene isomers present. After 1.5 hr, tlc for nitro compounds was very intense. The reaction was quenched by the addition of water (15ml). The organic phase was washed with 50% saturated sodium bicarbonate solution which gave very yellow aqueous extracts the color of which was discharged by the addition of hydrochloric acid to pH 4. The organic solution was dried over sodium sulfate and the solvent removed by distillation then under vacuum. TLC showed only two nitrotoluene products with the ortho-isomer predominating. The yield was 0.42 gm (22%).

The above experiment was repeated exactly except the total reaction time was 20 hr. The yield of mixed nitrotoluenes was 0.91 gm (47%).

Reaction of cupric nitrate trihydrate with acetic anhydride. Stirring cupric nitrate trihydrate ( 0.5 gm) with acetic anhydride (3-4 ml) at room temperature results after about 5-6 minutes in a rapid exotherm with temperature rising to ca. 40C. The reaction mixture undergoes a very noticeable change from a light blue suspension to a very deep blue – black mixture.

Reaction of toluene with cupric nitrate trihydrate in acetic anhydride. Acetic anhydride (4 ml) was chilled to about 7 C in an ice bath. Cupric nitrate trihydrate (0.56 g, 4 mmole) was added in one portion and the reaction mixture stirred for a few minutes. No exotherm was noted. The cooling bath was removed and the mixture allowed to warm slowly. At about 20 C a rapid exotherm occurred with a temperature rise to about 35 C at which point the cooling bath was reemployed and the temperature returned to 20 C. A solution of toluene ( 0.37 gm, 4 mmole) in acetic anhydride (4 ml) was added dropwise in about 1 minute. A strong exotherm resulted with the appearance of some brown fumes. After 1.5 hr TLC showed strong appearance of both nitro isomers. The reaction was filtered and the deep blue solid washed with a few ml ether. The filtrate and washings were stirred with water (20 ml) for about 30 min. The mixture was then extracted with ether (2x10 ml). The aqueous portion was made to pH 5-6 with solid sodium bicarbonate and extracted again with ether (2x10 ml). The combined ether extracts were washed with 50% saturated sodium bicarbonate solution giving a yellow aqueous layer as observed above. The ether was dried (sodium sulfate) and removed to give a yellow oil which showed only the two expected products (0.50 gm, 90%).

Reaction of toluene with cupric nitrate trihydrate and acetic anhydride in acetic acid. Cupric nitrate trihydrate (2.42 gm, 10 mmole) was added to acetic anhydride (7 ml, 73 mmole) chilled to about 7 C. As above, the cooling bath was removed and the stirred mixture allowed to slowly warm and the expected exotherm occurred when the temperature reached about 20 C. The dark reaction mixture was cooled to 15 C and a solution of toluene (3.2 ml, 30 mmole) in acetic acid (10 ml) was added dropwise over about 10 min. There was an initial slight temperature rise which was maintained until most of the toluene was added. Upon completion of the addition, the cooling bath was removed and the reaction stirred. TLC analysis after 20 min reaction time showed very strong product spots with the ortho isomer much more intense. After 1.5 hr the reaction was filtered and the solids washed with ether (3x10 ml). The filtrate was poured onto ice and 50% sodium hydroxide solution was added dropwise to pH 5. The green aqueous solution was extracted with ether (2x50 ml) and the ether extracts washed with 50% saturated sodium bicarbonate (6x40 ml). Washes 1 and 2 evolved considerable carbon dioxide. Washes 3-5 were very yellow and the final wash was colorless. TLC of the ether solution showed only the nitrotoluene products with the ortho isomer in strong predominance. After drying and solvent removal a light yellow oil was obtained (2.16 gm, 79%).

The above reaction was repeated exactly as described but ether was replaced by toluene for the extractions. The yield of nitrotoluenes was 1.83 gm (67%).

The above reaction was repeated but the acetic acid solvent was replaced with dichloromethane and the reaction was extracted with dichloromethane instead of toluene or ether. TLC results were identical and the yield of nitrotoluenes was 1.97 gm (72%).

A final experiment was done replacing cupric nitrate trihydrate/acetic anhydride in dichloromethane with sodium nitrate/acetic anhydride. Only a negligible amount of nitrotoluenes was observed.

Discussion

The nitration of toluene using the procedure described by Mellor did not match the yields reported in the paper. It is hard to believe that using sodium nitrate in place of potassium nitrate is the sole cause of the yield discrepancy. Therefore this method is somewhat suspect.

Cupric nitrate trihydrate provided the most encouraging result and did appear to yield a greater percentage of the desired ortho-isomer based on TLC analysis. The initial report of Mawardi as well as the report by Bell used acetic anhydride as the solvent. Not only is acetic anhydride somewhat difficult to obtain for many, its use leads to problems with reaction work up due to the large amounts of acetic acid produced. Switching to acetic acid as solvent provides equally good product yield, alleviates some of the acetic anhydride requirement but still leads to a large amount of acetic acid which must be neutralized for work up. The observation that the reaction proceeds and provides good yields using dichloromethane as solvent was most gratifying. The problem of acetic acid neutralization is also minimized. An excess of toluene was used in most reactions in order to ensure that all the nitrating species were consumed as much as possible.

Finally, there is the interesting question of what the yellow bicarbonate soluble material might be. The discharge of the yellow color upon acidification suggests a nitrophenolic compound. Such a compound would be quite soluble in bicarbonate. Unfortunately, I was not able to isolate anything from the acidified washes. It is interesting to note that dilute nitric acid, which may be present in these reactions, selectively oxidizes p-nitrotoluene to p-nitrobenzoic acid [6]. Could phenols be also formed more rapidly from the p-isomer and might this be the reason for the predominance of the o-nitrotoluene in these reactions?

Overall, the utility of cupric nitrate and acetic anhydride for small scale nitrations of some aromatic compounds may be quite good. The reaction is fast and, in the case of toluene, quite clean. There is a considerable literature on these types of reactions in general, unfortunately most of it behind paywalls. Scalability is a question due to the need for increasing amounts of acetic anhydride and, of course, the dangers inherent in generating acetyl nitrate. Because of the presence of acetyl nitrate, the reactions must be worked up with water rather than simply distilling off the reaction solvent to obtain the crude products.


Finally, I should point out that all attempts to nitrate phenolic compounds using the above procedures leads to tarry, multiple product mixtures. This may be a starting point for further investigation.

References

1. Mellor, JM, et al., Tetrahedron 2000, 56, 8019-8024.
2. Mawardi, R, Pertanika 1982, 5, 7-11.
3. Bell, KH, Aust. J. Chem., 1978, 31, 2567-2569.
4. Sandler, SR and Karo, W, Organic Functional Group Preparations, Vol. 1, p498 ff.
5. www.oc-prakticum.de, page 6.
6. US 2 815 373.


Attachment: Nitration of toluene with analytics.pdf (166kB)
This file has been downloaded 2340 times

Attachment: Nitrotoluene oxidation HNO3+O2 US2815373.pdf (339kB)
This file has been downloaded 938 times

Attachment: Ceric ammonium nitrate nitration of heterocycles.pdf (135kB)
This file has been downloaded 926 times

Attachment: Cupric Nitrate Nitration of 2-Phenylcyclohexanone.pdf (166kB)
This file has been downloaded 1029 times

Attachment: Regioselective nitrations w metal nitrates+Ac2O.pdf (248kB)
This file has been downloaded 1138 times

Attachment: Cupric Nitrate and toluene nitration.pdf (1.8MB)
This file has been downloaded 1267 times

Nicodem - 23-8-2015 at 00:49

AvBaeyer, thank you so much for the beautifully written contribution. I think this should be in the Prepublication section. If you allow I will move it there. A discussion can be had there as well.
Quote: Originally posted by AvBaeyer  
The nitration of toluene using the procedure described by Mellor did not match the yields reported in the paper. It is hard to believe that using sodium nitrate in place of potassium nitrate is the sole cause of the yield discrepancy. Therefore this method is somewhat suspect.

The use of potassium salt might make a difference, given the reaction depends on the formation of HNO3 from the NaNO3/H2SO4 slurry. Only HNO3 is significantly soluble in dichloromethane so that the nature of the cation should make no difference to the rate of nitration in organic phase. However, since the nitric acid might not liberate equally efficiently from the two salts, and might not get equally efficiently extracted from the inorganic slurry, I would expect that NaNO3 and KNO3 not only do behave differently, but also that stirring rate has a huge impact.
I would suggest either the use of more H2SO4, or use of acetic acid as a solvent instead of dichloromethane, possibly at a smaller amount to make the mixture more concentrated.
Quote:
Finally, there is the interesting question of what the yellow bicarbonate soluble material might be. The discharge of the yellow color upon acidification suggests a nitrophenolic compound. Such a compound would be quite soluble in bicarbonate.

I can't immediately see a viable mechanism for nitrophenols formation, though chemistry is all about surprises. Reactions at the methyl group which occur via nitrogen oxides might be the cause of impurities giving a yellow color in basic media. For example, the alpha, ortho/para-dinitrotoluene should be enough C-acidic to dissolve in sodium bicarbonate (I would estimate its pKa is somewhere in the range of 5 to 8). Its carboanion is probably intensely yellow. Another possibility is alpha-nitrosation of nitrotoluenes yielding the nitrobenzaldoximes which have the pKa of about 10. I assume these would also have a strongly yellow deprotonated form. Actually, once the methyl group is oximated, it more easily goes further to the dinitromethyl with further reaction with N(IV) oxides (via Ponzio reaction). Therefore, it only takes a little impurities derived from the reactions on the methyl group to give highly yellow colored impurities.
In any case, the impurity profile of toluene nitrations is well studied (due to industrial importance of TNT manufacturing). I know there are some articles and I remember reading at least one many years ago, its just that I can't be bothered searching them now.
Quote:
Unfortunately, I was not able to isolate anything from the acidified washes. It is interesting to note that dilute nitric acid, which may be present in these reactions, selectively oxidizes p-nitrotoluene to p-nitrobenzoic acid [6]. Could phenols be also formed more rapidly from the p-isomer and might this be the reason for the predominance of the o-nitrotoluene in these reactions?


I think ortho-nitrotoluene undergoes oxidations with hot dilute nitric acid similarly easily, but anyway, I don't think the observed increased ortho selectivity is due to side reactions specifically consuming the para isomer. Your limiting reagent were the nitrates which means that yields would be significantly affected if say 1/4 would be consumed in side reactions and you would probably observe the impurities on TLC. The yellow impurities are probably present at less than 1% amounts. It can take a few ppm of highly colored substances to give intense colorations. If you still have the developed TLC plates, you can try spraying them with sodium bicarbonate solution for the yellow spots to develop. If they become visible, but were not visible under UV, then you are dealing with insignificant amounts.

CuReUS - 23-8-2015 at 03:16

magpie,amazing example for a regioselective reaction.If you have time,see this paper,it seems o-nitrotoluene can be prepared in 70% yield
http://benthamscience.com/journals/letters-in-drug-design-an...

AvBaeyer - 23-8-2015 at 19:24

Nicodem,

Thank you for your kind comments and thoughts. These are much appreciated. If you wish to move this post, I have no objection.

Regarding the yellow "mystery" bicarbonate soluble product, I did notice (and I failed to mention in my haste- my apologies) that a uv-active, quite polar spot (Rf less than 0.1) appears on the tlc almost as fast as the appearance of the nitrotoluene spots. It is always a very minor spot and is independent of the reaction conditions. It is there with the sodium nitrate/sulfuric acid reaction as well as reactions employing cupric nitrate. It does seem to be larger in the sodium nitrate reactions than in the cupric nitrate reactions. This spot is present in the initial reaction extracts but is gone after the bicarbonate washes. Without further work, it is not possible to say that this new spot is the mystery compound, but it is a suspect. Your suggestion to spray with dilute bicarbonate is quite appropriate and might provide further insight. Unfortunately, I do not have any developed plates left, only my sketches and notes. I do not think I will be running any more of these reactions in the near term. Having run so many of them already (each reaction was run twice), I have enough mixed nitrotoluenes to supply more o-nitrotoluene than I need.

I do like your idea about some sort of reaction at the methyl group. A reaction at this site is quite consistent with the data in reference 6 and could be related to some partially oxidized intermediate as you have posited. I read somewhere that no one really knows the mechanism for the oxidation of nitrotoluenes to nitrobenzoic acids by dilute nitric acid or reaction products derived from nitric acid. Another project presents itself!

Again, thank you for your thoughts,

AvB

PHILOU Zrealone - 24-8-2015 at 07:27

Quote: Originally posted by AvBaeyer  


I do like your idea about some sort of reaction at the methyl group. A reaction at this site is quite consistent with the data in reference 6 and could be related to some partially oxidized intermediate as you have posited. I read somewhere that no one really knows the mechanism for the oxidation of nitrotoluenes to nitrobenzoic acids by dilute nitric acid or reaction products derived from nitric acid. Another project presents itself!


Maybe take a look at Green solution tread in Energetic materials.