Sciencemadness Discussion Board - Benzaldehyde synthesis using aqueous nitric acid

Benzaldehyde synthesis using aqueous nitric acid

Amos - 3-2-2017 at 20:28

Introduction

The procedure below describes the oxidation of benzyl alcohol to benzaldehyde in high yield using aqueous nitric acid as an oxidizer. Other methods of oxidizing benzyl alcohol to benzaldehyde are well-known, including those using chlorochromates, persulfate, or activated manganese dioxide. The method is advantageous in that it gives a high yield while using relatively simple equipment and more common, easy-to-obtain reagents than others.

Synthesis

50g (714mmol) of 90% nitric acid (homemade, concentration determined by titration, density 1.48g/mL) were placed in a 250mL round-bottom flask. The nitric acid was cooled in a salted ice bath and a couple of mL of benzyl alcohol (technical grade, KYAntec) were added to the flask using a Pasteur pipette.

The flask was manually swirled by hand to mix the reagents, and an immediate change in color from pale yellow to a bright yellow-green was noted. As more benzyl alcohol was added, an upper layer separated and the color of this gradually deepened to an intense blue-green. The release of brown nitrogen oxide vapors was also noted. The classic almond extract of aroma was already heavy upon the air between additions by this point.

Upon each further addition of benzyl alcohol and with stirring of the mixture, the color disappeared and became milky yellow, but after being allowed to react, the blue color returned; the color change was used to follow the reaction, with each addition of benzyl alcohol occurring after the upper layer had regained the unusual coloration. With some difficulty I managed to get enough light passing through the deep green mixture for a picture.

A total of 66g (610mmol) of benzyl alcohol were added over the course of about 4 hours. While an ice bath was used to regulate temperature at the beginning of the experiment, as the reaction continues and the concentration of nitric acid falls, the reaction slows significantly and it is sufficient to carry out the last third or so of the reaction at room temperature. Because a goal of this synthesis was to avoid benzyl alcohol contamination in the product, addition of benzyl alcohol ceased when the blue-green color didn’t return, at least to its full strength, after allowing the reaction mixture to sit for a period of 30 minutes.

After being left to sit overnight in an airtight container, the two-layer mixture was placed in a separatory funnel and the lower aqueous layer removed. The top layer was washed twice with saturated sodium bicarbonate solution, followed by distilled water, and finally partially dried with a wash of saturated sodium chloride. The pale green upper layer as well as the aqueous layer became immediately red-orange in color after the sodium bicarbonate solution was added for the first time, though the color of the aqueous layer is noticeably less with each successive wash. The brine wash came out colorless, though the benzaldehyde layer is still strongly colored.

The crude benzaldehyde, a translucent reddish-orange liquid, was placed into a pre-weighed container with a little anhydrous sodium sulfate and the weight was recorded as 55.3 grams. If pure benzaldehyde was assumed, this would correspond to an 85% yield! Upon prolonged storage over the sodium sulfate, the color actually lightened to a golden yellow color, though this isn't pictured.

My plan with this mostly pure benzaldehyde is to purify it using the standard bisulfite adduct formation seen elsewhere. I'm not going to include it as a part of this write-up due to the fact that I don't have the highest hopes for it; I don't own vacuum filtration equipment. A good example of this can be found on Chem Player's YouTube channel or elsewhere on this forum.

Analysis

I’m fortunate enough to work in a lab with a GC-MS (gas chromatograph with built-in mass spectrometer) to use at my disposal, so the benzaldehyde came to work with me the following day. The resulting chromatogram obtained (I’ll be attaching it when I get to work on Monday) contained 3 major peaks. The largest peak, benzaldehyde, was measured at 92% of the peak area, with an additional peak for residual benzyl alcohol making up another 3%. The slightly more worrying peak could not be confidently identified from the machine’s database, but the compound suggested it to be benzyloxyurea (MW: 166.18). This peak didn’t show up in earlier small-scale tests, so I’m not sure if the impurity was introduced as a side product or if I accidentally introduced it through sloppy lab practices (certainly not out of the question in my case).

Discussion

First I'd like to say that I came upon this method entirely by accident a while back, originally looking to make 3-nitrobenzaldehyde. Hence I didn't reference a paper for the reaction, nor do I know the mechanisms involved. I can update the thread as per your suggestions, but in the meantime it serves simply as a guide for preparing benzaldehyde via this method.

Assuming the GC-MS is at all accurate (and it generally has been in the right ballpark where my work is concerned), the yield of this synthesis in pure benzaldehyde came out to about 79%, which I'm more than happy with. The "benzyloxyurea" peak is troubling to me, as I didn't encounter this peak in earlier test runs despite both of them being rushed and encountering thermal runaways during the preps. If anyone has any idea what this compound could be, or what the blue-green intermediate is during the synthesis, I'm all ears.

Given my limited equipment and space, this synthesis, while proceeding without any major hitches or obstacles, could likely be improved in a few ways. Magnetic stirring and an addition funnel would almost certainly be preferable to hovering over the reaction mixture and swirling it manually, for example. But the fact that I made do with what I had is a testament to the ease and accessibility of this pathway for the less well-equipped amateur.

Since this is more-or-less my first real write-up on the forum, I'd love your feedback about how to improve my posts in the future, or any concerns or critiques of the synthesis itself.

[Edited on 2-4-2017 by Amos]

DraconicAcid - 3-2-2017 at 21:07

Benzyloxyurea? That seems like an odd impurity. Could it be benzoic acid?

ETA: No, you have a mass spec on the GC, so that wouldn't be it.

[Edited on 4-2-2017 by DraconicAcid]

Melgar - 4-2-2017 at 00:54

I found this yesterday, in the course of trying to find a different article that I was looking for:

http://www.tandfonline.com/doi/abs/10.1080/00304940509354991...

I'm guessing that this reaction produces water, hence the need for the P2O5 in the reaction I linked to. The nitric acid is obviously the oxidizer. It seems oxidation of benzylic alcohols is right up there with "reduction of nitroarenes to anilines" as far as organic reactions that aren't particularly difficult.

Really, you can do this reaction with bleach as well, as long as you maintain good stirring and slow addition. A nonpolar layer to trap the aldehyde also helps.

The green color is probably the brief appearance of nitrous acid (which is blue) combined with the yellow-orange of the NOx fumes. Nitric acid minus one oxygen is nitrous acid, so the stoichiometry works out, although nitrous acid is usually pretty reactive with alcohols. Maybe benzyl nitrite decomposes into benzaldehyde after some time? Or maybe it's just one of those weird exceptions that always seem to affect anything attached to the benzyl carbon?

[Edited on 2/4/17 by Melgar]

unionised - 4-2-2017 at 03:36

It's the right mass for phthalic acid- which gets everywhere due to the common use of its esters as plasticisers.
Does it have a fragment at M/Z =149?

https://www.agilent.com/cs/library/applications/si-02100_pht...

[Edited on 4-2-17 by unionised]

Dr.Bob - 4-2-2017 at 06:11

Here is other article with some good background, references, and suggested mechanisms. They also show benzylnitrite as an intermediate, but that mass is off from your impurity, benzyl nitrite would be mw 137 and benzyl nitrate would be mw 153.

The experiment looks pretty good, much simpler than using PDC or another metal based oxidant. And the yield is fine for most work. Good job on the write up.

https://www.scribd.com/doc/78113590/Benzyl-Alcohol-to-Benzal...

[Edited on 4-2-2017 by Dr.Bob]

[Edited on 4-2-2017 by Dr.Bob]

Amos - 4-2-2017 at 06:13

Phthalic acid gets dehydrated to the anhydride on the GC with this method. It's most likely a compound that is too obscure to be in the library in the first place, as the MS identifies most common molecules like phthalic anhydride or benzaldehyde with greater than 60% confidence. This was 13% if I recall correctly.

unionised - 4-2-2017 at 09:27

Is nitrobenzoic acid plausible?
http://file.scirp.org/Html/7-1020108/8fa13627-51c3-40cf-a2c5...
I must admit the loss of hydride looks a bit fishy to me.

Wiki tells me that "3-Nitrobenzoic acid (m.p. 142 °C) is a precursor to 3-aminobenzoic acid, which in turn is used to prepare some dyes. It can be prepared by nitration of benzoic acid. It also can be prepared by treating benzaldehyde under nitration conditions, a process that initially converts the aldehyde to the acid."

If there's a mixture of two or more materials that co-elute on the GC the MS will give funny results.
Is the mass spectrum the same from the front of the peak to the tail? (And I'd expect some tailing with an acid like this).

I don't suppose you kept the bicarbonate washes? They might have been worth working up for benzoic acid(s)

[Edited on 4-2-17 by unionised]

AvBaeyer - 4-2-2017 at 18:56

Amos,
Some of your observations are similar to my own while studying the nitration of toluene with cupric nitrate in acetic anhydride or sodium nitrate with sulfuric acid. These reactions also gave an intensely yellow colored extract with sodium bicarbonate. The color was discharged by the addition of acid. The thread can be seen at:

http://www.sciencemadness.org/talk/viewthread.php?tid=63393#...

AvB

Chlorine - 4-2-2017 at 19:09

I'm almost sure it's nitrobenzoic acid+ a minute amount of benzoic acid.

Texium - 4-2-2017 at 22:08

Quote: Originally posted by Chlorine

I'm almost sure it's nitrobenzoic acid+ a minute amount of benzoic acid.

You have a reference for that?

wg48 - 5-2-2017 at 05:18

Quote: Originally posted by Melgar

I found this yesterday, in the course of trying to find a different article that I was looking for:

http://www.tandfonline.com/doi/abs/10.1080/00304940509354991...

I'm guessing that this reaction produces water, hence the need for the P2O5 in the reaction I linked to.

[Edited on 2/4/17 by Melgar]

That paper is an odd and curious synthesis It uses 65% nitric acid and phosphorous pentoxide (PP) supported on silica gel. Now 65% nitric acid is 35% water and the given stoichiometry by my calculation of 18mmol of H2O per 0.9mmol of PP all PP will be converted to phosphoric acid.

Ignoring the reaction of the PP with the silica gel and the alcohol, presumably phosphoric acid could be used in place of the PP.

What am I missing? The PP is supposed to be a catalyst

chemplayer... - 5-2-2017 at 05:23

Well done and great write-up.

It would be good to follow-up if possible and let us know how much benzaldehyde you recovered after going through the bisulfite purification process (distillation without a good vacuum is not really feasible and you end up with a huge amount of oxidation)?

Another method we read about was the use of copper (II) nitrate solution to do the oxidation. It definitely works but we found it always over-oxidises to too much benzoic acid, or in the case of diluting the mixture, then suddenly causes the reaction to stop and fail. Your method could be a really convenient and easy one if the yield checks out...

Praxichys - 5-2-2017 at 10:17

Since you have access, would it be possible for you to run GC/MS on the initial benzyl alcohol? I have some USP stuff and also some of the KYAntec stuff. The USP grade is water clear but as you know the KYA stuff is an amber color. I'm wondering what the impurities are in that, and if that contributed to the byproducts in the product?

EDIT:

If anyone is interested in the potential of this method, check out this paper:

Looks like you can pretty much mix 1:0.625 BnOH:HNO3 of 10% concentration, heat it for a couple hours with a pinch of NaNO2, and get like 95% BzH yield. Nice!

Attachment: scribd-download.com_benzyl-alcohol-to-benzaldehyde-oxidation-w-nitric-acid-92-yield-certified-diy.pdf (176kB)
This file has been downloaded 1274 times

The paper shows that high concentrations of nitric acid reduce yield. At low concentrations of nitric acid the reaction does not require cooling, and in fact needs additional heat.

A procedure from the paper:

0.3 mol benzyl alcohol is charged into a flask with 1g NaNO2 and 118mL of 10% aqueous nitric acid preheated to 90°C. This is stirred as rapidly as possible and kept at 90°C for 4 hours. The final product shows >86% conversion of the benzyl alcohol to benzaldehyde and benzyl nitrite, with a yield of >95% for benzaldehyde and <1% for the nitrite, with <5% unconverted alcohol.

[Edited on 5-2-2017 by Praxichys]

Amos - 5-2-2017 at 13:13

Quote: Originally posted by Praxichys

Since you have access, would it be possible for you to run GC/MS on the initial benzyl alcohol? I have some USP stuff and also some of the KYAntec stuff. The USP grade is water clear but as you know the KYA stuff is an amber color. I'm wondering what the impurities are in that, and if that contributed to the byproducts in the product?

EDIT:

If anyone is interested in the potential of this method, check out this paper:

Looks like you can pretty much mix 1:0.625 BnOH:HNO3 of 10% concentration, heat it for a couple hours with a pinch of NaNO2, and get like 95% BzH yield. Nice!

The paper shows that high concentrations of nitric acid reduce yield. At low concentrations of nitric acid the reaction does not require cooling, and in fact needs additional heat.

A procedure from the paper:

0.3 mol benzyl alcohol is charged into a flask with 1g NaNO2 and 118mL of 10% aqueous nitric acid preheated to 90°C. This is stirred as rapidly as possible and kept at 90°C for 4 hours. The final product shows >86% conversion of the benzyl alcohol to benzaldehyde and benzyl nitrite, with a yield of >95% for benzaldehyde and <1% for the nitrite, with <5% unconverted alcohol.

[Edited on 5-2-2017 by Praxichys]

I've actually already run it but haven't taken it off the machine onto my computer yet. IIRC there were small amounts of both benzaldehyde and benzyl alcohol in the starting material, but any impurities amounted to less than 5%. Probably what you'd expect after storing any benzyl alcohol for a few years.

Amos - 5-2-2017 at 13:15

Quote: Originally posted by Chlorine

I'm almost sure it's nitrobenzoic acid+ a minute amount of benzoic acid.

Nope, this is the same retention time as benzaldehyde I've made via other methods, the product isn't a solid at room temperature (this would probably be your first clue...?), and essentially the entire mass of the product converted to a cream-colored solid after treatment with sodium bisulfite solution. Not to mention I still had a product at the end of the base washes. You might want to read the write-up.

This stuff is benzaldehyde.

Melgar - 5-2-2017 at 15:20

Quote: Originally posted by wg48

Phosphorus pentoxide on silica gel is a common heterogeneous reagent for solventless reactions, since P2O5 by itself isn't very pleasant to work with. P2O5 coats the silica gel grains, and also traps nitric acid as it gradually turns into gel-like polyphosphoric acid. I'm pretty sure its only purpose here is to act as a desiccant and a support.

Also, keep in mind, silica gel is a pretty good desiccant as well.

Amos - 6-2-2017 at 08:55

So here's the original GC-MS readout for my crude benzaldehyde. The peaks are in order: Benzaldehyde, the unknown compound, benzyl alcohol, and 3 other tiny peaks each less than 1%: p-nitrobenzyl alcohol, benzoic acid, and 3-nitrobenzaldehyde. It would seem that I was not as considerate in this preparation as I had been in my earlier test runs. The mass spectrum pictured is from the second peak, which has still yet to be identified.

Here's what the mass spec has to say about the second peak. As stated earlier, the compound with the highest probability out of those in the directory is benzyloxyurea, but it would seem more likely that it is a compound the software doesn't recognize. The other candidates given actually have a variety of different molecular weights, making it even harder to discern what it might be.

Finally, I wanted to attach this readout from my preliminary test of the oxidation with nitric acid. This was on the scale of 2 grams benzyl alcohol and 5mL or so of nitric acid; a good deal more acid than I used later on. Over 99% is benzaldehyde with the balance of the peak area as either p-nitrobenzyl alcohol or noise.

I'm making another go at this synthesis for consistency's sake tonight or soon thereafter on a 1/5 scale of the one in the writeup; I'll post the results as soon as I've got them.

[Edited on 2-6-2017 by Amos]

Nicodem - 6-2-2017 at 13:07

It could be a nitrobenzaldoxime. It has M=166.1, it would not be washed away by bicarbonate washes, but I would not expect it to come out of GC so close to benzaldehyde, and if this is present, then I would expect an even higher presence of benzaldoxime as well (M=121). So this still does not look as the best candidate.
nitrobenzaldoxime.gif - 1kB

Obviously, the isomeric nitrobenzamide also has M=166.1, but that would surely need way more time to come out of GC. And besides, for it to form, nitrobenzaldoxime would have to be an intermediate.

If anything isomeric, then I would rather consider a nitroxime as the most likely candidate:
nitroxime.gif - 1kB

That would explain the absence of plain benzaldoxime. However, I'm not so sure it would not be washed away by bicarbonate washes (I would expect its pKa is somewhere between 4 and 8). You could try washing a sample with 1M aq. NaOH and run it on GC-MS to check if that peak gets diminished.

Can you post the MS of the second peak? The MS you posted is for the peak at 6.8 min, which is the benzaldehyde peak.

[Edited on 6/2/2017 by Nicodem]

Just found out that nitroximes are called "nitrolic acids".

[Edited on 7/2/2017 by Nicodem]

Amos - 7-2-2017 at 07:08

Ack, my bad. Here's the MS of the unknown peak; I don't know how useful it will be.

aga - 7-2-2017 at 09:08

Beautiful and inspiring work Amos !

Nicodem - 7-2-2017 at 13:01

Quote: Originally posted by Amos

Ack, my bad. Here's the MS of the unknown peak; I don't know how useful it will be.

That looks like benzyl alcohol over the tail of benzaldehyde's peak. It is not the mass spectra of the unidentified M=166 impurity you talked about. Are you sure you didn't mix up the order for the second and the third peak, and the unknown impurity is actually the third peak?

Amos - 7-2-2017 at 13:24

Quote: Originally posted by Nicodem

Quote: Originally posted by Amos

Ack, my bad. Here's the MS of the unknown peak; I don't know how useful it will be.

I'm 100% positive the 3rd peak is benzyl alcohol; I've seen it show up at this retention time in several samples before. There's nothing conclusive that says the molar mass of peak 2 is 166; that's simply the mass of benzyloxyurea, which was the "best match" the MS's directory contains. As you can tell from this problematic-looking mass spectrum and from the list of suggested compounds (which have wildly varying masses) on the readout I posted earlier, the MS really has no idea what molar mass the compound has. Hence why I would like to re-do the experiment and verify that the synthesis can be done without producing this extra impurity, as I managed in previous attempts at smaller scales.

[Edited on 2-7-2017 by Amos]

[Edited on 2-7-2017 by Amos]

Nicodem - 7-2-2017 at 14:06

Quote: Originally posted by Amos

I misunderstood. I thought you deducted the mass peak from the spectra (N-benzyloxyurea is no reasonable candidate for an impurity, so I could not understand why you brought it up, unless it fitted the mass).

Then the obvious choice for the second peak is benzyl nitrite (M=137.1). The mass peak is there, although very small, but that is to be expected, as benzyl nitrite should easily fragment to the same fragments as benzyl alcohols (91, 51). It is an intermediate in the synthesis, so its presence is normal. The retention time is also reasonable for its physical properties (in between benzaldehyde and benzyl alcohol).

Amos - 8-2-2017 at 07:37

Quote: Originally posted by Nicodem

Quote: Originally posted by Amos

Great! I appreciate your expertise on the matter; if benzyl nitrite is the impurity, perhaps the benzaldehyde yield can be increased and made more pure if part of the workup includes an acid hydrolysis step. When I've got more nitric acid perhaps I'll give this another go.

Eddygp - 9-2-2017 at 05:45

Ogata Y, Sawaki Y, Matsunaga F, Tezuka H. Kinetics of the nitric acid oxidation of benzyl alcohols to benzaldehydes. Tetrahedron, 1966: 22(8), 2655-2664.

"A probable mechanism is discussed, which involves a rapid and reversible coupling of α-hydroxybenzyl radical with nitrogen dioxide followed by the hydrolysis of α-hydroxybenzyl nitrite"

http://www.sciencedirect.com/science/article/pii/S0040402001...

This was in conditions with 10% HNO3. A high concentration implies that not as much water will be available for the hydrolysis of the nitrite. Basically... this confirms Nicodem's comment.

Praxichys - 9-2-2017 at 07:02

Yes, the paper I posted a few days ago gives the mechanism and the rate constants for each step at various temperatures. Another key factor is phase transfer since initially the acid and the alcohol are immiscible.

It is no surprise that you either need conc. HNO3 or a nitrite initiator with dilute acid since nitrite plays a catalytic role. You either get it from decomp. of the acid or by adding some directly. Using dilute acid is clearly preferable since the aldehyde forms through hydrolysis and you will end up with less BnONO and ring-nitrated products as impurities.

kmno4 - 9-2-2017 at 09:10

See interesting article, posted long time ago on the board :
http://www.sciencemadness.org/talk/files.php?pid=125689&...

LD5050 - 26-2-2017 at 18:25

Is it important the concentration be at 90% or can higher concentrations be used? I'm performing this experiment right now and I have some pretty concentrated acid probably around or over 95%. I'm getting a very strong reaction with the addition of Benzyl alcohol and a hint of Benzaldehyde smell that quickly dissipates. I'm not getting a separation of layers or green coloring yet but I only added a few ml's at this point.

Also you say that when left to sit over night you then put into a seperatory funnel and remove the lower aqueous layer. For me it seems that the lower layer is oily and im thinking that it is the benzaldehyde/benzyl alcohol layer. could for some reason my benzaldehyde be on the bottom?

[Edited on 2-27-2017 by LD5050]

CaptainMolo - 13-9-2017 at 08:50

I've been after a good way to do this lately so I thought I'd throw in my two cents, although depending on it's usefulness, the market value of my input my vary.

Quote:

For me it seems that the lower layer is oily and im thinking that it is the benzaldehyde/benzyl alcohol layer. could for some reason my benzaldehyde be on the bottom?

During my most successful trial to date, I had this exact thing happen, so yes, in my experience it very much can depending on the density of the aqueous layer.

I was trying an oxidation using a dilute piranha solution, which appeared to work reasonably well, the place where I messed up appears to be during the workup.

I formed the bisulfite adduct, and filtered it off, but when trying to break the adduct I boogered it up somehow and ended up with a mess. It was a very small scale run, 20 milliliters of BnOH to start with so I scrapped it to try a different approach.

I'm going to prepare some nitric acid in the next few days and give this a whirl, and I'll report back with my results.

[Edited on 13-9-2017 by CaptainMolo]

Corrosive Joeseph - 13-9-2017 at 17:37

Posted recently on another forum -

"Method 1

In a 2L flask, 800 ml of a 10% solution of HNO3 was added, to which 150 ml of benzyl alcohol was added. The flask is placed on medium heat, reflux with a reflux condenser (put on the hose and vent it). Continuous stirring, after about 40 min rm boils and begins a self-sustaining reaction, the mixture turns orange, apparently the oxides of nitrogen. It's time to remove the flask from the heating and help stirring. After about 30 minutes the reaction subsides, and it must be maintained, put on average heating, reflux 2-2.5 hours. After a certain time, remove the reaction mass from the heating, cool and rinse 3 * 500 ml of 15% K2CO3 solution. Carbonate washing allows to get rid of HNO3 residues, if this is not done, the yields decrease to 40-50% due to oxidation of part of BA to benzoic acid. Next, we separate the lower phase of the dirty benzoic aldehyde and put it on distillation with steam (you can not do this, and so it condenses well). The yield of the ready-pure benzoic aldehyde in the region of 80-84%

Method 2

Nitric acid was diluted to 1.59M (10% by weight). 4.6 g (42 mmol) of benzyl alcohol were added to a round bottom flask, and 18 g of 10% acid (28 mmol) was added, maintaining the molar ratio of benzyl alcohol to acid, approximately as in article 1: 0.625. About 0.3 g of NaNO2 was added to the eye ** and heated to 90 ° in a water bath at high agitator speeds.

The entire mixture was heated for 4 hours (70-90 degrees) in a bath with a reflux condenser with running water. During the reaction, brown nitric oxide is released, so an extract (or at least an open balcony) is necessary. The organic phase became yellowish.

The organic phase is separated on a separatory funnel, the water is extracted with DCM a couple of times. The organic phase is fused with DCM and washed with a small amount of weak NaOH (while it is strongly yellowed), then saturated with the r-rum salt. The hydrochloric solution is extracted with DXM just in case. After evaporation of DCM, 3.9 g (85%) of dirty benzaldehyde was finally obtained. Further purification was not carried out, because, as practice shows, impurities of benzyl alcohol do not interfere with condensation.

** The amount of NaNO2 is not important, since it is only needed to initiate the process."

/CJ

Dennis A - 19-5-2018 at 02:06

Has this method only tried with un-subsituted benzyl alcohols ?

Regards

Dennis A - 22-5-2018 at 00:34

Error i meant to say substituted benzyl alcohols

sumatra - 30-8-2018 at 04:46

Must the nitric acid be 90%? Will 58% suffice?

Fyndium - 10-11-2020 at 08:03

Quote: Originally posted by Praxichys

0.3 mol benzyl alcohol is charged into a flask with 1g NaNO2 and 118mL of 10% aqueous nitric acid preheated to 90°C. This is stirred as rapidly as possible and kept at 90°C for 4 hours. The final product shows >86% conversion of the benzyl alcohol to benzaldehyde and benzyl nitrite, with a yield of >95% for benzaldehyde and <1% for the nitrite, with <5% unconverted alcohol.

I tested the reaction and it was about to turn a big mess. I pre-heated the 10% HNO3 to 90C, added the nitrite which fizzed a little and immediately created a brown cloud, and then I dropped in the benzyl alcohol in during a couple of minutes instead of just pouring. At first it was looking ok, but then proggressively more red gas was emitted and it started to fume through the inverted funnel in a runaway type scheme, so I quenched the reaction by quickly pouring in a liter of cold water. Why this happened, even though I followed the above conditions?

Should the benzyl alcohol be added into the acid solution slowly, because if it is poured in fast, large amounts of brown nitrogen dioxide gas is formed?

Or should they first be mixed cold, and then slowly heated to 90C?

It is certain that BzH was formed because the odor of it was pervasive. I vacuum distilled some of the benzyl alcohol and it appeared and smelled the same, so it's mostly itself.

[Edited on 10-11-2020 by Fyndium]

Fyndium - 11-11-2020 at 11:24

I think I'll try to alter the reaction so that I first mix and heat the 10% HNO3 and nitrite to 90 and then drop in the BnOH during 4 hours to control reaction rate. Dumping all in sounds too uncontrollable to me. I don't know how they did this in study, but I managed to just create a blast of NO2 brown cloud and got exposed to it, and inverted funnel over conc NaOH was a mere joke absorbing the gas, such was the rate of formation.

nzlostpass - 12-11-2020 at 00:27

It seems to always take off when temp hits about 80 to 90 degrees and all reactants are in.......last time, I heated the BnOH first then added in the sodium nitrite in one go then the nitric acid in small bursts......that seemed to work but I guess it would make more sense to heat the acid first and add the BnOH but I was worried about the ratio of nitric to alcohol being too high if adding the BnOH last.

Fyndium - 12-11-2020 at 01:32

I suppose the method works without big issues because the OP did so and even with 90%, not 10% HNO3.

Fyndium - 12-11-2020 at 14:56

Well, whatever I did, the main composition was not benzaldehyde, although the smell was there.

At least a significant part of it appears to be benzoic acid. Upon cooling to room temp, crystalline mass formed and made the entire reaction a mush, and dark gunk accumulated at the bottom, which did only partially react with sodium bicarbonate and it had very powerful yellow staining ability to fingers.

There were no recoverable layer or form of benzaldehyde, everything seem to have turned into benzoic acid and something else alongside.

How could it have overoxidized?

Amos - 12-11-2020 at 16:37

Quote: Originally posted by LD5050

Is it important the concentration be at 90% or can higher concentrations be used? I'm performing this experiment right now and I have some pretty concentrated acid probably around or over 95%. I'm getting a very strong reaction with the addition of Benzyl alcohol and a hint of Benzaldehyde smell that quickly dissipates. I'm not getting a separation of layers or green coloring yet but I only added a few ml's at this point.

Also you say that when left to sit over night you then put into a seperatory funnel and remove the lower aqueous layer. For me it seems that the lower layer is oily and im thinking that it is the benzaldehyde/benzyl alcohol layer. could for some reason my benzaldehyde be on the bottom?

[Edited on 2-27-2017 by LD5050]

This will depend on the concentration of leftover nitric acid that you end with, since nitric acid is far more dense than benzaldehyde or benzyl alcohol. Your intuition about the benzaldehyde being on the bottom is likely correct, which would indicate that you used up most of the nitric acid during the course of your reaction, or that your initial concentration was lower than estimated. I know this was a 3-year-old question but I wanted to clarify this to anyone who decides to attempt the reaction in the future.

[Edited on 11-13-2020 by Amos]

nzlostpass - 12-11-2020 at 19:47

Quote: Originally posted by Fyndium

Well, whatever I did, the main composition was not benzaldehyde, although the smell was there.

At least a significant part of it appears to be benzoic acid. Upon cooling to room temp, crystalline mass formed and made the entire reaction a mush, and dark gunk accumulated at the bottom, which did only partially react with sodium bicarbonate and it had very powerful yellow staining ability to fingers.

There were no recoverable layer or form of benzaldehyde, everything seem to have turned into benzoic acid and something else alongside.

How could it have overoxidized?

That happened to me when I heated it for too long.....once its up to 90 degrees, turn down heating and let the reaction take place, it finally cools down after 2-4 hours and then proceed.
The first time i tried it, I did it in an open beaker and heated/stirred it too long and got what you describe. Another time I started it and had to turn it off and go to sleep and when I awoke it looked like a good amount of benzaldehyde had been produced but I was greedy and heated it again to try get more as it hadnt had the full 4 hours....bad idea....it did what you described.
Now i just let it do its thing after it gets to 90 degrees and it seems to work great.
The yellow stains are a bastard but you cant seem to avoid that.
I'm still a beginner myself so take what I say with a grain of salt but that is my observations/experience.

Fyndium - 12-11-2020 at 23:21

Hmm, thanks for the input. Funny how the research paper states to heat for 240min instead of just starting a reaction and letting it cool. I suppose I gotta try your method since you state it works, apparently mine doesn't.

What I was looking into was this vid on Tube:

https://www.youtube.com/watch?v=FYZpsOsUxnU

nzlostpass - 13-11-2020 at 00:52

Quote: Originally posted by Fyndium

Hmm, thanks for the input. Funny how the research paper states to heat for 240min instead of just starting a reaction and letting it cool. I suppose I gotta try your method since you state it works, apparently mine doesn't.

What I was looking into was this vid on Tube:

https://www.youtube.com/watch?v=FYZpsOsUxnU

Well it kinda goes for that sort of time by itself....I may have added a small amount of heat to keep it at 90ish degrees but it definately has exothermic properties.....and prolonged heating definately caused the benzoic acid
formation.

And I too, followed that Toms Lab video and the posts that have been mentioned.

[Edited on 13-11-2020 by nzlostpass]

Fyndium - 13-11-2020 at 09:58

Just wonder if NO2 catalyzes benzyl nitrite formation, could nitrate residues cause formation of benzyl nitrate, hence affecting yield?

Fyndium - 13-11-2020 at 14:16

I replicated the Tom's Lab method by the book and got 26.3g yield of crude BzH from 31.5mL of BnOH. Only difference was using only 0.5g of NaNO2, which is self-made and appears to contain significant amount of carbonate based on strong effervescence upon mixing with HNO3. Being an initiator catalyst, the amount does not matter as long as there is, difference between 0.1 and 1g was negligible in the research.

I mixed up the ingredients and heated up to 90C for 4h and an extra 30 minutes because I hesitated at the middle for ½h and let it cool for a while for not being able to decide which method I try first, but I ended up using the Tube/Research paper method. The BzH conversion from nitrite occurs strongly at the last hours of the reaction according to the research paper, hence running it hot seems to do the job. Using water bath the temp hovers steadily between 90-94C, adding water a couple of times. I used a 150mL fritted funnel as a trap because it allowed significant negative pressure buildup without letting oxygen in the system. No benzoic acid crystals were noticed at any given point, compared to the previous reaction which ended up as a crust of BzOH and some yellow gunk, probalby unreacted nitrite. I cooled down the reaction completely to ntp before opening the system, separated the bottom layer, washed it once with water, twice with conc bicarb solution for zero CO2 emissions and then with brine, bottled it as a slightly cloudy brown liquid and added a pinch of CaSO4 and it cleared up. The layer darkened as it was washed, it started as orange, and ended up as dark brown.

Gonna do a few more reactions before I vacuum this and add to my bottle of previous vacuum BzH extract from bitter almond extract that consisted mostly of oil and BzH.

Is there a danger of detonations from any possible remaining nitro compounds within the crude BzH upon distillation? I generally add a dose of mineral oil into the reaction flask so I can distill it till the end without going dry. Would it be less risky to strip it by steam distillation instead?

PS. The yellow staining property of this process is annoying. Wearing gloves is mandatory, unless one wants to be a goldfinger. I suppose it originates from some nitro compounds, as I remember that nitrophenols appear to be quite stainy too. It is very easy to rinse off from glass and hard surfaces, but it just sticks to the skin and nails.

[Edited on 13-11-2020 by Fyndium]

[Edited on 13-11-2020 by Fyndium]

nzlostpass - 13-11-2020 at 17:10

I steam distilled mine as im not very confident doing vacuum distillations yet...let us know what you decide and how you got on please.

Fyndium - 14-11-2020 at 03:01

Absolutely I will. I should have time to do further processing during next week so I know better.

I would love to test out steam distillation as I haven't never done it. Vacuum distillation is more of a routine to me nowadays, whatever cannot be distilled with bleeder tube(because haven't got no inert gas) goes with a stirrer.

Fyndium - 16-11-2020 at 16:02

I prepared another batch, which is 205 grams from 250g of BnOH, after washing twice with saturated NaHCO, once with pure water and finally with brine. The result is much darker than the previous batch, dried with CaSO4.

Another curious notice is that the first batch I put in freezer has crystallized out something. Not sure if it could be BzOH, BnOH or some other impurity? I put the second batch also in freezer to see if same happens. BnOH has mp of -15.4C so it could easily be residual which should be removed by distillation.

I'm still not confident distilling the substance as I'm not sure if there are anything left that could decompose violently?

morganbw - 16-11-2020 at 16:26

A byproduct of this synthesis is benzyl nitrite. It has a close boiling point to benzaldehyde. It seems doubtful that distillation would separate the two.
By the way it should not be a huge amount but it will be there.

Perhaps making the adjunct, filtering, rinse the organics from the adjunct.
Carry on from there?

Just thinking through my keypad.
Good luck.

Fyndium - 16-11-2020 at 16:44

Adduct is an option to separate it really pure, yes, I gave a thought for that and got bisulfite at hand, but keep it as a last resort because it is a high loss process.

If the BnONO amount is small enough and it does not interact with the possible further uses of the BzH, it could just be left in. The content of it is reported in the research paper as few % max when going at 90C. It's BP is almost identical to BzH so separation by distillation does not work. On the other hand, the concentration of BnONO should hence remain so marginal as it co-distillates that no complications should occur. If the mere heating or enrichment in the reaction vessel could cause issues, that's more of a question.

I'm not sure if BnONO react with NaHCO3 or would it decompose back to HNO3 and BnOH during processing?

morganbw - 16-11-2020 at 16:53

Looks like you knew already what I was cautioning you about.
The questions you now have seem to be as mine are.

My main problem is that I do not know benzyl nitrite well enough to know how it would behave in certain reactions.

Fyndium - 17-11-2020 at 06:01

Wait a minute. There are two sources that state BzH has -26(Sigma, Britannica) or -54C(wiki, chem sites) mp. If -26 is correct, my solid could be it instead of an impurity. My freezer goes down -35 at best.

Btw, how quickly BzH oxidizes when handling, like separating layers, pouring into flasks, etc? Is the oxidation much slower at cold temps as it is fast in hot temps?

And finally, just as in persulfate method, could BzH be steam distilled directly from the reaction vessel or would HNO3 and other stuff co-distill?

EDIT: Prepared another test run, from 25mL BnOH I got 22g of BzH.

[Edited on 17-11-2020 by Fyndium]

Fyndium - 18-11-2020 at 05:32

This article appears interesting:

https://doi.org/10.1039/JR9600003932

It studies pyrolysis of benzyl nitrite alone, with BzH and various other stuff. There appears to be many references that BnONO is vacuum distilled, and it decomposes to nitrous or nitric oxide and CO2, BzH and other stuff, and decomposition half-life for ex in 190C vapor phase is 70min. Hence there appears to be of no danger of sudden explosions with this process.

If someone else notices potential issues, please refer. I thought that I could first vacuum a small amount and see how it behaves, and put some paraffin in the reaction flask to prevent going dry.

Fyndium - 19-11-2020 at 06:32

I successfully vac'ed the BzH. I got total yield of pure distillate from crude synthesis product in a bit over 80% yield, but I did not use fractioning column and I stopped the distilling when the temperature started to rise significantly. I collected most at 54C, where it climbed quickly and stayed most of the run, then it creeped to 60 and I stopped. Meanwhile the boiler temp was constantly 5C higher, but upon depletion it started to create larger margin and ended up to all the way to 80C.

It has a slight yellow tint, but whatever causes it, contaminates everything in trace amounts. A bucket of washing water with traces of yellow in ground joints turns all the water to pale yellow. Sodium percarbonate seems to intensify the color largely.