Sciencemadness Discussion Board

Is it normal to get low yields all the time, or am I a bad organic chemist?

Cou - 24-12-2020 at 20:23

My yields tend to be around 50%. I got a 42% yield making isobutyl bromide from isobutyl alcohol using red P and Br2, even though the Vogel's practical organic chemistry procedure says 90% is possible. It's still fun to make stuff, but this isn't good for multistep synthesis.

DBX Labs - 24-12-2020 at 20:39

You simply haven’t prayed hard enough to the gods of organic chemistry and the deities of good yields (Peace be upon them).

Idk there’s a good chance that the fault isn’t on you as much as the fact that a synthesis with multiple reagents has reasonably more variables to it, hence more areas for imperfection. You might have all the measurements right, but it’s likely that the people that reported a 90% yield have done the synthesis countless times, making personal tweaks each time.

Take my words with a grain of salt.

UC235 - 24-12-2020 at 20:54

Depends what you're doing and scale. Mechanical losses on small preparations can cause large drops in yield. Sometimes 50% yield is fantastic. Sometimes 80% is terrible. We do this for fun. Optimization is rarely the goal like it is for industry or (sometimes) for academic purposes. Vogel's is largely cribbed from Orgsyn, I prefer going to the source.

http://www.orgsyn.org/demo.aspx?prep=CV2P0358

55-60% with distilled PBr3 on a 7mol scale. I think 42% sounds just fine. The HBr (or NaBr)/H2SO4 method ( http://www.orgsyn.org/demo.aspx?prep=cv1p0025 ) method is often very high yielding but only works on simple alcohols well that do not suffer from competing elimination or carbocation rearrangements. I can get 90% yield of ethyl or isopropyl bromide with the latter, but isobutyl bromide would be a bit hairy. I'd expect a lot of isobutylene to be formed and contamination with tert-butyl bromide.

teodor - 26-12-2020 at 07:32

@Cou, I think what is most important is to understand why the yield is low in each particular case. Think about it as about the first thing to do to become a better chemist.
Some procedures from Vogel book always give a good yield, like esters. But some others like sec-alcohol chlorides with HCl or ethers require some lessons to be learned. In this sense some of the purpose of any educational book is to rise questions which motivate to self exploration of the subject. It would be not educational at all if everything will be just by a receipt. I think the most important aspect of this particular book is what is out of its scope.

To understand why yield is poor we need to start experimenting doing the same procedure several times with variations. The book also has a lot of cognate preparation to study reaction with variations.

By my personal experience 2 factors can greatly change the yield:
1. Mixing. Most reactions will not give anything except wetting of flasks walls if you don't stir it quite energetically. Also it can result in wrong product because changes temperature distribution also.
2. Separation/purification without understanding principles of particular method. You can get enough product but the purification procedure you follow is not perfect or doesn't work in your case for some reason (see, for example my post about di-n-butyl ether yield: https://www.sciencemadness.org/whisper/viewthread.php?tid=84...)

[Edited on 26-12-2020 by teodor]

And about P + Br2 method, it is not as straightforward as HBr that is why it follows HBr method in the book (as well as those sec-alcohol chlorides which allow you to study how mixing is important). I studied the topic a little, you can find in Internet research about yield of this particular procedure with conclusion that it is much better to get PBr3 first and then use it in the reaction. That study contained some consideration why P + Br2 method can give a low yield. One of the point is requirement about quite dry reagents and glassware which requires some additional efforts.


[Edited on 26-12-2020 by teodor]

Maurice VD 37 - 26-12-2020 at 09:02

I have the tendency to think that Vogel's book reports exaggerated yields. Personally I have always found lower yields than published in Vogel's book.


[Edited on 26-12-2020 by Maurice VD 37]

arkoma - 26-12-2020 at 11:00

Cou, don;'t lose heart. at times i don't get ANY yield. Carbon is a complicated mess.

Texium - 26-12-2020 at 14:21

A certain part of it comes down to experience as well. During undergrad, there were a few times my PI would run the same reaction using the same reagents as me, and he might get a ~70% yield while I would get ~50%. Even though I was trying my best, there was still an experience gap between us of over 30 years, and he had internalized many subtle techniques for minimizing mechanical losses and knowing exactly where to cut the column fractions that can't be learned from anything besides decades of practice.

Also, keep in mind that in published procedures, they probably ran the reaction numerous times, and published the best results of the bunch. It doesn't mean that the stated yield is what they got on their first try, or that it's consistently attainable. It could even be a fluke. They may have run the synthesis five times and gotten 55, 67, 75, 90, 76 for their yields. That doesn't stop them from reporting a 90% yield.

teodor - 27-12-2020 at 04:46

From Vogel's textbook:

"
2P + 10ROH + 5Br 2 — • 2H 3 P0 4 + lORBr + 2H 2 0
The reaction is of general application : with primary alcohols (n-propyl to
n-hexadecyl) the yields are over 90 per cent, of the theoretical, but with
secondary alcohols the yields are 50-80 per cent.; in the latter case a small
quantity of high boiling point by-product is also formed which can, however,
be readily removed by fractional distillation. The reaction is conveniently
carried out in a special all-glass apparatus."

So, it is not 90%, it is 50-80.

But where are your losses? Is it that high boiling point by-product (how much did you get @Cou in the high boiling fraction)?
Is it impurity in P? Is it escaping of reagents? Can you titrate the residue to measure H3PO4 contents?

If you combine both you can get less than 50%. Also some substance escaped through condenser on your video, you missed the hole to put your phosphorus in (and I see it as black, not red on your video, is it so impure? by the way you don't need a glass funnel, use just PP "powder" funnel, it's dirty cheap, also pour the alcohol after P in the same funnel to get all your P inside the reaction flask), also you poured bromine not exactly inside the reaction, also you didn't use the special construction of the equal pressure funnel with a condenser on a back line, so why do you ask?

You mentioned the reaction was very exothermic, so did you try to put it on magnetic stirrer to dissipate the heat on the whole volume or regulate the speed of bromine addition?

I think your question has answers.

Update: I see, iso-butanol is a primary alcohol, I was thinking by iso-propanol analogy... What was happen to me once - I was making sec-butyl chloride and got very poor yield. I've made a check of the alcohol (from "ZEOS", marked as "pure") by fractional distillation and found that it contains quite a big fraction of bp 87-91C (2-butanol should be 98-100C).

[Edited on 28-12-2020 by teodor]

Fyndium - 27-12-2020 at 06:38

Sometimes a common sense approach has saved me from low yields or disasters. Such things as solubilities, temperatures, extraction solvents and reagent purity have played a role. I have recently been getting textbook yields from most all of my reactions, some have even surprised me giving a higher yield than expected.

Overheating can cause side reactions, stuff evaporates off, and in extreme cases, stuff foam or even eject out of the vessel. Underheating can stall the reaction, for example one reaction has an intermediate which is produced at all temperatures and appears similar to the final, but the final product needs a lot of heat.

About stirring, I've seen many processes describe as "stir occasionally", while I generally stir my reactions as fast as my magstir can hold up, and now that I got my overhead working, I can really push them if I need, especially the more voluminous or viscotic stuff.

One hydrolysis reaction, for example, will give exactly twice higher yield (quantitive, to be exact) when stirred for 36 hours instead of 2. The difference is huge, and I suppose the majority of the hydrolysis would occur within the prime hours, but that was the result.

Solubility of stuff in reaction mass, for example crashing stuff out: small solubility can become a big loss factor when the total volume is high. This is a case in potassium chlorate making, as was discussed in it's thread.

HeYBrO - 11-1-2021 at 14:46

Quote: Originally posted by Texium (zts16)  


Also, keep in mind that in published procedures, they probably ran the reaction numerous times, and published the best results of the bunch. It doesn't mean that the stated yield is what they got on their first try, or that it's consistently attainable. It could even be a fluke. They may have run the synthesis five times and gotten 55, 67, 75, 90, 76 for their yields. That doesn't stop them from reporting a 90% yield.


Exactly right, as much as I love the fact that organic chemistry is a bit of an art sometimes as well as a science, it completely eludes me how as a field (especially in synthesis or methodology) there is a distinct lack of any reproducibility studies given in papers. It is very rare for an author to have run a reaction in a paper 3+ times and then give a measure of spread. Biologist are better at it than we are!

Whilst I won't comment on Tomas Hudlicky's recent Angewandte, he does have some constructive papers discussing reproducibility in chemistry. This has of course been discussed on this amazing forum before! http://www.sciencemadness.org/talk/viewthread.php?tid=17621

macckone - 12-1-2021 at 14:33

Practice helps a lot.
But some synthesis are poor yielding no matter what, regardless of what the published results say.

For example a friedel-crafts alkylation with aluminum chloride.
An ideal yield is often just over 30% or so.
That assumes all reagents are bone dry and the aluminum chloride is fresh.
A trace of moisture can reduce that yield by 50% depending on the substrate and chloro compound.

Inorganic chemistry tends to be more predictable until you get into coordination complexes.
Then all bets are off.

Keeping a sacrificial rubber chicken around for a ritual to the chemistry gods cannot hurt.
Unless you get nitric acid on it and it catches fire .. that could hurt.