PBr3 and SOCl2

by James

in Alcohols, Organic Chemistry 1, Organic Chemistry 2

Lately we’ve been talking a lot about converting alcohols to good leaving groups [halides] [tosylates and mesylates] You might think we’d be done already, but not yet! In today’s post we talk about a third important method for converting alcohols to good leaving groups – by using the reagents phosphorus tribromide (PBr3) and thionyl chloride (SOCl2) .


Making Alcohols Into Good Leaving Groups, Part Three.

[Before we get too far into this, let me say that there’s some differences as to how the mechanism of the reaction of SOCl2 with alcohols is taught. Most schools teach inversion, but it is also (rarely) taught as retention via a different mechanism. I cover that whole discussion here. ]

So far we’ve covered two different ways of making alcohols into good leaving groups.

Conversion of alcohols to alkyl halides with strong acid. This works well for tertiary alcohols when nothing “bad” can happen (i.e. no side reactions). However, when certain secondary alcohols are used, rearrangements can occur.

Conversion of alcohols into tosylates or mesylates – here, we break O-H and “cap” the oxygen with a “sulfonyl” group (“tosyl” and “mesyl” are popular choices). Very simple. No rearrangements.   This does not affect the stereochemistry.

2-pbr3 and socl2

So why might we need more than these two ways to make alcohols to good leaving groups? Isn’t two methods enough? 

Fair question!

We mentioned that strong acid (HCl, HBr, HI) can lead to rearrangements with certain secondary alcohols. So an alternative that doesn’t lead to rearrangements would be useful from that perspective. Secondly, strong acid is a pretty blunt instrument, like a sledgehammer. It gets the job done, but can lead to some collateral damage if you have a molecule containing functional groups with various levels of acid sensitivity (esters, alkenes, alkynes). Using a milder, more targeted reagent would help us avoid undesired side reactions in more complex situations.

A harder point to address is this: why not just, for example, always make alcohols into mesylates or tosylates if we want to make them good leaving groups? This is actually a great idea most of the time!  As for exceptions, I can think of at least one situation where when you would need to make a halide. For example, if you haven’t already, you will learn about Grignard reagents at some point. These can be made from alkyl halides but not from mesylates or tosylates, so an alternative to what we’ve already learned is good to know.

OK. Let’s dig in.

Phosphorus Tribromide (PBr3) and Thionyl Chloride (SOCl2)

The reagents we’ll talk about today are thionyl chloride (SOCl2) and phosphorus tribromide (PBr3). These are two representatives of a family [note 1] of reagents that can convert alcohols to alkyl halides (Later on, when you learn about carboxylic acids, you’ll see that these can also be used to convert carboxylic acids to acyl halides).

Here’s examples of each of these reagents in action.

3-pbr3 in action

What do you notice?

  • First of all, check out the bonds formed and bonds broken: break C-OH, form C-Br or C-Cl
  • Note the change in stereochemistry. Both occur with inversion.
  • Note the lack of rearrangement. Had we used HCl or HBr, it would have led to a ring expansion.

Nice and clean way to convert alcohols to alkyl halides.

How Do They Work? Activation, Then Substitution

So how do they work? Let’s look at PBr3.


This reaction proceeds in two steps that you can think of as “activation” and “substitution”. In the “activation” step,  the alcohol is converted into a good leaving group by forming a bond to P (O-P bonds are very strong) and displacing Br from P [note that this is essentially nucleophilic substitution at phosphorus].

Now that the oxygen has been “activated” (i.e. converted to a good leaving group) a substitution reaction at carbon can occur. The bromide ion that was displaced from phosphorus attacks carbon via backside attack, forming C-Br and breaking C-O and we are left with a new alkyl bromide (with inversion of configuration) and the Br2P-OH leaving group.


The reaction of thionyl chloride with alcohols similarly goes through an “activation” step and a “substitution” step. In the first step, oxygen attacks sulfur, displacing chloride ion. In the second step the chloride ion attacks carbon in an SN2 reaction, leading to inversion of configuration. [Note 2]

For our purposes, the mechanism ends here, but it’s worth noting that the sulfur byproduct (HO-S(O)-Cl) can further break down to SO2 gas and HCl through the mechanism shown [not dissimilar to the breakdown of carbonic acid to CO2 and water]. Removal of SO2 from the reaction vessel renders this reaction irreversible and helps drive the reaction to completion.


[I recall TA’ing a lab where a student dropped a round bottom flask with 5 mL of SOCl2 into a rotovap bath – there was immediate bubbling and the stench of SO2 made us have to evacuate the entire lab of about 120 people outside for fresh air. We were lucky it was a pleasant day and not in the depths of Montreal’s epic winters]

The process shown works well for primary and secondary alcohols. A process that goes through an SN2 mechanism shouldn’t work so well for tertiary alcohols.  I find textbooks extremely vague as to how they cover the use of these reagents with tertiary alcohols, so I’m not going to go into more detail on this point. [note 3]. Ask your instructor.

The bottom line for today is to learn about these two methods for converting alcohols into alkyl halides, and pay particular attention to their stereochemistry. Extremely testable! 

I think that’s about all we have to say about converting alcohols to good leaving groups!

Next time – Elimination

There’s just one more thing here. We’ve finished covering substitution reactions of alcohols. But what about elimination reactions of alcohols? How would we go about making alkenes? (aka “dehydration”). Many of the steps will look familiar – but there will be new wrinkles too.

Next Post – Elimination Reactions Of Alcohols


[Note 1] By family of reagents, I mean that there are related reagents that go through the same mechanisms , that we won’t talk about today (PCl3, SOBr2, PCl5, PBr5) ]

[Note 2] again, things in “real life” are a bit more complicated. You might want to double check that your instructor follows this mechanism. If not, check out this post on the SNi mechanism.

[Note 3] March mentions that SOCl2 can be used to convert tertiary alcohols to tertiary alkyl chlorides. So “in the lab”, things are a bit more complex than alluded to here.

Related Posts:

{ 5 comments… read them below or add one }


Woww. Good article :) I have a couple of questions though.
1. Let’s say we have a 2-bromopentane, can SOCl2 (and PCl3) replace the Br substituent with Cl?
2. What software do you use to produce those nice structures? :))



In my book it says that SOCl2 resembles SN1 and that there is retention of configuration.



It’s called SNi (nucleophilic substitution with internal return). Post here: https://www.masterorganicchemistry.com/2014/02/10/socl2-and-the-sni-mechanism/



What accounts for the difference in step 2 btw this rxn and the mesylate/tosylate route for making alkyl halides from alcohols (i.e., why does the free Cl- deprotonate in the latter but substitute in the former)?



That’s an excellent question and one that isn’t immediately obvious. One thing I neglected to add in the MsCl / TsCl example is that a weak base (e.g. pyridine) is usually added to mop up any HCl formed, which could prevent the substitution for occurring.
Another is that in the case of the PCl3, once the phosphorus attacks oxygen you could have formation of a partial P-O double bond, which would put a formal charge of +1 on oxygen and make it a better leaving group, more easily displaced by a halide.
I’m sorry if this seems like an unsatisfying answer!


Leave a Comment