Tertiary alcohols can be converted to tertiary carboxylic acids with CO and H2SO4 (Koch-Haaf hydrocarboxylation). The source of CO can be formic acid or oxalic acid. See attached document...

It'd be simple enough to convert this into the ethyl ester by working up with EtOH. It would be similar structurally to the propionic ester, except the the carbonyl carbon in the carboxyl would be bonded to the main loperamide structure as opposed to the ethyl.

It would be much easier to prepare the des-chloro compound without that benzylic hydroxyl to worry about.

So... this could also be used, perhaps, to convert ephedrine to methylphenidate analogs? Or would the nitrogen somehow interfere?

Also, this:

http://en.wikipedia.org/wiki/Carbon_monoxide_poisoning#Signs_and_symptoms

I don't understand you statement though: "It would be much easier to prepare the des-chloro compound without that benzylic hydroxyl to worry about."

Namely, being able to use reductions to strip off the chlorine that would otherwise remove the benzylic hydroxyl as well.

So... this could also be used, perhaps, to convert ephedrine to methylphenidate analogs? Or would the nitrogen somehow interfere?

So long as ephedrine doesn't rearrange, yes - but this is a hot and acidic environment. It would take experimental data or a review paper to be sure, but there's good reason to believe that the alpha methyl would just scoot over, providing an inactive product.

My preferred synthesis of deschloro-Lop.  Please read the paper if you're interested. 

To a 500 ml flask are added: 0.01 mol Lop HCl (5.135 g), ~250 ml EtOH/IPA (whichever or both, EtOH will make the reaction go a bit faster), (consider converting to the freebase either prior to reaction or while in solution with equal molar KOH before Pd/C-remember water generation for below calc's),1 equivalent of formic acid,  1 equivalent of KOH, .1 grams Pd/C 10% (not sure what mol% that actually worked out to, maybe could be less, the paper calls for 0.2% Pd).  Ideally, one wants the ratio of water to formate to be 3:1 but the water from the 90% formic acid plus the water from making the potassium formate is going to make the amount of water greater than for optimal reaction rate.  If one wanted to, they could just increase the KOH and formic acid to get the optimal ratio.  If one were using a larger amount of Lop HCl then eventually water would need to be added to get the 3:1 ratio.  Anywayz, heat to reflux, or if concerned about the hydroxyl, at least to a temp. where all materials are dissolved (>50-60 deg.C).  If using the optimized water:formate ratio and EtOH then the reaction would be complete by 1 hour of reflux.  If using IPA then it would be 93% complete at an hour. Factor in your solvent and water ratio and heating as needed.  (The paper doesn't call for reflux even, just 50 degrees, but in my case higher heat was needed for me to get everything dissolved.)

Jon- I told you once that I preferred potassium formate over ammonium formate because it was considered the best CTH hydrogen donor.  Methylammonium formate is going to be present in an imine reduction with an excess of the amine, methylamine in my case, but I would never recommend ammonium formates over potassium formates personally.

I don't know what to say except I have only done this deschloropropionyl-Lop synthesis ONE time.  Therefore, this is the same reaction where I used 1:1 potassium formate.  I had CO₂ generation.  If the HCl from the lop HCl neutralized the KOH which was added at the end then formic acid can, although not as well, do CTH on it's own.  Shit, the loperamide ammonium formate could have gotten the job done.  There's just too many unknowns.  At the same time that the aromatic Cl hydrogenolysis was going on, there was the benzylic tertiary hydroxyl could have been removed as well.  So while I normally like to use an excess of reagent and thus why I wrote 1.2 excess in the above instructions, you have helped me realize that in my actual synthesis I'm glad that there wasn't an excess (as well as the presence of a Lewis acid like FeCl₃ or AlCl₃)/ This conversation has actually given me hope that I didn't waste my propionyl chloride on a hydroxyl that was no longer there.  But yeah, I have only done this reaction once.  So, hopefully the 1:1 formate to Lop favored the hydrogenolysis of the aromatic Cl and left that hydroxyl well alone.     Please note the correction for the above (and I will go fix it now...), it may bee more prudent to keep the formate to halide ratio at 1:1.

BTW, indisputably I did get ROCKED by the deschloropropionyl-Lop once I got the dose high enough plus factoring in the methadone-like slow increase in effects...   And that was while I certainly had aftereffects from methadone still present in my neurons.  I'm sure if I had kept to classic opiates for a few weeks and then dropped the DCP-L it would have been even more pronounced.

jon managed to duplicate his sucess evidenced by the vigourous co2 evolution.
it subsided about an hour later.

so this is how this went .081
gm 95% formic acid and equimolar KOH 70 mg were mixed with 95 mg's of water
.00148 moles of loperamide base (.5gram) pd/c 5
5% pd/c 48 mg were added
were dissolved in 30 ml ipa
in and the reagents added and refluxed until co2 evolution stoped, ridiculously easy.
filtered of palladium i reclaimed the free based good's with some ultra dry rubbing alcohol,filtered and proceded to weigh my base it was quantitative.

fi

oh another trick for us poor fellows who cant afford pure loperamide is to get it in methanol and use hot methanol.
filter add about 4 volumes of water and make basic the bases with come crashing out as a white powder.
easiest extraction i ever done.
be sure to add a pinch of salt to salt it out and use a hirsh funnel to filter it, cone filters always plug up.