Shake and bake is the pits as far as meth manufacture goes. It really is, forget scraping the bottom of the barrel, that meth-od is re-using a barrel last in the employ of a lazy cunt thats prepared to simply grind up PFED pills and throw them in a birch and call the garbage that comes out the other end 'meth', then setting up a deep-sea oil well drilling rig over the top and going so far through the bottom of the barrel that once you've finished, assuming you do so in a manner which involves retention of one's eyeballs in the sockets that they came in when the cook was born, as opposed to having new ones punched through some or other bodypart which was never meant to host a pair of half-dissolved eyeballs propelled by the overpressure of an exploding PET bottle filled with a mixture of salts and solvents rapidly generating ammonia gas and pressure along with it, you've done it, and found you scraped so damn hard and deep that you've just won a race with the devil himself trying to tunnel his way out of the anus-end of perdition.
And the ironic thing is, the pressurization isn't needed. The birch-benkeser, which this scatophagous obscenity is an incompetent, dangerous as hell and altogether shite abortion of, can be run by just drying the gaseous NH3 and using anhydrous ammonia GAS, in anhydrous ether (thats what the starting fluid is for, I recognize the brand. Depending on the date it was purchased, that starting fluid was originally diethyl ether, some heptane and a little bit of very high boiling cylinder-head lubricant for when its actually used in cars, and then IIRC thats the one that changed to a mixture of diethyl ether, diisopropyl ether (the DIPE is MUCH more prone, btw, to peroxide formation, and believe you me, ether peroxides are no fucking joke. Sensitive as hell and absolutely notorious amongst organic chemists for their tendency to be the guilty party in events which involve things like fridges being turned into the equivalent of a shotgun blast, pointed at the ceiling and windows being blown out of rooms.
Along with, IIRC xylene and acetone, possibly some sort of alcohol, I'd have to check on the alcohol, pretty sure I've some of the later formulation around somewhere.
The undesired parts can be separated out via distillation (its alright to use the diisopropyl ether, in fact to toss it would be senseless when an etherial solvent system is being used, and diethyl ether is the usual solvent, its just not so safe to store, or to handle if you don't know what your doing with it due to peroxidation tendency, although the reducing environment of the Birch-benkeser would of course destroy peroxides as soon as they were formed)
And the acetone is the biggest nuisance, although unlike the ethers, its water soluble, and additionally, it forms a solid adduct with sodium bisulfite or metabisulfite (the latter being used in beer/wine making stores as a preservative) and then it can be left behind whilst the ether is fractionally distilled off.
As for pressure, anyhow, it isn't needed, nor is the ammonium salt required specifically to bee ammonium nitrate. ANY ammonium salt will do, one simply adds a stronger base, as a strong base added to a solution of a salt of a weaker base displaces the weaker base to form the corresponding salt of the stronger one.
(E.g adding sodium hydroxide to fr.ex ammonium sulfate or phosphate, sold in B&Q as fertilizer, would displace ammonia and leave behind sodium sulfate)
Although calcium oxide (quicklime) is a better base, as then only the powders need be mixed and a small quantity of H2O added to allow reaction, and the CaO doesn't release H2O as it reacts in the above displacement reaction, and its also a powerful dessicant (drying agent) so it helps dessicate the gaseous ammonia, which is then lead through a drying tube packed with CaO or other drying agent suitable for drying amines, and the ammonia, still in its gaseous form is passed through a length of plastic tubing, ideally using an aquarium airstone to disperse it and assist in uptake of the NH3 into the ether, in which it has some solubility. This has the pre-weighed quantity of lithium metal added (or sodium, potassium, either are less powerful reducing agents [whilst reactivity with water/air increases pretty drastically as one goes down the first column of the periodic table from lithium (least violence) to caesium (lets just say, whilst I've never done it myself, I've seen a video where some guy tossed some caesium metal into a bathtub full of water, outside, and the result was a tsunami, shot in an upwards direction and liberally mixed with the fragments of what was, in its fairly recent past incarnation, a bath tub; the trend as far as power as a reducing agent however is the reverse with lithium being the most powerful, and is capable of overreduction of the pseudo/ephedrine to form a cyclohexadiene or even if a lithium-flavoured pudding was massively over-egged in that respect, all the way to dearomatization to a cyclohexane ring, which would result instead in propylhexedrine)
Sodium or potassium are less powerful reducing agents and don't have the same tendency to overreduce the feedstock.
Anyhow, the metal of choice, finely divided as best possible, into little shreds is protected with some argon (not nitrogen, at least not with lithium, as Li is unusual in that its reactive with nitrogen gas, forming a very strong base, lithium nitride. And besides, argon is heavier and helps blanket the contents of the flask), and the tube dispensing the dried ammonia gas placed under the surface to bubble away for a few hours, its slower than the classic Birch-Benkeser as would be used in a professional chemistry lab, but it nevertheless forms the deep, dark blackish-blue of solvated electrons as the electride complex between the alkali metal and ammonia is formed, the solvated electrons are what does the reducing with this type of reaction.
The substrate to be reduced is then added, once the Li-blue or other alkali metal version is formed in the ether is done beeing formed, in the calculated quantity as a slurry in an alcohol to serve as a proton donor, and within moments the solvated electrons have reacted and reduced the feedstock, in the case of PSE/ephedrine, to meth. No pressure, no plastic bottles, doesn't go and blow your tits through your brainstem.
Although personally the route from benzaldehyde/nitroethane/amine base as acetate salt as catalyst (knoevanagel condensation) is far superior IMO, and its much better suited to scale-up, and this can even be done in an improved (vastly so) version whereby the heating isn't for -7 hours on a steam or hot water bath but carefully irradiated in a microwave oven for a total of about 30 minutes (for a synthesis of P2NP, 1-phenyl-2-beta-nitropropene on a half molar basis each of nitroethane and benzaldehyde using a 900 watt MW oven), a thermometer being kept in the flask, and just a bit of bog roll in the top, rather than a stopper, so pressure does not build up and blow the flask to hell, as if the thing WAS to be stoppered, it would either build up enough pressure to blast the stopper out like a bullet, or if the thing was keck-clipped (don't even think about it!) it would simply build up so much pressure it'd shatter and blast broken glass everywhere. The bog roll or kitchen towel is actually beneficial though, in that it helps to stop there being a release of P2NP vapors into the room, as these are irritating, like teargas. And after a simple workup (hint to get you started-without going into too much detail-, P2NP is more or less totally insoluble in water)
And from there, the uses are legion. Direct reduction to amphetamine with LAH, or with aluminium amalgam, or reaction at 80 'C (the MW rxn should be kept at a peak temp of 70 'C, takes about a half hour) for two hours with an excess of fine iron powder in glacial acetic acid, two and a half hours is also acceptable, then extracted with dichloromethane, and the DCM distilled off to give a crude P2P (phenylacetone, 1-phenyl-2-propanone) which is purified at least by first several washes of the DCM containing the P2P with 10% aqueous sodium hydroxide to remove some of the impurities, then preferably vacuum-distilled, or if this is impossible, then the hydroxide-washed DCM solution of ketone (4 washes IMO is about right, 5 isn't a bad thing either) washed with ice-cold H2O, and the DCM distilled off leaving behind the much higher boiling phenylacetone. A bisulfite adduct is then made if vac distillation can't be done and this, washed with various solvents and H2O, being pretty poorly soluble in many solvents, this being regenerated after this is done with a weak base such as sodium or potassium carbonate in water, and the freed P2P taken up into DCM again, and washed again, DCM removed, and then reductive amination using one of the legion of methods to be found online, or for straight amphetamine, formation of P2P ketoxime using hydroxylamine in alcoholic solution (using a base like sodium acetate to deprotonate the NH2OH HCl and form the freebase hydroxylamine in-situ, or else carbonate works for this also) and after rigorous drying of the oxime, (also insoluble in H2O) being taken up in anhydrous ethanol, under a cover of inert gas and the oxime reduced to the amine with sodium metal (Beauveault-Blanc reduction), affording amphetamine, the freebase of this then of course taken up in the solvent of choice (nonpolar) and a little isopropyl alcohol containing a small amount of sulfuric acid is added, this is tapped off, and once evaporated, the result is amphetamine, needing recrystallization, then ready to go.
For meth, or N-ethylamphetamine (which is rather nice, smoother than meth, not as weight-potent but the rough edges have been filed off, and its more potent than straight up phet), amalgam reductions with methylamine or ethylamine, or nitromethane or nitroethane respectively are usually used (these require hazardous mercuric [ Hg(II) ] salts and with amalgam reductions they don't scale too well, its very exothermic, finnicky and the Hg residues MUST be carefully removed via distillation, and washing of the amalgam with boiled water (the boiling is to de-gas it and removed dissolved O2) beforehand several times (warning-the washings contain highly toxic mercury salts) first to remove as much Hg as possible.
Reductive amination can also be done by forming the imine, using a suitable dessicant with the amine and the ketone, and then reducing the 1-phenyl-2-alkyliminopropane with borohydride, which is a lot cleaner, no highly noxious, environmentally filthy wastes either. Formation of the imine generates H2O, so dessicant is needed, as the imine is water sensitive, so it must be trapped as its formed. A lot nicer and less of a worry and a general cunt than Al/Hg amalgam reductions.