Manual Archive
Manual Archive
Benzaldehydes of Interest
a
Queensland Health Forensic and Scientific Services (QHFSS), 39 Kessels Rd, Brisbane
Australia
b
School of Biomolecular and Physical Sciences, Griffith University, Brisbane Australia
Forensic Chemistry, Queensland Health Forensic and Scientific Services, 39 Kessels Rd,
Brisbane 4108, Queensland, Australia
Introduction
The l-Phenylacetylcarbinol (l-PAC) process is a biotransformation process using the
fungus yeast. The mechanism, which was initially investigated by Neuberg et al.1,2,
involves glycolysis of glucose to produce pyruvic acid which is then decarboxylated
by pyruvate decarboxylase to produce acetaldehyde. The resulting acetaldehyde
then undergoes a condensation reaction with benzaldehyde to produce l-PAC [(R)-1-
hydroxy-1-phenylpropan-2-one]. During this process, by-products are formed due to
the action of alcohol dehydrogenase on benzaldehyde and l-PAC resulting in benzyl
alcohol and (1R,2S)-1-phenylpropan-1,2-diol (PAC-diol) respectively.3,4 (Refer to
Scheme 1)
O
Glycolysis
Glucose OH
H3C
O
Pyruvate
Pyruvate
decarboxylase
TPP, Mg++ CO2
O
O
OH
H3C H
Acetaldehyde 1
Benzoic Acid O OH
Pyruvate
CH3
decarboxylase
Oxidation H TPP, Mg++
O
Benzaldehyde 2 l-PAC 3
NADH + H+ NADH + H+
Alcohol Dehydrogenase
NAD+ NAD+
OH
CH3
OH
OH
Benzyl Alcohol 4 PAC-diol 5
Scheme 1: Mechanism of the fermentation of benzaldehyde to produce l-PAC and various by-
products3,4
The l-PAC process has previously been employed by the pharmaceutical industry
due to the ease with which l-PAC can be chemically converted to
pseudoephedrine/ephedrine. In recent years, a number of clandestine drug
laboratories have been located which have exploited the commercial process for the
production of pseudoephedrine/ephedrine for use in the manufacture of
methylamphetamine3.
2.2 Instrumentation
Gas Chromatography-Mass Spectrometry (GC-MS) analyses were performed on an
Agilent HP-6890N Network GC System using an Agilent HP-5MS capillary column
(30 m x 0.25 µm) fitted with an Agilent HP-5973 mass selective detector. The carrier
gas was helium at a constant flow of 1.0 mL/min and a split ratio of 25:1 with an
injection volume of 0.2 µL. The injector temperature was set to 100oC, with an initial
oven temperature of 100oC held for 1 min, then ramped at 30oC/min to 280oC and
held there for 10 mins. The mass selective detector operated between m/z= 40 and
450 in electron impact mode with an ionization energy of 70eV.
Nuclear Magnetic Resonance (NMR) analysis was carried out using a Varian 400
MHz Unity INOVA spectrometer operating at 400 MHz (1H) and 100 MHz (13C).
Samples were run in a mixed solvent system of d6DMSO and CDCl3 at 298 K. 1H
spectra were referenced to the d6DMSO solvent residual taken as 2.49 ppm and 13C
spectra were referenced to the d6DMSO solvent residual taken at 39.50 ppm at the
temperature quoted.
Table 1: Reagents and fermentation times used in each substituted benzaldehyde fermentation
Whilst our intent was to only perform minimal purification in order to observe
reaction marker by-products at each stage, the corresponding l-PAC products were
purified prior to NMR analysis by the bisulfite adduct purification technique
previously employed by Neuberg and Ohle5. Each crude sample was taken up in 10
mL diethyl ether and extracted with saturated sodium carbonate solution (2 x 5 mL)
to remove excess benzoic acid. The diethyl ether layer was then washed with
deionised water (2 x 5 mL), the aqueous extracts combined and further extracted
with diethyl ether (2 x 5 mL). The diethyl ether extracts were then combined and 15
mL of a freshly made saturated sodium hydrosulfite solution was added. The
solution was left to react for 60 mins, being shaken every few minutes. Any benzyl
alcohol in the sample was then removed through extraction with diethyl ether (3 x 10
mL). To the retained aqueous layer, solid sodium bicarbonate was added until the
evolution of carbon dioxide ceased. This aqueous layer was then extracted with
diethyl ether (3 x 10 mL), dried over anhydrous sodium sulphate, filtered and
evaporated on a rotary evaporator. A sample of each product was dissolved in
chloroform and assessed for purity by GC-MS prior to NMR analysis.
Excess benzyl alcohol that had formed as a by-product was then removed through an
acid-base extraction. The sample was acidified using 10 mL of 10% hydrochloric
acid, shaken briefly and extracted with dichloromethane (3 x 10 mL) to remove the
neutral components. The aqueous layer was then basified with a 10% solution of
sodium hydroxide in water and extracted with dichloromethane (3 x 10 mL). The
combined extracts were dried over anhydrous sodium sulphate, filtered and
evaporated on a rotary evaporator. A sample of each product was collected,
dissolved in chloroform and analysed by GC-MS and GC-IRD. Specific quantities of
each reagent can be found in Table 2.
Table 2: Reagents used in the reductive amination of each l-PAC analogue.
Starting Material
Sodium
Methylamine Ethanol
Grams Borohydride
Crude l-PAC (mL) (mL)
(% l-PAC (g)
analogues
Analogue)
6.672
5.6 3.934
l-PAC (21%) 8.4
(65 mmol) (104 mmol)
(9 mmol)
15.773
3,4-Methylenedioxy- 2.5 2.045
(10%) 15
l-PAC (29 mmol) (54 mmol)
(8 mmol)
10.769
1.5 0.900
4-Methoxy-l-PAC (19%) 15
(17 mmol) (24 mmol)
(11 mmol)
10.473
2.5 1.577
4-Fluoro-l-PAC (23%) 15
(29 mmol) (42 mmol)
(14 mmol)
2.025
2 1.726
4-Methyl-l-PAC (14%) 6
(23 mmol) (46 mmol)
(2 mmol)
11.526
3 3.185
4-Methylthio-l-PAC (14%) 20
(35 mmol) (84 mmol)
(8 mmol)
Starting Material
Hypophosphorous
Crude Grams Iodine
Acid
Pseudoephedrine/ (% Pseudoephedrine/ (g)
(mL)
ephedrine ephedrine
Analogue Analogue)
0.497
0.496 2
Pseudoephedrine/ephedrine (88%)
(2 mol) (37 mmol)
(3 mmol)
3,4-Methylenedioxy-α-[1- 0.531
0.456 2
(methylamino)ethyl]- (29%)
(2 mol) (37 mmol)
benzenemethanol (0.7 mmol)
4-Methoxy-α-[1- 0.201
0.507 2
(methylamino)ethyl]- (23%)
(2 mol) (37 mmol)
benzenemethanol (0.2 mmol)
4-Fluoro-α-[1- 0.83
0.959 2
(methylamino)ethyl]- (62 %)
(4 mol) (37 mmol)
benzenemethanol (3 mmol)
4-Methyl-α-[1- 0.192
0.103 2
(methylamino)ethyl]- (32%)
(0.4 mmol) (37 mmol)
benzenemethanol (0.3 mmol)
4-Methylthio-α-[1- 0.061
0.188 2
(methylamino)ethyl]- (45%)
(0.7 mmol) (37 mmol)
benzenemethanol (0.1 mmol)
Methcathinone has a tendency to form a pyrazine dimer when in the free base form6.
For stability purposes the analogues were converted to the corresponding
hydrochloride salt by the addition of hydrogen chloride gas to each of the substituted
methcathinone samples in chilled isopropyl alcohol. Once the pH of the solution had
reached pH 3 or lower, it was placed on a heater block and the solvent evaporated
under a stream of nitrogen.
Starting Material
Sodium
Grams Dichromate
Crude Pseudoephedrine/ (g)
(% Pseudoephedrine/
ephedrine Analogue
ephedrine Analogue)
0.450
0.613
Pseudoephedrine/ephedrine (88%)
(2 mmol)
(2 mmol)
3,4-Methylenedioxy-α-[1- 0.843
1.032
(methylamino)ethyl]- (29%)
(3 mmol)
benzenemethanol (1 mmol)
4-Methoxy-α-[1- 0.257
0.673
(methylamino)ethyl]- (23%)
(2 mmol)
benzenemethanol (0.3 mmol)
4-Fluoro-α-[1- 0.167
0.526
(methylamino)ethyl]- (55%)
(2 mmol)
benzenemethanol (0.5 mmol)
4-Methyl-α-[1- 0.220
0.297
(methylamino)ethyl]- (16%)
(1 mmol)
benzenemethanol (0.2 mmol)
4-Methylthio-α-[1- 0.106
0.516
(methylamino)ethyl]- (45%)
(2 mmol)
benzenemethanol (0.2 mmol)
H
R1
Substituted
Benzaldehyde
Fermentation
OH
CH3
R1
O
Substituted
l-PAC
Reductive Amination
OH
CH3
R1
HN
CH3
Substituted
Pseudoephedrine/Ephedrine
Reduction Oxidation
O
CH3 CH3
R1 R1
HN HN
CH3 CH3
Substituted Substituted
Methylamphetamine Methcathinone
Scheme 2: The substituted benzaldehydes of interest investigated and the pathway followed.
3.1 Fermentation
Initially, a number of small scale fermentations were conducted to examine the l-
PAC process parameters using benzaldehyde. The parameters of buffer, buffer
volume, temperature and acetaldehyde addition were examined. It was found that a
pH 5 0.1 M citrate buffer made from dissolving trisodium citrate and citric acid in
deionised water gave a higher yield of l-PAC than when a deionised water buffer
was employed. The buffer volumes were then altered to produce a benzaldehyde
concentration of 12 g/L and 6 g/L with the 6 g/L fermentation broth found to
produce the highest l-PAC yield. It was also found that heating the fermentation
broth to 30oC in a water bath as opposed to being left at room temperature
contributed to an increased l-PAC yield. Acetaldehyde addition at various ratios of
1:1 benzaldehyde to acetaldehyde and 1:2 benzaldehyde to acetaldehyde also
produced an increased l-PAC yield with each substituted benzaldehyde's preferred
ratio noted.
Substituted Calc'd
l-PAC Analogue
Benzaldehyde R1 % MS IR
Produced
Fermented Yield
164,
2872 (vC-H al), 2933 (vC-H al), 1363
1-Hydroxy-1-(4- 121,
(δC-H al), 3028 (vC-H ar), 1730
4-Methyl methylphenyl)-2- 4- CH3 21.3% 93, 91,
(vC=O), 3446 (2vC=O), 3508 (vO-H),
propanone 77, 65,
3651 (vOH:O=C), 1176 (vC-OH)
43
H
H O b
a' OH
H CH CH3 b'
b' a c' a' a b CH3
c
c
O d'
F H f' O
d' F
H e'
c'
Figure 1: The protons (left) and the carbons (right) of the 4-fluoro analogue of l-PAC.
1
H NMR (400 MHz, DMSO-d6) δ 7.38-7.28 (m, 2H, Ha', Hd'), 7.05-6.91 (m, 2H,
Hb', Hc'), 5.63 (d, J= 4.3 Hz, 1H, Hb), 4.99 (d, J= 4.2 Hz, 1H, Ha), 3.19 (s, H2O),
2.00 (s, 3H, Hc);
13
C NMR (100 MHz, DMSO-d6) δ 207.88 (Cb), 161.87 (d, J= 245.6, Cd'), 134.81
(Ca'), 128.30 (Cb', Cf'), 115.06 (Cc', Ce'), 78.43 (Ca), 24.7 (Cc) (refer to Figure 1 for
proton and carbon labels).
H
H O b
a'
H CH CH3
b' a c
O
H3C H
d d'
H
c'
To confirm the production of 4-methyl-PAC 22, the sample which was purified via
the bisulfite adduct method was analysed by 1H NMR; 1H NMR (400 MHz, DMSO-
d6) δ 7.20-7.11 (m, 2H, Ha', Hd'), 7.11-6.99 (m, 2H, Hb', Hc'), 5.14 (d, J= 4.3 Hz,
1H, Hb), 4.95 (d, J= 4.2 Hz, 1H, Ha), 2.25 (s, 3H, Hd), 1.98 (s, 3H, Hc) (refer to
Figure 2 for proton labels). A gCOSY (gradient correlation spectroscopy) spectrum
was also obtained to confirm the assignment of each proton signal.
Figure 3: The protons (left) and the carbons (right) of the 4-methylthio analogue of l-PAC.
1
H NMR (400 MHz, DMSO-d6) δ 7.24-7.20 (m, 2H, Ha', Hd'), 7.18-7.08 (m, 2H,
Hb', Hc'), 5.28 (d, J= 4.2, 1H, Hb), 4.95 (d, J= 4.0 Hz, 1H, Ha), 2.39 (s, 3H, Hd),
1.99 (s, 3H, Hc);
13
C NMR (100 MHz, DMSO-d6) δ 138.02 (Cd'), 134.94 (Ca'), 126.83 (Cb', Cf'),
125.84 (Cc', Ce'), 78.93 (Ca), 24.58 (Cc), 14.91 (Cd) (refer to Figure 3 for proton
and carbon labels).
An additional peak at δ 207 was assigned to the carbon of the aldehyde group of 4-
(methylthio)benzaldehyde but was suspected to also be masking Cb of the 4-
methylthio analogue of l-PAC. This assignment was based upon the chemical shifts
that Cb was found at in the previous 13C NMR's of l-PAC and the 4-fluoro analogue
of l-PAC. Additional peaks were also encountered in the 1H NMR at δ 9.82, 7.69,
7.26 and 2.46 and the 13C NMR at δ 132.25, 129.37, 147.31, 124.57 and 13.98 which
were attributed to remaining 4-(methylthio)benzaldehyde in the sample.
The bisulfite adduct purification technique employed for the purification of the
fermentation products was accompanied by a number of problems such as loss of
large amounts of starting material and carryover of substituted benzaldehyde. As a
result, samples pure enough for NMR analysis could not be obtained for the 3,4-
methylenedioxybenzaldehyde and 4-methoxybenzaldehyde fermentation products.
The assignment of these compounds was therefore made using GC-MS and GC-IRD
analysis.
l-PAC
Pseudoephedrine/ Calc'd
Analogue
ephedrine Analogue R1 % MS IR
Reductively
Produced Yield
Aminated
In each reductive amination where the crude l-PAC analogue used contained
substituted benzaldehyde, the corresponding N-methyl-benzylamine analogue was
produced, formed through the amination of residual benzyl alcohol. The presence of
N-methyl-benzylamine analogues can be used as an indicative marker that the
product was synthesised from an initial substituted benzaldehyde fermentation
followed by reductive amination.
Pseudoephedrine/ Methylamphetamine
Calc'd %
ephedrine analogue Analogue R1 MS IR
Yield
reduced Produced
H
a' H b' a
H H2C b CH3 a' b CH3
b' a c c'
c
HN d'
NH
F H d CH3 F f'
CH3
d' e e'
H d
c'
Figure 4: The protons (left) and the carbons (right) of the 4-fluoro analogue of
methylamphetamine.
1
H NMR (400 MHz, DMSO-d6) δ 3.14-3.05 (m, 1H, Hb), 3.03-2.55 (m, 2H, Ha),
13
2.48 (s, 3H, He), 1.07 (d, J= 6.3 Hz, 3H, Hc); C NMR (100 MHz, DMSO-d6) δ
162.76-159.75 (Cd'), 132.72 (Ca'), 130.39 (Cb', Cf'), 114.86 (Cc', Ce'), 55.69 (Cb),
31.13 (Cd), 16.18 (Cc) (refer to Figure 4 for proton and carbon labels).
A proton signal for proton d was not observed in the 1H NMR spectrum. It is
proposed that this signal would have been present at approximately 9-10 ppm,
however, this cannot be confirmed as this region was not viewable in the 1H NMR
spectrum.
The shifts assigned to the aromatic proton signals would also have been contributed
to by the aromatic protons of N-methyl-4-fluorobenzylamine. It was extremely
difficult to separate the signals of these two compounds due to their overlapping
nature. Therefore δ in the region of 7.37-6.83 were attributed to the aromatic H's of
both 4-fluoromethamphetamine and N-methyl-4-fluorobenzylamine.
Pseudoephedrine/ Methcathinone
Calc'd
ephedrine analogue Analogue R1 MS IR
% Yield
reduced Produced
H O O
a' H b'
H a CH3 a' a b CH3
b' b c'
c
HN d'
NH
F H c CH3 F f'
CH3
d' d e'
H d
c'
Figure 5: The protons (left) and the carbons (right) of the 4-fluoro analogue
of methcathinone
1
H NMR (400 MHz, DMSO-d6) δ 9.69 (s, 1H, Hc), 8.07-7.97 (m, 2H, Ha', Hd'),
7.27-7.14 (m, 2H, Hb', Hc'), 5.05 (dt, J= 7.3, 5.9 Hz, 1H, Ha), 2.57 (q, J= 4.6, 3.9 Hz,
3H, Hd), 1.47 (d, J= 7.2 Hz, 3H, Hb);
13
C NMR (100 MHz, DMSO-d6) δ 194.08 (Ca), 167.08 (Cd'), 131.43 (Cb', Cf'),
129.06 (Ca'), 115.89 (Cc', Ce'), 58.18 (Cb), 30.64 (Cd), 15.38 (Cc) (refer to Figure 5
for proton and carbon labels). A number of additional small peaks were also
observed in the 1H NMR and were attributed to impurities such as N-methyl-4-
fluorobenzylamine.
H O
a' H
H a CH3
b' b
HN
H3C H c CH3
e d' d
H
c'
1
H NMR (400 MHz, DMSO-d6) δ 9.32-9.22 (m, 1H, Hc), 7.79 (d, J= 8.2 Hz, 2H,
either Ha' and Hd' or Hb' and Hc'), 7.22 (d, J= 8.0 Hz, 2H, either Ha' and Hd' or Hb'
and Hc'), 4.99 (q, J= 6.5 Hz, 1H, Ha), 2.59 (q, J= 6.4, 5.8 Hz, 3H, Hd), 2.31 (s, 3H,
He), 1.49 (d, J= 7.0 Hz, 3H, Hb) (refer to Figure 6 for proton labels ).
Shifts at 6.80, 3.81 and 1.01, and 9.45, 7.34, 7.07, 3.95, 2.46 and 2.21 were also
1
observed in the H NMR, attributed to isopropyl alcohol and N-methyl-4-
methylbenzylamine.
It was anticipated that difficulties would be had during the reduction of the 3,4-
methylenedioxy and 4-methoxy analogues of pseudoephedrine/ephedrine due to the
tendency of hydriodic acid to cleave ethers and less commonly thioethers. An
absence of the expected hydroxyl substituted methylamphetamine products was
attributed due to deprotonation during basification and subsequent loss during
extraction.
4.0. Conclusion
The fermentation of the substituted benzaldehydes was found to be a viable route for
the production of the corresponding l-PAC analogues, except in the case of 3,4-
methylenedioxybenzaldehyde. Reductive amination of these l-PAC analogues was
able to generate viable amounts of the 4-fluoro and 4-methyl analogues of
pseudoephedrine/ephedrine but was unsuccessful at producing useable quantities of
the 3,4-methylenedioxy, 4-methoxy and 4-methylthio analogues.
In conclusion, this manufacturing pathway was found to be a viable route for the
synthesis of the 4-fluoro and 4-methyl analogues of methylamphetamine and
methcathinone. Whilst a reasonable yield of the 4-methylthio analogue of
methylamphetamine could be produced following reduction of the corresponding
pseudoephedrine/ephedrine analogue, synthesis of the pseudoephedrine/ephedrine
analogue was difficult. Therefore, this manufacturing pathway was deemed
inadequate for the production of the 4-methylthio analogue of methylamphetamine
and methcathinone as well as for the manufacture of the 3,4-methylenedioxy and 4-
methoxy analogues of methylamphetamine and methcathinone.
Acknowledgments
The authors which to acknowledge the contribution of Peter Vallely of the Australian
Crime Commission for constructive comments during the development of this
project, and Dr Sue Boyd of Griffith University for NMR analysis.
References
1. Neuberg, C.; Hirsch, J., An enzyme which brings about union into carbon
chains (Carboligase). Biochemische Zeitschrift 1921, 115, 282-310.
2. Neuberg, C.; Liebermann, L., Carboligase II. Biochemische Zeitschrift 1921,
121, 311-325.
3. Cox, M.; Klass, G.; Koo, C., Forensic Aspects of the Biotransformation of
Benzaldehyde used in the Synthesis of Methamphetamine, Part 1: Reaction
Conditions, Stereochemical Outcomes, and the use of Other Substituted
Benzaldehydes. Journal of the Clandestine Laboratory Investigating
Chemists Association 2009, 19 (4), 20-37.
4. Shukla, V.; Kulkarni, P., L-Phenylacetylcarbinol (L-PAC): biosynthesis and
industrial applications. World Journal of Microbiology and Biotechnology
2000, 16 (6), 499-506.
5. Neuberg, C.; Ohle, H., Carboligase IV. Biosynthetic carbon chain union in
fermentation process. Biochemische Zeitschrift 1922, 128, 610-618.
6. Szendrei, K., The Chemistry of Khat. Bullentin on Narcotics 1980, XXXII
(3).
7. Shin, H.; Rogers, P., Production of L-phenylacetylcarbinol (L-PAC) from
Benzaldehyde Using Partially Purified Pyruvate Decarboxylase (PDC).
Biotechnology and Bioengineering 1996, 49 (1), 52-62.
8. Overman, L., Organic Reactions, Volume 59. John Wiley & Sons, Inc.:
Hoboken, New Jersey, 2002.
9. UNODC, Amphetamines and Ecstasy: 2011 Global ATS Assessment. United
Nations Office on Drugs and Crime, United Nations Publication, Sales No.
E.11.XI.13.