http://pubs.acs.org/cgi-bin/article.cgi/jcchff/2002/4/i06/pdf/cc0200282.pdf
J. Comb. Chem. 2002, 4, 576-583
Use of Statistical Design of Experiments in the Optimization of Amide
Synthesis Utilizing Polystyrene-Supported N-Hydroxybenzotriazole
Resin
Owen W. Gooding,* Lanchi Vo, Sukanta Bhattacharyya, and Jeff W. Labadie
Two fields that routinely perform reaction optimization studies are chemical development (prior to scaleup)
and medicinal or combinatorial chemistry (prior to analogue synthesis or library production). To date,
the use of statistical design of experiments (DoE) in conjunction with automated synthesizers has been
applied in process research to a greater extent than in the medicinal or combinatorial laboratories. We have
applied DoE in conjunction with an automated synthesizer to optimize the synthesis of amides employing
resin-bound N-hydroxybenzotriazole (PS-HOBt) active esters as intermediates. This methodology allowed
the rapid development of an improved protocol for the parallel synthesis of amides by conversion of carboxylic
acids to PS-HOBt esters followed by treatment with appropriate amines. Product isolation involved only
simple filtration and evaporation.
In summary, statistical analysis of this data set revealed
the following (in order) as the only significant variables: (1)
solvent ratio, with lower DMF/DCM ratios preferred; (2)
amount of DIC, with 4.4 equiv giving maximum loading.
Time and acid type are predicted to be insignificant. It is
interesting to note that an amount greater than 4.4 equiv of
DIC is deleterious to the reaction. This goes against
conventional “solid-phase” chemistry wisdom that a large
excess is always better. It was gratifying to see that the results
were more or less the same with either acid type (alkyl or
aryl). This was desirable from the standpoint of a robust
process for library synthesis. The optimum conditions for
realizing maximum resin loading were predicted to be 4.4
equiv of DIC, 20:80 DMF/DCM, and reaction time in the
range of 1-12 h.
yada yada yada...
Table 8. Optimal General Conditions for Amide Synthesis
time, h 2
solvent ratio 20:80 DMF/DCM
amount of carboxylic acid, equiv 1.5
amount of DIC, equiv 4.4
order of addition acid before DIC
looks nice.
experimental:
carboxylic acids: 0.375M in DMF, 10.00 mL
DIC: 20.823 g of 1,3-diisopropylcarbodiimide
in 100.00 mL of DCM (1.60 M)
DMAP: 2.199 g of
4-(dimethylamino)pyridine in 400.00 mL of dichloromethane
(0.044 M)
amine + cat.: 0.5626 g of benzylamine and 1.9388 g of DIEA
in 100.00 mL of THF (0.0525 M benzylamine and 0.150 M
DIEA), 0.8518 g of 1-phenylpiperazine and 1.9388 g of
DIEA in 100 mL of THF (0.0525 M 1-phenylpiperazine and
0.150 M DIEA), 0.4889 g of aniline and 1.9388 g of DIEA
in 100.00 mL of THF (0.0525 M aniline and 0.150 M DIEA).
reaction vessels charged with 0.15 g of PS-HOBt(HL) resin (~0.15
mmol)
The DMAP solution (2.00 mL) was added first followed by the appropriate acid
solution (0.60 mL) and the DIC solution (0.40 mL). The ester
formation was allowed to proceed with agitation for 3 h at
25 °C, and the vessels were
automatically drained and washed three times with 4 mL
portions of DMF, THF, DMF, and DCM to remove side
products and unreacted starting materials.
Coupling was conducted by adding the amine/
DIEA solution (2.00 mL) for 3 h at 25 °C.
The products were collected and concentrated in vacuo
(Genevac) at 25 °C. The residues were weighed to determine
mass yield and were characterized by HPLC (purity) and
1H NMR (identity).
I was kind of pressed for time when i shat this out... but if any bees wanna turn this into a large scale LSA to LSD synth, BEE (pun intended) my guest.
