Exact Mass GC-MS Analysis of Amine Monomers Used in Combinatorial Library Production

In this application note, monomer confirmation using the GCT orthogonal Time-of-Flight (TOF) MS detector combined with OpenLynx high throughput software is described. The GCT produces exact mass data for elemental composition calculation and strucural elucidation. OpenLynx provides a true ‘walk up and use’ interface with flexible data browsing and reporting.

Benefits

The combination of GCT and OpenLynx has been shown to be ideal for the high throughput screening of monomers for combinatorial library production.

Monomers are the low molecular weight building blocks used in the construction of combinatorial libraries. Confirmation of the identity and purity of these compounds, prior to using them in synthesis, is important in order to ensure that a synthetic route to an identified ‘active’ can be recreated. In addition, the requirement for post-purification of the final product is significantly reduced if not eliminated.^1,2,3 In many cases, however, the production of combinatorial libraries has sometimes proceeded with little regard to the purity of starting material. This has more to do with the complexity of analyzing large numbers of compounds than with a lack of desire from the synthetic chemist. While much time has been spent in developing LC-MS techniques and software to analyze the products of a combinatorial synthesis, little has been done to speed the process of analyzing small molecules not easily seen by LC-MS. GC-MS analysis in both electron impact (EI) and chemical ionization (CI) modes has long been used for the analysis of small, volatile compounds.

In this note, monomer confirmation using the GCT orthogonal Time-of-Flight (TOF) MS detector combined with OpenLynx high throughput software is described. The GCT produces exact mass data for elemental composition calculation and strucural elucidation.

OpenLynx provides a true ‘walk up and use’ interface with flexible data browsing and reporting.

Ninety-five (95) primary amines from the Monomers Store facility at GlaxoSmithKline RTP were analyzed by exact mass measurement using the GCT acquiring both EI and CI mass spectra.

Experiments were performed using a Micromass GCT oa-TOF mass spectrometer operated in positive ion electron impact (EI+) and positive ion chemical ionisation (CI+) modes with a one second acquisition over a mass range of 35-500Da. The source temperature was set to 180 °C in EI+ and 100 °C in CI+. Ammonia reagent gas at a pressure of 2 x 10^-4 mbar was used for the CI analysis.

GC analyses were performed using a HP6890 gas chromatograph split/splitless injector. A J & W Scientific DB5-MS, 15 m x 0.53 mm ID column with a 1 m x 0.1 mm ID fused silica restrictor at the GC transfer line was used in constant flow mode with 1 mL/min Helium flow. The GC temperature program was 150 °C (2 mins) to 250 °C (4 mins) at 50 °C /min. This gave a run time of 8 minutes and injection-to-injection time of 11 minutes.

The EI and CI mass spectra of the 95 amines revealed only eight that did not contain the desired compound (vial 74 was broken). The amount of the desired compound present determined from the CI Total Ion Chromatogram (TIC), ranged from less than a few percent to 100%. This calculation was made using OpenLynx, comparing the peak area of the compound of interest to the total area under the chromatogram. The results shown in Figure 1 were extracted from the OpenLynx Browser file shown in Figure 2.

In addition to this purity estimate, the identity of the impurities can also be determined from the data as shown in the following examples.

Example 1

Cyclopentylamine, m/z 86.097, is present at 63% (Figure 3). Several impurities are present with a nominal mass of 102, but with different elemental compositions. Two of these impurities were identified as saturated oxidized impurities of amines (Figure 4). Impurities at 104 and 106 were also identified as saturated oxidized impurities (Figure 4).

Example 2

Three isomeric fluorobenzylamines were analyzed in vials (4-fluorobenzylamine), (3-fluorobenzylamine), (2- fluorobenzylamine), and showed vastly different impurity profiles. The accurate mass data confirmed that the indicated GC peaks at 4.42 minutes retention time (Figure 5) were due to the desired compounds. The amount of desired compound was 31%, 27%, and 15% respectively. A combination of CI and EI spectra and accurate mass data were used to determine the structure of these impurities. The compounds were from three different vendors and contained one common impurity at approximately 7.5 minutes retention time shown in Figure 5. This compound was determined to be an isomer of the difluorobenzylimine shown in Figure 6. The electron impact spectrum showed only the molecular ion at m/z 231 and a fragment at m/z 109 due to the fluorobenzyl ion. Since these three compounds are from three different vendors, this impurity is probably due to degradation of the amine.

The major impurity in 4-fluorobenzylamine was determined to be an amide as shown in Figure 7 (CI spectrum) and Figure 8 (EI spectrum). The two impurities present in 2- fluoro-benzylamine (Figure 9) are not related to the desired compound - Figure 10 shows the CI spectra and the elemental composition results. From the exact mass measured fragments in the EI spectra, the structures were determined to be as shown in Figure 11.

Of the 95 compounds analyzed in this experiment, only eight were found not to contain the desired compound. The accuracy of masses measured for the compounds and impurities presented was calculated. (Figure 12). These masses were measured using a single lock mass in both EI and ammonia CI. The exact mass data can also be used to calculate elemental composition, thus enabling the identification of impurities in the starting material. The use of purer starting materials can reduce the time necessary for purification and quantitation of final library members. Time can also be saved if it can be demonstrated that the impurities are not likely to lead to impurities in the final product. In the case of the benzylamines, for example, the amine present in all three samples would not present a problem if the primary amine was linked to a solid phase resin. The impurities would not react and could be rinsed from the resin after the coupling reaction had been performed. This type of analysis is easily automated using OpenLynx, a sophisticated batch-processing engine allowing chemists easy access to GCT. This streamlines the analysis of large batches of samples for screening and automatically verifies that a compound of the desired formula is present at each well location and that the purity of each targeted compound exceeds a user-defined threshold.

The combination of GCT and OpenLynx has been shown to be ideal for the high throughput screening  of monomers for combinatorial library production.

1. Van Hijfte, Luc; Marciniak, Gilbert; Froloff, Nicolas. J. Chromatogr., B: Biomed. Sci. Appl. (1999), 725(1), 3–15.
   
2. Volonterio, Alessandro; Bravo, Pierfrancesco; Zanda, Matteo. Tetrahedron Letters (2001), 42, 3141–3144.
   
3. Bookser, Brett C.; Zhu, Shirong. J.Comb. Chem., (2001), 3(2), 205–215.