Can reversed-phase flash chromatography compete with prep-HPLC?

Flash chromatography competing with prep-HPLC? Why and when would flash outperform or even be equal to prep HPLC? Well, it depends on your purification needs and goals.

Flash chromatography is a staple technique within the organic and medicinal chemistry world because it is fast, efficient, provides good loading capacity, and can generate highly purified reaction products. Prep HPLC, on the other hand, is efficient at providing very high purity compounds but is slow and typically has lower loading capacity than larger-particle flash columns of similar size. Also, flash chromatography is typically used for intermediate compound purification while prep HPLC is primarily used for final product or really challenging reaction mixture purification.

What I have seen in my 22+ years at Biotage, however, is a meaningful and sustained movement of chemists towards reversed-phase flash and prep-HPLC for routine purifications. There must be a reason for this and I believe it has a lot to do with the increased research into new chemical entities which then increases the number of novel compounds, and their by-products, being generated. Often, these new small molecules have increased polarity and complexity compared to synthetic compounds of the past, which complicates their purification using traditional silica normal-phase chromatography techniques. So, chemists are using reversed-phase to address these purification challenges.

Some chemists perform reversed-phase purifications themselves while others utilize prep-HPLC capabilities, either in their own lab or in a prep lab in another part of their facility. But why, I wonder. I realize that for final products where purity targets exceed 98% you may need this capability, but for intermediates…?

Well, I would not be writing this post if I did not have data, using a real synthesized reaction mixture, supporting the performance of flash chromatography as essentially equal to semi-prep HPLC.

My reaction was hippuric acid and benzylamine (2 mM each) in DMSO creating N-[2-(benzylamino)-2-oxoethyl]benzamide, Figure 1.

Figure 1. Synthesis of N-[2-(benzylamino)-2-oxoethyl]benzamide from the reaction of hippuric acid with benzylamine in DMSO.

By using DMSO as the synthesis solvent I was able to load directly onto both my flash and prep-HPLC columns without work-up. To see how these two preparative columns compared, I tested them at different sample loads - 100 mg and 200 mg using the same gradient.

My first comparison was the 100 mg load (0.5 mL). My flash column was a Biotage® Sfär C18, 60 gram, 30 µm column operated at 50 mL/min. The prep-HPLC column was a 20 x 250 mm, 15 µm Shimadzu Shim-Pak C18 run at 10 mL/min, essentially matching linear velocities.

By the way, this reaction generated a TON of by-products but both columns separated the target compound from its closest eluting by-products quite well, as can be seen in Figure 2.

Figure 2. Comparison of reaction mixture purifications with a 100 mg load. Top - 60 gram Biotage® Sfär C18 flash column. Bottom - 20x250 mm Shimadzu prep-HPLC column. The chromatograms were normalized to the programmed gradient column volumes.

Because of media differences (typical with C18 media from various vendors), the Shimadzu column provided slightly different selectivity, retaining the product and its trailing by-products longer than the flash column. This is an important distinction as a purification is only as good as its resolution of the target product and its closest eluting by-product.

I then decided to challenge both columns by doubling the load to see if either column showed any separation degradation; neither did, Figure 3.

Figure 3. Purification comparison at 200 mg load. Top - 60 g Sfär C18 flash column. Bottom - Shimadzu prep-HPLC. Chromatograms normalized to the programmed gradient volume.

What is important to recognize with the 200 mg load comparison is that the resolution between the product and the trailing by-product remained the same on both columns while the leading impurity still was still well resolved on the flash column, even at a 2x load. I imagine a 300 mg load may show some minor degradation but at 200 mg, the purifications looked good. But how good?

I dried the product fractions to determine yield (Table 1) and analyzed them for purity, Figure 4.

So, both columns get the job done. However, the biggest reason to use flash over prep HPLC is the time required to perform the purification. The HPLC column required over an hour to purify each load while the flash column needed less than 25 minutes.

Same product purity with less time, that is how reversed-phase flash competes with prep HPLC.

Learn More