Hydroxymethylfurfural (HMF) Analysis of Beer by HPLC-UV using a 4.6 mm ID Monolithic Silica Column

Anita Piper, Scientist
MilliporeSigma

Hydroxymethylfurfural (HMF)
(5-(hydroxymethyl)furfural)

Section Overview

• Introduction
• Experimental
  
• Results & Discussion
• Conclusion
• Related Products

Introduction

Hydroxymethylfurfural (HMF) is generated by thermal decomposition of carbohydrates or sugars. It can be detected in many heat-treated food and beverage samples. In the latter, low levels of HMF indicate its freshness and natural finish, while long-term storage or exposition to heat leads to high HMF content, caused by fructose decomposition. 

HMF is suspected to have negative health effects on animals and humans¹ hence more attention is given to this by official institutions like the European Union², where e.g. limit values for honey are defined.³

This application focusses on the analysis of HMF in the different beer samples, smoke beer, wheat beer dark, and dark beer by HPLC with UV detection using a matrix tolerant monolithic silica column, the Chromolith^® HighResolution RP-18 endcapped. 

Experimental

Quantitation of HMF in three different beer samples was performed on a Chromolith^® HighResolution RP-18 endcapped 100 x 4.6 mm column using UV detection without extensive sample preparation performed (Table 1). The samples were just degassed and then injected.

Table 1. Conditions for HMF HPLC determination in beer

HPLC Conditions
Column:	Chromolith® HighResolution RP-18 endcapped 100x4.6mm (1.52022)
Mobile phase:	[A] 0.01% Trifluoroacetic acid in water; [B] Acetonitrile
Gradient:	Time (min)	A%	B%	Flow (mL/min)
	0	95	5	0.5
	7	95	5	0.5
	8	40	60	0.5
	12	40	60	0.5
	13	95	5	0.5
	19	95	5	1.0
	20	95	5	0.5
Flow rate:	See gradient table
Pressure drop:	15-34 bar (218-493 psi)
Column temp.:	25° C
Detection:	UV at 280 nm (micro flow cell; 1.4 µL/7 mm); 5 Hz
Injection:	5 µL
Samples:
Diluent:	Water
Standard solution (200 mg/L):	20 mg of homogenized HMF was weighed into a 100 mL volumetric flask, dissolved in diluent and filled up to mark with diluent. Standard solution needs to be prepared on the day of use.
Working solution (10 mg/L):	0.5 mL of the standard solution were pipetted into a 10 mL volumetric flask and filled up with diluent to mark. Working solution needs to be prepared on the day of use.
Sample preparation beer:	After degassing 1 L beer for 1 h in an ultrasonic bath the beer samples were filled into glass vials and injected directly.
Sample preparation spiked beer sample:	After degassing the beer samples were spiked with 20 mg/L of HMF. After the samples were filled into glass vials and injected directly.

Results

Figure 1. HMF standard working solution 10 mg/L (Peak 2: HMF).

Table 2. Chromatographic data - Working solution (10 mg/L)

Peak no.	Compound	Retention time (min)	S/N	Area (mAU*min)	Tailing factor
1	Void volume	3.1
2	HMF	6.5	1930.2	13.7558	1.1

Calibration

An external calibration curve was established using 7 concentrations (Table 3 & Figure 2)

Table 3. LOD, LOQ and linearity of external calibration for HMF

Conc. (mg/L)	Mean Area (mAU*min)
2.58	3.351
5.15	6.658
10.31	13.586
20.61	26.368
41.22	52.188
STEYEX	0.226
Slope	1.262
LOD (mg/L)	0.59
LOQ (mg/L)	1.79
R2	0.9999

Figure 2. Calibration curve of HMF standards.

Sample Measurements

As examples for beer samples, the results for the smoke beer unspiked and spiked (20 mg/L) are presented below (Figures 3 & 4, Tables 4 & 5). Reproducibility was determined by injecting each beer sample 5x (Table 6). In Table 7 an overview of the measure unspiked and spiked is displayed.

Figure 3. Injection of a smoke beer sample (Peak 2: HMF).

Table 4. Chromatographic data - Smoke beer

Peak no.	Compound	Retention time (min)	S/N	Area (mAU*min)	Tailing factor
1	Void volume	3.1
2	HMF	6.4	379.7	7.0099	0.9

Figure 4. Injection of a spiked (20 mg/L HMF) smoke beer sample (Peak 2: HMF).

Table 5. Chromatographic data - Smoke beer spiked with 20 mg/L HMF

Peak no.	Compound	Retention time (min)	S/N	Area (mAU*min)	Tailing factor
1	Void volume	3.1
2	HMF	6.4	1997.7	30.2236	1.1

Table 6. Sample repeatability of different unspiked beer samples in area (mAU*min)

Peak no.	Smoke beer	Wheat beer dark	Dark beer
Sample 1	7.0791	3.4538	4.9731
Sample 2	7.0316	3.5016	4.9713
Sample 3	6.9511	3.5393	4.9973
Sample 4	7.0099	3.5446	4.9828
Sample 5	7.0695	3.4996	4.9704
Mean	7.0282	3.5078	4.9790
Standard deviation	0.0514	0.0366	0.0114
(%) RSD	0.7	1.0	0.2

Table 7. Overview of HMF amounts detected in unspiked and spiked (20 mg/L) beer samples using the external calibration

Sample	Conc .HMF (mg/L)	Spike Recovery (%)
Smoke beer	5.34
Smoke beer spiked	23.74	92.0
Wheat beer dark	2.55
Wheat beer dark spiked	20.34	88.9
Dark beer	3.48
Dark beer spiked	22.29	94.0

Conclusion

The developed simple HPLC-UV method allows for a 5-HMF test in three different beer samples on a Chromolith^® HighResolution RP-18e 100x4.6 mm column with an LOD of 0.6 mg/L and LOQ of 1.8 mg/L. The recoveries for spiked beer samples were between 88.9% and 94%.

The bimodal pore structure of the monolithic silica Chromolith^® HighResolution column provided the matrix tolerance that allowed us to reduce the sample preparation to a minimum (just degassing). By this, the workflow could be kept short, which otherwise typically requires more elaborate sample preparation steps like SPE to, e.g. avoid column clogging.

Another way of HMF determination can be performed by a rapid photometric test that can be found in our Manual on Analysis Methods for the Brewery Industry. This covers a range of other parameters for the brewery industry and is complemented by e.g. our application note on Sulfur Dioxide in Beer Photometric Determination with Ellmann’s Reagent.

More applications on Food & Beverage Testing

REFERENCES

1. Shapla UM, Solayman M, Alam N, Khalil MI, Gan SH. 2018. 5-Hydroxymethylfurfural (HMF) levels in honey and other food products: effects on bees and human health. Chemistry Central Journal. 12(1): https://doi.org/10.1186/s13065-018-0408-3

2. Schrenk D, Bignami M, Bodin L, Chipman JK, del Mazo J, Grasl‐Kraupp B, Hogstrand C, Hoogenboom L(, Leblanc J, Stefano Nebbia C, et al. 2022. Evaluation of the risks for animal health related to the presence of hydroxymethylfurfural (HMF) in feed for honey bees. EFS2. 20(4): https://doi.org/10.2903/j.efsa.2022.7227

3. European commission. Council Directive 2001/110/EC of 20 December 2001 relating to honey. . [Internet]. Available from: http://data.europa.eu/eli/dir/2001/110/2014-06-23j