Green chromatography-carbon footprint of columns packed with core–shell materials

Abstract

Although acetonitrile and methanol are the most popular organic solvents employed in reversed-phase HPLC, it is important to minimize the environmental impact of organic solvent usage during chromatographic analyses and the use of environmentally friendly solvents. Greener solvents, such as ethanol, are good organic modifiers and environmentally preferable solvents that may replace toxic solvents in many RP-HPLC applications, with good chromatographic properties. The present study describes a critical evaluation of core–shell columns packed with different particle sizes (5.0, 2.6, 1.7, and 1.3 μm), which was carried out to show the possibility of the reduction of solvent consumption and to show the possibility of replacing acetonitrile by ethanol in liquid chromatography. By comparing different columns packed with core–shell particles it was proven that the organic solvent consumption may be reduced six times obtaining the sufficient parameters of the analysis.

---

1. Introduction

High performance liquid chromatography (HPLC) is an analytical technique used for the separation, quantification, and identification of chemical compounds. In recent years, improvements in both instrument and stationary phase technologies have greatly enhanced the performance of analytical laboratory in the industry and academic sections, leading to improvement in the performance of many users. Nowadays, green analytical chemistry has been developed to reduce or remove the use of environmentally hazardous organic solvents and reagents.^1–6 Recent improvements in HPLC technology suggest that greener, but less effective, solvents, such as ethanol, can be used without a significant loss in analytical capacity.

Different alternatives have been suggested to reduce the use and generation of harmful solvents in liquid chromatography, including micro flow HPLC,^6,7 or the use of stationary phases allowing a high water proportion in the mobile phases.⁸ Traditional reversed-phase high performance liquid chromatography (RP-HPLC) thus become an attractive eco-separation technique, using conventional stationary phases under simple and user-friendly experimental conditions.

In addition to the well-known principles of green chemistry,⁹ the three R's (Reduce, Replace, and Recycle) are commonly mentioned in connection with green analytical chemistry. In recent years, most of the efforts at greening analytical chromatography have focused on either the replacement of hazardous solvents with green alternatives or, in general, on the reduction of the amount of hazardous waste produced per unit of manufactured product.¹⁰ In addition, both from an environmental and an economic point of view, it is reasonable to remove the severe amounts of hazardous organic solvents, for instance acetonitrile and methanol, from the eluent of liquid chromatography and replacing them with nontoxic components and “green” solvents, i.e. ethanol.^11–14 Furthermore, pure water mobile phase used in reversed-phase chromatography is usually associated with phase collapse in case of chromatography on C18 phases.¹⁵ Recently, some directions towards green analysis developed in chromatography are as follows:

- replacement of hazardous organic solvents with non-toxic and greener ones, i.e. ethanol¹⁶

- reducing the usage of hazardous solvent in HPLC with decreasing the column dimension (e.g. column length, internal diameter, and/or particle size)

- operating at an elevated temperature¹⁷

- use of mobile phase additives such as cyclodextrins,¹⁸ and surfactants to reduce the proportion of organic solvents¹⁹

- use of shorter chain alkyl groups of 1 and 8 carbon atoms bonded silica phases²⁰ and more polar reversed phases, instead of long-chain alkyl groups, such as C18 and C30 bonded phases²¹

- core–shell column technology has been developed as an environmentally friendly stationary phase and evaluated to achieve rapid separations with high efficiencies by using the shorter diffusion path of particles to reduce the axial dispersion of solutes.

In recent years, significant advanced techniques are performed in the new analysis methods for improving the speed of chromatographic analysis, which are mostly based on the increased use of core–shell and stationary phases with smaller particles (d_p = 1.3 μm), operating at higher back-pressures up to 1000 bar.^22–24 Theoretically, considerably higher flow rates can be used with smaller particles to improve the speed of separation in HPLC because the optimum linear velocity is inversely proportional to the particle size, which can be defined as follows:²³

where d_c is the internal diameter of the column, u is the linear velocity and t_R represents the retention time.

In general, these advantages can be obtained irrespective of the separation method or analyte structure, while some variation in the savings may be observed. It is evident from the UHPLC fast analysis technologies using smaller particles (d_p = 1.3 μm) that higher pressures (Δp = 1000 bar) and shorter column lengths (L = 50 mm) with smaller diameters (ID = 2.1 mm) offer a greener alternative to conventional HPLC by an organic solvent usage of 0.61 ml for one run separation of the test mixture.

The aim of this study is to determine the carbon footprints of core–shell stationary phases with different particle size and reduced column dimensions. The principal advantage of these materials is the use of available environmentally friendly solvents and reagents for liquid chromatography analyses and extractions to follow the first principle of green chemistry that emphasizes waste prevention instead of remediation. In this work, we investigate the parameters involved in the use of EtOH–H₂O mixture for greener analytical RP-HPLC applications by comparing it with mixtures of MeOH–H₂O and ACN–H₂O mobile phases.

2. Experimental

2.1. Chemicals and materials

All standards, benzene, ethylbenzene, naphthalene, phenanthrene, and pyrene (concentration range 10–40 μg ml⁻¹), acetonitrile, and methanol (HPLC grade) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Ethanol (HPLC grade) was purchased from J.T. Baker (Deventer, The Netherlands). Ultra-pure water used for mobile phase preparation was purified through a Milli-Q (Millipore, Bedford, MA). Four Kinetex-C18 columns (Phenomenex, Torrance, CA, USA) were tested. The physico-chemical properties of stationary phases are listed in Table 1.

Table 1 Physico-chemical properties of the Kinetex column given by the manufacturer

Packing material	Parameter
	Total particle size (μm)	Column length (mm)	Column diameter (mm)	Pore size (Å)	Effective surface area (m^2 g^−1)	Effective carbon load (%)	pH stability	Pressure stability (bar)
Kinetex C18	5.0	150	4.6	100	200	12	1.5–8.5	1000/600
Kinetex C18	2.6	150	4.6	100	200	12	1.5–8.5	1000/600
Kinetex XB-C18	1.7	100	2.1	100	200	10	1.5–8.5	1000
Kinetex C18	1.3	50	2.1	100	200	12	1.5–8.5	1000

---

2.2. Apparatus and methodology

The analyses were performed on a UHPLC Shimadzu Nexera chromatographic apparatus (Shimadzu Corporation, Kyoto, Japan) equipped with solvent delivery systems LC-30AD, a SIL-30AC autosampler, CTO-20-AC column oven, DGU-20A3 on-line degasser, and SPD-M20A detector equipped with a semi micro detection cell. The system control, data acquisition, and data evaluation were performed by Shimadzu “Lab-Solution” software (Shimadzu Corporation, Kyoto, Japan).

The isocratic mobile phase was prepared by on-line mixing in HPLC gradient grade organic solvents (ACN, MeOH or EtOH) and water.

The mobile phase composition for each column was optimized to get a resolution (R_s) value of about 1.6 for the ethylbenzene and naphthalene pair. The flow rate was adjusted to obtain the highest efficiency for each column by the van Deemter curve. Upon switching from a 4.6 mm ID column (5.0 and 2.6 μm of particle size) to 2.1 mm ID column (1.7 and 1.3 μm of particle size), in the case of acetonitrile and methanol solvents, 4.6 mm column was often operated at 1.0 ml min⁻¹, and 2.1 mm column was operated at 0.3 ml min⁻¹. However, in the case of ethanol, because of higher viscosity that causes high backpressure, flow rate for columns with particles with the sizes of 5.0, 2.6, and 1.7 μm (or 1.3 μm) was 1.0, 0.8, and 0.1 ml min⁻¹, respectively. A major advantage that evidently arises is the reduced solvent consumption with a factor of 4.8 without compromising the separation. Switching from a 4.6 mm ID to a 2.1 mm column is no problem on state-of-the-art LC instrumentation, although analysis time can slightly increase because of the gradient delay caused by the volume of both the pump and injector. Decrease in the internal diameter is often accompanied with an increase in sensitivity. This is a direct consequence of the reduced dilution of the solutes in the mobile phase and the appearance of more concentrated bands at the detector.

The column temperature was set as 30 °C, and the injected volumes were 1.0 and 0.1 μl for the column with 4.6 mm and 2.1 mm internal diameters, respectively. Three parallel injections were performed at each flow rate, and photometric UV detection at λ = 254 nm was applied. The peak profile data were acquired at the frequency of 100 Hz.

3. Results and discussion

3.1. Solvent replacement in RP-HPLC

The target of green analysis involves the use of less or non-toxic, renewable and green solvents,²⁵ which are an attractive feature, that ongoing and includes switching to materials that need low organic solvent content and those that reduce the consumption of organic solvents. The most common form of analytical separation technology performed in laboratories is RP-HPLC, using a hydrophobic stationary phase and a mobile phase comprising acetonitrile with water containing additives to adjust pH and ionic strength. HPLC separations require organic solvents in high proportions generating a large volume of waste. Water is the most environmentally friendly solvent, and in some cases, the chromatographic separation can be achieved by the use of pure water as the mobile phase.²⁶

Acetonitrile and methanol are mainly used as organic solvents in analytical HPLC, but they suffer from several drawbacks from the viewpoint of green sustainable chemistry. Methanol has some disadvantages compared to acetonitrile in the HPLC applications. The column pressure drop will be increased because of the higher viscosity of methanol,²⁷ which gives lower efficiency and broad peaks. Because methanol has a higher UV cut-off wavelength (λ = 205 nm) than acetonitrile (above λ = 195 nm), it thus limits its application in the low ultraviolet (UV) region.²⁸ The different analyte selectivities may be the most important factors that can occur when changing the solvent from acetonitrile to methanol or ethanol. Because of the toxicity of acetonitrile, these mixtures have to be treated as hazardous waste. Thus, there is growing interest in the replacement of toxic solvents with non-toxic solvents, mainly alcohols, as an alternative to solve some specific chromatographic problems.^29,30 The replacement of acetonitrile is a critical task because of its unique properties, e.g. lower viscosity and high transparency in the UV region, coupled with better chromatographic efficiencies. Therefore, it is not easy to replace acetonitrile with another solvent. However, hazardous waste streams containing acetonitrile as chemical waste in the HPLC can cause great environmental pollution, thus it should be disposed. Because acetonitrile is considerably more expensive than other solvents in HPLC, due to its limited availability, its cost dramatically increased in 2009.³¹

To follow the first principle of green chemistry, greener mobile phases (i.e. methanol) or less toxic solvent (i.e. ethanol as the sole organic solvent) have been proposed in this work. Although changing to methanol has a lower pressure than ethanol, because of the more amount of volume required for separation, ethanol is preferred. This study clearly showed that ethanol performs reasonably well as an RP-HPLC solvent and may be suitable for replacing acetonitrile in certain instances. The use of the greener solvent, i.e. ethanol, may therefore be an option for some analytical laboratories, in particular those where the poorer elutropic strength and UV cut-off would not be a significant problem.

Although the EtOH–H₂O mixture, as an organic modifier, has the same elution strength compared with ACN–H₂O solutions at room temperature;³² however, because of higher viscosity, it is less favourable. As a result, the related pressure required for a typical chromatographic separation tends to be higher.³³

The impact estimation of the effects of solvent change on retention volume of run time and percentage of organic solvent usage in similar conditions are summarized in Table 2. Because of the variation of acetonitrile solvent with ethanol by approximately constant resolution of 1.6, the retention factors of naphthalene for different column dimensions, namely, d_p = 5.0, 2.6, 1.7, and 1.3 μm, are 0.27, 0.23, 0.39, and 0.62, respectively. The proper retention volume, satisfied resolution, and separation can be accomplished with a nontoxic solvent. However, the disadvantages of the changes caused by higher pressure is reduction of the column lifetime limit and the flow rate. Therefore, with the change in solvent from acetonitrile to ethanol, in the case of d_p = 5.0 μm particle size, pressures increased from 85 to 288 bar. Although flow rate was the same for d_p = 5.0 μm particle when the solvent changes from acetonitrile to ethanol, for other particle sizes, the changes were as follows: for d_p = 2.6 μm flow rate was from 1.0 to 0.8 ml min⁻¹, and for d_p = 1.7 and 1.3 μm, the flow rate was from 0.3 to 0.1 ml min⁻¹. Under such conditions, for d_p = 2.6, 1.7, and 1.3 μm particle sizes, the pressure changes from 188, 268, and 395 bar to 548, 302 and 411 bar, respectively.

Table 2 Effect of different solvent and particle size on separation

Particle size	Solvent
	ACN			MeOH			EtOH
	Organic solvent (%)	Run time (min)	Vol. organic solvent (ml)	Organic solvent (%)	Run time (min)	Vol. organic solvent (ml)	Organic solvent (%)	Run time (min)	Vol. organic solvent (ml)
a Due to lower efficiency in MeOH the resolution 1.6 was not achieved.
5.0 μm	80	4.5	3.6	78.5	10.3	8.1	75	4.8	3.6
2.6 μm	82	3.8	3.1	82	6.5	5.3	78	4.5	2.8
1.7 μm	73.5	3.2	0.71	76	6.2	1.4	70.5	8.0	0.6
1.3 μm	62	3.3	0.61	70^a	7^a	1.47^a	60	9.9	0.6

---

The most glaring change was the elution time of these compounds under limiting pressure conditions for the separation of a test mixture when acetonitrile was used as the organic modifier compared to methanol and ethanol as a solvent for the column with 5.0 μm particle size. The constant resolution (R_s) value of about 1.6 for the ethylbenzene and naphthalene pair (peaks 2 and 3) is illustrated in Fig. 1.

Fig. 1 Comparison of the experimental chromatograms of test mixture separation (1-benzene, 2-ethylbenzene, 3-naphthalene, 4-phenanthrene, 5-pyrene) on the Kinetex C18 column (150 mm × 4.6 mm, 5 μm), column temperature 30 °C: (A): 75/25_v/v EtOH–H₂O, (B): 78.5/21.5_v/v MeOH–H₂O, and (C): 80/20_v/v ACN–H₂O.

3.2. Solvent reduction by using fast and green separation

3.2.1. Fast separation by reduce the run time. The easiest way to reduce the solvent consumption in a separation process is to reduce the run time in the HPLC instrument. The mobile phase consumption (V_m) is given by the following expression:

V_m = F × t_R (2)

where F denotes the volumetric flow rate, and t_R represents run time of the separation.

By reducing the column dimensions (i.e. column diameter, length and particle size), a rapid and efficient separation can be achieved. However, pressure drop for the column (50 × 2.1 mm, 1.3 μm) is considerably higher than the column (150 × 4.6 mm, 5.0 μm) even though a lower flow rate was used. The column length, L, required for a given separation is related to the particle size by the following equation:³⁴

L = 2N_req × d_p (3)

where N_req is the required column efficiency for a given separation, and d_p denotes the particle size of the packing material.

It can be seen that, though columns provide similar separation, the separation speed for the shorter column (50 × 2.1 mm, 1.3 μm particles) is nearly two times faster than the longer column (150 × 4.6 mm, 5 μm particles) simply because the former is three times shorter. Consequently, by reducing the column length and particle diameter size, a solvent saving of approximately six-fold can be achieved (see Table 2).

3.2.2. Green separation with core–shell column technologies. A number of approaches exist to reduce the solvent use for HPLC. Column dimensions and/or particle size and column equilibration time can be altered to achieve significant solvent savings along with more efficient, sensitive, and faster analyses.³⁵ The core–shell columns implement faster separation and offer lower back-pressure compared to traditional columns that are typically employed in UHPLC. Thus, the waste generated is typically lower than traditional columns.

During the study scale-down column dimensions were performed. The initial analysis was conducted on an HPLC column with standard analytical scale of 150 mm × 4.6 mm ID × 5.0 μm. To increase the speed of the analysis, the internal diameter of the column was decreased to ID = 2.1 mm and the particle size was reduced to d_p = 1.3 μm, as well as column length was decreased (Table 2). By the reduction of column diameter, the acetonitrile consumption can be reduced by approximately 34%, and the run time was shortened by 37%. Decreasing the column length from 150 mm to 100 mm and 50 mm decreases the ACN consumption from 3.1 ml to 0.71 ml and 0.61 ml, respectively. Improved resolution and a decrease in run time and equilibration times were additional benefits. All the data for MeOH, ACN and EtOH are summarized in Table 2. In Fig. 2 an example of a greener separation, using different core–shell columns, is presented.

Fig. 2 Comparison of separations performed on different column dimensions, (A) 150 mm × 4.6 mm, 5 μm, flow = 1.0 ml min⁻¹, 80/20_v/v ACN–H₂O; (B) 150 mm × 4.6 mm, 2.6 μm, flow = 1.0 ml min⁻¹, 82/18_v/v ACN–H₂O; (C) 100 mm × 2.1 mm, 1.7 μm, flow = 0.3 ml min⁻¹, 73.5/26.5_v/v ACN–H₂O, (D) 50 mm × 2.1 mm, 1.3 μm, flow = 0.3 ml min⁻¹, 62/38_v/v ACN–H₂O for test mixture (1-benzene, 2-ethylbenzene, 3-naphthalene, 4-phenanthrene, and 5-pyrene) on Kinetex C18 columns at the column temperature of 30 °C.

To minimize the solvent consumption, a smaller column diameter can be used with d_p = 1.3 μm particles to achieve higher optimum linear velocity. Reducing the internal diameter of HPLC columns can significantly lead to the reduction of hazardous waste solvent generations are presented in the following equation. Therefore, a flow rate of 0.3 ml min⁻¹ can be used for a column packed with d_p = 1.7 μm and 1.3 μm particles to achieve the same linear velocity, of about 2.5, separation speed and column efficiency, affording about six-fold additional solvent savings (see Fig. 2). As seen in Table 2, a reduction in solvent consumption and waste generation is observed only with minor differences in selectivity observed between the separations by comparing the volumes of mobile phase for elution time in an HPLC column.

As seen in Fig. 3, ACN can be replaced by EtOH. This change allows the sufficient separation of test mixture. Unfortunately, because of the higher viscosity of EtOH–water mixtures, the flow rate has to be reduced for columns with lower particle size. As a result, the time of the separation increases in comparison with the ACN as an organic modifier. It should be noted that the volume of the EtOH became slightly lower than the volume of ACN used. The results confirm that liquid chromatography may be carried out with a green solvent, such as ethanol.

Fig. 3 Comparison of separations performed on different column dimensions, (A) 150 mm × 4.6 mm, 5 μm, flow = 1.0 ml min⁻¹, 75/25_v/v EtOH–H₂O; (B) 150 mm × 4.6 mm, 2.6 μm, flow = 0.8 ml min⁻¹, 78/22_v/v EtOH–H₂O; (C) 100 mm × 2.1 mm, 1.7 μm, flow = 0.1 ml min⁻¹, 70.5/29.5_v/v EtOH–H₂O; (D) 50 mm × 2.1 mm, 1.3 μm, flow = 0.1 ml min⁻¹, 60/40_v/v EtOH–H₂O for test mixture (1-benzene, 2-ethylbenzene, 3-naphthalene, 4-phenanthrene, and 5-pyrene) on Kinetex C18 columns at the column temperature of 30 °C.

4. Conclusions

This work confirms that by using new core–shell Kinetex C18 columns, replacement of acetonitrile by ethanol in mobile phase has very similar performance in RP-HPLC applications. Because of the reduction in the particle size, (from 5.0 to 1.3 μm) in combination with the shortening of the column length (from 150 to 50 mm) and also with decrease in the internal diameter (from 4.6 to 2.1 mm), the new core–shell Kinetex C18 column provides a rapid throughput design, lower solvent consumption, low detection level, and high reproducibility. Therefore, by changing core–shell particle size from 5.0 to 1.3 μm, the column void volume for the acetonitrile and ethanol solvents are decreased from 1.95 to 0.2 ml and 2.11 to 0.21 ml, respectively. Under these conditions, the possibilities offered by this column technology are efficient, and the 1.3 μm core–shell particles are particularly attractive and can be exclusively used on a powerful state-of-the-art UHPLC system possessing an upper pressure limit of approximately 1000 bar. The separation times (3.3 min) and reduction in acetonitrile consumption (0.61 ml) can be considerably increased as compared to longer columns packed with 5.0 μm (3.6 ml). Moreover, it was proven that toxic solvent used in liquid chromatography can be replaced by ethanol.

Acknowledgements

Authors are grateful to ShimPol A. M. Borzymowski for kindly donating the Kinetex columns used in the study.

References

1. A. S. Pereira, F. David, G. Vanhoenacker and P. Sandra, J. Sep. Sci., 2009, 32, 2001.
2. S. A. Schuster, B. E. Boyes, B. M. Wagner and J. J. Kirkland, J. Chromatogr. A, 2012, 1228, 232.
3. F. Gritti and G. Guiochon, J. Chromatogr. A, 2011, 1218, 907.
4. R. Berky, S. Fekete and J. Fekete, Chromatographia, 2012, 75, 305.
5. K. Broeckhoven, D. Cabooter and G. Desmet, J. Pharm. Anal., 2013, 3, 313.
6. F. Gritti and G. Guiochon, J. Chromatogr. A, 2013, 1280, 35.
7. S. A. Schuster, B. M. Wagner, B. E. Boyes and J. J. Kirkland, Lecture L-01-04 35th International Symposium and Exhibit on High Performance Liquid Phase Separations and Related Techniques HPLC2012, Anaheim, CA, USA, June 17–22, 2012.
8. L. Pereira, An Overview of Core Enhanced Technology for Fast, High Efficiency HPLC, Thermo Fisher Scientific, Chromatography Today, May/June 2012.
9. P. T. Anastas and J. C. Warner, Green Chemistry: Theory and Practice, Oxford University Press, Oxford, UK, 2000.
10. C. J. Welch, N. Wu, M. Biba, R. Hartman, T. Brkovic, X. Gong, R. Helmy, W. Schafer, J. Cuff, Z. Pirzada and L. Zhou, TrAC, Trends Anal. Chem., 2010, 29, 667.
11. Y. Yang, Z. Strickland, B. Kapalavavi, R. Marple and C. Gamsky, Talanta, 2011, 84, 169.
12. K. Chen, F. Lynen, M. De Beer, L. Hitzel, P. Ferguson, M. H. Brown and P. Sandra, J. Chromatogr. A, 2010, 1217, 7222.
13. N. Furusawa and K. Kishida, LCGC North Am., 2004, 22, 1092.
14. N. Furusawa and K. Kishida, LC GC Eur., 2005, 18, 600.
15. D. Šatínský, I. Brabcová, A. Maroušková, P. Chocholouš and P. Solich, Anal. Bioanal. Chem., 2013, 405, 6105.
16. D. Raynie and R. Majors, LC GC Eur., 2011, 24, 78.
17. P. Sandra, G. Vanhoenacker, F. David, K. Sandra and A. Pereira, LC GC Eur., 2010, 23, 240.
18. V. González-Ruiz, A. G. León, A. I. Olives, M. A. Martin and J. C. Menéndez, Green Chem., 2011, 13, 115.
19. L. Zhu, L. Ding, Q. Zhang, L. Wang, F. Tang, Q. Liua and S. Yao, Green Chem., 2009, 11, 132.
20. D. Raynie and R. Majors, LC GC Eur., 2011, 24, 78.
21. D. Šatínský, I. Brabcová, I. Maroušková, P. Chocholouš and P. Solich, Anal. Bioanal. Chem., 2013, 405, 6105.
22. N. Wu and A. M. Clausen, J. Sep. Sci., 2007, 30, 1167.
23. H. Chen and C. Horvath, J. Chromatogr. A, 1995, 705, 3.
24. J. R. Mazzeo, U. D. Neue, M. Kele and R. S. Plumb, Anal. Chem., 2005, 77, 460A.
25. R. E. Majors, Continuing Innovations in Reversed-phase HPLC Column Technology, Pittcon, Orlando, Florida, USA, 2010.
26. Sz. Bocian, A. Nowaczyk and B. Buszewski, Anal. Bioanal. Chem., 2012, 404, 731.
27. R. E. Majors, LCGC North Am., 2009, 27, 458.
28. P. C. Sadek, The HPLC Solvent Guide, Wiley, New York, 1996.
29. F. Gritti, I. Leonardis, D. Shock, P. Stevenson, A. Shalliker and G. Guiochon, J. Chromatogr. A, 2010, 1217, 1589.
30. F. Gritti and G. Guiochon, Chem. Eng. Sci., 2012, 72, 108.
31. F. Gritti and G. Guiochon, LCGC North Am., 2012, 30, 7.
32. H. Chen and C. Horvath, J. Chromatogr. A, 1995, 705, 3.
33. L. R. Snyder and J. J. Kirkland, Introduction to Modern Liquid Chromatography, Wiley, New York, 2nd edn, 1979.
34. J. C. Giddings, Unified Separation Science, John Wiley, New York, USA, 1991.
35. C. J. Welch, N. Wu, M. Biba, R. Hartman, T. Brkovic, X. Gong, R. Helmy, W. Schafer, J. Cuff, Z. Pirzada and L. Zhou, TrAC, Trends Anal. Chem., 2010, 29, 667.

---