This lesson focuses on two key practical aspects of capillary electrophoresis (CE): sample introduction and detection. It compares hydrodynamic and electrokinetic injection methods and examines the limitations of on-capillary UV detection. In addition, the lesson introduces micellar electrokinetic chromatography (MEKC), a powerful variant of CE that enables the separation of neutral analytes through micelle-mediated partitioning. With a clear understanding of injection effects, detector considerations, and the operational principles of MEKC, students will gain valuable insight into how CE techniques can be adapted to broaden the scope of analyte classes.
Learning Goals #
After this lesson you should be able to
Differentiate between hydrodynamic and electrokinetic injection in CE and describe their impact on peak shape and quantification.
Identify the main limitations of UV detection in CE and suggest strategies to mitigate them.
Define the factors that influence injection- and detection-related band broadening in CE.
Explain how micellar electrokinetic chromatography (MEKC) enables the separation of neutral analytes.
Evaluate how surfactant type, concentration, and buffer composition affect selectivity in MEKC.
READ SETION 5.2.2
Injection Methods
1. Injection techniques #
In the previous class we have discussed some effects that the sample and the injection volume and concentration may have on the peak widths (efficiency) and peak shape in CE. We discussed sample stacking and electromigration dispersion. We will now take a closer look at injection methods, as well as detectors for CE.
1.1. Hydrodynamic injection #
In hydrodynamic injection the capillary inlet is immersed in the sample vial and an excess pressure (\Delta P_{\text{inj}}) is applied for a short period of time (t_{\text{inj}}). We can use Darcy’s law (Equation 1.70 in the book) to estimate the length of the resulting sample zone (L_{\text{inj}})
Equation 5.25: L_{\text{inj}}=u_{\text{inj}} \cdot t_{\text{inj}}=\frac{d^2_{\text{c}} \cdot \Delta P_{\text{inj}}}{32 \eta \cdot L_{\text{tot}}} \cdot t_{\text{inj}}
where d_{\text{c}} is the inner diameter of the capillary and L_{\text{tot}} its total length, and \eta is the viscosity of the BGE solution. For a dilute aqueous BGE solution
with L_{\text{inj}}, d_c and L_{\text{tot}} in m, t_{\text{inj}} in s, and \Delta P_{\text{inj}} in Pa. As a rule of thumb, the length of the resulting sample zone should be less than 1% of the effective length of the capillary from inlet to detector (L_{\text{inj}} < L_{\text{eff}}/100).
In Section 3.3.3 it is explained that in order for the extra-column band broadening to contribute less than 10% to the observed bandwidth, the standard deviation due to extra-column effects should be less than about half that of the bandwidth due to the separation, i.e. \sigma_{\text{ex}} \leq \sigma_{\text{col}}/2. If we want the effect to be less than 5%, this becomes about one third, i.e. \sigma_{\text{ex}} \leq \sigma_{\text{col}}/3.
If we assume that due to the Poiseuille flow profile \sigma_{\text{inj}} \approx L_{\text{inj}}/2, we get if we measure the injection band broadening against the ideal (molecular diffusion) band broadening in CE, we get
EXERCISE 1
Enter your answer mm. Round to one decimal.
This is correct!
This is incorrect. Use the equation above, i.e. L_{\text{inj,max}} \approx \frac{L_{\text{eff}} \cdot \sqrt{D_{\text{m}}}}{\sqrt{\mu_{\text{app}} \cdot V}}
What is the anticipated number of plates if diffusion is the only contribution to band broadening? Enter an integer in the field (i.e. no decimals).
This is correct!
Unfortunately, this is not correct. Hint: N_{\text{diff}}=\frac{\mu_{\text{app}} \cdot V}{2D_{\text{m}}}
1.2. Electrokinetic injection #
In electrokinetic injection the capillary inlet is immersed in the sample vial and an injection voltage (V_{\text{inj}}) is applied for a short period of time (t_{\text{inj}}). Usually, the sample is introduced at the side of the positive electrode. The length of the injection zone is different for each analyte, i.e.
Equation 5.26: L_{\text{inj},i}=\mu_{\text{app}} \cdot E \cdot t_{\text{inj}}\approx (\mu_{\text{EOF}} + \mu_i) \cdot \frac{V_{\text{inj}}}{L} \cdot t_{\text{inj}}
Note that substituting for is appropriate at best, because the field strength in the sample zone will depend on the conductivity of the sample. Component-dependent and sample-dependent injection amounts may complicate quantitative analysis.
READ SECTION 5.2.3
Detection Methods
2. Detection #
The most common detection principle is UV absorbance. It can be applied for most analytes, but sensitivity varies.
EXERCISE 2
Which factors complicate UV detection in capillary electrophoresis?
This is correct! UV-absorbance of the capillary is not an issue as used silica is not UV-absorbing. The outer coating of the capillary often is a UV-absorbing polyimide, but this can be removed to create a “detection window”. Alternatively, non-UV-absorbing coating may be used (e.g. a polyacrylate).
The geometry is, however, a problem: The circular cross section of the capillary is sub-optimal. This is also true for the short optical path length, which is arguably the most serious limitation. “Bubble cells” have somewhat longer pathlength at the location of the detection window.
The UV absorbance of the BGE is usually not so much an issue. Extra-capillary band broadening is also not an issue because detection takes place inside the capillary.
This is not correct. UV-absorbance of the capillary is not an issue, because the fused silica is not UV-absorbing. The outer coating of the capillary often is a UV-absorbing polyimide, but this can be removed to create a “detection window”. Alternatively, non-UV-absorbing coating may be used (e.g. a polyacrylate).
The geometry is, however, a problem: The circular cross section of the capillary is sub-optimal. This is also true for the short optical path length, which is arguably the most serious limitation. “Bubble cells” have somewhat longer pathlength at the location of the detection window.
The UV absorbance of the BGE is usually not so much an issue. Extra-capillary band broadening is also not an issue because detection takes place inside the capillary.
For non-chromophores indirect UV detection is an option that allows the use of standard commercial CE instrumentation. Fluorescence detection offers lower detection limits than UV absorbance, but it often requires analyte derivatization (or “labelling”).
The maximum length of the detection window can be calculated in a manner similar as for the maximum injection-zone length. However, the rule of thumb is now \sigma_{\text{det}} \approx 0.3 L_{\text{det}} (instead of \sigma_{\text{inj}} \approx 0.5 L_{\text{inj}}). This results in
The area of a peak in CE varies with the migration velocity or the migration time. This is irrelevant when comparing areas obtained for the same analyte with reference solutions, but to avoid effects of variations in, for example, the EOF velocity, you may use peak heights or “corrected peak areas”, defined as
The detection techniques are discussed more extensively in Section 5.2.3.
READ MODULE 5.4
Micellar Electrokinetic Chromatography (AND SECTION 5.1.4.6)
3. Micellar Electrokinetic Chromatography #
3.1. Concept #
Capillary electrophoresis is a high-resolution separation method for ions. It cannot be used for separation neutral molecules, but the related micellar electrokinetic chromatography (MEKC) does just that. In MEKC an ionic surfactant is added to the BGE in a concentration above its critical micelle concentration (\text{cmc}), so that micelles are formed. If an anionic surfactant is used (and the wall of the capillary is also negatively charged), the mobility of the micelles is in a direction opposite to that of the EOF.
Figure 1. Schematic illustration of the concept of micellar electrokinetic chromatography (MEKC).
Neutral molecules partition between the liquid BGE phase and the inside of the micelles. Very polar analytes will not enter the micelles and they will migrate with the EOF velocity. Highly non-polar molecules will be found exclusively in the micelles and they will migrate with a velocity equal to
As long as the absolute mobility of the micelles (|\mu_{\text{mic}}|) is smaller than the EOF mobility (\mu_{\text{EOF}}), all neutral analytes will appear in a window t_{\text{EOF}}<t_i<t_{\text{mic}}.
3.2. Retention #
We can define a retention factor (k_{\text{MEKC},i}) analogous to that in chromatographic separation methods, i.e.
where q{\text{mic},i} is the amount of analyte i in the micelles and q{\text{BGE},i} the amount in the BGE. It follows from Equation 1.4 (Lesson 1) that the migration velocity is equal to
or in terms of migration times
or
The equation for determining the retention factor from the electropherogram becomes a bit different from that used in chromatography, i.e.
3.3. Efficiency #
Now let’s look at the kinetics. In comparison with capillary electrophoresis, mass transport of analyte between the micelles and the BGE may add to the band broadening (C_{\text{s}} term), but the factors that dominate the band broadening in LC (eddy diffusion, A term, and variations in mobile-phase velocity, C_{\text{m}} term) are absent. As a result, MEKC is a high-resolution separation technique, as is evident from the chromatogram below for the screening of forensic drugs.
Figure 2. Chromatogram showing the separation of illicit drugs in a forensic screening. Reproduced from [1].
EXERCISE 3
In the above chromatogram t_{\text{EOF}} = 3.5 min and t_{\text{mic}} = 41 min. What is the retention factor of fentanyl, for which t_{\text{fen}} = 21 min? Enter your answer in minutes and round to two decimals.
This is correct! This implies that at t time there is about 10 times more fentanyl in the micelles than in the BGE, resulting in it appearing many times later than the EOF.
This is not correct. Use the equation above!
Concluding remarks #
Injection and detection play a critical role in the performance and reliability of capillary electrophoresis experiments. While hydrodynamic injection offers simplicity and reproducibility, electrokinetic injection introduces analyte-specific selectivity. On-capillary UV detection remains widely used despite geometrical and optical limitations. Importantly, this lesson also introduced micellar electrokinetic chromatography (MEKC), a versatile technique that extends the applicability of CE to neutral molecules through the use of micelles. By understanding both instrumental limitations and methodological extensions, students are now equipped to explore more advanced and application-specific uses of CE.
References #
[1] R. Weinberger, I. Lurie, Micellar electrokinetic capillary chromatography of illicit drug substances, Anal. Chem. 1991, 63, 8, 823-827, DOI: 10.1021/ac00008a018
