Capillary Electrophoresis

Capillary  Electrophoresis         4.1: Introduction     The  separation  of  compounds  based  on  their  movement  when  exposed  to  an  electric field  was  first  observed  in  1807  by  Ferdinand  Friedrich  Reuß,  who  noticed  the movement  of  clay  particles  in  water  when  a  constant  electric  field  was  applied.  The theory  of  electrophoresis  was  refined  in  the  early  1900s  by  Marian  Smoluschowski, and further  refined  in  1937  by  Arne  Tiselius,  who  won  the  1948  Nobel  Prize  in Chemistry  for  his  work.     Electrophoresis  was  initially  conducted  in  polyacrylamide  or  agarose  gels  on  a slab—a  technique called  slab  gel  electrophoresis—where  charged  molecules  are applied  to  one  end  of  the  slab  and  an  electric  field  is  applied  over  the  length  of  the slab.  The  molecules  migrate  down  the  slab  at  different  rates  according  to  the  charge-­‐ to-­‐size (m/z) ratio.  Slab  gel  electrophoresis  is  still  widely  used  in  the  fields  of biology  and  biochemistry  on  large  molecules  such  as  nucleic  acids  and  proteins  to show  relative  concentrations  of  molecules,  the  purity  of  a  sample,  or,  when  used  in conjunction  with  a  standard,  to  identify  compounds.    Slab  gel  electrophoresis generally  has  long  analysis  times  (often  20-­‐40  minutes  per  sample),  low  efficiency, difficulties  in  analysis,  and is  unable  to  definitively  identify  compounds  in  a  sample as  migration time  is  not  necessarily  unique  to  each  compound.  Additionally,  slab-­‐gel electrophoresis  is  difficult  to  automate,  making  it  very  time-­‐intensive  to  run multiple  samples.     Capillary  electrophoresis  (CE)  is  widely used  as  an  alternative  to  slab  gel electrophoresis.  Gel  media  are  not  necessary  in  capillary  electrophoresis  as  capillary tubes are  themselves anti-­‐convective.  Stellan  Hjérten  preformed  the  first  work  in capillary  electrophoresis  in  1967,  using  millimeter-­‐diameter  capillary  tubes.  By  the early  1980s,  the  diameter  of  capillary  tubes  had  been  reduced  to  75µm. Capillary electrophoresis  generally  runs  faster  than  slab  electrophoresis,  provides  better precision  and  accuracy,  uses fewer  reagents  and  is  more  easily  automated.  Capillary electrophoresis  can  also  analyze  smaller  molecules  than  slab  electrophoresis,  thus expanding  the  range  of possible analytes.    By  pairing  capillary  electrophoresis  with mass  spectroscopy,  it  is  possible  to  obtain  confirmatory  results.  This  makes capillary  electrophoresis  an  intensely  useful and  powerful instrument  for  many scientific  disciplines.    An  overview  of  a  CE  system  is  shown  in  Figure  4.1.     Figure  4.1    Overview  of  a  modern  CE  system.     4.2:  Electrophoresis  and  Capillary  Electrophoresis     The  separation  of  compounds  in  capillary  electrophoresis  depends  on  the  velocity  of   individual  compounds.  The  velocity  is  given  by  the  electrophoretic  mobility  (µe)  of   the  compound  and  the  applied  electric  field,  (E)     Eqn  4.1 velocity = µ e E € The  electrophoretic  mobility  is  a  measure  of  the  particle’s  tendency  to  move   through  the  medium  given  the  applied  electrical  field  and  thus  changes  with  the   medium  and  the  particle;  tabulated values of electrophoretic mobility often differ from the experimentally observed electrophoretic mobility. Consequently, experimentally determined mobilities are called “effective mobilities,” and can change radically with different solvents and with different solution pH. The  electrophoretic  mobility   depends  on  the  frictional  drag  (Ff)  the  medium  exerts  on  the  particle  and  the   electrical  force  (Ef)  exerted  to  move  the  particle  such  that   Ef                   Eqn  4.2   µe = Ff   € € € € where  the  electrical  force  is  dependent  on  the  charge  of  the  ion  (q)  and  the  strength   of  the  electrical  field  (E)     E f = qE                     Eqn  4.3     The  frictional  force  can  be  described  as     F f = −6πηrv                   Eqn  4.4     where  r  is  the  ion  radius,  v  is  the  ion  velocity,  and  η  is  the  solution  viscosity.       In  slab  electrophoresis,  the  sample  is  placed  on  the  gel,  and  then  the  electrical  field   is  applied.  The  force  of  the  electrical  field  causes  the  sample  to  progress  down  the   slab,  accelerating  until  the  drag-­‐-­‐or  friction-­‐-­‐experienced  by  intermolecular   interactions  with  the  medium  equals  the  force  caused  by  the  electrical  field,  and   then  the  sample  proceeds  at  a  constant  rate.  Once  the  sample  has  reached  a  steady   velocity,  it  can  be  shown  that     q µe =                   Eqn  4.5   6πηr   This  equation  shows  that  highly  charged,  small  particles  move  quickly—especially   with  a  low  viscosity  fluid—whereas  less-­‐charged,  larger  particles  move  more  slowly,   especially  through  a  more  viscous  medium.     4.2.1: Electro-Osmotic Flow (EOF) Also called electroendosmotic flow, electro-osmotic flow is the movement of liquid in a porous material (such as a capillary tube) caused by a difference in potential across the material. When a charged particle is put in contact with a liquid in a capillary tube, a double-layer-or electrical double layer (EDL)--forms at the wall of the capillary (see Figure 4.2); this occurs at the interface of the glass capillary wall and the bulk solution. The first layer is surface charge, and can be positive or negative depending on the material. As capillaries are generally borosilicate glass, the numerous silanol (SiOH) groups cause the charge of the first layer to be negative. This layer is also sometimes called the Stern layer or Helmholtz layer. The second layer is made up of ionic particles in solution that are electronically attracted to the charge of the capillary surface. As the particles in this layer are not fixed, but move as a result of electrical and thermal energy, it is called the diffuse layer. Figure 4.2: Capillary double-layer The diffuse layer has a net charge, whereas the bulk, uncoordinated solution generally does not. When an electrical potential is placed across the capillary tube, the diffuse layer is pulled to one side. As the diffuse layer progresses to one side of the capillary tube, it drags the bulk solution along with it, creating a flow (specifically, the electro-osmotic flow) of the solution through the cathode. Because the species in the solution are charged, they separate in solution as discussed above; for most applications*, positively-charged analytes move toward the anode given normal (negatively charged capillary) conditions. The electro-osmotic flow (EOF) can be described in terms of velocity or mobility. The mobility can be calculated from Equation 4.6, where: ε (unitless) is the dielectric constant, a measure of a material’s ability to respond to a dialectric’s ability to store kg charge; ζ (V) is the zeta potential, discussed later; and η ( Pa s = ) is the solution sm viscosity. µ EOF = € ες η Eqn 4.6 Since the mobility is the ability to respond to electric potential, the velocity is the mobility multiplied by the electric potential: VEOF = E € € ες η Eqn 4.7 The zeta potential (ζ) is the potential due to the double layer on the capillary wall, and is directly correlated with the charge on the capillary. Because the charge on the capillary varies as a function of pH, the zeta potential also varies with pH, meaning the mobility and velocity of the EOF are highly pH dependent. At high pH, the silanol groups are deprotenated, meaning the net charge of the capillary wall is greater. Thus, the EOF is significantly greater at high pH than at low pH; in fact, the EOF can sometimes vary more than an order of magnitude between a pH of 2 and a pH of 12. Figure 4.3: Pump induced flow vs. electro-osmotic induced flow                                                                                                                 *  For  small  monoatomic  ions,  the  solute  mobility  can  sometimes  be  greater  the  EOF;  for  these   conditions,  anions  and  cations  will  move  in  opposite  directions.     A marked difference between electro-osmotic flow and flow induced by a pump is the cross section of the flow. Because a pump applies a force across the entire cross sectional area of the liquid in a capillary or tube, the friction between the wall and the liquid will cause the liquid at the center of the cross section to move faster than the liquid closer to the wall. Thus, the liquid velocity gradient flows parabolically down through the capillary. However, if a liquid is moved by electro-osmotic flow, the bulk flow is caused primarily by the acceleration of the cations near the capillary wall, and the force on the liquid is uniformly distributed. This results in a flat flow profile. As seen in (Figure 4.3), the flat flow profile drops off sharply at the capillary wall; this is caused by friction against the direction of flow, but to a much lower extent than observed in laminar flow caused by pumps. An even (or flat) flow results in compact peaks whereas a parabolic laminar flow results in a broad peak because the analyte is spread over a larger area. “Tighter” peaks are easier to integrate and facilitate the isolation of peaks that elute at similar times. Table 4.1 summarizes the variables that control electro-osmotic flow. Variable pH Ionic Strength Temperature Electric Field Organic Modifiers Hydrophilic Polymer Covalent Coating Effect on EOF directly proportional inversely proportional inversely proportional directly proportional usually inversely proportional usually inversely proportional dependent on coating Possible Problems can alter structure and charge of analyte sample heating and adsorption changes in viscosity can cause heating can alter selectivity of instrument dependent on polymer coating often unstable Table 4.1: EOF variables 4.2.2: Peak Resolution and Solute Zones Peak resolution in capillary electrophoresis is dependent on many variables, including longitudinal diffusion, injection length, sample adsorption to capillary walls, electrodispersion, Joule heating, and detector cell size. The loss of peak resolution is often referred to as “zone broadening” in CE. As in other forms of chromatography, this results in reduced separation of compounds and increased difficulty in accurate peak integration. 4.2.2.1: Longitudinal Diffusion Under normal conditions, diffusion along the capillary tube ( longitudinal diffusion) is one of the more substantial contributors to zone broadening. A solute’s propensity to diffuse throughout the solvent depends on the solute and the solvent. A longitudinal diffusion coefficient can be provided for a given system, with a lower coefficient meaning the solute will form narrower zones. The equation for the number of theoretical plates in a capillary tube is given in Equation 4.8, where N is the number of theoretical plates, µa is the solute mobility, E is the electric field, I is the effective capillary length, and D is the diffusion coefficient of the solute. N= € µ a EI 2D Eqn 4.8 This equation only accounts for longitudinal diffusion, although there are often other variables that affect zone broadening. 4.2.2.2: Injection Length When introducing a sample to a CE system, the volume of liquid that enters the capillary is called the sample plug (See Figure 4.4). If the sample plug is very spread out, analytes will not separate into tight peaks and tailing or broadening can occur. A practical injection should be between 1% and 2% of capillary length. Figure 4.4: Diagram of sample introduction and plug length 4.2.2.3: Sample Adsorption Interactions of the analyte with the capillary wall negatively affect zone broadening. In some cases, complete adsorption of the analyte to the capillary wall can occur. As most capillary tubing is made of borosilicate glass, this is most substantial with positively charged analytes that can interact strongly with the hydroxide groups on the surface of the glass. Prolonged sample-surface interactions can result in peak tailing and zone broadening. 4.2.2.4: Electrodispersion The capillary in a CE system contains an ambient level of ions, called the background electrolyte (BGE). For ideal separation, the BGE should be uniform in and between all solute zones. Therefore, the background solvent is generally always a buffer. 4.2.2.5: Joule Heating Joule heat is generated by passing electric current through a conductor and is directly proportional to the amount of power used. Joule heating can result in non-uniform temperature gradients throughout the capillary tube. Temperature increases are often around 10ºC, although increases of up to 70ºC or higher can occur. While uniform increases in temperature generally do not affect zone broadening, uneven increases in temperature leading to temperature gradients can cause substantial peak broadening and/or tailing. Heat can relatively easily dissipate through capillary walls, which can result in a large difference in temperature between the center of the capillary and the internal edges of the capillary. These gradients can result in differences in solvent viscosity, and thus air or liquid cooling of the capillaries is often necessary. Using thicker capillary tubes with a large outer radius and a smaller inner radius can also help combat this problem. Smaller sample volumes and higher surface-to-volume ratios limit the Joule heating, even when hundreds of volts per centimeter are applied. 4.2.2.6: Detector Cell Size Detectors on CE systems average a signal over a finite volume. Naturally, if the volume averaged is too large, it can combine signals from different analytes. In this way, zone broadening due to the size of the detector cell is similar to zone broadening due to a long injection plug. Generally, any variance due to detector cell size can be ignored if the length of the detection cell is roughly one third the sample zone length. 4.3: Samples 4.3.1: Sample Introduction For loading a sample, a stable DC power supply of about 30kV (200-300mA) is required. Instruments are made with the ability to switch polarity to facilitate a variety of methods of operation. Usually, current is monitored and maintained to compensate for some temperature-induced changes in buffer viscosity. Sample volumes are generally incredibly small. Volumes as low as 5µL can be sufficient for several injections. This can be disadvantageous if the sample is very dilute, because being limited on the loading volume of a dilute sample can reduce sensitivity. Loading too much of a sample can result in bad peak shapes due to field inhomogeneities if the conductivity of the sample is different from that of the buffer. Additionally, if the injection length is greater than the diffusion-controlled zone, peak broadening will occur. 4.3.1.1: Types of Sample Injection There are several methods for introducing samples into CE systems. It is important to note that some sample is always injected when inserting a capillary into the sample reservoir, due to capillary action. This volume is called the Zero-Injection, and the effect of this initial injection on sample migration and analysis is called the Zero-Injection Effect. In most cases, this phenomenon is insignificant, but it can be important in systems being employed for quantitative analysis. 4.3.1.1.1: Hydrodynamic Injection Hydrodynamic injection introduces a sample by applying a pressure on the sample reservoir. This physically forces solution through the capillary. The volume of sample loaded is independent of the sample matrix; instead, the volume depends on the details of the system, such as capillary dimensions, the viscosity of the buffer, the pressure applied, and the amount of time that pressure was applied. Pressures and times generally range from 25-100mbar and 0.5-5 seconds. Figure 4.5: Hydrodynamic injection setup Similarly, hydrodynamic injection can be performed by siphoning, or raising the input reservoir 5-10cm above the output reservoir while keeping the levels of each equal. This technique is generally employed when the equipment necessary for pressure injection is unavailable. 4.3.1.1.2: Electrokinetic Injection In electrokinetic injection, a voltage 3-5 times lower than that required for separation is applied between the input and output reservoirs for approximately 10-30 seconds. Sample is loaded according to the difference in voltage across the capillary and the amount of each individual analyte that is loaded is dependent on the analyte’s electrophoretic mobility. As such, electrokinetic injection cannot be used quantitatively. This technique is useful in situations where the media in capillary is very viscous making hydrodynamic injection difficult. Additionally, it requires no extra instrumentation. Figure 4.6: Electrokinetic injection setup 4.3.1.1.3: Field Amplified Sample Stacking (FASS) and Field Amplified Sample Injection (FASI) Samples zones are sharpened when the conductivity of the sample matrix is substantially lower than that of the buffer. Essentially, this is the reverse of electro-dispersion, described previously. Thus, samples with low salt content can be actively sharpened and/or large quantities of them can be loaded onto a capillary before zone-broadening effects begin to occur. When this phenomenon is utilized to allow larger sample lengths to be loaded hydrodynamically, the concentration is called field amplified sample stacking (FASS). When the injection is performed electrokinetically, it is called field amplified sample injection (FASI). In the case of FASI, if the conductivity of the sample varies or is unstable, results can be unreliable. In such a case, FASS can be used instead and, while peak resolution may be affected by variability in sample conductivity, the peak areas are still reliable. Often 100 times more sample can be loaded by FASS before separation is greatly affected. The actual amount of sample that can be loaded by FASS is regulated by the sample’s conductivity and factors such as Joule heating. As the conductivity of the sample must necessarily be much lower than that for the running buffer, the applied electric field can induce a large degree of heating within the sample plug, which can change the viscosity of the sample and result in degassing and/or boiling. Additionally, due to the difference in ionic strength between the sample plug and the running buffer, the electro-osmotic mobility will vary between the sample plug and the buffer. This can result in a parabolic flow profile between the plug and the running buffer that may cause zone broadening. Similarly, a substantially different pH between the sample plug and the buffer can also result in differences in analyte mobility, so a buffer with a similar pH to the sample solution should be chosen. 4.3.2: Sample Purification CE is attractive as a preparatory purification source because of its separatory strength. However, CE can only be used to purify small amounts of analyte, as the sample loading limit is low. However, small-volume fractions can be collected for sequencing DNA fragments, establishing biological activity of substances, and ascertaining the purity of a collected peak by reinjection. Autosamplers can be programmed to run a sequence of samples so that, in the event that multiple runs must be made and fractions collected and combined, the instrument does not need to be attended for the entire period. 4.4: Methods of Operation   4.4.1:  Capillary  Zone  Electrophoresis  (CZE)     Capillary  Zone  Electrophoresis  is  the  standard  method  of  CE.  Separations  are   performed  by  a  difference  in  migration  time  as  previously  described.         Figure  4.7    Diagram  of  a  CZE  system.       4.4.2:  Miceller  Electrokinetic  Chromatography  (MEKC)     Miceller  electrokinetic  chromatography  is  a  combination  of  electrophoresis  and   chromatography.  If  one  dissolves  surfactants  (for  example,  8-­‐9mM  SDS)  in  the   running  buffer,  micelles-­‐-­‐often  anionic-­‐-­‐form  and  travel  against  the  EOF.  The  EOF,   however,  is  still  faster  and  the  net  motion  of  the  micelles  is  in  the  same  direction  as   the  EOF.    Solutes  partition  between  the  micelles  and  the  buffer  through  hydrophobic   and  electrostatic  interactions.    In  this  way,  the  micelles  act  as  a  stationary  phase   which  is  not  necessarily  stationary,  but  slower  with  respect  to  the  EOF.  The  more   hydrophobic  compounds  migrate  more  slowly  through  the  capillary  due  to  their   interactions  with  the  micelles.  All  neutral  compounds  will  elute  between  the  EOF,   which  elutes  first,  and  the  micelles,  which  elute  last.  (See  Figure  4.8).       The  surfactants  can  be  anionic,  cationic,  non-­‐ionic,  zwitterionic,  mixtures  of  these,   salts,  or  microemulsions.  Deciding  which  surfactant  to  use  is  based  on  the  specific   interactions  necessary  to  separate  target  compounds.         4.8:  Example  miceller  electrokinetic  chromatogram     Surfactants  may  affect  the  EOF  and  solute-­‐wall  interactions.  Generally,  using  a  high   pH  buffer  will  maintain  the  EOF  and  diminish  changes  in  solute-­‐wall  interactions.   Additionally,  a  consistent  temperature  is  especially  important  as  changes  in   temperature  will  affect  the  concentrations  of  surfactant  required  to  form  micelles  as   well  as  the  partition  coefficient(s)  of  the  system.             4.4.3:  Capillary  Electro-­‐Chromatography  (CEC)     Capillary  electro-­‐chromatography  is  miniaturized  liquid  chromatography.  An   applied  electric  field  pumps  liquid  through  a  packed  capillary  column.  The  silica   capillaries  are  packed  with  silica-­‐based  reversed-­‐phase  particles  of  1-­‐5µm  diameter.   The  migration  times  of  different  solutes  vary  as  a  function  of  their  partitioning   between  the  mobile  and  stationary  phases.  If  the  packed  phase  is  charged,  there  is   also  an  electrophoretic  component  to  the  partitioning.    On  average,  there  are   100,000  theoretical  plates  per  peak  in  a  CEC  system.       As  the  only  difference  in  a  CEC  setup  is  the  capillary  tubing  used,  CZE  systems  can  be   easily  set  up  for  CEC  analysis.             4.4.4:  Capillary  Gel  Electrophoresis  (CGE)     Capillary  gel  electrophoresis  is  essentially  gel  electrophoresis  conducted  in  a   capillary  tube.  The  “gel”  is  not  necessarily  stiff  and  solid  as  it  is  in  slab  gel   electrophoresis.  To  avoid  confusion,  it  is  usually  referred  to  as  a  polymer  matrix.   Many  polymer  matrices  are  available  and  offer  different  separating  properties.       The  polymer  is  dissolved  in  buffer  and  loaded  into  the  capillary  tube.  Generally,  the   polymer  concentration  necessary  for  adequate  separation  is  inversely  proportional   to  the  size  of  the  analyte.  As  low  viscosity  polymer  solutions  do  not  experience  the   same  capillary  action  as  most  regular  buffer  solutions,  the  capillary  gel  is  often   pumped  with  pressure  and  the  capillary  wall  is  coated  to  eliminate  EOF.       There  are  generally  10  million  theoretical  plates  per  meter  of  capillary  tubing.    CGE   is  most  commonly  used  to  separate  single-­‐stranded  oligonucleotides,  proteins,  and   DNA  fragments.       As  in  CZE,  it  is  possible  to  add  agents  such  as  chiral  selectors  and  ion-­‐pairing  agents   by  covalently  bonding  them  to  the  gel  or  the  running  buffer.  This  can  greatly   increase  the  separation  possibilities  of  a  single  system.             4.4.5:  Capillary  Isoelectric  Focusing  (CIEF)       Capillary  isoelectric  focusing  separates  peptides  and  proteins  based  on  their   isoelectric  point  (pI)-­‐-­‐the  point  at  which  they  carry  no  net  charge.  A  pH  gradient  is   formed  within  the  capillary  using  ampholytes—zwitterionic  molecules  with  both  an   acidic  and  basic  moiety.     The  capillary  is  filled  with  a  mixture  of  solutes  and  ampholytes,  and  a  pH  gradient  is   formed  with  a  basic  solution  at  the  cathode  and  an  acidic  solution  at  the  anode.   When  an  electric  field  is  applied,  ampholytes  and  solutes  migrate  through  the  pH   gradient  until  they  reach  the  pH  that  corresponds  to  their  isoelectric  point,  where   they  become  uncharged.  The  solute  zones  will  maintain  a  narrow  breadth  as  any   solute  which  migrates  into  the  zone  of  a  different  pH  becomes  charged  and  migrates   back  into  the  pH  of  its  isoelectric  point.  After  this  equilibrium  occurs,  a  steady  state   is  reached  and  hydraulic  pressure  is  used  to  pass  the  zones  through  to  the  detector.       EOF must be minimized to prevent flushing ampholytes through the capillary. 4.4.6: Capillary Isotachophoresis (CITP) Capillary isotachophoresis is a moving boundary technique using two buffer systems. The buffer systems trap the analyte zones and all zones lay in between “leading” and “terminating” electrolytes, where the leading electrolytes are those in the buffer with a higher EOF and the terminating electrolytes are those in the buffer with the lower EOF. This bracketing affects the electric field experienced by each zone, creating different electric fields for each analyte. The electric field varies such that each analyte moves at the same velocity as that established by the leading electrolyte. Because velocity is defined as the product of the mobility and the electric field (velocity=mobility•field), this varying field acts to compliment the mobility such that all analytes move at the same velocity. Zones are formed and focused much in the same way as they are in CIEF; as soon as an analyte drifts away from its zone, it experiences a new electric field such that it either accelerates or decelerates until it migrates back to its zone and the steady state is reestablished. According to the Kohlrausch Regulating Function, the concentration of analyte in a zone is determined by the concentration of the leading electrolyte. Analytes with a higher concentration than the leading electrolyte experience broadened until the concentrations match, whereas analytes with a lower concentration than the leading electrolyte sharpen. In this way, CITP can be used as a preconcentration step prior to other methods of analysis. Generally, up to 30-50% of the capillary can be filled with sample while still maintaining good separation and concentration. Also due to the Kohlrausch Regulating function, increasing the concentration of the leading and/or trailing electrolytes can result in better separation and can improve detection of analytes that can be difficult to separate. However, when applying CITP, it can be difficult to identify a buffer system with leading and trailing electrolytes that have the appropriate mobilities while still maintaining the appropriate pH for the sample analysis. Additionally, only cations or anions can be analyzed by a given buffer system. 4.5: Detectors Detection in CE can be a significant challenge, as the small injection volumes require the samples to be relatively concentrated for reasonable detection. Many of these techniques are similar to those employed in liquid chromatography. 4.5.1: UV-vis Spectrophotometry UV-visible absorption is the most common detection technique for CE. Detection is nearly universal, and with fused-silica as the sample holder, wavelengths from 200nm through the visible spectrum can be used without interference. Generally, the capillary itself is used as the sample holder and passes through a UV-vis detector. Because the capillary tube is so thin compared to conventional path lengths, care needs to be taken in the design of the detector. The beam of light must be of small radius and focused directly on the capillary for accurate results. A disadvantage of this is that, correspondingly, few photons reach the detector and a generally high baseline is obtained. The optical beam is generally passed through a slit to restrict its width and center it on the capillary tube. One approach to decrease the detection limit is by increasing the path length of the flow cell. This can be done by increasing the diameter of the capillary, but by doing this there is also an increase in the current and resulting Joule heating. Two ways of increasing path length without increasing capillary diameter are depicted in Figure 4.10. The method shown in 4.10A, called the “bubble cell” does not require an increase in current, as the detection occurs after separation. When a zone enters the bubble, the velocity decreases as it expands to fill the capillary, it contracts along the capillary and overall concentration remains the same, even with increasing path length. There is a third method, which is a combination of 4.10a and 4.10b; the capillary bends to increase the path length, but the capillary is also bubbled along the path. 4.10: a, left) Capillary bubble cell for UV-Vis detection; b, right) capillary cell. When trying to detect a given analyte, care must be taken to ensure the running buffer is not optically active in the range of the analyte. Most peptides and carbohydrates absorb around 200nm, and consequently many biological buffers, such as HEPES, CAPs, and tris, should not be used in systems analyzing peptides and carbohydrates. A normal S/N for a 75mm id capillary is 62.5, and up to 650 on an increased path length cell.     4.5.2:  Indirect  Photometry       Indirect  photometry  utilizes  a  monitoring  ion  that  is  UV-­‐active  to  trace  the  presence   of  a  non  UV-­‐active  compound.  An  ion  of  the  running  buffer  is  chosen,  generally  to   have  a  high  UV  absorbance  intensity,  and  its  absorbance  is  monitored.  When  the   sample  zone  passes  through  the  detector,  it  will  be  accompanied  by  a  corresponding   drop  in  the  UV  absorption  due  to  the  monitoring  ion.       A  very  simplified  explanation  for  this  phenomenon  is  that  the  monitoring  ion  is   displaced  by  an  analyte  with  the  same  charge  based  on  electroneutrality.  However,   the  displacement  ratio  can  be  greater  than  one  and  analytes  with  charges  opposite   to  the  monitoring  ion  can  be  detected.  This  can  be  better  explained  by  a  closer   analysis  of  the  Kohlrausch  Regulation  Function.             4.5.3:  Laser-­‐Induced  Fluorescence  (LIF)     Laser-­‐induced  fluorescence  adapts  fluorescence  spectrometry  to  CE  in  the  same   manner  as  UV-­‐vis  spectrometry.  A  laser  is  used  to  excite  the  analyte,  and   fluorescence  is  measured  via  a  PMT.  Often  the  laser  is  aligned  non-­‐collinearly,  and  a   ball  lense  is  used  to  focus  the  light  on  the  center  of  the  capillary.  Generally,  a   reflective  ellipsoid  is  glued  to  the  capillary  to  direct  fluorescence  toward  the  PMT.     A  drawback  of  LIF  is  that  most  compounds  of  interest  in  CE  are  not  fluorescent.   However,  a  number  of  derivatizing  agents  can  be  used  to  act  as  markers  for  the   compound.  These  are  generally  organic  aromatics  and  require  excitation  between   250nm  and  500nm.             4.5.4:  Contactless  Conductivity     Contactless  conductivity  detectors  (CCDs)  place  an  actuator  electrode  and  receiver   electrode  along  the  capillary  tube,  as  depicted  in  Figure  4.11.  An  oscillating   frequency  is  applied  at  the  actuator  electrode  and  detected  at  the  receiver  electrode   with  the  liquid  inside  acting  as  a  resistor.  As  a  sample  zone  passes,  the  conductivity   changes.  Since  analytes  are  charged,  the  conductivity  generally  increases-­‐-­‐except  for   when  samples  have  a  lower  ionic  strength  than  the  running  buffer.         Figure  4.11    CCD  setup.     Contactless  conductivity  detectors  are  universal  and  work  well  for  inorganic   analytes  that  are  difficult  to  measure  by  other  methods.  CCDs  function  well  with  low   sample  volume,  which  is  advantageous  for  use  in  CE.  Additionally,  it  generally  has  a   detection  limit  in  the  ppm-­‐ppb  range  for  small  organic  ions.         4.5.5:  Mass  Spectrometry       Mass  spectrometry  is  a  particularly  attractive  detecting  strategy  as  it  is   confirmatory  and  can  give  valuable  structural  information  about  analytes.  However,   it  is  difficult  to  transition  from  a  liquid-­‐phase  separatory  technique  to  a  gaseous,   vacuum-­‐based  detection  technique.  To  an  extent,  this  problem  is  alleviated  by  the   small  amount  of  solvent  and  the  low  flow  rate  associated  with  CE.           4.5.5.1:  Electro-­‐Spray  Ionization  (ESI)       Electro-­‐spray  ionization  is  an  ideal  technique  for  interfacing  CE  and  MS,  as  it  accepts   small  volumes  at  a  low  flow  rate  and  results  in  gaseous,  ionized  particles  (for  a   detailed  description  go  to  section  5.3.1).  Additionally,  ESI  is  well-­‐suited  to  larger,   biological  molecules,  which  are  frequently  analyzed  by  CE.  However,  CE  and  ESI  are   both  dependent  on  electric  fields.  In  order  to  couple  them  in  the  simplest  of  cases,   the  end  electrode  of  the  CE  instrumentation  is  shared  with  the  ESI  electrode.   However,  in  some  situations  the  difference  in  the  magnetic  field  can  be  several   orders  of  magnitude  and  the  electrodes  may  be  of  the  opposite  sign.       Some  situations  can  be  remedied  by  grounding  the  end  electrolyte  of  the  CE  system,   and  providing  voltage  to  the  electro-­‐spray  system  from  the  MS.  If  the  potential   applied  to  the  ESI  is  negative,  then  positive  ion  will  enter  the  MS  and  this  is  called   positive  ion  mode.  If  the  CE  voltage  is  much  greater  than  the  ESI  voltage,  a  ground   path  for  the  current  is  provided  by  a  resistor  sink.  In  general,  when  the  ESI  voltage  is   much  different  from  the  CE  voltage,  the  variability  in  the  electric  field  can  affect  the   sample  separation  in  the  CE  system.           4.5.5.1.1:  Hydraulic  Interfacing     A  hydraulic  flow  rate  is  necessary  to  form  the  electro-­‐spray.  The  amount  of   hydraulic  flow  necessary  is  dependent  on  many  variables,  including  the  EOF  and   capillary  dimensions.  In  general,  there  are  two  methods  for  introducing  a  hydraulic   flow:  sheath  flow  and  sheathless  flow.             4.5.5.1.1.1:  Sheath  Flow  Interface       Smith,  et  al.  invented  the  tri-­‐axial  capillary  sprayer,  similar  to  the  one  shown  below.           Figure  4.12:  Sheath  flow  interface       Solvent  flows  between  the  capillary  and  the  sprayer  tip,  and  combines  with  the   capillary  solution  immediately  before  the  nebulizing  gas.  In  this  way,  a  hydraulic   flow  independent  of  the  EOF  is  provided  by  the  sheath  solvent.       The  major  disadvantage  of  the  sheath  flow  interface  is  the  fact  that  the  sample  is   necessarily  diluted  by  the  sheath  solvent.  Despite  this,  the  sheath  flow  interface  is   one  of  the  most  reliable  methods  of  interfacing  CE  and  MS,  although  several   alternative  methods  exist  but  are  not  “as  robust,”  according  to  Agilent  literature.  (It   should  be  noted  that  Agilent  possesses  the  distribution  rights  on  the  tri-­‐axial   sheath.)             4.5.5.1.1.2:  Sheathless  Flow  Interface       Several  methods  of  uniting  the  CE  capillary  with  the  ESI  system  exist    However,  in   all  of  these  systems,  electrochemical  processes  such  as  oxygen  and  hydrogen   formation  can  occur.  The  formation  of  gas  bubbles  that  are  not  frequently  vented   can  cause  increased  resistance.  In  the  tri-­‐axial  system,  evolved  gases  are  frequently   flushed  due  to  sheath  flow  and/or  the  nebulizer  gas,  but  in  sheathless  systems   increased  resistance  can  form  a  substantial  problem.         In  2007,  Moini  published  a  technique  that  does  not  require  sheath  flow  and  does  not   create  electrochemical  processes  at  the  electrodes.  A  thin,  porous  segment  of  glass   acts  as  a  liquid  and  electrical  contact  outside  the  capillary.  The  tube  is  either   grounded  or  placed  at  the  ESI  voltage.  This  system  has  recently  become  available   commercially  through  Beckman-­‐Coulter.             4.5.5.2:  Atmospheric  Pressure  Chemical  Ionization  (APCI)  and                                        Atmospheric  Pressure  Photo  Ionization  (APPI)       ESI  is  highly  efficient  for  many  systems.  However,  for  systems  where  the  running   buffer  is  not  especially  volatile  and/or  the  molecules  are  not  strongly  charged,   alternative  ionization  techniques  may  be  more  efficient.  Atmospheric  pressure   chemical  ionization  (APCI)  and  atmospheric  pressure  photo  ionization  (APPI)  can   both  be  used  for  systems  like  these.       APCI  relies  on  the  evaporation  of  an  analyte-­‐containing  aerosol,  where  a  charged   reagent  gas  collides  with  analyte  molecules  and  confers  the  charge.  In  APPI,  an   analyte  aerosol  is  again  evaporated  to  a  vapor,  and  then  photons  ionize  the  analyte.   In  situations  where  the  analyte  is  not  easily  photoionized,  a  buffer  gas  may    be     photoionized  and  then  transfer  the  charge  to  the  analyte.       APCI  and  APPI  both  require  substantial  heating  of  the  sample  and  therefore  are  not   appropriate  for  analytes  that  easily  degrade  at  higher  temperatures.  Typically,  APCI   and  APPI  systems  are  heated  to  between  200-­‐300ºC.    Additionally,  low  molecular   weight  (<1kD)  analytes  are  better  suited  to  APPI  and  APCI  as  they  are  more  volatile.       4.6:  Cited  and  Consulted  References     Dolnik, V. DNA sequencing by capillary electrophoresis (review). J. Biochem. Biophys. 1999, 41, 103-119. Lauer, H.H.; Rozing, C.P. High Performance Capillary Electrophoresis; Agilent Technologies: Germany, 2010. Petersen, J; Okorodudu, A.; Mohammad, A.; Payne, D. Capillary Electrophoresis and its application in the clinical laboratory. Clinica Chimica Acta. 2003, 330, 1-30. Tagliaro, F.; Manetto, G.; Crivellente, F.; Smith, F. A brief introduction to capillary electrophoresis. Forensic Science International. 1998, 92, 75-88. Tagliaro, F.; Turrina, S.; Smith, F. Capillary electrophoresis: principles and applications in illicit drug analysis. Forensic Science International. 1996, 77, 211-229.