Fundamental LC-MS Electrospray Ionisation Instrumentation · use fused silica capillaries with 20 m i.d. that can be used in the flow rate range 100-1000 nl/min.[19] The sample solution

i Wherever you see this symbol, it is important to access the on-line course as there is interactive material that cannot be fully shown in this reference manual.

Mass Spectrometry

Fundamental LC-MS

Electrospray Ionisation – Instrumentation

Aims and Objectives

Aims and Objectives

Aims Explain the function of the major components of an electrospray Interface

Investigate methods of optimising signals using electrospray Ionisation

Objectives At the end of this Section you should be able to:

List and describe the most important components of an electrospray Ionisation interface

Demonstrate an understanding of the principles of optimising instrument response when using electrospray ionisation

© Crawford Scientific www.chromacademy.com

2

Content Electrospray source design 3 Introduction 3 Electrospray capillary design 3 Sprayer –Sampling Plate Configuration 6 Cluster Ions 9 Prevention of Cluster Ion Sampling 10 Source Cleaning 12 Ion Optics 13 Ring Electrode 13 Ion Bridges 14 Collision Induced Dissociation 14 Ion Declustering 16 References 16


3

Electrospray source design Introduction Electrospray is the dispersion of a liquid into electrically charged droplets, combining the two processes of droplet formation and droplet charging. The process of droplet charging is affected by three main variables:

Eluent flow rate

Liquid surface tension

Electrolyte concentration If these parameters are not maintained at an optimised minimum level, the electrospray process will become unstable. If any of the variables increases significantly it maybe difficult for the electric field to produce the desired charged aerosol necessary for ion production in the API interface. Any effects observed due to a significant increase in any of the variables may be countered to a certain degree by increasing the capillary voltage (and hence the effective field strength at the capillary tip), but electrical discharge may occur, resulting in a decrease in instrument response and an unstable electrospray. Electrospray capillary design Standard electrospray capillaries are constructed from stainless steel or a coaxial arrangement of fused silica and stainless steel. If the capillary is of fused silica design, electrical contact is usually made by clamping the capillary in a metal union, however, capillaries with silver or gold deposits upstream from the tip have also been employed for improved electrical contact.

Conventional Electrospray Capillary (practical upper flow rate limited to 10-20μL/min)


4

The metal composition used for stainless steel capillaries is of great importance in the oxidative and reductive processes occurring during droplet charging and as such continuity of composition for the metal capillaries should be ensured to give long term robustness to analytical determinations.

Drying gas The practical upper limit to eluent flow in pure electrospray is 10-20 μL/min depending upon the solvent composition. Capillary design may be modified to increase the tolerance of the electrospray process to increases in eluent flow rate, liquid surface tension or electrolyte concentration. One successful approach used in order to increase electrospray flow rate is the introduction of a nebulising gas via a concentric tube around the capillary (pneumatically assisted ESI).[1]

High-flow electrospray sources (>5-10 L/min.) are normally combined with a supply of heat within the API source housing to assist the evaporation of solvents. The evaporation of large amounts of solvent is important to ensure that ion evaporation occurs in the optimum position within the API source to ensure maximum transmission and reduction of ion solvent clusters in the nozzle-skimmer region of the source. Source heating needs to be optimized for each analytical determination to ensure the production of the maximum amount of analyte ions in the source. Practically the use of heated nitrogen blown into the source housing is normally employed to aid desolvation in


5

high flow electrospray source designs. Perkin Elmer (Sciex) (Applied Biosystems, Foster City, CA), introduced one of the earliest instruments to incorporate pneumatically assisted electrospray in their IonSpray design. This design used a silica capillary within a stainless steel needle that was housed within a concentric PTFE tube. The design of the PTFE tube is such that gas flow rates at the capillary tip are around 200 m/s. When physically connecting the t-piece or HPLC column to the API interface housing, PEEK (polyetheretherketone) (0.1mm i.d.) tubing is preferred to fused silica tubing due to the possible adsorption of analyte species to residual silanol species on the inner surface of the silica capillary. Most modern instruments employ stainless steel or platinum capillaries for electrospray to avoid similar problems with fused silica that can adversely affect the quantitative response of the instrument. For low flow rate applications micro-electrospray needles are available which consist of either coated or uncoated silica capillaries with a drawn tip to allow micro-droplet formation.[18] Filters in the needle

assembly help to prevent blocking of the fine capillary tip (<0.5m), and the coating of the needle helps electrical contact, negating the need for coaxial sheath liquids. Potentials of between 0.5 and 1kV are normally sufficient to produce efficient electrospray at eluent

flow rates of less than 1 l/min. Nanoflow capillary instruments are also available which

use fused silica capillaries with 20 m i.d. that can be used in the flow rate range 100-1000 nl/min.[19] The sample solution is fed to these capillaries via nitrogen-pressurised nanovials and can provide stable electrospray for numbers of hours with sample volumes of 1ml and less. Application areas of nanoflow electrospray systems include the study of proteins and other solutions where sample size is extremely limited.[2.3]


6

Sprayer –Sampling Plate Configuration The position of the sprayer (capillary) relative to the ion sampling orifice within the sampling plate is of significant importance. By avoiding directly spraying at the sampling plate orifice the maintenance interval required for API interface types may be extended and the instance of charged droplet being sampled (as opposed to gas phase ions) is lowered –hence increasing instrument response.[3] The core of the electrosprayed aerosol contains larger diameter droplets than in the extremities of the spray. It is expected that the efficiency of ion production (via ion evaporation), will be higher at the perimeter of the aerosol. The off-axis response is also more stable than the one observed when using on-axis.[3,4] In most modern instruments the sprayer position may be adjusted in one, two or three axes by means of micrometer screws that will allow optimisation relative to the sampling orifice.

On-axis spray

Off-axis spray

Diagonal


7

The distance and potential difference between the tip of the capillary and the sampling plate determine the electric field that creates the electrospray and will influence the performance of the spray. Optimisation of the sprayer position and capillary voltage are interrelated and should be optimised empirically together.

Pepperpot (FISONS-Micromass)

Cross Flow (Waters)


8

Perhaps the most important practical advance in sprayer positioning has come with the introduction of orthogonal source design, where the sprayer is positioned orthogonally to the sampling orifice. This design has several advantages including, reduced down-time of the source with decreased coating of the source elements, sampling of fewer charged droplets relative to ions and the ability to tolerate higher flow rates. In orthogonal spray, neutrals and non-volatile materials collide with a plate perpendicularly located to the spray axis. Orthogonal design can be combined with a second orthogonal extraction (LCZ or ‘Z-Spray’ from Waters) or a with an off-axis extraction, such as in the aQa (Thermo-Finnigan) disposition.[5] Orthogonal designs allow the use of eluent flow rates up to 1ml/min with electrospray and gives the advantage that eluent systems containing non-volatile buffers may be used for extended numbers of samples before source cleaning is required. Orthogonal sources are available from several other manufacturers.


9

Cluster Ions Polar molecules in the gas phase (water, solvent and eluent molecules), tend to form clusters with ions.[23] Cluster ions will appear in the mass spectrum and they are often so large that they are far outside the detectable mass range of a typical quadrupole (3000-6000 Da):

nOHXOHnX )( 22

Cluster ions may collide with the source elements in the early stages of the spectrometer and give rise to ion ‘bursts’ which can result in noisy baselines in the Total or Selected Ion Chromatogram (TIC or SIC). One commonly applied solution to this problem allows ions to pass into the sampling orifice but excludes water vapour and other neutral species from the entrance to the vacuum system. This is achieved by forcing ions and neutrals in the opposite direction by the action of an electric field and/or a flow of dried gas.

Cluster formation

i


10

Cluster ions passing into the vacuum region of the source may be de-clustered by collision with rapidly moving background gas molecules, imparting enough energy to break the hydrogen bonds between the ion and the solvating cluster molecules. Increasing the temperature and the potential difference in the first vacuum region of the spectrometer in order to accelerate ions will both assist with ion declustering. Prevention of Cluster Ion Sampling A Sciex (Applied Biosystems, Foster City, CA), API source is shown. The region between the interface plate and the sampling orifice plate is continuously flushed with dry nitrogen that flows into the API source as well as into the vacuum region of the spectrometer. The gas flow into the source helps to repel water, neutrals and other potential contaminants, such as dirt and buffer salts, away from the sampling orifice, increasing the intervals between source maintenance. Ions in the region of the sampling orifice are driven into the vacuum region by the gas flow combined with a 600V potential difference between the interface plate and the sampling plate. The nitrogen employed in this sense is often referred to as a ‘Curtain Gas’.[6,7]

Gas curtain As the nitrogen curtain gas and ions undergo expansion into the nozzle-skimmer region of the source, significant cooling occurs. If the source and curtain gas are not heated, there is a significant possibility of cluster ion formation occurring in this region. Therefore most modern sources use both heated drying and curtain gas. A method preventing cooling effects experienced through expansion and the re-formation of cluster ions uses a heated transfer tube.[25] This device will preheat the mixture of ions and neutrals prior to expansion using a heated tube of approximately 20 cm long (100 – 200oC). A further advantage of this type of device is the ability to generate ions (via ion evaporation) from droplets that may be sampled into the nozzle-skimmer region. An Agilent Technologies (Palo Alto, CA, USA), source is shown which employs an ion transfer tube with metallised ends (also known as a 'dielectric capillary'), that allows the application of an accelerating voltage that can be used to promote in-source dissociation of cluster ions.

i


11

Orthogonal source with dielectric capillary (heated transfer tube)


12

Source Cleaning Contamination of the sampling orifice or tube can prove to be detrimental to the performance of the instrument, in some cases leading to very frequent source cleaning when dealing with samples in dirty matrices or when using non-volatile solvents and buffers. The layer of contamination inside an orifice or tube will attract a build up of charge that can effectively stop the passage of ions, while the flow of neutrals is not affected, thus significantly decreasing the instrument response. Cleaning regimes will differ for instruments from different manufacturers but may include some physical abrasion (aluminium powder is a popular choice) and / or wipe cleaning of the sampling cone and other source components followed by sonication in a range of solvents matched to the polarity of the contaminants –(hexane, acetone and methanol are all popular choices for source cleaning).

Dirty layer formation The use of curtain gas, off axis spray and/or orthogonal spray can significantly increase the interval between source cleaning, by directing the deposition of contaminant species away from the sampling plate, orifice or ion transfer tube.

i


13

Ion Optics Ring Electrode In the nozzle-skimmer region of the mass spectrometer ions are transported in the free-jet expansion of gas, some of which will pass through the skimmer into the higher vacuum region of the mass spectrometer. Focusing of the ions into a narrow beam is not possible due to the effects of free jet expansion, which tends to direct the ions and entrained neutrals and gases into a barrel type shockwave, away from the axis of the spectrometer. However, ions may be forced to remain closer to the axis of the beam if a tube or ring lens is placed between the nozzle (sampling plate) and skimmer.[8] A voltage applied to the ring will reduce the spread of ions (but not water vapour or neutrals), away from the axis with a subsequent increase in instrument sensitivity (mainly due to a reduction in the noise).

Ring electrode operation

i


14

Ion Bridges In the high vacuum region of the spectrometer after the skimmer lenses, the pressure is still not low enough to focus ions using traditional ion lenses. In most cases the ions are focused using a radio frequency (RF) only hexapole or octapole ion bridge, that guides the ion beam into the analyzing quadrupole system. Ion bridges work on a similar principle to a quadrupole mass analyser but operate with radio frequency potential only. The quadrupole DC component is removed, and the application of a radio frequency potential allows ions of all m/z values to pass from one region of the spectrometer to another.[9] The multipole ion bridge has the advantage of slightly focusing the ion beam, therefore increasing transmission as well as allowing the removal of a significant amount of neutral species which are not held within the multipole ion bridge, so increasing signal to noise ratio.

Ion bridges operation Collision Induced Dissociation Classical API spectra tend to show very little fragmentation due to the 'soft' nature of the ionisation processes. That is, during the formation of ions, the analyte molecules do not receive enough energy to break the intra-molecular bonds. However, fragmentation may easily be induced in one of the higher-pressure regions and structural information can be gathered. Acceleration of ions between the sampling orifice and the skimmer, or between the skimmer and the RF-only multipole results in collisions of ions with the background gas. This process is known by various different names depending upon the instrument manufacturer and includes 'in-source collision induced dissociation (CID)', 'nozzle-skimmer fragmentation', 'cone-voltage fragmentation', etc.

By increasing the potential difference between the skimmer and the quadrupole V(S-Q)

or between the nozzle and skimmer V(N-S)), the energy imparted to the analyte molecule through increasing frequency and energy of collisions can be enough to cause intra-molecular bonds to be broken and for fragmentation to occur.[10,11] In the case of larger molecules, such as peptides and proteins, the excess energy can often be absorbed in several vibrational modes and high potential differences are required to fragment these kind of molecules.

i


15

CID working principle The major advantage of this technique for the production of spectra containing a greater amount of structural information is its simplicity. Only one voltage needs to be adjusted and there is no need for switching or adjustment of collision gases and no retuning of ion optics. It is possible to test the degree of CID using probes.[11] In negative ion mode the degree of CID can be estimated using the drug naproxen that normally shows the [M-H]- ion as the only at m/z 229. With small nozzle-skimmer potential differences the naproxen molecule readily loses CO2 giving rise to a base peak at m/z 185. In the positive ion mode the [M+H]+ ion of dibutyl phthalate at m/z 279 can be used as a test ion. Fragmentation to m/z 149 takes place readily even under mild CID conditions, making declustering of ions impossible with these labile sample ions.

Mass spectrum of naproxen (used to estimate the degree of CID in negative ion mode)

Fragmentationi

i


16

Ion Declustering Collision Induced Dissociation can also be used to improve the baseline noise and increase the signal-to-noise ratio in LC-MS experiments.[9,12] When ions pass through the sampling orifice into the vacuum region the background density of neutral gas ions falls rapidly. If ions moving in a low-density gas are accelerated by the nozzle-skimmer region mild-CID may be affected, which will be enough to cleave the hydrogen bonds inside the ion-water or ion-neutral gas clusters. Further, heating of the ion clusters, which occurs during collisions, will also aid desolvation of the cluster ions. Moderate acceleration of clusters is effective and widely used to decluster ions, leading to a reduction in baseline noise and the numbers to cluster ions detected. A disadvantage of this type of approach is the moderate scattering of the ion beam that is associated with ion-neutral gas collisions that may lead to a small reduction in the numbers of ions passing through the skimmer element and into the mass analyser (i.e. reduced spectrometer transmission).

Ion declustering process by using CID References 1. A. P. Bruins, T. R. Covey and J. D. Henion, Anal. Chem. 59, (1987), 2642. 2. M. R. Emmett and R. M. Caprioli. J. Am. Soc. Mass Spectrom. 5, (1994), 605. 3. M. S. Wilm and M. Mann. Anal. Chem. 68, (1996), 1. 4. A. P. Bruins, T. R. Covey and J. D. Henions. Anal. Chem. 59, (1987), 2642. 5. K. Hiraoka. Rapid Commun. Mass Spectrom. 6, (1992), 463. 6. J. Abian. “The Coupling of Gas and Liquid Chromatography with Mass Spectrometry.” J. Mass Spectrom. 34, (1999), 157 – 168. 7. C. K. Meng and J. B. Fenn. Org. Mass. Spectrom. 26, (1991), 542. 23. C. M. Whitehouse, R. N. Dreyer, M. Yashamita, J. B. Fenn. Anal. Chem. 57, (1985), 675. 8. I. C. Mylchreest, M. E. Hail and J. R. Herron. United States Patent 5, 157, 260, October 20, 1992. 9. M. E. Hail and I. C. Mylchreest. Presented at the 41st ASTM Conference on Mass Spectrometry and Allied Topics. May 31 – Jun 1, 1993, San Francisco, CA, 745. 10. R. D. Smith, J. A. Loo, C. J. Barinaga, C. G. Edmonds and H. R. Hudspeth. J. Am. Soc. Mass Spectrom. 1, (1990), 53. 11. R. D. Smith and C. J. Barinaga. Rapid Commun. Mass Spectrom. 4, (1990), 54. 12. A. P. Bruins in “Electrospray Ionisation Mass Spectrometry” R. B. Cole [ed.] John Wiley and Sons Int. 1997, 132-133.

Declustering

i

Fundamental LC-MS Electrospray Ionisation Instrumentation · use fused silica capillaries with 20 m i.d. that can be used in the flow rate range 100-1000 nl/min.[19] The sample solution

Documents