EP2710624B1

EP2710624B1 - Method and device for electrospraying a sample or a solvent containing the sample

Info

Publication number: EP2710624B1
Application number: EP12732867.2A
Authority: EP
Inventors: Anneli KRUVE; Koit HERODES; Rünno LÕHMUS; Ivo LEITO; Ants LÕHMUS
Original assignee: Tartu Ulikool (University of Tartu); Estonian Nanotechnology Competence Centre
Current assignee: Tartu Ulikool (University of Tartu); Estonian Nanotechnology Competence Centre
Priority date: 2011-04-27
Filing date: 2012-04-26
Publication date: 2015-06-17
Anticipated expiration: 2032-04-26
Also published as: EP2710624A1; WO2012146979A1; EE201100030A; EE05649B1

Description

TECHNICAL FIELD

The present invention relates to mass spectrometry and electrospray-ionisation (ESI) source used in mass spectrometers to spray the analyte or effluent containing the analyte into mass spectrometer. In particular, the invention relates to a novel method for spraying the analyte from ionisation source into mass spectrometer and a novel construction of the injector of the ionisation source, which ensures the formation of smaller droplets with more uniform size distribution of effluent comprising the analyte, which results in more efficient evaporation of the solvent, which in turn enables to enhance the sensitivity of the mass spectrometer and the stability of the results. The same principle may also be applied to improve the two other ionisation sources - atmospheric-pressure chemical ionisation and atmospheric-pressure photoionisation.

BACKGROUND ART

Electrospray ionisation (ESI) has been used for more than two decades as a solvent for entering various samples into the mass spectrometer. ESI conditions are mild, enabling also to analyse thermally unstable compounds.
In ESI, the effluent is directed into a metal capillary. This capillary is subject to high potential (2-5 kV) in terms of the input of mass spectrometer. Both positive and negative potentials could be applied to the capillary. Herein, we relate to the formation of positive ions, in which case positive potential is applied to the capillary of the mass spectrometer. The potential creates an electric field in the space of the ESI ionisation source. Due to the electric field, positive and negative ions are emitted from the surface of the droplets of the liquid exiting the capillary. If the electric field is sufficiently strong, positively charged droplet escapes the surface of the solution. Formation of charged droplets caused by the electric field is called electrospray.
Field strength necessary for the operation of electrospray process: $E_{0} = {(\frac{2 γ \cdot \cos θ}{ε})}^{1 / 2}$
where γ is the surface tension of the effluent, θ is the Taylor cone angle, ε₀ is dielectric constant of vacuum and r _c is the radius of the metal capillary. [Kebarle, 2000] For example, when V_c = 2000 V, r_c = 5·10^-4 m, d = 0.02 m, the electrospray requires the field strength of E₀=1.6·10⁶ V/m. [Kebarle, 2009]
And to achieve this situation, the voltage to be applied to the capillary V _on is: $V_{on} \approx {(\frac{r_{c} \cos θ}{2 ε_{0}})}^{1 / 2} \ln (\frac{4 d}{r_{c}})$

where d is the distance between the capillary end and the mass spectrometer (opposite electrode). Depending on the solvent of the analyte and distance of the capillary V _on values are in the range of 2200 V (e.g. methanol) and 4000 V (water). [Kebarle 2009].
Due to the voltage applied to the capillary negatively charged ions in the solution are drawn to the surface of the capillary, where the electrochemical reaction takes place. Blades et al [Blades, 1991] showed that electrochemical oxidation occurs in the surface of the capillary, resulting in the excess of positive charges in the effluent. For example, by using a zinc capillary, the recorded mass-spectrum may demonstrate Zn²⁺ ions. The same applies to Fe²⁺ ions in the case of a stainless-steel capillary. Furthermore, it has been demonstrated that electrochemical reaction takes place in the end of the capillary.
The droplet exiting the capillary forms the so-called Taylor cone, which in turn is sprayed due to the Coulombic forces into a uniform mist of tiny droplets (diameter ca 1 µm). The radius R of the formed droplets and charge q thereof may be calculated as follows: $R \approx {(\frac{V_{f} \cdot ε}{K})}^{1 / 3}$
$q \approx 0.7 ⌊ 8 π {(ε_{0} γ R^{3})}^{1 / 2} ⌋$

where V _f is the flow rate of the analyte, K is the conductivity of the analyte and ε is the dielectric constant of the effluent containing the analyte.
The charges generated in the course of electrospray - the so-called ion current - may be calculated as follows: $I = {(\frac{4 π}{ε})}^{\frac{3}{7}} {(9 γ)}^{\frac{2}{7}} {ε_{0}}^{\frac{5}{7}} Q^{\frac{4}{7}} {(K \cdot E_{c})}^{\frac{3}{7}}$

where ε₀ is the dielectric constant of vacuum, Q is the flow rate of the analyte, ε is the dielectric constant of the solvent, γ is the surface tension of the analyte and E _c is the electric field applied. [Tang&Kebarle 1991]
Owing to the electric field, charged droplets move towards the mass spectrometer (opposite electrode).
ESI chamber uses the heated gas flow blown by the mass spectrometer - the so-called drying gas. This gas protects the mass spectrometer against the entry of neutral molecules and facilitates evaporation of solvent from the droplets formed in the course of electrospray. As a result of evaporation of solvent the droplet radius decreases to its Rayleigh limit: $q_{Ry} = 8 π {(ε_{0} γ R^{3})}^{1 / 2}$

where q_Ry is droplet charge and R its radius [Pramanik, 2002].
When reaching the Rayleigh limit smaller droplets escape from the initial droplet, undergoing the so-called Coulomb explosion. New droplets created by each explosion carry about 2% of the initial mass and 15% of initial charge [Kebarle, 2009].
Tang and Smith [Tang & Smith, 1999] have developed a model for estimating the charge and radius of the created droplets and the time of reaching the Rayleigh limit: $t = 0.181 \cdot \frac{d_{0}^{2}}{k}$

where d ₀ is the initial diameter of the droplet and k is the evaporation rate of the solvent.
Wortmann et al showed that the concentration of the analyte increases in the formation processes of smaller droplets from larger droplets by at least 100 times [Wortmann, 2006]. Furthermore, it has been demonstrated that evaporation of solvent and formation of smaller droplets may result in a considerable change in the composition of the solvent, mainly due to preferential evaporation of more volatile solvents [Wang, 2010].
The small droplets returned are also subject to analogous solvent evaporation and distribution into smaller droplets via Coulomb explosion upon exceeding the Rayleigh limit. This process is repeated until due to the density of charges the field on the surface of the droplet is so strong to give way to desorption of ions from the droplet surface to the gas phase. [(Fenn, 1989); (Pramanik, 2002)] It has been suggested that ion desorption from the surface of the droplets starts when the droplet radius has decreased below 10 nm [Kebarle, 2009].
Such model of ion formation is called Iribarne- Thomson model as well as ion evaporation model, which describes the formation of ions from smaller molecules. [(Fenn, 1989); (Pramanik, 2002)]
ESI is also capable for creating ions, including multiply charged ions, from very large molecules, e.g. proteins. These ions are formed according to the charge residue model. According to this model, droplets divide into smaller droplets until the entire solvent has evaporated and the remaining charged analyte is analysed using a mass-spectrometer. [Kebarle, 2000]
For a bicomponent system, which comprises ions A⁺ created from the analyte and ions E⁺ created from other components of the effluent, the mass-spectrometric signal of ion A⁺ depends on the ion current generated in electrospray and the regeneration efficiency of gas-phase ions k _A: $I (A^{-}, ms) = Pf \frac{k_{A} [A^{+}]}{k_{A} [A^{+}] + k_{E} [E^{+}]} I$

where [E⁺] is the concentration of ions of other substance contained in the analyte, P is the processing efficiency of ions in mass-spectrometer and f is the fraction of charges emitted from the capillary and turned into gas-phase ions. k _A and k _E express relative efficiency with which A⁺ and E⁺ are turned into respective gas-phase ions. [(Iribarne, 1976); (Kebarle, 2009); (de Hoffmann, 2004)]
It has been shown that almost linear dependency exists between the mass-spectrometer signal and analyte concentration up to the concentration of 10^-5 M. Kostiainen et al [Kostiainen, 1994] showed that limits of the dynamic area of the analyte signal - already as ions of the analyte in the solution - are not caused by the limits resulting from the ion current. It has been suggested that the cause lies in the saturation of the droplet surface with the analyte. In addition, it has been demonstrated that by using two capillaries instead of one, it is possible to enlarge the dynamic area [Kostiainen, 1994].
Furthermore, the same group showed that the dynamic area of the signal depends on the solvent used. It was found that the widest dynamic area pertained to the solvent with the highest surface tension. In the case of solvent with high surface tension, the droplets formed in Coulomb explosion are relatively large, and there is more surface area available for the analyte. On the other hand, the solvents with lower surface tension resulted in greater mass-spectrometric signal. Thus the smaller droplets provide greater sensitivity. [Kostiainen, 1996]
Tang et al [Tang & Smith, 2004] showed that dynamic area depends on the complexity of the analyte, and that in the case of mixtures the dynamic area of the analytic signal may be limited by the presence of free additional charges in the droplet.
ESI also enables to analyse compounds, which, which are not charged in the effluent. These companies can undergo ionisation with the addition of a proton, metal cation (Na⁺, K⁺) or ammonium cation. Usually a proton is added. Protonation occurs in a greater extent in electrospray than it would in the effluent as a result of pH reduction, as additional charges are generated in the course of electrochemical process. The mass-spectrometric signal of ions returned from protonation of analytes with high pK _b value decreases with a reduction in pK _b. [Ehrmann, 2008]
Furthermore, it has been found that gas-phase proton affinity impacts on the generation of ions from analyte. For example, protonated water cluster can give a proton to the molecule of gaseous methanol, the proton affinity of which is greater than the proton affinity of water. Thus, the analytes may be analysed using ESI mass-spectrometer only when the gas-phase basicity of the analyte is greater than that of the solvent. [Amad, 2000]
Moreover, pneumatically assisted ESI has also been applied, which is also known as ionspray. The advantage of this system is the easier coupling with liquid chromatography. Kostiainen et al [Kostiainen, 1994] demonstrated that ion current is lower for pneumatically assisted ESI than for ordinary ESI.
As to the instrumental side, Maxwell et al [Maxwell, 2010] have shown that the shape of the end of the metal capillary used in ESI source has a strong impact on the strength of the mass-spectrometric signal.
Document US 5,879,949 describes a method and apparatus for combining electrochemical techniques and mass-spectrometry and for quick analytic study of substances. In the apparatus described in the said documents, the internal electrode of the capillary has been added to the ESI source. This approach ensures greater area for electrochemical reaction to occur, but does not increase the efficiency of droplet spray.

SUMMARY OF THE INVENTION

The object of the present invention is firstly to develop a method how to decrease the size of the droplets of the analyte or the effluent containing the latter, which are sprayed from ionisation source into mass-spectrometer. Decreasing the droplet size in turn enables better ionisation of the analyte particles, resulting in increased sensitivity of the mass-spectrometer.
The invention is also aimed at providing such construction of the injector of the ionisation source, which enables to implement the method according to the invention in the ionisation source, which is used for spraying the analyte into mass-spectrometer.
The most critical features of the ESI source are sensitivity to the analyte content in the solution and insensitivity (robustness) to the remaining characteristics of the sample. One of the most important characteristics for assessing robustness is the matrix effect.
Matrix effect is a change in the ionisation efficiency of analyte due to the influence of other sample components analysed therewith. The matrix effect may cause both signal suppression and amplification, which may result in under- or overestimation of the analysis results, respectively. [(Taylor, 2005), (Niessen, 2006)]
Matrix effect, ME%, may be assessed quantitatively as follows: $ME % = \frac{A_{Post_extraction_Spike}}{A_{Standard}} \cdot 100 %$

where A is the peak area of the analyte in the standard and sample extract, which has been enriched with the analyte at the same concentration level. When ME% values are close to 100%, it indicates a lack of matrix effect. The difference of %ME from 100% shows either the suppression (ME% below 100%) or amplification of ionisation (ME% over 100%).
Andrews et al found that the use of nanospray (decrease in the flow rate from 200 µl/min to 0.1 µl/min) prior to the ESI source increases the sensitivity of mass-spectrometer to 1000%. [Andrews, 2004]
Bahr et al compared the micro- and nano-sources of ESI (flow rates were 1 µl/min and 30 nl/min respectively). It was found that nano-source provides a considerably greater absolute signal - depending on the compound from 3 up to 600 times. This effect was explained by smaller radius, but greater charge to mass ratio, formation of droplets in the nano-source. [Bahr, 1997]
Thus, the formation of smaller droplets leads to increased sensitivity and lower matrix effect. Therefore, the intensity achieved using ESI ionisation source would increase further by modifying the spraying process so that the dimensions (radius, diameter) of the created droplets would be smaller than in the earlier solutions.
There are two hypotheses on the emergence of matrix effect. Firstly, the absence of additional charges in the droplet, which would enable the ionisation of all ionisation-prove compounds in the droplet. Secondly, the formation of gas-phase ions is limited by access of ions to the surface of the droplet.
Consequently, it may be possible to increase the ionisation efficiency of analytes by a) increasing the number of charges in the droplets created by .ESI, and b) increasing the surface area of ESI droplets (i.e. reducing the radius of the droplet).
The efficiency of electro-chemical reaction depends on the voltage applied to the electrode and the surface area of the latter. It is known that in ESI the metal capillary acts as one electrode, whereas the electrochemical reaction takes place at the end of the capillary.
Thus, to increase the number of available charges, the metal area of the end of the capillary should be increased on the account of the surface area of the electrode. By lengthening the capillary it is not possible to achieve the increasing effect desired, as the electrochemical reaction takes place only at the end of the capillary. Consequently, an additional construction needs to be fixed at the end of the capillary to increase the quantity of ions.
To form droplets with larger surface area and smaller dimensions in the ESI source, spraying has to be made more efficient. To achieve this, the system is supplemented with a capillary for directing an additional nebulising gas system. If such an additional capillary has been made from metal, and when the effluent containing the analyte is entered in the capillary, the additional capillary carrying the nebulising gas functions as an electrode by participating in generation of additional charges.
To attain the objective of the invention, i.e. optimisation of the size of gas droplets, a method is provided, according to which the additional spray of nebulising gas is directed into the spray of analyte and effluent containing the latter, which results in a circular-done shaped spray of analyte or effluent containing the latter, which is between inner and other spray of nebulising gas. To attain further objective of the invention, a piping of three (or more) capillaries is provided and the main parameters of the construction are optimised. This invention has been used for experimentation with pesticides in terms of sensitivity, decision limit and matrix effect. The construction according to the invention results in, for example, a unique combination of electrospray-ionisation source spray, which provides the mass-spectrometer with better sensitivity and greater insensitivity to the rest of the features of the sample. It has been demonstrated earlier that by changing the parameters of ESI itself it is possible to increase robustness only on the account of sensitivity.

LIST OF DRAWINGS

The present invention is described in more detail below with respect to certain embodiments, which to not limit the scope of invention, referring to the accompanying drawings, where

Figure 1 is a schematic diagram of the injector three-capillary system of the device used in the ionisation source according to the invention, where inside capillary B containing the analyte or effluent comprising the latter, housed in the external capillary A, is another, third, inner capillary C,
Figure 2 is a cross-sectional view of the sprays of the nebulising gas and analyte or the effluent containing the latter directed from the injector of the device according to the invention into the mass-spectrometer immediately after exiting the injector,
Figure 3 depicts the results of optimisation of gas pressure in the third, inner capillary C,
Figure 4 is a chromatogram, which characterises the use of the device according to the invention in the electrospray-ionisation source, and where the concentrations of analytes correspond to their detection limit,
Figure 5 depicts the injector of the device according to one embodiments of the invention, which may be used, for example, in the ESI device,
Figure 6 is a cross-sectional view of an alternative injector according to the invention, where the capillary with the analyte or effluent containing the latter houses a number of inner capillaries.

DETAILED DESCRIPTION

The method according to the invention comprises spraying the analyte or effluent comprising the analyte into mass-spectrometer and creating an outer jet of nebulising gas around it, and additional spraying of nebulising gas into the jet of the analyte or effluent containing the latter so that the jet of the analyte or effluent containing the latter is forced between the jets of two nebulising gases, i.e. the outer and inner jet, thus forming a circular-cone shaped jet of the analyte or effluent containing the latter. Such a method also enables to use various nebulising gases, whereas if the outer nebulising gas is usually nitrogen, the inner nebulising gas may also be nitrogen or other inert gas, which is necessary for transporting the particles of the analyte into the mass-spectrometer.
The device according to the invention is an injector used, for example, in electrospray-ionisation source (ESI), which is designed for spraying the analyte or effluent comprising the latter from the ionisation source 1 to mass-spectrometer 2, where the sample is analysed. The device according to the invention, which comprises a three-capillary system, may be used in various ionisation sources, for example, in atmospheric-pressure ionisation sources (APCI), atmospheric-pressure photoionisation sources (APPI), and the like.
The injector 3, depicted in figure 5, comprises a system of capillaries, consisting of an external capillary A and the capillary B with the analyte, which has been placed inside the external capillary A. One end of the external capillary A is connected with the source of nebulising gas 4. One end of the analyte capillary B is connected with the source of analyte or effluent containing the latter 5 and through the space between the internal wall of the external capillary A and the external wall of the analyte capillary B the nebulising gas is directed into the injector 3, which sprays the gas into the mass-spectrometer. Moreover, the capillary system also comprises an additional inner capillary C, which is placed inside the capillary B comprising analyte. The inner capillary C has been made of non-metal material, for example, quartz, but alternatively, it could also be made of metal or non-metal material, in which the external surface of the inner capillary is covered with metal making the inner capillary conductive. In comparison with the potential of the mass-spectrometer, the potentials of the inner capillary (C) and the analyte capillary (B) are equal. Alternatively, the potentials of capillaries (C) and (B) differ from that of the mass-spectrometer. Thus, if necessary, better ionisation of analyte particles is possible.
One end of the inner capillary C is connected with the nebulising gas source 6, which may be the same nebulising gas source 4, which feeds the nebulising gas to external capillary A, but the presence of more than one source of nebulising gas enables to use different nebulising gases in the experiments. Through the space between the inner wall of the analyte capillary B and the external wall of the inner capillary C the analyte or the effluent containing the latter is directed into the injector 3, where it is sprayed into the mass-spectrometer. At the same time, the nebulising gas is directed via the inner capillary C into the injector, where the nebulising gas is sprayed into the jet 15 of the analyte or the effluent containing the latter, which results in forcing the analyte spray 15 between the two, external 14 and inner spray 16 of nebulising gas (the external jet 14 exits the injector 3 via the space formed between capillary A and the analyte capillary B and the internal jet 16 exits the injector 3 from the inner capillary C) so that the analyte jet has a circular cross-section (see figure 5). In the device according to the invention the outlets of the capillaries on the side of the mass-spectrometer (external, analyte and inner capillaries) are aligned.
In an alternative embodiment, the outlets of the capillaries may be shifted axially from one another so that, for example, the jet of the analyte coming out of the injector 3 via the space between the inner capillary C and the analyte capillary B would not mix immediately with the nebulising gas jet coming via the space between the external capillary and the analyte capillary B, thus the analyte capillary B has been moved outwards from the end of the external capillary in the injector 3. So that the nebulising gas spray 16 directed from the inner capillary C into the spray of the analyte or the effluent containing the analyte 15 would not, immediately after exiting the injector, mix with the analyte spray 15, the inner capillary C has been shifted outwards from the analyte capillary B, towards the mass-spectrometer in the injector 3 (see figure 4).
In an alternative embodiments, the analyte capillary houses an additional number of inner capillaries (C), which form a capillary bundle and the inlet of which has been connected with the nebulising gas source and the outlet of which is for spraying the nebulising gas into the spray of the analyte or effluent containing the latter. Whereas in such an embodiment the inner capillaries are arranged in such a manner that the nebulising gas could be directed into the analyte or effluent containing the latter uniformly via all inner capillaries (C) or sequentially by the capillaries, which enables even more efficient breaking of the effluent spray so that the ions of the analyte of a very small quantity would reach the inlet of the mass-spectrometer. In this solution the outlets of all the inner capillaries (C) in the bundle are aligned or shifted slightly with respect to one another. Similarly, the nebulising gas is sprayed from the inner capillaries (C) into the jet of the analyte and the mass-spectrometer from the inner capillaries continuously or sequentially by capillaries.

Testing device

The testing device was constructed from three stainless-steel capillaries with the inner and outer diameter 4 and 2 mm, 0.8 and 0.55 mm and 0.203 and 0.089 mm, respectively. These capillaries are below referred to as capillaries A, B and C respectively. The positions of capillary A and capillary B are fixed. The length of capillary A was 53 mm, and the extension of capillary B from inside capillary A was 0.27 mm (i.e. 1/3 of its outer diameter). The capillary C was removable and its length in relation to capillary B may be modified.
In the course of experiments, three constructions were used:

1. The first construction was an ordinary two-capillary structure, where capillary B contained the sample solution (effluent exiting the liquid chromatograph) and was surrounded by capillary A, via which the nebulising gas was directed at 15 psi.
2. The second construction had an additional capillary C contained in capillary B (figure 1). Capillary C was empty - the sample did not reach it. All three capillaries were electrically connected.
3. The third construction was analogous to the second construction, but via capillary C additional nebulising gas - nitrogen - was directed. Nitrogen pressure was modifiable in the capillary.

In an alternative solution, the injector nozzle of the ionisation source is a system of capillaries or pipes, where the external capillary is connected with the nebulising gas source and the analyte capillary holds a number of inner capillaries directing the nebulising gas in the jet of the analyte. This solution enables to feed the nebulising gas into the mass-spectrometer via inner capillaries continuously, or alternatively, sequentially via individual capillaries, which would further break the jet of the nebulising gas and the analyte ions would enter the mass-spectrometer.

Liquid chromatograph

Chromatographic separation is achieved using 250 mm Zorbax Eclipse XDB-C18 column, having an inner diameter of 4.6 mm and particle size 5 µm. The separation also uses Eclipse XDB-C18 12.5 mm pre-column, diameter 4.6 mm and particle size 5 µm. 10 µl of the sample was injected into the column. Elution gradients were methanol and acetate (pH = 2.8). The buffer consisted of 1 mM ammonium acetate in 0.1% formic acid. The linear gradient from 20% methanol to 100% methanol took 15 minutes; thereafter the column was eluted with methanol 7 minutes. Then the methanol content was returned to 20% in 3 minutes. The flow rate of chromatographic eluent was 0.8 ml/min. The retention times of carbendazim, thiabendazole, imazalil and methiocarb were 8.3, 9.5, 13.7 and 16.5 min respectively.

Mass-spectrometry

The parameters of mass-spectrometer were optimised for the devices of the invention at the flow rate of the chromatographic solution. It was found that the optimum MS parameters were rather similar in all three devices, wherefore the same parameters were used for all constructions (Table 1).

Table 1 Parameters of the mass-spectrometer

	Time, min
Parameter	6.56-10.59	13.18-15.89	15.89-20.00
Capillary V	-5000	-5000	-5000
Skimmer V	59.6	35.9	27.5
Cap Exit V	148.4	123.8	152.5
Oct 1 DC	8.52	16.27	11.39
Oct 2 DC	0.98	0.61	0.86
Trap Drive	30.9	43.8	31.0
Oct RF	78.7	142.6	83.6
Lens 1 V	-5.9	-5.9	-5.9
Lens 2 V	-55.7	-76.4	-61.6

The constructed ionisation source was compared with the conventionally available Agilent Inc. ESI ionisation source in terms of sensitivity, decision limit and the matrix effect.
For all pesticides used the commercial source returned a higher increase in the calibration graph, but also a greater variability in accordance with the standard deviation the rise of the calibration curve. The ratio of the curve and its standard deviation was comparable in both constructions - the commercially available device as well as that of the invention.
Furthermore, it is possible to compare the limits of detection (LoD). For the commercial source the limits of detection were 10.0, 10.0, 10.0 and 5.0 µg/kg for carbendazim, thiabendazole, imazalil and methiocarb respectively. In the ionisation source according to the invention these were 5.0, 0.5, 1.0 and 5.0 µg/kg respectively. Thus, the use of an alternative source enabled to considerably lower the limit of detection for the three pesticides.
Furthermore, the matrix effects were examined using a commercially available source. Paired t-test revealed that the alternative ionisation source returned 100% closer %ME values that were statistically important. For imazalil, for example, the ionisation with commercially available source was suppressed by 49% in the sample on average. The device according to the invention gives the average suppression of 26%.
All ion currents referred to herein correspond to the eluent content of 47% methanol and 53% buffer.

Preparation of samples

The samples used were purchased from the local market and their pesticide content was checked prior to use. The extracts obtained after the sample preparation were enriched with the standard mix of pesticides.
The buffered QuEChERS method was used for sample preparation [Lehotay, 2005]. 15 g of homogenised sample was placed in 50 ml polyethylene centrifuge tube. 15 ml 1% acetic acid in acetonitrile, 6 g anhydrous magnesium sulphate and 1.5 g anhydrous sodium acetate were added to the reaction. The tube was shaken intensively for 1-2 minutes, whereafter the tube was spun at 300U-3100 p/min (900×g) for 1-1.5 min. From the upper extract layer 1-1.1 ml was subjected to further purification, and 15 ml was dried using a centrifuge tube, which contained 50 mg PSA sorbent and 150 mg anhydrous magnesium sulphate. The tube was shaken intensively for 0.4-0.6 minutes, and spun thereafter at 3000-3100 p/min. The extract was collected and enriched using a standard solution of pesticides.

Optimisation of the inner capillary position

In the first stage the position of capillary C was optimized in relation to the analyte capillary. The optimisation was also carried out for the second and third construction. Capillary positions (extension) - length L with respect to capillary B in figure 1 - were 0, 0.5 and 1.0 mm. Optimisation was carried out in chromatographic conditions and the standard solution used contained pesticides at the concentration of 0.45-0.5 mg/kg.
For the second construction it was revealed according to the chromatographic peak areas that the more capillary C extends out of capillary B, the higher the sensitivity according to the pesticide peak areas. The highest sensitivity was achieved when L = 1 mm. Moreover, it was noted that regardless of the position of capillary C a liquid meniscus was formed around the capillary, wherefore the analyte was in contact with the end of capillary C.
However, for the third construction, the optimum capillary position was with L = 0.5 mm in accordance with the peak areas of pesticides analysed.
In addition to peak areas of analysed pesticides, the electrospray also registered the generated ion current. It was found that the highest ion current corresponds in both construction with a situation where L = 1 mm. With L decreasing the corresponding ion current also decreased. Furthermore, it was noted that when L = 0 mm, ion current was higher than in the first construction. Thus one may presume that electrochemical reaction takes place on the surface of capillary C and creates additional changes in the analyte solution (effluent).
The different optimum positions of capillary C for the second and third construction may be explained via the formation of droplets. When capillary C is in its most extended position (L = 1 mm), the liquid meniscus is already relatively stretched, and, in the third construction, the additional gas can no longer considerably impact on the spraying process. But when capillary C is inside the sprayed liquid meniscus (e.g. L = 0.5 mm), the spraying is made more efficient, i.e. smaller droplets are formed.
On the basis of these results, the position of L = 0.5 mm was used for both constructions.
In the case of the third construction, the gas pressure in capillary C was optimised also. Gas pressures were optimised for both positions of capillary C, i.e. L = 0 mm and L = 0.5 mm. For both positions a non-linear relationship was found between the sensitivity of the method and gas pressure, but the optimums found for the positions were different. For position L = 0 mm, optimum gas pressure was 2 bar, but for L = 0.5 mm the pressure was 4 bar. Respective graph for position L = 0.5 mm has given in figure 2.
The non-linear relationship could be explained by the fact that pneumatically assisted spraying is very effective, when capillary C is inside the liquid meniscus. Thus, it only little gas was to be applied on capillary C to induce the formation of markedly smaller droplets. At the same time, too much gas pressure leads to blowing away of the formed droplets so that considerably less gas-phase ions reaches the mass-spectrometer.
Furthermore, it was found that in optimum conditions, L = 0.5 mm would result in higher sensitivity than L = 0 mm in its optimum conditions. Consequently, L = 0.5 mm was used as the capillary position and gas pressure of 4 bar in capillary C in the final analysis.
Following the optimisation of parameters for the constructions, the latter were compared using the standard pesticide solutions in concentration range from 5.00 to 0.01 mg/kg, 11 solutions in total. Moreover, 6 pesticide solutions were made using standard-enriched garlic sample in 1.5-0.25 mg/kg concentration range. The garlic samples were used was it was known from previous experiments that these samples were very complicated in terms of the analysis [Kruve, 2010]. The injection sequence of the samples were random, likewise the construction used.
Comparison of calibration graphs revealed considerably higher sensitivity for thiabendazole, imazalil and methiocarb correlating with the rising curve. In addition, it was noted that the three constructions provide slightly different linear areas in the graph.
For the first construction the linear area was the smallest, being up to 0.01-1.5 mg/kg, 0.01-0.5 mg/kg, 0.01-0.9 mg/kg and 0.01-2.5 mg/kg for carbendazim, thiabendazole, imazalil and methiocarb, respectively. In the case of the second construction, the linear areas were 0.01-1.5 mg/kg, 0.01-1.5 mg/kg, 0.01-2.5 mg/kg and 0.01-2.5 mg/kg, respectively, thus slightly wider than in the first construction. For the third construction - with an additional gas capillary C - the linear areas were 0.01-1.5 mg/kg, 0.01-3.5 mg/kg, 0.01-1.5 mg/kg and 0.01-1.5 mg/kg respectively.
One possible cause for changes in the width of linear and dynamic area may be the changing size of droplet surface area (according to source [Kostiainen, 1994]). Therefore, the formation of smaller droplets induced by the additional gas capillary - and the greater surface area of the droplets - should enlarge the linear and dynamic area towards higher analyte concentrations.
Ionisation of the neutral molecule M provides balance to the solution: $M + H^{+} < - > {MH}^{+},$

which depends on the excess protons in the solution. At the same time, the number of protons depends on the efficiency of counterion removal. Counterions were removed as a result of the electrochemical reaction at the metal-liquid boundary surface at the end of the capillary. Thus, it could be said that more counterions are removed when the metal-liquid boundary surface is larger - for example, as a result of addition of inner capillary C. Thus, more additional protons results in more protonated analyte molecules while the factor limiting the dynamic area is the lack of protons.
In addition to the enlargement of the linear area both additional protons and the enlarged droplet surface area should facilitate increase in sensitivity. The rises in the calibration graphs were used to compare the sensitivities attained with the three constructions.
For all calibration graphs, linear regression was used at the concentration range of 0.01-0.9 mg/kg (which corresponded to the linear area of the signal). The rises were compared using the t-test. At the confidence level 5% it was found that thiabendazole, imazalil and methiocarb gave a considerably higher sensitivity for the third construction compared to the first construction. The rises in calibration graph for the third construction were 25%, 63% and 64% (relative standard deviations 11 %, 10% and 4% respectively) higher, respectively, compared to the first construction. The differences between the second and the first construction were not statistically important in the experimentation conditions.
It could be presumed that the construction that provides the highest sensitivity also gives lower limits of detection. Thus, standard solutions of low pesticide content at 0.5 - 50 µg/kg were analysed in terms of all constructions, and the signal-to-noise ratio was established for the analytes. The limit of detection was the analyte concentration, which returned the signal-to-noise ratio of at least 10 for the respective ionisation source construction (the limits of detection were not given precision and the signal-to-noise ratio was computed according to the peak shape). The results are provided in table 2 below. Table 2 Limits of detections of analytes (µg/kg) in different ESI constructions.

First construction Second construction Third construction

Carbendazim 10.0 10.0 5.0

Thiabendazole 0.5 0.5 0.5

Imazalil 5.0 5.0 1.0

Methiocarb 10.0 5.0 5.0
Table 2 shows that the lowest limits of detections were established for the third construction.

Matrix effect

To test the robustness of ESI constructions, 4 garlic and 2 onion extracts were analysed, which extracts has been enriched with the standard mix of pesticides (1.5 up to 0.25 mg/kg). Matrix effect %ME values were calculated for all samples according to formula 6. The least matrix effect - %ME values closest to 100% - was established for carbendazim and thiabendazole in the third ESI source construction. In t-test it was found that these matrix effect values were statistically considerably closer to 100% than in the first construction. There were no statistically significant differences between imazalil and methiocarb.
Carbendazim and thiabendazole elute in a relatively watery environment (53% buffer and 47% methanol), wherefore the droplets created in the course of electrospray are quite large. In the case of large droplets the droplet area is relatively small and the analytes have to compete with the matrix components for the droplet surface area. Thus, the third ESI construction, which enables the formation of smaller droplets, creating more droplet surface area, decreases the competition between the analyte and the matrix, resulting in formation of more gas-phase ions of the analyte.

REFERENCES

1. Ehrmann, B.M.; Henriksen, T.; Cech, N.B. J Am Soc Mass Spectrom 2008, 19, 719-728
2. Wortmann, A.; Kistler-Momotova, A.; Zenobi, R.; Heine, M.C.; Wilhelm, O.; Pratsinis, S.E. JAm
3. Tang, K.; Smith, R.D. Int J Mass Spectrom 1999, 185/186/187, 97-105
4. Blades, A.T.; Ikonomou, M.G.; Kebarle, P. Anal Chem 1991, 63, 2109-2114
5. Karas, M.; Bahr, U.; Dülcks, T. Fresenius J
6. Tang, L.; Kebarle, P. Anal Chem 1991, 63, 2709-2715
7. Kebarle, P.; Verkerk, U.H. Mass Spectrom Rev 2009, 28, 898-917
8. Cech, N.B., Enke, C.G. Mass Spectrom Rev 2001, 20, 362-387
9. Enke, C.G. Anal Chem 1997, 69, 4885-4893
10. Kebarle, P.; Tang, L. Anal Chem 1993, 65, 972-986
11.Ahadi, E.; Konermann, L. J Am Chem Soc 2010, 132, 11270-112777
12. Wang, R.; Zenobi, R. J Am Soc Mass Spectrom 2010, 21, 378-385
13. Kostiainen, R.; Bruins, A.P. Rapid Commun Mass Spectom 1994, 8, 549-558
14. Kostiainen, R.; Bruins, A.P. Rapid
15. Maxwell, E.J.; Zhong, X.; Chem, D.D.y. Anal Chem 2010, 82, 8377-8381
16.Amad, M.H.; Cech, N.B.; Jackson, G.S.; Enke, C.G.
17. Tang, K.; Page, J.S.; Smith, R.D. J Am
18. Kruve,A.; Herodes, K.; Leito, I.; Rapid Commun Mass Spectrom 2010, 24, 919-926
19.Niessen, W.M.A., Manini, P., Andreoli, R., Mass Spectrom Reviews 25 (2006) 881-899
20. Taylor, P.J., Clinical Biochemistry 38 (2005) 328-334
21. Gangl, E.T., Annan, M., Spooner, N., Vouros, P., Anal Chem 73 (2001) 5635-5644
22.Andrews, C.L., Yu, C.-P., Yang, E., Vouros, P., J Chromatogr A 1053 (2004) 151-159
23. Bahr, U., Pfenninger, A., Karas, M., Anal Chem 69 (1997) 4530-4535

Claims

A method for spraying the analyte or effluent containing the latter from ionisation source to mass-spectrometer, which comprises spraying the analyte or effluent containing the analyte into mass-spectrometer and creating an outer nebulising-gas jet around it, characterised in that
a. inside the jet of the analyte or effluent additional inner nebulising gas is sprayed so that the jet (15) of the analyte or effluent containing the latter is forced between the inner jet (16) and outer jet (14) of the nebulising gas, and that

b. a circular-cone shaped jet of the analyte or effluent containing the latter is formed.
The method according to claim 1, characterised in that the outer nebulising gas and the inner nebulising gas are different.
The method according to claim 1, characterised in that for spraying the analyte or the effluent containing the analyte into mass-spectrometer, a system of capillaries is used, which comprises an external nebulising gas capillary, intermediate analyte capillary and at least one inner nebulising gas capillary.
The method according to claim 3, characterised in that the inner nebulising gas is sprayed into the jet of the analyte or the effluent containing the latter via several inner capillaries.
The method according to claim 4, characterised in that the inner nebulising gas is sprayed into the jet of the analyte or the effluent containing the latter from all the inner capillaries at once or from different inner capillaries at different points of time.
A device for spraying the analyte or the effluent containing the latter from ionisation source (1) to the mass-spectrometer (2) comprising
- an injector (3) with a capillary system, which comprises an external capillary (A) and the analyte capillary (B), which is placed inside the external capillary, whereas the capillaries have an outlet, through which the analyte or effluent containing the latter and the nebulising gas is sprayed into the mass-spectrometer (2),

- an external capillary (A), the inlet of which is connected with the nebulising gas source (4) and the inlet of the analyte capillary (B) is connected with the source of the analyte or the effluent containing the latter (5), whereas

- between the inner wall of the external capillary (A) and the outer wall of the analyte capillary (B) is a space, via which the nebulising gas is directed into the mass-spectrometer, characterised in that

- inside the analyte capillary (B) there is at least one additional inner capillary (C), which has a nebulising gas inlet, which is connected with the nebulising gas source (6) and an outlet, which is configured to spray the nebulising gas inside the jet of the analyte or the effluent containing the latter (16) and that

- between the inner wall of the analyte capillary (B) and the outer wall of the inner capillary (C) is a space, via which the analyte or the effluent containing the latter is directed into the mass-spectrometer (2).
The device according to claim 6, characterised in that the outlets of all capillaries (A, B, C) are arranged at the same distance from one another inside the ionisation source.
The device according to claim 6, characterised in that the inner capillary (C) is placed inside the analyte capillary (B) so that the outlet of the inner capillary (C) is extending from the cross-sectional plane of the outlet of the analyte capillary (B) towards the mass-spectrometer.
The device according to claim 6, 7 or 8, characterised in that the inner capillary (C) is made of non-metal material.
The device according to claim 6, 7 or 8, characterised in that the inner capillary (C) is made of metal or metal-covered non-metal material.
The device according to any preceding claim 6-10, characterised in that the potential of the inner capillary (C) differs from that of the analyte capillary (B) and that of the mass-spectrometer.
The device according to any preceding claim 6-10, characterised in that the potentials of the inner capillary (C) and the analyte capillary (B) are equal in comparison with that of the mass-spectrometer.
The device according to claim 6, characterised in that inside the analyte capillary (B) there is a number of inner capillaries (C), which form a bundle of capillaries, which are connected with the nebulising gas source (6) via the inlet and outlet thereof, which is meant for spraying the nebulising gas into the jet of the analyte or effluent containing the latter (16).
The device according to claim 13, characterised in that the nebulising gas outlets of all inner capillaries (C) of the bundle are aligned.
The device according to claim 13, characterised in that the nebulising gas outlets of the bundle of inner capillaries (C) are shifted with respect to one another.
The device according to claim 13, characterised in that spraying the nebulising gas from the inner capillaries (C) to the jet of the analyte and the mass-spectrometer is continuous or sequentially according to the inner capillaries.