On the Mechanism of Ion Formation from the Aqueous Droplet Irradiated with 3μm IR Laser Pulses under Atmospheric Pressure

MassTechVictor V. Laiko, Vladimir M. Doroshenko
Columbia, MD, USA (www.apmaldi.com)

 

INTRODUCTION

In 2002, the possibility of atmospheric pressure matrixless ionization from aqueous solutions under the influence of IR radiation was shown [1]. Although few labs in the world are equipped with IR AP-MALDI sources, the technique was recently successfully applied for MS analysis of oligosaccharides [2-5].

The understanding of the mechanism of ion production is important for the application and optimization of any technique. The first insight into the mechanism of ion formation/transport/losses for UV AP-MALDI was reported recently [6]. The suggested mechanism includes the release of both positive/negative molecular ions into a gas phase under the influence of UV laser pulse followed by the separation of positive from negative ions under the influence of applied electrical field. The recombination of positive/negative ions at this stage decreases the ion yield. The ions then are transported toward the MS instrument inlet under the influence of both electrical field and gas flow. The understanding of the mechanism has enabled to improve the transport of ions, generated by UV AP-MALDI, into MS instrument, using newly developed Pulsed Delay Focusing technique [7].

Unfortunately, the adequate understanding of processes of matrixless ionization from aqueous solutions under the influence of IR radiation is absent. Instead, there are several experimental observations that enable to reject some possibilities and speculate for the other.

The experimental study of the mechanism of IR-AP ionization (at 3-µm wavelength) of aqueous solution in comparison with UV AP-MALDI is the subject of the present research. Several specifically tailored experiments answer for several questions, bringing some peaces of general picture together:

Why water-glycerol solution gives ~10 higher signal compared with purely water solution?

What is the influence of electrical field and the droplet surface charge on peptide ion formation?

What is the portion of neutral peptide molecules, produced from the Aqueous Droplet Irradiated with 3µm IR Laser Pulses under Atmospheric Pressure?

In the Discussion, we try to suggest the model of ionization mechanism that could accounted for the most experimental data.

EXPERIMENTAL SETUP

Experimental Setup
Thermo Finnigan LCQ quadrupole ion trap Thermo Finnigan LCQ quadrupole ion trap

A Thermo Finnigan (San Jose, CA) LCQ quadrupole ion trap was integrated with modified MassTech Inc. (Columbia, MD) AP/MALDI source. In-house Yb:YAG-pumped OPO infrared laser provided 2.95 µm wavelength, 30 ns duration, 0.6-0.8 mJ pulses with a frequency of 5 Hz. The optical fiber of AP/MALDI was replaced with aluminum mirror and calcium fluoride lens (f=50mm) to focus the laser at the target surface. The position of the focusing lens was shifted off the focus to tune the energy density of the laser radiation. CCD camera and sample positioning from the commercial AP/MALDI source were used. External HV source was connected to the target plate through HV switch for quick voltage switching.

RESULTS

Why water-glycerol solution gives ~10 higher signal compared with purely water solution?

Figure 2
Figure 3
Figure 4

What is the influence of electrical field and droplet surface charge on the peptide ion formation?

The experiment: The dependence of Angiotensin II peak intensity vs. the voltage applied to the sample plate is measured with the precision of 10-15%. The following problems were accounted for during the experiment.
• Big droplet-to-droplet signal variation. Solution: the RATIO of the peak intensity at the given voltage to the intensity at the "reference" voltage (Uref=0V) is measured. Both intensities were measured for the SAME droplet. Droplets were carefully centered vs. the focal spot using CCD camera.
• The water is evaporating from the sample droplet over a short time of the order 5-10 minutes (see the previous section). Solution: a mechanical HV switch (home-made) was used to switch between 0 and HV quickly. 2- µL droplet is deposited to the plate. The cycle: (signal accumulation for 5 scans at zero volts - signal accumulation for 5 scans at variable voltage) was repeated 3 times for every droplet. Experiment time was 3-4 minutes. The 3 values for zero volts and for given voltage were averaged. The ratio of the average values was calculated. The whole experiment was repeated 5 times (using 5 sample droplets) for every voltage. The average ratio and standard deviation were calculated for every voltage.
• The external HV power supply was used to apply both positive and negative voltages in positive ion mode.

Figure 2

Conclusion: The measured voltage dependences for 1:1 water/glycerol is similar to that for 3:1. Peptide signal profile is almost flat for target voltages in a broad range (from +4,000 to -1,000V) and drops if the voltage is further decreased to -2,000V. Peptide signal in pure water solution varies strongly with the position of irradiated spot on the droplet surface. The irradiated droplet oscillates and moves during the measurement, which results in pure precision of the measurement. The absence of voltage is preferable for water samples.


Discussion: The initial formation of relatively big, weakly charged droplets directly during and after the laser pulse duration could explain the striking difference in voltage signal dependences between IR AP-MALDI of water solutions and UV AP-MALDI. The big droplets with low electrical mobility propagate toward the MS instrument entrance mainly by the influence of the gas flow. This speculation can explain both relatively flat voltage dependence for positive voltages and the presence of relatively big signal for negative target voltages (in positive ion mode). If this hypothesis is accepted, the following questions still have no answers:
• What is the mechanism of ion release from the almost uncharged big droplets?
• Glycerol has very low volatility. What is the mechanism of water-glycerol droplet evaporation, that could lead to the release of analyte ions?
• If electrical field weakly influences on the motion of big weakly charge droplets, why approximately zero voltage gives the best signal both for water/glycerol and water solutions?

What is the portion of neutral peptide molecules, produced from the Aqueous Droplet Irradiated with 3µm IR Laser Pulses under Atmospheric Pressure?

Figure 2

Conclusion: The irradiation of water sample with 3µm Laser Radiation produces both peptide ions and neutral molecules. Combined AP-MALDI-APCI ion source enables to ionize neutral peptide molecules. The increase in peptide peak intensity (20-50%) indicates that the portion of neutral molecules is not too big.

CONCLUSIONS

•Water-glycerol solutions demonstrate much better MS signal vs. pure water samples. Glycerol plays a “passive” role decreasing the speed of water evaporation and increasing the solution viscosity, thus preventing sample splashing during the experiment. Purely glycerol analyte solutions give marginal MS signal.
•The time of water evaporation from a typical 2µL water-glycerol droplet under a typical lab condition is 6-10 minutes. The laser IR irradiation has almost no influence on the water evaporation time. Thus the time between the droplet deposition and MS measurement should be short, while the measurement time is limited with approximately 5 minutes.
•Glycerol droplet can be “revitalized” after the water evaporation during MS measurement. The pipetting of few µL of water into the “exhausted” droplet leads to essential signal recovery.
•The method for measurements of quantitative dependences of MS signal should be tolerant to big sample-to-sample and experiment time signal variations.
•The dependences of peptide peak intensity on the target plate voltage were measured both for water-glycerol mixture and for purely water solution, in the positive ion mode. For water-glycerol, the signal is almost independent on the applied voltage in 0÷4,000V range, and slowly drops if the voltage of the opposite sign in the range from 0 to -3,000V is applied. For water solution, the dependence shows more sharp maximum at zero voltage.
•In contrast to IR-AP-MALDI with 10-µm wavelength [8], AP-MALDI at 3- µm generates analyte gasphase molecules mostly in the form of molecular ions together with a relatively small portion of neutral molecules.

REFERENCES

  1. Laiko V.V., Taranenko N.I., Berkout V.D., Yakshin M.A., Prasad C.R., Lee H.S., Doroshenko V.M., J Am Soc Mass Spectrom. 2002 Apr;13(4):354-61.
  2. Tan P.V., Taranenko N.I., Laiko V.V., Yakshin M.A., Prasad C.R., Doroshenko V.M., J Mass Spectrom. 2004 Aug;39(8):913-21.
  3. Von Seggern C.E., Moyer S.C., Cotter R.J., Anal Chem. 2003 Jul 1;75(13):3212-8.
  4. Von Seggern C.E., Cotter R.J., J Am Soc Mass Spectrom. 2003 Oct;14(10):1158-65.
  5. Von Seggern C.E., Cotter R.J., J Mass Spectrom. 2004 Jul;39(7):736-42.
  6. Berkout V., Kruchkov S., Doroshenko V., “Modeling of AP-MALDI processes”, Proc. of 52nd ASMS, 2004.
  7. Tan P.V., Laiko V.V., Doroshenko V.M., Anal Chem. 2004 May 1;76(9):2462-9.
  8. Coon J.J., McHale K.J., Harrison W.W., RCM, 2002;16(7):681-5.