Oleg 90 Desorption Dynamics Chem Asian J 2011

8/4/2019 Oleg 90 Desorption Dynamics Chem Asian J 2011

1/12

DOI: 10.1002/asia.201100419

Desorption Dynamics, Internal Energies, and Imaging of Organic Moleculesfrom Surfaces with Laser Desorption and Vacuum Ultraviolet (VUV)

Photoionization

Oleg Kostko,[a] Lynelle K. Takahashi,[a, b] and Musahid Ahmed*[a]

Introduction

Imaging the chemical composition of material that compris-es our world is of great interest to those who seek to under-stand and preserve it. Developing new microscopies todetect and identify organic molecules with high spatial reso-lution will afford the necessary molecular specificity toprobe complex nanoscale chemistry. For instance, visualizingthe organic composition of atmospheric aerosol particle sur-faces can promote understanding of the critical roles aero-

sols play in global climate change

[1]

and on the radiative bal-ance and chemistry of the troposphere. Biological molecules

are critical determinants in the structure and dynamics atboundaries of different cellular structures.[2] These moleculesplay important roles in the regulation of both intra- and in-tercellular processes, including metabolic activity, proteinsynthesis, hostpathogen interactions, and cell motility.Many of these cellular processes involve reactions over dif-ferent spatial locations or chemical gradients, for example,within a lipid bilayer.[3] The development of chemically sen-sitive methods to image the distribution of small moleculesin cells would provide a key experimental tool.

Biological systems are now being viewed both as possiblesources of alternative energy and environmental bioreme-diation. For instance, plant-based lignocellulosic materialsare being seen as potential biofuels,[4] whereas different bac-teria are being considered for environmental bioremedia-tion.[5] Interestingly, similar types of molecules that are pres-ent on the surfaces of different biological systems such asfatty acids, long-chain linear and branched hydrocarbons,and a variety of oxidized hydrocarbon species[6,7] are alsothought to be present on the surfaces of tropospheric aero-sols.[810] Furthermore, similar molecules are thought to beprevalent in soil organic matter[11,12] and it would be benefi-cial to have access to a set of tools that will allow visualiza-

Abstract: There is enormous interest invisualizing the chemical composition oforganic material that comprises ourworld. A convenient method to obtainmolecular information with high spatialresolution is imaging mass spectrome-try. However, the internal energy de-posited within molecules upon transferto the gas phase from a surface canlead to increased fragmentation and tocomplications in analysis of mass spec-tra. Here it is shown that in laser de-sorption with postionization by tunablevacuum ultraviolet (VUV) radiation,

the internal energy gained during laser

desorption leads to minimal fragmenta-tion of DNA bases. The internal tem-perature of laser-desorbed triacontanemolecules approaches 670 K, whereasthe internal temperature of thymine is800 K. A synchrotron-based VUV pos-tionization technique for determiningtranslational temperatures reveals thatbiomolecules have translational tem-peratures in the range of 216346 K.

The observed low translational temper-atures as well as their decrease with in-creased desorption laser power is ex-plained by collisional cooling. An ex-ample of imaging mass spectrometryon an organic polymer by using laser-desorption VUV postionization shows5 mm feature details while using a30 mm laser spot size and 7 ns pulse du-ration. Applications of laser-desorptionpostionization to the analysis of cellu-lose, lignin, and humic acids are brieflydiscussed.

Keywords: desorption lasers mass spectrometry nucleic acids scanning probe microscopy

[a] Dr. O. Kostko, L. K. Takahashi, Dr. M. AhmedChemical Sciences DivisionLawrence Berkeley National LaboratoryMS 6R-2100, 1 Cyclotron RoadBerkeley, CA 94720 (USA)Fax: (+1)510-486-5311E-mail: [email protected]

[b] L. K. TakahashiDepartment of ChemistryUniversity of California at BerkeleyBerkeley, CA 94720 (USA)

Chem. Asian J. 2011, 00, 0 0 2011 Wiley-VCH Verlag GmbH & C o. KGaA, Weinheim 1& &&

These are not the final page numbers!
http://www.interscience.wiley.com/http://www.interscience.wiley.com/


2/12

tion of these molecules. To this end, laser-desorption massspectrometry coupled with synchrotron vacuum ultraviolet(VUV) postionization is being developed. Single-photonionization (SPI) with VUV radiation, when combined withhigh spatial resolution, could offer unprecedented chemicalspecificity for mass-spectral imaging of organic structures

and surfaces.In an effort to understand VUV photoionization of thekinds of molecules expected to be important in biologicalsystems, aerosols, and soil organic matter, a number of testmolecules that range from DNA bases, environmental ex-tracts (humic acids), plant biopolymers, and long-chain hy-drocarbons, were selected for study. To fully exploit thechemical specificity of SPI for mass-spectral imaging, thefragmentation of the parent cation must be minimized. Pre-serving parent molecular ions in a mass spectrum is essentialfor generating chemically detailed maps of complex samples.It is well known that the amount of fragmentation in a massspectrum is extremely sensitive to the amount of internal

energy within the parent ion or neutral molecule.[13,14]

Clear-ly, threshold ionization reduces the amount of fragmentationin a mass spectrum by minimizing the internal excitation ofthe molecules during the ionization step.[15]

For large molecules (long-chain hydrocarbons or biomole-cules), the internal energy imparted to the neutral moleculeto introduce it into the gas phase is equally as important asthe energy imparted upon photoionization. To preserve thefragile molecular ion, supersonic jet cooling of the internaldegrees of freedom is usually required.[16,17] Since supersonicjet cooling cannot be easily incorporated into surface-imag-ing experiments, the internal energy of the ion or laser-des-orbed neutrals could pose a significant problem in imaging

mass spectrometry. To evaluate this potential obstacle, ex-periments were conducted by combining laser desorptionwith VUV postionization. These experiments will show thatthe neutral molecule can be detected through its parent mo-lecular ion with high signal-to-noise ratios and that thedegree of fragmentation can be finely controlled or, in somecases, eliminated entirely.

Early reviews of attempts to quantify energies impartedinto molecules formed in laser desorption are provided byLevis and Hanley.[18,19] Molecular-level understanding of de-sorption processes, particularly as applied to MALDI, byusing molecular dynamics simulations and experiments have

been attempted by Knochenmuss, Zhigilei, Garrison, andco-workers.[2022] Desorption dynamics have also been exam-ined by using a number of different techniques, particularlywith relevance to organic aerosols, in which the ionizationstep is separated from the desorption step of the experi-ment.[2326] Very recently, the internal energy of tryptophanand thymine formed upon ion sputtering was quantified,and similar methods will be used here to ascertain internalenergies from laser desorption.[14]

Results and Discussion

Internal Energy

Previous VUV secondary neutral mass spectrometry (VUV-SNMS) experiments have revealed that neutral biomole-cules released from a surface after sputtering by 25 keV

Bi3+

ions have about 2.5 eV of internal energy.

[14]

Althoughsuch high-energy ion beams are well suited for high spatialresolution imaging down to 100 nm, this energy can yieldsignificant amounts of molecular fragmentation. In thisstudy, the energies imparted into neutral molecules desor-bed by a 349 nm laser are investigated using two differentmethods: one by utilizing fragmentation ratios and thesecond by measuring the appearance energies of fragments.

Internal Energy from Fragmentation Ratios

For the measurement of internal energy of hydrocarbons,the unbranched, long-chain hydrocarbon triacontane(C30H62) was chosen. Previously it was observed that the

level of fragmentation of the parent molecule from pure tri-acontane aerosols strongly depends on the temperature atwhich the aerosols were flash vaporized.[15] The dependenceof the parent molecule fractional population on temperaturewith 10.7 eV photoionization was also determined (seeFigure 2 in Ref. [15]). Using this dependence, triacontanecould be used as a molecular thermometer by observing theratio of the parent and fragments and fitting the obtainedfractional population into the calibration curve.

Owing to the very different experimental conditions usedin Ref. [15] and in this work (evaporation of aerosols versusdesorption from a bulk surface), the applicability of the aer-osol curve that relates the parent molecule fractional popu-

lation to the molecular temperature was verified. A few mil-ligrams of triacontane were placed on a heating/coolingstage inside of the experimental apparatus and heated to333, 353, and 373 K. The mass spectrum, recorded for tria-contane evaporated at 373 K and postionized with 10.5 eVsynchrotron VUV radiation, is shown in Figure 1 a. Themass spectrum has a strong peak at m/z 422, which corre-sponds to the triacontane parent, and a number of peakswith lower m/z, the intensities of which are much smallerthan the intensity of the parent peak. The elevated baselinefollowing the parent peak is due to the continuous postioni-zation of evaporating triacontane by the quasi-continuous

synchrotron VUV radiation.For an evaporation temperature of 373 K, the parent con-stitutes 73% of the total signal and the fragments make up27%. The good correlation of these values, as well as thoseat 333 and 353 K, with the values obtained from the aerosolexperiment is shown in Figure 2. This assures the applicabili-ty of the temperature dependence of the triacontane frac-tional population obtained in Ref. [15] to the current experi-ment.

The mass spectrum obtained for laser-desorbed triacon-tane is shown in Figure 1b. For the collection of the massspectrum, a 5 ms extraction pulse was applied to the ionoptics 1 ms after the desorption laser pulse. Synchrotron

&2& www.chemasianj.org 2011 Wiley-VCH Verlag GmbH& Co. KGaA, Weinheim Chem. Asian J. 0000, 00, 0 0&&


FULL PAPERS


3/12

VUV radiation of 10.5 eV was used for postionization. Inthis case, most of the triacontane parent is fragmented, andthe mass spectrum exhibits a fragmentation pattern domi-nated by sequential elimination of CH2. The fractional pop-ulations in this case are 0.09 for the parent molecule and0.91 for the fragments. These values are fit to the fractionalpopulation dependence curve shown in Figure 2. The tem-perature that corresponds to the observed laser-ablated tria-contane fractional population corresponds to an internaltemperature of 670 K.

To determine the corresponding energy for an internaltemperature of 670 K, the average vibrational energy of tria-

contane is approximated by a collection of harmonic oscilla-tors [Eq. (1)]:[13,14]

Evib Xs

i1

hviehvi=kT 1

1

The molecular structure of triacontane was optimized andthe vibrational frequencies were computed by using Gaussi-an 03[27] structure calculation software at the B3LYP/6-311Glevel of theory. Using this approximation, it was found thatthe temperature of 670 K corresponds to an internal energyof 4.6 eV. This value of internal energy arises due to thelarge number of vibrational modes for triacontane (270versus 39 vibrational modes for thymine).

Internal Energy from Appearance Energies of Fragments

For laser-desorbed thymine (C5H6N2O2), a different ap-proach was used to measure the internal energy. In this ap-

proach, the energy tunability of the synchrotron VUVsource was utilized to collect mass spectra at various photonenergies. The signal intensities from the molecular parent(m/z 126) and the main fragment of thymine (m/z 83) arethen plotted versus photon energy to produce their photoio-nization efficiency (PIE) curves. PIE curves of the parentand fragment of thymine are shown in Figure 3.

The first set of data in Figure 3, shown by a black solidline, corresponds to PIE curves of low-internal energy thy-mine molecules generated in a molecular beam.[28,29] Briefly,thymine was heated to 490 K and expanded together with

Figure 1. Photoionization mass spectra (10.5 eV) of triacontane at variousexperimental conditions: a) thermally evaporated at 373 K and b) laser

desorbed.

Figure 2. Temperature-dependent fractional populations of triacontaneparent ions (solid line) and fragment ions (dashed line). Open and filledcircles correspond to the fractional population obtained using a heating/cooling stage. The fractional population corresponding to laser-desorbedtriacontane is shown by squares: fragment ions are indicated by the blackopen square and parent ions by the black filled square.

Figure 3. Photoionization efficiency curves for a) the thymine parent andb) its fragment m/z 83. Solid black lines correspond to the molecular-beam experiment, filled black circles correspond to thermal evaporationat 313 K, and open black squares correspond to the laser-desorption pos-tionization experiment. The arrows indicate ionization or appearance en-ergies.

Chem. Asian J. 2011, 00, 0 0 2011 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.chemasianj.org 3& &&


Vacuum Ultraviolet Photoionization


4/12

argon (backing pressure of 34.7 kPa) through a 100mmnozzle into vacuum, which led to adiabatic cooling of thethymine molecules. The experimental conditions in the mo-lecular-beam experiment were roughly comparable to theconditions used by Amirav et al., who observed a vibration-al temperature below 50 K for molecules with similar

weight.

[30]

The onset of the PIE curve of the thymine parent(Figure 3a) corresponds to an adiabatic ionization energy of(8.950.05) eV,[31] which correlates well with previous mass-analyzed threshold ionization[32] ((8.91780.0010) eV) andVUV-SPI ((8.870.05) eV[33] and (8.820.03) eV[34]) experi-ments. Two signal spikes, observed in the region from 11.6to 11.9 eV are due to two absorption transitions of argonused in the gas filter (at 11.62 and 11.83 eV). A correctionof signal intensity by using measured VUV photon fluenceleads to the appearance of the observed discontinuity.

The thymine fragment C4H5NO (m/z 83) appears in themass spectrum upon loss of HNCO (isocyanic acid). Its ap-pearance energy is (10.950.05) eV (see Figure 3b).

To obtain PIE curves of thymine that have more definedtemperature conditions, the thymine film sample was placedon a heating/cooling stage and heated to 313 K inside theimaging apparatus. PIE curves obtained in this case by usingVUV ionization are shown in Figure 3 by filled black circles.A good correlation of the PIE curves of thermally evaporat-ed thymine to those from the molecular beam was immedi-ately observed, thus demonstrating the small effect that thetemperature increase to 313 K has on the PIE curves.

Laser-desorption PIE curves for thymine and its fragmentm/z 83 are shown in Figure 3 by open squares. The ion-ex-traction pulse was applied 2 ms after the laser-desorptionpulse. Desorbed neutral thymine molecules were postion-

ized with tunable VUV radiation. The PIE curve for theparent is very similar to PIE curves for the molecular beamand thermally evaporated thymine in the energy range be-tween 8.010.6 eV. Above 10.6 eV, the PIE curve for laser-desorbed thymine starts to deviate from the PIE curve ofthermally evaporated thymine. It has been observed previ-ously that the increase in internal energy of different mole-cules has very low if any influence on the ionization energyof the parent molecule.[13,35,36] Such negligible thermal ef-fects on the ionization energy have also been observed forthymine.[14]

The ion signal curve for the laser-desorbed thymine frag-

ment m/z 83 differs from those observed in the thermalevaporation and molecular-beam experiments. The onset ofthe signal for laser desorption is around 10.2 eV, and is dif-ferent from the 11.05 eV of thermally evaporated moleculesand 10.95 eV for molecules in the molecular beam. Thecurves have similar shapes but in the case of laser desorp-tion, the onset is redshifted by 0.85 eV. This shift correlateswell with the PIE curve shape observed for the parent start-ing from 10.6 eV and is due to the fragmentation of theparent ions into m/z 83 mass channel.

Although the increase of internal energy has a negligibleeffect on the ionization energies, it leads to a noticeablechange in the appearance energy of the fragment.[3537] It has

been shown that the thermal (internal) energies stored inthe molecule and the energy added by photons are equiva-lent with regards to dissociative photoionization.[37] Thiswould suggest that the increase of temperature in the parentmolecule will eventually lead to a shift of the appearanceenergy of the molecular fragment by a value equal to the

change in internal energy.To determine the internal energy that corresponds to ashift of 0.85 eV in the appearance energy relative to ther-mally evaporated appearance energy, an approximationshown in Equation (1) is used. The vibrational frequenciesof thymine were computed by using the B3LYP/6-311+G(d,p) level of theory with optimized molecular structures ob-tained with the same level of theory. The internal energy ofthymine thermally evaporated at 313 K according to Equa-tion (1) is 0.15 eV. Similarly, the internal energy of laser-des-orbed thymine is 1.00 eV, which corresponds to an internaltemperature of 800 K. One should note here that the tem-perature of 800 K is a lower limit of the internal tempera-

ture, which corresponds to the case when 100% of vibra-tional excitation goes into the dissociation of thymine mole-cules. When this process is less effective, a higher internalenergy is required to explain the observed shift in the exper-imental PIE curve.

Desorption Dynamics and Translational Temperature

The translational temperatures of laser-desorbed neutralmolecules were also measured to provide a better under-standing of the desorption dynamics. In Figure 4, time-of-flight (TOF) distributions measured for adenine (C5H5N5),thymine (C5H6N2O2), cytosine (C4H5N3O), and uracil

(C4H4N2O2) are shown. All four distributions have a similarshape, but different starting times that vary from around44 ms for cytosine to around 50 ms for adenine. These differ-ences in starting times correlate well with the differentmasses of the molecules. Adenine has a mass of 135 Da, thy-mine 126 Da, cytosine 111 Da, and uracil 112 Da.

For analysis of the TOF distributions, a half-space Max-wellBoltzmann distribution was used. To directly fit the ob-tained experimental TOF curves, the MaxwellBoltzmanndistribution was modified to account for the TOF conditions[Eq. (2)]:

It $L

t t0 4

exp m

2kTL

t t0 2 !

baseline 2

Here L is a distance between the surface and ionizing VUVbeam, m is the mass of the molecule, k is Boltzmanns con-stant, T is the molecular plume translational temperature, tcorresponds to time of flight, and t0 appears in the equationbecause the starting point of the TOF distribution is shiftedduring the flight in a TOF mass spectrometer and is notequal to 0 ms. A baseline term appears in the equation to ac-count for background signal and for the broad peaks pro-duced by molecular fragments that overlap with the ana-lyzed parent peak.



FULL PAPERSM. Ahmed et al.


5/12

The fitted curves obtained for the experimental data byusing Equation (2) are shown in Figure 4. The excellent fitof the measured spectra with the idealized MaxwellBoltz-mann distribution suggests that the molecules in the desor-bed molecular plume have well-defined translational tem-perature. The temperatures obtained from the fits are:216 K for adenine, 329 K for thymine, 346 K for cytosine,and 298 K for uracil. By using these temperatures, one canderive the mean and most probable translational velocities.The values of the velocities for the molecules investigated inthis work are shown in Table 1 together with their transla-tional energy values. It is immediately apparent that thetranslational energies are quite low and are in a range of a

few tens of meV. The mean translational velocities for allfour bases are comparable, and for three of them the valueslie in range of 235257 ms1. Only adenine has a lower ve-locity of 184 ms1.

Comparably low velocities were observed previously forlaser-desorbed biomolecules. Spengler et al. found that thetranslational temperature for tryptophan (m/z 204) in theirexperiment was about 300 K,[38] which corresponds to a

mean velocity of 176 ms1

. Tsai et al. demonstrated that intheir experiment the velocity distributions of tryptophanand 2,4,6-trihydroxyacetophenone (THAP) cannot be fittedby a single MaxwellBoltzmann distribution.[39] Instead, theyused three MaxwellBoltzmann distributions to fit the ex-perimental curve. The most probable velocities for these dis-tributions were 61, 154, and 287 ms1 for tryptophan and127, 258, and 503 ms1 for THAP. In the investigation oflaser ablation of tryptophan-glycine (Trp-Gly), Elam andLevy observed a variety of average translational velocitiesand temperatures, depending on the experimental condi-tions. They noticed that the average velocity was in a rangeof 440 to 570 ms1 for Trp-Gly film with the addition of

0.1% of R6G laser dye. For the neat Trp-Gly films, the aver-age velocities were found to be larger, in the range of 670 to820 ms1.[40] Engelke et al. determined the velocity distribu-tion of phenylthiohydantoin-glycine (m/z=192), which has amean translational velocity of (1777) m s1.[41] Huth-Fehreand Becker measured the velocity distributions of neutralgramicidin S (m/z 1141) and ferulic acid (m/z 194), the latterof which was used as a matrix in their study.[42] They ob-served almost identical velocity distributions for both ofthese substances, despite their very different masses, withthe maximum of the distribution peaking between 300 and400 ms1.

As is apparent from these previous works, there is no de-pendence observed in the translational velocities of differentdesorbed molecules on their mass. The translational veloci-ties found in previous investigations and in the presentstudy lie in range of several hundred meters per second, andas shown in Ref. [40] and discussed below, they depend con-siderably on the experimental conditions.

Collisional Cooling in Laser Desorption

To understand the effect of desorption laser power on thetranslational temperature of molecules, the TOF distribu-tions were measured for adenine coatings at four different

Figure 4. Time-of-flight (TOF) spectra for: a) adenine, b) thymine, c) cy-tosine, and d) uracil. The spectra are shown together with half-spaceMaxwellBoltzmann fit curves. For adenine, the desorption peak poweris 451 MWcm2.

Table 1. Mean and most probable translational velocities and translation-al energies for biomolecules studied in this work.

Molecule Ttr [K][a] m/z vmean [m s

1][b] vprob [m s1][c] Etr [eV]

[d]

adenine 216 135 184 163 0.019thymine 329 126 235 208 0.028cytosine 346 111 257 228 0.030uracil 298 112 237 210 0.026

[a] Translational temperature. [b] Mean translational velocity. [c] Themost probable translational velocity. [d] Translational energy.





6/12

laser peak-power densities. Three of the resulting TOF spec-tra are shown in Figure 4a and Figure 5a and b. The laserpeak-power densities used in these measurements are451 MWcm2 for Figure 4a, 294 MWcm2 for Figure 5 a,and 194 MWcm2 for Figure 5b. The fit of the obtainedTOF distributions with the MaxwellBoltzmann distribution

[Eq. (2)] revealed translational temperatures of 140, 199,and 208 K for high, average, and low laser peak-power den-sities, respectively. The translational temperatures of ade-nine molecules at four different desorption laser peak-power densities are shown in Figure 5c. The dependence isobtained for two sets of data and is shown together with the

standard deviation of the measurements. A roughly lineardecrease in translational temperature with an increase in de-sorption laser peak-power density is observed. This counter-intuitive behavior when the lower translational temperaturecorresponds to higher desorption laser power could be ex-plained by the collisional cooling in the laser-desorbed mo-lecular plume, when the stronger laser pulse desorbs alarger number of molecules, which subsequently leads to alarger number of intermolecular collisions. It is also appar-ent that the fragmentation peak located between 45 to 50 msdecreases with increasing laser peak power, thereby suggest-ing again that there is collisional cooling in the desorptionprocess. There could also be coupling between vibrational

and translational temperatures, which could skew fragmen-tation distributions.

In previous molecular-beam experiments it was found thatincreasing the pressure of the gas that will expand leads toan increase in the number of collisions between the gasatoms or molecules and improved cooling of translational,vibrational, and rotational degrees of freedom.[18,43,44] Sincea similar cooling effect may be occurring during laser de-sorption, it is instructive to estimate the number of adeninemolecules desorbed by a single laser shot from the samplesurface and compare it to a model molecular-beam experi-ment during a time interval typical for laser desorption.

To estimate the number of laser-desorbed molecules, one

can use the following equation [Eq. (3)]:

N1V

MNA 3

in which N is the number of molecules of the substance,which has volume V, density 1 and molar mass M. NA corre-sponds to the Avogadro constant. To find the volume of ad-enine, which is desorbed by a single laser pulse, a combina-tion of optical and atomic force microscopies and opticalsurface profilometry was used. During sample preparation,it was discovered that adenine did not form a homogeneous

layer within the samples but rather tended to coalesce intomicrocrystals with a size of several microns. The volume ofthe crystals and the density with which they cover thesample surface was estimated by using the techniques men-tioned above. These techniques revealed that the averageheight of the adenine microcrystals was 0.9 mm and that onlyabout 30% of the sample surface was covered by adeninecrystals. The size of the laser spot was determined with opti-cal microscopy to have a diameter of 30 mm. From this, thenumber of adenine molecules that can be desorbed from thesample surface by a 30 mm laser-spot size was determined tobe 1.361012 molecules. The desorbed molecule signal decaytime profile suggests that approximately ten laser shots were

Figure 5. TOF spectra for adenine, desorbed under various conditions:desorption laser peak-power density of a) 294 and b) 194 MWcm 2. Thegray lines represent the fit of the experimental data with the MaxwellBoltzmann distribution. The dashed lines correspond to the fit of adeninefragment m/z 119, and the main features peaking around 51.5 ms corre-spond to the adenine (m/z 135) signal. Translational temperatures of theadenine molecules, obtained from the fit of experimental TOF distribu-tions for different laser peak-power densities, are shown in panel (c). Thesolid line in panel (c) corresponds to the linear fit of the data and isshown to guide the eye.





7/12

required to desorb all adenine from the sample surface.Hence 1.361011 adenine molecules were desorbed per lasershot.

To model the laser-desorption experiment as a molecularbeam, the following approximation can be used. For a per-fect gas (the conditions inside of the gas container with a

nozzle can be approximated as a perfect gas) the equationof state is [Eq. (4)]:

p N

VkT 4

in which p is a pressure of the gas, N is number of moleculesof the gas, V is the volume, T is the temperature, and k isBoltzmanns constant. By using Equation (4), the number ofmolecules N can be estimated. Ambient conditions will beused for the pressure and temperature of the gas (101 kPaand 300 K, respectively). To find the volume of the gas V,the nozzle is approximated as a cylinder. In this case, the di-

ameter of the cylinder (nozzle) corresponds to the 30mm di-ameter of the laser-desorption spot. The height of the cylin-der equals the product of gas velocity and time. A desorp-tion laser pulse width of 7 ns is used in this calculation. Themost probable speed (vprob=(2kT/m)

1=2), obtained from theMaxwellBoltzmann distribution for room temperature ade-nine gas is used as the velocity and is 192 ms1. This calcula-tion estimates 2.31010 adenine molecules escaping througha 30 mm nozzle over 7 ns.

These two rough estimations give approximately the sameorder of magnitude for the number of molecules that escapefrom either the surface of a sample or through a nozzle overa similar time. Thus collisional cooling, widely observed in

molecular beams, should play a significant role in the caseof laser desorption as well. This cooling leads to a decreaseof the translational temperature and should subsequentlylead to a reduction of internal energy.

Previously it was observed that increasing the laser flu-ence led to an increase in the number of desorbed species,and this agrees well with the collisional cooling in the laserdesorption plume proposed above. Puretzky et al. andErmer et al. observed an increase in the ion-desorptionyield of two MALDI matrices, 3-hydroxypicolinic and 2,5-dihydroxybenzoic acids, correspondingly, with an increase inlaser fluence.[45,46] In his investigation of laser sputtering of

highly oriented pyrolytic graphite,

[47]

Krajnovich observedan increase in the flux of desorbed neutral atomic carbonand carbon dimers and trimers with increasing laser fluence.He also observed that the mean translational energy ofthese species grew with an increase in laser fluence. A simi-lar behavior of translational energy was observed by Geor-giou et al. for ablation of chlorobenzene films.[48] On thecontrary, Grivas et al. stated that the Knudsen-layer temper-ature (translational temperature in the vicinity of the sur-face area of the laser-irradiated material) of polyarylsulfonefilms decreased with increasing laser fluence.[49] They ex-plained this behavior as we do here, which is to say, an in-creasing number of collisions among the desorption prod-

ucts when fluence was increased led to enhanced collisionalcooling.

Imaging Mass Spectrometry

The ultimate aim of these dynamics and energy studies is to

provide a knowledge base and an experimental platform toperform imaging mass spectrometry using laser desorptionwith VUV postionization of fragile organic molecules. Tothis end, a surface scan of a test sample was undertaken.The optical microscope image of the sample surface isshown in Figure 6a. The sample is a silicon wafer covered

Figure 6. a) Optical microscope and b) laser-desorption postionizationimages of photoresist features on a silicon wafer. In the optical image,the dark features correspond to photoresist and white features to siliconsubstrate. In (b), the intensity of m/z 165 signal (the lighter the signal,the higher the intensity) is shown. A and B show positions of signal pro-files shown in Figure 7b and c.





8/12

with a photoresist polymer etched with features of differentsizes. The sample field of view shown in Figure 6a is700 mm700 mm, and the smallest features are 5 mm in size.The mass spectrum of the laser-desorbed photoresist withphotoionization with 10.5 eV synchrotron radiation is shownin Figure 7a. The spatial distribution of a peak of average in-

tensity in the mass spectrum (m/z 165) is shown in Fig-ure 6b. For the scan, 256 laser shots per surface spot and5 mm step sizes between spots were used. The laser powerwas chosen such that no silicon ablation was observed. Forthe detection of the Si signal in the mass spectrum, signifi-cantly higher laser intensity is required.

Comparison of the optical microscope image of thesample (Figure 6a) with the laser-desorption image (Fig-

ure 6b) shows very good agreement and reproducibility ofeven the smallest features. Optical microscopy revealed thatthe size of the laser-desorption spots over the polymer isabout 10 mm. This can also be observed in cross-sections Aand B (shown in Figure 6b) of the laser-desorption image,intensity profiles of which are shown in Figure 7b and c. For

comparison, idealized profiles of the photoresist on the sur-face of the silicon wafer are also shown in Figure 7b and c.In Figure 7b, one may see that the experimental profileshave approximately a Gaussian shape. To test the size of thelaser beam, the profile of an ideal feature (shown by a blacksolid line in Figure 7b) was convoluted by a Gaussian with10 mm full-width at half-maximum. The resulting curve(dashed line in Figure 7b) demonstrates perfect agreementwith the experimental data and proves that the laser-desorp-tion spot size in this experiment reaches 10 mm.

This may be surprising since a 30 mm laser-spot size, ob-served for biomolecules, is reduced to 10 mm over the poly-mer film. This is due to the Gaussian spatial profile of the

laser-irradiation intensity, which leads to the following de-pendence of the squared spot diameter D2 on the maximumlaser pulse peak power P0 [Eq. (5)]:

[50,51]

D2 2w2 lnP0Pth

5

in which w is the actual beam radius and Pth is the thresholdfluence for the material. Figure 7d shows a linear depend-ence of the experimental squared spot diameter on the laserpeak power in a semilog plot and could be fit with Equa-tion (5). This means that choice of low laser powers willlead to a decrease in laser-spot size and subsequently to an

increase of imaging spatial resolving power. The Pth value isdependent on the particular compound being investigated,and can account for the larger spot sizes observed over thebiomolecules relative to the polymer photoresist.

It is interesting to note that previously in laser ablation,laser craters as small as 0.7mm were observed on high-ther-mal-conductivity samples upon 248 nm excimer laser irradia-tion with a 5.6 mm spot and 25 ns pulse.[52] The experimentsdescribed here, although conducted on low-thermal-conduc-tivity samples, do show a similar behavior, and featuressmaller than the laser-spot size can be resolved. We believethat with a judicious combination of laser-spot size and flu-

ence, VUV postionization imaging mass spectrometry willnot be limited by diffraction-spot size and submicron resolu-tion can be achieved. Efforts in this direction are underway.There is, however, a caveat, as has been shown in the previ-ous discussion on collisional cooling: increasing peak-powerdensity leads to enhanced cooling and lower translationalenergies. However this will lead to larger spot sizes.

A new laser-desorption VUV postionization imaging tech-nique and its applicability for imaging small features on apolymer surface have been demonstrated. This will find ap-plication in imaging a number of biological and geologicalsystems. Recently it has been used to study the effect of an-tibiotics on bacterial biofilms.[53] With the methods described

Figure 7. a) Mass spectrum of a photoresist, covering the sample used forimaging; b) and c) m/z 165 signal profiles, which correspond to sectionsA and B in Figure 6b, correspondingly. Solid lines in (b) and (c) are ide-alized feature profiles. Gray dashed line in (b) corresponds to the fit tothe experimental data accounting for the Gaussian profile of the laser in-tensity. d) Squared diameter of the desorbed area versus laser peakpower. The dashed line in (d) corresponds to fit of the experimental datawith Equation (5).





9/12

here, it is anticipated that chemical composition and changewithin real-world environmental samples can be visualizedwith molecular specificity.

To illustrate the capabilities of this method for chemicalanalysis, representative mass spectra from lignin, cellulose,and an extracted environmental sample (Suwannee River

humic acid) using laser desorption with VUV postionizationare shown in Figure 8. This chemical information, when

combined with molecular imaging,[54] will provide valuablenew information for studying chemically heterogeneous sys-tems. These systems were chosen to show that tunable VUVcoupled with laser desorption is capable of detecting molec-ular masses between 200500 Da. Although MALDI hasbeen utilized to study many biological systems, there are

few techniques that can be applied to small molecules innative environments, as pointed out recently in a review.[55]

Furthermore, previous attempts at laser-desorption ioniza-tion on similar systems showed extensive pyrolysis-type phe-nomena and fragmentation of molecules.[5661] A recentimaging study that used MALDI to investigate cellulose inpoplar stem has shown that biological matrix effects fromlignin and hemicelluloses reduced the sensitivity towards de-tection of cellulose.[62] It is believed that the techniques de-scribed here will provide a new way to visualize organicmatter with molecular specificity.

Conclusion

Imaging mass spectrometry that utilizes laser desorption ofmolecules and subsequent synchrotron VUV postionizationhas been developed at the Chemical Dynamics Beamline atthe Advanced Light Source. We characterized this technique

by the analysis of internal and translational temperatures oflaser-desorbed molecules. Two independent techniques showthat the internal temperatures of a linear long-chain hydro-carbon and DNA and RNA bases range from 670800 K.The translational temperatures for DNA and RNA basesare lower and lie in the range of 216346 K. A dependenceof translational temperature on desorption laser power wasobserved and was explained by effective collisional coolingof desorbed molecules. The number of molecules desorbedby a single laser pulse is estimated and compared to thenumber of molecules expanded during a typical molecular-beam experiment under comparable conditions. Estimationof a similar number of molecules under both cases suggests

efficient cooling during laser desorption. Initial applicationof laser-desorption imaging mass spectrometry to a polymerpattern on silicon demonstrates its ability to resolve 5mmfeatures on the sample surface with a 30 mm laser-spot sizeand 7 ns duration.

Experimental Section

The experiment was performed on a secondary ion mass spectrometer(SIMS) coupled to the Chemical Dynamics Beamline (9.0.2) at the Ad-vanced Light Source. The experimental apparatus was essentially thesame used recently for the investigation of fragmentation mechanism of

lignin monomers coniferyl and sinapyl alcohols. [63] That apparatus wasequipped with an desorption laser which allowed for synchrotron VUVlaser-desorption postionization mass spectrometry (VUV-LDPI MS). Thelaser used for the experiments was a 349 nm Nd:YLF laser, which wastriggered externally by the SIMS apparatus master clock and typicallyruns at 2500 Hz.

The laser and its focusing optics were mounted directly on the SIMS ap-paratus so that the laser beam irradiated the sample surface at an angleof 45 degrees. In the experiments described here, the size of the laserspot was approximately 30 mm in diameter. To achieve this spot size, aset of lenses was used. First a diverging lens (focal length F=15 mm)was used to increase the size of beam exiting the laser. The divergingbeam was then focused by a set of two converging lenses (F=175 and200 mm). The distance from the focusing lenses to the sample surfacewas about 22 cm.

The VUV-LDPI mass spectrometer works as follows: A laser pulse de-sorbs molecules from the sample surface. The desorbed neutral molecularplume starts to spread perpendicularly from the sample surface; theplume is intersected by a synchrotron VUV beam, which is directed par-allel to the sample surface and positioned approximately 2050 mm abovethe surface. The molecules, after being ionized by the VUV light, contin-ue to spread unaffected until application of an extraction electrical-fieldpulse. The extraction pulse is applied 1 to 5 ms after the desorption lasershot. This delay is used to accumulate more ions in the interaction regionand eventually obtain mass spectrum with a better signal-to-noise ratio.

To gather better statistics for the measurements, the sample was rasteredand the signal for each presented data set was the sum of mass spectracollected from several fresh spots (typically 2040) on the sample surface.For each surface spot, the data was collected for approximately 10100laser shots.

Figure 8. Photoionization mass spectra (10.5 eV) for laser desorbed a) Su-wannee River humic acid, b) lignin, and c) cellulose.





10/12

The cellulose, DNA, and RNA base samples (all from Sigma Aldrich)used in the current work were prepared by dispersing the chemical com-pounds in methanol. Only a few drops of the smallest, dispersed particlesin the solution were deposited onto a piece of a silicon wafer and airdried. The lignin sample was prepared by placing alkali lignin (Sigma Al-drich) into a beaker and adding methanol. Methanol did not dissolve allof the lignin; however, a few drops of the dissolved solution was deposit-ed onto the silicon wafer and air dried. The humic acid sample (Interna-

tional Humic Substances Society) was prepared by directly placing


11/12

[38] B. Spengler, U. Bahr, M. Karas, F. Hillenkamp, Anal. Instrum. 1988,17, 173 193.

[39] S. T. Tsai, C. H. Chen, Y. T. Lee, Y. S. Wang, Mol. Phys. 2008, 106,239247.

[40] J. W. Elam, D. H. Levy, J. Phys. Chem. B 1998, 102, 81138120.[41] F. Engelke, J. H. Hahn, W. Henke, R. N. Zare, Anal. Chem. 1987, 59,

909912.[42] T. Huth-Fehre, C. H. Becker, Rapid Commun. Mass Spectrom. 1991,

5, 378 382.[43] D. H. Levy, Science 1981, 214, 263 269.[44] S. Y. T. Van De Meerakker, H. L. Bethlem, G. Meijer, Nat. Phys.

2008, 4, 595 602.[45] A. A. Puretzky, D. B. Geohegan, Chem. Phys. Lett. 1998, 286, 425

432.[46] D. R. Ermer, M. Baltz-Knorr, R. F. Haglund, J. Mass Spectrom.

2001, 36, 538545.[47] D. J. Krajnovich, J. Chem. Phys. 1995, 102, 726743.[48] S. Georgiou, A. Koubenakis, J. Labrakis, M. Lassithiotaki, Appl.

Surf. Sci. 1998, 127, 122 127.[49] C. Grivas, H. Niino, A. Yabe, Appl. Phys. A 1999, 69, S159S163.[50] J. M. Liu, Opt. Lett. 1982, 7, 196 198.[51] S. Martin, A. Hertwig, M. Lenzner, J. Kruger, W. Kautek, Appl.

Phys. A 2003, 77, 883 884.

[52] I. Avrutsky, D. G. Georgiev, D. Frankstein, G. Auner, G. Newaz,Appl. Phys. Lett. 2004, 84, 23912393.

[53] G. L. Gasper, L. K. Takahashi, J. Zhou, M. Ahmed, J. F. Moore, L.Hanley, Anal. Chem. 2010, 82, 74727478.

[54] R. Y. Tsien, Nat. Rev. Mol. Cell Biol. 2003, Suppl, SS16 SS21.[55] A. Svatos, Trends Biotechnol. 2010, 28, 425 434.[56] S. M. Mugo, C. S. Bottaro, Rapid Commun. Mass Spectrom. 2004,

18, 23752382.[57] T. Ferge, F. Muhlberger, R. Zimmermann, Anal. Chem. 2005, 77,

45284538.

[58] V. Samburova, R. Zenobi, M. Kalberer, Atmos. Chem. Phys. 2005, 5,21632170.

[59] Z. Li, L. Q. Chu, J. V. Sweedler, P. W. Bohn, Anal. Chem. 2010, 82,26082611.

[60] S. Reale, A. Di Tullio, N. Spreti, F. De Angelis, Mass Spectrom. Rev.2004, 23, 87126.

[61] E. M. Pea-Mndez, D. Gajdosova, K. Novotna, P. Prosek, J. Havel,Talanta 2005, 67, 880890.

[62] S. Jung, Y. F. Chen, M. C. Sullards, A. J. Ragauskas, Rapid Commun.Mass Spectrom. 2010, 24, 32303236.

[63] L. K. Takahashi, J. Zhou, O. Kostko, A. Golan, S. R. Leone, M.Ahmed, J. Phys. Chem. A 2011, 115, 32793290.

Received: April 29, 2011Published online:&& &&, 0000



http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1021/jp111437ehttp://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/rcm.4757http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1002/mas.10072http://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.1021/ac100026rhttp://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.5194/acp-5-2163-2005http://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1021/ac050296xhttp://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1002/rcm.1635http://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1021/ac101667qhttp://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1063/1.1688995http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1007/s00339-003-2213-6http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1364/OL.7.000196http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1016/S0169-4332(97)00621-1http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1063/1.469186http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1002/jms.155http://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1016/S0009-2614(98)00013-Xhttp://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1038/nphys1031http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1126/science.214.4518.263http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1002/rcm.1290050811http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/ac00133a026http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1021/jp982545+http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/00268970701779671http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672http://dx.doi.org/10.1080/10739148808543672


12/12

FULL PAPERS

Desorption Dynamics

O. Kostko, L. K. Takahashi,M. Ahmed* &&&&&&&&

Desorption Dynamics, Internal Ener-gies, and Imaging of Organic Mole-cules from Surfaces with LaserDesorption and Vacuum Ultraviolet(VUV) Photoionization

Dont blow me up : An imaging massspectroscopy technique that uses laserdesorption and vacuum ultraviolet(VUV) postionization has been char-

acterized. It promises good spatial res-olution and leads to minimal molecularfragmentation.



Documents

Oleg 90 Desorption Dynamics Chem Asian J 2011