close

Вход

Забыли?

вход по аккаунту

?

Multiphoton-Ionization-Mass Spectrometry (MUPI-MS) [New Analytical Methods (34)].

код для вставкиСкачать
Volume 27 - Number 4
April 1988
Pages 447-592
International Edition in English
Multiphoton-Ionization-Mass Spectrometry (MUPI-MS)**
Methods
By Jurgen Grotemeyer* and Edward W. Schlag
(34)
Mass spectrometry is one of the most important analytical tools in chemistry, biology, medicine and related areas. During the past 30 years, methods have been developed, both for
the qualitative as well as the quantitative analysis of a wide variety of substances. The introduction of lasers into chemistry has also profited mass spectrometry, since its nonlinear
properties and its tunability open u p fundamentally new frontiers. Multiphoton ionization
mass spectrometry combines UV-spectroscopy and mass spectrometry, thus providing a
two-dimensional method that enables substance-specific and even state-specific analyses.
This progress report presents the fundamentals and possibilities of MUPI mass spectrometry and discusses investigations on amino acids, peptides, chlorophylls and sugars. For each
substance, multiphoton ionization can be tuned in such a way that only the molecular ion is
formed. An increase of laser intensity induces substance-specific fragmentations of the molecule, thus contributing to a quick and easy identification of the substance.
1. Introduction
Analytical tasks, i.e., the identification of a substance
and the elucidation of its structure, have always been the
chief concern of mass spectrometry. On the one hand, a
compound can be identified via its molecular ion, which
leads to the exact determination of the molecular weight
and thus to the elemental composition and the empirical
formula of the molecule, while on the other, a fragmenta-
tion of the molecule giving a specific fragmentation pattern is also important for identification and structure elucidation."]
Informative mass spectra are only to be expected, however, if the four principal technical prerequisites of mass
spectrometry are met with:
-
-
[*] Ilr. J. Grotemeyer, Prof. Dr. E. W. Schlag
lnstitut fur Physikalische und Theoretische Chemie der
Technischen Universitat Munchen
Lichtenbergstrasse 4, D-8046 Garching (FRG)
I**]
A number of abbreviations exist for photon ionization. All terms like
MPI, REMPI, R2PI o r R E M P I I F describe a two-photon excitation
leading to ionization. For multiphoton ionization we prefer to use the
generic term MUPI, because, as described in this paper, three-, four- o r
multiphoton ionizations via resonant intermediate states as well as nonresonant absorptions can lead to ionization.
A n y e w G e m . Inr.
Ed. Engl. 27 11988) 447-459
-
-
The substance under investigation must either be evaporated as an intact neutral molecule and ionized later or
must be vaporized and ionized in one step.
Evaporation and ionization should be soft enough to
ensure that the signal of the molecular ion is detectable
with adequate intensity. However, the signals of substance-specific fragmentation should also be observable.
The ions produced must be separated according to mass
with the highest possible accuracy.
Finally, these ions must be detected without any discrimination due to their absolute mass.
0 VCH Verlagige.seN.schaft mhH.
0-6940 Weinheim. 1988
0570-0833/88/0104-00447 S 02.5OiO
441
During the past decade, several techniques have been
developed that can fulfill these demands, at least in part,
and the mass range for routine investigations has been increased from some 100 Da to some 1000 Da. The driving
force thereby has been the ever increasing need to analyze
larger molecules, especially in biochemical investigations.
In this progress report it will be demonstrated that the
use of a laser light source for "optical ionization" is a new,
generally useful ionization method in mass spectrometry.
Coupling mass spectrometry with the tunable laser ionization results in a two-dimensional analytical method that
gives absorption spectroscopic as well as mass spectrometric data. This technique is somewhat equivalent to other
well known two-dimensional methods such as the coupling
of mass spectrometry with gas chromatography (GC/
MS),['] with liquid chromatography (LC/MS),[31 or even
with mass spectrometry itself (MS/MS).I4'
impact ionization (El), ionization via charge exchange
(CE), and one-photon ionization (PI).[*1 Hence, MUPI
must be compared with these methods.
Tunable multiphoton ionization was first used by Boesl
et al.Li21
and Zundee et a1.['31in 1978, and since then has
been discussed in several review^."^-"^
In the conventional ionization methods, an energy that
is higher than the ionization energy is supplied to the molecule in one step, so that fragmentations occur as well as
ionization, whereas in MUPI the energy of a single photon
is lower than the ionization energy (Fig. 1). The wavelength of the photon is tuned to the energy difference between the ground state and an excited state of the neutral
molecule. For ionization, a multiple absorption of photons
in the molecule has to take place to accumulate the necessary energy.
a1
2. Fundamentals
The methods that have been developed in the last few
years for investigating involatile and large molecules[*l by
mass spectrometric techniques are often only applicable to
special problems. Of general use are the so-called particle
impact methods that utilize atomic particles to desorb ions
from a matrix into the gas phase: secondary ion mass
spectrometry (SIMS),IS1fast atom bombardment (FAB),@]
252Cf-plasmadesorption (25zCf-PD),[71
and laser desorption
(LD).@]In addition, direct chemical ionization (DCI)I9] and
field desorption (FD)[''] are employed for the investigation
of large molecules.
Although these methods differ widely in their technical
details they have one principal feature in common: The
desorption and the ionization are coupled directly."" A
steering of the ionization process, and thus of the degree of
fragmentation, is therefore impossible. In general, the molecular ion is not detectable, but several adduct ions are
found which stem from the addition of protons and/or alkali cations to the molecule. Often, there is also an ionization of the matrix molecules, which leads to a considerable
amount of non-substance-specific ions and thereby to further complications in the interpretation of the mass spectrum. To achieve a tunable ionization, the evaporation and
ionization must be carried out stepwise.
In a MUPI experiment the substance is first vaporized
by some method (see Section 2.2), then the molecules are
ionized in the gas phase with laser light and finally mass
analyzed. Such a two-step process aims at much higher ion
yields and is more flexible than one-step direct ionization.
2.1. Multiphoton Ionization
A molecule that undergoes multiphoton ionization becomes a radical cation. Radical cations also result from all
conventional impact ionization methods such as electron
[*I
The classification of the molecules discussed in this report into small
( <300 Da), middle sized (300-1000 Da), and large (> 1000 Da) molecules
is arbitrary.
448
bi
tI -
I tl l i
F?
F?
M
L
I
hv
M
Fig. I . The difference between a ) electron impact ionization (El) and b) mul9 = fragtiphoton ionization (MUPI). El =first ionization energy,
mentation ions. Other than in El, in MUPI the molecular ions are formed
with a narrow energy distribution and little excess energy.
e,e,
In general, the absorption of photons into a molecule is
a process of low efficiency. If, however, the wavelength of
the absorbed photons just corresponds to the energy difference between the molecular ground state and an excited
molecular state, the efficiency of the absorption process is
increased by several orders of magnitude (resonance enhancement),"', 19] particularly if the wavelength of the photon corresponds to a maximum in the electronic spectrum
of the molecule.
In the most simple case, the absorption of two UV-photons leads to ionization of the molecule. But a three- or
four-photon ionization is also possible (2b, c).[20.221
A multiphoton ionization with photons of different wavelength
(Fig. 2d) can also be realized. Figure 2e shows the special
case of a nonresonant two-photon ionization. This process
needs much higher laser intensities than the resonance-enhanced excitation.
MUPI succeeds in state selective formation of molecular
ions. As demonstrated by several molecules,[z21MUPI
leads to molecular ions that populate a well defined vibrational level, predominantly the vibrational ground state.
[*] In general, SI-, FAB-, CI-, 25zPD-and LD-mass spectrometry produce
adduct ions containing the molecule and a proton and/or an alkali cat-
ion.
Angew. Chem. Int. Ed. Engl. 27 11988) 447-459
a1
bl
cl
dl
el
Fig 2. Some features of multiphoton ionization (MUPI). S, indicates “allowed’’ intermediate state, mostly the S, state. But the usage of higher states
I S also possible. a ) “Normal” resonant two-photon ionization. b) Resonant
three-photon ionization via two intermediate states. c) Four-photon ionization via one intermediate state. d ) Resonant two-photon ionization with photons of different wavelength. e) Nonresonant two-photon ionization.
The excess of energy supplied in the ionization step is removed as kinetic energy by the electron leaving the molecule, so the molecular ions have equal or only slightly different energies (Fig. 1). In contrast, in the case of EI the
excess energy remains in the ions; this leads to a wide energy distribution, reflecting the energy distribution of the
impact electrons. With increasing laser intensity, absorption of photons by the molecular ions induces fragmentation reactions. Several mechanisms have been put forward‘”-’71 to explain these fragmentations. Generally accepted is the “ladder switching model”, which is supported by experimental findings’281and theoretical studies.[z’l As shown in Figure 1, the neutral molecule absorbs
two photons on its “absorption ladder” and is ionized.
Here the internal energy is only slightly higher than the
ionization energy. The molecular ion then absorbs one or
several photons, thereby starting the first fragmentation
reaction. The system then switches from the molecular absorption ladder to a fragment ion, which can absorb further photons on its absorption ladder, inducing further
fragmentation reactions. This absorption-fragmentation
mechanism may occur several times during typical laser
pulses of 6-10 ns and, with sufficient laser densities, result
in a total breakdown of the molecule into the atomic particles.
Because of the narrow energy distribution in the molecular ions in MUPI experiments, the absorption-fragmentation mechanism leads to an increased amount of metastable ions,”0.75.X7j Th‘is aspect is of importance in energetic
and dynamic investigations of elementary reactions in the
gas phase (see Section 3).
2. I . I . So3 Ionization
The absorption yield is clearly dependent on the intensity of the laser light. By keeping the intensity low, e.g. lo6
W/crnZ, the absorption of photons can be limited to form
exclusively molecular ions. I n Figure 3a, this is demonstrated with the example of p-xylene. Since an absorption
of photons by the molecular ions is almost impossible at
this low laser light intensity, no fragmentation takes place.
This possibility of inducing fragmentations in a controlled
way is unique in mass spectrometry. Even for large molecules the ionization can be tuned to form only molecular
ions.
Anyea’ Chem lnr. Ed En@ 2711988) 447-459
50
100,
c)
100
39,L1
t12
51
f
27
50-
n OQ
2L
I l%l
91
65
I,
I
,Id
1
1
,
m/z
-
I
I
106
I.
Fig. 3. MUPI-mass spectra of p-xylene. a) Soft ionization without frdgmentation of the molecular ion, laser intensity about lo6 W cm-2. b) Partially hard
ionization of p-xylene with little fragmentation, laser intensity about 10’ W
cm-l. c ) Very hard ionization with strong fragmentation, laser intensity about lo9 W cm-’ I=relative intensity. Excitation wavelength I =272 nm.
2.1.2. Hard Ionization
Increasing the laser intensity leads to an additional absorption of photons by the molecular ion, which then undergoes the energetically and kinetically most favored
fragmentation. Such a moderate ionization/fragmentation
is referred to as “partially hard ionization”. For the mass
spectrum in Figure 3b, the density of the photons was adjusted to a level that allowed observation of both the signals of the molecular ion as well as of those of the products of the first fragmentation, e.g. m/z 91.
Tuning to higher laser light intensities induces further
fragmentations, and, depending on the photon density, the
system can switch fragmentation ladders until a decomposition of the molecule into its atomic particles occurs. Figure 3c demonstrates this. Here ‘the fragmentation has advanced to the formation of atomic carbon. This means that
the p-xylene molecular ion must have absorbed about nine
photons of wavelength 272 nm. N o other mass spectrometric ionization method can accumulate such a high energy
in a molecule. The advantages from the analytical viewpoint, of this possibility of controlling the amount of fragmentation by tuning the laser light intensity are demonstrated in Section 3.
2.1.3. Wavelength Selectivity
In multiphoton ionization, not only the intensity but
also the wavelength of the light can be varied. The absorption maximum of a molecule and thus the position of the
449
resonant intermediate state are intrinsic properties of the
molecule and therefore substance specific. This is of great
importance for the investigation of mixtures, since the
component that has an intermediate state in resonance
with the wavelength used is preferably ionized; therefore it
is possible to detect traces in a mixture with increased sensitivity. Thus, in blood serum, for example, several metals
and organometallic compounds can be measured quantitatively in concentrations of about 1 ng per 1 mL
Thereby, both the electronic spectrum, measured for one
substance-specific mass, as well as the mass spectrum of
the substance, measured at a fixed wavelength, can serve
for the identification of the components. This method of
analysis has been used very s u c c e s s f ~ l l y , [ both
~ ~ - ~on
~ ~mixtures of several substances as well as mixtures of structural
One example of a selective excitation from a mixture is
shown in Figure 4.[391Here a sample of p-xylene with natural isotopic composition ( '2C8Hlo/'3C'2C7Hlo:100/8.8%)
was investigated. Recording a mass selected optical absorption spectrum of p-xylene ( m / z 106) and the '3C-isotopomer (m/z 107) yields absorption lines of the resonant
intermediate state shown on the left-hand side of Figure 4.
It can be seen that the positions of the absorption maxima
of the two molecules differ slightly.
I
18
106
m/z
-
107
Fig. 4. Left-hand side: mass selective absorption specmun ofp-xylene; top:
spectrum of the species m / z 106; bottom: spectrum of the species m / z 107.
Normally, the intensity ratio in the mass spectrum of p-xylene is m / z 106'
m/z 107=100:8.8. As the mass spectra show (right), this ratio is drastically
influenced (100:0.5 in A ; 50: 100 in B) by the choice of the excitation energy
for B).
for the MUPI-experiment (A,for A,
By choosing the appropriate wavelength from these absorption spectra, the signal of the mono- l3C-xylene molecular ion in the mass spectrum can be either amplified or
suppressed (Fig. 4, right).
The improvement of the mass spectrum of a substance
present in only trace amounts in a mixture is a possibility
which only multiphoton ionization can offer.
450
2.2. Vaporization of Molecules
The gas-phase UV/VIS absorption spectra of polyatomic molecules at room temperature normally show
broad and unstructured bands, since a large number of vibrational and rotational states are populated. To investigate molecules with MUPI-mass spectrometry, their thermal excitation should be as low as possible, so that the
broad absorption bands become sharp peaks and a selective excitation is possible. Cooling can be achieved by
isentropic expansion of a jet of noble gas into the vacuum.
Small amounts of the sample under investigation, which
must have a sufficiently high vapor pressure, are mixed
into the gaseous ~ t r e a m . [ ~ ~ - ~ ~ ]
In typical "seeded supersonic beam expansion" experiments, a gaseous mixture of argon and the sample is allowed to lead from a reservoir through a nozzle of diameter 100-500 pm. The temperature obtainable thereby is
about 2 K for the translational degrees of freedom, and
20 K and 50 K respectively for the rotational and vibrational degrees of
This method of injection resembles that of a G U M S coupling. The reservoir can easily be replaced by a gas chromatograph or a HPLC instrument. Several authors have described a GC/MS coupling
under the conditions of supersonic beam expansion and
multiphoton i o n i ~ a t i o n . [ ~Johnston
~ - ~ ~ I et al. have recently
reported the coupling of liquid chromatography with
MUPI-mass spectrometry.[471But these methods, using
MUPI, are still in the development stage.
Chromatography with supercritical gases, which has
found increasing interest during the past few years, is especially suitable for use with MUPI, since highly pressurized
gases are used as solvents. In the expansion step, the separation of these gases from the sample is very easy and no
clustering with the sample molecules occurs, a problem
that is often encountered in liquid chromatography experiments. Up to now, only molecules smaller than 250 Da
have been investigated with this r n e t h ~ d . [ ~ " . ~ ' ~
In an other approach, the thermospray
was
modified by Levy et aI.L541
to vaporize involatile samples
such as tryptophan as neutral molecules.
All these injection methods are well suited for the separation of mixtures and the evaporation of small, thermally
stable molecules, but they cannot be used for the investigation of large, thermally labile molecules. The methods developed for such compounds aim to evaporate and ionize
the molecules in one step. But in these methods more neutral molecules are desorbed than there are ions produced.
Thus, investigations by Cofferet al.[551have shown that in
laser desorption with an IR-laser, the yield of neutral molecules is much larger than the ion yield, as has also been
found to be the case by Budzikiewicz et al.["] and Campana
et al.[571for the FAB-method. Typical values for the ratio
of neutral molecules to ions is about lo4 : 1 in laser desorption experiments.[551Theoretical studies have been undertaken in an attempt to clarify the processes of desorption," '.'I but many details still remain unexplained.
To make use of the large ratio of neutral molecules to
ions we have utilized since 1983 a tandem arrangement of
laser evaporation of intact neutral molecules (LEIM) into a
supersonic beam followed by multiphoton i o n i z a t i ~ n . [ ~ ~ ~ ~ ~ '
Angew. Chem. ini. Ed. Engl. 27 (1988) 447-459
lotm Ar
b)
Fig. 5. Typical experimental set-up for the laser desorption into a supersonic
beam. a ) Pulsed valve with nozzle; b) probe tip mounted in the vacuum in
front of the nozzle: c) beam of laser light (produced by a C02-IR-laser); d)
skimmer. e) ion source.
In a typical LEIM-MUPI experiment (Fig. 5) the sample
is dissolved in a suitable solvent, the solution placed on a
probe tip, and the solvent evaporated. The probe tip is thus
coated with hundreds to thousands of layers of molecules,
and is now mounted in the vacuum close to the nozzle for
the supersonic beam. The molecules are then desorbed
from the surface by IR-photons into the expanding supersonic beam. As described above, the desorbed particles are
cooled in their internal degrees of freedom and at the same
time transported into the ion source. The pulsed system of
this experiment consists of a pulsed valve for the production of the supersonic beam and a small C0,-laser for the
desorption of the neutral molecules.
2.3. Mass Spectrometric Detection
In principle, the ions produced by pulsed MUPI can be
analyzed with any type of mass spectrometer. However, instruments that were developed for a continuous cyclic
scanning of the mass range, such as magnetic sector instruments, are much less suitable for coupling with MUPI than
quadrupole instruments, ICR-spectrometers,"' and timeof-flight mass spectrometers.
2.3.1. Quadrupole and ICR-Mass Spectrometry
In the first studiesr'2.61.62*
concerning the mass spectrometric investigation of ions produced by MUPI, quadrupole
instruments were used for the mass separation. Quadrupole instruments are often chosen in G U M S coupling, because they are easy to handle and have a rapid scanning
cycle. However, due to their limited mass range, about
2000 Dalton, and their low mass resolving power,'"] their
suitability for coupling with MUPI is somewhat limited.
Also the ion transmission of quadrupole fields is not very
high.
The coupling of ICR-mass spectrometry with MUPI has
also been d e ~ c r i b e d . [ ~ It
~ -offers
~ ~ I the advantage of high
mass resolution,lb6' but u p to now the mass range is still
limited to about 4000 Dalton. For technical reasons, the
production of ions should be very low in ICR-experiment~.'~''But MUPI produces, even with low laser intensities, considerable ion yields, so that the coupling of this
ionization method with an ICR-instrument is still complicated. There might be some possibilities in the future,
when ICR-mass spectrometry can be operated with high
mass resolving power and simultaneous recording of the
complete mass spectrum.
2.3.2. Time-of-night Mass Spectrometry
The mass spectrometric method that really takes advantage of the pulsed nature of ion production by MUPI, is
the technique of time-of-flight (TOF) mass spectrometry.16x1
T O F mass spectrometers are simple instruments with extremely high transmission. As opposed to all other mass
spectrometers, in these systems a complete mass spectrum
is recorded with every laser pulse. The fundamental parts
of a TOF mass spectrometer are the ion source, the drift
region (flight tube), and the detecting system. On coupling
with an EI-ion source, a mass resolving power of only 200400 could be achieved because of the broad kinetic energy
distribution of the ions generated. This low resolution is
the reason for T O F mass spectrometers having seldom
been used, despite being known for more than 40 years.
Because of the development of pulsed ion sources like
SIMS!691 o r plasma des~rption!~"]
(but also because of
MUPI[391), the interest in T O F mass spectrometers has
grown during the last few years. By using a two-step acceleration field ( W Z l e y - M c L ~ r e n ~and
~ ' ~ multiphoton
)
ionization, a mass resolving power of 400 was reached in a linear
T O F mass spectrometer for the simultaneous detection of
all
Several
to increase the mass resolution in
T O F instruments have been made, but with limited success. The most promising attempt was made by Mumyrin et
al."41 using an electrostatic field that reflects the ion beam
at the end of the drift section. This reflecting field not only
lengthens the flight distance, but is also able to compensate for most of the differences in the kinetic energy of
ions of one nominal mass and thereby the differences in
the flight time (Fig. 6). In our experiment, the ions produced by MUPI are accelerated by a constant field of
about 800 V and then drift through a length of 80 cm. At
the end of this field-free drift region, the mass separated
ion packets are slowed down by a short deceleration field
and reflected with only low energy in the following reflec-
['I I f R = ion Cyclotron Resonance.
[*"I Mass resolving power means the
ability to distinguish between two
neighboring masses. The signals of two masses are resolved when they
are separated by a "valley"; two definitions are usual in mass spectrometry. "lOo% valley" and "50% valley". I n this report the "50% valley"
definition is always used. For example a mass resolving power of 1000
separates the masses 500.00 and 500.50.
Angew Chem Inr. Ed Engl 27/1988) 447-459
Fig. 6. Schematic representation of the time-of-flight mas5 spectrometer. I )
Flight path of ions with low kinetic energy; 2) flight path of ions with the
same mass but higher kinetic energy. A) Evaporation chamber: B) beam of
ionizing laser light; C ) reflecting field; D) detector.
45 1
tor field. After leaving the reflector and deceleration field,
the ions fly through the drift region in the opposite direction till they impinge on the detector.
Ions with higher kinetic energy penetrate deeper into the
reflector field and therefore have a somewhat longer flight
path than ions with lower kinetic energy. As a consequence
the differences in the flight time of ions of the same mass
are compensated, simultaneously, for all masses. In 1982,
Boesl et al.[’’] succeeded in combining the multiphoton ion
source with a reflection time-of-flight (RETOF) mass spectrometer. In these first experiments, a mass resolving power of 3800 could already be reached. Lubrnun et al.[761and
El-Suyed et aI.[”] have also reported a coupling of MUPI
with a reflection T O F instrument, but with different experimental results.
By optimizing the parameters, considerably higher mass
resolutions are possible. In particular, it could be shown
that the grids used in a classical RETOF-mass spectrometer to define the electrostatical fields lead to a dispersion
of the ions and thereby affect the transmission and the
mass resolving power. In contrast a gridless ion mirror allows a clean reflection of the ion beam and is autofocussing. Boesl et al. succeeded in increasing the mass resolving
power to more than 10000139.7R-801
by incorporating a supersonic beam to cool the neutral molecules in their translational degrees of freedom, shortening the laser pulse
from 5-8 ns down to 1.5 ns, and lowering the laser intensity. Even further enhancement seems to be possible. Theoretical values of 100000 u p to 1000000 have been discussed for the mass resolving power in a RETOF-mass
spectrometer.[s01
In the mean time, the reflecting electrostatical field
method has also been used in combination with other ionization techniques (SIMS; 1985[”’ and 25ZCf-PD;
1987L821).
The advantages of time-of-flight mass spectrometers
compared to other mass spectrometers are manifold: first,
this technique offers an almost unlimited mass range; with
every laser pulse a complete mass spectrum is detected. A
registration of u p to 100 mass spectra per second is possible; this number is limited only by the capacity of the computer. Today, the mass resolving power and transmission
of RETOF instruments are already better than in the case
of quadrupole o r small magnetic sector instruments.’s31
Moreover, time-of-flight mass spectrometry has one physical feature that is unique in comparison to other techniques : the mass resolving power increases with increasing
mass. The detection of metastable ions[751(especially important for MS/MS investigations) is also possible because
of the incorporation of the reflecting field.
The location of the evaporation is confined to only a few
pm, which is especially important for the investigation of
surfaces. The time window of the sampling is limited by
the laser pulse to a few ns. Due to the high repetition rate
10-100 spectra can be measured for a signal lasting one
second, e.g. from a gas chromatograph.
3. Mass Spectra
MUPI-mass spectrometry was for a long time only used
for kinetic investigations of the fragmentation processes of
452
small molecules in the gas phase.[841It is well suited for this
purpose, not only because of the very sharply defined energy for the ionization and fragmentation of molecules, but
also because reactions of metastable ions can be investigated.
Since 1983, however, MUPI has found increasing application as a general method of mass spectrometry for the
investigation of organic and biochemical samples. A number of examples of such investigations will be outlined in
the following sections.
3.1. Small Molecules
Small arenes, such as ben~ene,[”~alkylbenzenes,[86-881
halogen-substituted aromatic corn pound^,^^^^ aromatic ami n e ~ , ” ~ - phenol‘9z1
~‘’
and aromatic aldehydes,[931as well as
organometallic compounds and several di- and triatomic
compounds have been investigated in detail with the help
of MUPI. In the main, these experiments were aimed at
elucidating the physical processes of ionization by multiphoton excitation or details of certain fragmentation reactions. Interesting reviews of such investigations are to be
found in articles by El-Sayed[”] and N e u ~ s e r . ~All
* ~ ]the
measured samples were volatile and their introduction into
the ion source was simple. In this report, we shall concern
ourselves with some mass spectra of small, biochemically
interesting substances that are either thermally labile or involatile. Amino acids, with their variety of polar, hydrophilic, and hydrophobic residues, are good models in this
respect.
The EI mass spectra of free amino acids have been well
d ~ c u m e n t e d . [ ~ ~In- ~most
’ ] cases, the signal for the molecular ion is either small or not detectable at all. Figure 7
a1
HO~CH,-CH-COOH
AH,
1107
I
t
54
I PA
L.. ‘..
MOO
181
I
u
200
l o0
I181
‘O0I
,107
1001
t
I I%l
501
d)
,;
!l,
,i’
,I jll
100
,
200
I M I HI@
’0°1
Fig. 7. Comparison of mass spectra of L-tyrosine. a) kl-mass spectrum; note
the intensity of the molecular ion signal at n d z 181. b) Soft MUPI-mass spectrum; exclusively the molecular ion is formed. c) Mass spectrum after partially hard MUPI: ionization wavelength I = 272 nm. d) CI-(methane)-mass
spectrum, corrected for the methane background.
Angew. Chern. Inr. Ed. Engl. 27 (1988) 447-459
shows two MUPI mass spectra of the amino acid L-tyrosine obtained after laser evaporation, and, for comparison,
an EI spectrum and a C I spectrum. The advantages of
MUPI are obvious. In the spectrum obtained after soft
MUPI (Fig. 7b), the molecular ion signal is the base peak,
whereas in the EI spectrum (Fig. 7a), the relative intensity
of this signal is only 5%. After background correction, the
C I mass spectrum (Fig. 7d) affords the signal of the Hadduct molecular ion with relatively high intensity, but the
identification of this ion is clearly complicated by the appearance of the signals of larger adduct ions. Increasing
the intensity of the laser light (partially hard ionization)
gives rise to the spectrum shown in Figure 7c. Despite pronounced fragmentation, the molecular ion signal is still remarkably intense. It has been shown[981from the mass
spectra of the amino acids that the fragmentation pathways remain virtually unchanged upon tuning the ionization wavelength by u p to 40 nm, but the relative intensities
of these fragmentations might change drastically. This behavior can be attributed to the different absorption cross
sections of the fragmentations. Extreme changes in the relative intensities are observed, for example, in the case of
tryptophan.'"''
In the case of aromatic amino acids, the wavelength
used for the ionization normally corresponds to an absorption in the aromatic ring system. But the amino acid arginine has no chromophoric group absorbing above 250 nm.
Therefore, in this case only a nonresonant absorption can
occur (see also Section 2.1). Arginine is known to be particularly thermolabile and neither an EI-spectrum nor a
CI-mass spectrum has been published u p to now. Field
desorption methods give a mass spectrum, but only the
protonated molecular ion is detectable.['001In the MUPI
mass spectrum (Fig. S), the signal of the molecular ion of
arginine has been measured for the first time. This ion is of
special interest, because arginine is extremely polar and is
one of the strongest organic acids.
tides containing aromatic amino acids. As example, we
shall discuss here phenylalanyl-tyrosine (Fig. 9). According to Lubman et aI.,"O3' the mass spectrum shows only a
HzN-
CH - CO
I
- N H - CH - COOH
I
C"2
Fig. 9. The dipeptide
phenylalanyl-tyrosine
OH
signal stemming from the loss of an hydroxy radical from
the molecule. This fragmentation seems to be an artifact of
the special experimental setup of the evaporation and ionization and can easily be avoided. Figure 10 shows the
mass spectra of this compound obtained by soft and by
partially hard MUPI under various experimental condit i o n ~ . [ ' ~In~ the
. ' ~soft
~ ~ ionization spectrum, only the signal
of the molecular ion is detected. With increasing laser light
intensity, a number of fragmentations occur. The important fragments have m / z 236 and m/z 208. They correspond to the formation of an acylium ion (m /z 236) on
rupture of the peptide bond and a further loss of CO from
this ion ( m / z 208). The cleavage of the peptide bond was
found to be the favored fragmentation in more than 50 investigated d i p e p t i d e ~ . [ ' ~ ~ ~ ' ~ ~ '
Figure 11 shows a mass spectrum of saccharose that was
obtained by nonresonant multiphoton ionization. Saccharose is a classical example for compounds that d o not give
molecular ions upon electron impact ionization. There
f "'1
I
A
I%]
'
100
200
MOO
H2N - C -NH-CHz CHI
AH
CH? $H-COOH
I
[M-NHJO
328
NHZ
30 0
LOO
B
big. 8. Mars spectrum of i-arginine after nonresonant MUPI. Ionization
wavelength 1 = 250 nm.
9
200
100
Of the amino acid derivatives the phenylhydantoins['O'l
have already been investigated by MUPI-spectrometry.[9x.1021 It was possible to detect less than subfemtogram
amounts of these compounds in the gas
Some
phenylhydantoins, however, proved to be impossible to detect by way of the molecular ion. In contrast, we were abIe
to showigx1that, by optimal tuning of the wavelength, the
ionization can be achieved softly enough to obtain exclusively molecular ions.
Lubman et al.L'031also reported that the molecular ion
was not always detectable in the investigation o f dipepAngew. Chem. Inr.
Ed. Engl. 27 (1988) 447-459
MOB
I 311
-
300
m/z
i
328
400
Fig. 10. Comparison of the mass spectra of phenylalanyl-tyrosine obtained
by soft (A) and the partially hard (B) MUPI. Ionization wavelength
A=272 nm. The signal in B at m / z 209 stems from the formation of the X;
fragment (see Fig. 13).
453
I lO0l
3.2. Middle Sized Molecules
I4 3
126
Il%l
60
50-
J
J,
144
i7'
I
The MUPI-mass spectra of several protected tripeptides
have also been recorded[1151
and compared with EI-spectra.["'l I n the mass spectra obtained under partially hard
ionization conditions the rupture of the peptide bond is
again dominating. Complete series of the A and B fragments are detected (Fig. 13). The nomenclature of the
163 179
96
b
1
I,.
I1
I/
MOO
I
-
I,
xm
,Y
2,
r ' ii'
,
4b0
300
m/z
a1
342
Fig. 1 I . Mass spectrum of saccharose after partially hard, nonresonant
MUPI. Ionization wavelength ,l=252 nm. The signal at m/z 43 stems from
an acylium ion.
bl
have been controversial discussions in the literature['071on
whether molecular ions of sugars actually exist under EI or
F D conditions. On the other hand, under MUPI conditions the molecular ion of saccharose is clearly detectable.['onl
The fragmentation reactions of saccharose (Fig. 12) are
well known from pyrolysis-FI mass
The main
fragmentation is the rupture of the glycosidic bond followed by the successive lost of water molecules.
8,
f-
R 2 ~ - ~ ~ - c ~ - - - - - - - - ~ Y
~*.; - c ~ - ~ 0 o
;73
R"
H 2 N - CI H - C O - N H - C H - C = 0
R'
A2
Fig. 13. a) Principal fragmentations of a peptide backbone and the nomenclature of the fragments (Roepstorffet al. [ I 171). The products of proton migrations are labeled, e.g., as A', A:, or A:' indicating the number of protons isomerized prior to the fragmentation. b) Main fragmentations of peptides that have been observed in MUPI-mass spectra.
H
HO
1
H
Hd
M = 3L2
C6H1206
k
\\\
acylium ions (B) and the acyliminium ions (A) follows the
concept of Roepstorff and Fohlrnan.["'] Besides the A and
B fragments containing the N-terminal end of the peptide
chain, also fragments of the Y'-series (Fig. 13) are detectable. These fragment ions contain the C-terminal end of the
peptide.
-C6Hl10g
m/z
-
Fig. 14. Mass spectrum of Leu-enkephalin after partially hard MUPI. Ioniza
tion wavelength ,l=272 nm.
M.126
Fig. 12. Fragmentation scheme of saccharose. Some of the decomposition
paths have been confirmed by the occurrence of metastable ions.
Further biochemically important small molecules whose
MUPI-mass spectra have been obtained include stecatecholamines,l"il neuroleptic drugs and
r o i d ~ ) " ~'07,1101
.
vitamins," 12. ' l 3 ] and purine
454
Figure 14 shows the mass spectrum of leucine-enkephalin, an unprotected pentapeptide)' '*I after partially hard
multiphoton ionization; the molecular ion is clearly recognizable. Other intense signals are those of the fragment
ions originating from rupture of the peptide chain (Fig.
15). Signals indicating fragmentations in the side chains
are only of minor intensity, as is characteristic for MUPI.
Once again, besides the complete series of A and B fragAngew. Chem. Inr. Ed. Engl. 27 (1988) 447-459
136
y’
P’
I
I
I
I
l a , X = M g , R1 = H
H , ,’
Tyr-Gly-Gly-Phe-Leu
H3C”
ments, the signals of some Y ‘ ions (Fig. 13) are also observed. It should be mentioned here that these complementary fragments are necessary in order to unequivocally
prove the sequence of a peptide.
Various smaller peptides, such as pentagastrin,’”91 the
“sleep inducing peptide” (SLP)[’061and others,”z01 have
been successfully analyzed by MUPI-mass spectrometry.
The favored fragmentation reaction in all cases is the
cleavage of the peptide bond.
The spectra of other classes of compounds, for example
of erythromycin, have also been recorded.”061 The molecular ions are always formed with high efficiency, and substance-specific ions can be detected without any protection
of the molecule prior to the measurements. MUPI enabled
the first successful mass spectrometric detection of porphyrins[’2’1and natural chlorophylls[’z21without protecting
groups. Natural chlorophyll and its derivatives could b e
detected in a biological sample containing about 30 different substances, including, inter alia, lipids and glycosides.
Figure 16 shows the EI- and MUPI-mass spectra of this
sample. It can clearly be seen that the wavelength selectivity of MUPI (Section 2.3.1) enables a separation of the
123
A
2L8
I
1
100
200
1
!
300
LOO
500
laO@
600
700
008,z, ,
900
Fig. 16. Comparison of the El- (A) and MUPI- (B) mass spectra of a biological sample from Spimlina geirlene. In the El-spectrum, ions of masses below
m / z 100 are not shown. The ionization wavelength was A = 2 8 1 nm in the
MUPI-mass spectrum. Identifiable are the following molecular ions: m/z
870 = pheophytin a: m / z 892 = chlorophyll a ; m/z 908 = 10-hydroxychlorophyll a.
Angew. Chem. I n t . Ed. Engl. 27 (1988) 447-459
C
/
Fig. IS. Structure and fragmentations of Leu-enkephalin
inn,
w
,
V
H
3
l b , X = M g , R1 = O H
1
lC,X=H.H, R :H
Fig. 17. Structures of the chlorophylls l a - c detected mass spectrometrically
in a sample of Spimlino geitierie.
substances, and three different chlorophyll derivatives
(Fig. 17) can easily be identified. These are pheophytin a
( l c ) at mass m / z 870, 10-hydroxychlorophyll a ( l b ) at
mass m/z 908, and chlorophyll a ( l a ) at mass m/z 892.
3.3. Large Molecules
Very informative mass spectra can also be obtained by
MUPI combined with LEIM in the mass range above 1000
Da, which hitherto was reserved to SIMS, the FAB method, or plasma desorption. Thus, it could be shown that a
soft ionization and controlled fragmentation of the unprotected decapeptide angiotensin I (Fig. 18) is possible by
multiphoton excitation.[’231Figure 19 shows the mass spectrum of this compound obtained after partially hard ionization. Fragmentations close to the peptide bonds lead to
intense signals; the complete series of A and B fragments
are clearly recognizable, and once again the important signals of the Y’ ions containing the C-terminus are also detected. As was found in the smaller peptide chains, fragmentations in the amino acid side chains are only of minor
importance.
Figure 21 enables a comparison of the results of peptide
investigations with the FAB-MS/MS-method14,
using
collisional activation and MUPI-mass spectrometry. The
FAB-MUMS experiment on substance P (Fig. 20) was carried out with a triple sector instrument (Biemann et al.[’251).
The protonated molecule was mass separated from the
fragment ions in the first stage and then fragmentations
were induced by collisional activation. By this method predominantly sequence ions of the A series are formed. The
signals of the ions of the B series are only of minor intensity. Ions containing the C-terminus are not detectable. In
addition, a large number of signals are observed that cannot be definitely assigned (Fig. 21a).
In contrast, the MUPI-mass spectrum (Fig. 21b) of this
compound shows signals for all ions of the A series and B
series with good intensities. Signals stemming from sidechain fragmentations are of minor significance. As in all of
455
A s p - A r g - V a l - T y r - I le - H i s -Pro -Phe- H i s - L e u
6L8
924
r-I
r
r-
L14
H*NCOOH
1 1 1 6
88
1000
619
C-NH
1137
756
506
1
2L L
Fig. 18. Structure and fragmentations of angiotensin I ( M = 1295)
I [%I
107
50.
81
116
91
A2
2CL
B2
272
A3
3 L3
r l
88
'3
371
AL
506
5
415
Bf
A5
667
y's
619
6L9
L
1
1
-
88
1028
78L
MOO
87
*6
756
L
I
A7
853
A
89
1165
v;
925
A8
1000
A9
1137
I ,
1
,
1295
m/z
1277
L
I .
-
Fig. 19. Mass spectrum of angiotensin I after partially hard MUPI. Ionization wavelength A =272 nm.
Arg - P r o -Lys -Pro - G I u -GIu -Phe - Phe - G ly - Leu- Me t - NH2
867
61 1
L6 4
NH
CH
1001
1171
1302
35L
129
t i p 20 Structure dnd fragmentations of substance P.
the above described MUPI-mass spectra of peptides, signals for fragments of the Y' series are also detectable;
hence, the peptide sequence can be unambiguously deduced from the spectrum.
Thus, both methods enable the sequencing of peptides
in the gas phase, but the FAB-MUMS experiment is much
more expensive than the investigation by LEIM-MUPImass spectrometry.
456
In Figure 22, a FAB-mass spectrum and a MUPI-mass
spectrum of an unknown sample are compared. The aim of
the experiment was to investigate whether the compound
shown in Figure 23, which should have been formed in a
chemical reaction,"261 was present in the sample. MUPI
leads to an intense signal corresponding to the molecular
ion of the compound, while FAB produces only the signals
of the fragmentation products. Thus, together with other
Angew. Chem. Inr. Ed. Engl. 27 (1988)447-4.59
7
analytical methods, MUPI-MS was able to give evidence
for the existence of the compound.
As a final exampie of the potential of MUPI-mass spectrometry in the investigation of large molecules, Figure 24
displays the soft ionization spectrum of bovine insulin."271
800
900
1000
1200
1100
1300
1400
100
23%
1
I
[%I
1000
A, 882
II%/-l
07
868
,
Bm
09
A 1086
Am1199
1088, 1129 1171
A,
1
1
NO@
1346
1330
1302,
Fig. 24. MUPI-mass spectrum of bovine insulin. Ionization wavelength
2 = 212 nm.
Fig. 21 Comparison of the mass spectra of substance P (a) from a FAB
MS/MS experiment and b) after MUPI
The most intense signal after multiphoton ionization is
that of the molecular ion at m / z 5729, which is far less
noticeable in the mass spectra recorded under FAB or PDconditions. It should be mentioned again that, unlike in
other methods, no background correction is necessary in
the case of the MUPI-mass spectra. Figure 25 shows the
experimental and calculated patterns of the isotopic distribution for the signal groups of the molecular ion and of
the A and B chains.
I
95.
90.
85.
80.
75.
70.
65.
25
20
15
10
5
0
300
.,
3000
2000
613
a)
A chain
B chain
m/z 233&
m/z 3395
M 00
m/z 5729
I
400
b) 1001
1 cnl
628
L -
200
100
300
400
500
600
I
760
MOO
11260
1001
t 501
I l%l
L-
1175
. .
Fig.
25. Compariaun of 1
the experimental intensity ratius (a) for the molecular
.
-
~
800
900
-
1100
1000
m/z
1200
1300
1400
ion of bovine insulin and the fragment signals for the A and the B chains
with the isotopic pattern (b) calculated from the elemental composition.
Fig. 22. t'ompariaon of the FAB-mass spectrum (a) and the MUPI-mass
spectrum (h) of the product shown in Figure 23.
(
COOCH3
{
f
COOCH3 COOCH3
<
COOCH3
big. 23. Pustulated btruclure of a compound in the sample used for the mass
spectra i n Figure 22.
A n g e w Chem. In(. Ed. Engl. 27j1988) 447-459
The superimposability of these two patterns for the signal group of the molecular ion proves that the insulin is
desorbed into the gas phase as intact molecule without any
addition of protons or alkali cations and is then ionized.
The main fragmentations can easily be identified o n the
basis of the signal for the A chain at m/z 2334 and the
signal for the B chain at m/z 3395. These results demonstrate that very large molecules can also be analyzed by
MUPI-MS. In the meantime, numerous interesting rnolecules, like gramicidin and
have also been investigated by this technique.
457
4. Conclusion and Outlook
The present review has outlined some of the advantages
that are gained by combining the laser evaporation of intact neutral molecules with multiphoton ionization and
mass spectrometry.
The MUPI-mass spectra of organic and biochemical
compounds differ in some particular characteristics from
the spectra obtained by other mass spectrometric methods.
MUPI-mass spectra always show an intense signal for the
molecular ion, thus allowing an easy identification. Since
the occurrence of fragmentations can be controlled by appropriately tuning the excitation energy, MUPI-mass spectroscopy enables very detailed analyses.
Investigation of the possibilities opened u p by this tunability, not only for mass spectrometric but also absorption
spectroscopic analysis, has just begun.
One important aspect of the method described here is
that it can be used for nonvolatile and labile compounds;
this is possible because of the spatial and temporal separation of the evaporation step from the ionization step. Thus,
the intrinsic difficulties caused by simultaneous vaporization and ionization in other mass spectrometric techniques
can be avoided. This has a positive effect on the signal to
noise ratio, so that, in contrast to other mass spectrometric
methods, no background correction is necessary in MUPIMS.
The very high sensitivity of multiphoton ionization is a
further advantage of MUPI-mass spectrometry. Femtogram and subfemtogram amounts of substances can easily
be detected by MUPI-MS, thus opening u p new applications in trace analysis.
The generation of ions by the influence of light is still
more expensive than their production by conventional
methods,”281but the advantages of this excitation method
are evident. On the other hand, analysis of the ions in a
time-of-flight instrument is very simple, and the overall
costs are comparable with those for other types of mass
spectrometers.
The range of potential applications of MUPI with respect to the identification of biological material is virtually
limitless. The intrinsic features of MUPI, such as soft ionization and the ionization efficiency as a function of the
wavelength, the different classes of compounds that can be
ionized, and the absolute size of molecules that can be examined, all point to areas of future research and exploitation.
For the investigation of molecules with masses between
20000 and 50000 Da, the main limitation is detection: all
commercially available systems have an upper mass limit
in this range. Solutions to this problem have been discussed in the literature,‘izY1
but further developments have
to be made. Although MUPI-mass spectrometry is still a
relatively young method, its future course of development
is already forseeable. Parallel with improvements in mass
resolving power and higher sensitivity MUPI-mass spectrometry will find increasing importance, in particular, for
the elucidation of the structure of biomolecules of higher
masses, but also for surface studies and material analyses.
Furthermore, the ability of the system to rapidly and selectively (by wavelength changes) mass analyze a sample
45 8
makes it well suited for the monitoring of technological
processes, with the additional dimension of time-selected
sampling.
The technique of producing ions with light is still in a
state of rapid development, but it is already clear that it
will find increasing application during the next few years,
not only in mass spectrometric analysis, but also in the
quality control of pharmaceutical products and in trace
analysis, as well as in resolving diverse clinical and pharmacological problems.
The authors thank the “Deutsche Forschungsgerneinschaft”, the “Fonds der Chemischen Industrie” and rhe
“Bundesministerium fur Forschung und Technologie ” f o rjinancial support of this work and Dr. U . Boesl and K . Walter
for their cooperation in the development of multiphoton ionization and the set-up of the time-of-flight mass spectrometer.
Received: July 27, 1987 [A 664 IE]
German version: Angew. Chem. 100 (1988) 461
Translated by Dr. B. Grotemeyer a n d Dr. R . C. Beauts, Garching (FRG)
[I] F. W. McLafferty. Interpretarron of Mass Spectra. University Science
Books, Mill Valley, C A , U S A 1980.
121 G. M. Message: Practical Aspects of Gas Chromatography Mass Spectrometry. Wiley-Interscience, New York 1984.
[31 C. G. Edmonds, 1. A. McCloskey, V. A. Edwards, Blomed. Mass Spectram. 10 (1983) 237.
[41 F. W. McLafferty (Ed.): Tandem Mas.? Spec/rornerr,v, Wiley-lnterscience, New York 1983.
[51 A. Benninghoven. W. Sichtermann, Anal. Chem. 50 (1978) 1180.
[6] M. Barber, R. S. Bordoli, G. J. Elliot, R. D. Sedgewick, A. N. Tyler,
Anal Chem 54 (1982) 645 A.
171 B. Sundqvist, R. D. Macfarlane, Mass. Specrrom. Reu. 4 (1985) 421.
[8] a ) P. G. Kistemaker. G. J. Q. van d e r Peyl, J. Haverkamp in H. R. Morris (Ed.). Sojr lonizafion Biologtcal Mucs Spectrometrv. Heyden, Lond o n 1981, p. 120; b) M. A. Posthumus, P. G. Kistemaker, H. L. C . Meuzelaar, Anal. Chem. X I (1978) 985.
191 M A. Baldwin, F. W. McLafferty, Org Mass Spectrom. 7 (1973) 1353.
[lo] H. D. Beckey, Field Ionization Mass Spectrometry, Pergamon Press, Oxford 1971.
[ I I] 1. S. T. Tsong, S. H. Lin, B. V. King in C . J. McNeal (Ed.): Mass Spectrometry in the Analvsis o/ Large Molecules. Wiley, Chichester 1986, p.
67.
1121 U. Boesl, H. J. Neusser, E. W. Schlag, Z. Naturforsch. A33 (1978) 1546.
1131 L. Zandee, R. B. Bernstein, D. A. Lichtin, J . Chem. Phys. 69 (1978)
3274.
[I41 E . W. Schlag, H. J Neusser, Acc. Chem. Res. 16 (1983) 335.
(151 H. Kuhlewind, ti. Boesl, H. J. Neusser, E. W. Schlag, Nachr. Chem.
Tech. Lab 32 (1984) 205.
1161 D. A. Gobeli, J. J. Yang, M. A. El-Sayed, Chem. Reo. 85 (1985) 529.
[ I71 V. S. Letokhov (Ed.): Laser Photoionization Spectroscopy, Academic
Press, New York 1987.
1181 G. Petty, C. Tai, F. W. Dalby, Phys. Reu. Lett 34 (1975) 1207.
(191 P. M. Johnson, C . E. Otis, Annu. Reu. Phvs. Chem 32 (1981) 139.
[20] H. Zacharias, H. Rottke, K. H. Welge, Appl. Phys. 24 (1981) 24.
1211 J. C. Miller, Anal. Chem. 58 (1986) 1702.
1221 Review: K. Kimura, Adu. Chem. Phys. 60 (1985) 161, and references
cited therein.
1231 D. H. Parker, R. B. Bernstein, J . Phys. Chern. 86 (1982) 60
[24] H. Kuhlewind, H. J. Neusser, E. W. Schlag, J . Phy.7. Chem. 89 (1985)
5600.
[25] J. Silberstein, N. Ohmici, R. D. Levine, J Phys. Chem. 89 (1985) 5606.
1261 F. Rebentrost, K. L. Kompa, A. Ben-Shaul, Chem. Phys. Lett. 77 (1981)
394.
1271 F. Rebentrost, A. Ben-Shaul, J . Chem. Phys. 74 (1981) 3255.
1281 a) U. Boesl, H. J. Neusser, E. W. Schlag, J . Chem. Phys. 72 (1980) 4327;
b) Chem. Phys. Let/. 87 (1982) I.
1291 W. Dietz, U. Boesl, H. J. Neusser, E. W. Schlag, S. H. Lin, Chem. Phys.
66 (1982) 105.
1301 a ) R. G. Cooks, J. H. Beynon, R. M. Caprioli, G. R. Lester: Metastohle
Ions. Elsevier, Amsterdam 1973; b) H. Kuhlewind, H. J . Neusser, E W.
Schlag, Laser Chem. 3 (1983) 3 ; c ) J . Chem. Phys. 82 (1985) 5452.
1311 a ) L. J. Moore, J. E. Hurst, E. H. Taylor, D. W. Beekman, M. T. Spaar,
Int. Phys. Con/. Ser. 84 (1986) 239; b) D. E. Pratt, J. Eagles, R. Self,
Masc Spectrom. 9 (1987) 407, and references cited therein.
Anqew. Chem. Int. Ed. Engl. 27 (1988) 447-459
1321 U. Boesl, H. J. Neusser, E. W. Schlag. J . Am Chem. Soc. 103 (1981)
5058.
1331 S. Leutwyler, U. Even, Chem. Phvs. Left. 81 (1981) 578.
[341 D. A. Lichtin, S. Ddtta-Ghosh, K. R. Newton, R. B. Bernstein, Chem
Ph.v.7. L e u 75 (1980) 214.
1351 R. Trembreull, D. M. Lubman, Anal. Chem. 54 (1984) 1962.
I361 R. Trembreull. C. H. Sin, P. Li, H M. Pang, D. M Lubman, Anal.
Chem. 57 (1985) 1186.
I371 D. M. Lubman, Anal. Chem. 59 (1987) 31A, and references cited therein.
[381 I>. M. Lubman, R. Trembreull, C. H. Sin, Anal. Chem. 57 (1985) 1084.
[391 U. Boesl, J. Grotemeyer, K. Walter, E. W. Schlag, Anal. Insfrum. (NY)
16 (1987) 151.
I401 R. E. Smalley, L. Wharton, D. H. Levy, Acc. Chem. Res. lO(1977) 139,
and references cited therein.
1411 J. B. Anderson, R. P. Andres, J. B. Fenn, Adu. Chem. Phys. 10 (1966)
275.
[421 J. M. Hayes, G. J. Small, Anal. Chem. 55 (1983) 565A.
[431 C . M. Klimcak, J. E. Wessel, Anal. Chem 52 (1980) 1233.
[441 R. B. Opsal, Thesis. Indiana University, Bloomington (USA) 1985.
1451 S. W. Stiller, M. V. Johnston, Anal Chem. 59 (1987) 567.
I461 G. Rhodes, R. B. Opsal, J. T. Meek, J. P. Reilly, Anal. Chem. 55 (1983)
280.
[47] G. R. Kinsel, K. R. Segar, M. V. Johnston, Pror. 35th Annu. Conf: Mass
Spectrom. Allied Top. (Denver, C O , USA 1987), p. 405.
[481 E. Klesper, A. H. Corvin, D. A. Turner, J . Org. Chem. 27(1962) 700.
[491 C. M. White, R. K. Houck, H R C CC J . High Resoluf. Chromatugr.
Chromafogr. Commun. 9 (1986) 3.
[SO] C. H. Sin, H. M. Pang, D. M. Lubman, Anal. Chem. 58 (1986) 490.
[Sl] H. M. Pang, C . H. Sin, D. M. Lubman, Anal. Chem. 58 (1986) 1581.
1521 C R. Blakley, M. L. Vestal, Anal. Chem. 55 (1983) 750.
I531 M. L. Vestal, G. J. Fergusson, Anal. Chem. 57 (1985) 2373.
I541 T. R. Rizzo, Y. D. Park, D. H. Levy, J . Am. Chem. Sot. 107 (1985) 277.
I551 a ) R. B. van Bremen, M. Snow, R. J. Cotter, l n f . J. Mass Spectiom. lun
Phys. 49 (1983) 35; b) R. J. Cotter, Anal. Chim. Acfa 195 (1987) 45.
1561 E. Schroder, H. Munster, H. Budzikiewicz, Org. Mass Specfrum. 21
( 1 986) 707.
I571 R. B. Freas, M. M. Ross, J. E. Campana, J. Am. Chem. SUC.107(1985)
6195.
[58] a ) R. N. Zare, R. D. Levine, Chem Phys. Leff. 136 (1987) 593; b) R. R.
Lucchese in C. J. McNeal (Ed.): Mass Specrrometry in the Analysis of
large Molecules. Wiley, Chichester 1986, p. 151, and references cited
therein.
1591 W. E. Henke, H. von Weyssenhoff, H. L. Selzle, E. W. Schlag, Verh.
Dfsch. Pfiys. Ges. 3 (1983) 139.
I601 H. von Weyssenhoff, H. L. Selzle, E. W. Schlag, 2. Nafurforsch. A 4 0
(1985) 674.
(611 L. Zandee, R. B. Bernstein, J . Chem. Phys 70 (1979) 1359, 2574.
[62] G . J. Fisanick, T. S. Eichelberger, B. A. Heath, M. B. Robin, J . Chem.
Phys. 72 (1980) 5571.
1631 M. P. Irion, W. D. Bowers, R. L. Hunter, F. S. Rowland, R. T. Mclver,
Jr., Chem. Phys. Leu. 93 (1982) 375.
[64] T. J. Carlin, B. S. Freiser, Anal. Chem 55 (1983) 955.
1651 T. M. Sack, D. A. McCrery, M. L. Gross, Anal. Chem. 57 (1985) 1291.
1661 Reported was a mass resolving power of 10" at mass 40: H. Bamberg,
K. P. Wancek, Proc. 35fh Annu. ConJ Mass Spectrum. Allied Top.
(Denver, CO, USA 1987), p. 1 1 16.
[67] K. P. Wancek, l n f . J Mass Specfrum. /on Processes 60 (1984) I I
1681 a ) W. E. Stephens, Bull. A m . Phys. Soc. 21 (1946) 22; b) A. E. Cameron,
D. F. Eggers, Jr., Reo. Sci lnstrum. 19 (1948) 605; c) M. M. Wolff, W.
E. Stephens, ibid. 24 (1953) 616; d ) H. S. Katzenstein, S. S . Freidland,
ibrd. 26 (1955) 324.
[69] A. Benninghoven, Z . Phys. 230 (1970) 403.
1701 D. F. Torgerson, R. P. Skowronski, R. D. Macfarlane, Biochem. Biophys Res. Commun. 60 (1974) 616.
[71] W. C Wiley, I. M. McLaren, Rev. Sri. lnstrum 26 (1955) 1150.
[72] R. B. Opsal, K. G. Owens, J. P. Reilly, Anal. Chem. 57 (1985) 1984.
[73] a ) W. R. Poschenrieder, I n f . J . Mass Spectrom. /on Phys. 9 (1972) 357;
b) M. L. Muga, Anal Insfrum.( N Y ) 16 (1987) 31.
[74] B. A. Mamyrin, V. 1. Karataev, D. K. Shmikk, V. A. Zagulin, Sou. Ph.vs
JETP (Engl. Transl.) 37 (1973) 45.
[75] U. Boesl, H. J. Neusser, R. Weinkaui, E. W. Schlag, J . Phys. Chem. 86
(1982) 4857.
[76] D. M. Lubman, W. E. Bell, M. N. Kronick, Anal. Chem. 55 (1983)
1437.
[77] T. L. Tai, M. A. El-Sayed, Chem. Phys. Leff.130 (1986) 224.
[78] K. Walter, U. Boesl, E. W. Schlag, Inf. J . Mass Specfrom Ion Processes
71 (1986) 309.
1791 R. Frey, G. Weiss, R. Kaminski, E. W. Schlag, Z . Noturforsch. A 4 0
(1985) 1349.
[SO] M. Yang, J. P. Reilly, Anal. Insfrum. INY) 16 (1987) 133.
[Sl] E. Niehuis, T. Heller, H. Feld, A. Benninghoven in A. Benninghoven
(Ed ): Ion Formation from Organic Solids IFOS 111. Springer, Berlin
1986, p. 192.
Angew. Chem. I n f . Ed. Engl. 27 11988) 447-459
1821 a ) R. D. Macfarlane, Pruc. 35fh Annu. Conf: Mass Specfrom. Allied Top.
(Denver, CO. USA 1987), p. 1146; b) P. W. Geno, R. D. Macfarlane,
Inf. J . Mass Specfrom. Ion Processes 77 (1987) 75.
[831 C. Brunnee, I n / . J . Mass Spectiom. Ion Processes 76 (1987) 125.
[MI See also H. J. Neusser, Inr. J . Mass Spectrum. Ion Processes 79 (1987)
141.
[851 H. Kuhlewind, H. J. Neusser, E. W. Schlag, I n ! . J. Mass Spectrom. lun
Phys. 51 (1983) 255.
L861 D. W. Squire, M. P. Barbalas, R. B. Bernstein, J Ph.vs. Chem. 87(1983)
1701.
1871 K. R. Newton, R. B. Bernstein, J. Phys. Chem. 87 (1983) 2246.
[SS] J. Grotemeyer, U. Boesl, K. Walter, E. W. Schlag, Adu. Mass Spectrom.
I 0 B (1986) 945.
[89I B. D. Kopiltz, J. M. McVey, J. Chem. Phys. 8 0 (1984) 2271.
1901 J. H. Catanzarite, Y. Haas, H. Reisler, C. Wittig, J. Chem Ph-ys. 78
(1983) 5506.
[911 D. A. Lichtin, D. W. Squire, M. A. Winnik, R. B. Bernstein, J . Am.
Chem. SOC. 105 (1983) 2109.
1921 R. S. Pandolfi, D. A. Gobeli, J. Lurie, M. A. El-Sayed, Laser Chem. 3
(1983) 29.
1931 J. J. Young, D. A. Gobeli, R. S. Pandolfi, M. A. El-Sayed, J. Phys.
Chem. 87 (1983) 2255.
[94] S . R. Heller, G. W. A. Milne (Ed.): EPA/NINH - Mass Specrral Data
Base, NSRDS, Department of Commerce, MA, USA 1978.
[951 W. Vetter in G. R. Waller, 0. C. Dermer (Ed.): Biochemical Applications
of Mass Specfromefry,First Suppl. Vol.. Wiley-lnterscience, New York
1980, p. 439ff.
[961 K. Biemann, J. A. McCloskey, J . A m . Chem. Sot. 84 (1962) 3 192.
[97] G . Junk, H. Svec, J . Am Chern Soc. 85 (1963) 839.
[981 J. Grotemeyer, U. Boesl, K. Walter, E. W. Schlag, I n f . J Mass Specfrom. /on Processes 78 (1987) 68.
[99] T. R. Rizzo, Y. D. Park, L. Petenau, D. H. Levy, J . Chem. Phys 83
(1985) 4819.
[I001 H. U. Winkler, H. D. Beckey, Org. Mass Spectrum. 6 (1972) 655.
[loll P. Edman, Acra Chem. Scand. 4 (1950) 283.
[I021 F. Engelke, J. H. Hahn, W. Henke, R. N. Zare, Anal Chem. 57 (1987)
909.
[I031 R. Trernbreull, D. M. Lubman, Anal. Chem. 59 (1987) 1003.
[I041 J. Grotemeyer, K Walter, U. Boesl, E. W. Schlag, Proc. 35th Annu.
Cunf Mass Spectrom. Allied Top. (Denver, CO, USA 1987). p. 191
[I051 J. Grotemeyer, J. Lindner, K. Walter, E. W. Schlag, 2. Narurforsch.. in
press.
[I061 J. Grotemeyer, E. W. Schlag, unpublished results.
[I071 a ) H. J. Veith, F. W. Rollgen, Org. Mass Specfrom. 2U (1985) 689; b) D.
E. C. Rogers, P. J. Derrick, ibid. 19 (1984) 490; c) P. J. Derrick, N.
Tang-Trong, D E. C. Rogers, ihid 20 (1985) 690.
[IOS] J. Grotemeyer, E. W. Schlag, NATO-AS1 Symp. Ser.. in press.
[I091 H. R. Schulten, W. Gortz, Anal. Chem. 5 0 (1978) 428.
[I101 J. H. Hahn, R. Zenobi, R. N. Zare, J. Am Chem. Soc. 109 (1987)
2843.
[ I I I] R. Trembreull, D. M. Lubman, Appl. Specfrosc 41 (1987) 431.
11121 R. Trembreull, D. M. Lubman, Anal. Chem. 59 (1987) 1082.
[ I 131 K Fujiwara, E. Voigtman, J . D. Winefordner, Spectrosc. Lett. 17 (1984) 9.
[ I 141 R. Trembreull, D. M. Lubman, Anal. Chem. 58 (1986) 1299.
[ I 151 J. Grotemeyer, U. Boesl, K. Walter, E. W. Schlag, Org. Mass Specfrom.
21 (1986) 645.
11161 H. W. Ronge, H. F. Grutzmacher, Anal Chem. Symp. Ser. 7 (1981)
255.
[ I 171 P. Roepstorff, J . Fohlman, Biumed. Mass Spectrum. 11 (1984) 601.
[ I 181 J. Grotemeyer, E. W. Schlag in A. Benninghoven (Ed.): /on Furmalion
from Organic Solids 1V. Springer, Berlin, in press.
[ I 191 R Frey, G Weiss, H. Kaminski, U. Boesl, J. Grotemeyer, K. Walter, E.
W. Schlag, Pruc. 351h Annu. Con5 Mass Specfrom. Allied Top. (Denver,
C O , USA 1987), p. 538.
[I201 J. Grotemeyer, E. W. Schlag, Org. Mass Spectrum., in press.
[I211 U. Boesl, J. Grotemeyer, K. Walter, E. W. Schlag, Adu. Mass Specfrom.
1UB (1986) 985.
[I221 J. Grotemeyer, U. Boesl. K. Walter, E. W. Schlag, J . Am. Chem. Soc.
108 (1986) 4233.
[I231 J. Grotemeyer, U. Boesl. K. Walter, E. W. Schlag, Org. Mass Specfrom.
21 (1986) 595.
[ 1241 a ) K. Biemann, S. A. Martin, Mass Sperfrom. Reu. 6 (1987) 1 ; b) B. N.
Green, R. S. Bordoli in S. J. Gaskell (Ed.): Mass Spectrometry in Brumedical Research. Wiley, Chichester 1986, p. 235; c) M. M. Sheil, P. J.
Derrick, ibid.. p. 251
[I251 H. A. Scoble, S. A. Martin, K. Biemann, Biochem. J . 245 (1987) 621
[I261 Kindly supplied by S. Schneider and E. Braslawski, Max-Planck-Instirut fur Strahlenchemie, Mulheim.
[I271 J. Grotemeyer, E. W. Schlag, Org. Mass Specrrom 22 (1987) 758.
[I281 See also K. Kleinermanns, J. Wolirum, Angew. Chem. 99 (1987) 38:
Angew. Chem. lnt. Ed. Engf. 26 (1987) 38.
[I291 H. Jungclas. H. Danigel. L. Schmidt. l n f . J. Mass Spectiom. Ion Phys.
52 (1983) 223.
459
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 204 Кб
Теги
analytical, mass, method, mupi, ionization, multiphoton, new, spectrometry
1/--страниц
Пожаловаться на содержимое документа