Fenselau 1983
Fenselau 1983
Fenselau 1983
ABSTRACT
INTRODUCTION
The technology, principles and ideas which were transferred a decade ago
by MacFarlane and Torgerson from nuclear to analytical chemistry have evolved
into the set of diverse yet related desorption techniques represented in this
vol une. Some of these phenomena have been better described by theoreticians
than others. Some of these methods have enjoyed more analytical success than
others. As a whole. these techniques enjoy the privilege of concomitant
interest from nuclear physicists, chemists and biologists. Much of the credit
for this unusually wide breadth of interaction goes to Professor R. D.
MacFarl ane.
The scope and analytical potential of techniques which use particle
induced desorption are illustrated in the first figure. On the left is the
molecular ion profile of bovine insulin (termed “the phthalate of the 80’s” by
*Opening Remarks from the Texas Symposium on Particle Induced Desorption held
May 18, 1983, in honor of Professor R. D. MacFarlane on the occasion of his
50th birthday,
1271
MacFarl ane) measured in a time-of-f1 ight mass spectrometer ionized by
(+20 charge state) parti cl es accelerated to 90 MeV in the Uppsal a accelerator
(ref.1). On the right is the protonated molecular ion envelope of porcine
insulin analyzed with a double focussin,g (energy and momentum) i nstrusent
of Nei r Johnson geometry ionized by xenon atoms accelerated to 8 KeV (ref.2).
Both spectra were acquired by particle induced desorption and computer support
was critical in both cases. The spectra can be differentiated by the lag time
of approximately one year between their acquisition, by the cost of the
i nstrunental systems used, by the characteristics of the bombarding particles,
by presentation of the sample in a sol id matrix in one case and a liquid
matrix in the other, and by the differences in the resolution provided by the
two analyzers.
Bovine
Insul in PORCINE INSULIN
Molecular
Ion
1
57302 IO
I
-----r-- \ n. ..
. . .
4469 i-99
m/z
If we lose the ability to distinguish and report the monoisotopic ion, can
we report instead the mast abundant molecular,ion as characteristic of an
unknown sample? Examination of Fig. 2 and 3 suggests that as the compound
becomes heavier it will be difficult to readily distinguish any particular
peak as most abundant. The envelopes become more nearly symmetrical. We
suggest that the most characteristic molecular weights which can be deduced
from spectra of peptides weighing more than 5000 are the average molecular
weights determined with good precision.
The question has been asked if high resolution accurate mass measurements
will provide the additional information on elemental compositions of middle
and heavy molecules as they do for species below 2000. The information in
TABLE 1 suggests that this will not be the case. TA8LE 1 lists the thirteen
most abundant isobaric ions which contribute to the ion current in the most
abundant peak of the insulin cluster. In total , more than 20 isobars
contribute to thi s peak.
TABLE 1
Composition of the most abundant+peak in the molecular ion cluster of porcine
insulin C256H382N65076S6 = (M+H) .
b 10253
5779.00 5778.59
A= + .41 Y8*- + g6.37
s=.53 s=.22
(8) (5)
SODlull HEPTYLSULFOlU4TEN-15
_-
C
the third category, and techniques for studying gas phase ion stabilities are
under development to address the fourth problem. The first , third and fourth
categories‘are addressed by other authors in this volune.
In considering desorption mechanisms it is important to distinguish the
nature of the ions observed. Quantitative measurements support the following
order of ease of formation:
R,N+ > (M+Na)+ B (M+H)+ > Mf
Preexisting ions, of which quaternary ammonium ions are a major example, are
most readily desorbed by 1 aser, fast atom bombardment, SIMS and field
deso,rption (ref -8). odd electron ions are formed by removal of an electron
under fast atom bombardment only when the sample has a very low ionization
potent i al , as in quinones and quinonediimines which readily undergo one
electron oxidations.
MacFarlane has included four “variations” in his definition of particle
induced desorption (ref .9) :
- heavy ion induced desorption
- secondary ion mass spectromctry
- fast atom bombardment
- laser desorption.
1249
t
IIONOISOTOPIC
!I+ 8110.5
8110
TABLE 3
Typical primary particle flux.
The term fast atom bombardment actually obscures the real innovation of
the two UMIST groups, which is the presentation of the sample in a liquid
matrix (ref .lO,ll) . Bombardment with atoms accelerated between 1 and 10 Kev
was reported in the literature at least as early as 1962 (ref.12) and more
recent studies in several labs (e.g. ref.13-15) suggest that the charge state
is not critical in this energy regime. In this context the term FAB would
appear to be unnecessarily redundant with the term SIMS. However, the physics
and chemistry of the mobile matrix are sufficiently distinct from those of the
solid matrix as to merit distinctive terminology.
The various contributions of the liquid matrix are summarized in Fig. 7
(ref.16). Chemical equilibria exist, of course, by which samples may be
i oni zed, and active protons may be exchanged for sodium cations, etc.
Solvation separates ions in solution and lowers the energy required to remove
the ions from the matrix (ref.17). The mobility of the liquid matrix
continually brings new sample ions to the surface and carries away the
radiolysis-1 i ke products of impact damage (ref .18). The analytical
effectiveness of the liquid matrix has led to its rapid acceptance world-wide.
However, mechanistic and theoretical considerations have addressed the solid
15
DESORPTION
SPUTTERING
-
.
GLYCEROL
bombardment.
16
NH2
1
O=P-OH 80
I
1
377
Hb OH Hd 6H
M = 874
363
(MH-79)+
796
+
(M+ Nal
I
a97
i I . II, 1
0 790 800 820 840 860 8430 So0 920
about 400. In the mass range of interferon (20,000) 1 part in about 1400
would need to be resolved. The ability to distinguish even smaller molecular
weight changes would permit recognition of disulfide reduction, stable isotope
labels, cationization with lithium ions, etc. Although the analytical chemist
must mediate wttat is technologically and economically feasible, the
sophisticated biochemist can fully utilize the best sensitivity, dynamic
range, mass range and resolution we can provide.
ACKNOWLEDGEMENTS
The work at Hopkins was supported by a grant from the National Science
Foundation, PCM 8209954.
20