SECTION: ELECTRON AFFINITIES of the ELEMENTS
Electron affinities of the elements (and molecules) are one fundamental property that
has been of special interest to workers who use negative ions for various applications, and
that are listed in the Handbook of Chemistry and Physics. Electron affinities (EAs) have been
calculated using several approaches and the magnitude of the results vary significantly, as can
be learned by studying the literature on this subject. Some of this literature is listed in the
references and bibliography of this section. Some values of EA have been measured
experimentally, especially using various photodetachment techniques. These measurements
have been made especially for the elements with high values of EA (from 1.5 to 3.5 eV), and
those values are stated with good accuracy and agreement among various workers is fairly
good. However, there are many elements that have only various calculated values or values
estimated from extrapolations, and these values are not consistent among workers. This is
especially true for elements that have lower values of EA, below about 1 eV. The elements for
which values of EA are established are primarily those from columns 16 and 17, plus some
light elements, plus some heavy/noble metals, namely H, C, N, O, F, S, Cl, Se, Te I, Pt, Au).
The quoted value of EA for several elements is zero (actually negative), from calculations
primarily, but also from some experimental measurements. These elements include N, Mg,
Mn, Zn, Cd, Hg, and the noble/rare gases, except for He, which has a very small EA of 0.078
eV. Low or unknown values of EA are quoted especially for the alkaline earth and rare earth
elements.
Zollweg1 has estimated the electron affinities (EAs) for the main groups of the third,
fourth, and fifth rows of the periodic table by horizontal analysis. Hotop and Lineberger 2 note
that his results agree to within 0.2 eV of the then available experimental data. Hotop and
Lineberger note that few experimental data existed as of 1975 for low EA elements. We
searched the literature between 1975 and 1995 for data, especially more recent data in J.
Chem Phys. and Prog. in Physics and found very few new experimental data. This was
another motivation for applying the different and sensitive technique of secondary ion mass
spectrometry (SIMS) for detecting negative ions. We can compare our experimental results
using SIMS with the calculations of Zollweg1 as now defined by Hotop and Lineberger.2
The calculations described above assume that the ground state of the negative ion has
dks2 configuration, but calculations are also made and reported for other configurations.
These quoted (calculated) values of EA are for isolated atoms of the elements or 'vacuum'
EAs, with no adjacent atoms of similar or dissimilar kind.
CHAPTER: ELECTRON AFFINITIES and SECONDARY ION MASS
SPECTROMETRY MEASUREMENTS
In the surface analysis technique of secondary ion mass spectrometry (SIMS), negative
ions (together with positive ions and neutrals) are sputtered from a solid surface, focused into
an ion beam, mass analyzed, and collected and measured accurately at a collector. Details of
this technique are given in the following chapters, along with experimental results for nearly all
elements, sputtered from a single matrix, such as Si or GaAs, or from their own matrices.
Advantages of this experimental approach include: 1) all elements of the periodic table can be
studied, and under identical conditions, 2) the detection sensitivity for ions is very large, so low
values of EA can be determined (0.3 to about 0.03 eV, depending on the element and its
SIMS sensitivity), and 3) ion implantation can be used to implant all elements into any solid
matrix with good accuracy of density so that SIMS measurements can be quantified with the
same good accuracy, allowing all elements to be sputtered from the same well characterized
matrix.
Values of a 'solid state' electron affinity can be determined from these measurements
and these values can be compared with the calculated (and sometimes measured) values of
'vacuum' EA. Issues of accuracy and uniformity associated with this approach are described
and discussed in the following chapters.
We have been able to measure the negative ion intensities for most of the elements for
which EA is not zero, sputtered from a common matrix (e.g., Si or GaAs) or from a matrix that
contains a known stoichiometric amount of the element of interest. In this work, we use
elements and compound materials as targets for SIMS, which uses a Cs primary ion beam
and negative secondary ion mass spectrometry. The use of a Cs primary ion beam mixes the
low ionization potential (IP) element Cs into the surface atom layers of the target element (or
compound material that contains the element of interest) thereby providing Cs atoms at the
dynamic sputtered surface as atoms are sputtered from that surface into the secondary ion
mass spectrometer. The low IP Cs atoms (3.9 eV) provide a source of electrons that are
easily attached to atoms being sputtered from that surface, thereby forming negative ions the
quantity of which can then be accurately measured by the secondary spectrometer. The use
of Cs enhances the formation (yield) of negative ions by three or four orders of magnitude over
the use of many commonly used primary ion beams of elements like the noble gases,
nitrogen, and oxygen. The rate of recorded secondary ion emission can be as high as 10 10
cts/s for the CAMECA 3, 4, and 5f sector magnet instruments used in this work. One issue in
this work is that the elements (atoms) may exist in a different configuration than the above
stated dks2, as they exist here in a solid matrix and not isolated in vacuum.
Hopefully, this work provides new insight into the nature of electron affinities of the
elements and molecules, and the magnitudes of electron affinity for many elements for which
low or zero values have been predicted.
Experimental errors can be introduced in this approach from several sources, which are
described below and which we attempted to minimize. The total experimental error from these
effects was estimated by combining all of these errors and is estimated to be ±15%. These
errors contribute to the spread of the relative yields given here. While our experimental errors
may be significant, we are able to estimate values of EA for stable or metastable negative ions
that have so far been difficult to determine using less sensitive techniques. We can compare
our estimated EAs with prior theoretical and experimental vacuum values for each element
and we do so here using the help of the comprehensive work of Hotop and Lineberger, 2 and
others.
When SIMS relative ion yields (inverse of RSFs) for negative atomic ions for emission
from the same 20 matrices are plotted versus electron affinity (EA) (published values for
isolated atom, or 'vacuum' values), a pattern results that is the same among the various
matrices, but which does not have a general single exponential dependence on EA. Some
elements yield no measurable negative atomic ions when sputtered from these solid matrices
in which they are dilute impurities (<0.5%) - - expected to be those with published values of EA
of zero or very small, less than about 0.2 eV. However, some elements that have zero value
of published 'vacuum' EA or values of 0.3 eV or less emit copious quantities of negative
secondary ions when sputtered from solid matrices. The elements of the periodic table have
been introduced into these several matrices by implantation as ions, and thus the total fluence
in the near surface region, which can be converted to atom density and concentration, is
accurately known. All elements can be implanted into any solid material to the concentrations
dealt with here. Ambiguities about whether secondary ions of a selected mass are those of
the element of concern or some molecular interference are essentially eliminated by the
demonstration of an implanted depth profile of the expected depth and shape at the selected
mass, and the use of isotope fingerprints (natural abundances of the isotope of the elements).
If mass interferences are still an issue, high mass resolution SIMS can be performed, subject
to the concomitant loss of sensitivity of about an order of magnitude.
The values of ion current for compound materials were adjusted for the fraction of each
element of interest contained in the compound material used for the analysis.
Such relative secondary ion emission data have been reported before for limited
numbers of elements in a few matrices, for example by Storms et al.3, Stevie and Wilson,4
Wilson and Novak5, and Wilson.6
The consistency of the pattern for relative ion yields (inverse of RSFs) for so many
materials (matrices), that now include metals, insulators, semiconductor, and organics/plastics,
confirm that the relative negative atomic secondary ion yields are universal and not matrixdependent and are related to the electron attachment of atoms immersed in a solid matrix of
similar or different atoms, where they can interact with nearest neighbor atoms that possibly
may have some effect on their solid state electron affinity or ionization potential when
sputtered from the surface of that solid matrix.
References
1. R.J. Zollweg, "Electron affinities of the heavy elements." J. Chem. Phys. 50, 4251-61 (1969)
2. H. Hotop and W.C. Lineberger, "Binding energies in atomic negative ions," J. Phys. Chem
Ref. Data 4, 539-76 (1975)
3. H.A. Storms, K.F. Brown, and J.D. Stein, Anal. Chem. 49, 2023-30 (1977)
4. F.A. Stevie and R.G. Wilson, "RSFs for positive molecular ions sputtered from Si and
GaAs," J. Vac. Sci. Technol. A9, 3064-70 (1991)
5. R.G. Wilson and S.W. Novak, "Systematics of SIMS RSFs vs IP and EA for
Semiconductors and Insulators," J. Appl. Phys. 69, 466-74 (1991)
6. R.G. Wilson, "O2 and Cs SIMS of rare earths and low EA elements implanted into
semiconductors and their EAs as estimated from SIMS measurements," Secondary
Ion Mass Spectrometry - SIMS VII, Benninghoven, Evans, McKeegan, Storms, and
Werner, Eds. [WIley, NY, 1990], pp.131-143
Other references
R.G. Wilson, F.A.Stevie, and C.W. Magee, Secondary Ion Mass Spectrometry [Wiley,
NY, 1989]
S.W. Novak and R.G. Wilson, "Ion yields from Si and SiO2 using Ar and O2 bombardment,"
SIMS VII, Benninghoven, Evans, McKeegan, Storms, and Werner, Eds. [Wiley, NY,
1990] pp. 87-90
F.A. Stevie and R.G. Wilson, SIMS VII ibid. pp. 683-86
S.W. Novak and R.G. Wilson, "AlGaAs" SIMS VI, Benninghoven, Huber, and Werner, Eds.
[Wiley, NY, 1990] pp. 303-06
R.G. Wilson and S.W. Novak, SIMS VI ibid. pp. ??
R.G. Wilson, "HgCdTe" J. Appl. Phys. 63, 5121 (1988)
D.S. Simons, P.H. Chi, P.M. Kahora, G.E. Lux, J.L. Moore, S.W. Novak, C. Schwartz, S.A.
Schwarz, F.A. Stevie, and R.G. Wilson, "Are RSFs transferable among SIMS
instruments?," SIMS VII ibid. pp. 111-14
Other Bibliography
Doty and Mayer, (Br) J. Chem Phys. 12, 323-328 (1944)
Pritchard, (many EAs) Journal?? 530-563 (1953)
Branscomb, (negative ions) Adv. Electronics Electr. Phys. 9, 43-94 (1957)
Branscomb, (photodetachment) Atomic and Molecular Processes [Academic Press,
NY/London, 1962], pp. 100-140
Moiseiwitsch, (EAs) Adv. Atomic Molec. Phys. 1, 61-83 (1965)
Kaiser, et al., (general) Z. Physik 243, 46-59 (1971)
Christophorou, (lifetimes of metastable negative ions) Adv. Electronics and Electr. Phys.
46, 55-129 (1978)
Massey, (negative ions) Adv. Atomic Molec. Phys. 15, 1-36 (1979)
CHAPTER: 'Solid State' Electron Affinities
and
Negative Atomic and Molecular Secondary Ion Mass Spectrometry Relative Sensitivity
Factors (RSF)
Some elements cannot be analyzed as solid matrices using SIMS because of their high
vapor pressures, or required low temperatures in the solid phase. Therefore, it is very difficult
to measure the relative negative atomic ion emission/yields from these elements in their own
solid matrix where their atom density is greater than 10 22 cm-3 and from which low yields
would be the most detectable. Even if this were accomplished, the question arises as to the
relative secondary ion yield enhancements/variations among the different matrices comprising
the different elements. An alternative is to measure their relative negative atomic ion
emission/yields from a solid (at room temperature) matrix of which they are a component and
have an atom density of the order of 10 22 cm-3. The questions still arises as to the yield
enhancement or variations among different compound matrices. Another alternative is to
measure the relative secondary ion yields of all of the elements implanted into one common
matrix that provides relatively good secondary ion yields, for example, Si, measuring their
relative sensitivity factors under Cs ion bombardment under a standard set of secondary ion
mass spectrometry (SIMS) conditions. Then an issue encountered is that the atom density in
the implanted profile must be kept below the density that might cause any alteration of the ion
yield from Si, typically 1% atomic or about 5x10 20 cm3, reducing the potential detectivity by
two orders of magnitude compared with the pure element. However, this approach ensures
that the relative ion yields do not vary significantly because of "matrix effect" variations among
the individual elemental matrices.
When the (universal) relative negative ion yields are plotted versus published values of
EA (measured for isolated atom or 'vacuum' values) for a common matrix, e.g. Si, for which
the most complete set of data exists (at least 82 elements), a single linear curve upon which
even most elements lie does not exist. This observation leads to the question as to whether
there is one value of EA for an atom in vacuum or not in immediate proximity to one or more
other atoms, and another value of EA for the same atom when it is immersed in a solid where
there are other atoms, nearest neighbor and more distant, that could alter the effective EA (or
IP) of that atom, or whether some published values of EA are not accurate. In this regard, we
note that values of electron affinity (vacuum) of the elements are apparently still evolving,
because a new list of values was published 1 that caused us to change a significant number of
them in a recent publication (most notably Sc, Ti, Fe, Y, Zr, Nb, Mo, Ta, W, and Pb),
compared with the values published in the Handbook of Chemistry and Physics. The values
for Sc and Y were changed from zero to the significant values of 0.19 and 0.31 eV,
respectively. Zero value of EA was still listed for some elements for which significant
quantities of negative secondary ions of elements sputtered from a matrix in which they are a
major constituent or as in impurity in a different matrix (e.g., Si) are measured, e.g., Be, Ca,
Sr, Ba, and Hf. The data presented here cause us to question some of those changes,
because some relative ion yield data are more consistent with the former values, while some
are more consistent with the new values. Still other data strongly contradict zero values for EA
(for example, Be, Ca, Sc, Ba, and Hf). Taken all together, these data indicate that additional
theoretical calculations of the EAs for atoms of the elements in vacuum and immersed in a
matrix of similar or different atoms may be necessary (solid state EAs). The data presented
here may offer experimental evidence to assist in such calculations, and certainly will offer
quantitative help to research.
The process of sputtering ions (atoms) from a surface involves the presence of
significant numbers (density) of the sputtering species in the surface from which the ions are
sputtered. In the cases we are discussing here, the sputtering element is Cs, the most
electropositive element, for purposes of enhancing the emission of negative secondary ions
from the composite surface, via a change of the work function of the surface and the resulting
effect on the ionization probability. The exact concentration of the sputtering (primary) ion in
the composite emitting surface is a function of certain other factors such as the angle of
incidence (but which is constant for the work reported here), but because the RSF is relative to
the matrix ion intensity, the consistent pattern of ion yield vs EA or IP is not affected. The
observation of a saturation in both the data for negative ion yield vs EA for elements with
values of EA greater than about 2 eV and for positive ion yield for elements with IP less than
about 5 eV, leads to the tentative conclusion that in both these cases 100% of the emitted
atoms are ionized, causing the observed saturation in RSF/ion yield. This conclusion is
consistent with the data obtained here, which exhibit a maximum EA of about 2.2 eV, while the
vacuum values for columns 16 and 17 are 2.2 to 3.4 eV. The RSF for secondary ion emission
is the ratio of intensities of the positive or negative ions of any element in a matrix to one
selected element that is a constituent (high concentration) of that matrix multiplied by the atom
density of that reference element in the matrix, under any set of instrument conditions, and is
essentially the inverse of the ion yield. Reference 2 gives a detailed explanation. Any change
of instrument conditions that effects the emission of all ions in the same way does not change
the RSFs. Therefore, RSFs are transferable among instruments,3 except when the instrument
has nonlinearities, such as quadrupole instruments, for which modest, but consistent,
differences are noted.
We implanted about 80 elements into Si and measured their relative negative ion yields
under the standard set of SIMS conditions in CAMECA 3f or 4f instruments. We also
measured the normalized relative atomic ion yields for 50 elements as solid element matrices.
We have also studied some of the elements that are difficult to analyze as solid elements, as
components of other solid materials, including Li, Na, K, Sr, Ba, Y, La, N, P, As, O, S, Se, F,
Cl, Br. The energies of the implanted “standards’ were between 200 and 700 keV in order to
place the peaks of the resulting depth profiles beneath the surface equilibration depths of
SIMS, to reduce any error in measured ion yield caused by this equilibration zone.
Simultaneous bombardment with electrons to provide charge compensation of dielectric
surfaces was used in all appropriate cases (dielectric materials).
(BaSr)TiO4 was implanted to study Ba and Sr in stable solid high density matrices.
Similarly, LiYF4 was used for Li and Y, Si3N4 for N, and rare earth fluoride glasses for some of
the lanthanum rare earth elements. Some lanthanide rare earths were studied as pure metals,
namely Nd, Gd, and Dy. Two actinide rare earths were studied as pure metals, namely Th
and U. Improved charge-compensation techniques were used for dielectric materials like
Si3N4. .
The SIMS instrument conditions for the measurement of relative sensitivity factors
(RSFs) are described in references 2 and 3. Higher fluences were implanted for the elements
with quoted zero or low values of electron affinity, namely Ca, Sr, Ba, Sc, Y, La, Mn, Hf, and
some of the lanthanide rare earths (Ce, Nd, Sm, Tb, Dy, Ho, Er, Tm, and Yb). Of these, Mn,
Hf, Nd, Gd, Th, and U were also studied both as elemental matrices and implanted into Si.
(Many of the elements were studied both ways.) Some observations for alkaline earth and
rare earth elements follow.
Be- is easily detected in Si, GaAs, and other matrices, and copious quantities of Be - are
sputtered from the Be matrix.
The detection of 40Ca- and 42Ca- in Si requires high mass resolution (HMR) because of
interferences from 28Si12C-, and 30Si2C- and 28Si14N-, respectively, which decreases the
detection sensitivity enough to preclude the detection of Ca - in Si with certainty. A lower limit
on the RSF is >3x1026 cm-3. Both 40Ca and 42Ca can be detected using CaSi-. For moderate
fluences of Ca, Ca- cannot be detected in GaAs either, apparently because of the presence of
enough Si and C/N to cause interferences from the same SiC/SiN masses. Again, Ca can be
detected using CaAs-. Ca- is easily detected when Ca- is sputtered from Ca metal.
The detection of 45Sc in Si requires HMR because of interference from 29Si16O-, which
decreases the detectivity enough to preclude the detection of Sc - in Si with certainty. A lower
limit on the RSF is >4x1025 cm-3. Sc is well detected, however, using ScSi-. See Fig. Ba-Sc.
45Sc-
also suffers from an interference in GaAs, although of lesser intensity than in Si; the
interference is probably the same as for Si. The lower limit for Sc- in GaAs is >5x1023 cm-3.
Sc is well detected in GaAs using ScAs-. Large numbers of Sc- are sputtered from Sc metal.
The detection of 86,87,88Sr- in Si requires HMR because of interferences from Si3. A
lower limit on the RSF is >1x1026 cm-3. Sr can be detected well in Si using SrSi-. 86,87,88Sr
have interferences in GaAs from 69Ga16O1H-, 71Ga16O-, and 71Ga31P1H-, respectively,
precluding the detection of Sr- in GaAs without HMR. Sr- is detected in both GaP and InP,
however.
In Si, Y requires HMR because of the Si3 interference.
In Si and in GaAs, Ba work well. See Fig. Ba-Sc.
In Si, La works well. In GaAs, 139La requires HMR because of the strong interference
from 69Ga2H.
When no negative secondary atomic ions could be detected (< 3 cts/s), from either an
implant in Si or from sputtering of the elemental matrix, the electron affinity is considered to be
zero or  in this work. From extrapolation of plots of yield vs known electron affinity, the upper
limit of this  determination is probably 0.03 eV. The elements for which no secondary ions
were detected under either condition are N, Mg, Mn, Zn, Cd, Dy, Hg, and all of the noble
gases except He. Mg, Mn, Zn, Cd, and Dy were studied as both solid elements and as
implants into Si. Hg was studied as HgTe and HgCdTe and as an implant into Si. N was
studied as an implant into Si and as BN and Si 3N4. The five noble gases (less Rn) were
studied as high fluence implants into Si (3x10 15 to 5x1017 cm-2). No secondary negative ions
were observed for any of the noble/rare gases (See below). Significant negative ion intensities
were measured for Be, Ca, Sr, Ba, Sc, Y, La and many of the lanthanide rare earths, Hf, Th,
and U. The existence of He- was first documented by Hiby as reported by Moiseiwitch (1965),
Section 3 (See Bibliography of preceding chapter). The value for EA of He - is 0.078 eV.
The lifetime for 'existence' criterion here is the time from creation of a negative ion at or
near the sputtered surface to impact at the detector in a CAMECA SIMS instrument, which is
about 1 s. Negative atomic ions with significantly shorter lifetimes would not be measured in
these experiments.
Relative Sensitivity Factors (for SIMS), the inverse of ion yield
The SIMS RSF (relative sensitivity factor) is defined via the simple expression
impurity = (impurity/matrix) RSF
where  is the secondary ion intensity. When the impurity is treated as the matrix, RSF for the
matrix is the atom density of the matrix, for example, 5.0x10 22 cm-3 for Si or 2.2x1022 cm-3 for
either As or Ga in GaAs, by definition of the RSF.
The different ion yields for different matrices ("matrix effect") result from the different
sputtering rates under constant sputtering conditions (atoms sputtered per unit time), and the
different ionization rates for the impurities (per unit area) in the different matrices (I i - Im) where
Ii is now constant and Im varies (in contrast to the usual constant Im and varying Ii for the RSF
patterns of Ref. 2). and the varying intensity of cesium or other yield-enhancing atoms that
may be involved with the sputtering surface from which the negative ions are emitted. The
relative sputtering rates under various SIMS conditions can be measured. 2 Values of relative
sputtering rate are known for most elements under various experimental conditions and can
be adjusted for most elements when a few experimental points are measured. The varying
values of Im in the  ~ exp[Ii - Im] expression are known and can be substituted. The only
remaining factor and the one that is most difficult to quantify is the effect of the third element
yield enhancer, cesium. Good vacuum and pure elements help here.
The classical definition of electron affinity (EA here) is a measure of the affinity of an
atom or molecule for electrons - a measure of the strength of the bond associated with the
attachment of an electron to the atom or molecule in vacuum; that is, with no nearby atoms or
molecules. In the case that we consider here, the EA is associated with the probability of an
electron from a solid surface being attached to an atom or molecule when that particle
escapes through the surface of a solid that has electrons available for this process. The
question arises as to how this latter electron attachment compares with that in a vacuum, and
whether they are the same; and, if not the same, how they differ.
Secondary ion mass spectrometry Relative Sensitivity Factors (RSF) are basically the
inverse of relative ion yields. RSFs are determined from experimental measurements - often
from ion implanted standards. RSFs are determined from measurements of the relative
secondary ion yields from bombardment of a surface with ions. For negative secondary ions,
the surface is usually bombarded with Cs ions, which then are incorporated into the surface of
the bombarded material and enhance the emission of negative ions sputtered from the
composite surface (target matrix plus Cs atoms). This enhancement is made possible by the
low ionization potential atoms of Cs mixed into the surface through which the sputtered atoms
(becoming negative ions) pass to escape from the surface. The relative yield of secondary
negative ions is a function of the electron affinity of the sputtered ions (atomic and molecular).
Thus measurement of the relative yields of negative secondary ions might be a measure of the
electron affinities of the sputtered species - directly or strongly related.
We have made measurements and calculations that address this issue.
We have
calculated these 'solid state' electron affinities (EA) from data for nearly all of the elements (as
atoms and as certain molecules) sputtered from the surface of a variety of materials,
especially Si. In order to do this, we determined a relationship between EA and yield of
secondary negative ions by using published values of vacuum EAs. We then related these
EAs to measured relative negative ion yields via the solid matrix (plus Cs, which supplies most
of the electrons for attachment). We found that a good relationship (that fits the published
values of vacuum EAs for the elements) is that the secondary negative ion yield varies as the
EA to the power 4.5. Thus we write the relationship as: EA(element)4.5 = [C/RSF of the
element]. For the case where Si is the solid matrix, the value of C is determined from the
value of the RSF for the matrix Si, which is the atom density of Si (5.0x10 22 cm-3), and the
published vacuum value of EA for Si, which is 1.39 eV. The units of RSF are also cm -3. The
value of C for Si is then 2.2x1023 cm-3, and the relationship becomes: EA(element) =
[2.2x1023 cm-3/RSF of the element in Si]1/4.5, or EA(element) = [2.2x1023/RSFSi]0.222. Here,
element can also be a molecule (negative molecular ion), for example ESi-. The value of 4.5
for the dependence of secondary ion yield on EA is not definite, but the calculated values of
EA are changed little if the values of 4.0 or 5.0 (or even 3.5 or 5.5) are substituted for 4.5.
The values calculated in this manner can then be compared with the published values
of EA for the elements in a vacuum, which we do in this section. Our values are thus
measured (experimental) values of the relative 'solid state' electron affinities for elements and
molecules. We consider certain molecules of interest for SIMS because of the enhanced yield
of negative ions and therefore increased detection sensitivities for SIMS analyses. We also
consider other matrices for which we have numerous experimental data, such as GaAs.
Published values of vacuum EAs are not all the same and are given in various literature
sources (reference - journals). The values used here were determined from the literature as
described in Ref. 2.
Saturation - An upper limit on the magnitude of our measurements is caused by
saturation of negative ion emission - when all of the secondary atoms are ionized - at an EA of
~2. Vacuum EAs go as high as 3.4 eV; the elements here are F, Cl, Br, I, Pt, and Au.
We estimate that the accuracy of our measurements is ±0.15 eV.
We implanted the elements that have low values of electron affinity into Si to fluences
between 1 and 5x1015 cm-3, which produce peak atom densities in the Si matrix of about 10 20
cm-3 (just less than 1% concentration), and measured their relative sensitivity factors for
negative atomic secondary ions and selected molecular ions under Cs ion bombardment for a
standardized set of SIMS conditions.
These conditions produce 1x108 cts/s of 28Si- ions under the condition of 75V offset,
used to essentially eliminate interference from all secondary ions at the masses of the
negative atomic ions of weak intensity that we are attempting to identify and quantify. For
elements with more than one isotope, identification and quantification are aided by the isotope
fingerprints of the elements. We assumed that all atomic ions have approximately the same
ion energy distribution, so that only small errors may result from the use of relative atomic ion
intensities under the condition of 75V offset.
References
1. R.D. Mead, A.E Stevens, and W.C. Lineberger, "Electron affinities," Gas Phase Ion
Chemistry, vol. 3, M.T. Bowers, Ed. [Academic Press, Orlando, 1984]
2 R.G. Wilson, F.A.Stevie, and C.W. Magee, Secondary Ion Mass Spectrometry [Wiley,
NY, 1989]
3. D.S. Simons, P.H. Chi, P.M. Kahora, G.E. Lux, J.L. Moore, S.W. Novak, C. Schwartz, S.A.
Schwarz, F.A. Stevie, and R.G. Wilson, "Are RSFs transferrable among SIMS
instruments?," SIMS VII, Benninghoven, Evans, McKeegan, Storms, and Werner, Eds.
[Wiley, NY, 1990] pp. 111-14
Table EA-Si. SIMS Relative Sensitivity Factor (RSF, cm-3), Ion Yield Enhancement (), and
Estimated 'Solid State' Electron Affinity (EA, eV), for Atomic and Molecular Secondary
Negative Ions of Elements (E) from 14.5-keV Cesium Ion Bombardment of Si. E
represents any element.
______________________________________________________
E
EESiPubl. vac. Estm.
RSF
EA
RSF
EA

EA____EA___
1 H
23
23
3.6x10
0.90
6.8x10
0.78 <
0.75
1
2 H
23
3.8x10
0.89
1
1.2x1025
0.404 1.2x1024 0.69 10
0.62
3Li
25
1.8x10
0.365
3Li
4Be
2.7x1027
2.6x1024
5B
4.9x1022
6C
ND
7N
2.7x1022
8O
5.1x1021
9F
25
11Na 5.6x10
12Mg ND
25
13Al 1.4x10
0.123
0.58
1.39
0*
1.59
2.31
0.29
0*
0.40
[5.0x1022]
23
15P 1.4x10
22
16S 1.0x10
21
17Cl 8.2x10
26
19K 1.1x10
27
20Ca 1.1x10
26
21Sc 2.1x10
26
22Ti 2.5x10
3.2x1025
23V
1.39
1.10
1.99
2.08
0.25
0.15
0.22
0.21
0.33
14Si
24Cr
25Mn
26Fe
27Co
28Ni
29Cu
30Zn
4.1x1024
ND
1.6x1025
4.3x1024
6.3x1023
5.0x1023
ND
3.0x1023
1.5x1023
9x1022
2.0x1022
1.6x1025
1.0x1023
3.6x1024
6.2x1023
6.1x1023
6.7x1022
3.5x1023
4.2x1023
6.9x1024
1.7x1024
1.0x1024
1.9x1024
1.4x1024
0.31
3.1x1023
<0.15
1.7x1024
0.38
1.5x1024
0.52
1.5x1024
0.82
2.9x1023
0.83
3.1x1023
<0.1
1.6x1025
0.93 9000 ?
1.09 17
0.28
1.22 <
1.26
1.70 large 0
0.39 <
1.46
1.19 <
3.40
0.48 15
0.55
0.82 large 0
0.83 23
0.44
1.30
0.90
0.87
0.47
0.64
0.70
0.62
0.66
2
<
<
16
647
210
131
23
1.39
0.75
2.08
3.62
0.50
? 0.15
0.19
0.08
0.53
0.93
0.64
0.65
0.65
0.94
0.93
0.39
14
large
40
3
2
1.6
large
0.67
0
0.16
0.66
0.83
1.23
0
0.12
31Ga
9.7x1025
32Ge
1.5x1023 1.09
3.6x1023 0.89
6.4x1021 2.19
6.6x1021 2.18
>1x1026 <0.25
6.8x1026 0.17
1.5x1026 0.23
1.0x1025 0.43
4.6x1024 0.89
33As
34Se
35Br
37Rb
38Sr
39Y
40Zr
41Nb
0.26
1.1x1024
0.70
88
0.30
7.1x1022
2.0x1022
1.7x1023
3.0x1023
1.5x1025
2.3x1024
2.9x1024
1.7X1023
1.5X1023
1.29
1.70
1.06
0.93
0.37
0.59
0.56
1.06
1.10
2
18
<
<
>10
296
52
59
31
1.2
0.81
2.02
3.36
0.49
?
0.31
0.43
0.89
0.75
1.05
Interf = Si4 -> HMR
0.17
42Mo
2.0x1025
24
44Ru 4.3x10
0.37
0.52
2.3X1023
4.7x1023
0.99
0.84
87
9
1.5x1024
24
46Pd 6.3x10
23
47Ag 2.0x10
ND
48Cd
26
49In 2.8x10
23
50Sn 1.9x10
23
51Sb 2.6x10
21
52Te 9.6x10
0.65
0.48
1.02
5.0x1023
6.5x1023
2.6x1023
1.3X1025
~1x1024
1.0x1023
2.6x1022
1.3x1023
0.83
0.79
0.96
0.40
0.71
1.19
1.61
1.12
3
10
1.4x1023
1.10
<
4.6x1024
4.0x1024
2.0x1024
7.3x1023
0.50
0.53
0.61
0.77
4.8x1024
3.4x1024
3.0x1024
0.50 l20+ ?
0.54 1500 ?
0.56 5
?
<0.15
0.09
0.38
0.67
0.66
0.58
0.55
0.64
0.64
0.61
0.18
0.16
<0.15
0.12
0.17
<0.11
<0.11
45Rh
53I
55Cs
56Ba
57La
58Ce
59Pr
60Nd
4.6x1021
2.36
Interf = Cs beam
4.0x1026 0.19
3.0x1025 0.33
7x1025
0.26
>8x1026 <0.14
>6x1026
27
62Sm 5x10
25
63Eu 1.5x10
64Gd
65Tb
66Dy
67Ho
68Er
68Er
69Tm
0.20
1.03
0.96
2.00
1.4x1026
4.3x1026
>6x1026
1.5x1027
3.5x1026
>2x1027
>2x1027
<0.15
0.089
0.382
0.178 1.3x1024
0.165 1.4x1024
<0.15 2.5x1024
0.121 3.3x1024
0.173 1.6x1024
<0.11 1.6x1024
<0.11 2.1x1024
1.14
0.56
<
1.30
large 0
280 0.30
2
1.15
10
1.07
<
1.97
20V offset
3.06
0.47
87
?
0.19
7.5 0.5
35
?
0.26
1000+ ?
<0.14
108 ?
307 ?
240+ ?
450 ?
220 ?
1200+ ?
950+ ?
70Yb
71Lu
72Hf
73Ta
74W
75Re
76Os
77Ir
78Pt
79Au
3x1026
0.181
2.2x1024
0.60
136
4.4x1026
0.164 5.7x1024
0.49 77
7x1026
0.17
8.7x1023 0.74 800
1.4x1026
0.24
2.5x1023
0.97 560
4.0x1024 0.53
9.4x1022
1.21 42
>2x1027 <0.13 >1.4x1026 <0.24 ~10
1.4x1024 0.66
1.8x1023
1.04 8
8.4x1022
1.31
1.0x1023
1.19 <
6x1022
1.33
4.8x1023
0.84 <
1.1x1022 1.94
1.4x1023
1.10 <
?
0.18
?
?
0.32
0.82
0.12
1.12
1.57
2.13
2.31
0.16
0.17
>1.5x1028 <0.08
27
0.15
81Tl 1.0x10
4.7x1025
6.7x1024
0.30
0.47
300+ 0
150 0.30
2.9x1025
24
83Bi 1.9x10
25
90Th 6.5x10
26
92U 1.1x10
2.6x1023
1.1x1023
3.2x1023
9,4x1023
096
1.19
0.92
0.72
110
17
200
115
80Hg
82Pb
0.34
0.62
0.29
0.25
0.37
0.95
?
?_
0.29
0.25_
ND means not detected
> means greater than
< means no ion yield enhancement; molecular RSF is greater than atomic RSF
- means approximately no ion yield enhancement; molecular RSF is approximately the same
as atomic RSF
estimated from SIMS RSF measurement - negative ion emission from a solid surface, vs
value for an atom in vacuum
Estimated accuracy of electron affinity: within ~0.2 eV of stated value
Table EA-GaAs. SIMS Relative Sensitivity Factor (RSF, cm-3), Ion Yield Enhancement (),
and Estimated 'Solid State' Electron Affinity (EA, eV), for Atomic and Molecular
Secondary Negative Ions of Elements (E) from 14.5-keV Cesium Ion Bombardment of
GaAs
____________________________________________________________________
E
EEAsEGa- or other
Publ.
H
Li
Be
B
C
RSF
3.3x1022
1.1x1024
7.3x1025
2.0x1023
3.2x1021
EA
0.74
0.33
0.13
0.49
1.24
RSF
2.0x1022
6.5x1023
1.0x1022
6.7x1020
1.8x1020
EA
0.96
0.47
1.12
2.05
2.74

RSF
1.7
1.0x1022
1.7
7300
ND
300
ND
18
EA
1.12
___ EA__
3
0.75
0.62
small
0.28
1.26
N
ND
0
3.3x1023
0.52
large
4.2x1022
large 0
O
F
Na
Mg
Al
Si
P
S
Cl
4.2x1020
1.5x1020
3.1x1024
ND
4.0x1023
5.5x1021
5.6x1021
2.7x1020
7x1019
1.96
2.45
0.27
0
0.42
1.10
1.10
2 .15
2.90
1.0x1022
1.6x1022
7.1x1023
4.4x1022
3.1x1021
3.0x1020
2.2x1022
1.0x1021
1.1x1022
1.13
1.01
0.44
0.81
1.46
2.45
0.94
1.88
1.10
<
<
4.3
large
129
18
<
<
<
K
Ca
8x1024
1.0x1025
0.22
0.21
1.3x1024
2.4x1023
0.38
0.56
6
42
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
1.2x1025
2.4x1025
3.0x1024
1.3x1024
ND
2.2x1024
4.1x1023
1.4x1023
0.20
0.17
0.27
0.33
0*
0.29
0.42
0.54
5.0x1022 0.79 52
2.6x1023 0.55 92
1.4x1023 0.63 21
6.0x1023 0.45 2
1.8x1020 ? 3.56 large
6.2x1023 0.45 3.5
1.4x1023 0.63 3
2.1x1023 0.57 <
0.19
0.08
0.53
0.67
0
0.16
0.66
1.16
Cu
Zn
Ga
Ge
As
Se
Br
Rb
Sr
4.1x1023 0.42
ND
0
[2.2x1022] 2.3X1022 0.94
[2.2X1022] 0.81
2.0X1020 2.29
1.6X1020 2.42
Interf = GaO
3.1x1023
0.45
4x1023
3.0x1022
0.50 1
0.88 large
3.9x1020
2.31
1.1X1021 1.84 <
1.5X1022 1.03 <
Interf = GaAsO
Interf = GaAsO
1.23
0
0.30
1.20
0.81
2.02
3.36
0.49
?
Y
Zr
Nb
Mo
Ag
Cd
In
~4x1024
5.0x1023
Interf
2.2x1024
4.9x1022
ND
2.0x1024
4.7x1022
4.8x1021
2.3X1022
6X1023
5.0x1022
1.8x1023
6.0x1021
0.25
0.40
0.29
0.68
0
0.30
1.25
1.32
0.94
0.45
0.79
0.59
1.26
59
85
104
large
4
1
large
333
1.2x1021
0.82
3.5x1024
8.7x1023
2.0x1023
6.3x1022
7.2x1020
7x1024
1.80 <
0.31
0.62
0.96
0.75
2.02
1.46
3.40
0.55
large 0
0.5
0.44
<
1.39
<
0.75
<
2.08
3.62
0.26 1.1
5.5X1022
0.77
<
2.4X1020
8X1022
2.57
0.71
<
<
7.9x1023
0.43
<
6x1023
0.45
3.7
5.7x1023 0.46
3.5
0.50
?
0.31
0.43
0.89
0.75
1.30
0
0.30
Sn
1.3x1022
0.91
3.8x1020
2.32
90
1.15
Sb
Te
I
Cs
Ba
La
Nd
Tb
Dy
2.6x1022 0.78
2.6x1020
2.17
2.6x1020 2.17
Interf=beam
>2x1025 <0.18
1.8x1024 0.30
>1x1024 <0.35
ND
?
ND
?
1.0x1022
1.5x1021
1.3x1022
1.12
2.05
1.06
2.6
<
<
1.07
1.97
3.06
3.1x1023
1.5x1023
4.8x1022
6.1x1022
7.6x1022
0.52
0.62
0.79
0.75
0.72
>7
12
>20
large
large
?
0.51
?
?
?
1.1x1023
5.4x1022
0.66
0.77
large
large
?
?
0.59
0.44
1.15
1.25
1.00
0.32
0.42
large
large
large
ND
large
23
9.6x1023 0.41 <
>50
large
~200 Interf at 208
<
Interf=Ga4
?
?
?
0.32
0.82
0.12
0.30
0.37
0.95
Ho
Er
Interf
Interf
?
?
Tm >2x1027
<0.06 1.9x1023
Yb >1.2x1026 <0.12 7x1023
Hf
ND
?
9x1021
Ta
ND
?
6.2x1021
W
3.9x1023
0.43
1.7x1022
Re >1.4x1026 <0.11 2.8x1024
Tl
Interf=GaCs ?
4.9x1022
Pb
1.3x1024 0.33
7.5x1021?
Bi
3.7x1023? 0.43? 1.2x1024?
0. 39
3.9x1020
3.9x1022
2.31
0.83
<
<
0.47
Th
1.5x1024
0.32
?
U
3x1024
0.27______________
?_____
ND means not detected
> means greater than
< means no ion yield enhancement; molecular RSF is greater than atomic RSF
Estimated from SIMS RSF measurements - negative ion emission from a solid surface vs
from an atom in vacuum
Estimated accuracy of electron affinity: ±0.15 eV of stated value
EA-Summary: Electron Affinities (Solid State) of the Elements Sputtered from
Si and GaAs and Vacuum Values from Literature (all in eV)
_____________________________________________________________________
Elem Si
GaAs Vac.
Elem Si GaAs Vac. Elem Si GaAs Vac.__
H
0.89 0.74 0.75
Zn
0
0
0
Pr <0.14
? <0.5a
Li
0.40 0.33 0.62, 0.39b Ga
0.26 0.30
Nd <0.15f ? <0.5a
Be 0.12 0.13 ?
0.24a Ge
1.09 0.94 1.2
Sm 0.09
? <0.5a
B
0.58 0.49 0.28, 0.57d As
0.89 0.81 0.81
Eu 0.38
? <0.5a
C
1.39 1.24 1.26
Se
2.19 2.29 2.02
Gd 0.18g? <0.5a
N
0
0
0
Br
2.18 2.42 3.36#
Tb 0.17
? <0.5a
? <0.5a
? <0.5a
O
F
1.59
2.31
1.96
2.45
1.46
3.40#
Rb <0.25
Sr
0.17
?
0.45
0.49
? 0c
Dy <0.15
Ho 0.12
Na
Mg
Al
Si
P
S
Cl
K
0.29
0
0.40
1.10
1.99
2.08
0.25
0.27
0
0.42
1.10
1.10
2.15
2.90
0.22
0.55
0
0.44
1.39
0.75, 1.12b
2.08
3.62#
0.50
Y
Zr
Nb
Mo
Ru
Rh
Pd
Ag
0.23
0.43
0.89
0.37
0.52
0.65
0.48
1.02
0.25
0.40
0.29
0.68
0.31
0.43
0.89
0.75
1.05
1.14
0.56
1.30
Er
Tm
Yb
Lu
Hf
Ta
W
Re
<0.11h
? <0.5a
<0.11 ? <0.5a
0.18
? <0.5a
0.16
? <0.5a
0.17e
? >0a
0.24
0.32
0.53 0.43 0.82
<0.13 <0.11 0.12
Ca
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
0.15
0.22
0.21
0.33
0.31
0
0.38
0.52
0.82
0.21
0.20
0.17
0.27
0.33
0
0.29
0.42
0.54
?*
0.19
0.08, 0.2a
0.53
0.67
0
0.16, 0.25a
0.66
0.83
Cd
In
Sn
Sb
Te
I
Cs
Ba
La
0
0.20
1.03
0.96
2.00
2.36
0.19
0.33
0
0.30
0.91
0.81
2.53
2.53
<0.21
0.35
0
0.3
1.15
1.07
1.97
3.06#
0.47
?*
0.5
Os
Ir
Pt
Au
Hg
Tl
Pb
Bi
Th
0.66
1.31
1.33
1.94
<0.08
0.15
0.34
0.62
0.29
?
0.33
0.43
0.32
1.12
1.57
2.13#
2.31#
0
0.30
0.37
0.95
?
Cu 0.83 0.42 1.23
Ce
0.26 <0.40 ?
U 0.25 0.27 ?_____
* reported detected in Hotop and Lineberger, Ref. 2 of first chapter of this section
# limited by saturation
a Hotop and Lineberger, Ref. 2 of first chapter of this section
b reported in Branscomb, Ref. 6 of the first chapter of this section
c calculated
d <0.3 eV from sputtering from bulk Dy
e 0.11 eV from sputtering from bulk Hf
f 0.064 eV from sputtering from bulk Nd
g 0.21 eV from sputtering from bulk Gd
h <0.037 eV from sputtering from ErF
J 1.70 eV from sputtering from bulk Pt
3
Comparisons between the data of Table EA-Summary (Solid State) and published (vacuum)
values of Electron Affinity
Agree on EA = 0: (10) N, Mg, Mn, Zn, Cd, Hg, Ne, Ar, Kr, Xe
Agree on EA not equal to zero: (11) Li (*), B (*), C, P (*), S, Sc, As, Zr, Nb, Pd, Hf
Close agreement (within about 20%): (17) H, O, Al, Si, Ti (*), Ni, Ga, Ge, Se, Y, In (?), Sn, Sb,
Te, Re, Ir, Pb
Large differences (>20%)(all less than published values): (16) Be, Na, K, Cu, Ag, [Au (sat)], V,
Cr, Co, Ru, Rh, La, Ta, W, Os, Tl, Bi
Note that these elements include those from columns 1 and 11 (1A)
Large difference (>20%)(greater than published value): Fe, an anomaly in this sense
Saturated (in SIMS sputtering - measurement): (6) F, Cl, Br, I, Pt, Au
Special: (1) He - zero here, but known to be 0.078 eV
No answer: (2) Rb (SIMS interferences), Cs (cannot do because is sputtering ion beam)
New values given here: (18) See below.
Observations and conclusions from the data in Table EA-Summary:
Elements that have zero vacuum EA and also zero solid state EA are: N, Mg, Mn, Zn, Cd, Hg,
and probably Dy (See Chapter on Stoichiometric EAs.)
Values of EA (solid state) from this work for elements for which there are no published
vacuum EAs are:
Non rare earths:
Be 0.12 to 0.15 eV
Hf 0.17 eV
Alkaline earths:
Ca 0.15 (0.24) eV
Sr 0.17 eV
Ba 0.19 eV
Lanthanide rare earths: Ce 0.26 eV
Pr <0.14 eV
Nd <0.15 eV
Sm 0.09 eV
Eu 0.38 eV
Gd 0.18 eV
Tb 0.17 eV
Dy ~0 eV
Ho 0.12 eV
Er <0.11 eV
Tm <0.11 eV
Yb 0.18 eV
Lu 0.16 eV
Actinide rare earths: Th ~0.3 eV
U ~0.3 eV
The element Eu stands out for having the highest value of EA in this last list. No signal
was detected for Pr, Nd, Er, or Tm, but the limits of detection are given and are not sensitive
enough to say that the EAs are zero. The data from Chapter: Stoichiometric EAs are sensitive
enough to say that the EA for Dy is probably zero.
Molecules
Under Cs bombardment of Si, the species 75As28Si- has a significantly higher
secondary ion yield than the atomic ion 75As-, and provides a better detection limit for As. A
molecular ion may also provide a better detection limit when the mass of the atomic ion
contains a background contribution, such as the limitation for C analysis caused by residual
gas contributions, which can be minimized by profiling CAs - instead of C- for analysis in GaAs
using Cs primary ion bombardment. Another application of molecular ions is the use of ECs +
where E is the impurity element of interest, under Cs ion bombardment to carry out major and
trace element analyses with less secondary ion intensity variation than is encountered for
other modes of SIMS analysis, and to provide improved detection for the noble gas elements
Molecular enhancement
For negative secondary ions from Si under cesium bombardment: data are given in
Table EA-Si for 75 elements. 36 have large enhancement (50 to 9000), and 26 are enhanced
somewhat (2 to 50). 62 of 75 are enhanced, which is more than 80%.
For negative secondary ions from GaAs under cesium bombardment: data are given in
Table GaAs-EA for 58 elements. 27 are greatly enhanced (50 to 9000), and 18 are enhanced
somewhat (2 to 50). 45 of 58 are enhanced, which is more than 75%.
III-V combinations are greatly enhanced as EAs- or have high electron affinities in the
III-V matrix, GaAs, e.g.:
Ion Enhancement
BAs300
AlAs
130
InAs330
Est. EA
2.05 eV
1.46 eV
1.26 eV
CHAPTER: Negative secondary ion mass spectrometry of elements sputtered from their
own bulk solid state matrices, and their
estimated 'stoichiometric' electron affinities
Elements with published zero or low value of electron affinity (EA) do not form
significant numbers of negative ions in vacuum by electron attachment. Studies of negative
secondary ion yields during sputtering of these same elements from a matrix in which they are
immersed, and therefore in an environment in which they can interact with atoms of the same
or different elements during the ionization/sputtering process, indicate that some of these
elements produce more substantial quantities of negative secondary ions than might be
expected from published values of vacuum EAs. For example, some elements with quoted
zero values of vacuum EA produce measurable quantities of negative secondary ions when
sputtered from surfaces that contain atoms of other elements and the element of the
sputtering ion, or even from their own bulk material. The electron attachment process may be
different during the energetic interaction of the atoms of the element of interest and the other
dissimilar atoms of the matrix and sputtering ion beam because of the potentials (chemical,
work function) of the complex surface from which they are emitted during the sputtering
process.
In secondary ion mass spectrometry (SIMS), the detection and quantification of impurity
atoms in a matrix depends on their ability to form positive or negative ions. Knowledge of the
relative ability to form negative ions during sputtering from different or complex surfaces
(various matrix atoms plus atoms of the sputtering beam) is important in SIMS applications,
and ways to enhance the detection of these low yield elements during negative ion mass
spectrometry is important. For example, measurement of negatively charged molecules that
include atoms of the element of interest and that may more readily form negative ions may
enhance the detection of these elements.
The energies and path lengths of ions between creation and detection place an
approximate lower limit of 1s on the lifetime of these negative ions. True self-sputtering,
where the matrix and the sputtering ion are the same is difficult or impossible to carry out for
all elements of the periodic table. Solid matrices of the gaseous, liquid, and high vapor
pressure elements are difficult to study.
The data presented here can serve as an aid to the user of SIMS for detection
sensitivity and quantification of negative SIMS, as well as a measure of values of electron
affinity of elements when sputtered from the surface of a solid (that is, in the presence of other
atoms as opposed to the vacuum values of an isolated atom), to within about 0.15 eV.
Here we call these values of EA the 'solid state' EAs to differentiate them from the
published vacuum EAs. How these two sets of values compare can be determined through a
study of the data presented here.
Elements of special concern here are those with quoted zero value of vacuum EA and
values of EA below about 0.3 eV, that is, elements that would not be expected to produce
large quantities of negative secondary ions. In the preceding chapter, the atoms of the
elements of interest are immersed in a matrix (<1%). We have also sputtered these elements
from a solid material that is either 100% the atoms of interest or a compound that contains a
significant fraction of the atoms of interest, for example, Si 3N4 to study N. In this case, the
negative yields were standardized by measuring Si with each other element under identical
analysis conditions, thereby allowing data for all elements to be intercompared via the
common yield for Si.
In no case did we observe negative ions of Mg, Zn, Cd, Hg, Mn, N, Dy, Er, or the noble
gases. Pr and Tm are also possibilities, with an upper limit of 0.05 eV for their EAs.
Atomic negative ions could not be detected for some elements in this work because
practical implantation fluences did not produce sufficient atom density in the solid target.
While some elements could be implanted to stoichiometric densities (>10 22 cm-3), others
could be implanted conveniently only to densities of 10 20 or 1021 cm-3. When atoms are
sputtered from a solid matrix of the pure element or a compound that contains more than
about 10% of the element of interest, then the atom density is of the order of 10 22 cm-3 and
maximum detectivity was possible.
When elements with EA greater than 1 eV, e.g. Si, are sputtered using Cs primaries
with 75V offset, the negative ion intensity can be 3x10 8 ct/s in the CAMECA sector magnet
instruments used in this work. When Zn or Cd are sputtered under identical conditions, no
secondary negative ions were detected. If about 3 cts/s is considered to be a valid detected
signal, then the secondary ion signals of Zn, Cd, Mg, N, Dy, etc., are less than 10 -8 compared
with Si, etc. This ratio can be reduced to less than 10 -9 by eliminating the 75V offset, but with
the risk of allowing a molecular interference to cause an error.
The relative intensities of negative secondary ions of elements sputtered from their own
matrix should probably be treated in the same manner to estimate values of EA because the
secondary intensities/yields are a function of the relative sputtering rates and the different
enhancements of negative ions from each of the elemental matrices (the matrix effect). The
sputtering rates of each element would need to be measured, which would be difficult for the
ones without well polished surfaces in which a crater depth could be accurately measured
using surface profilometry.
The detection of negative ions sputtered from elemental targets does establish their
existence, and allows an estimate to be made of their electron affinities, especially for the rare
earth elements, assuming that all rare earth elements have approximately the same
enhancement properties for negative ions, because their chemistry is approximately the same.
As an example of such data, values of relative intensities of elemental and selected simple
molecules for a few selected elements (Sc, Gd, Dy, Hf) with quoted low values of electron
affinity are given in Table Low EA, obtained under experimental conditions that produce about
1x108 cts/s of Si-. Sc is seen to produce significant Sc-, so should have a relatively high value
of EA. The yield from Gd is less, from Hf, smaller still, and from Dy, undetectable, implying
that the EA of Dy is probably near zero (<0.03 eV). In cases where E- (E being any element)
is not detected, EH- is almost always detected, as is seen for Dy in Table Low EA. The
negative ion mass spectra for Sc metal, Gd metal, Dy metal, and Hf metal are shown in figures
included in Chapter: Rare Earths of Section: Mass Spectra.
The data obtained here for sputtering of atomic negative ions and the EH - and EOmolecules (in some cases) are given in Table: Stoichiometric EAs. These data are corrected
for the fraction of the element of interest compound material used, but have not been adjusted
for any sputtering yield (rate) differences from Si, the material used to adjust all SIMS
conditions to be identical for the other materials as they were studied (different times and
instruments).
Table Low EA. Secondary negative ion intensities (cts/s) for selected rare earth and low EA
elements sputtered from their own matrices (relative) to 1x10 8 ct/s of Si
___________________________________________________
E
EEH-_______EOEA (eV)__
5
5
3
1.7x10
1.0x10
7x10
0.39
21Sc
1.3x104
7.6x102
3x104
0.22
64Gd
ND*
2.4x103
3x103
<0.04
66Dy
___ 72Hf
5.4x102
2x102
7x101
0.11_____
*<2x100
The relative magnitudes of the intensities of negative secondary ions in the columns of
Table Low EA might lead to conclusions about the (relative) affinities of these elements for H
and O. Sc has a stronger affinity for H, compared with the other three. For O, Gd has an
order of magnitude greater affinity for O than Sc and Dy, and Hf has little at all. Hf is an
apparently relatively inert element.
Table Stoichiometric EAs. Secondary negative ion intensities (cts/s) and estimated 'solid
state' electron affinities (EA, eV) for selected rare earth and low electron affinity
elements (E) plus others sputtered from their own matrices or from a material that
contains a stoichiometric amount of the element (all relative to ~1x10 8 Si-). The
materials are indicated. Units of EA are eV and are cts/s for all intensities (EX).
_____________________________________________________________________
E
Mat'l
EEHEOEA____Other Vac. EA_
3Li
4Be
LiYF4
LiNbO3
Be
2.7x105
1.9x105
2.1x105
1.5x105
1.9x105
2.2x105
2.0x105
graphite
5.3x107
6C
BN
NDa
7N
Si3N4
NDc
7N
KTaNbO3 8.8x107
8O
BaSrTiO3 9.2x107
NDc
12Mg Mg
13Al
20Ca
21Sc
22Ti
22Ti
23V
24Cr
25Mn
Al
Ca
4.4x101
NDa
3.8x105
2.2x103
Sc
1.6x103
1.5x105
BaSrTiO3
9x104
1.7x105
8x105
Ti
6.3x105
3.4x104
V
Cr
Mn
8.1x104
6x105
1.0x104
NDb
54FeH
6.8x106
2.0x102
2.6x103
6.3x101
0.43
0.40
0.41
0.62
0.62
?
0.38
0.40
0.41
0.40
1.39
<0.028
<0.034
1.56
1.58
<0.034
?
?
?
?
1.27
0
0
1.46
0
0
0.44
?
0
1.4x104
2x102
0.06
<0.028
0.47
0.15
1.1x105
5x103
0.14
0.39
?
0.19
2.5x105
0.34
0.39
0.55
0.19
0.19
0.08
2.0x101
3.0x103
.47
.27
0.08
0.08
5.7x104
Interf
0.33
0.52
0.35
<0.031
0.08
0.53
0.67
0
interf
26Fe
26Fe
30Zn
Fe
6.9x105
2.1x104
Kovar
Zn
8.2x105
1.7x105
NDb
NDa
NDa
1.1x104
4.5x102
3.2x102
0.53
0.16
0.55
0.39
<0.031
0.16
0.16
0
0.24
0
0
0
0.3
0.21
0.18
0.32
0.13
0.3
0.3
0.3
?
0.15
0.24
0.38
0.41
0.20
<0.05
?
0.31
0.43
0.43
0.75
0
Ga
~2x104
8.3x100
GaAs
SrBaTiO3
1.1x104
5.6x103
7.5x104
1.2x103
1.7x101
LiYF4
Zr
BaZrThF4
Mo
Cd
2.2x103
1.8x104
1.5x105
2.3x105
8.3x103
<1.3x101
48Cd
CdS
<2x101
NDb
<0.031
0
0
49In
In
4.5x103
0.17
0.30
5.0x101
InP
SrBaTiO3
7.3x103
3.9x103
2.3X103
1.0x104
0.19
0.17
0.15
0.21
0.30
0.30
0.30
?
8.3x101
ZrBaThF4
2.1x104
1.8x104
0.24
0.23
?
?
31Ga
31Ga
38Sr
39Y
40Zr
40Zr
42Mo
48Cd
49In
56Ba
58Ba
2.0x104
3.2x101
5.8x104
5.0x101
57La
LaF3
7x104
1.2x101
5x100
0.32
58Ce
CeF3
Nd
4.5x104
5.2x101
2x101
5.9x102
<1x100
5.2x101
0.31
0.064
Gd
5.3x101
1.1x104
2.7x104
0.065
0.21
60Nd
64Gd
1.4x104
1.1x104
3.6x104
9.5x102
9.5x102
8.2x103
2.7x104
0.22
0.20
0.22
EF8x103
1.0x104
0.5
66Dy
Dy
NDb
2.5x103
3.3x103
<0.03
68Er
ErF3
Hf
NDa
6.5x102
4.3x101
7.5x102
<0.037
0.11
Ta
3.7x102
1.3x103
1.5x104
W
Re
Pt
1.3x104
4.8x105
7.4x104
1.3x108
81Tl
Au
TlIBr
82Pb
Pb
72Hf
73Ta
74W
75Re
78Pt
79Au
90Th
90Th
92U
Th
ZrBaThF4
U
b
<2
?
9.8x103
0.10
0.13
0.23
?
?
0.32
8x104
1.0x103
4.6x103
2.8x103
0.22
0.49
.32
1.70
0.32
0.82
0.12
2.13
1.7x107
4.8x102
1.1
0.11
2.31
0.3
4.8x102
0.10
0.37
5.8x102
5.0x102
1.5x103
5x104
~1.2x104
~6x103
0.11
0.11
0.14
0.30
0.22
0.18
0.37
0.37
0.37
?
?
?
0.13
?
1.2x103
a <1
5.5x103
1.0x104
c <3 d <4
6x103
1.0x105
1.3x103
2x104
0.21____________?_______